Citation
"The challenges are many"

Material Information

Title:
"The challenges are many" women's perceptions of cervical cancer and cervical cancer screening in rural Tanzania
Creator:
Chasco, Emily E. ( author )
Language:
English
Physical Description:
1 electronic file : ;

Subjects

Subjects / Keywords:
Cervix uteri -- Cancer ( lcsh )
Cervix uteri -- Cancer -- Prevention -- Tanzania ( lcsh )
Cervix uteri -- Cancer ( fast )
Cervix uteri -- Cancer -- Prevention ( fast )
Tanzania ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
This dissertation explores cervical cancer and cervical cancer screening within the context of a rural health care setting in Tanzania. In particular it is concerned with how cervical cancer is perceived and understood by Tanzanian women as well as how these perceptions influence attitudes towards cervical cancer screening and treatment. Cervical cancer is a major women s health issue in Tanzania and other developing countries and the resources needed to adequately address the magnitude of the problem are scare. This is slowly changing as cervical cancer prevention programs are implemented in these regions often by international non governmental organizations alone or in partnership with local government and other agencies. These programs have frequently struggled with low utilization and access barriers even where issues of cost have been addressed. Based on six months of ethnographic fieldwork conducted largely within the established cervical cancer prevention program of a rural referral hospital this research takes a layered view that examines the historical social and political environment in which cervical cancer came to be seen as a health priority in Tanzania as well as current efforts for cancer control. It seeks to understand first how Tanzanian women think about cervical cancer and cervical cancer screening but also examines issues affecting utilization related to the broader context of the cervical cancer prevention program and its place within the Tanzanian health care system. This dissertation adds to the small but growing body of literature that seeks to better comprehend prevention program utilization with the goal of increasing the adoption and expansion of screening programs in developing countries.
Thesis:
thesis (Ph.D.)--University of Colorado Denver.
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Emily E. Chasco.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
930187033 ( OCLC )
ocn930187033

Downloads

This item has the following downloads:


Full Text
THE CHALLENGES ARE MANY: WOMENS PERCEPTIONS OF CERVICAL
CANCER AND CERVICAL CANCER SCREENING IN RURAL TANZANIA
by
EMILY E. CHASCO
B.A., University of Michigan, 2004
M. A., University of Colorado Denver, 2009
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences
2015


2015
EMILY E. CHASCO
ALL RIGHTS RESERVED


This thesis for the Doctor of Philosophy degree by
Emily E. Chasco
has been approved for the
Health and Behavioral Sciences Program
by
Jean Scandlyn, Chair
Sara Yeatman, Advisor
John Brett
Deborah Thomas


Chasco, Emily E. (Ph.D., Health and Behavioral Sciences)
The Challenges Are Many: Womens Perceptions of Cervical Cancer and Cervical
Cancer Screening in Tanzania
Thesis directed by Assistant Professor Sara Yeatman
ABSTRACT
This dissertation explores cervical cancer and cervical cancer screening within the
context of a rural health care setting in Tanzania. In particular, it is concerned with how
cervical cancer is perceived and understood by Tanzanian women, as well as how these
perceptions influence attitudes towards cervical cancer screening and treatment. Cervical
cancer is a major womens health issue in Tanzania and other developing countries, and
the resources needed to adequately address the magnitude of the problem are scare. This
is slowly changing as cervical cancer prevention programs are implemented in these
regions, often by international non-governmental organizations alone or in partnership
with local government and other agencies. These programs have frequently struggled
with low utilization and access barriers, even where issues of cost have been addressed.
Based on six months of ethnographic fieldwork conducted largely within the established
cervical cancer prevention program of a rural referral hospital, this research takes a
layered view that examines the historical, social, and political environment in which
cervical cancer came to be seen as a health priority in Tanzania, as well as current efforts
for cancer control. It seeks to understand first how Tanzanian women think about cervical
cancer and cervical cancer screening, but also examines issues affecting utilization
related to the broader context of the cervical cancer prevention program and its place
within the Tanzanian health care system. This dissertation adds to the small but growing


body of literature that seeks to better comprehend prevention program utilization with the
goal of increasing the adoption and expansion of screening programs in developing
countries.
The form and content of this abstract are approved. I recommend its publication.
Approved: Sara Yeatman
IV


DEDICATION
This work is dedicated to my parents, David and Beth Chasco, whose willingness
to pack up three children under the age of ten and travel through Europe with only two
suitcases for four months quite literally made me the person I am today. Upon returning
home, when their fifth-grader announced she planned to be an anthropologist someday,
they responded not with dismissal or amusement, but encouragement. Most importantly,
their love and support has been without limits over the years.


ACKNOWLEDGEMENTS
Dissertations are rarely completed without a great deal of support, advice, and
encouragement, and in my case I was fortunate to have all three in abundance. I would
first like to express my immense gratitude to the Tanzanian women who participated in
this research project. They frequently sat down to our interviews with a long wait for
health services either just behind or in front of them, yet often their sense of humorI
received more than one teasing offer to arrange my marriage to a brother or cousin
made the time pass by quickly. Their willingness to answer our many questions with
patience and grace made this project what it is.
I also wish to thank St. Francis Hospital and St. Francis University College of
Health and Allied Sciences (SFUCHAS) for their cooperation and support. From my first
visit with Prof. Cassian Magori, the Principal of SFUCHAS, he welcomed me with open
arms, smoothing the way forward, orienting me to the various hospital departments, and
making introductions. The St. Francis administration was instrumental in assisting me
with the IRB approval process, as was my contact person at the National Institute for
Medical Research in Dar es Salaam. The staff overall were incredibly welcoming and,
particularly in RCH where space for interviews was limited, extremely accommodating.
I owe a huge debt to my friends and mentors both in the Obstetrics/Gynecology
Department and CECAP. The head of the department was a supporter from the start,
throwing me into the deep end on my first visit to St. Francis by encouraging me to come
and observe screenings that same day. The staff members of CECAP were inspiring to
watch. It could not have been easy to have an mzungu under their feet all day, but they
never made me feel an inconvenience. They constantly challenged my assumptions and
vi


pushed me to consider new angles. My contact at Jhpiego was also very generous with
her time and our conversations were thought-provoking.
A special thank you goes to my Research Assistants, Paula Madata and Rachel
Bwoki. When a friend said he had a potential RA for me, and when that RA mentioned
she was bringing along her friend to work with me too, I had no idea what to expect. But
they quickly proved themselves up to the task and became not only invaluable to the
process, but friends and guides in navigating day to day life in Tanzania.
Finding your way in a foreign country can be daunting, so on a personal level I
am so grateful to have had my ELCT Language School family (both Tanzanians and
wazungu) in Morogoro, Tanzania, and my wonderful host family in Ifakara to guide me.
My parents, sisters Erica and Claire, and friends (in particular Bethany, Alisa, Tony, and
Piya) offered incredible love and care from afar and saved my sanity during the writing
process when I arrived back in the States.
Over the many years I have spent at the University of Colorado Denver while
working on my M.A. and Ph.D., my committee members and mentors both in the Health
and Behavioral Sciences program and in Anthropology have been incredibly generous
with guidance, advice, patience, and oftentimes, much-needed nudges. I have had
wonderful opportunities to travel and grow as a researcher while working with incredible
people. In particular, I want to thank Sara Yeatman, Jean Scandlyn, John Brett, Deborah
Thomas, and Charles Musiba, who pushed me to go to Tanzania in the first place. When
my project nearly fell through, when I encountered road blocks in the field, their
encouragement and support meant a great deal. For their much-needed guidance through
the processes of graduate school and for offering moral support, I also want to thank the
vii


administrative staff at UCD with whom I have worked closely over the years, Abby Fitch
and Connie Turner.
Finally, I would like to express my gratitude to the National Security Education
Program (Boren Fellowship), Rotary International, the William Robinson/Peter Durst
Scholarship, and the Department of Health and Behavioral Sciences for the funding that
made this research possible.
viu


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION.........................................................1
Research Questions....................................................3
Theoretical Approach..................................................3
Organization of the Dissertation......................................5
II. SETTING THE SCENE....................................................8
Morogoro City and Region.............................................10
Ifakara and the Surrounding Area.....................................11
The Local Economy...............................................12
Population......................................................13
Ifakara as a Field Site..............................................14
A Center of Health Research.....................................14
St. Francis Hospital............................................15
Cervical Cancer Prevention Program...................................16
Location........................................................16
Screening Services..............................................17
III. METHODS & ANALYSIS..................................................22
Methodological Approach..............................................23
Data Collection......................................................27
Participant Observation/Observation.............................27
Participant Interviews..........................................28
Patient Information Forms.......................................33
Key Informants and Specialized Informants.......................34
Screening Map Forms.............................................35
Supplementary Information.......................................37
Analysis.............................................................38
Qualitative Data................................................38
Quantitative Data...............................................40
IV. CHANGING PRIORITIES.................................................44
ix


A Global Picture......................................................45
Health Transitions...............................................48
The Macro-Policy Environment.....................................50
Transferable Technologies........................................52
Cervical Cancer in Tanzania...........................................55
NGO and Government Partnerships.......................................57
V. LOCAL CONSTRUCTIONS OF CERVICAL CANCER................................61
Understanding Cervical Cancer in the Local Context....................62
Accessing Familiar Biomedical Models of Infection.....................65
Infections, Microbes, and other Causes of Cervical Cancer........67
Perceived Symptomology...........................................73
Explanations of Risk..................................................76
Sexual Behavior..................................................77
Childbirth.......................................................80
Prevention and Prognosis..............................................84
HIV, Cervical Cancer and Ugonjwa wa Kisasa............................90
VI. UTILIZING SCREENING SERVICES..........................................95
Cervical Cancer: A Disease of Poverty.................................97
The Benefits of Screening.............................................99
Barriers to Screening................................................103
Barriers Coming to the Hospital.................................104
Barriers at the Hospital........................................109
Overcoming Barriers..................................................113
VII. STIGMA, SHARING, AND SUPPORT.........................................117
Cervical Cancer and Stigma...........................................119
Sharing the Screening Decision.......................................122
The Role of Husbands............................................124
Relatives, Neighbors, and Fellows...............................127
Sharing a Cervical Cancer Diagnosis..................................132
Fear of Stigma or Rejection.....................................132
Marshaling Resources............................................135
x


A Heart of Sharing.......................................140
Negotiating Stigma and Support...............................142
VIII. COMPETING PRIORITIES........................................146
CECAPs Challenges...........................................148
Low Utilization..........................................148
A Program in Transition..................................152
On the Ground................................................153
Hospital Context.............................................157
Larger Program Goals.........................................160
Shared Goal, Different Priorities............................162
IX. CONCLUSION..................................................166
Limitations..................................................170
Future Recommendations.......................................172
Moving Forward...............................................175
REFERENCES..........................................................178
APPENDIX
A. LIST OF PARTICIPANTS.........................................197
B. PARTICIPANT INTERVIEW GUIDES.................................201
C. PARTICIPANT INFORMATION FORM.................................205
D. SCREENING MAP FORM...........................................206
xi


LIST OF TABLES
TABLE
1. Comparison of CECAP and GD Interview Groups.............................33
2. Specialized Informants..................................................35
3. Final Screening Map Form Variables......................................36
4. Interview Participants'Occupations.....................................108
5. Clients Per Visit Type Per Year........................................152
xii


LIST OF FIGURES
FIGURE
1. Map of Tanzania.........................................................9
2. The CECAP Room.........................................................18
3. Screening Supplies.....................................................19
4. Parity of CECAP Clients, October 2010-October 2013....................107
5. Number of Clients Screened Per Month in Clinic, May 2010-October 2013.150
6. Total Number of Clients Screened Per Month, May 2010-October 2013.....151
7. Cervical Cancer Posters...............................................158
xiii


LIST OF ABBREVIATIONS
ART Antiretroviral Therapy
CECAP Cervical Cancer Prevention Program
CHW Community Health Worker
CTC Care and Treatment Clinic
GAVI Global Alliance for Vaccines and Immunizations
GD Gynecology Department
HBM Health Belief Model
HPV Human Papillomavirus
mi Ifakara Health Center
LEEP Loop Electrosurgical Excision Procedure
MAISHA Mothers and Infants, Safe, Healthy and Alive
MOHSW Ministry of Health and Social Welfare
NGO Non-Governmental Organization
NIMR National Institute for Medical Research
PIF Patient Information Form
RA Research Assistant
RCH (MCH) Reproductive and Child Health (Maternal and Child Health)
RTI Reproductive Tract Infection
SFUCHAS St. Francis University College of Health and Allied Sciences
SMF Screening Map Forms
STI Sexually Transmitted Infections
USAID United States Agency for International Development
VIA Visual Inspection with Acetic Acid
xiv


NOTE TO THE READER
During the course of my fieldwork, I interviewed women who had come to St. Francis
Hospital, women I refer to as research participants. In this dissertation, when discussing
my encounters with these women or reproducing their words, I refer to them using
pseudonyms in order to protect their confidentiality. Some details, such as references to
specific doctors they have seen or villages they live in, have been left out. Brief
descriptions of these women, along with their pseudonyms, can be found in Appendix A.
I also interviewed specialized informants and came into contact with other hospital
personnel. The informants I interviewed were often less concerned with confidentiality,
though we discussed it during the consent process, and were in several cases eager for
their words to be heard. St. Francis is a relatively small hospital; it would not be too
difficult for a resolute individual to determine who the informants I discuss are.
Nonetheless, I have chosen to refer to all informants and hospital personnel using initials
(not their real initials) here in the dissertation to provide a degree of confidentiality.
xv


CHAPTER I
INTRODUCTION
Cervical cancer, public health experts and propaganda leaflets repeat, is a
'fully preventable disease'. There is, however, a big difference between
preventabilityhypothetical variableand the construction of efficient
prevention programmes. In the twentieth century medical sciences devised
efficient ways to reduce the number of women who suffer from advanced
cervical malignancies. The persistence of the disease in many parts of the
world reflects to an important extent a failure to act upon economic and
social conditions that hamper the diffusion of preventive measures.
(Lowy, 2011, p. 173)
Cervical cancer is a major issue for women in the developing world, despite a
dramatic reduction in incidence and mortality in developed nations over the course of the
20th century. This disparity is largely due to a discrepancy in access to and utilization of
screening programs in low- and middle-income countries. The international community
has increasingly recognized the importance of cervical cancer screening in these regions
and the expansion of resources targeted towards the disease, however, services, when
they are available, are often under-utilized (Agurto, Sandoval, Rosa, & Guardado, 2006;
Tsu, Jeronimo, & Anderson, 2013). As stakeholders have implemented cervical cancer
screening programs in settings with few resources and high incidence of the disease,
interest in understanding how to increase utilization of these programs has likewise
increased.
The cervical cancer burden is especially severe in sub-Saharan Africa, where it is
the most common female cancer in many countries. The incidence rate for the region is
one of the highest of any in the world, at 34.8 per 100,000, with 93,225 new cases in
2012, the most recent year for which reliable statistics are available (Castellsague et al.,
2012; Ferlay et al., 2013). Cervical cancer is the single largest contributor to years of life
1


lost from cancer there (Yang, Bray, Parkin, Sellors, & Zhang, 2004). This picture is
daunting, and made further so by the fact that the disease disproportionately affects
young adult women in this region, who are often integral to their families health and
economic security. Yet the majority of women in this region live in rural areas where
screening and treatment services may be unavailable or difficult to access (Anorlu, 2008;
Castellsague et al., 2012).
This dissertation examines the issues of cervical cancer and cervical cancer
screening in the context of a rural health-care setting in the sub-Saharan African country
of Tanzania. Tanzania has in recent years partnered with the international non-profit
health organization Jhpiego to organize cervical cancer prevention programs at various
sites around the country; this dissertation focuses on a program located at a Referral
Hospital in the town of Ifakara. The initiative was begun with the idea that after a set
number of years under Jhpiegos supervision, administration and funding of these
programs would transfer to the Tanzanian government. As with screening services in
many developing nations, low utilization of the program has been an issue since its
inception and was a source of tension between the program staff and Jhpiego. Despite
attempts to mitigate this low utilization, the program has been unable to raise their
number of clients served in any consistent way.
Some of the issues that influence poor utilization in developing nations have been
well documented and are widespread, particularly critical issues with health
infrastructure, transportation infrastructure, and cost. Solving these remains a large and
important undertaking. However, these issues do not account for the picture entirely.
Only more recently have researchers begun to examine how local knowledge and
2


perceptions of the disease promote or inhibit health-seeking behaviors among women in
populations who may be targeted for screening. Research into how cervical cancer is
culturally situated within broader health issues, how women think about and give
meaning to the disease, and sociocultural factors that might influence the decision to
undergo cervical cancer screening have been largely understudied, particularly in
Tanzania and other sub-Saharan African nations. This research seeks to address this gap
by exploring these issues in relationship to the larger background of cervical cancer
policy in Tanzania.
Research Questions
This project was guided by a primary research question: How do perceptions of
cervical cancer among Tanzanian women influence their uptake of cervical cancer
screening services, and what implications does this have for the adoption and expansion
of cervical cancer screening programs in Tanzania? In order to address this overarching
question, four sub-questions were developed to guide the research.
(RQ1) What is the population currently attending screening programs and who
is missing from this population?
(RQ2) How is cervical cancer understood and given meaning among Tanzanian
women seeking care?
(RQ3) What influence do cultural constructions of cervical cancer have on
womens attitudes towards cervical cancer screening?
(RQ4) How does the screening program operate on a daily basis, and how does
it fit within the overall structure of the hospital?
Theoretical Approach
This dissertation is organized in a somewhat non-traditional manner, in that there
is no one individual chapter devoted to an examination of the relationship between theory
and the subject at hand. Theory is instead incorporated throughout the dissertation
chapters where it is most appropriate and relevant to specific themes. However, here I
3


would take a moment to outline briefly the general approach guiding my research.
In seeking to understand cervical cancer screening in Tanzania, I have been
influenced by previous scholarship in critical medical anthropology that has examined
health with an eye towards larger power relationships (H. A. Baer, Singer, & Johnsen,
1986; Farmer, 1999; Singer, 1989, 1998; Turshen, 1998). Traditional anthropological
scholarship has been criticized for focusing too much on the micro-level, and in the case
of medical anthropology, this has often meant on the individual/patient or the patient-
physician interaction (Dressier, 2001; Singer, 1989). I take a political economy of health
approach, a perspective which recognizes that economic, political, and historical
processes influence how disease spreads and who is affected (Doyal, 1995; Farmer, 1999;
Singer, 1989, 1998). This approach emphasizes multiple levels of influence and
interaction, from the macro to the micro, and is therefore particularly useful in examining
issues tied to globalization and health (H. Baer, Singer, & Susser, 2003).
Yet at the same time, I do not abandon more interpretive approaches to medical
anthropology completely. Discerning how local populations conceptualize and assign
meaning to disease and how individuals understand their own illness experiences are
important. This is not simply an academic exercise; it has significant implications for
health-seeking behavior (Kleinman, Eisenberg, & Good, 2006; Nichter, 2008).
Bourdieus (1980) work, which attempted to tie together more structural approaches with
what he called a cultural constructivist approach, and Dressiers (2001) subsequent
argument for a synthesis of these perspectives were therefore influential. In practice, this
meant that while I was concerned with local perspectives on the disease of cervical
cancer and how these might influence screening utilization, I also sought to understand
4


how structural forces and power relationships both influenced these phenomena and were
influenced by them.
Organization of the Dissertation
In Chapter II, I begin by providing background on how I came to do my research
in Ifakara, and introduce the field setting, including the town and its surrounding
environs. A description and brief history of St. Francis Hospital, the specific facility at
which I made most of my observations is included here. Finally, I introduce the cervical
cancer prevention program, including its setting within the hospital, and the services
provided.
In Chapter III, I outline the methodological approach that guided this research,
providing a summary of the data collection procedures and analysis. Here I also describe
the specialized informants and the participants from the cervical cancer prevention
program and the Gynecology Department who were so generous with their time and
whose interviews were a key part of the qualitative data collected for this project.
In Chapter IV, I focus on the macro-policy environment in which cervical cancer
became a health concern at the national level in Tanzania, corresponding to larger
changes in global health priorities and policy. This shift then resulted in increased interest
and funding from foreign governments, non-governmental organizations (NGOs), and the
Tanzanian government to implement screening programs, as well as an increase in media
attention. I also tie this to the efforts made at the local level in Ifakara to implement
screening and increase awareness.
In Chapter V, I move to an exploration of local knowledge of cervical cancer,
including its causes, potential risk factors, and prognosis, among the St. Francis patients I
5


interviewed in the course of my fieldwork. I explore how the disease is perceived by
interview participants, and how these perceptions relate to familiar but general
biomedical explanations of infection and immunity, in particular those related to HIV,
STIs, and other sexual or reproductive health issues. I argue that these beliefs have
implications both for patient education and outreach, as well as more mid-level concerns,
such as how the cervical cancer screening program is situated within the overall hospital
setting.
In Chapter VI, I summarize findings on how participants understand benefits and
barriers to screening and treatment at the cervical cancer screening program in Ifakara, as
well as for more extensive treatment in Morogoro or Dar es Salaam. I review the barriers
in light of a fairly extensive literature on challenges to screening in developing nations
that nonetheless is thin on regionally-specific data in East Africa. Understanding these
challenges can provide insight into how to improve services, particularly for programs
that have already attempted to address, at least in part, the cost issue.
In Chapter VII, I turn my attention first to the notion of stigma, both as it relates
to cervical cancer and cervical cancer screening, before examining whether or not women
choose to discuss screening behavior or a positive diagnosis with others. Although
privacy is an important cultural value in this regard, it must be weighed against the need
to access the types of resources and support required to follow through on a decision to
seek health care services. Furthermore, some women expressed a strong desire to be a
source of information and support to other women, which influenced their decision to
share their own experiences with this aspect of their health.
In Chapter VIII, I return to more mid- and macro-level concerns by examining
6


some of the challenges facing the cervical cancer prevention program and the
relationships among the different stakeholders I encountered while conducting this
research at St. Francis. Although these stakeholders all ostensibly desire the same goal
to improve womens access to cervical cancer screeningin reality their priorities were
sometimes at odds with one another. These tensions reflect larger debates in the global
health literature, such as the role of NGOs working within health care systems in
developing countries and the prioritization of vertical versus horizontal health care
structures. Although I did not go into this project intending to study these issues, it was
impossible to escape their discussion and influence while in the field.
Finally, I conclude with a brief discussion of some of the overall themes of the
dissertation, as well as limitations and several areas for future research.
7


CHAPTER II
SETTING THE SCENE
I first became interested in conducting my research into cervical cancer at St.
Francis Hospital in Ifakara, Tanzania through mentors at the University of Colorado
Denver. I had originally intended to study cervical cancer screening in northern Tanzania,
in the city of Mwanza, where I had made my first visit to the country in 2010. However,
the hospital there was located in a major urban center and was one of the largest in the
nation. I was interested, but I knew that the urban population and wider screening and
treatment procedures available there were unusual when compared to the health facilities
in most of Tanzania. Additionally, their screening program was already in development
and plans were in place to conduct their own evaluation in-house. As I began to search
for another opportunity, preferably in a more rural location, I was invited to visit St.
Francis Hospital and observe their screening program by one of my mentors colleagues,
who worked at the facility.
In June 2011,1 visited Ifakara for two weeks. During that time, I was able to
speak with faculty and personnel at St. Francis, observe the activities of the screening
program, and become acquainted with Ifakara. St. Francis was an interesting and ideal
choice for this study for several reasons. It served a primarily rural and semi-urban
population, reflecting the circumstances of the majority of Tanzanian women, and its
population was likely to face similar challenges as those in other rural areas. As a
Referral Hospital (though until recently only a District Hospital), it was better equipped
than most dispensaries and small health clinics in rural areas; however, women were still
referred to larger health centers for any positive diagnosis beyond a small cervical lesion
8


treatable with cryotherapy. This, I felt, would give me the opportunity to study not only
the one-visit screen-and-treat approach practiced by the screening program (discussed
further in Chapter IV), but also the referral process for more advanced cases and the
potential challenges women would face during this process. Furthermore, St. Franciss
program is similar to many others in Tanzania, in that it was initiated largely through the
work of an international NGO working within the Tanzanian health care system. I
returned in May 2013 to begin my fieldwork, following two and a half months of
Kiswahili study in the city of Morogoro. In this chapter, I begin by describing the region
in which both Ifakara and Morogoro are located, also called Morogoro. I then focus the
lens on Ifakara and St. Francis Hospital before introducing the cervical cancer prevention
program itself.
Figure 1: Map of Tanzania
9


Morogoro City and Region
The Morogoro region, located in southeastern Tanzania, is divided into seven
districts. Near the northern end, its major urban hub, the city of Morogoro, makes up the
smallest district by area (referred to as the Morogoro Urban District so as to differentiate
it from the adjacent Morogoro Rural District) and is connected by tarmac highway to the
commercial capital of Dar es Salaam, 194 km away (National Bureau of Statistics, 2006).
Because of its proximity to Dar es Salaam and location on the major east-west highway,
Morogoro is a busy thoroughfare for those traveling to interior cities and towns, including
the political capital Dodoma, Arusha, and Iringa, as well as for safari tourists on their
way to the popular Mikumi National Park.
Of the approximately 2.2 million people living in the region, 315,866 call the
municipal district of Morogoro home, though not all live in the town proper (National
Bureau of Statistics, 2013). The overall region has had an average population growth rate
of 2.4% per year, which is close to the national average of 2.7%, and is one of only two
regions in the country outside of Dar es Salaam to have a level of urbanization higher
than that of the nation as a whole (Ministry of Planning, Economy, and Empowerment,
2007; National Bureau of Statistics, 2013). Tanzania has seen a trend of rural-urban and
urban-urban migration, often tied to employment opportunities. Morogoro Region has
experienced both trends, with those from rural areas moving to the city of Morogoro and
those in the city migrating to other urban areas outside of the region (Ministry of
Planning, Economy, and Empowerment, 2007).
Situated in the southern highlands at the feet of the Uluguru Mountains, Morogoro
is home to Sokoine University of Agriculture (among other educational institutions) and
10


is a major center for agricultural production, including a wide range of both food and
cash crops (Ministry of Planning, Economy, and Empowerment, 2007; National Institute
for Medical Research, 2012). During my first months in Tanzania while studying
Kiswahili in Morogoro and traveling back and forth to Dar es Salaam frequently for
meetings, the vast sisal fields stretching out in every direction on approach from the east
were the first signs I was nearing home. In addition to agriculture, many of the religious
and non-profit organizations, foreign and domestic, operating in the region are based in
Morogoro.
For the purposes of this study, the primary importance of the city of Morogoro is as
the location of a referral health facility for cervical cancer screening clients requiring a
specific procedure (known as Loop Electrosurgical Excision Procedure, or LEEP) that
cannot be obtained locally at the field site but whose cases are not advanced enough to
necessitate a trip to the major medical facilities in Dar es Salaam. This will be discussed
in more detail in a later chapter. The remaining six districts of the Morogoro region are
larger and more rural in nature (National Bureau of Statistics, 2013).
Ifakara and the Surrounding Area
On the good, though congested, tarmac highway, the distance between Dar es
Salaam and Morogoro can be covered in approximately three hours by bus. If one
continues on through Morogoro, the highway eventually winds its way west directly
through Mikumi National Park, where baboons linger by the roadside hoping for food
scraps thrown by passengers, until it reaches the town of Mikumi on the other side. After
a quick stop at the chips mayai stand (a popular street food best described as a French fry
omelet) in Mikumi, the bus turns southwest and continues for 25 km before the tarmac
11


gives way to a bumpy narrow dirt road that often floods during the rainy season.
Eventually, approximately ten hours and 400 km after departure, one arrives in Ifakara,
the town in which I lived and conducted my fieldwork.
Ifakara is a semi-urban town of 55,956 people near the Udzungwa Mountain range
in Kilombero District, the second largest district in Morogoro region at 14,918 km2. The
town is set down near the Kilombero River in a valley of the same name (National
Bureau of Statistics, 2006, 2013). As the largest urban area in the Kilombero District,
Ifakara is home to the district government, several NGOs, and important health and
educational institutions, including St. Francis Hospital (Ifakara Health Institute, 2014;
National Bureau of Statistics, 2006). A long narrow tract of land, the district wraps in part
around the larger but much less densely populated Ulanga District, and despite primarily
serving the residents of Kilombero, St. Francis is the closest hospital for many of those
living across the river in Ulanga as well (National Bureau of Statistics, 2006). With the
Udzungwa Mountains to the north and lush vegetation fed by the river, the valley is a
scenic place, though at this point, with the exception of Udzungwa Mountains National
Park, there is relatively little to tempt tourists here off the beaten path.
The Local Economy
Agricultural work is the primary economic activity in the fertile Kilombero
Valley, as it is in both Morogoro Region and Tanzania as a whole, with rice cultivation
particularly important (Hetzel et al., 2008; National Bureau of Statistics, 2006, 2011).
The valley has a large floodplain and when the rainy season comes, the river overflows
its banks. During this time, many individuals or households relocate to the shamba (farm)
for rice paddy cultivation (Hetzel et al., 2008; Kato, 2007). Others, like the family with
12


whom I lived, had access to a shamba close enough to walk to from our house in town
and return again at night.
The people of Kilombero also engage in fishing and the raising of livestock,
including chickens and cattle. However, an increase in the population of the area in recent
decades has precipitated a corresponding increase in land use for both farming and
pastoral purposes, leading to concerns regarding the potential environmental impacts
such as water scarcity and deforestation (Kangalawe & Liwenga, 2005). After farming,
the most common occupations include craftsmen, venders, and small business owners,
particularly in more urban areas like Ifakara. The presence of the government, non-profit,
health, and educational sectors provide a significant number of professional, technical,
and administrative positions as well.
Population
The lingua franca in Tanzania is Kiswahili, part of the Bantu language group,
although various tribal languages are spoken as well. At my field site, Kiswahili was most
commonly spoken in day to day life, though on occasion an individual would tell me they
couldnt remember the proper Kiswahili word and would substitute a term from another
language they spoke. The dominant tribal or ethnic groups in the Morogoro region are
similarly of Bantu origin. In Kilombero, these primarily include the Wambunga,
Wandamba, Wabena, and Wahehe, though migration has brought other groups into the
area (National Bureau of Statistics, 2006). In Kiswahili, wa is the plural prefix for a
class of nouns denoting humans or types of people (McGrath & Marten, 2003).
Kilombero District is overall primarily rural, with only 28.2% of its population
living in urban areas. This is slightly higher than the 23% found across Tanzania as a
13


whole (National Bureau of Statistics, 2006). The average household size of 4.3 is
representative of both the Kilombero District, and the region of Morogoro, though
slightly lower than the national average of 4.8 (National Bureau of Statistics, 2013). The
sex ratio of men to women in Ifakara is 91, slightly lower than the district (99), region
(97), and the nation (95) as a whole (National Bureau of Statistics, 2013). Approximately
86% of the population of Kilombero is below the age of 45, and 40% is under 15 years of
age, compared to 45% nationally (National Bureau of Statistics, 2006; Population
Reference Bureau, 2012). According to these descriptors, Kilombero does not differ
greatly from the nation as a whole.
Ifakara as a Field Site
A Center of Health Research
Ifakara has historically been a center for health research in the region, dating back
to the founding of the Swiss Tropical and Public Health Institute in 1956. The Institute
has undergone multiple evolutions and name changes since then, before settling on
Ifakara Health Institute (IHI) in 2008. Although originally founded by the Swiss, the
Institute has had Tanzanian leadership since 1993, and became formally independent
from the Swiss in 1996. It is an important training ground for Tanzanian scholars and
scientists. Originally focused on malaria research, IHFs research portfolio has expanded
to include other public health issues important to the region and the nation (Ifakara
Health Institute, 2014).
IHI shares a campus with St. Francis Hospital, as well as St. Francis University
College of Health and Allied Sciences (SFUCHAS), and the Tanzanian Training Center
for International Health. St. Francis is a teaching hospital and the medical students are a
14


frequent sight in its corridors. This arrangement has allowed the various institutions to
share resources and form research partnerships. In addition to students and researchers
from Tanzania and other parts of Africa, IHI and St. Francis draw a significant number of
foreign medical personnel who come to work for limited periods of time at the hospital,
as well as foreign researchers.
St. Francis Hospital
There has been a dispensary in Ifakara since 1927; however the history of St.
Francis Hospital itself extends back to 1955 when an Austrian surgeon, Dr. Schoepf,
working at the small Swiss-run mission hospital in Ifakara (which opened in 1931)
initiated its construction. It was completed in 1960 and the surgeon eventually returned to
Austria in 1969, though the hospitals administration remained in the hands of European
health professionals until its first Tanzanian director in 1993. While it eventually waned
in the years following Tanzanias independence from colonial rule in 1961 due to lack of
resources, Dr. Schoepf undertook efforts in 1995 to organize aid, supplies, and training
for its personnel (St. Francis Hospital, 2014).
Since then, the hospital has rebounded. Although it began as a Catholic
institution, in the past it also served as a Designated District Hospital under agreement
with the Tanzanian government. That changed in 2010 when it became one of nine
Referral Hospitals in the country. The Diocese of Mahenge and the Tanzanian
government share both financing and governance of the hospital. Between Outpatient and
Inpatient Departments, St. Francis treats approximately 117,000 patients per year. While
it currently has 371 beds, it is undergoing expansion because it has outgrown its current
buildings. Construction was ongoing while I was conducting my research there.
15


Cervical Cancer Prevention Program
Location
Within the hospital, most of my time was spent with the cervical cancer
prevention program, or CECAP, which had been in existence since May 2010. This
program was currently housed in the Maternal and Child Health (MCH) building. Though
the name had officially been changed to Reproductive and Child Health (RCH), in order
to be more inclusive of men, the sign outside still read MCH and most people, including
myself, referred to it using the old acronym. A small, single-story building sitting on a
cement platform in a low-lying area across from the main hospital building, it was
reached by a long raised cement walkway that reached all the way to the street. This
walkway lay directly across the road from the gate to the HIV clinic, also called CTC (for
Care and Treatment Clinic).
The interior of the building contained a small courtyard, where on busy days a
long line of women in brightly colored kangas (cloth wraps) holding babies would form
as they each waited patiently to deposit their child into the fabric harness attached to a
scale hanging from the ceiling. A nurse would direct this process, and then note the
weight on a small graph on the back of the childs record card, before giving it back to
the mother. A small patio sat just off this courtyard with two doors facing the back of the
building. One opened into a room where daily classes for expectant mothers were held.
The second, in the very back corner of the building, was CECAP.
I explain this here to illustrate where the clinic was located within the larger
context of the hospital campus. While the program was originally conducted out of the
Gynecology Department in the main hospital building when I visited in 2011, it had been
16


moved here so that it could have its own designated space. However, as a result, it was
rather tucked away and not something likely to be seen by women coming to the hospital
for other purposes. There was no passing traffic other than the class attendees next door,
and very little likelihood of someone finding it unless they were directly looking for it.
Screening Services
The CECAP schedule followed a general routine. Screening took place two days of
the week; in 2013 it was Mondays and Wednesdays. The program opened in the morning,
generally between 9-10am, though there was no set time. The two CECAP staff members
had many other duties in the hospitalone was a Medical Officer and the other a nurse
and so its opening time depended somewhat on their activities elsewhere. While they
typically finished with screening by the early afternoon, at which time they would return
to the main hospital, occasionally there were enough women to keep them busy longer.
Typically one of the two staff members, MN or AL, not both, would be present at the
clinic, though again, a large number of women or specific CECAP tasks would
sometimes require that they both be there.
While CECAP services typically began around 9am, the main hospital gates and
RCH both opened earlier. By the time I arrived each morning then, at least one or two
women were typically already there waiting. Often more were gathered on the bench or
standing on the patio. Women attending the classes for pregnant and new mothers held
next door also gathered here, so there was nearly always a small crowd to greet as I
walked up.
The screening room itself was small and its furnishings simple (Figure 2). A large
wooden cupboard containing all the necessary supplies, forms, and records lined the wall
17


to the right as one entered. An exam table with stirrups took up the center of the room,
while a second that served as a makeshift storage surface stood up against another wall. A
small wooden desk sat under the window, with cryotherapy tanks bearing stickers from
the United States Agency for International Development (USAID) on either side in the
corners. The clinic was a corner room, and large windows in two walls served as the
primary source of light, with the overhead florescent bulb rarely turned on. Fluttery
drapes were twisted up and tucked behind the window bars on arrival to let in the light
and a breeze, with the consequence that anyone outside could see in, though given that it
faced the back of a building on the edge of campus this was not a common issue.
Figure 2: The CECAP Room
Preparing the room for screening was a quick process that involved the staff
member on duty retrieving screening forms and records from the cabinet, disinfecting
surfaces and the exam table, and setting out all of the supplies needed for screening
(Figure 3). These supplies were few and simplecotton swabs, a flashlight, small metal
containers of acetic acid and water, and speculums. Once everything was prepared, the
staff member would invite the waiting women in for an education session.
18


Figure 3: Screening Supplies
There were several small wooden footstools scattered in the room, and clients
would gather around, sitting on these or standing, while the staff member delivered a
short lecture. During the talk, AL and MN not only explained the basics of cervical
cancer and cervical cancer screening, but also demonstrated quickly and briefly on their
own bodies how one would do a breast self-exam. Both are short but imposing women
who project an air of knowledge, but also good humor. MN in particular is always
smiling and occasionally breaks out into song. They regularly checked in with women
during the lecture after finishing each main point with the question, Sawa? (Okay?)
basically shorthand for Do you understand? This was less a true question and more a
call-and-response, to which the groups inevitable answer was a chorus of Sawa!
Questions were in fact usually saved for the end of the lecture.
Once MN or AL had answered all of the womens questions, the group would file
back outside to wait, while the first woman to be screened remained, standing or seated
on a stool while a staff member filled out her paperwork. If she was HIV positive, she
would produce her blue CTC card with her history of testing and treatment to show to the
staff member. If she was unsure of her HIV status, she was referred for testing. In some
19


cases, though quite rarely, a woman had been screened for cervical cancer before. If so,
she would have a passport in her possession, a small folded cardboard form that
recorded the dates and results of previous VIA screenings (VIA will be discussed in more
detail in Chapter IV). She gave this to the staff member so that it could be updated after
the present screening.
Finally, the staff member would rise and indicate that the client should remove her
clothes from the waist down and climb up onto the exam table while the staff member
washed her hands and put on gloves. During my observations, it became apparent to me
without them saying so that many women had never had a gynecological exam using
stirrups before, because it was not unusual for the client to climb onto the table and then
lie down backwards, with their head between the projecting stirrups and their feet where
their heads should be. Or, in other cases, to put their heels into the stirrups rather than
resting their ankles on the supports and to lie with legs stick straight and their rear ends
resting in the center of the exam table, rather than scooted to the edge of it. If the woman
had a sense of humor about the process, this awkwardness sometimes actually served to
break the ice a little bit because what would follow would be an ungainly scooting down
the table while MN or AL waved them on and the woman laughed or giggled.
The screening process itself only took about five to ten minutes (the details of
which are explained in Chapter IV) and I observed many screenings in the course of my
research. The sight of foreign doctors was not unusual at St. Francis and most clients did
not appear surprised to see me in the room. Some clients found it amusingOh!
Mzungu! (Oh! Foreigner!), they would laugh upon seeing memost largely ignored me
after greetings were exchanged, and two particular women were determined to introduce
20


me to their brothers as a potential marriage candidate. Throughout the field period
though, I tried to be very cognizant of the potential discomfort some might feel in my
presence and on more than one occasion stepped out of the room if the client indicated
she was uncomfortable, or if I or the staff members noticed behavior that gave us pause.
In these matters I tried to take my cues from AL and MN.
21


CHAPTER III
METHODS & ANALYSIS
This study took place over the course of six months in 2013, during which I lived
and conducted research primarily at St. Francis Hospital in Ifakara, Tanzania. Although I
was in Tanzania for a total of ten months and intended to conduct my research over a
nine month period following a month spent studying Kiswahili at a language school in
Morogoro, difficulties in obtaining the proper research permits delayed the start of my
project. As a result, I spent an additional month in Morogoro (its proximity to the permit
offices in Dar es Salaam made this a practical choice) and six weeks living in Ifakara
prior to the start of data collection. I was able to use this time to become acquainted with
my new home and the cervical cancer prevention program, and to engage with
community members and hospital staff.
The focus of my research was on CECAP, the cervical cancer prevention program
at St. Francis Hospital, its clients, and another subset of women who came to the hospital
seeking care in the Gynecology Department (GD) but who did not utilize the prevention
program. Through methods that will be discussed in more detail below, such as semi-
structured interviews, participant observation, patient record reviews, and the collection
of grey literature, I collected both qualitative and quantitative data to address the research
questions outlined at the beginning of this dissertation and reiterated below:
(RQ1) What is the population currently attending screening programs and who is
missing from this population?
(RQ2) How is cervical cancer understood and given meaning among Tanzanian
women seeking care?
(RQ3) What influence do cultural constructions of cervical cancer have on
womens attitudes towards cervical cancer screening?
(RQ4) How does the screening program operate on a daily basis, and how does it
fit within the overall structure of the hospital?
22


I established the first three questions before I left for the field, but upon
commencing my research I soon realized that any understanding of who came to CECAP
for screening, and how at least some of the participants perceived cervical cancer and
cervical cancer screening, needed to consider how CECAP itself was situated both within
the hospital itself, and within the larger intersection of the Tanzanian health care system
and the international NGOs that work in the country. Thus I developed the fourth
research question.
Methodological Approach
This study took an ethnographic approach, a method which emphasizes holism; in
this case, the goal was to not only illuminate processes present in the site of interest, but
also to provide the context in which these processes were grounded (Marcus & Fischer,
1986; Neuman, 2003). Participant observation has long been a hallmark of ethnography
as practiced by cultural anthropologists, to the degree in fact that to some anthropologists
participant observation is the ethnographic field method (DeWalt & DeWalt, 2010). The
classic version of this method was complete immersion in the society under study,
usually small-scale, for long periods of time. To not only investigate but also to
participate in the daily lives of those one was studying with the goal, as Marcus and
Fischer write, not to make universally valid statements, but to represent a particular way
of life as fully as possible (1986, p. 33). Such an undertaking is increasingly difficult for
researchers today, who instead tend to spend shorter periods of time in the field and focus
on more narrow topics (LeCompte & Schensul, 2010). In the case of this study, classic
participant observation was also somewhat hampered by a lack of fluency on my part in
Kiswahili and the nature of the hospital setting.
23


As an anthropologist however, participant observation as a method of data
collection is part of my training and my nature. Without a medical degree or medical
training, my opportunities to fully immerse myself at the hospital were obviously
restricted, but I frequently found myself assisting in what ways I could, such as helping
with office tasks, holding the flashlight during screenings, and folding cotton for swabs
used during procedures. These types of activities allowed me to fade into the background
to a certain degree and observe without drawing attention to myself. In addition to my
work at the hospital, I lived with a Tanzanian family near the hospital, helped with their
rice harvest, shopped at the nearby market, rode on the back of pikipiki (motorcycle
taxis), and in general, lived my life in the community. During these daily encounters,
both at the hospital and outside, I frequently had conversations that made their way into
my field notes. Therefore, the way that I used participant observation in this study is
more in line with its treatment by DeWalt and DeWalt (2010) as one tool in the
qualitative toolbox, rather than as the sole ethnographic method.
Instead of classic participant observation, I utilized a methodological approach
called polymorphous engagement:
Polymorphous engagement means interacting with informants across a
number of dispersed sites, not just in local communities, and sometimes in
virtual form; and it means collecting data eclectically from a disparate
array of sources in many different ways. Polymorphous engagement
preserves the pragmatic amateurism that has characterized anthropological
research, but displaces it away from a fetishistic obsession with participant
observation.. .as the communities we study disperse and link up across the
globe, as all sorts of elites now are, and as we try to match the emergence
of global cultures with a global anthropology, we must develop new
research strategies. [Gusterson, 1997, p. 116]
This approach is appropriate given that while this project is focused on
understanding how women view cervical cancer and how these views in turn influence
24


their utilization of screening services, this process is situated within larger political,
economic, and cultural systems. Polymorphous engagement is particularly useful for
looking at situations where the macro- and micro-levels interact, and just as importantly,
situations where it can be difficult to separate the two into discrete entities (Markowitz,
2001). While the bulk of fieldwork took place in a local setting, the CECAP clinic, the
environment the clinic is situated in, from its immediate location on the St. Francis
Hospital complex adjacent to the HIV clinic, to its broader place within the Tanzanian
healthcare system and as a recipient of funding and oversight from an international NGO,
all influence how screening is utilized and, in some cases, perceived. Therefore, in
addition to the primary site, the CECAP clinic, I also visited the HIV clinic, other areas
of the hospital campus including the Gynecology Department, Ocean Road Cancer
Institute, and when opportunities arose, smaller dispensaries.
Furthermore, polymorphous engagement arose in part out of a concern that
participant observation did not translate well to settings where the anthropologist wanted
to study up (Gusterson, 1997, p. 116). Studying office settings, corporations, hospitals,
and other organizations or elites, may involve issues that make classic participant
observation difficult or impossible. Gusterson notes of participant observation that this
technique may not be readily portable to elite contexts in the U.S. where ethnographic
access is by permission of people with careers at stake, where loitering strangers with
notebooks are rarely welcome, and where potential informants are too busy to chat
(1997, p. 116).
Though not in the U.S., this was a concern for me going into my study as well and
at times I became acutely aware of how these concerns can play out in the field. My visit
25


to the Ocean Road campus took the form of a structured and guided tour, for example,
where the doctors I met briefly were primarily interested in showing me the new
technologies available for treatment at the facility. While I was typically allowed free
reign to observe and assist with menial office tasks in the CECAP clinic, there were also
days where the staff were simply too busy to answer questions or to have me underfoot.
Healthcare workers in the larger RCH building the clinic was housed in were perfectly
happy to assist me with an interview room and chairs during their quieter hours, but were
not shy about opening the door and heavily hinting we needed to move along when they
needed the space; several also expressed concerns to the Research Assistants (RAs) when
I was not present that they were reluctant to be interviewed about their jobs, despite the
fact that we did not in fact plan to do this.
However, in other cases I certainly experienced what Markowitz (2001) observed
in her own fieldwork, that by virtue of my being an American researcher I was granted a
degree of elite status by some with whom I interacted. For example, I was very often
accorded the respectful greeting of Shikamoo in the hospital setting by patients despite
my being younger than many, and was certainly given latitude when it came to gendered
behavioral expectations. Almost no one questioned my presence in the hallways of St.
Francis, in the clinical waiting areas, in meetings between hospital staff and medical
students, or in exam rooms. It was typically assumed that if I was there, I must have
permission to be, although I certainly also tried to always be sensitive and respectful to
patient privacy and to the healthcare workers going about their tasks.
Polymorphous engagement also stresses the use of multiple and diverse research
techniques. As noted before, participant observation was one such tool and I certainly
26


used it when and where it was possible. In other cases my role was limited to just that of
observer, with no participation possible. In the following section, I outline the different
techniques and tools used for data collection in this research project.
Data Collection
Participant Observation/Observation
During the six-month field period, I typically spent 3-5 days per week at the St.
Francis Hospital facilities. While Mondays and Wednesdays were devoted to CECAP, on
alternate days I would often sit and observe other areas of the hospital and speak
informally with those I encountered, including SFH and SFUCHAS personnel, and
foreign medical students and doctors who were working there. In addition, at various
times I traveled outside of the local area to visit smaller dispensaries and Ocean Road
Cancer Institute. I took field notes on the observations and experiences that occurred
during these activities, as well as on daily life in Ifakara. This generally took the form of
a pen and small notebook that fit in my bag, which I could pull out to jot down notes as
the day went by. In some situations, where this proved disruptive or distracting, I would
record my observations as soon as I was able following the conclusion of the activity.
These field note jottings were then fleshed out in the evening or early the next day to
capture as many remembered details as possible. While all quotes from formal interview
participants are taken directly from digital voice recordings, I did not carry a tape
recorder with me otherwise. After a few attempts, I found that it was awkward to use and
that some of the Tanzanians I interacted with expressed discomfort with it. In many cases
I was able to write down directly what an individual said to me, but when I could not
capture their exact words I paraphrased them in my notes. I then followed that same
27


convention in this dissertation, using quotation marks only to express those instances
when I am using an individuals exact words.
Participant Interviews
I interviewed two groups of women at St. Francis Hospital for this study, for a
total of 49 participants (Appendix A). The first were CECAP clients (n=24), women who
either came to St. Francis specifically for cervical cancer screening or were referred for
screening by a health care provider in a different department at St. Francis. A second
group of women were recruited from the GD (n=25). These women were seeking care for
reproductive or gynecological reasons outside of cervical cancer screening. By using a
hospital-based population in this study, I planned to focus on women who were able to
overcome some of the more well-known structural and economic barriers to screening,
such as transportation to an appropriate facility and cost of transportation and care
(Anorlu, 2008; Bingham et al., 2003), to more closely examine ideas around perception
and knowledge of the disease, and how these interact with the use of screening services.
This does not mean however that these barriers did not exist for the women interviewed;
indeed, they were frequent topics of conversation. Furthermore, both groups of women
were likely to be sexually active and therefore exposed to some of the same risk factors
for cervical cancer.
Each participant took part in one semi-structured interview, during which they
were asked questions about their general health, their awareness of cervical cancer and
knowledge about the disease, how cervical cancer and screening were discussed in their
communities (if at all), and whether or not they had prior experience with screening. The
CECAP clients were also asked questions about their decision to seek out these services,
28


while those in the GD were instead asked to discuss if they were aware of the program
and if so how, if they would consider being screened, and what might inform that
decision (for Participant Interview Guides, please see Appendix B).
Given the aims of this project, purposive sampling was used to identify interview
participants because it allows the researcher to identify cases that will provide a deeper
understanding of types (Bernard, 2006; Neuman, 2003, p. 213). In this case, there were
two larger types as outlined in the previous paragraph, however efforts were also made
to balance each group across a range of factors that have been shown to influence risk for
cervical cancer, specifically age, parity, and HIV status. Participants were recruited on
the day they sought services at SFH and without exception participated in interviews that
same day in order to eliminate the financial and time burden that would likely ensue if
they had to return on a different day. Recruitment occurred through several mediums.
Some participants were approached by the RAs while they were waiting to see a health
care provider, given a brief introduction to the project, and asked if they would be
interested in hearing more about the study. In a few cases, nurses or other medical
personnel referred eligible participants to the RAs. Finally, the CECAP staff conducted
informational meetings about cervical cancer screening for women who came to the
clinic on Monday and Wednesday mornings and some of the women were informed of
the study there as well.
All of the women who expressed an interest in hearing more were invited to a
private room where they were given a description of the study and asked a few questions
to determine their eligibility. In order to be eligible, women had to be over the age of 18,
speak Kiswahili, and currently be seeking care at either CECAP or the GD. If they
29


confirmed their desire to participate and were eligible to do so, they were then taken
through the informed consent process. The nursing staff of SFH exhibited a great deal of
patience and graciousness with our need for private rooms a resource in short supply in
the crowded hospital and so we tried to be conscious of both the time asked of our
participants as well as the fact that we were taking up valuable space when conducting
our interviews.
Interviews were conducted in Kiswahili by two female Tanzanian RAs, Paula and
Rachel, both of whom were fourth-year medical students at the local university,
SFUCHAS. Rachel and Paula were highly proficient in English, had some prior
experience with the research techniques used, and took part in a training period that
included discussions of human subjects concerns and interview techniques. I sat in on the
majority of interviews in order to help clarify questions or issues that arose and ensure
study protocol was followed. Furthermore, I was able to observe the appearance and body
language of the women as they were interviewed, and the interactions between the RAs
and participants.
These types of observations can be useful for many reasons, for example, clothing
may provide markers as to a persons social status, profession, or religion, and demeanor
may give clues as to a persons emotional state during the interview. Given that this was
a health setting, some of the women were experiencing symptoms or feeling poorly. If I
had not been in the room, I would not have been able to observe them pointing to various
parts of their body or acting out certain symptoms. Sometimes they themselves did not
have the medical vocabulary needed to talk about specific health problems, so these
observations were particularly important in discerning what type of ailment was
30


bothering them. Furthermore, when reviewing the interview notes later in conjunction
with the interview transcripts, the RAs and I were better able to understand why some
interviews went the way that they did. As an example, we spoke with one woman who
provided only very short quiet answers to each question. In some cases, we might have
attributed this to nervousness or perhaps ignorance about the matter at hand, but in this
case, both the RA conducting the interview and I noticed that she kept placing her hand
on her stomach throughout the interview. After she had done it several times, the RA and
I made eye contact, and she asked the client about her behavior. The client revealed that
she had a lot of stomach pain and just wanted the interview to go by quickly. Of course
we obliged, but also took this into consideration when examining her interview data later.
During our training, we discussed the possibility that a participant might not feel
comfortable with me in the room and came up with a non-verbal code to use in the event
that the RA conducting the interview felt it would be best for me to leave. It was not
often necessary, but there were a very few cases in which a participant either directly
expressed discomfort with my being there, or the RA felt that the participant appeared
distracted during the interview by my presence, in which case they gave the signal and I
left to stand outside the room where I could be reached if there were questions. All
interviews were audio recorded, and were, on average, 25 minutes in length (though there
was a wide range, from 10-52 minutes individually).
Sample
We conducted forty-nine interviews in total: 24 women from CECAP and 25
women from the GD. The average age of women in the CECAP group was 42.5 years
(range 25-65), while the GD patient group was younger on average at 34.5 years (24-45).
While attempts were made to recruit women across a wide age range for each sample,
31


this was complicated by the fact that the GD clinic attracted a younger population due to
its focus on fertility issues and pregnancy. In addition, while women as young as 25 were
interviewed for each group, given that older age is a risk factor for cervical cancer, it was
logical that CECAP would attract a clientele that skewed older, while reproductive
concerns would alternatively skew the GD patient pool towards women still in their
child-bearing years, particularly given fertility rates in Tanzania (ICF International, 2004-
2012).
This explanation likely also accounts for another difference seen between the two
groups: Women interviewed from CECAP exhibited higher parity than their fellows in
the GD group, though again there was variety within each group. None of the CECAP
women were nulliparous; on average they had 4.2 children. While the majority of
CECAP clients had fewer than five children, four had between 6-10 children compared
with only one in the GD group. In contrast, the GD group had seven nulliparous women
and an average parity of 2.
32


Table 1: Comparison of CECAP and GD Interview Groups
CECAP GD
(n=24) (n=25)
Mean Age (Range) 42.5 (25-65) 34.5 (24-45)
Mean Age at Sexual Initiation (Range) 17.6(14-23) 19.7 (14-25)
Mean Parity (Range) 4.2(1-10) 2 (0-7)
HIV Status:
Positive 7 (29.2%) 2 (8.0%)
Negative 12 (50.0%) 16 (64.0%)
Unknown 5 (20.8%) 7 (28.0%)
While at least two HIV positive, two HIV negative, and two unknown status
women were interviewed for each group, because during many weeks of the field period
the hospital was out of the necessary testing supplies, it was in fact impossible to know
the HIV status of every woman interviewed. While CECAP and GD patients at the
hospital would typically be asked for their status and referred to the HIV clinic for testing
if it was unknown as a matter of course, this was not possible during periods of supply
shortage. The majority of women we interviewed self-reported their HIV status, though
patients who were positive often visibly carried their CTC cards with them (which
tracked visits for HIV care and treatment).
Patient Information Forms
In order to capture basic demographic information about each interview
participant, the RAs used a Patient Information Form (PIF) (Appendix C) during the
33


interviews. While initially this form was to be completed at the beginning of each
interview, we found that it was preferable to move the process to the end both because
the participants would often offer some of the information asked for in the course of the
interview, and because a few of the questions asked on the form were somewhat sensitive
(e.g., Age of Initiation of Sexual Activity and Number of Sexual Partners). Participants
generally appeared more comfortable addressing these sensitive questions after they had
been speaking for a while. The questions on the form were usually asked in a
conversational manner so as not to interrupt the flow of the interview, but by recording
the data on a specific sheet it ensured that the same data was collected for each
participant.
Key Informants and Specialized Informants
H. Russell Bernard distinguishes between two types of informants in his writing
on anthropological methods: key informants and specialized informants. The former,
according to him, are people who know a lot about their culture and are, for reasons of
their own, willing to share all their knowledge with you, while specialized informant are
those who have particular competence in some cultural domain (2006, p. 196). While I
spoke with many people in the course of my day-to-day interactions, there were those
who became important sources of information and guidance when it came to navigating
both daily life in Tanzania and my work life at the hospital; these I considered to be key
informants. My encounters with them were primarily informal interactions and
discussions that arose either in response to a specific event or experience, or more
naturally out of shared social interactions. In contrast, I conducted formal sit-down
interviews with five specific specialized informants (Table 2), chosen for their unique or
34


in-depth knowledge regarding issues related to my research. Three of these were
employees of St. Francis Hospital, one worked for Jhpiego, and one was a foreign doctor
working for an extended period of time at St. Francis on HIV issues.
Table 2: Specialized Informants
Specialized Informant Affiliation
Physician Specializing in Obstetrics/Gynecology St. Francis Hospital
Assistant Medical Officer in Obstetrics/Gynecology St. Francis Hospital
Inpatient Nurse St. Francis Hospital
Visiting (Foreign) Physician Specializing in HIV St. Francis Hospital
Physician and Public Health Specialist Jhpiego
Screening Map Forms
This project was primarily qualitative in nature, however, I also collected data
from CECAPs Screening Map Forms (SMFs) to be analyzed statistically in order to
address the first research question: What is the population currently attending the
screening program and who is missing from this population? Furthermore, I hoped it
would allow me to summarize data on the screening, treatment, and referral practices of
the screening program, though as I will discuss in a later section, this was complicated by
the sometimes incomplete nature of these forms. In addition, CECAP wanted these forms
to be digitized for its own use.
The SMF is filled out for each client when they come in for screening. In addition
to important client information, the form asks for information on HIV status and ART
use, and records details of the visit, screening results, and any follow up activities or care
required. At the time of this study, SMFs existed only in paper form, with no digital
35


record. This made it difficult to examine the patient data as a whole. During my time with
CECAP, I created an Excel spreadsheet that contained the SMF data of every client
CECAP had seen since the program began in 2010. In all, data from 2440 SMFs were
collected, spanning the period from May 2010-October 2013. This database was then
imported into SPSS for statistical analysis and a copy of the Excel file was also then
given to the staff of the screening program for use in their own program reports and
potential future funding applications. The final variables used for this project from the
SMFs can be found in Table 3 below; please see Appendix D for the original document.
Table 3: Final Screening Map Form Variables
SMF Section Variables Collected
Client Information VIA Screening Date Age Parity District of Residence
HIV Status HIV Status ART Status Source of HIV Status PITC Offered? PITC Accepted? HIV Results from PITC
Visit Type Visit Type
VIA VIA Conducted Today? VIA Results (If VIA conducted today)
Treatment Cryotherapy performed today Cryotherapy postponed/ Reason
Referral Referred? Referred to Reason for Referral
(ART= Antiretroviral Therapy, HIV=Human Immunodeficiency Vims,
PITC= Provider-Initiated HIV Testing and Counseling, VIA= Visual
Inspection with Acetic Acid)
36


Supplementary Information
Ethnographic research can benefit greatly from examining grey literature that may
only be accessible while at the field site or at the very least difficult to obtain, such as
government documents or agency reports (Schensul & LeCompte, 2012). These
documents can provide background or historical information, as well as context and
descriptive data. During the field period I spent time examining various archival
documents associated with the screening program, including training materials, meeting
notes, and reports. These documents were all in English and were kept in storage by
CECAP. Initially I undertook this task simply to better acquaint myself with the history
of the program and to provide a richer description of the prevention program, but after
discussions with specialized informants, and reviewing the early participant transcripts, I
developed the fourth research question (How does the screening program operate on a
daily basis, and how does it fit within the overall structure of the hospital?) and began to
collect this information in a more deliberate manner. I also conducted a thorough
examination of archival literature and secondary data available at the SFUCHAS library
in order to gain a better contextual understanding of cervical cancer screening and
treatment both in Ifakara and Tanzania as a whole. This literature included research
reports of previous related projects conducted at SFH, government reports and
publications, as well as reports published by various NGOs, community groups, and
global health organizations.
37


Analysis
Qualitative Data
I took notes on potentially relevant archival documents and secondary qualitative
data, and in some cases a digital camera was used as a makeshift photocopier to record
documents for later reference; these notes were typed up and stored with the images on
an external hard drive. Throughout the field period, I fleshed out and typed up field notes;
these are useful as a daily record, but also provide a place for reflexivity during the
fieldwork process and serve as a first round in analysis as the researcher digests what
they have observed and experienced (DeWalt & DeWalt, 2010). In four out of five cases,
the specialized informants were quite proficient in English and so I conducted and
transcribed the interviews myself. In the fifth, an RA conducted the interview in
Kiswahili because the informant was more comfortable in that language; the transcription
and translation was done by an RA as well. I then uploaded all of these documents into
Atlas.ti, a qualitative data analysis program (in the case of archival research, the notes
and any available PDFs were uploaded to Atlas.ti, rather than the documents themselves)
for organization and coding when relevant.
After each participant interview, the recording was assigned to an RA for
translation and transcription. Ethnography is an inductive and iterative approach
(LeCompte & Schensul, 2010) and so this process began immediately after the first day
of interviews so that early transcripts could inform future interviews. According to the
original study design, the interviews were to be spread out over eight months, however
the period allotted for interviews had to be greatly condensed. This was in part due to the
delay in obtaining study permits, but primarily because of unexpected changes in the
38


RAs school schedules requiring them to leave the field site prematurely. Because of this,
there was not adequate time to review each transcript prior to conducting the next
interview. Instead we worked with small batches of transcripts at a time. I read each
transcript alone first, making notes where I had questions or suggestions for future
interviews and then the RAs and I met for a read-through and discussion of any point I
had marked. While we would discuss their impressions of each interview immediately
after its completion, these read-throughs also gave the RAs an opportunity to share any
additional thoughts that were triggered by rereading the transcript itself. Following these
meetings, the final transcripts with any additional notes resulting from said meetings
were uploaded into Atlas.ti.
Certain themes related to screening perceptions and decision-making have been
identified by previous researchers, however, I was cognizant of the fact that relatively
few of these studies were conducted in this geographic and cultural setting. I therefore
first coded the interview transcripts inductively, using what Glaser and Strauss (1977)
call constant comparison, to generate a preliminary code structure. Inductive codes, or
codes derived from the data themselves, are useful in that they come without the
preconceived ideas and definitions that can weigh down deductive concepts borrowed
from the literature (Patton, 1990; Strauss & Corbin, 1990). Strauss and Corbin write of
inductive codes, they are loaded with analytic meaning and may already be considerably
well developed in their own right (p. 68). While deductive concepts are useful, and as I
outline below, were eventually added to this analysis, using them from the beginning in
exploratory research can also cause the researcher to inadvertently impose a structure on
the data that is not in evidence.
39


As I developed the preliminary code structure, it was continually refined as new
chunks of text were compared to existing categories and novel ones were added. The
coded chunks of text within each category were reviewed with a critical eye to determine
inclusion and exclusion criteria, and excerpts were at times reassigned with new codes.
Eventually the preliminary code structure was compared to the existing literature and
potential deductive codes considered and added when relevant. Through this process I
developed a final coding scheme and codebook (Schensul & LeCompte, 2012). The data
was then recoded using the tools provided in Atlas.ti and memos kept throughout the
coding process were used to develop the analysis and resulting ideas and insights, as well
as track potential new directions for research (Bernard, 2006).
Quantitative Data
I reviewed all SMFs collected from the beginning of the screening program in
May 2010 through October 2013: 2440 forms in total. This included all SMFs resulting
from outreach events conducted by SFH as well. Although I collected and digitized data
from these forms, for this dissertation I chose to focus only on those 1340 that resulted
from on-site St. Francis clients. The primary reason for this choice was that this study
focuses on utilization of the services provided by CECAP at St. Francis. Outreach events
involved large numbers of women from one single location; usually this location was the
village in which the event was being held and the surrounding area. These villages were
often some distance from St. Francis, which was the primary reason for taking screening
services to them, rather than the reverse. The events were heavily advertised within the
area to ensure high turnout. For these reasons, outreach clients were not necessarily
typical of the average CECAP user and did not face the same barriers to screening.
40


Furthermore, outreach events usually involved supplementary staff members and
cryotherapy was not always provided. Instead women were referred to St Francis for
treatment. Therefore outreach events also were not typical of standard CECAP
operations.
The data from the remaining 1340 forms had some important limitations. First,
the form itself had undergone several minor changes since the programs inception. Most
important for this study, the variables Parity and Source of HIV Status were not added
until the October 2010 version of the form; the relevant data was therefore missing for
the first 196 clients. Second, certain variables were entered more faithfully by CECAP
staff than others. Clients HIV status, for example, was nearly always written down;
however, questions about PITC were not and were therefore not very useful in the
analysis. The reason for a clients referral to other health services was tracked, but where
the client was referred was often not. I had hoped to use this information but in the end
did not include that variable in my analysis.
Finally, for a period of time during my fieldwork, St. Francis was out of the
reagent needed to test for HIV. Although all clients in the screening program were asked
about their HIV status and, if needed, referred for testing, during this time period testing
was not actually available. It was not the first time during CECAPs tenure that this had
happened either. This explains the high prevalence of unknowns, despite CECAP
typically offering PITC. The final variables uploaded into SPSS can be found in Table 3
above.
The data was entered into an SPSS spreadsheet and cleaned line by line to ensure
accurate translation from paper form to digital medium. All variables were then recoded
41


into numerical values for analysis in SPSS and a codebook was created. Once in SPSS,
descriptive statistics were run to explore the data and describe the sample.
Sample
Women in this sample had an average age of 39.6 (range 18-76) and age had a
normal distribution. The mean parity was 3.8 children (range 0-17) and the mode was 3.0
children. The majority of women (88.9%) were from the Kilombero district, with 5.2%
coming from neighboring Ulanga. The remaining women hailed from Kilosa, Morogoro,
Iringa, Dar es Salaam, Shinyanga, Njombe, Handeni, or Kilolo districts. While 30.8% of
the sample was HIV positive and 43.3% was HIV negative, the 25.9% with unknown
status include those who could not be tested due to the missing reagent. Of those who
were HIV positive, 86.7% were on ART at the time of their cervical cancer screening.
This is not surprising as the government provides free access to ART for individuals
below a specific CD4 count at CTC clinics throughout the country. For those women
with a known HIV status, 70.9% self-reported, while 29.1% brought their CTC cards
with them as confirmation of status.
Four years into the CECAP program, the clients who come to utilize its services
are still primarily those seeking an Initial VIA Screening, as indicated by the SMF forms.
However, this simply notes whether or not it is their initial visit to CECAP at St. Francis.
It is possible they have been screened elsewhere in the past. Only 12 (0.9% of visits) HIV
positive women have returned for their One-Year Follow Up screening, while 4.1% and
0.3% of visits were for Routine VIA Screening or Postponed Cryotherapy respectively.
Every SMF form indicated that VIA had been done on the day the client attended the
program, however, having observed in the clinic I know that the staff typically simply did
42


not fill out a form for women who could not go through with screening for whatever
reason, rather than check No on the form in this box.
There was no missing data for VIA results; a result was lodged for each and every
client. Since the inception of the program, 86.4% (n=l 158) of women had been screened
and found negative. Positive results, meaning small lesions that could be treated with
cryotherapy, were found in 3.1% of cases (n=41), while the staff found large lesions in
3.0% (n=40) and suspected cancer in 7.5% (n=101). In other words, there was some type
of cervical abnormality in 13.6% of clients. For those women who were positive and
eligible for cryotherapy, 78.0% had cryotherapy done the same day. The other cases were
postponed due to other gynecological issues (cervicitis or pelvic inflammatory disease,
for example), or for personal reasons (such as wanting their husband with them). In one
case, the clinic was temporarily without a working cryotherapy tank. One hundred sixty-
one women were given referrals for other services (12.0% of the total client cases). The
most common place of referral was to Ocean Road for suspected cancer (n=94). Another
31 women were referred either to Morogoro or to KCMC in the north of the country for
LEEP, a procedure used to treat larger lesions, depending on which site was more
convenient for them. The remaining women were referred to different areas of St. Francis
for other gynecological issues or for biopsy services.
43


CHAPTER IV:
CHANGING PRIORITIES
Dr. M is one of the most senior physicians on staff at St. Francis
Hospital. As the head of the Gynecology Department, he has overseen the
cervical cancer prevention program since its inception in 2010. But his years
of experience stretch back much further than that and I had been looking
forward to sitting down with him today. Wed had a number of casual
conversations in this, the large room between the administrative block and
the dental clinic where he had both his consulting desk and exam table, over
the course of my time in Ifakara. But today was the first time we touched
on his own perceptions of the growing interest in cervical cancer in
Tanzania.
I started my medical practice in 1982, he told me, peering over his
glasses, and I happened to work at Muhimbili National Hospital. MNH is
the large national referral hospital in Dar es Salaam, and a center for both
teaching and research. The cases of cancer were not as many as it is now.
Now we see a lot of cancers. I have heard variations on this theme from
physicians, patients, and laypersons with whom I have shared my research
agenda while in Tanzania, but a few have also expressed the opinion that
the ability to screen for cancer is simply what has improved, and that both
health care providers and the public are more aware. But Dr. M is firm on
his feelings: I think there are more cases, not that the women are coming
in, but that there are more cases.
Given his long tenure of 30-plus years as a physician, both in the
large city of Dar es Salaam and in rural Ifakara, Dr. M had seen many shifts
in health priorities in Tanzania and in government policy-making. I asked
him to share if and how he felt the government had become more interested
or involved in addressing cervical cancer over the course of his career. Dr.
M thought back to early initiatives to screen for breast cancer, and said he
believed they were responsible for creating an awakening in the government
that cancer was a problem in Tanzania, which eventually led to the current
interest in cancer of the cervix:
So the government has also taken some steps to make sure
women are screened against cancer...I was reading the
parliament from the Ministry of Health, [which] was saying
that this will be one of the things, to screen the women. The
women will be screened as other countries are doing. And
this was repeated by the president, I think two, three days
ago.
His mention of the presidents comments reminded me that I had
read the same article in the Tanzania Daily News last week (Semaya 2013).
It profiled President Jakaya Kikwete and former U.S. President George W.
Bush, who had spoken to the African First Ladies Summit about the
governments five-year strategic plan to reduce cervical cancer and to
44


implement a major immunization program with the help of both public and
private partners. In his speech, President Kikwete called cervical cancer a
silent killer and spoke of the burden of cancer on women in Tanzania. I
mentioned I had read the article too, and Dr. M said that he thought the
number of women they screened at St. Francis Hospital was a bit higher
now than in the previous couple of years because awareness of cervical
cancer was increasing. Its not only news through the radio and that, even
the mass media speaks about cancer of the cervix.
- Adapted from field notes and interview transcript
(July 10, 2013)
Dr. M had spent over 30 years in the medical profession in Tanzania, but only
recently had cervical cancer become an important focus for health education and
intervention. As noted above, he had seen the first cervical cancer screening program
implemented at St. Francis only three years before (as it turned out, this implementation
came just four years before his retirement). His personal experiences reflected broader
changes not only in public health priorities within Tanzania, but also in the arena of
global health. This chapter details the emergence of cervical cancer as an important
global health issue, part of a broader new focus on chronic and non-communicable
diseases, and in particular, how a new screening method changed the way screening
programs could be implemented in developing nations. Corresponding changes within
Tanzania, at both the national and local levels are then examined within this context,
setting the stage for later chapters.
A Global Picture
Approximately 528,000 women developed cervical cancer in 2012 worldwide.
Compared to the 1,677,000 who developed breast cancer that same year, cervical cancer
might at first glance appear to be a less critical global health cancer issue (IARC, 2014).
However, in this case, context is essential. To look at the picture of cervical cancer as a
45


whole is to miss important nuances in the distribution of the disease across regions of the
world. In fact, the burden of cervical cancer falls largely on the developing world, where
85% of cervical cancer cases and 87% of cervical cancer deaths are found (Ferlay et al.,
2013). In many of these regions, cervical cancer is the number one female cancer and the
largest, or one of the largest, contributors to years of life lost from cancer (Arbyn et al.,
2011; Yang etal., 2004).
Due to this disparity between different regions of the world according to
development status, cervical cancer is considered to be a disease of poverty. Tsu and
Levin (2008) argue that the burden of cervical cancer is not simply an important public
health issue, it is also an issue of health inequity. It should therefore be discussed not only
in terms of actual objective disparities, but also in terms of how poverty contributes to
and shapes the distribution of disease:
While every untimely death, illness and suffering is of concern, a widely
accepted principle of social justice says those who are most vulnerable to
illness or less able to access services, because of social or demographic
characteristics over which they have little control, are entitled to special
consideration. Poverty is the most commonly identified source of inequity,
but other aspects include gender, ethnicity, religion, geography, age,
education and social status, especially when these cluster and magnify
negative effects. These concepts of inequity, both in health status and
access to health services, resonate strongly with regard to cervical cancer.
[Tsu & Levin, 2008, p. 104-5]
Furthermore, this disease primarily affects a critical portion of the population:
young to middle-aged women with children. These women are crucial members of
society and may be the only source of economic support and caregiving for their families
(Anorlu, 2008; Franco & Harper, 2005). In sub-Saharan Africa, the region in which
Tanzania is located, this is particularly concerning given the impact of the HIV
pandemic, where older women are increasingly stepping into these roles as their younger
46


counterparts fall victim to the disease (Knodel, Watkins, & VanLandingham, 2003;
Schatz, 2010). In addition, as has often been noted in the literature, women are
responsible for the majority of agricultural work and food production in much of the
developing world (Ellis, Blackden, Cutura, MacCulloch, & Seebens, 2007; Ishikawa,
1999). This is certainly true in Tanzania, where according to the 2006 World Economic
Forum Global Gender Gap report, Tanzania was ranked number 1 globally, out of 115
countries, in terms of womens economic participation (Ellis et al., 2007).
As I noted in Chapter I, the statistics for a woman in sub-Saharan Africa when it
comes to cervical cancer are discouraging to say the least. Sub-Saharan African women
are not only more likely to develop cervical cancer than their counterparts in Europe or
the United States, their cases are more likely to be fatal (Ferlay et al., 2013; Franco,
Schlecht, & Saslow, 2003). Once diagnosed with cervical cancer, they have a probability
of survival of just 21%, compared to 42-70% in other areas of the world (Anorlu, 2008;
Okonofua, 2007). The situation is particularly bad when the geographic area in question
is narrowed to East Africa, which has a higher incidence rate than any other sub-Saharan
region (Ferlay et al., 2013). While the overall mortality rate for women globally is only 9
per 100,000, it is 34 per 100,000 for the region (Perng et al., 2013; Peters, Soliman,
Bukori, Mkuchu, & Ngoma, 2010).
When compared to both the region of East Africa and sub-Saharan Africa as a
whole, Tanzania itself still stands out as having one of the highest age-standardized rates
of cervical cancer at 50.9 cases per 100,000, and it is the most common cancer to affect
young women between the ages of 18-44 (Castellsague et al., 2012; Hutubessy et al.,
2012). A Tanzanian woman under the age of 64 has a cumulative risk of cervical cancer
47


nearly four times that of women worldwide (Okonofua, 2007). She is also more likely to
have her disease progress unchecked and to present for care at a later stage of cancer than
her counterparts in developed nations (Kidanto, Kilewo, & Moshiro, 2002; Mosha et al.,
2009). Despite these disheartening facts, cervical cancer as a public health concern is a
fairly new arrival to the national scene.
Health Transitions
The word cancer often evokes certain images and ideas. Traditionally we think of
cancer as a chronic or non-communicable condition, often one that develops in middle- or
advanced-age, though there are certainly cancers that strike children or young adults. As
an umbrella term, it is often lumped in with heart disease, diabetes, and other so-called
lifestyle diseases that affect developed nations where life expectancy is high and
sanitation and other public health interventions have made the threat posed by infectious
diseases (in particular, infectious childhood diseases) far less likely. The term
epidemiological transition (or health transition) is often used to describe this changing
pattern in causes of disease and mortality: As nations undergo development, infectious
diseases decline and chronic and degenerative diseases take their place (Frenk, Bobadilla,
Stern, Frejka, & Lozano, 1991; Omran, 1971).
Chronic diseases are on the rise globally, including in low- and middle-income
countries (Kanavos, 2006). This trend, termed the globalization of chronic disease, has
not been seen to the same extent in sub-Saharan Africa as it has in other developing
regions, however, it is occurring and is influenced in part by changing demographics and
the increasing impact of a westernized lifestyle (Kanavos, 2006, p. viii 15; Morhason-
Bello et al., 2013). But the trend has not corresponded everywhere to a reduction in
48


infectious diseases. Instead, regions like sub-Saharan Africa are left with a double
burden and health care systems that are ill equipped to deal with this (Maher, Smeeth, &
Sekajugo, 2010, p. 943).
The situation is further complicated by the fact that certain types of cancer
muddle the seemingly neat categories of communicable and non-communicable diseases.
While cancer itself may be non-communicable, the past three decades have greatly
expanded our understanding of the role viruses play in the development of some
malignancies, for example Hepatitis B/Hepatitis C and liver cancer, and Helicobacter
pylori and stomach cancer (Franco, Duarte-Franco, & Ferenczy, 2001; Jones, 1999). In
the case of cervical cancer, the role of the human papillomavirus is essential in
understanding both how cervical cancer develops, and how it can be prevented.
Human papillomavirus, or HPV, is the most common sexually transmitted
infection (STI) and is what is called a necessary cause of cervical cancer (Bosch, Manos,
Munoz, & al., 1995; Centers for Disease Control, 2004; Dailard, 2003; Franco et al.,
2001; Walboomers et al., 1999). This relationship was only confirmed in the late 1990s
(Walboomers et al., 1999). Specifically, cervical cancer is caused by particular strains, or
types, of HPV that infect the anogenital tract; although 40 HPV types are known to infect
this region, less than half actually cause cancer (Franco & Harper, 2005). Being termed a
necessary cause indicates that while infection with HPV alone is not sufficient for
cervical cancer to develop in an individual, the virus has been implicated in 99.7% of
cervical cancer cases, i.e., in nearly all cases the presence of HPV is necessary for an
individual to develop cervical cancer (Franco & Harper, 2005; Walboomers et al., 1999).
While the average person may be more aware of the association between smoking and
49


lung cancer, in fact, this association is weaker than that between HPV and cervical cancer
(Franco et al., 2001).
While HPV is a necessary cause, it is not a sufficient cause of cervical cancer
(Franco & Harper, 2005). In other words, infection with HPV is not enough by itself to
cause the development of cervical cancer. In the majority of cases, infection is cleared by
the individuals own immune system before causing any harm, however persistent
infection with particular types can lead to eventual cancer of the cervix (Centers for
Disease Control, 2007; Krishnan, 2008). At this point in time, attempts to discover
exactly what the threshold is after which an infection can be considered persistent have
been unsuccessful (Trottier & Franco, 2006). Researchers have been more successful in
determining factors that increase a womans risk of contracting HPV and of having that
infection progress to cervical cancer; this will be discussed further in Chapter V.
The Macro-Policy Environment
These categories, communicable and non-communicable, may be problematic
when applied to cervical cancer, but they are important when it comes to understanding
health priorities at the national and international levels. Farmer (1999) has pointed out
that the health transitions model held a certain appeal for the powerful international
financial institutions that shape development programs. This appeal had multiple
components. First, developed nations that had already reduced their burden of infectious
diseases provided successful models for development. Second, infectious diseases had the
potential to cross borders and cause epidemics. Furthermore, through the use of vaccines,
certain infectious diseases could be prevented, or even potentially eradicated (Anderson
& Chu, 2007; Davies, 2010). These were also coupled with certain prevalent ideas around
50


chronic conditions: that they primarily occurred in high-income nations, that tackling
them would be more expensive, and that they were largely a problem of the elderly
(Anderson & Chu, 2007).
The infectious disease burden is acute in the developing world, along with, for
women, what Tsu et al. have called the crushing burden of morbidity and mortality
associated with reproduction (2013, p. 683). This cannot be denied. However, as noted
above, non-communicable diseases are on the rise. In the last decade, there have been
increasing calls for renewed emphasis on both non-communicable conditions themselves,
and non-communicable conditions that interact or arise from communicable diseases,
such as cervical cancer (Anderson & Chu, 2007; Fuster & Voute, 2005; Maher et al.,
2010; Seffrin et al., 2009). Development agency funding focused on addressing public
health outcomes in the developing world has also not kept pace with its non-
communicable disease burden (Seffrin et al., 2009).
There are limited resources to address the many health needs that exist in low-
and middle-income countries. HIV, tuberculosis, and malaria in particular are often the
center of much international aid, along with maternal mortality (Anderson & Chu, 2007;
Morhason-Bello et al., 2013). However, in some cases, programs that address non-
communicable diseases may in fact be a better use of limited global health funds, even if
their actual disease burden is not as large. Cervical cancer has taken a huge toll on
women in the developing world. Technologies currently exist that could alleviate much
of this burden and they are transferable to low-resource settings, unlike other important
causes of morbidity and mortality for which comparable treatments and technologies may
51


not be available, or may not have the potential for mass implementation (Yang et al.,
2004).
Transferable Technologies
In fact, cervical cancer appears at first glance to be an ideal target for public
health intervention in the developing world. It can be prevented in the majority of cases
using already developed screening techniques, with some nations reporting a reduction in
incidence of the disease as high as 70-90% following the implementation of high-
coverage screening programs (Gustafsson, Ponten, Zack, & Adami, 1997). It is also
typically slow to develop once an individual is infected with the HPV virus, often taking
years or decades for persistent HPV infection to develop into invasive cancer, and during
much of that time early medical intervention is possible.
Routine use of cytology in the form of the Pap smear has played a major role in
dramatically reducing the incidence of cervical cancer in developed nations since its
development in the early 20th century, and remains an important part of prevention
strategies there today, along with the newer HPV testing (Crum, Abbott, & Quade, 2003 ;
U.S. Preventive Services Task Force, 2012; Vilos, 1998). Unfortunately, these screening
tools, and standard follow-up treatment methods such as LEEP, cryotherapy, or cold-
knife conization, require medical infrastructure and financial resources that are simply
not always available in resource-poor regions (Drain, Holmes, Hughes, & Koutsky,
2002). Furthermore, a lack of trained staff and the mismanagement of what few resources
do exist have also played a role in stifling cervical cancer programs (Anorlu, 2008;
Bingham et al., 2003; Megevand, Van Wyk, Knight, & Bloch, 1996).
52


With this in mind, researchers developed new methods for screening and treatment
in low resource settings, focusing on using visual inspection with either acetic acid (VIA)
or Lugols iodine (VILI). Both solutions are utilized in a similar manner. For VIA, the
healthcare worker administering the screening uses a cotton swab to spread a solution of
3-5% dilute acetic acid across the surface of the cervix. After waiting for a minimum of
one minute, the worker then examines the cervix using a flashlight for any lesions, also
known as acetowhite lesions, which will have turned white during the intervening time
(Sankaranarayanan, Gaffikin, Jacob, Sellors, & Robles, 2005). During the screening
process, the worker will also examine the cervix and vaginal canal for bleeding, unusual
growths, inflammation, or discharge. Unlike the Pap smear, which requires a waiting
period for the results and typically a follow-up appointment, VIA allows treatment to be
conducted using cryotherapy during the same visit if the visible lesion is below a certain
threshold, leading to the name one-visit screen-and-treat approach. Larger lesions or
lesions which have spread beyond the cervix will still typically require more extensive
follow up treatment.
VIA has now been evaluated in comparison with the Pap smear and HPV testing
(Denny, 2005; Sankaranarayanan et al., 2005). All methods have limitations, for
example, the Pap smear does not directly test whether or not a woman is positive for
HPV, but rather looks for changes to the cervix brought on by the virus (Krishnan 2008).
It therefore is not surprising that while it has a high specificity (98-99%), it has only a
moderate sensitivity (50-75%) (Krishnan, 2008; Sankaranarayanan et al., 2005; Urasa &
Darj, 2011). On the other hand, HPV testing has a high sensitivity, but a less impressive
specificity and positive predictive value; this high sensitivity means that many women
53


who receive a positive result are likely to spontaneously clear the infection before any
cytological abnormality develops (Franco et al., 2003; Jeronimo & Schiffman, 2006;
Krishnan, 2008).
The reported sensitivity and specificity values of VIA have varied, with one meta-
analysis demonstrating ranges from 67% to 79% and from 49% to 86%, respectively
(Sankaranarayanan et al., 2005). However, in comparative studies around the world, VIA
has typically been shown to have a higher or similar sensitivity to the Pap smear and
lower specificity (Bhatla et al., 2010; Bomfim, Santana-Franco, & Bahamondes, 2005;
Doh et al., 2005; Goel et al., 2005; Shastri et al., 2005). In other words, VIA effectively
detects true positives, but it does not detect true negatives as well as the Pap smear. As a
result, one concern with VIA is that it results in more false positives and therefore
overtreatment, given that there is no biopsy or confirmation process for acetowhite
lesions in a one-visit screen-and-treat approach. Additionally, while VIA is useful for one
of the two types of cervical lesions, squamous lesions, it is much less effective in
identifying glandular precursor lesions (Sankaranarayanan et al., 2005).
There are many benefits, however, to using one-visit visual inspection methods.
Cost is typically low and the resources needed to perform the screeningcotton swabs,
vinegar, a flashlight, a private room or privacy screen, a speculum, and an examination
tableare often available in or near clinics, even in low-resource settings. While training
is necessary to properly carry out the screening and to recognize the potential conditions
one might encounter, VIA can often be done by paramedical, or even nonmedical,
personnel who have undergone short training programs (approximately 5-10 days) rather
than requiring physicians to carry it out (Ghosh, Gandhi, Kochhar, Zutshi, & Batra, 2012;
54


Sankaranarayanan et al., 2005). Furthermore, VIA reduces the burden on the patient in
terms of the cost, time, and transportation requirements of a multiple visit strategy. As
Sankaranarayanan et al. (2005) write, The immediate availability of results after visual
testing is a major logistic advantage (p. S8). Despite the potential for overtreatment,
VIA is considered to be a viable and valuable screening tool in low-resource settings
(Belhadj, Rasanathan, Denny, & Broutet, 2013).
Cervical Cancer in Tanzania
Although the first cancer control program in Tanzania was initiated in 1978,
through funding provided by USAID, a report by that same agency evaluating the
sustainability of the program noted that, the project began at a time when few believed
that cancer was a problem in Africa (Dunlop et al., 1990, pp. F-l). This view has slowly
changed in the decades since, corresponding to the larger changes in global health
priorities outlined in the previous section. Ocean Road Cancer Institute, a public
institution devoted solely to the issue of cancer, has been open since 1996 and treatment
for cervical cancer is theoretically free, assuming a space is available and the patient can
reach the hospital (Ocean Road Cancer Institute, 2014; Peters et al., 2010). Additionally,
the Strategic Plan of the National Institute for Medical Research (NIMR) in Tanzania,
which traditionally focused on infectious diseases, also included non-communicable
diseases such as cervical cancer for the first time in 2008. After several fits and starts
during the latter half of the 20th century, and despite difficulties with the quality of data
available, a new attempt to build a national cancer registry began in 2010 (African
Cancer Registry Network, 2015). Preliminary HPV vaccine trials and cost analyses have
taken place, and larger-scale programs began last year in 2014 with the assistance of the
55


Global Alliance for Vaccines and Immunization (GAVI) (Cagney, 2013; GAVI, 2013;
Hutubessy et al., 2012; Quentin et al., 2012). However, while the fourth edition of the
Standard Treatment Guidelines for Mainland Tanzania issued by the Ministry of Health
and Social Welfare (MOHSW) has updated its entry for cervical cancer, national cervical
cancer service policy guidelines still do not exist (Lyimo & Beran, 2012; Ministry of
Health and Social Welfare, 2013; Urasa & Darj, 2011).
Few research studies have been done on cervical cancer specifically in Tanzania,
though the general regions of sub-Saharan Africa or East Africa have been part of wider
discussions in the literature on the disease in the developing world for some time now
(Drain et al., 2002; Jones, 1999; Louie, de Sanjose, & Mayaud, 2009; Sankaranarayanan,
Budukh, & Rajkumar, 2001). However, when the relationship between HPV and cervical
cancer was confirmed in the late 1990s, attention quickly turned to the potential
implications for vaccination as a tool for primary prevention. The few studies related to
cervical cancer in Tanzania done both just before this confirmation and during the decade
that followed therefore unsurprisingly focused on the distribution of HPV types present
(Bosch et al., 1995; Clifford, Smith, Plummer, Munoz, & Franceschi, 2003; Mayaud et
al., 2001; Mayaud, Weiss, Lacey, Gill, & Mabey, 2003; Munoz et al., 2004; Ter Meulen
et al., 1992). Often these studies were not restricted to Tanzania, but included the nation
in a broader grouping of developing countries. In addition, there was also an interest in
evaluating the existing resources for screening and treatment, but these resources were
limited and primarily confined to large health centers in cities (Chirenje et al., 2001;
Ngwalle et al., 2001). Nor is Tanzania unusual in this respect; rather, with a few
exceptions, particularly in Latin and South America, this was the norm.
56


While vaccination held great promise for reducing the overall incidence of the
disease over time, it could not protect women who were already infected with HPV.
There, screening remained the primary hope. Visual inspection methods were developed
and tested in the mid-1990s, and after early trials verified their usability, they opened up
possibilities for more effective and efficient cervical cancer screening programs in low-
resource settings. Unfortunately, given the already burdened and limited health care
infrastructure, implementation of screening programs in Tanzania was fairly non-existent
outside of large health centers and mobile units (Pemg et al., 2013; Peters et al., 2010).
NGO and Government Partnerships
International donors and NGOs have since broadened the availability of cervical
cancer screening, though the infrastructure is still quite limited (Peters et al., 2010;
WHO/ICO, 2010). These types of actors in the health care arena are not an uncommon
phenomenon in Tanzania, where vertical intervention programs addressing specific health
issues have frequently been implemented to address gaps in services that cannot be
adequately provided by the struggling national health care system (Hanson, 2000).
During previous trips to Tanzania and during my field period, I heard stories of mobile
vans or teams of doctors who had come through the area to screen for breast or cervical
cancer, but along with these came complaints that they rarely came on a set schedule and
sometimes did not return at all.
Over the past decade, partnerships between the government and these entities
have increased, in an effort to build more sustainable programs. Ocean Road, for
example, began a screening program in 2003 at its main campus in Dar es Salaam, as
well as introduced an outreach program that includes education and training for clinics
57


around the country (Peters et al., 2010). During my visit to Mwanza in 2010,1 witnessed
another example of this when I spoke with members of a foreign non-profit working in
partnership with the Bugando Hospital Obstetrics/Gynecology Department to increase
their outreach screening services in the surrounding area, and Bugando itself was
building their own clinic-based prevention program. The scope of need for screening
services however is still overwhelming. Perng et al. (2013) observed while there were 74
active cervical cancer screening sites in Tanzania by March of 2012, they were still
concentrated in urban areas, despite the fact that three quarters of women at risk for
cervical cancer in Tanzania actually live in rural locations.
The cervical cancer prevention program that served as the field site for this
dissertation research project was begun in 2010 by the U.S-based non-profit Jhpiego and
its consortium partners, including the Tanzanian government, with funding provided by
USAID. It was developed as an addition to the MAISHA (Mothers and Infants, Safe,
Healthy, and Alive) program. MAISHA included multiple previous program components
focused on maternal and child health. According to a Jhpiego representative, at CECAPs
inception, the organization was already involved with St. Francis Hospital, the nearby
Kibaoni Health Center, and Morogoro Regional Hospital, having implemented program
components at each of these focused on obstetric emergency care and PMTCT
(Prevention of Mother-to-Child Transmission of HIV). Cervical cancer prevention was
the next step. Jhpiego provided the training for screening providers, as well as supplies,
and administers funding for the program.
An early advocate for the one-visit screen-and-treat approach using VIA, Jhpiego
works on womens health issues all over the world and has an office in Dar es Salaam.
58


Since turning their focus to cervical cancer in Tanzania, Jhpiego has taken a multi-
pronged approach. In addition to supporting 21 cervical cancer prevention programs sites
around the country, they have also chosen to partner with Tanzanian government
agencies to provide technical support, education and training, and a variety of advocacy
tools. Some of that technical support is being directed towards an effort by the MOHSW
to correct the lack of national service delivery guidelines noted in the previous section
(Jhpiego, 2015).
While the cervical cancer prevention program does target one disease, Jhpiego has
attempted to build on this approach by incorporating the MOHSW guidelines calling for
Provider-Initiated HIV Testing and Counseling (PITC) into their services. The program
intake forms include a section detailing whether or not PITC was offered and accepted,
and if so, its result (HIV status). They encourage reciprocal referral systems between HIV
testing and cervical cancer screening as well, and encourage program staff to build
referral relationships with other clinics within the same institutions.
The CECAP program at St. Francis provide single-visit VIA and cryotherapy, but
women with more advanced cervical lesions or invasive cancer cannot be treated there.
Certain central sites were given the training and resources to provide LEEP for larger
lesions. For women in Ifakara, the closest site was Morogoro, a half-day trip by bus.
Invasive cancer cases were treated in Dar es Salaam at Ocean Road. When I first arrived,
St. Francis could not even provide biopsies, though they had a surgical theatre available
in the hospital and a staff member capable of conducting the procedure, because the
program did not have its own punch biopsy device. This meant women had to travel to
Dar es Salaam once for a biopsy, and then return if the results indicated it was necessary.
59


Eventually program staff appealed to Jhpiego and obtained their own device. But the
biopsy then had to be sent to Dar es Salaam, typically by bus, for analysis, the results of
which were transmitted by phone and email back to staff at St. Francis before being
conveyed to the patient.
Jhpiegos intention in creating the cervical cancer prevention programs within
existing CTC or health care facilities in partnership with the Tanzanian government was
not for them to remain independent units indefinitely, but rather for the services to
eventually be integrated into the health care system after a period of oversight. Jhpiego
provided this oversight for the first four years of the program at St. Francis, and funding
was provided by USAID. However, in 2014, Jhpiego planned to turn over the
administration of the program to the government, who would also take responsibility for
its financing. This dissertation focuses on one cervical cancer prevention program in one
hospital in Tanzania, yet the implementation and utilization of this program have broader
implications, particularly given that CECAP was standing on the threshold of fuller
integration into the overall hospital structure and health care system. The next three
chapters examine the issue of utilization from the perspective of the clients; in Chapter
VIIII will return to this issue and examine utilization in the context of the programs
experiences over its first three years.
60


CHAPTER V
LOCAL CONSTRUCTIONS OF CERVICAL CANCER
Ummi
Ummi was the first woman we spoke with in the Gynecology
Department, and she came into the room a little shyly, hesitating on the
threshold and looking around as her eyes adjusted to the shadowed room.
We had closed the curtains for privacy since the room faced a courtyard
between two wards, and as usual, the overhead light was off. A married
mother of two, Ummi worked as a nursery school teacher. She had traveled
to St. Francis from her home in a village near the Udzungwa Mountains
National Park, approximately 35 miles away. In a hoarse, quiet voice, she
told us that in addition to hearing radio and television announcements about
cervical cancer, she had just that day seen an educational poster in one of
the doctors rooms about the disease, warning women that when they saw
or experienced some changes that were not ordinary, they should hurry to a
health center. From these sources, Ummi knew that cervical cancer was
related to sexual intercourse and that having sex at a younger age or with
many partners were particularly problematic. We asked her who was likely
to get cervical cancer, and in her response she reiterated this belief:
Women. Women, starting from a child who has already
matured. According to the way I understand, I think sexual
intercourse, especially to girls, or unsafe sex. Also the
number of pregnancies, as one may have many deliveries
that are consecutive. This may cause the uterus to get loose,
or to get certain diseases that get into the uterus.
Ummi placed great stock in the power of hospital medicine, telling
us that she trusted doctors because they had studied all the conditions that
are disturbing us... like malaria, fever, abdominal disturbance, and typhoid.
The brief health history she gave us bore that out, as she recounted several
different health concerns that had led her to the clinic. As a Muslim, when
sick she would pray to God for the strength to get to the clinic, but, she told
us, she did not involve herself in traditional medicine: I am usually 100%
sure that when I get to a health center and consult a health attendant, I will
get better. I mean I will be healed. However, Ummi told us, it was up to
the individual to recognize when they were sick and seek care, and then to
follow the resulting medical instructions. In the case of cervical cancer, she
shared what a woman should look out for and what would happen if she
chose to ignore those symptoms:
According to the way I understand, I think a person will feel
low abdominal pain and also may have frequent urination or
may be unable to control the passing of urine so the urine
just overflows...If a person diagnosed to have cervical
cancer is going to be reluctant to go to the place where they
61


are supposed to, I know at the end that the uterus will rot and
it will have to be removed. And this person will not be able
to have children again. So it is important to go quickly to a
health center so that they get immediate treatment.
I was curious what her response would be when we asked her if she had
herself been screened in the past, given that she had heard a fair amount
about the disease. To my surprise, she wasnt sure.
Maybe it was done to me, but I did not know the name if it
was cervical cancer screening. Because I have been cleaned1
before in the year 1993 during Dr.______s time before I had
a child, and in the year 1994 I had a child. I have not been
cleaned again since then but I have just had an operation for
the appendix and am now on treatment for a swelling.
Did you know what was being done to you?
At that time I did not know because I was only told that my
reproductive tubes had blocked.
Ummi was only our third interview participant; my surprise at her response
later faded as I came to understand that it was not uncommon for many of
the women we interviewed to be unsure as to their diagnosis or the specific
medical procedures they had undergone. Sometimes this was due to a
forgetful memory or an unclear understanding, other times they said that
they had never actually been told to begin with. While Ummi wasnt sure
of her screening history, she was very interested in being screened, asking
me to tell her how to go about it, and said that she would have done it sooner
if she lived near to St. Francis. She pointed out that screening was the only
way to know for sure what was wrong: ...this is a disease that is inside, so
even when I get abdominal pain, I will not know whether its cancer or not.
Understanding Cervical Cancer in the Local Context
According to a western biomedical understanding, cervical cancer has a fairly
clear etiology. That risk for the disease is related to sexual activity has been postured by
physicians since at least the nineteenth century, and was confirmed with the identification
of HPV as a necessary cause in the 1990s (Bosch et al., 1995; Lowy, 2011; Walboomers
et al., 1999). Early age at first intercourse, multiple sexual partners, sex with a partner
1 The use of the term cleaned here is a little ambiguous. According to the RA who sat
in on the interview, and several informants, this term is most likely referring to curettage,
which may be a veiled reference to abortion in this particular context.
62


who has had multiple partners, and frequent sex with an infected partner all increase a
womans risk of contracting HPV (Franco et al., 2003; Krishnan, 2008). Furthermore,
HPV is spread through skin contact, which means that condoms are not as effective in
preventing its transmission as they are for infections spread through bodily fluids
(Centers for Disease Control, 2004; Krishnan, 2008).
Once a woman is infected with HPV, there are other factors that increase the
likelihood the infection will become persistent and develop into cancer. High parity, for
example, has been identified as a co-factor for cervical cancer due to repeated trauma to
the cervix during labor. Poor hygiene and nutrition, tobacco use, oral contraceptive use,
infection with other STIs, and HIV infection are all factors that may play a role
(Castellsague & Munoz, 2003; Trother & Franco, 2006). Finally, social and cultural
practices and norms, as discussed in this chapter, may influence the degree to which these
risk factors are a concern in specific settings.
The symptoms of cervical cancer are also known, though they may be ambiguous
in presentation. Women may not experience any symptoms at all during early stages of
the disease (Castellsague, 2008; Krishnan, 2008). This is problematic, as it is well-
established that detection and treatment of the disease in its early stages can dramatically
improve survival among cervical cancer patients (Castellsague et al., 2012). Adding to
these difficulties, some symptoms, such as blood spotting, bleeding between menstrual
cycles, and vaginal discharge may mimic other sexually transmitted infections or fall
within the range of what women consider normal for their own bodies.
Yet biomedical explanations do not always correspond to local understandings of
illness, which can in turn influence whether or not individuals (and those around them)
63


consider themselves sick and in need of medical care, and if so, what specific types of
medical care they seek out. For this reason, medical anthropologists typically draw a
distinction between disease, used to denote a biomedical condition, and illness, indicating
the individuals personal experience of ill health, an experience which Kleinman argues
is always culturally shaped (1988, p. 5). Illness therefore involves not only the types of
physical sensations that an individual or group qualify as cause for concern, but also how
a person with those symptoms should behave, the potential cause of those symptoms, and
the type of treatment or practitioner that should be consulted (Kleinman, 1988; Kleinman
et al., 2006). This relationship, between a biomedically-defined condition and the social
context in which it sits, is reciprocal and should be examined holistically (Jordan, 1983).
The questions that elicited the responses Ummi gave us in the opening of this
chapter arose from my desire to better understand how women in rural Tanzania
conceptualize and assign meaning to cervical cancer. Lakoff (1987) has written that an
understanding of how we categorize is central to any understanding of how we think and
how we function (p. 6). While classification is an activity practiced by all human
cultures, local systems of categorization have important implications for health and
illness (Kleinman, 1988; Nichter, 2008). For example, the symptoms that women identify
as related to cervical cancer, or at the very least, identify as potentially problematic, are
important to know because they shape care-seeking behavior, as do those behaviors or
traits that are seen to put women at risk of cervical cancer in the first place (Bradley et al.,
2006; Chirwa et al., 2010 ). Furthermore, cultural views on treatment and prognosis may
impact whether or not women view screening as a benefit (Bingham et al., 2003; Drew &
Schoenberg, 2011), which will be explored further in Chapters VI and VII. This chapter
64


examines how cervical cancer is conceived of and understood among the participants we
interviewed. Chapter VII will then explore how this conceptualization interacts with the
challenges women face in seeking screening and treatment services in rural Tanzania, and
the benefits they perceive to come from screening.
Accessing Familiar Biomedical Models of Infection
Very basic biomedical ideas regarding infection were familiar to most Tanzanian
women I met, given decades of public health campaigns around common infectious or
communicable conditions, such as malaria and HIV. In my own daily life, I was
frequently warned to take care regarding potential microbes or parasites in unsafe food
and water, and whenever I felt the slightest bit ill, regardless of where in my body the
illness was centered, I was advised to go to the clinic to check for malaria parasites.
Because I was a mzungu, I was warned, I was even more susceptible because my body
was not used to these elements; I had no built up resistance. Furthermore I conducted
this research among a clinical population who largely expressed familiarity with the
health care system, as well as with the importance of hygiene, a good diet, and
protecting oneself from infections or risk, whether that protection came in the form of
mosquito nets, condom use, or clean water. Thus, it was not surprising that many of the
participants we interviewed from both CECAP and the GD communicated a general
understanding of the idea that infectious agents could put an individual at risk for cervical
cancer and certain activities increased that risk. This knowledge base was typically
shallow however, and lacking in specific details.
65


We opened our discussions about cervical cancer simply by asking women what
they could tell us about cervical cancer, or sarataniya shingoya kizazi,2 but this was
rarely enough to elicit responses about the actual cause. The things that first came to
mind for participants were radio announcements they had heard, or symptoms they knew,
or personal anecdotes about family members and neighbors with cancer. When we asked
about the specific cause of cervical cancer, nearly half (n=22) of the women we
interviewed told us they did not know.3 The matter-of-fact response of Jesca, a 30 year-
old mother of three in the GD was fairly representative: My dear, to be honest, I dont
know anything about cancer. She went on to say, I know cancer because I hear that
there is a disease called cancer.. .but I dont have any knowledge. But those who did
have a response drew on generalized biomedical terms and their familiarity with other
medical conditions (particularly those they saw as related in some way) in describing the
causes, symptoms, and risk factors of cervical cancer. Particularly common were
2 Saratani is the Kiswahili word for cancer (sometimes also called kansa). Shingoya
kizazi, or neck of the womb/uterus, is used to refer to the cervix; another name is
mlango wa kizazi, or Door/entrance of the womb/uterus. Both phrases were used
interchangeably to refer to cervical cancer, however, for the sake of consistency, we used
sarataniya shingoya kizazi in all study documents.
3 The Patient Information Forms did not collect data on education level, so it is difficult
to draw definitive conclusions as to whether these individuals had less education than
those who were able to share their knowledge of the causes of cervical cancer. Using
occupation as a rough proxy for education, however, 18 of these 22 I dont knows were
farmers or tailors, occupations that do not require formal education, and the 19th was a
self-described business woman (an ambiguous term in this setting where that may mean
anything from a formal professional position to a person who cooks food at home to sell
on the street). Additionally, one was unemployed, and another worked with the local
office of an HIV non-profit, but she had been recruited due to her HIV status and local
knowledge rather than a high educational background. The last was a secondary school
teacher. In contrast, while the group of participants who did respond affirmatively to this
question included many farmers as well, and four businesswomen, it also included seven
teachers and health care workers.
66


descriptions of the risks posed by sexual activity and childbirth. In fact, it was difficult to
separate womens understanding of the biological causes of cervical cancer from their
recognition of the risk factors, as they were tightly wound together during our interviews.
Infections, Microbes, and other Causes of Cervical Cancer
The most commonly cited causes of cervical cancer drew upon models of disease
causation pulled from biomedicine, but without specifically referencing human
papillomavirus or indeed stating that one specific virus was responsible for the disease.
This is in keeping with previous research from both Tanzania and other parts of sub-
Saharan Africa showing that knowledge of cervical cancer is typically quite low in local
populations (Abotchie & Shokar, 2009; Kidanto et al., 2002; Mutyaba, Faxelid, Mirembe,
& Weiderpass, 2007; Ports, Reddy, & Rameshbabu, 2013). Even those participants
employed as health care workers, including two nurses and a ward attendant, did not
make this type of reference, though the ward attendant, Hellena, did outline a model of
infection for us: ... cervical cancer is the pathogens which one acquires during sexual
intercourse or by having multiple sexual partners. So the pathogens are transferred from
men to women and inhabit the uterus. While we interviewed most CECAP women
before they sat through the educational session that opened each screening day, I initially
thought (incorrectly) that perhaps those who were interviewed afterwards would be able
to be more specific. After observing a number of these sessions, however, I learned that
the staff typically only referenced HPV by name once at the beginning of each talk.
67


Thereafter they commonly used the non-specific word dudu4 with the clients to describe
the virus.
Instead, the small number of women (four from CECAP and four from the GD)
who specifically referred to a microbial cause of cervical cancer used generalized terms,
discussing pathogens and microbes as commonly causing infection rather than a specific
virus. Salome, a 52 year-old CECAP client who lived nearby and had heard from several
women she knew about the screening program, came perhaps the closest when she told
us, From what I know, cancer is the disease when certain bacteria attack a certain part of
the body.. .the microbes start slowly and eventually are untreatable. Yet even she only
went so far as to imply that one specific type of pathogen was at fault. Three of these
eight women specifically stated that these disease agents were deposited by men during
sexual intercourse, indicating an awareness of the sexual transmission of the disease.
They were not alone in linking cervical cancer to sexual activity, however, most women
framed this in terms of risk rather than causation.
Only eight women mentioned infectious agents like pathogens or microbes, but
many more indirectly referenced biomedical models of infectious disease. However,
these models were interpreted through a cultural lens in which certain activities,
conditions, or injuries were seen to place the body at risk and to result in certain
symptoms. Medical anthropologists have often pointed out that biomedical diseases can
rarely be divorced from the social and cultural contexts in which they exist when
4 Dudu is the Kiswahili word for insect or bug (plural: madudu), and a form of the word
is also used for bacteria/germs, vijidudu. There is a Kiswahili word for virus, virusi,
however during my time at CECAP the word dudu was typically used with clients
instead.
68


attempting to understand the meanings attached to them (Jordan, 1983; Kleinman, 1988;
Lock & Scheper-Hughes, 1990). Kleinman (1988) devised the term explanatory models
to explain how individuals suffering from illness made sense of their experiences. These
models, he argued, enable sick persons to order, communicate, and thereby symbolically
control symptoms (p. 49). In this study, none of our participants had cervical cancer, but
they were asked to discuss their perceptions of the diseaseperceptions that were
influenced by their experiences with other medical conditions. The models proposed by
participants were principally influenced by their knowledge of other sexually transmitted
or gynecological conditions, the HIV pandemic, and high rates of complications from
childbirth in Tanzania, and will be discussed further in the next section.
While alcohol and tobacco use are known to increase the risk that a womans
HPV infection will develop into cervical cancer (Sussman et al., 2007; Trottier & Franco,
2006), only one participant spoke of this connection as a potential cause of the disease.
Tobacco use among women in Tanzania is quite low according to the limited data
available, however it is increasing in the developing world in general (Sussman et al.,
2007; WHO/ICO, 2010). Among this population at least, it did not appear to be
associated with cervical cancer either as a cause or a risk factor in the same way it might
be for other cancers.
Within Tanzania, and elsewhere in sub-Saharan Africa, witchcraft has been
included in lay models for other reproductive and sexual health issues, such as infertility
or miscarriage (Kielman, 1998; Okonofua, 2007; Plummer et al., 2011; Upton, 2001), as
well as for cervical cancer and its symptoms (Awodele et al., 2011; van Schalkwyk,
Maree, & Wright, 2008; White et al., 2012). In our interviews, only five woman (three
69


CECAP clients and two GD patients) related bewitchment directly to cervical cancer.
Each did so, not as a matter of conviction, but as an example of what someone else in
their community might believe to be the cause of their ailment. Sandra, for example, told
us that those in her village sometimes refused to believe they had cervical cancer: They
may claim to be bewitched, while as a matter of fact they were supposed to go to the
hospital for further consultations and investigations. Witchcraft wasnt necessarily seen
as causing cervical cancer specifically; rather individuals who believed they had been
bewitched blamed their symptoms on this rather than recognizing that it was cancer.
Sandra herself had struggled with repeated miscarriages, and so had been referred
to CECAP for screening. She believed that seeking care at the clinic was the correct
course of action when she or her children were taken ill; she claimed not to use traditional
medicine either. It is difficult to say for certain whether the observations Sandra and her
four fellows made regarding other members of their communities highlighted widespread
beliefs that were simply underrepresented in this clinical population, or if they were in
fact unconventional beliefs. Presumably, women blaming their illness symptoms on
witchcraft would be less likely to seek cervical cancer screening or other health services,
choosing instead to seek out traditional healers or practitioners.
Furthermore, while women in both CECAP and the GD were quick to tell us that
they did not themselves use traditional medicine or believe in witchcraft, both my RAs
and AL felt strongly (based on their own experiences in the hospital and as members of
the community) that at least some were simply reluctant to discuss the matter in a clinical
setting where they thought such beliefs would be disapproved of. As far back as 1985,
Alexander observed that Tanzanian cancer patients likely down-played their use of
70


traditional medicine or self-treatment when in a hospital setting in deference to the
medical personnel. An informant of mine in the CTC clinic also told me that many of
those they see for HIV use traditional medicine and that if they are asked directly the
patients will admit to using it. He attributed this to the trust the CTC clinic had built up
with their patient base and the non-judgmental attitude they try to portray. Given this, it is
likely that there were amongst our interview participants those who used traditional
medicine more frequently but were unwilling to discuss it with us.
Before I began my interviews, Dr. M and I had discussed some of the knowledge
barriers around cervical cancer. Dr. M felt that cultural concerns about how cervical
cancer screening and treatment might affect fertility were a deterrent to women coming in
for screening. Social values and concerns beyond the mere biological are often embedded
in the way certain illnesses are viewed, in the meanings ascribed to them and the risks
they are seen to pose (Setel, 1999; Sontag, 1977). In much of sub-Saharan Africa, despite
changing trends in social practices including increasing use of contraception and delaying
marriage and procreation, fertility is still highly valued and the role of mother an
important, even essential, part of womens social identity (Hollos & Larsen, 2008; ICF
International, 2004-2012; Plummer et al., 2011; Setel, 1999). Even those who do not
currently wish to become pregnant may desire to safeguard their long-term fertility and
therefore may be apprehensive about undergoing a procedure they may view as
jeopardizing this (Castle, 2003). Therefore, any threat to fertility is likely to be a cause
for concern and infertility an important reason to seek medical care, particularly if a
woman has not conceived within the first year or two of marriage (Plummer et al., 2011).
71


None of the women we interviewed in either group mentioned a fear that the
screening process itself might interfere with their fertility. In fact, it was rather the
opposite: Many of the women who came to CECAP were referred there to rule out
cervical cancer as a cause of infertility. Dr. Ms views that this concern is present in the
community or is a deterrent to screening may very well be true, but this fear was not
represented in this studys data. Yet despite this specific concern not being mentioned by
participants, threads of cultural importance placed on fertility wove through the discourse
on cervical cancer while I was in the field. As mentioned, several participants thought
infertility was associated with cervical cancer itself, and had come to determine whether
their own struggles to conceive could be due to the disease. Two women also believed
that family planning drugs, which have sometimes themselves been viewed as a threat to
long-term fertility, could cause cervical cancer. This belief has also been noted in other
studies and is concerning given the already low rates of contraceptive use in Tanzania
(Anasel & Mlinga, 2014; ICF International, 2004-2012). Finally, treatment was known to
potentially involve removal of the womb, which would of course cause infertility;
avoiding this possibility was seen as a reason to go for screening.
Although not related to screening, I also anecdotally heard from health care
providers and researchers that some women were concerned the HPV vaccine might
impact fertility. Of course, the belief that vaccines inhibit fertility or in fact cause
sterilization has a long and unfortunate history hindering immunization programs in some
developing nations (e.g., Feldman-Savelsberg, Ndonko, & Schmidt-Ehry, 2000; Renne,
2006). We did not ask questions about the vaccine in this study, but if this is indeed a
concern dissuading Tanzanian women from vaccination, it could be a quite potent one.
72


Furthermore, it could hinder the governments plans to expand preliminary vaccination
programs among schoolgirls to broader areas of the country.
Perceived Symptomology
The great promise of cervical cancer screening lies in its ability to detect the
disease in its earliest stages, when the chances for survival are highest and treatment is
relatively simple; it seeks to prevent cervical lesions from becoming invasive cancer.
However, preventive care is not standard practice for most Tanzanians. For example,
Kidanto et al. (2002) found that a majority of women in their clinic-based sample
believed hospitals could successfully treat cervical cancer while a traditional healer could
not, results replicated in this study. Yet they also felt that routine gynecological visits
were unnecessary. In Dar es Salaam, 31% of women surveyed expressed reluctance at the
idea of a preventive test when they were otherwise in good health (Kahesa et al., 2012).
Similar findings have been found in developing countries in Central and South America
(Agurto, Bishop, Sanchez, Betancourt, & Robles, 2004) and in sub-Saharan Africa
(Bingham et al., 2003); women often do not go to the clinic when they are symptom-free,
either because they do not see the need or do not have the resources for preventive care.
This is even more problematic when one considers that cervical cancer is often
asymptomatic in its early, more treatable stages.
When symptoms do occur, they often mimic other gynecological conditions.
Many Tanzanian health facilities lack the necessary laboratory equipment and staff to
perform standard of care clinical tests to rule out other STIs or reproductive tract
infections (RTIs). Instead, national guidelines call for a syndromic approach when
laboratory tests are not possible, in which the focus is first on the clinical presentation or
73


description of the patients symptoms and the provider then follows a flow chart to
determine the likely disorder and appropriate treatment (Lyimo & Beran, 2012; Ministry
of Health and Social Welfare, 2013). There are no lab test results to receive in these cases
and patients are not always given their specific diagnosis before a treatment is prescribed.
As a result, women may not perceive themselves to be at risk for cervical cancer if they
believe they are being treated for an STI or other RTI. Cervical cancer may not even
enter into their discussions with health care providers until and unless treatment for a
different condition fails and they continue to experience symptoms.
In our interviews, womens descriptions of cervical cancer symptomology were
related to and overlapped with their understanding of symptoms commonly associated
with sexual or gynecological conditions like HIV, RTIs, and STIs. When asked about the
symptoms of cervical cancer, women provided a broad range of potential indicators that
were cause for concern, including vaginal discharge, vaginal itching, bad smells in the
genital area, abdominal pain, painful intercourse, pain during menstruation or urination,
irregular bleeding, swelling or tumor, frequent urination, rashes, lack of sexual desire,
headache, troubles with fertility, diarrhea, weight loss, general poor health or weakness,
urinary or fecal incontinence, and a general feeling that things were not normal.
Of these, the three most frequently mentioned were vaginal discharge (n=10),
pain of some type (n=24, abdominal pain or stomachache being the most common
responses), and irregular bleeding (n=19), all of which are in fact potential symptoms of
cervical cancer. Women noted that a watery or brown discharge was worrisome, as well
as that with an offensive smell. Pain that potentially indicated cervical cancer included
abdominal pain or stomachache, pain during intercourse, and pain in the genital area.
74


Irregular bleeding encompassed anything from non-stop bleeding, to bleeding outside of
menses, to passing blood clots, to bleeding after menopause. Eight participants were also
sensitive to the fact that a swelling, lump, or tumor in the abdominal cavity or in/on the
genital area could possibly indicate cancer (with one additional woman making the same
observation about a lump in the breast). In one concerning interview, a woman told us
that she could feel a painful mass when she inserted a finger into her vaginal cavity,
while Hellena, a Ward Attendant, shared that some women, when they bend down
during washing their genitals, they may see something is protruding from the
genitals.. .when you insert the fingers you may feel it.5
However, participants also cited other symptoms, such as rash, itching, fever,
headache, frequent urination or pain during urination, fecal incontinence, even a mucus
discharge, that are more likely to indicate other sexually transmitted infections or
reproductive conditions such as obstetric fistula, rather than cervical cancer.6 A few
mentioned infertility as a potential symptom or outcome of cervical cancer. Though the
disease is not the most likely cause of infertility in this setting, they were correct that
these could be associated (Ombelet, Cooke, Dyer, Serour, & Devroey, 2008). In fact,
many of the women I encountered at CECAP had been referred there from the GD to rule
5 Despite Hellena sharing her anecdote in response to a question on cervical cancer
symptomology, what she was referencing was in fact likely vaginal prolapse and not a
tumor resulting from cervical cancer; however, in a setting where women often do not
seek care until their cases are quite advanced, very large tumors may in fact develop
before a woman is diagnosed with cancer. I did have occasion to see, while observing in
the clinic, one woman come into CECAP with a genital mass that had grown to the size
of a small avocado, which hung down between her legs. In that particular case though,
the mass originated on the vulva, and not internally on the cervix.
6 Though urinary and fecal incontinence can occur in very advanced stages of cervical
cancer, particularly if a fistula develops as a result of the cancer.
75


out cervical cancer as a reason for their difficulty conceiving. Weakness, weight loss, and
a feeling that things were not normal are fairly vague and could be true of a number of
diseases; in this context the first two were frequently tied to HIV as well. Despite their
potential ambiguity, women viewed all of these as symptoms to take seriously, and
certainly reasons to seek medical care.
That cervical cancer was thought to share symptoms with these other
conditionsRTIs/STIs, HIV, and to a lesser degree, fistula or other birth
complicationstells us something about where women placed cervical cancer within the
constellation of health concerns. Cervical cancer, in terms of causes, risks, and
symptomology, was largely discussed as an issue of sexual and reproductive health, tied
to other conditions that affected that area of the body, rather than in relation to other
cancers, their causes, risk factors, and symptoms (the exception being the mention of
tumors and tumor growth).
Explanations of Risk
Health and illness must be examined with an eye not only towards the
physiological basis of the disease, but also towards the larger social and cultural context
in which it exists and the meanings assigned to it (Douglas & Wildavsky, 1982; Geertz,
1985; Jordan, 1983). I noted earlier that in regard to cervical cancer, womens lay models
included conceptualizations of disease causation and risk that were bound together. These
models also reflected concerns with larger social processes. Medical anthropologists and
other scholars have long pointed to the usefulness of metaphor and meaning-making in
describing and understanding illness (Douglas, 1970; Farmer & Kleinman, 1989;
Scheper-Hughes & Lock, 1987; Sontag, 1977). Here I will explore how women called on
76


what they knew about infections, both sexually transmitted and otherwise, to explain the
risk of contracting or developing cervical cancer and in so doing, illuminate larger
cultural concerns around sexuality and childbirth.
Sexual Behavior
In Tanzania, the HIV pandemic and high rates of STIs have contributed to the
conceptualization of certain sexual behaviors as risky (Hanson, 2000; Outwater et al.,
2001; Setel, 1999). This is reinforced by the information people have heard repeatedly in
public health campaigns and from medical personnel about the risks of unprotected sex
and of having multiple partners. Given that a large proportion of CECAPs clients were
HIV positive (31.4% during the period from May 2010-October 2013), and that many
had been referred from CTC where they did health education on the disease, I expected to
find that many in the CECAP group would be familiar with the risks posed by sexual
activity. But in fact, most of the women we spoke to, in both groups of participants,
echoed these concerns. As Hanson has noted, despite the fact that STI public education
efforts in Tanzania have often concentrated on broader national media campaigns rather
than on more focused local education programs, impressive levels of knowledge of
AIDS and its transmission have been reached (2000, p. 344).
The most risky activity, according to the women we interviewed, was having
multiple sexual partners. Although we were told this specifically in regards to questions
about cervical cancer risk, the repetition of this theme during discussions about protecting
ones health generally suggests that it was also seen as risky to an individuals well-being
in a broader sense. Repeatedly, women told my RAs and me that those having many
lovers and those changing men from time to time, were at risk for cervical cancer.
77


Having multiple partners was a well-known risk factor for HIV and other STIs, and
Shazia for one, a Community Health Worker (CHW), told us she had actually heard on
the radio that cervical cancer resulted from STIs. However, in most cases, it was less
clear whether participants knew that this was specifically a risk factor for cervical cancer
or rather were making an intuitive leap that non-monogamy was a risk factor for some
sexually transmitted infections, and since cervical cancer impacted the reproductive
system, it likely was a risk factor for that disease as well. Some actually verbalized this
thought, including Anna, who said, I dont know. When you picture HIV, we know it
can be transmitted from one person to the other. I havent known of the cervical, if that is
the case or not.
Of course, as Setel (1999) points out, more important to the spread of HIV and
other STIs is unprotected sex with multiple partners. While in reality condoms are not as
effective in preventing the transmission of HPV as they are with other STIs, they still
provide some protection (Castellsague, 2008; Centers for Disease Control, 2007). Along
these lines, women correctly believed that unsafe or unprotected sex and prostitution
could increase a womans risk of cervical cancer. As Dinah, a young woman of 25 who
had had seven partners in as many years (the second highest of any in either CECAP or
GD group) but was now married warned us, nowadays there is a Swahili saying, which
goes, umdhaniaye stye kumbe ndiye \never judge by looks]. Rather than HPV or a
specific pathogen, women instead focused on these behaviors themselves as both a risk
factor and a cause of cervical cancer. Abortion and rape were only mentioned by two
participants and one participant respectively, and were not typically part of discussions
about risk.
78


Participants were particularly concerned about sexual activity as a young woman.
This concern was not necessarily tied to a belief in a higher physiological risk of
infection but rather due to a period of time when specific behaviors or a specific type of
relationship were common. Anna explained, You know, we are ladies, or some of us are
still young ladies, and there are those who still want to enjoy life, but after some time has
gone, they may decide that it is enough of the starehe and yet be able to protect their
health. As used in this context, the term starehe can have two meanings, both of which
were implied by different participants at different times. The first is its more literal
meaning, which is to enjoy oneself or to live luxuriously, the implication when used to
refer to sexual relationships being that young women more interested in (or in need of)
material things may seek out relationships with wealthier men that involve gift giving or
transactional sex (Haram, 2005; Wamoyi, Wight, Plummer, Mshana, & Ross, 2010).
Starehe can also refer to casual, short-term (sometimes one-time) affairs that may
comprise a young persons initial foray into a sexual relationship (Setel, 1999).
According to Setel, who studied these types of relationships in northern Tanzania, starehe
relationships are typically brief and sexual activity is the focus. He notes that young
people would typically differentiate between a starehe partner and a boyfriend or
girlfriend, with whom they had a relationship outside of sexual relations, though others
have questioned the strict typology he uses to define this category of relationships
(Wamoyi et al., 2010). Although a starehe relationship can develop into a long-term
relationship, desire is a more important component at its initiation than tabia (good
character) (Setel, 1999).
79


In either usage of starehe, fidelity is not presumed, particularly on the part of the
man (Haram, 2005; Setel, 1999). These types of relationships therefore pose risks both to
the physical body and to a womans reputation. Young women, according to Marjani,
were at risk because we like doing sex and might get infections like gonorrhea and
syphilis that would contribute to the disease, while Jesca shared how her mother had
warned her against her wild behavior when she was younger (she was 30 when
interviewed). Her mother told her that her drinking in the streets would result in her
getting men who have problems, and blamed these youthful indiscretions for Jescas
current health struggles. Sexuality and fidelity expectations are highly gendered in
Tanzania, and anxiety around changing social norms related to sexual behavior and
gender roles has been noted repeatedly, particularly when it comes to young women
(Dilger, 2003; Haram, 2005; Setel, 1999). And regarding cervical cancer risk, there is
reason to be concerned. While both male and female Tanzanian youth tend to experience
their first sexual encounter at approximately the same age, females are more likely to
have this experience with a significantly older partner, which is a risk factor for the
disease because an older partner typically has had a number of previous partners (Urassa,
Moshiro, Chalamilla, Mhalu, & Sandstrom, 2008).
Childbirth
Given that women knew cervical cancer to be a womans disease, specifically a
disease that affected the reproductive system, it is perhaps not surprising that their
notions of risk centered on activities tied to their reproductive organs: sexual activity, as
outlined above, and childbirth. Yes, sexual activity could introduce infection into the
body, but so too could internal wounds resulting from childbirth or other reproductive
80


medical care. As an opening to the body, the vaginal canal was an entry point for disease
during both sexual activity, when a mans sexual organ or his sperm could leave behind
disease, as well as childbirth.
In reality, childbirth is a dangerous process for many women in Tanzania,
particularly rural women. In a nation where barely 50% of women give birth in a health
facility, maternal mortality is high and birth complications frequent.7 High parity is a
biomedical risk factor for cervical cancer, and in Tanzania, where the total fertility rate is
5.4 (ICF International, 2004-2012) and cervical cancer incidence is likewise high among
young and middle-aged women specifically, the perception that having many children
can lead to cervical cancer is both understandable and true. While most women could not
explain the actual mechanism through which this risk materialized, giving birth
frequently or many times during ones life were particularly thought to increase the
likelihood a woman would develop cervical cancer. Debora, who had given birth to two
children herself, gave a response that was illustrative of this tension between belief and
lack of specific knowledge: From my view, I think cervical cancer is caused by giving
birth.. .1 dont know if this cancer is caused by having very many children or very close
pregnancies, but I think if a person has already given birth they can get cervical cancer.
How one became pregnant, for instance, if one conceived within marriage or without, or
if one conceived several children from ones primary partner versus from multiple
partners, was not a part of the risk discussion.
7 Maternal mortality is 454 out of every 100,000 live births (ICF International, 2004-
2012).
81


Women across the age and parity spectrum of those we interviewed, from both
CECAP and the GD, reiterated the idea that frequent or numerous births were risky.
However, though a biomedical understanding of cervical cancer tells us that it begins
with HPV infection of the cervix, the organ of most concern to participants here was in
fact the womb or uterus,8 (though this may be because women lacked an understanding of
the cervix as a separate entity; even the Kiswahili name for the cervix, which translates to
entrance to the womb refers to it in terms of its relationship to the uterus). The actual
process of childbirth, women shared, could cause internal wounds or lacerations through
which microorganisms could enter and cause infection. Likewise, metal tools used during
delivery, abortions, or miscarriages might scrape the internal organs,9 making them more
susceptible to disease. These wounds or scratches provided a vulnerable point of entry to
the body. Another concern was that the uterus itself could become loose or tired
through frequent or numerous deliveries, a condition that, again, made that organ
susceptible to disease, including cervical cancer.
Fears of lacerations or wounds during labor, taken in conjunction with some of
the symptoms thought to be associated with cervical cancer in the previous section (such
as urinary and fecal incontinence) may be related to womens experiences with obstetric
fistula10 and other birth complications. Approximately half of all women in Tanzania give
8 The uterus was often referred to as the mfuko wa uzazi (reproductive bag) or simply
mfuko (bag), rather than the more formal uterasi.
9 This may in fact relate to a concern voiced by a few women regarding the use of the
speculum in conducting cervical cancer screenings discussed in a later chapter, though no
participant specifically stated they personally were afraid of lacerations from the device.
10 Obstetric fistula often occurs during prolonged labor, when the babys head presses for
too long against the mothers sensitive pelvic tissues. The lapse in blood supply caused
by this pressure can cause necrosis and a hole may form either between the vaginal canal
82


birth at home or outside of a health center, a situation due in part to the long distances
and poor roads women must travel in rural areas to reach health care. Those who do reach
health centers often find they do not receive adequate care (Mselle et al., 2013). As a
result, prolonged labor and complications may arise and obstetric fistula result. Marjani
told us a story of how she had been warned about this during a previous cervical cancer
screening experience (other types of health information are sometimes included in
CECAP information sessions, particularly if a woman asks a question about a specific
topic or symptom), and she appeared to have integrated this information with what she
had learned about cervical cancer:
What causes cervical cancer?
I dont know, but when I came for screening in 2011 we were given a
lesson that having many sexual partners can also lead to its transmission,
abortions, and home deliveries. They told us many things, that you may
deliver at home using a lot of energy and get a laceration without
knowing, and so continue staying at home. This can also cause some
problems.
Women utilized their knowledge and experiences of birth complications to
construct an explanatory lay model of cervical cancer risk that included childbirth. These
models accessed concepts and language used in biomedical explanations of infection, in
addition to more cultural explanations of ill health, such as the susceptibility of tired or
injured wombs. No doubt at least some of these women had heard that high parity was a
risk factor for cervical cancer from a third party, whether a poster, radio announcement,
or provider. But they interpreted this basic piece of information through the lens of a
and the bladder, or the vaginal canal and the rectum (Mselle, Moland, Mvungi, Evjen-
Olsen, & Kohi, 2013).
83


reality in which childbirth is, despite its common occurrence, a dangerous journey for
many women.
Prevention and Prognosis
In keeping with previous research in Tanzania (Kidanto et al., 2002; Lyimo &
Beran, 2012; Peters et al., 2010), the women we spoke with did not have deep or specific
knowledge of cervical cancer as a disease, but many understood that it could be sexually
transmitted and that certain sexual behaviors and childbirth played a role in risk for the
disease. However, they struggled to name specific tactics to prevent cervical cancer,
perhaps because the primary risk factors they perceivedsexual activity and childbirth
were such a common part of daily life. Shany, a young woman in the GD, for example,
knew some of the risk factors, but really believed all women were at risk:
Any woman can get the disease because they have many sexual partners or
give birth to many children, maybe more than five times. On a quick look,
every woman would like to have children in their lifetime, as well as to
have a sexual partner, and so women are prone to get the disease.
Gregg found such attitudes in the Brazilian favela as well, where women felt that sexual
activity was a given, and not a risk for which anyone could actually assume
responsibility (2003, p. 96). In that case, women had limited agency in their sexual
relationships and so chose instead to view the Pap smear, or more specifically, the lack of
one, as the factor that increased their risk for cervical cancer.
At St. Francis, the women we spoke with could be divided into two types
regarding this issue. The first, made up of thirteen women, simply didnt know how it
could be prevented under present conditions. Debora confessed, I dont know if it can be
prevented or not because people are still giving birth. Hellena, while pointing out the
importance of the education she does in schools as a Ward Attendant, acknowledged that
84


Full Text

PAGE 1

CANCER AND CERVICAL CANCER SCREENING IN RURAL TANZANIA by EMILY E. CHASCO B.A., University of Michigan, 2004 M.A., University of Colorado Denver, 2009 A thesis submitted to the Faculty o f the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Health and Behavioral Sciences 2015

PAGE 2

2015 EMILY E. CHASCO ALL RIGHTS RESERVED

PAGE 3

ii This thesis for th e Doctor of Philosophy degree by Emily E. Chasco has been approved for the Health and Behavioral Sciences Program by Jean Scandlyn, Chair Sara Yeatman, Advisor John Brett Deborah Thomas July 23 2015

PAGE 4

iii Chasco, Emily E. (Ph.D., Health and Behavioral Sciences) Cancer Screening in Tanzania Thesis directed by Assistant Professor Sara Yeatman ABSTRACT This dissertation explores cervical cancer and cervical cancer screening wit hin the context of a rural health care setting in Tanzania. In particular, it is concerned with how cervical cancer is perceived and understood by Tanzanian women, as well as how these perceptions influence attitudes towards cervical cancer screening and t reatment. Cervical the resources needed to adequately address the magnitude of the problem are scare. This is slowly changing as cervical cancer prevention programs are implemented in these regions, often by international non governmental organizations alone or in partnership with local government and other agencies. These programs have frequently struggled with low utilization and access barriers, even where issues of co st have been addressed. Based on six months of ethnographic fieldwork conducted largely within the established cervical cancer prevention program of a rural referral hospital, this research takes a layered view that examines the historical, social, and pol itical environment in which cervical cancer came to be seen as a health prior ity in Tanzania, as well as current efforts for cancer control. It seeks to understand first how Tanzanian women think about cervical cancer and cervical cancer screening, but als o examines issues affecting utilization related to the broader context of the cervical cancer prevention program and its place within the Tanzanian health care system This dissertation adds to the small but growing

PAGE 5

iv body of literature that seeks to better comprehend prevention program utilization with the goal of increasing the adoption and expansion of screening programs in developing countries. The form and content of this abstract are approved. I recommend its publication. Approved: Sara Yeatman

PAGE 6

v DEDIC ATION This work is dedicated to my parents, David and Beth Chasco, whose willingness to pack up three children under the age of ten and travel through Europe with only two suitcases for four months quite literally made me the person I am today. Upon return ing home, when their f ifth grader announced she planned to be an anthropologist someday they responded not with dismissal or amusement, but encouragement. Most importantly, their love and support has been without limits over the years.

PAGE 7

vi ACKNOWLEDGEMENT S Dissertations are rarely completed without a great deal of support, advice, and encouragement, and in my case I was fortunate to have all three in abundance. I would first like to express my immense gratitude to the Tanzanian women who participated in th is research project. They frequently sat down to our interviews with a long wait for health services either just behind or in front of them, yet often their sense of humor I received more than one teasing offer to arrange my marriage to a brother or cousin made the time pass by quickly. Their willingness to answer our many questions with patience and grace made this project what it is. I also wish to thank St. Francis Hospital and St. Francis University College of Health and All ied Sciences (SFUCHAS) for their cooperation and support. From my first visit with Prof. Cassian Magor i, the Principal of SFUCHAS he welcomed me with open arms, smoothing the way forward, orienting me to the various hospital departments a nd making introductions. The St. Francis ad ministration was instrumental in assisting me with the IRB approval process, as was my contact person at the National Institute for Medical Research in Dar es Salaam. The staff overall were incredibly welcoming and, particularly in RCH where space for inte rviews was limited, extremely accommodating. I owe a huge debt to my friends and mentors both in the Obstetrics/Gynecology Department and CECAP The head of the department was a supporter from the start, throwing me into the deep end on my first visit to S t. Francis by encouraging me to come and observ e screenings that same day The staff members of CECAP were inspiring to watch. It could not have been easy to have an mzungu under their feet all day, but they never ma de me feel an inconvenience. T hey const antly challenged my assumptions and

PAGE 8

vii pushed me to consider new angles. My contact at Jhpiego was also very generous with her time and our conversations were thought provoking. A special thank you goes to my Research A ssista nts, Paula Madata and Rachel Bwok i When a friend said he had a potential RA for me, and when that RA mentioned she was bringing alo ng her friend to work with me too, I had no idea what to expect. But they quickly proved themselves up to the task and became not only invaluable to the proc ess, but friends and guides in navigating day to day life in Tanzania Finding your way in a foreign country can be daunting, so on a personal level I am so grateful to have had my ELCT Language School family (both Tanzanians and wazungu ) in Morogoro, Tanz ania, and my wonderful host family in Ifakara to guide me My parents, sisters Erica and Claire, and friends (in particular Bethany, Alisa, Tony, and Piya) offered incredible love and care from afar and saved my sanity during the writing process when I arr ived back in the States Over the many yea rs I have spent at the University of Colorado Denver while working on my M.A. and Ph.D. my committee members and mentors both in the Health and Behavioral Sciences program and in Anthropology have been incredibly generous with guidan ce, advice, patience, and oftentimes much needed nudges. I have had wonderful opportunities to travel and grow as a researcher whil e working with incredible people In particular, I want to thank Sara Yeatman, Jean Scandlyn, John Bret t, Debo rah Thomas, and Charles Musiba, who pushed me to go to Tanzania in the first place. When my project nearly fell through, when I encountered road blocks in the field, their encouragement and support meant a great deal. For their much needed guidance through the processes of graduate school and for offering moral support, I also want to thank the

PAGE 9

v iii administrative staff at UCD with whom I have worked closely over the years Abby Fitch and Connie Turner. Finally, I would like to express my gratitude to the National Security Education Program (Boren Fellowship), Rotary International, the William Robinson/Peter Durst Scholarship, and the Department of Health and Behavioral Sciences for the funding that made this research possible.

PAGE 10

ix TABLE OF CONTENTS CH APTER I. INTRODUCTION ................................ ................................ ............................... 1 Research Questions ................................ ................................ .............................. 3 Theoretical Approach ................................ ................................ ........................... 3 Organization of the Dissertation ................................ ................................ .......... 5 II. SETTING THE SCENE ................................ ................................ ...................... 8 Morogoro City and Region ................................ ................................ ................ 10 Ifakara and the Surrounding Area ................................ ................................ ...... 11 The Local Economy ................................ ................................ .................. 12 Population ................................ ................................ ................................ 13 Ifakara as a Field Site ................................ ................................ ......................... 14 A Center of Health Research ................................ ................................ .... 14 St. Francis Hospital ................................ ................................ ................... 15 Cervical Cancer Prevention Program ................................ ................................ 16 Location ................................ ................................ ................................ .... 16 Screening Services ................................ ................................ .................... 17 III. METHODS & ANALYSIS ................................ ................................ ............... 22 Methodological Approach ................................ ................................ ................. 23 Data Collection ................................ ................................ ................................ .. 27 Participant Observation/Observation ................................ ........................ 27 Participant Interviews ................................ ................................ ............... 28 Patient Information Forms ................................ ................................ ........ 33 Key Informants and Specialized Informants ................................ ............. 34 Sc reening Map Forms ................................ ................................ ............... 35 Supplementary Information ................................ ................................ ...... 37 Analysis ................................ ................................ ................................ .............. 38 Qu alitative Data ................................ ................................ ........................ 38 Quantitative Data ................................ ................................ ...................... 40 IV. CHANGING PRIORITIES ................................ ................................ ................ 44

PAGE 11

x A Global Picture ................................ ................................ ................................ 45 Health Transitions ................................ ................................ ..................... 48 The Macro Policy Environment ................................ ............................... 50 Transferable Technologies ................................ ................................ ........ 52 Cervical Cancer in Tanzania ................................ ................................ .............. 55 NGO and Government Partnerships ................................ ................................ .. 57 V. LOCAL CONSTRUCTIONS OF CERVICAL CANCER ................................ 61 Understanding Cervical Cancer in the Local Context ................................ ....... 62 Accessing Familiar Biomedical Models of Infection ................................ ........ 65 Infections, Microbes, and other Causes of Cervical Cancer ..................... 67 Perceiv ed Symptomology ................................ ................................ ......... 73 Explanations of Risk ................................ ................................ .......................... 76 Sexual Behavior ................................ ................................ ........................ 77 Ch ildbirth ................................ ................................ ................................ .. 80 Prevention and Prognosis ................................ ................................ ................... 84 HIV, Cervical Cancer and Ugonjwa wa Kisasa ................................ ................. 90 VI. UTILIZING SCREENING SERVICES ................................ ............................ 95 Cervical Cancer: A Disease of Poverty ................................ ............................. 97 The Benefits of Screening ................................ ................................ .................. 99 Barriers to Screening ................................ ................................ ........................ 103 Barriers Coming to the Hospital ................................ ............................. 104 Barrie rs at the Hospital ................................ ................................ ........... 109 Overcoming Barriers ................................ ................................ ........................ 113 VII. STIGMA, SHARING, AND SUPPORT ................................ ......................... 117 Cervical Cancer and Stigma ................................ ................................ ............. 119 Sharing the Screening Decision ................................ ................................ ....... 122 The Role of Husbands ................................ ................................ ............. 124 Relatives, Neighbors, and Fellows ................................ .......................... 127 Sharing a Cervical Cancer Diagnosis ................................ .............................. 132 Fear of Stigma or Rejection ................................ ................................ .... 132 Marshaling Resources ................................ ................................ ............. 135

PAGE 12

xi A Heart of Sharing ................................ ................................ .................. 140 Negotiating Stigma and Support ................................ ................................ ...... 142 VIII. COMPETING PRIORITIES ................................ ................................ ............ 146 ................................ ................................ ....................... 148 Low Utilization ................................ ................................ ....................... 148 A Program in Transition ................................ ................................ ......... 152 On the Ground ................................ ................................ ................................ .. 153 Hospital Context ................................ ................................ .............................. 157 Larger Program Goals ................................ ................................ ...................... 160 Shared Goal, Different Priorities ................................ ................................ ..... 162 IX. CONCLUSION ................................ ................................ ................................ 166 Limitations ................................ ................................ ................................ ....... 170 Future Recommendations ................................ ................................ ................ 172 Moving Forward ................................ ................................ .............................. 175 REFERENCES ................................ ................................ ................................ ............... 178 APPENDIX 197 A. LIST OF PARTICIPANTS ................................ ................................ .............. 197 B. PARTICIPANT INTERVIE W GUIDES ................................ ......................... 201 C. PARTICIPANT INFORMATION FORM ................................ ...................... 205 D. SCREENING MAP FORM ................................ ................................ ............. 206

PAGE 13

xii LIST OF TABLES TABLE 1. Comparison of CECAP and GD Interview Groups ................................ .............. 33 2. Specialized Informants ................................ ................................ .......................... 35 3. Final Screening Map Form Variables ................................ ................................ ... 36 4. Inter view Participants' Occupations ................................ ................................ .... 108 5. Clients Per Visit Type Per Year ................................ ................................ .......... 152

PAGE 14

xiii LIST OF FIGURES FIGURE 1. Map of Tanzania ................................ ................................ ................................ ..... 9 2. The CECAP Room ................................ ................................ ................................ 18 3. Screening Supplies ................................ ................................ ................................ 19 4. Parity of CECAP Clients, October 2010 October 2013 ................................ ..... 107 5. Number of Clients Screened Per Month in Clinic, May 2010 October 2013 ..... 150 6. Total Number of Clients Screened Per Month, May 2010 October 2013 .......... 151 7. Cervical Cancer Posters ................................ ................................ ...................... 158

PAGE 15

xiv LIST OF A BBREVIATIONS ART CECAP CHW CTC GAVI GD HBM HPV IHI LEEP MAISHA MOHS W NGO NIMR PIF RA RCH (MCH) RTI SFUCHAS SMF STI USAID VIA Antiretroviral Therapy Cervical Cancer Prevention Program Community Health Worker Care and Treatment Clinic Global Alliance for Vaccines and Immunizations Gynecology Department Health Belief Model Human Papillomavirus Ifakara Health Center Loop Electrosurgical Excision Procedure Mothers and Infants, Safe, Healthy and Alive Ministry of Health and Social Welfare Non Governmental Organization National Institute for Medical Research Patient Informatio n Form Research Assistant Reproductive and Child Health (Maternal and Child Health) Reproductive Tract Infection St. Francis University College of Health and Allied Sciences Screening Map Forms Sexually Transmitted Infections United States Agency for Inter national Development Visual Inspection with Acetic Acid

PAGE 16

xv NOTE TO THE READER During the course of my fieldwork, I interviewed women who had come to St. Francis Hospital, women I refer to as research participants. In this dissertation, when discussing my encounters with these women or reproducing their words, I refer to them using pseudonyms in order to protect their confidentiality. Some details, such as references to specific doctors they have seen or villages they live in have been left out. Brief desc riptions of these women, along with their pseudonyms, can be found in Appendix A. I also interviewed specialized informants and came into contact with other hospital personnel The informants I interviewed were often less concerned with confidentiality, th ough we discussed it during the consent process, and were in several cases eager for their words to be heard. St. Francis is a relatively small hospital; it would not be too difficult for a resolute individual to determine who the informants I discuss a re. Nonetheless, I have chosen to refer to all informants and hospital personnel using initials (not their real initials) here in the dissertation to provide a degree of confidentiality

PAGE 17

1 CHAPTER I INTRODUCTION Cervical cancer, public health experts and propag anda leaflets repeat, is a fully preventable disease There is, however, a big difference between preventability hypothetical variable and the construction of efficient prevention programmes. In the twentieth century medical sciences devised efficient wa ys to reduce the number of women who suffer from advanced cervical malignancies. The persistence of the disease in many parts of the world reflects to an important extent a failure to act upon economic and social conditions that hamper the diffusion of pre ventive measures. ( L wy, 2011, p. 173 ) Cervical cancer is a major issue for women in the developing world, despite a dramatic reduct ion in incidence and mortality in developed nations over the course of the 20 th century. This disparity is largely due to a discrepancy in access to and utilization of screening programs in low and middle income countries. The international community has increasingly recognized the importance of cervical cancer screening in these regions and the expansion of resources targeted towards the disease, however, services, when they are available, are often under utilized ( Agurto, Sandoval, Rosa, & Guardado, 2006 ; Tsu, Jeronimo, & Anderson, 2013 ) As stakeholders have implemented cervical cancer screening programs in settings with few resources and high incidence of the disease, interest in understanding ho w to increase utilization of the se programs has likewise increased. The cervical cancer burden is especially severe in sub Saharan Africa, where it is the most common femal e cancer in many countries. The incidence rate for the region is one of the highest of any in the world, at 34.8 per 100,000, with 93,225 new cases in 2012, the most recent year for which reliable statistics are available ( Castellsagu et al., 2012 ; Ferlay et al., 2013 ) Cervical cancer is the single largest contributor to years of life

PAGE 18

2 lost from cancer ther e ( Yang, Bray, Parkin, Sellors, & Zhang, 2004 ) This picture is daunting, an d made further so by the fact that the disease disproportionately affects economic security. Yet the majority of women in this region live in rural areas where screening and treatment services may be unavailable or difficult to access ( Anorlu, 2008 ; Castellsagu et al., 2012 ) This dissertation examines the issues of cervical cancer and cervical cancer screening in the context of a rural health care setting in the sub Saharan African country of Tanzania. Tanzania has in recent years partnered with the international non profi t health organization Jhpiego to organize cervical cancer prevention programs at various sites around the country; this dissertation focuses on a program located at a Referral Hospital in the town of Ifakara. The initiative was begun with the idea that aft er a set programs would transfer to the Tanzanian government. As with screening services in many developing nations, low utilization of the program has been an issue since its inception and was a source of tension between the program staff and Jhpiego. Despite attempts to mitigate this low utilization, the program has been unable to raise their number of clients served in any consistent way. Some of the issues that influence p oor utilization in developing nations have been well documented and are widespread, particularly critical issues with health infrastructure, transportation infrastructure, and cost. Solving these remains a large and important undertaking. However, these is sues do not account for the picture entirely. Only more recently have researchers begun to examine how local knowledge and

PAGE 19

3 perceptions of the disease promote or inhibit health seeking behaviors among women in populations who may be targeted for screening. Research into how cervical cancer is culturally situated within broader health issues, how women think about and give meaning to the disease, and sociocultural factors that might influence the decision to undergo cervical cancer screening have been largely understudied, particularly in Tanzania and other sub Saharan African nations. This research seeks to address this gap by exploring these issues in relationship to the larger background of cervical cancer policy in Tanzania. Research Questions This project was guided by a primary research question: How do perceptions of cervical cancer among Tanzanian women influence their uptake of cervical cancer screening services, and what implications does this have for the adoption and expansion of cervical cancer scr eening programs in Tanzania? In order to address this overarching question, four sub questions were developed to guide the research. (RQ1) What is the population currently attending screening programs and who is missing from this population? (RQ2) How is c ervical cancer understood and given meaning among Tanzanian women seeking care? (RQ3) What influence do cultural constructions of cervical cancer have on (RQ4) How does the screening program operate on a daily basis, and how does it fit within the overall structure of the hospital? Theoretical Approach This dissertation is organized in a somewhat non traditional manner, in that there is no one individual chapter devoted to an examination of the relationsh ip between theory and the subject at hand. T heory is instead incorporated throughout the dissertation chapters where it is most appropriate and relevant to specific themes. However, here I

PAGE 20

4 would take a moment to outline briefly the general approach guiding my research. In seeking to understand cervical cancer screening in Tanzania, I have been influenced by previous scholarship in critical medical anthropology that has examined health with an eye towards larger power relationships ( H. A. Baer, Singer, & Johnsen, 1986 ; Farmer, 1999 ; Singer, 1989 1998 ; Turshen, 1998 ) Traditional anthropological scholarship has been criticized for focusing too much on the micro level, and in the case of medical anthropology, thi s has often meant on the individual/patient or the patient physician interaction ( Dressler, 2001 ; Singer, 1989 ) I take a political economy of health approach, a perspective which recognizes that economic, political, and historical processes influence how disease spreads and who is affected ( Doyal, 1995 ; Farmer, 1999 ; Singer, 1989 1998 ) This approach emphasizes multiple level s of influence and interaction, from the macro to the micro, and is therefore particularly useful in examining issues tied to globalization and health ( H. Baer, Singer, & Susser, 2003 ) Yet at the same time, I do not abandon more interpretive approaches to medical anthropology completely. Discerning how local populations conceptualize and assign meaning to disease and how individuals understand their own illness experien ces are important. This is not simply an academic exercise; it has significant implications for health seeking behavior ( Kleinman, Eisenberg, & Good, 2006 ; Nichter, 2008 ) 0) work, which attempted to tie together more structural approaches with argument for a synthesis of these p erspectives were therefore influential. In practice this meant that while I was concerned with local perspectives on the disease of cervical cancer and how the se might inf luence screening utilization, I also sought to understand

PAGE 21

5 how structural forces and power relationships both influenced these phenomena and were influenced by them. Organization of the Dissertation In Chapter II I begin by providing background on how I came to do my research in Ifakara, and introduce the field setting, including the tow n and its surround ing environs. A description and brief history of St. Francis Hospital, the specific facility at which I made most of my observations is included here Finally, I introduce the cervical cancer prevention program, including its setting with in the hospital, and the services provided. In Chapter III I outline the methodological approach that guided this research, providing a summary of the data collection procedures and analysis. Here I also describe the specialized informants and the partici pants from the cervical cancer prevention program and the G ynecology D epartment who were so generous with their time and whose interviews were a key part of the qualitative data collected for this project. In Chapter IV I focus on the macro policy environ ment in which cervical cancer became a health concern at the national level in Tanzania, corresponding to larger changes in global health priorities and policy. This shift then resulted in increased interest and funding from foreign governments, non govern mental organizations (NGOs) and the Tanzanian government to implement screening programs, as well as an increase in media attention. I also tie this to the efforts made at the local level in Ifakara to implement screening and increase awareness. In Chapte r V I move to an exploration of local knowledge of cervical cancer, including its causes, potential risk factors, and prognosis, among the St. Francis patients I

PAGE 22

6 interviewed in the course of my fieldwork. I explore how the disease is perceived by intervie w participants, and how these perceptions relate to familiar but general biomedical explanations of infection and immunity, in particular those related to HIV, STIs, and other sexual or reproductive health issues. I argue that these beliefs have implicatio ns both for patient education and outreach, as well as more mid level concerns, such as how the cervical cancer screening program is situated within the overall hospital setting. I n Chapter VI I summarize findings on how participants understand benefits a nd barriers to screening and treatment at the cervical cancer screening program in Ifakara, as well as for more extensive treatment in Morogoro or Dar es Salaam. I review the barriers in light of a fairly extensive literature on challenges to screening in developing nations that nonetheless is thin on regionally specific data in East Africa. Understanding these challenges can provide insight into how to improve services, particularly for programs that have already attempted to address at least in part, the cost issue In Chapter VII I turn my attention first to the notion of stigma, both as it relates to cervical cancer and cervical cancer screening, before examining whether or not women choose to discuss screening behavior or a positive diagnosis with oth ers. Although privacy is an important cultural value in this regard, it must be weighed against the need to access the types of resources and support required to follow through on a decision to seek health care services. Furthermore, some women expressed a strong desire to be a source of information and support to other women, which influenced their decision to share their own experiences with this aspect of their health. In Cha pter VIII I return to more mid and macro level concerns by examining

PAGE 23

7 som e of t he challenges facing the cervical cancer prevention program and the relationships among the different stakehol ders I encountered while conducting this research at St. Francis Although these stakeholders all ostensibly desire the same goal to improve women in reality their priorities were sometimes at odds with one another. These tensions reflect larger debates in the global health literature, such as the role of NGOs working within health care systems in developing cou ntries and the prioritization of vertical versus horizontal health care structures. Although I did not go into this project intending to study these issues, it was impossible to escape their discussion and influence while in the field. Finally I con clude with a brief discussion of some of the overall themes of the dissertation, as well as limitations and several areas for future research

PAGE 24

8 CHAPTER II SETTING THE SCENE I f irst became interested in conducting my research into cervical cancer at St. Franci s Hospital in Ifakara, Tanzania through mentors at the University of Colorado Denver. I had originally intended to study cervical cancer screening in northern Tanzania, in the city of Mwanza, where I had made my first visit to the country in 2010. However, the hospital there was located in a major urban center and was one of the largest in the nation. I was interested, but I knew that the urban population and wider screening and treatment procedures available there were unusual when compared to the health f acilities in most of Tanzania. Additionally, their screening program was already in development and plans were in place to conduct their own evaluation in house. As I began to search for another opportunity, preferably in a more rural location, I was invit ed to visit St. Francis Hospital and obse colleague s, who worked at the facility. In June 2011, I visited Ifakara for two weeks. During that time, I was able to speak with faculty and personnel at St. Franci s, observe the activities of the screening program, and become acquainted with Ifakara. St. Francis was an interesting and ideal choice for this study for several reasons. It served a primarily rural and semi urban population, reflecting the circumstances of the majority of Tanzanian women, and its population was likely to face similar challenges as those in other rural areas. As a Referral Hospital (though until recently only a District Hospital), it was better equipped than most dispensaries and small hea lth clinics in rural areas; however, women were still referred to larger health centers for any positive diagnosis beyond a small cervical lesion

PAGE 25

9 treatable with cryotherapy. This, I felt, would give me the opportunity to study not only the one visit screen and treat approach practiced by the screening program (discussed further in Chapter IV ), but also the referral process for more advanced cases and the program is similar to many others in Tanzania, in that it was initiated largely through the work of an international NGO working within the Tanzanian health care system. I returned in May 2013 to begin my fieldwork, following two and a half months of Kiswahili study in the city of Morogoro. In this chapter, I begin by describing the region in which both Ifakara and Morogoro are located, also called Morogoro. I then focus the lens on Ifakara and St. Francis Hospital before introducing the cervical cancer prevention program it self. ( Source: Free Vector Maps ) Figure 1 : Map of Tanzania

PAGE 26

10 Morogoro City and Region The Morogoro region, located in southeastern Tanzania, is divided into seven districts. Near the northern end, its major urban hub, the city of Morogoro, makes up the smallest district by area (referred to as the Morogoro Urban District so as to differentiate it from the adjacent Morogoro Rural District) and is connected by tarmac highway to the commercial capital of Dar es Salaam, 194 km away ( National Bureau of Statistics 2006 ) Because of its proximity to Dar es Salaam and location on the major east west highway, Morogoro is a busy thoroughfare for those traveling to inte rior cities and towns, including the political capital Dodoma, Arusha, and Iringa, as well as for safari tourists on their way to the popular Mikumi National Park. Of the approximately 2.2 million people living in the region, 315,866 call the municipal dis trict of Morogoro home, though not all live in the town proper ( National Bur eau of Statistics 2013 ) The overall region has had an average population growth rate of 2.4% per year, which is close to the national average of 2.7%, and is one of only two regions in the country outside of Dar es Salaam to have a level of urbanizatio n higher than that of the nation as a whole ( Ministry of Planning Economy, and Empowerment 2007 ; National Bureau of Statistics 2013 ) Ta nzania has seen a trend of rural urban and urban urban migration, often tied to employment opportunities. Morogoro Region has experienced both trends, with those from rural areas moving to the city of Morogoro and those in the city migrating to other urban areas outside of the region ( Ministry of Planning Economy, and Empowerment 2007 ) Situated in the southern highlands at the feet of the Uluguru Mountains, Morogoro is home to Sokoine University of Agriculture (among other educational institutions) and

PAGE 27

11 is a major center for agricultural production, including a wide range of both food and cash crops ( Ministry of Planning Economy, and Empowerment 2007 ; National Institute for Medica l Research, 2012 ) During my first months in Tanzania while studying Kiswahili in Morogoro and traveling back and forth to Dar es Salaam frequently for meetings, the vast sisal fields stretching out i n every direction on approach from the east were the first signs I was nearing home. In addition to agriculture, many of the religious and non profit organizations, foreign and domestic, operating in the region are based in Morogoro. For the purposes of th is study, the primary importance of the city of Morogoro is as the location of a referral health facility for cervical cancer screening clients requiring a specific procedure (known as Loop Electrosurgical Excision Procedure or LEEP ) that cannot be obtain ed locally at the field site but whose cases are not advanced enough to necessitate a trip to the major medical facilities in Dar es Salaam. This will be discussed in more detail in a later chapter. The remaining six districts of the Morogoro region are la rger and more rural in nature ( National Bureau of Statistics 2013 ) Ifak ara and the Surrounding Area On the good, though congested, tarmac highway, the distance between Dar es Salaam and Morogoro can be covered in approximately three hours by bus. If one continues on through Morogoro, the highway eventually winds its way west directly through Mikumi National Park, where baboons linger by the roadside hoping for food scraps thrown by passengers, until it reaches the town of Mikumi on the other side. After a quick stop at the chips mayai stand (a popular street food best describe d as a French fry omelet) in Mikumi, the bus turns southwest and continues for 25 km before the tarmac

PAGE 28

12 gives way to a bumpy narrow dirt road that often floods during the rainy season. Eventually, approximately ten hours and 400 km after departure, one arri ves in Ifakara, the town in which I lived and conducted my fieldwork. Ifakara is a semi urban town of 55,956 people near the Udzung wa Mountain range in Kilombero D istrict, the second largest district in Morogoro region at 14,918 km 2 The town is set do wn n ear the Kilombero River in a val ley of the same name ( National Bureau of Statistics 2006 2013 ) As the largest urban area in the Kilombero District, Ifakara is home to the district government several NGOs, and important health and educational institutions including St. Francis Hospital ( Ifakara Health Institute, 2014 ; National Bureau of Statistics 2006 ) A long narrow tract of land, the district wraps in part around the larger but much less densely populated Ulanga District and despite primaril y serving the residents of Kilombero St. Francis is the closest hospital for many of those living across the river in Ulanga as well ( National Bureau of Statistics 2006 ). Wit h the Udzungwa Mountains to the north and lush vegetation fed by the river, the valley is a scenic place though at this point, with the exceptio n of Udzungwa Mountains National Park there is relatively little to tempt tourists here off the beaten path. The Local Economy Agricultural work is the primary economic activity in the fertile Kilombero Valley, as it is in both Morogoro Region and Tanza nia as a whole, with rice cultivation particularly important ( Hetzel et al., 2008 ; National Bureau of Stati stics 2006 2011 ) The valley has a large floodplain and when the rainy season comes, the river overflows its banks. During this time, many individuals or households relocate to the shamba (farm) for rice paddy cultivation ( Hetzel et al., 2008 ; Kato, 2007 ) Others, like the family with

PAGE 29

13 whom I lived, had access to a shamba close enough to walk to from our house in to wn and return again at night. The people of Kilombero also engage in fishing and the raising of livestock, including chickens and cattle. However, an increase in the population of the area in recent decades has precipitated a corresponding increase in lan d use for both farming and pastoral purposes, leading to concerns regarding the potential environmental impacts such as water scarcity and deforestation ( Kangalawe & Liwenga, 2005 ) After farming, the most common occupations include craftsmen, venders, and small business owners, particularly in more urban areas like Ifakara. The presence of the government, non profit, health, and educational sectors provide a significant number of professional, technical, and administrative positions as well. Population The lingua franca in Tanzania is Kiswahili part of the Bantu language group, although various triba l languages are spoken as well. At my field site, Kiswahili was most commonly spoken in day to day life, though on occasion an individual would tell me they Kiswahili wo rd and would substitute a term from another language they spoke. The do minant tribal or ethnic groups in the Morogoro region are similarly of Bantu origin. In Kilombero, these primarily include the Wambunga, Wandamba, Wabena, and Wahehe though migration h as brought other groups into the area ( National Bureau of Statistics 2006 ). In Kiswahili class of nouns denoting humans or types of people (McGrath & Marten 2003). Kilombero District is overall primarily rural, with only 28.2% of its population liv ing in urban areas. This is slightly higher t han the 23% found across Tanzania as a

PAGE 30

14 whole ( National Bureau of Statistics 2006 ). The average household size of 4.3 is representative of both the Kilombero District, and the region o f Morogoro, though slightly lower than the national average of 4.8 ( National Bureau of Statistics 2013 ). The sex ratio of men to women in Ifakara is 91, slightly lower than the district (99), region (97), and the nation (95) as a whole ( National Bureau of Statistics 2013 ). Approximately 86% of the population of Kilombero is below the age of 45, and 40% is under 15 years of age compared to 45% nationally ( National Bureau of Statistics 2006; Po pulation Reference Bureau 2012 ). According to these descripto rs, Kilombero does not differ greatly from the nation as a whole. Ifakara as a Field Site A Center of Health Research Ifakara has historically been a center for health research in the region, dating back to the founding of the Swiss Tropical and Public He alth Institute in 1956. The Institute has undergone multiple evolutions and name changes since then, before settling on Ifakara Health Institute (IHI) in 2008. Although originally founded by the Swiss, the Institute has had Tanzanian leadership since 1993, and became formally independent from the Swiss in 1996. It is an important training ground for Tanzanian scholars and to include other public health issues important to the region and the nation ( Ifakara Health Institute, 2014 ) IHI shares a campus with St. Francis Hospital, as well as St. Francis University College of Health and Allied Sciences (SFUCHAS), and the Tanzanian Training Center for International H ealth. St. Francis is a teaching hospital and the medical students are a

PAGE 31

15 frequent sight in its corridors. This arrangement has allowed the various institutions to share resources and form research partnerships. In addition to students and researchers from Tanzania and other parts of Africa, IHI and St. Francis draw a significant number of foreign medical personnel who come to work for limited periods of time at the hospital, as well as foreign researchers. St. Francis Hospital There has been a dispensary i n Ifakara since 1927; however the history of St. Francis Hospital itself extends back to 1955 when an Austrian surgeon, Dr. Schoepf, working at the small Swiss run mission hospital in Ifakara (which opened in 1931) initiated its construction. It was comple ted in 1960 and the surgeon eventually returned to health professionals until its first Tanzanian director in 1993. While it eventually waned in the years following Tan due to lack of resources Dr. Schoepf undertook efforts in 1995 to organize aid, supplies, and training for its personnel (St. Francis Hospital 2014). Since then, the hospital has rebounded. Although it bega n as a Catholic institution, in the past it also served as a Designated District Hospital under agreement with the Tanzanian government. That changed in 2010 when it became one of nine Referral Hospitals in the country. The Diocese of Mahenge and the Tanza nian government share both financing and governance of the hospital. Between Outpatient and Inpatient Departments, St. Francis treats approximately 117,000 patients per year. While it currently has 371 beds, it is undergoing expansion because it has outgro wn its current buildings. Construction was ongoing while I was conducting my research there.

PAGE 32

16 Cervical Cancer Prevention Program Location W ithin the hospital, most of my time was spent with the cervical cancer prevention program or CECAP, which had been in existence since May 2010 This program was currently housed in the Maternal and Child Health (MCH) building Though the name had officially been changed to Repr oductive and Child Health (RCH), in order to be more inclusive of men the sign outside still read MCH and most people including myself, referred to it using th e old acronym A small, single story building sitting on a cement platform in a low lying area across from the main hospital building, it was reached by a long raised cement walkway that r eached all the way to the street This walkway lay directly across the road from the gate to the HIV clinic also called CTC ( for Care and Treatment Clinic) The interior of the building contained a small courtyard, where on busy days a long line of women in brightly colored kangas (cloth wraps) holding babies would form as they each waited patiently to deposit their child into the fabric harness attached to a scale hanging from the ceiling. A nurse would direct this process, and then note the weight on a the mother. A small patio sat just off this courtyard with two doors facing the back of the building. One opened into a room where daily classes for expectant mothers were held. T he second, in the very back corner of the building, was CECAP I explain this here to illustrate where the clinic was located within the larger context of the hospital campus While the program was originally conducted out of the Gynecology Department in the main hospital building when I visited in 2011, it had been

PAGE 33

17 moved here so that it could have its own designated space. However, as a result, it was rather tucked away and not something likely to be seen by women coming to the hospital for other purposes There was no passing traffic other than the class attendees next door, and very little likelihood of someone finding it unless they were directly looking for it. Screening Services The CECAP schedule followed a general routine Screening took place two days of the week ; in 2013 it was Mondays and Wednesdays. The program opened in the morning, generally between 9 10am though there w as no set time. The two CECAP staff members had many other duties in the hospital one was a Medical Officer and the other a nurse and so it s opening time depended somewhat on their activities elsewhere While the y typically finished with screening by the early afternoon, at which time they would return to the main hospital, occasionally there were enou gh women to keep them busy longer. Typically one of the two staff members, MN or AL not both, would be present at the clinic, though again, a large number of women or specific CECAP tasks would sometimes require that they both be there. While CECAP services typically began around 9am, the main hospital gates and RCH both ope ned earlier. By the time I arrive d each morning then, at least one or two women were typically already there waiting. Often more were gathered on the bench or standing on the patio. W omen attending the classes f or pregnant and new mothers held next door also gather ed here so there was nearly always a small crowd to greet as I walked up. The screening room itself was small and it s furnishings simple (Figure 2 ) A large wooden cupboard containing all the necessary supplies, forms, and records lined the wall

PAGE 34

18 to the right as one entered An exam table with stirrups took up the center of the room, while a second that served as a makeshift storage surface stood up against another wall. A small wooden desk sat under the window, with cryotherapy tank s bearing stickers from the United States Agency for International Development (USAID) on either side in the corners. The clinic was a corner room, and large windows in two walls served as the primary source of light, with the overhead florescent bulb rarely turned on. Fluttery drapes were twisted up and tucked behind the window bars on arrival to let in the light and a breeze, with the consequence that anyone outside could see in though giv en that it faced the back of a build ing on the edge of campus this was not a common issue Figure 2 : The CECAP Roo m Preparing the room for screening was a quick process that involved the staff member on duty retrieving screening forms and records from the cabinet, disinfecting surfaces and the exam table, and setting out all of the supplies needed for screening (Figure 3) T hese supplies were few and simple cotton swabs, a flashlight, small metal containers of acetic acid and water, and speculums. Once everything w as prepared, the staff member would invite the waiting women in for an education session.

PAGE 35

19 Figure 3 : Screening Supplies There were several small wo oden footstools scattered in the room and clients would gather around, sitting on these or standing while the staff member delivered a shor t lecture. During the talk, AL and MN not only explain ed the basics of cervical cancer and cervical cancer screening, but also demonstrate d quickly and briefly on their own bodies how one would do a breast self exam. Both are short but imposing women who project an air of know ledge, but also good humor. MN in particular is always smiling and occasionally breaks out into song. They regularly check ed in with women during the lecture after finishing ea ch main point with the question, Sawa? (Okay?) b and more a call and Sawa! Questions were in fact usually saved for th e end of the lecture. Once MN or AL back outside to wait, while the first woman to be screened remained standing or seated on a stool while a staff member filled out her paperwork If she was HIV positive she would produce her blue CTC card with her history of testing and treatment to show to the staff member. If she was unsure of her HIV status, she was referred for testing. In some

PAGE 36

20 cases, though quite rarely a wom a n had been screened for c ervical cancer before. If so, a small folded cardboard form that recorded the dates and results of previous VIA screenings (VIA will be discussed in more detail in Chapter IV ) She gave this to the staff membe r so that it could be updated after the present screening. Finally, the staff member would rise and indicate that the client should remove her clothes from the waist down and climb up onto the exam table while the staff member washed her hands and put on g loves. During my observ ations, it became apparent to me without them saying so that many women had never had a gynecological exam using stirrups before because it was not unusual for the client to climb onto the table and then lie down backwards, with the ir head between the projecting stirrups and their feet where their heads should be. Or, in other cases, to put their heels into the stirrups rather than resting their ankles on the supports and to lie with legs stick straight and their rear ends resting in the center of the exam table, rather than scooted to the edge of it. I f the woman had a sense of humor about the process, this awkwardness sometimes actually served to break the ice a little bit because what would follow would be an ungainly scooting down the table while MN or AL wave d them on and the woman laughed or giggled. The screening process itself only took about five to ten minutes (the details of which are explained in Chapter IV ) and I observed many screenings in the course of my research. The s ight of foreign doctors was not unusual at St. Francis and most clients did not appear surprised to see me in the room. Some clients found it amusing Oh! (Oh! Foreigner!), they would laugh upon seeing me most largely ignored me after greetings we re exchanged, and two particular women were determined to introduce

PAGE 37

21 me to their brothers as a potential marriage candidate. Throughout the field period though, I tried to be very cognizant of the potential discomfort some might feel in my presence and on m ore than one occasion stepped out of the room if the client indicated she was uncomfortable or if I or the staff members noticed behavior that gave us pause. In these matters I tried to take my cues from AL and MN

PAGE 38

22 CHAPTER III METHODS & ANALYSIS This s tudy took place over the course of six months in 2013, during which I lived and conducted research primarily at St. Francis Hospital in Ifakara, Tanzania. Although I was in Tanzania for a total of ten months and intended to conduct my research over a nine month period following a month spent studying Kiswahili at a language school in Morogoro, difficulties in obtaining the proper research permits delayed the start of my pr oject. As a result, I spent an additional month in Morogoro (its proximity to the perm it offices in Dar es Salaam made this a practical choice) and six weeks living in Ifakara prior to the start of data collection. I was able to use this time to become acquainted with my new home and the cervical cancer prevention program and to engage wit h community members and hospital staff The focus of m y research was on CECAP, the cervical cancer prevention program at St. Francis Hospital, its clients, and another subset of women who came to the hospital seeking care in the Gynecology Department (GD) but who did not utilize the prevention program. Through methods that will be discussed in more detail below, such as semi structured interviews, participant observation, patient record reviews, and the collection of grey literature, I collected both qualit ative and quantitative data to address the research questions outlined at the beginning of this dissertation and reiterated below: (RQ1) What is the population currently attending screening programs and who is missing from this population? (RQ2) How is cervical cancer understood and given meaning among Tanzanian women seeking care? (RQ3) What influence do cultural constructions of cervical cancer have on attitudes towards cervical cancer screening? (RQ4) How does the screening program operate on a daily basis, and how does it fit within the overall structure of the hospital?

PAGE 39

23 I established the first three questions before I left for the field, but upon commencing my research I soon realized that any understanding of who came to CECAP for screenin g, and how at least some of the participants perceived cervical cancer and cervical cancer screening needed to consider how CECAP itself was situated both within the hospital itself, and within the larger intersection of the Tanzanian health care system a nd the international NGOs that work in the country. Thus I developed the fourth research question. Methodological Approach This study took an ethnographic approach, a method which emphasizes holism; in this case, the goal was to not only illuminate process es present in the site of interest, but also to provide the context in which these processes were grounded ( Marcus & Fischer, 1986 ; Neuman, 2003 ) Participant observation has long been a hallmark of ethnography as practi ced by cultural anthropol ogists, to the degree in fact that to some anthropologists participant observation is the ethnographic field method ( DeWalt & DeWalt, 2010 ) The classic version of this method was complete immersion in the society under study usually small scale, for long periods of time. To not only investigate but also to part icipate in the daily lives of those one was studying with the goal, as Marcus and of l 33). Such an undertaking is increasingly d ifficult for researchers today, who instead tend to spend shorter periods of time in the field and focus on more narrow topics ( LeCompte & Schensul, 2010 ) In the case of this study, classic participant observation was also somewhat hampered by a lack of fluency on my part in Kiswahili an d the nature of the hospital setting.

PAGE 40

24 As an anthropologist however, participant observation as a method of data collection is part of my training and my nature. Without a medical degree or medical training, my opportunities to fully immerse myself at the h ospital were obviously restricted, but I frequently found myself assisting in what ways I could, such as helping with office tasks, holding the flashlight during screenings, and folding cotton for swabs used during procedure s These types of activities all owed me to fade into the background to a certain degree and observe without drawing attention to myself. In addition to my work at the hospital, I lived with a Tanzanian family near the hospital, helped with their rice harvest, shopped at the nearby market rode on the back of piki piki (motorcycle taxis), and in general, lived my life in the community. During these daily encounters both at the hospital and outside I frequently ha d conversations that made their w ay into my field notes. Therefore, the way that I used participant observation in this study is more in line with its treatment by DeWalt and De W alt (201 0 ) as one tool in the qualitative toolbox rather than as the sole ethnographic method. Instead of classic participant observation, I utilized a m ethodological approach called polymorphous engagement: Polymorphous engagement means interacting with informants across a number of dispersed sites, not just in local communities, and sometimes in virtual form; and it means collecting data eclectically fr om a disparate array of sources in many different ways. Polymorphous engagement preserves the pragmatic amateurism that has characterized anthropological research, but displaces it away from a fetishistic obsession with participant nities we study disperse and link up across the globe, as all sorts of elites now are, and as we try to match the emergence of global cultures with a global anthropology, we must develop new resea rch strategies. [Gusterson 1997, p. 116] This approach is appropriate given that while this project is focused on understanding how women view cervical cancer and how these views in turn influence

PAGE 41

25 their utilization of screening services, this process is situated within larger political, economic, and cultural sys tems. Polymorphous engagement is particularly useful for looking at situations where the macro and micro level s interact, and just as importantly, situations where it can be difficult to separate the two into discrete entities ( Markowitz, 2001 ) While the bul k of fieldwork took plac e in a local setting, the CECAP clinic, the environment the clinic is situated in, from its immediate location o n the St. Francis Hospital complex adjacent to the HIV clinic, to its broader place within the Tanzanian healthcare system and as a recipient of funding and oversight from an international NGO, all influence how screening is utilized and in some cases, perceived. Therefore, in addition to the primary site, the CECAP clinic, I also visited the HIV clinic, other areas of the hospital campus includi n g the Gynecology Department Ocean Road Cancer Institute, and when opportunities arose, smaller dispensaries. Furthermore, polymorphous engagement arose in part out of a concern that participant observation did not translate well to settings where the a nthropologist wanted ( Gusterson, 1997, p. 116 ) Studying office setti ngs, corporations, hospitals, and other organizations or elites, may involve issues that make classic participant observation d ifficult or impossible. Gusterson notes of participant observation technique may not be readily portable to elite contexts in the U.S. where ethnographic access is by permission of people with careers at stake, where loitering strangers with not ebooks are rarely welcome, and where potential inform ants are too busy to chat (1997, p. 116). T hough not in the U.S., t his was a concern for me going into my study as well and at times I became acutely aware of how these concerns can play out in the fie ld. My visit

PAGE 42

26 to the Ocean Road campus took the form of a structured and guided tour, for example, where the doctors I met briefly were primarily interested in showing me the new technologies available for treatment at the facility. While I was typically al lowed free reign to observe and assist with menial office tasks in the CECAP clinic, there were also days where the staff were simply too busy to answer questions or to have me underfoot. Healthcare workers in the larger RCH building the clinic was housed in were perfectly happy to assist me with an interview room and chairs during their quieter hours, but were not shy about opening the door and heavily hinting we needed to move along when they needed the space; several also expressed concerns to the R esear ch Assistants (R As ) when I was not present that they were reluctant to be interviewed about their jobs, despite the fact that we did not in fact plan to do this However, in other cases I certainly experienced what Markowitz (2001) observed in her own fiel dwork, that by virtue of my being an American researcher I was granted a degree of elite status by some with whom I interacted. For example, I was very often accorded the respectful greeting of Shikamoo in the hospital setting by patients despite my being younger than many, and was certainly given latitude when it came to gendered behavioral expectations. Almost no one questioned my presence in the hallways of St. Francis, in the clinical waiting areas, in meetings between hospital staff and medical student s, or in exam rooms. It was typically assumed that if I was there, I must have permission to be, although I certainly also tried to always be sensitive and respectful to patient privacy and to the healthcare workers going about their tasks. Polymorphous en gagement also stres ses the use of multiple and dive rse research techniques. As noted before, participant observation was one such tool and I certainly

PAGE 43

27 used it when and where it was possible. In other cases my role was limited to just that of observer with no participation possible. In the following section I outline the different techniques and tools used for data collection in this re search project. Data Collection Participant Observation/Observation During the six month field period, I typically spent 3 5 days per week at the St. Francis Hospital facilities. While Mondays and Wednesdays were devoted to CECAP, on alternate days I would often sit and obse rve other areas of the hospital and speak informally with those I encountered, including SFH and SFUCHA S personnel and foreign medical students and doctors who were working there In ad dition, at various times I traveled outside of the local area to visit smaller dispensaries and Ocean Road Cancer Institute. I took field notes on the observations and exper iences that occurred during these activities, as well as on daily life in Ifakara. This generally took the form of a pen and small notebook that fit in my bag, which I could pull out to jot down notes as the day went by. In some situations, where this prov ed disruptive or distracting, I would record my obser vations as soon as I was able following the conclusion of the activity. These field note jottings were then fleshed out in the evening or early the next day to capture as many remembered details as possi ble. While all quotes from formal interview participants are taken directly from digital voice recordings, I did not carry a tape recorder with me otherwise After a few attempts, I fo und that it was awkward to use and that some of the Tanzanians I interac ted with expressed discomfort with it. In many cases I was able to write down directly what an individual said to me but when I could not capture their exact words I paraphrased them in my notes. I then followed that same

PAGE 44

28 convention in this dissertation, using quotation marks only to express those instances Participant Interviews I interviewed two groups of women at St. Francis Hospital for this study for a total of 49 participants (Appendix A) The first were CECAP clients ( n= 24) women who either came to St. Francis specifically for cervical cancer screening or were referred for screening by a health care provider in a different department at St. Francis. A second group of women were recruited from the GD ( n= 25) These women were seeking care for reproductive or gyne cological reasons outside of cervical cancer screening. By using a hospital based population in this study, I planned to focus on women who were able to overcome some of the more well known structu ral and economic barriers to screen ing such as transportation to an appropriate facility and cost of transportation and care ( Anorlu, 2008 ; Bingham et al., 2003 ) to more cl osely examine ideas around perception and knowledge of the disease, and how these interact with the use of screening services T his does not mean however that these barriers did not exist for th e women interviewed ; indeed, they were frequent topics of co nversation Furthermore, both groups of women were likely to be sexually active and therefore exposed to some of the same risk factors for cervical cancer. Each participant took part in one semi structured interview, during which they were asked questions about their general health, their awareness of cervical cancer and knowledge about the disease, how cervical cancer and screening were discussed in their communities (if at all), and whether or n ot they had prior experience with screening. The CECAP clients were also asked questions about their decision to seek out these services

PAGE 45

29 while those in the GD were instead asked to discuss if they were aware of the program and if so how, if they would con sider being screened, and what might inform that decision (for Participant Interview Guides, please see Appendix B) Given the aims of this project, p urposive sampling was used to identify interview participants because it allows the researcher to identif er ( Bernard, 2006 ; Neuman, 2003, p. 213 ) In this cas e, there were efforts were also made to balance each group across a range of factors that have been shown to influence risk for cervical cancer, specifically age, parity, and HIV status. Par ticipants were recruited on the day they sought services at SFH and without exception participated in interviews that same day in order to eliminate the financial and time burden that would likely ensue if they had to return on a different day. Recruitment occurred through several mediums. Some parti cipants were approached by the RAs while they were waiting to see a health care provider given a brief introduction to the project, and asked if they would be interested in hearing more about the study. In a fe w cases, nurses or other medical personnel referred eligible participants to the RAs. Finally, the CECAP staff conducted informational meetings ab out cervical cancer screening for women who came to the clinic on Monday and Wednesday mornings and some of th e women were informed of the study there as well. All of the women who exp ressed an interest in hearing more were invited to a private room where they were given a description of the study and asked a few questions to determine their eligibility. In order to be eligible, women had to be over the age of 18, speak Kiswahili and currently be seeking care at either CECAP or the GD. If they

PAGE 46

30 confirmed their desire to participate and were eligible to do so, they were then taken through the informed consent proce ss. The nursing staff of SFH exhibited a great deal of patience and graciousness with our need for private rooms a resource in short supply in the crowded hospital and so we tried to be conscious of both the time asked of our participants as well as the fact that we were taking up valuable space when conducting our interviews. Interviews were conducted in Kiswahili by two female Tanzanian RAs Paula and Rachel, both of whom were fourth year medical students at the local university, SFUCHAS. Rachel and Pau la were highly proficient in English, had some prior experience with the research techniques used, and took part in a training period that included discussion s o f human subjects concerns and interview techniques. I sat in on the majority of inter views in o rder to help clarify questions or issues that arose and ensure study protocol was followed. Furthermore, I was able to observe the appearance and body language of the women as they were interviewed, and the interactions between the RAs and participants. T hese types of observations can be useful for many reasons, for example, clothing status, profession, or religion and demeanor was a health setting, some of the women were experiencing symptoms or feeling poorly. If I had not been in the room, I would not have been able to observe them pointing to various parts of their body or acting out certain symptoms. Sometimes they themselv es did not have the medical vocabulary needed to talk about specific health problems, so these observations were particularly important in discerning what type of ailment was

PAGE 47

31 bothering them. Furthermore, when reviewing the interview notes later in conjunct ion with the interview transcripts, the RAs and I were better able to understand why some interviews went the way that they did. As an example, we spoke with one woman who provided only very short quiet answers to each question. In some cases, we might hav e attributed this to nervousness or perhaps ignorance about the matter at hand, but in this case, both the RA conducting the interview and I noticed that she kept placing her hand on her stomach throughout the interview. After she had done it several times the RA and I made eye contact, and she asked the client about her behavior. The client revealed that she had a lot of stomach pain and just wanted the interview to go by quickly. Of course we obliged, but also took this into consideration when examining her interview data later. During our training, we discussed the possibility that a participant might not feel comfortable with me in the room and came up with a non verbal code to use in the event that the RA conducting the interview felt it would be best for me to leave. It was not often necessary, but there were a very few cases in which a participant either directly expressed discomfort with my being there, or the RA felt that the participant appeared distracted during the intervie w by my presence, in w hich case they gave the signal and I left to stand outside the room where I could be reached if there were questions. All interviews were audio recorded, and were, on average, 25 minutes in length (though there was a wide range, from 10 52 minutes individu ally). Sample We conducted forty nine interviews in total: 24 women from CECAP and 25 wome n from the GD The average age of women in the CECAP group was 42.5 years ( range 25 65), while the GD patient grou p was younger on average at 34.5 years (24 45). Whil e attempts were made to recruit women across a wide age range for each sample,

PAGE 48

32 this was complicated by the fact that the GD clinic attracted a younger population due to its focus on fertility issues and pregnancy. In addition, while women as young as 25 we re interviewed for each group, given that older age is a risk factor for cervical cancer, it was logical that CECAP would attract a clientele that skewed older, while reproducti ve concerns would alternatively skew the GD patient pool towards women still in their child bearing years particularly given fertility rates in Tanzania ( ICF International, 2004 2012 ) This explanation likely also account s for another difference seen between the two groups: W omen interviewed from CECAP e xhibited higher parity than their fellows in the GD group though again there was variety within each group None of the CECAP women were nulliparous ; on average they had 4.2 chi ldren. While the majority of CECAP clients had fewer than five children, four had between 6 10 children compared with only one in the GD group. In contrast, the GD group had seven nulliparous women and an average parity of 2.

PAGE 49

33 Table 1 : Comparison of CECAP and GD Interview Groups CECAP ( n= 24) G D ( n= 25) Mea n Age (Range) 42.5 (25 65) 34.5 (24 45) Mean Age at Sexual Initiation (Range) 17.6 (14 23) 19.7 (14 25) Mean Parity (Range) 4.2 (1 10) 2 (0 7) HIV Status: Positive 7 (29.2%) 2 (8.0%) Negative 12 (50.0%) 16 (64.0%) Unknown 5 (20.8%) 7 (28.0 %) While at least two HIV positive, two HIV negative, and two unknown status women were interviewed for each group, because during many weeks o f the field period the hospital was out of the necessary testing supplies it was in fact impossible to know the HIV status of every woman interviewed. While CECAP and GD patients at the hospital would typically be asked for their status and referred to the HIV clinic for testing if it was unknown as a matter of course, t his was not possible during period s of supply shortage. The majority of women we interviewed self reported their HIV status though patients who were positive often visib ly carried their CTC cards with them (which tracked visits for HIV care and treatment) Patient Information Forms In order to capt ure basic demographic information about each interview participant, the RAs used a Patient Information Form (PIF) (Appendix C) during the

PAGE 50

34 interviews. While initially this form was to be completed at the beginning of each interview, we found that it was pre ferable to move the process to the end both because the participants would often offer some of the information asked for in the course of the interview, and because a few of the questions asked on the for m were somewhat sensitive (e.g. Age of Initiation o f Sexual Activity an d Number of Sexual Partners). P articipants generally appeared more comforta ble addressing these sensitive questions after they had been speaking for a while The questions on the form were usually asked in a conversational manner so as not to interrupt the flow of the interview, but by recording the data on a specific sheet it ensured that the same data was collected for each participant. Key Informants and Specialized Informants H. Russell Bernard distinguishes between two types of inf ormants in his writing on anthropological methods : key informants and specialized informants. The former, ile specialized informant are 196). While I spoke with many people in the course of my day to day interactions, there were those who became important sources of information and guidan ce when it came to navigating both daily life in Tanzania and my work life at the hospital ; these I considered to be key informants My encounters with them were primarily informal interactions and discussions that arose either in response to a specific ev ent or experience, or more naturally out of shared social interactions. I n contrast I conducted formal sit down in terviews with five specific specialized informants (Tab le 2) chosen for their unique or

PAGE 51

35 in depth knowledge regarding issues related to my re search. Three of these were employees of St. Francis Hospital, one worked for Jhp ie go, and one was a foreign doctor working for an extended period of time at St. Francis on HIV issues Table 2 : Specialized Informants Specialized In formant Affiliation Physician Specializing in Obstetrics/Gynecology St. Francis Hospital Assistant Medical Officer in Obstetrics/Gynecology St. Francis Hospital Inpatient Nurse St. Francis Hospital Visiting (Foreign) Physician Specializing in HIV St. Francis Hospital Physician and Public Health Specialist Jhpiego Screening Map Forms T his project was p rimarily qualitative in nature however I also collected data to be analyzed statistically in order to address the first research question : What is the population currently attending the screening program and who is missing from this population? Furthermore, I hoped it would allow me to summarize data on the screening, treatment, and referral practices of the scree ning program though as I will discuss in a later section this was complicated by the sometimes incomplete nature of these forms In addition, CECAP wanted these forms to be digitized for its own use. The SMF is filled out for each client when they come in for screening. In addition to important client information the form asks for information on HIV status and ART use, and records details of the visit, screening results, and any follo w up activities or care required At the time of this study, S MFs exis ted only in paper form with no digital

PAGE 52

36 record This made it difficult to examine the patient data as a whole. During my time with CECAP, I created an Excel spreadsheet that contained the SMF data of every client CECAP had seen since the program began in 2 010. In all, data from 2440 SMFs were collected, spanning the period from May 2010 October 2013. This database was then imported into SPSS for statistical analysis and a copy of t he Excel file was also then given to the staff of the screening program for u se in their own program reports and potential future funding applications The final v ariables used for this project from the SMFs can be found in Table 3 below ; please see Appendix D for the original document Table 3 : Final Screen ing Map Form Variables SMF Section Variables Collected Client Information VIA Screening Date Age Parity District of Residence HIV Status HIV Status ART Status Source of HIV Status PITC Offered? PITC Accepted? HIV Results from PITC Visit Type Visit Type VIA VIA Conducted Today? VIA Results (If VIA conducted today) Treatment Cryotherapy performed today Cryotherapy postponed/ Reason Referral Referred? Referred to Reason for Referral (ART= Antiretroviral Therapy, HIV=Human Immunodeficiency Virus, PITC= Provider Initiated HIV Testing and Counseling, VIA= Visual Inspection with Acetic Acid)

PAGE 53

37 Supplementary Information Ethnographic research can benefit greatly from examining grey literature that may only be accessible while at the field site or at the very least difficult to obtain, such as government documents or agency reports ( Schensul & LeCompte, 2012 ) These documents can provide background or historical information, as well as context and descriptive data D uring the field period I spent time examining various archival documents associated with the screening program, including training materials, meeting notes, and reports. These documents were all in English and were kept in storage by CECAP. Initially I undertook this task simply to bet ter acquaint myself with the history of the program and to provide a richer descr iption of the prevention program, but after discussions with specialized informants, and reviewing the early participant transcripts, I developed the fourth research question ( How does the screening program operate on a daily basis, and how does it fit within the overall structure of the hospital ? ) and b egan to collect this inform ation in a more deliberate manner I also conducted a thorough examination of archival literature a nd secondary data availabl e at the SFUCHAS library in order to gain a better contextual understanding of cervical cancer screening and treatment both in Ifakara and Tanzania as a whole. This literature included research reports of previous related projects conducted at SFH, government reports a nd publications, as well as reports published by various NGOs, community groups, and global health organi zations.

PAGE 54

38 Analysis Qualitative Data I took notes on potentially relevant archival documents and secondary quali tative data, and in some cases a digital camera was used as a makeshift photocopier to record documents for later reference; these notes were typed up and stored with the images on an external hard drive. Throughout the field period, I fleshed out and type d up field notes; these are useful as a daily record, but also provide a place for reflexivity during the fieldwork process and serve as a first round in analysis as the researcher digests what they have observed and experienced ( DeWalt & DeWalt, 2010 ) In four out of five cases, the specialized informants were q uite proficient in English and so I conducted and transcribed the interviews myself. In the fifth, an RA conducted the interview in Kiswahili because the informant was more comfortable in that language; the transcription and translation was done by an RA a s well. I then uploaded all of these documents into Atlas.ti, a qualitative data analysis program (in the case of archival research, the notes and any available PDFs were uploaded to Atlas.ti, rather than the documents themselves) for organization and codi ng when relevant. After each participant interview, the recording was assigned to an RA for translation a nd transcription Ethnography is an inductive and iterative approach ( LeCompte & Schensul, 2010 ) and so t his process began immediately after the first day of interviews so that early t ranscripts could inform future interviews. According to the original study design, the interviews were to be spread out over eight months, however the period allotted for interviews had to be greatly condensed This was in part due to the delay in obtainin g study permits but primarily because of unexpected changes in the

PAGE 55

39 requiring them to leave the field site prematurely. Because of this, there was not adequate time to review each transcript prior to conducting the next interview. I ns tead we worked with small batches of transcripts at a time. I re ad each transcript alone first, making notes where I had questions or suggestions for future interviews and then the RAs and I met for a read through and discussion of any point I had marked. While we would discuss their impressions of each interview immediately after its completion, these read throughs also gave the RAs an opportunity to share any additional thoughts that were triggered by rereading the transcript itself. Following these meeti ngs, the final transcripts with any add itional notes resulting from said meetings were upload ed into Atlas. ti C ertain themes r elated to screening perceptions and decision making have been ide ntified by previous researchers, however, I was cognizant of the fact that relatively few of these studies were conducted in this geographic and cultural setting. I therefore first coded the interview transcripts inductively, using what Glaser and Strauss (1977) e structure Inductive codes, or codes derived from the data themselves, are useful in that they come without the preconceived ideas and definitions that can weigh down deductive concepts borrowed from the literature ( Patton, 1990 ; Strauss & Corbin, 1990 ) Strauss and Corbin write of tic meaning and may already be considerably 68). While deductive concepts are useful, and as I outline below, were eventually added to this analysis, using them from the beginning in exploratory research can also caus e the researcher to inadvertently impose a structure on the data that is not in evidence.

PAGE 56

40 As I developed the preliminary code structure, it was continually refined as new chunks of text were compared to existing categories and novel ones were added. The c oded chunks of text within each category were reviewed with a critical eye to determine inclusion and exclusion criteria, and excerpts were at times reassigned with new codes. Eventually the preliminary code structure was compared to the existing literatur e and potential deductive codes considered and added when relevant. Through this process I developed a final coding scheme and codebook ( Schensul & LeCompte, 2012 ) The data was then recoded using the tools provided in Atlas.ti and memos kep t throughout the coding process were used to develop the analysis and resulting ideas and insights, as well as track potential new directions for research ( Bernard, 2006 ) Quantitative Data I reviewed all SMFs collected from the beginning of the screening program in May 2010 through October 2013: 2440 forms in total. This included all SMFs resulting from outreach events conducted by SFH as well. Although I collected and digitized data from these forms, for this dissertation I chose to focus only on those 1340 that resulted from on site St. Francis clients. The primary reason for this choice was that this study focuses on utilization of the services provided by CECAP at St. Francis. Outreach events involved large numbers of women from one single location; usually this location was the village in which the event was being held and the surrounding area. These villages were often some distance from St. Francis, which was the primary reason for taking screening service s to them, rather than the reverse. The events were heavily advertised within the area to ensure high turnout. For these reasons, outreach clients were not necessarily typical of the average CECAP user and did not face the same barriers to screening.

PAGE 57

41 Furth ermore, outreach events usually involved supplementary staff members and cryotherapy was not always provided. Instead women were referred to St Francis for treatment. Therefore outreach events also were not typical of standard CECAP operations. The data fr om the remaining 1340 forms had some important limitations. First, important for this study, the variables Parity and Source of HIV Status were not added until the Octo ber 2010 version of the form; the relevant data was therefore missing for the first 196 clients. Second, certain variables were entered more faithfully by CECAP however, q uestions about PITC were not and were therefore not very useful in the where the client was referred was often not. I had hoped to use this information but in the end di d not include that variable in my analysis. Finally, for a period of time during my fieldwork, St. Francis was out of the reagent needed to test for HIV. Although all clients in the screening program were asked about their HIV status and, if needed, refer red for testing, during this time period testing typically offering PITC. The final va riables uploaded in to SPSS can be found in Table 3 above. The data was entered into an SPSS spreadsheet and cleaned line by line to ensure accurate translation from paper form to digital medium. All variables were then recoded

PAGE 58

42 into numerical values for ana lysis in SPSS and a codebook was created. Once in SPSS, descriptive statistics were run to explore the data and describe the sample. Sample Women in this sample had an average age of 39.6 (range 18 76) and age had a normal distribution. The mean parity wa s 3.8 children (range 0 17) and the mode was 3 .0 children. The majority of women ( 88.9 %) were from the Kilombero district, with 5.2% coming from neighboring Ulanga. The remaining women hailed from Kilosa, Morogoro, Iringa, Dar es Salaam, Shinyanga, Njombe, Handeni, or Kilolo districts. While 3 0.8 % of the sample was HIV positive and 43.3 % was HIV negative the 25.9 % with unknown status include those who could not be tested due to the missing reagent. Of those who were HIV positive 86.7% were on ART at the t ime of their cervical cancer screening. This is not surprising as the government provides free access to ART for individuals below a specific CD4 count at CTC clinics throughout the country. For those women with a known HIV status 70.9% self reported, whi le 2 9 .1% brought their CTC cards with them as confirmation of status. Four years into the CECAP program, the clients who come to utilize its services are still primarily those seeking an Initial VIA Screening, as indicated by the SMF forms. However, this s imply notes whether or not it is their initial visit to CECAP at St. Francis. It is possible they have been screened elsewhere in the past. Only 12 ( 0 .9% of visits) HIV positive women have returned for their One Year Follow Up screening, while 4.1% and 0 .3 % of visits were for Routine VIA Screening or Postponed Cryotherapy respectively. Every SMF form indicated that VIA had been done on the day the client attended the program, however, having observed in the clinic I know that the staff typically simply did

PAGE 59

43 not fill out a form for women who could not go through with screening for whatever There was no missing d ata for VIA results; a result was lodged for each and every client. Since the inception of the program, 86.4% ( n= 1158) of women had been screened and found negative. Positive results, meaning small lesions that could be treated with cryotherapy, were found in 3.1% of cases ( n= 41), while the staff found large lesions in 3.0% ( n= 40) and suspected canc er in 7.5% ( n= 101). In other words, there was some type of cervical abnormality in 13.6% of clients. For those women who were positive and eligible for cryotherapy, 78 .0 % had cryotherapy done the same day. The other cases were postponed due to other gyneco logical issues (cervicitis or pelvic inflammatory disease, for example), or for personal reasons (such as wanting their husband with them). In one case, the clinic was temporarily without a working cryotherapy tank. One hundred sixty one women were given r eferrals for other services (12 .0 % of the total client cases). The most common place of referral was to Ocean Road for suspected cancer ( n= 94). Another 31 women were referred either to Morogoro or to KCMC in the north of the country for LEEP a procedure u sed to treat larger lesions, depending on which site was more convenient for them. The remain ing women were referred to different areas of St. Francis for other gynecological issues or for biopsy services.

PAGE 60

44 CHAPTER IV : CHANGING PRIORITIES Dr. M is one of the most senior physicians on staff at St. Francis Hospital. As the head of the Gynecology Department, he has overseen the cervical cancer prevention program since its inception in 2010. But his years of experience stretch back much further than that and I had been looking conversations in this, the large room between the administrative block and the dental clinic where he had both his consulting desk and exam table, over the course of my t ime in Ifakara. But today was the first time we touched on his own perceptions of the growing interest in cervical cancer in Tanzania. peering over his glasses, the large national referral hospital in Dar es Salaam, and a center for both physician s, patients, and laypersons with whom I have shared my research agenda while in Tanzania, but a few have also expressed the opinion that the ability to screen for cancer is simply what has improved, and that both health care providers and the public are mo re aware. But Dr. M is firm on Given his long tenure of 30 plus years as a physician, both in the large city of Dar es Salaam and in rural Ifakar a, Dr. M had seen many shifts in health priorities in Tanzania and in government policy making. I asked him to share if and how he felt the government had become more interested or involved in addressing cervical cancer over the course of his career. Dr. M thought back to early initiatives to screen for breast cancer, and said he believed they were responsible for creating an awakening in the government that cancer was a problem in Tanzania, which eventually led to the current interest in cancer of the cerv ix: So the government has also taken some steps to make sure parliament from the Ministry of Health, [ which ] was saying that this will be one of the things, to screen the women. The women will be screene d as other countries are doing. And this was repeated by the president, I think two, three days ago. read the same article in the Tanzania Daily News last week (Semaya 2013) It profiled Presid ent Jakaya Kikwete and former U.S. President George W. Bush, who had spoken to the African First Ladies Summit about the year strategic plan to reduce cervical cancer and to

PAGE 61

45 implement a major immunization program with the help of both pub lic and private partners. In his speech, President Kikwete called cervical cancer a I mentioned I had read the article too and Dr. M said that he thought the number of women they scre ened at St. Francis Hospital w as a bit higher now than in the previous couple of years because awareness of cervical Adapted f rom field notes and interview transcript (July 10, 2013) Dr. M had spent over 30 years in the medical profession in Tanzania, but only recently had cervical cancer become an important focus for health education and intervention. As noted above, he had s een the first cervical cancer screening program implemented at St. Francis only three years before (as it turned out, this implementation came just four years before his retirement). His personal experiences reflected broader changes not only in public hea lth priorities within Tanzania, but also in the arena of global health. This chapter details t he emergence of cervical cancer as an important global health issue part of a broader new focus on chronic and non communicable diseases and in particular, how a new screening method changed the way screening programs could be implemented in developing nations. Corresponding changes within Tanzania, at both the national and local levels are then examined within this context, setting the stage for later chapters. A Global Picture Approximately 528,000 women developed cervical cancer in 2012 worldwide. Compared to the 1,677,000 who developed breast cancer that same year, cervical cancer might at first glance appear to be a less critical global health cancer issue ( IARC 2014). However, in this case, context is essential. To look at the picture of cervical cancer as a

PAGE 62

46 whole is to miss important nuances in the distribution of the disease across regions of the world. In fact, the burden of cervical cancer falls largely on the developing world, where 85% of cervical cancer cases and 87% of cervical cancer deaths are found ( Ferlay et al., 2013 ) In many of these regions, cervical cancer is the number one female cancer and the largest, or one of the largest, contributors to years of life lost from cancer ( Arbyn et al., 2011 ; Yang et al., 2004 ) Due to this disparity between different regions of the world according to development status, cervica l cancer is considered to be a disease of poverty. Tsu and Levin (2008 ) argue that the burden of cervical cancer is not simply a n important public health issue, it is also a n issue of health inequity It should therefore be discussed not only in terms of actual objective disparities, but also in terms of how poverty contributes to and shapes the distribution of disease: While every untimely death, illness and suffering is of concern, a widely accepted principle of social justice says those who are most vulnerable to illness or less able to access services, because of social or demographic characteristics over which they have little control, are entitled to special consideratio n. Poverty is the most commonly identified source of inequity, but other aspects include gender, ethnicity, religion, geography, age, education and social status, especially when these cluster and magnify negative effects. These concepts of inequity, both in health status and access to health services, resonate strongly with regard to cervical cancer. [Tsu & Levin 2008, p. 104 5 ] Furthermore, this disease primarily affects a critical portion of the population: young to middle aged women with children. The se women are crucial members of society and may be the only source of economic support and caregiving for their families ( Anorlu, 2008 ; Franco & Harper, 2005 ) In sub Saharan Africa, the region in which Tanzania is located, this is particular ly concern ing given the impact of the HIV pandemic, where older women are increasingly stepping into these roles as their younge r

PAGE 63

47 counterparts fall victim to the disease ( Knodel, Watkins, & VanLandingham, 2003 ; Schatz, 2010 ) In addition as has often been noted in the literature, women are responsible for the majority of agricultural work and food production in much of the developing world ( Ellis, Blackden, Cutura, MacCulloch, & Seebens, 2007 ; Ishikawa, 1999 ) This is certainly true in Tanzania, where acc ording to the 2006 World Economic countries, in terms of wo ticipation ( Ellis et al., 2007 ) As I no ted in Chapter I the statistics for a woman in sub Saharan Africa when it comes to cervical cancer are discouraging to say the least. Sub Saharan African women are not only more likely to develop cervical cancer than their counterparts in Europe or the Un ited States, their cases are more likely to be fatal ( Ferlay et al., 2013 ; Franco, Schlecht, & Saslow, 2003 ) Once diagnosed with cervical cancer, they have a probability of survival of just 21%, compared to 42 70% in other areas of the world ( Anorlu, 2008 ; Okonofua, 2007 ) The situation is particularly bad when the geogra phic area in question is narrowed to East Africa, which has a higher incidence rate than any other sub Saharan region ( Ferlay et al., 2013 ) While the overall mortality rate for women globally is only 9 per 100,000, it is 34 per 100,000 for the region ( Perng et al., 2013 ; Peters, Soliman, Bukori, Mkuchu, & Ngoma, 2010 ) When compared to both the region of East Africa and sub Saharan Africa as a whole, Tanzania itself still stands out as having one of the highest age standardized rates of cervical cancer at 50.9 cases per 100,000, and it is the most common cancer to affect young women between the ages of 18 44 ( Castellsagu et al., 2012 ; Hutubessy et al., 2012 ) A Tanzanian woman under the age of 64 has a cumulative risk of cervical cancer

PAGE 64

48 nearly four times that of women wor ldwide ( Okonofua, 2007 ) She is also more likely to have her disease progress unchecked and to present for care at a later stage of cancer than her counterparts in developed nations ( Kidanto, Kilewo, & Moshiro, 2002 ; Mosha et al., 2009 ) Despite these disheartening facts, cervical cancer as a public health concern is a fairly new arrival to the national scene. Health Transitions The word cancer often evokes certain images and ideas. Traditionally we think of cancer as a chronic or non communicable condition, often one that develops in middle or advanced age though there are certa inly cancers that strike children or young adults A s an umbrella term, it is often lumped in with heart disease, diabetes, and other so called lifestyle diseases that affect developed nations where life expectancy is high and sanitation and other public h ealth interventions have made the threat posed by infectious diseases (in particular, infectious childhood diseases) far less likely. The term epidemiological transition (or health transition ) is often used to describe this changing pattern in causes of di sease and mortality: As nations undergo development infectious diseases decline and chronic and degenerative diseases take their place ( Frenk, Bobadilla, Stern, Frejka, & Lozano, 1991 ; Omran, 1971 ) Chronic diseases are on the rise globally, including in low and middle income c ountries ( Kanavos, 2006 ) This trend, not been seen to the same extent in sub Saharan Africa as it has in other developing regions, however, it is occurring and is influenced in part by changing demographics and the increasing impact of a wes ternized lifestyle ( Kanavos, 2006, p. viii15 ; Morhason Bello et al., 2013 ) But the trend has not c orresponded everywhere to a reduction in

PAGE 65

49 infectious diseases. Instead, regions like sub ( Maher, Smeeth, & Sekajugo, 2010, p. 943 ) The situation is further complicated by the fact that certain types of cancer muddle the seemingly neat categories of communicable and non communicable diseases. While cancer itself may b e non communicable, the past three decades have greatly expanded our understanding of the role viruses play in the development of some malignancies, for example Hepatitis B/Hepatitis C and liver cancer, and Helicobacter pylori and stomach cancer ( Franco, Duarte Franco, & Ferenczy, 2001 ; Jones, 1999 ) In the case of cervical cancer, the role of the human papillomavir us is essential in understanding both how cervical cancer develops, and how it can be prevented. Human papillomavirus or HPV, is the most common sexually transmitted infection (STI) and is what is called a necessary cause of cervical cancer ( Bosch, Manos, Munoz, & al., 1995 ; Center s for Disease Control, 2004 ; Daila rd, 2003 ; Franco et al., 2001 ; Walboomers et al., 1999 ) This relationship was only confirmed in the late 1990s ( Walboomers e t al., 1999 ) Specifically, cervical cancer is caused by particular strains, or types, of HPV that infect the anogenital tract; although 40 HPV types are known to infect this region, less than half actually cause cancer ( Franco & Harper, 2005 ) Being termed a necessary cause indicates that while infection with HPV alone is not sufficien t for cervical cancer to develop in an individual, the virus has been implicated in 99.7% of cervical cancer cases, i.e. in nearly all cases the presence of HPV is necessary for an individual to develop cervical cancer ( Franco & Harper, 2005 ; Walboomers et al., 1999 ) While the average person may be more aware of the association between smoking and

PAGE 66

50 lung cancer, in fact, this association is weaker than that between HPV and cervical cancer ( Franco et al., 2001 ) While HPV is a necessary cause, it is not a sufficient cause of cer vical cancer ( Franco & Harper, 2005 ) In other words, infection with HPV is not enou gh by itself to cause the development of cervical cancer. In the majority of cases, infection is cleared by m, however persistent infection with particular types can lead to eventual cancer of the ce rvix ( Center s for Disease Control, 2007 ; Krishnan, 200 8 ) At this point in time, attempts to discover been unsuccessful ( Trottier & Franco, 2006 ) Researchers have been more successful in in fection progress to cervical cancer; this will be discussed further in Chapter V The Macro Policy Environment These categories, communicable and non communicable may be problematic when applied to cervical cancer, but they are important when it comes to understanding health priorities at the national and interna tional levels. Farmer (1999 ) has pointed o ut that the health transitions model held a certain appeal for the powerful international financial institutions that shape development programs. This appeal had multiple components First, developed nations that had already reduced their burden of infecti ous diseases provided successful model s for development. Second, infectious diseases had the potential to cross borders and cause epidemics. Furthermore, through the use of vaccines, certain infectious diseases could be prevented, or even potentially eradi cated ( Anderson & Chu, 2007 ; Davies, 2010 ) These were also coupled with certain prevalent ideas around

PAGE 67

51 chroni c conditions: that they primarily occurred in high income nations, that tackling them would be more expensive, and that they were largely a problem of the elderly ( An derson & Chu, 2007 ) The infectious disease burden is acute in the developing world, along with, for associated with reproduction (2013, p. 683). This cannot be denied. H owever, as noted above, non communicable diseases are on the rise. In the last decade, there have been increasing calls for renewed emphasis on both non communicable conditions themselves, and non communicable conditions that interact or arise from communi cable di seases, such as cervical cancer ( Anderson & Chu, 2007 ; Fuster & Vote, 2005 ; Maher et al., 2010 ; Seffrin et al., 2009 ) Development agency funding focused on addressing public health outcomes in the developing world has also not kept pace with its non communicable disease burden ( Seffrin et al., 2009 ) There are limited resources to address the many health needs that exist in lo w and middle income countries. HIV, tuberculosis, and malaria in particular are often the center of much international aid, along with maternal mortality ( Anderson & Chu, 2007 ; Morhason Bello et al., 2013 ) However, in some cases, program s that address non communicable diseases may in fact be a better use of limited global health funds, even if their actual disease burden is not as large. Cervical cancer has taken a huge toll on women in the developing world. Technologies currently exist that could alleviate much of this burden and they are transferable to low resource settings, unlike other important causes of morbidity an d mortality for which comparable treatments and technologies may

PAGE 68

52 not be available, or may not have the pot ential for mass implementation ( Yang et al., 2004 ) Transferable Technologies In fact, c ervical cancer appears at first glance to be an ideal target for public health intervention in the developing world. It can be prevented in the majority of cases usin g already developed screening techniques, with some nations reporting a reduction in incidence of the disease as high as 70 90% following the implementation of high coverage screening programs ( Gustafsson, Pontn, Zack, & Adami, 1997 ) It is also typically slow to develop once an individual is infected with the HPV virus, often taking years or decades for persistent HPV infection to develop into invasive canc er, and during much of that time early medical intervention is possible Routine use of cytology in the form of the Pap smear has played a major role in dramatically reducing the incidence of cervical cancer in developed nations since its development in th e early 20 th century and remains an important part of prevention strategies there today, along with the newer HPV testing ( Crum, Abbott, & Quade, 2003 ; U.S. Preve ntive Services Task Force, 2012 ; Vilos, 1998 ) Unfortunately these screening tools, and standard follow up treatment methods such as LEEP cryotherapy, or cold knife conization require medical infrastructure and financial resources that are simply not always available in resource poor regions ( Drain, Holmes, Hughes, & Koutsky, 2002 ) Furthermore, a lack of trained staff and the mismanagement of what few re sources do exist have also played a role in stifling cervical cancer programs ( Anorlu, 2008 ; Bingham et al., 2003 ; Megevand, Van Wyk, Knight, & Bloch, 1996 )

PAGE 69

53 With this in mind, researchers developed new methods for screening and treatment in low resource settings focusing on using visual inspection with either acetic acid (VIA) (VILI) Both solutions are utilized in a similar manner. For VIA, the healthcare worker administering the screening uses a cotton swab to spread a solution of 3 5% dilute acetic acid across the surface of the cervix. Aft er waiting for a minimum of one minute, the worker then examines the cervix using a flashlight for any lesions also known as acetowhite lesions which will have turned white during the intervening time ( Sankaranarayanan, Gaffikin, Jacob, Sellors, & Robles, 2005 ) During the screening process, the worker will also examine the cervix and vaginal canal for bleeding, unusual growths, inflammation, or discharge. Unlike the Pap smear, which requires a waiting period for the results and typically a follow up appointment, VIA allows treatment to be conducted using cryotherapy during the sam e visit if the visible lesion is below a certain threshold lea ding to the name one visit screen and treat approach Larger lesions or lesions which have spread beyond the cervix will still typically require more extensive follow up treatment. VIA has now been evaluated in comparison with the Pap smear and HPV testing ( Denny, 2005 ; Sankaranarayanan et al., 2005 ) All methods have limitations, for example, the Pap smear does not directly test whether or not a woman is positive for HPV, but rather looks for changes to the cervix brought on by the virus (Krishnan 2008). It therefore is not surprising that while it has a high specificity (98 99%), it has only a moderate sensitivity (50 75%) ( Krishnan, 2008 ; Sankaranarayanan et al., 2005 ; Urasa & Darj, 2011 ) On the other han d, HPV testing has a high sensitivity, but a less impressive specificity and positive predictive value; this high sensitivity means that many women

PAGE 70

54 who receive a positive result are likely to spontaneously clear the infection before any cytological abnorma lity develops ( Franco et al., 2003 ; Jeronimo & Schiffman, 2006 ; Krishnan, 2008 ) The reported sensitivity and specificity values of VIA have varied with one meta analysis demonstrating ranges from 67% to 79% and from 49% to 86%, respectively ( Sankaranarayanan et al., 2005 ) However, in comparative studies around the world, VIA has typically been shown to have a higher or similar sensitivity to the Pap smear and lower specificity ( Bhatla et al., 2010 ; Bomfim, Santana Franco, & Bahamondes, 2005 ; Doh et al., 2005 ; Goel et al., 2005 ; Shastri et al., 2005 ) In other words, VIA effectivel y detects true positives, but it does not detect true negatives as well as the Pap smear As a result, o ne concern with VIA is that it results in more false positives and therefore overtreatment, given that there is no biopsy or confirmation process for acetowhite lesions in a one visit screen and treat approach. Additionally, while VIA is useful for one of th e two types of cervical lesions, squamous lesions, it is much less effective in identifying glandular precursor lesions ( Sankaranarayanan et al., 2005 ) There are many benefi ts, however, to using one visit visual inspection methods. Cost is typically low and the resources needed to perform the screening cotton swabs, vinegar, a flashlight, a private room or privacy screen, a speculum, and an examination table are often availab le in or near clinics even in low resource settings. While training is necessary to properly carry out the screening and to recognize the potential conditions one might encounter, VIA can often be done by paramedical, or even nonmedical, personnel who hav e undergone short training programs (approximately 5 10 days) rather than requiring physicians to carry it out ( Ghosh, Gandhi, Kochhar, Zutshi, & Batra, 2012 ;

PAGE 71

55 Sankaranarayanan et al., 2005 ) Furthermore, VIA reduces the burden on the patient in terms of the cost, time, and transportation requirements of a multiple visit strategy. As Sankaranarayanan et a p. S8). Despite the potential for overtreatment, VIA is considered to be a viable and valuable screening tool in low resource settings ( Belhadj, Rasanathan, Denny, & Broutet, 2013 ) Cervical Cancer in Tanzania Although the first cancer control program in Tanzania was initiated in 1978, through funding provi ded by USAID, a report by that same agency evaluating the ( Dunlop et al., 1990, pp. F 1 ) This view has slowly changed in the decades since, corresponding to the larger changes in global health priorities outlined in the previous section. Ocean Road Cancer Institute, a public institution devoted solely to the issue of cancer, has been open since 1996 and treatment for cervical cancer is theore tically free, assuming a space is available and the patient can reach the hospital ( Ocean Road Cancer Institute, 2014 ; Peters et al., 2010 ) Additional ly, the Strategic Plan of the National Institute for Medical Research (NIMR) in Tanzania, which traditionally focused on infectious diseases, also included non communicable diseases such as cervical cancer for the first time in 2008. After several fits and starts during the latter half of the 20 th century, and despite difficulties with the quality of data available, a new attempt to build a national cancer registry began in 2010 ( African Cancer Registry Network, 2015 ) Preliminary HPV vaccine trials and cost analyse s have taken place, and larger scale programs began last year in 2014 with the assistance of the

PAGE 72

56 Global Alliance for Vaccines and Immunization (GAVI) ( Cagney, 2013 ; GAVI, 2013 ; Hutubessy et al., 2012 ; Quentin et al., 2012 ) However, while the fourth edition o f the Standard Treatment Guidelines for Mainland Tanzania issued by the Ministry of Health and Social Welfare (MOHSW) has updated its entry for cervical cancer, national cervical cancer service policy guidelines still do not exist ( Lyimo & Beran, 2012 ; Ministry of Health and Social Welfare, 2013; Urasa & Darj, 2011 ) Few research studies have been done on cervical cancer specifically in Tanzania, though the general regions of sub Saharan Africa or East Africa have been part of wider discussions in the literature on the disease in the deve loping world for some time now ( Drain et al., 2002 ; Jones, 1999 ; Louie, de Sanjose, & Mayaud, 2009 ; Sankaranarayanan, Budukh, & Rajkumar, 2001 ) However, when the relationship between HPV and cervical cancer was confirmed in the late 1990s, attention quickly turned to the potential implications for vaccination as a tool for primary prevention. The few studies related to cervical cancer in Tanzania done both just before this confirmation and during the decade that followed therefore unsurprisingly focused on the distribution of HPV types present ( Bosch et al., 1995 ; Clifford, Smith, Plummer, Muoz, & Franceschi, 2003 ; Mayaud et al., 2001 ; Mayaud, Weiss, Lacey, Gill, & Mabey, 2003 ; Muoz et al., 2004 ; Ter Meulen et al., 1992 ) Often th ese studies were not restricted to Tanzania, but included the nation in a broader grouping of developing countries. In addition, there was also an interest in evaluating the existing resources for screening and treatment but these resources were limited a nd primarily confined to large health centers in cities ( Chirenje et al., 2001 ; Ngwalle et al., 2001 ) Nor i s Tanzania unusual in this respect; rather, with a few exceptions, particularly in Latin and South America, this was the norm.

PAGE 73

57 While vaccination held great promise for reducing the overall incidence of the disease over time, it could not protect women who were already infected with HPV. There screening remained the primary hope. Visual inspection methods were developed and tested in the mid 1990s and after early trials verified their usability, they opened up possibilities for more effective and efficien t cervical cancer screening programs in low resource settings. Unfortunately, given the already burdened and limited health care infrastructure, implementation of screening programs in Tanzania was fairly non existent outside of large health centers and mo bile units ( Perng et al., 2013 ; Peters et al., 2010 ) NGO and Government Partnerships I nternational donors a nd NGOs have since broadened the availability of cervical cancer screening, though the infrastructure is still quite limited ( Peters et al., 2010 ; WHO/ICO, 2010 ). These types of actors in the health care arena are not an uncommon phenomenon in Tanzania, where vertical intervention programs addressing specific health issues have frequently been implemented to addre ss gaps in services that cannot be adequately provided by the struggling national health care system ( Hanson, 2000 ) During previous trips to T anzania and during my field period, I heard stories of mobile vans or teams of doctors who had come through the area to screen for breast or cervical cancer, but along with these came complaints that they rarely came on a set schedule and sometimes did not return at all. Over the past decade, partnerships between the government and these entities have increased, in an effort to build more sustainable programs. Ocean Road, for example, began a screening program in 2003 at its main campus in Dar es Salaam, as well as introduced an outreach program that includes education and training for clinics

PAGE 74

58 around the country ( Peters et al., 2010 ) During my visit to Mwanza in 2010, I witnessed another example of this when I spoke with members of a foreign non profit working in partnership with the Bugando Hospital Obstetrics/Gynecology Department to increase their o utreach screening services in the surrounding area, and Bugando itself was building their own clinic based prevention program. The scope of need for screening services however is still overwhelming. Perng et al. (2013 ) observed while there were 74 active cervical cancer screening sites in Tanzania by March of 2012, they were still concentrated in urban areas, despite the fact that three quarters of women at risk for cervical cancer in Tanzania actually l ive in rural locations. The cervical cancer prevention program that served as the field site for this dissertation research project was begun in 2010 by the U.S based non profit Jhpieg o and its consortium partners, including the Tanzanian government, with funding provided by USAID. It was developed as an addition to the MAISHA (Mothers and Infa nts, Safe, Healthy, and Alive) program. MAISHA included multiple previous program components focused on maternal and child health. A ccording to a Jhpiego representat inception, the organization was already involved with St. Francis Hospital the nearby Kibaoni Health Center, and Morogoro Regional Hospital, having implemented program components at each of these focused on obstetric emergency care and PMT CT (Prevention of Mother to Child Transmission of HIV). Cervical cancer prevention was the next step. Jhpiego provided the training for screening providers, as well as supplies, and administers funding for the program. An early advocate for the one visit screen and treat approach using VIA, Jhpiego in Dar es Salaam.

PAGE 75

59 Since turning their focus to cervical cancer in Tanzania, Jhpiego has taken a multi pronged approach. In addition to supporti ng 21 cervical cancer prevention programs sites around the country, they have also chosen to partner with Tanzanian government agencies to provide technical support, education and training, and a variety of advocacy tools. Some of that technical support is being directed towards an effort by the MOHSW to correct the lack of national service delivery guidelines noted in the previous section ( Jhpiego, 2015 ) While the cervical cancer prevention program does target one disease, Jhpiego has attempted to build on this approach by incorporating th e MOHSW guidelines calling for Provider Initiated HIV Testing and C ounseling (PITC) into their services. The program intake forms include a section detailing w hether or not PITC was offere d and accepted, and if so, its result (HIV status). They encourage reciprocal referral systems between HIV testing and cervical cancer screening as well, and encourage program staff to build referral relationships with other clinics with in the same institu tions The CECAP program at St. Francis provide single visit VIA and cryotherapy, but women with more advanced cervical lesions or invasive cancer cannot be treated there. Certain central sites were given the training and resources to provide LEEP for lar ger lesions. For women in Ifakara, the closest site was Morogoro, a half day trip by bus. Invasive cancer cases were treated in Dar es Salaam at Ocean Road. When I first arrived, St. Francis could not even provide biopsies, though they had a surgical theat re availabl e in the hospital and a staff member capable of conducting the procedure, because the program did not have its own punch biopsy device. This meant women had to travel to Dar es Salaam once for a biopsy, and then return if the results indicated i t was necessary.

PAGE 76

60 Eventually program staff appealed to Jhpiego and obtained their own device. But the biopsy then had to be sent to Dar es Salaam, typically by bus, for analysis, the results of which were transmitted by phone and email back to staff at St. Francis before being conveyed to the patient. existing CTC or health care facilities in partnership with the Tanzanian government was not for them to remain independent units i ndefinitely, but rather for the services to eventually be integrated into the health care system after a period of oversight. Jhpiego provided this oversight for the first four years of the program at St. Francis, and funding was provided by USAID. However in 2014, Jhpiego planned to turn over the administration of the program to the government, who would also take responsibility for its financing. This dissertation focuses on one cervical cancer prevention progr am in one hospital in Tanzania, yet the impl ementation and utilization of this program have broader implications, particularly given that CECAP was standing on the threshold of fuller integration into the overall hospital structure and health care system. The next three chapters examine the issue of utilization from the perspective of the clients; in Chapter VIII experiences over its first three years.

PAGE 77

61 CHAPTER V LOCAL CONSTRUCTIONS OF CERVICAL CANCER Ummi Ummi was the first woman we spoke with in the Gynecology Department, and she came into the room a little shyly, hesitating on the threshold and looking around as her eyes adjusted to the shadowed room. We had closed the curtains for privacy since the room faced a courtyard between two wards and as usual, the overhead light was off A married mother of two, Ummi worked as a nursery school teacher. She had traveled to St. Francis from her home in a village near the Udzungwa Mountains National Park, approximately 35 miles away. In a hoarse, quiet voice, she told us that in addition to hearing radio and television announcements about cervical cancer, she had just that day seen an educational poster in one of the disease warning women that when or experienced some changes that were not ordinary, they should hurry to a related to sexual intercourse and that having sex at a younger age or with many partners were parti cularly problematic. We asked her who was likely to get cervical cancer, and in her response she reiterated this belief: Women. Women, starting from a child who has already matured. According to the way I understand, I think sexual intercourse, especially to girls, or unsafe sex. Also the number of pregnancies, as one may have many deliveries that are consecutive. This may cause the uterus to get loose, or to get certain diseases that get into the uterus. Ummi placed great stock in the power of hospital med icine, telling The brief health history she gave us bore that out, as she recounted several different health concerns that had led her to the clinic. As a Muslim, when sick she would pray to God for the strength to get to the clinic, but, she told sure that when I get to a health c enter and consult a health attendant, I will the individual to recognize when they were sick and seek care, and then to follow the resulting medical instructions. In the case of cerv ical cancer, she shared what a woman should look out for and what would happen if she chose to ignore those symptoms: According to the way I understand, I think a person will feel low abdominal pain and also may have frequent urination or may be unable to control the passing of urine so the urine cancer is going to be reluctant to go to the place where they

PAGE 78

62 are supposed to, I know at the end that the uterus will rot and it will have to be removed. And th is person will not be able to have children again. So it is important to go quickly to a health center so that they get immediate treatment. I was curious what her response would be when we asked her if she had herself been screened in the past, given that she had heard a fair amount Maybe it was done to me, but I did not know the name if it was cervical cancer screening. Because I have been cleaned 1 had a child, and in the year 1994 I had a child. I have not been cleaned again since then but I have just had an operation for the appendix and am now on treatment for a swelling. Did you know what was being done to you? At that time I did not know becaus e I was only told that my reproductive tubes had blocked. Ummi was only our third interview participant; my surprise at her response later faded as I came to understand that it was not uncommon for many of the women we interviewed to be unsure as to their diagnosis or the specific medical procedures they had undergone. Sometimes this was due to a forgetful memory or an unclear understanding, other times they said that of her screening h istory, she was very interested in being screened, asking me to tell her how to go about it, and said that she would have done it sooner if she lived near to St. Francis. She pointed out that screening was the only this is a disease that is inside, so even when I get abdominal pain, I will not kn Understanding Cervical Cancer in the Local Context According to a w estern biomedical understanding cervical cancer has a fairly clear etiol ogy. That risk for the disease is related to sexual activity has been postured by physicians since at least the nineteenth century, and was confirmed with the identification of HPV as a necessary cause in the 1990s ( Bosch et al., 1995 ; Lwy, 2011 ; Walboomers et al., 1999 ) Early age at first intercourse, multipl e sexual partners, sex with a partner 1 in on the intervi ew and several informants, this term is most likely referring to curettage, which may be a veiled reference to abortion in this particular context.

PAGE 79

63 who has had multiple partners, and frequent sex with an infected partner all increase a ( Franco et al., 2003 ; Krishnan, 2008 ) Furthermore, HPV is spread through skin contact, which means that condoms are not as effective in preventing its transmission as they are f or infections spread through bodily fluids ( Center s for Disease Control, 2004 ; Krishnan, 2008 ) Once a woman is infected with HPV, there are other factors that increase the likelihood the infection will become persistent and develop into cancer. High parity, for example, has been identified as a co factor f or cervical cancer due to repeated trauma to the cervix during labor. Poor hygiene and nutrition, tobacco use, oral contraceptive use, infection with other STIs, and HIV infection are all factors that may play a role ( Castellsagu & Muoz, 2003 ; Trottier & Franco, 2006 ) Finally, social and cultural practices and norms, as discussed in this chapter, may influence the degree to which these risk factors are a concern in specific settings. The symptoms of cervi cal cancer are also known, though they may be ambiguous in presentation. Women may not experience any symptoms at all during early stages of the disease ( Castellsagu, 2008 ; Krishnan, 2008 ) This is problematic, as it is well established that detection and treatment of the disease in its early stages can dramatically improve survival among cervical cancer patients ( Castellsagu et al., 2012 ) Adding to these difficulties, s ome symptoms, such as blood spotting, bleeding between menst rual cycles, and vaginal discharge may mimic other sexually transmitted infections or fall within the range of what women consider normal for their own bodies. Yet b iomedical explanations do not always correspond to local understandings of illness, which can in turn influence whether or not individuals (and those around them)

PAGE 80

64 medical care they seek out For this reason, medical anthropologists typically draw a distinc tion between disease used to denote a biomedical condition, and illness, indicating 5). Illness therefore involves not only the types of physical sensations that an individual or group qualify as cause for concern, but also how a person with those symptoms should behave, the potential cause of those symptoms, and the type of treatment or practitioner that should be consulted ( Kleinman, 1988 ; Kleinman et al., 2006 ) This relationship, between a biomedically defined condition and the social context in which it sits is reciprocal and should be exa mined holistically ( Jordan, 1983 ) The questions that elicited the responses Ummi gave us in the opening of this chapter arose from my desire to better understand how women in rural Tanzania conceptualize and assign meaning to cervical cancer. Lakoff (1987 ) understa nding of how we categorize is central to any understanding of how we think and p. 6). While classification is an activity practiced by all human cultures, local systems of categorization have important implications for health and illness ( Kleinman, 1988 ; Nichter, 2008 ) For example, the symptoms that women identify as related to cervical can cer, or at the very least, identify as potentially problematic, are important to know because they shape care seeking behavior, as do those behaviors or traits that are seen to put women at risk of cervical cancer in the first place ( Bradley et al., 2006 ; Chirwa et al., 2010 ) Furthermore, cultural views on treatment and prognosis may impact whether or not women view screening as a benefit ( Bingham et al., 2003 ; Drew & Schoenberg, 2011 ) which will be explored further in Chapter s VI and VII This chapter

PAGE 81

65 examines how cervical cancer is conceived of and understood among the partici pants we interviewed. Chapter VII will then explore how this conceptualization interacts with the challenges women face in seeking screening and treatment services in rural Tanzania, and the benefits they perceive to come from screening. Accessing Familiar Biomedical Models of Infection Very basic biomedical ideas regarding infection we re familiar to m ost Tanzanian women I met given decades of pub lic health campaigns around common infectious or communicable conditions such as malaria and HIV. In my own daily life, I was frequently warned to take care regarding potential microbes or parasites in unsafe food and water, and whenever I felt the slight est bit ill, regardless of where in my body the illness was centered, I was advised to go to the clinic to check for malaria parasites. Because I was a mzungu, I was warned, I was even more susceptible because my body d no built up resistance. Furthermore I conducted this research among a clinical pop ulation who largely expressed familiarity with the health care system as well as with the importance of hygiene, a good diet, and isk, whether that protection came in the form of mosquito nets, condom use, or clean water. Thus it was not surprising that many of the participants we interviewed from both CECAP and the GD communicated a general underst anding of the idea that infectious agents could put an individual at risk for cervical cancer and certain activities increased that risk. This knowledge base was typically shallow however, and lacking in specific details.

PAGE 82

66 We opened our discussions about cervical cancer simply by asking wo men what they could tell us about cervical cancer, or saratani ya shingo ya kizazi 2 but this was rarely enough to elicit responses about the actual cause. The things that first came to mind for participants were radio announcements they had heard, or symp toms they knew, or personal anecdotes about family members and neighbors with cancer. When we asked about the specific cause of cervical cancer, nearly half ( n= 22) of the women we interviewed told us they did not know 3 The matter of fact response of Jesca a 30 year But t hose who did have a response drew on generalized biomedical terms and their familiarity with other medical conditions (particularly those they saw as related in some way) in describing the causes, symptoms, and risk factors of cervical cancer. Particularly common were 2 Saratani is the Kiswahili word for cancer (sometimes also called kansa ). Shingo ya kizazi he womb/uterus mlango wa kizazi interchangeably to refer to cervical cancer, however, for the sake of consistency, we used saratani ya shingo y a kizazi in all study documents. 3 The Patient Information Forms did not collect data on education level, so it is difficult to draw definitive conclusions as to whether these individuals had less education than those who were able to share their knowledge of the causes of cervical cancer. Using farmers or tailors, occupations that do not require formal education, and the 19 th was a self described business woman (an ambi guous term in this setting where that may mean anything from a formal professional position to a person who cooks food at home to sell on the street). Additionally, one was unemployed, and another worked with the local office of an HIV non profit, but she had been recruited due to her HIV status and local knowledge rather than a high educational background. The last was a secondary school teacher. In contrast, while the group of participants who did respond affirmatively to this question included many farme rs as well, and four businesswomen, it also included seven teachers and health care workers.

PAGE 83

67 descriptions of the risks posed by sexual activity and childbirth. In fact, i t wa s difficult to understanding of the biological causes of cervical cancer from their recognition of the risk factors as they were tightly wound t ogether during our interviews. Infections, Microbes, and other Causes of Cervical Cancer The most commonly cited causes of cervical cancer drew upon models of disease causation pulled from biomedicine but without specifically referencing human papillomav irus or indeed stating that one specific virus was responsible for the disease. This is in keeping with previous research from both Tanzania and other parts of sub Saharan Africa showing that knowledge of cervical cancer is typically quite low in local pop ulations ( Abotchie & Shokar, 2009 ; Kidanto et al., 2002 ; Mutyaba, Faxelid, Mirembe, & Weiderpass, 2007 ; Ports, Reddy, & Rameshbabu, 2013 ) Even those participants employed as health care workers including two nurses and a ward attendant did not make this type of reference though the ward attendant, Hellena, did outline a model of intercourse or by having multiple sexual partners. So the pathogens are transferred from me While we interviewed most CECAP women before they sat through the educational session that opened each screening day, I initially thought (incorrectly) that perhaps those who were interviewed afterwards would be able to be more specific. A fter observing a number of these sessions, however, I learned that the staff typically only referenced HPV by name once at the beginning of each talk.

PAGE 84

68 Thereafter they commonly used the non specific word dudu 4 with the clients to describe the virus. I nstead, the small number of women (four from CECAP and four from the GD) who specifically referred to a microbial cause of cervical cancer used generalized terms, discussing pathogens and microbes as commonly causing infection rat her than a s pecific virus Salome a 52 year old CECAP client who lived nearby and had heard from several women she knew about the screening program, came perhaps the closest when she told certain bacteria attack a cer tain part of went so far as to imply that one specific type of pathogen was at fault. Three of these eight women specifically stated that these disease agents were deposi ted by men during sexual intercourse, indicating an awareness of the sexual transmission of the disease. They were not alone in linking cervical cancer to sexual activity, however, most women framed this in terms of risk rather than causation. O nly eight w omen mentioned infectious agents like pathogens or microbes, but ma ny more indirectly referenced biomedical model s of i nfectious disease. However, these models were interpreted through a cultural lens in which certain activities, conditions, or injuries we re seen to place the body at risk and to result in certain symptoms Medical anthropologists have often pointed out that biomedical diseases can rarely be divorced from the social and cultural contexts in which they exist when 4 Dudu is the Kiswahili word for insect or bug (plural: madudu ), and a form of the word is also used for bacteria/germs, vijidudu There is a Kiswahili word for v irus, virusi however during my time at CECAP the word dudu was typically used with clients instead

PAGE 85

69 attempting to understand the meanings attached to them ( Jordan, 1983 ; Kleinman, 1988 ; Lock & Scheper Hughes, 1990 ) Kleinman (1988 ) to explain how individuals suffering from illness made sense of their experiences. These lically 49). In this study, none of our participants had cervic al cancer, but they were asked to discuss the ir perceptions of the disease p erceptions that were influenced by their experiences with other medical conditions. The models proposed by participants were principally influenced by their knowledge of other sexu ally transmitted or gynecological conditions, the HIV pandemic and high rates of complications from childbirth in Tanzania and will be discussed further in the next section While alcohol and tobacco use are known to increase HPV infection will develop into cervical cancer ( Sussman et al., 2007 ; Trottier & Franco, 2006 ) only one participant spoke of this connection as a potential cause of the disease. Tobacco us e among women in Tanzania is quite low according to the limited data available, however it is increasing in the developing world in general ( Sussman et al., 2007 ; WHO/ICO, 2010 ) Among this population at least, it did not appear to be associated with cervical cancer either as a cause or a risk factor in the same way it might be for other cancers. Within Tanzania and elsewhere in sub Saharan Africa, witchcraft ha s been included in lay models for other reproductive and sexual health issues, such as infertility or miscarriage ( Kielman, 1998 ; Okonofua, 2007 ; Plummer et al., 2011 ; Upton, 2001 ) as well as for cervical cancer and its symptoms ( Awodele et al., 2011 ; van Schalkwyk, Maree, & Wright, 2008 ; White et al., 201 2 ) In our interviews, only five woman (three

PAGE 86

70 CECAP clients and two GD patients) related bewitchment directly to cervical cancer E ach did so not as a matter of conviction, but as an example of what someone else in their community might believe to be th e cause of their ailment. Sandra, for example, told may claim to be bewitched, while as a matter of fact they were supposed to go to the hospital for further consulta as causing cervical cancer specifically; rather individuals who believed they had been bewitched blamed their symptoms on this rather than recognizing that it was cancer. Sandra herself had stru ggled with repeated miscarriages, and so had been referred to CECAP for screening. She believed that seeking care at the clinic was the correct course of action when she or her children were taken ill; she claimed not to use traditional medicine either. It is difficult to say for certain whether the observations Sandra and her four fellows made regarding other members of their communities highlighted widespread beliefs that were simply underrepresented in this clinical population, or if they were in fact un conventional beliefs. Presumably, women blaming their illness symptoms on witchcraft would be less likely to seek cervical cancer screening or other health services, choosing instead to seek out traditional healers or practitioners. Furthermore, while wom en in both CECAP and the GD were quick to tell us that they did not themselves use traditional medicine or believe in witchcraft, both my RAs and AL felt strongl y (based on their own experiences in the hospital and as members of the community) that at leas t some were simply reluctant to discuss the matter in a clinical setting where they thought such beliefs would be disapproved of. As far back as 1985, Alexander observed that Tanzanian cancer patients likely down played their use of

PAGE 87

71 traditional medicine or self treatment when in a hospital setting in deference to the medical personnel. An informant of mine in the CTC clinic also told me that many of those they see for HIV use traditional medicine and that if they are asked directly the patients will admit t o using it. He attributed this to the trust the CTC clinic had built up with their patient base and the non judgmental attitude they try to portray. Given this, it is likely that there were amongst our interview participants those who used traditional medi cine more frequently but were unwilling to discuss it with us. Before I began my interviews, Dr. M and I had discussed some of the knowledge barriers around cervical cancer. Dr. M felt that cultural concerns a bout how cervical cancer screening and treatme nt might affect fertility were a deterrent to women coming in for screening. Social values and concerns beyond the mere biological are often e mbedded in the way certain illnesses are viewed, in the meanings ascribed to them and the risks they are seen to p ose ( S etel, 1999 ; Sontag, 1977 ) In much of sub Saharan Africa, despite changing trends in social practices including increasing use of contraception and delaying marriage and procreation, fertility is still highly valued and the role of mother an ( Hollos & Larsen, 2008 ; ICF International, 2004 2012 ; Plummer et al., 2011 ; Setel, 1999 ) Even those who do not currently wish to become pregnant ma y desire to safeguard their long term fertility and therefore may be apprehensive about undergoing a procedure they may view as jeopardizing this ( Castle, 2003 ) Therefore, any threat to fertility is likely to be a cause for concern and infertility an important re ason to seek medical care, particularly if a woman has not conceived within the first year or two of marriage ( Plummer et al., 2011 )

PAGE 88

72 None of th e women we interviewed in either group mentioned a fear that the screening process itself might interfere with their fertility. In fact it was rather the opposite: Many of the women who came to CECAP were referred there to rule out cervical cancer as a ca community or is a deterrent to screening may very well be true, but this fear was not Yet despite this specific concern not being mentioned by participa nts, threads of cultural importance placed on fertility wove through the discourse on cervical cancer while I was in the field. As mentioned, several participants thought infertility was associated with cervical cancer itself, and had come to determine whe ther their own struggles to conceive could be due to the disease. Two women also believed that famil y planning drugs, which have sometimes themselves been viewed as a threat to long term fertility could cause cervical cancer. This belief has also been not ed in other studies and is concerning given the already low rates of contraceptive use in Tanzania ( Anasel & Mlinga, 2014 ; ICF International 2004 2012 ) Finally, treatment was known to potentially invol ve removal of the womb, which would of course cause infertility; avoiding this possibility was seen as a reason to go for screening. Although not related to screening, I also anecdotally heard from health care providers and researchers that some women wer e concerned the HPV vaccine might impact fertility. Of cour se, the belief that vaccines inhibit fertility or in fact cause sterilization has a long and unfortunate history hindering immunization prog rams in some developing nations (e.g., Feldman Savelsberg, Ndonko, & Schmidt Ehry, 2000 ; Renne, 2006 ) We did not ask questions about the vaccine in this s tudy, but if this is indeed a concern dissuading Tanzanian women from vaccination, it could be a quite potent one.

PAGE 89

73 programs among schoolgirls to broader areas of the coun try. Perceived Symptomology The great promise of cervical cancer screening lies in its ability to detect the disease in its earliest stages, when the chances for survival are highest and treatment is relatively simple; it seeks to prevent cervical lesions from becoming invasive cancer. However, preventive care is not standard practice for most Tanzanians. For example, Kidanto et al. (2002 ) found that a majority of women in their clinic based sample believed hospitals could successfully treat cervical cancer while a traditional healer coul d not, results replicated in this study. Yet they also felt that routine gynecological visits were unnecessary. In Dar es Salaam, 31% of women surveyed expressed reluctance at the idea of a preventive test when they were otherwise in good health ( Kahesa et al., 2012 ) Similar findings have been found in developing countries in Central and South America ( Agurto, Bishop, Sanchez, Betancourt, & Robles, 2004 ) and in sub Saharan Africa ( Bingham et al., 2003 ) ; women often do not go to the clinic when they are symptom free, either because they do not see the need or do not have the resources for preventive care. This is even more problematic when one considers that cervical cancer is often asymptomatic in it s early more treatable stages When symptoms do occur, they often mimic other gynecological conditions. M any Tanzanian health facilities lack the necessary laboratory equipment and staff to perform standard of care clinical tests to rule out other STIs o r reproductive tract infections (RTIs) Instead, national guidelines call for a syndromic approach when laboratory tests are not possible, in which the focus is first on the clinical presentation or

PAGE 90

74 th en follows a flow chart to determine the likely disorder and appropriate treatment ( Lyimo & Beran, 2012 ; Ministry of Health and Social Welfare, 2013 ) There are no lab test results to receive in these cases and patients are not always given their specific diagnosis before a treatment is prescribed. As a result, women may not perceive themselves to be at risk for cervical cancer if they believe they are being treated for an STI or other RTI. Cervical cancer may not even enter into their discussions with health care pr oviders until and unless trea tment for a different condition fails and they continue to experience symptoms. In our interviews, women descriptions of cervical cancer symptomology were related to and overlapped with their understanding of symptoms commonl y associated with sexual or gynecological conditions like HIV RTIs, and STIs. When asked about the symptoms of cervical cancer, women provid ed a broad range of potential indicators that were cause for concern, including vaginal discharge, vaginal itching, bad smells in the genital area, abdominal pain, painful intercourse, pain during menstruation or urination, irregula r bleeding, swelling or tumor, frequent urination, rashes, lack of sexual desire, headache, troubles with fertility, diarrhea, weight loss, general poor health or weakness, u rinary or fecal incontinence, and Of these, the three most frequently mentioned were vaginal discharge ( n= 10), pain of some type ( n= 24, abdominal pain or stomachache being the most common responses), and irregular bleeding ( n= 19), all of which are in fact potential symptoms of cervical cancer. Women noted that a watery or brown discharge was worrisome as well as that with an offensive smell. Pain that potential ly indicated cervical cancer included abdominal pain or stomacha che, pain during intercourse, and pain in the genital area.

PAGE 91

75 Irregular bleeding encompassed anything from non stop bleeding, to bleeding outside of menses, to passing blood clots, to bleeding after menopau se. Eight participants were also sensitive to the fact that a swelling, lump, or tumor in the abdominal cavity or in/on the genital area could possibly indicate cancer (with one additional woman making the same observation about a lump in the breast). In o ne concerning interview, a woman told us that she could feel a painful mass when she inserted a finger into her va ginal cavity, while Hellena, a during washing their genitals, they may see someth ing is protruding from the 5 However, participants also cited other symptoms, such as rash, itching, fever, headache, frequent urination or pain during urination, fecal incontinence, even a mucus disch arge that are more likely to indicate other sexually transmitted infections or reproductive conditions such as obstetric fistula, rather than cervical cancer 6 A few mentioned infertility as a potential symptom or outcome of cervical cancer Though the di sease is not the most likely cause of infertility in this setting, they were correct that these could be associated ( Ombelet, Cooke, Dyer, Serour, & Devroey, 2008 ) I n fact many of the women I encountered at CECAP had been referred there from the GD to rule 5 Despite Hellena sharing her anecdote in response to a question on cervical cancer symptomology, what she was referencing was in fact likely vaginal prola pse and not a tumor resulting from cervical cancer; however, in a setting where women often do not seek care until their cases are quite advanced, very large tumors may in fact develop before a woman is diagnosed with cancer. I did have occasion to see, wh ile observing in the clinic, one woman come into CECAP with a genital mass that had grown to the size of a small avocado, which hung down between her legs. In th at particular case though, the mass originated on the vulva, and not internally on the cervix. 6 Though urinary and fecal incontinence can occur in very advanced stages of cervical cancer, particularly if a fistula develops as a result of the cancer.

PAGE 92

76 out cervical cancer as a reason for their difficulty conceiving. Weakness, weight l oss, and a feeling that things we re not normal are fairly vague and could be true of a number of diseases; in this context the first two were frequently tied to HIV as well Despite their potential ambiguity, w omen viewed all of these as s ymptoms to take seriously, and certainly reasons to seek medical care. That cervical cancer was thought to share symptoms with these other conditions RTIs/ STIs, HIV, and to a le sser degree, fistula or other birth complications tells us something about where women placed cervical cancer within the constellation of health concerns. Cervical cancer, in terms of causes, risk s and symptomology, was largely discussed as an issue of se xual and reproductive health, tied to other conditions that affected that area of the body, rather than in relation to other cancers, their causes, risk factors, and symptoms (the exception being the mention of tumors and tumor growth). Explanations of Ri sk Health and illness must be examined with an eye not only towards the physiological basis of the disease, but also towards the larger social and cultural context in which it exists and the meanings assigned to it ( Douglas & Wildavsky, 1982 ; Geertz, 1985 ; Jordan, 1983 ) I noted earlier that in regard to cervic lay models included conceptualizations of disease causation and ri sk that were bound together. These models also reflect ed concerns with larger social processes. Medical anthropologists and other scholars have long pointed to the usefuln ess of metaphor and meaning making in describing and understanding illness ( Douglas, 1970 ; Farmer & Kleinman, 1 989 ; Scheper Hughes & Lock, 1987 ; Sontag, 1977 ) Here I will explore how women called on

PAGE 93

77 what they knew about infections, both sexually tran smitted and otherwise, to explain the risk of contracting or developing cervical cancer and in so doing, illuminate larger cultural concerns a round sexuality and childbirth Sexual Behavior In Tanzania, the HIV pandemic and high rates of STIs have contribu ted to the ( Hanson, 2000 ; Outwater et al., 2001 ; Setel, 1999 ) This is reinforced by the information people have heard repeatedly in public health campaigns and from medical personnel about the risks of unprotected sex and of having multiple partners. Gi ven th at a large proportion of CECAP HIV positive (31.4% during the period from May 2010 October 2013) and that many had been referred from CTC where they did health education on the disease, I expected to find that many in the CECAP group would be familiar with the risks posed by sexual activity. But i n fact, most of the women we spoke to, in both groups of participants, echoed these concerns. As Hanson has noted, despite the fact that STI public education efforts in Tanzania have often con centrated on broader national media campaigns rather 344) The most risky activity, according to the women we intervie wed was having multiple sexual partners. Although we were told this specifically in regards to questions about cervical cancer risk, the repetition of this theme during discussions about protecting being in a broader sense. Repeatedly women told my RAs and me

PAGE 94

78 Having multiple partners was a well known risk factor for HIV and other STIs, and Shazia for one, a Community Health Worker (CHW) told us she had actually heard on the radio that cervical cancer resulted from STIs. However, in most cases, it was less clear whether participants knew that this was specificall y a risk factor for cervical cancer or rather were making an intuitive leap that non monogamy was a risk factor for some sexually transmitted infections, and since cervical cancer impacted the reproductive system, it likely was a risk factor for that disea se as well. Some actually verbalized this thought, including n of the cervical, if that is Of course, as Setel (1999 ) points out, more important to the spread of HIV and other STIs is unprotected sex with multiple partner s. While in reality condoms are not as effective in preventing the transmission of HPV as they are with other STIs, they still provide some protection ( Castellsagu, 2008 ; Center s for Disease Control 2007 ) Along these lines, women correctly believed that unsafe or unprotected sex and prostitution As Dinah, a young woman of 25 who had had seven partners in as many years (the second highest of any in either CECAP or goes, umdhaniaye siye kumbe ndiye [ never judge by looks ] specific pathogen, w omen instead focused on the se behaviors themselves as b oth a risk factor and a cause of cervical cancer. Abortion and rape were only mentioned by two participants and one participant respectively, and were not typically part of discussions about risk.

PAGE 95

79 Participants were particularly concerned about sexual activ ity as a young woman. This concern was not necessarily tied to a belief in a higher physiological risk of infection but rather due to a period of time when specific behaviors or a specific type of ladies, or some of us are still young ladies, and there are those who still want to enjoy life, but after some time has gone, they may decide that it is enough of the starehe and yet be able to protect their As used in this context, the term star ehe can have two meanings, both of which were implied by different participants at different times. The first is its more literal meaning, which is to enjoy oneself or to live luxuriously, the implication when used to refer to sexual relationships being th at young women more interested in (or in need of) material things may seek out relationships with wealthier men that involve gift giving or transactional sex ( Haram, 2005 ; Wamoyi, Wight, Plummer, Mshana, & Ross, 2010 ) Starehe can also refer to casual, short term (sometimes one time) affairs that may ip ( Setel, 1999 ) According to Setel, who studied these types of relationships in northern Tanzania, starehe relationships ar e typically brief and sexual activity is the focus. He notes that young people would typically differentiate between a starehe partner and a boyfriend or girlfriend, with whom they had a relationship outside of sexual relations, though others have question ed the strict typology he uses to define this category of relationships ( Wamoyi et al., 2010 ) Although a starehe relationship can develop into a long term relationship, desire is a more important component at its initiation than tabia (good character) ( Setel, 1999 )

PAGE 96

80 In either usage of starehe, fidelity is not presumed, particularly on the part of the man ( Haram, 2005 ; Setel, 1999 ) These types of relationships therefore pose risks both to rding to Marjani, syphilis that would contribute t o the disease, while Jesca shared how her mother had warned her against her wild behavior when she was younger (she was 3 0 when current health struggle s. Sexuality and fidelity expectations are highly ge ndered in Tanzania, and anxiety around changing social norms related to sexual behavior and gender roles has been noted repeatedly, particularly when it comes to young women ( Dilger, 2003 ; Haram, 2005 ; Setel, 1999 ) And regarding cervical cancer risk there is reason to be concerned. While both male and female Tan zanian youth tend to experience their first sexual encounter at approximately the same age, females are more likely to have this experience with a significantly older partner, which is a risk factor for the disease because an older partner typically has ha d a number of previous partners ( Urassa, Moshiro, Chalamilla, Mhalu, & San dstrom, 2008 ) Childbirth disease that affected the reproductive system, it is perhaps not surprising that their notions of risk centered on activities tied to their reproduc tive organs: sexual activity, as outlined above, and childbirth. Yes, sexual activity could introduce infection into the body, but so too could internal wounds resulting from childbirth or other reproductive

PAGE 97

81 medical care. As an opening to the body, the vag inal canal was an entry point for disease during both disease, as well as childbirth. In reality, childbirth is a dangerous process for many women in Tanzania, particularly rural wo men. In a nation where barely 50% of women give birth in a health facility, maternal mortality is high and birth complications frequent 7 High parity is a biomedical risk factor for cervical cancer, and in Tanzania, where the total fertility rate is 5.4 ( I CF International, 2004 2012 ) and cervical cancer incidence is likewise high among young and middle aged women specifically the perception that having many children can lead to cervical cancer is both understandable and true. While most women could not explain the actual mechanism through which this risk materialized, giving birth likelihood a woman would devel op cervical cancer. Debora, who had given birth to two children herself, gave a response that was illustrative of this tension between belief and his cancer is caused by having very many children or very close How one became pregnant, for instance, if one conceived within marriage or without, or if one concei ved several child from multiple partners, was not a part of the risk discussion. 7 Maternal mortality is 454 out of every 100,000 live births ( ICF Interna tional, 2004 2012 )

PAGE 98

82 Women across the age and parity spectrum of those we interviewed, from both CECAP and the GD, reiterated the idea that frequent or numerou s births were risky. However, though a biomedical understanding of cervical cancer tells us that it begins with HPV infection of the cervix, the organ of most concern to participants here was in fact the womb or uterus 8 (though this may be because women l acked an understanding of the cervix as a separate entity ; even the Kiswahili name for the cervix, which translates to ). The actual process of childbirth, women shared, could ca use internal wounds or lacerations through which microorganisms could enter and cause infection. Likewise, metal tools used during delivery, abortions, or miscarriages might scrape the internal organs 9 making them more susceptible to disease. These wounds or scratches provided a vulnerable point of entry to through frequent or numerous deliveries, a condition that, again, made that organ susceptible to disease, including c ervical cancer. Fears of lacerations or wounds during labor, taken in conjunction with some of the symptoms thought to be associated with cervical cancer in the previous section (such s with obstetric fistula 10 and other birth complications. Approximately half of all women in Tanzania give 8 The uterus was often referred to as the mfuko wa uzazi (reproductive bag) or simply mfuko (bag), rather than the more formal uterasi 9 This may in fact relate to a concern voiced by a few women regarding the use of the speculum in conducting cervical cancer screenings discussed in a later chapter, though no participant specifically stated they personally were afraid of lacerations from the device. 10 for lapse in blood supply caused by this pressure can cause necrosis and a hole may form either between the vaginal canal

PAGE 99

83 birth at home or outside of a health center, a situation due in part to the long distances and poor roads women must travel in rural areas to reach he alth care. Those who do reach health centers often find they do not receive adequate care ( Mselle et al., 2013 ) As a result, prolonged labor and complication s may arise and obstetric fistula result. M arjani told us a story of how she had been warned about this during a previous cervical cancer screening experience (other types of health information are sometimes included in CECAP information sessions, particularly if a woman asks a question about a spe cific topic or symptom), and she appeared to have integrated this information with what she had learned about cervical cancer: What causes cervical cancer? lesson that having many sexual p artners can also lead to its transmission, abortions, and home deliveries. They told us many things, that you may deliver at home using a lot of energy and get a laceration without knowing, and so continue staying at home. This can also cause some problems Women utilized their knowledge and experiences of birth complications to construct an explanatory lay model of cervical cancer risk that included childbirth. These models accessed concepts and language used in biomedical explanations of infection, in ad injured wombs. No doubt at least some of these women had heard that high parity was a risk factor for cervical cancer from a third party, whether a poster, radio a nnouncement or provider. But they interpreted this basic piece of information through the lens of a and the bladder, or the vaginal canal and the rectum ( Mselle, Moland, Mvungi, Evjen Olsen, & Kohi, 2013 )

PAGE 100

84 reality in which childbirth is, despite its common occurrence, a dangerous journey for many women. Prevention and Prognosis In keeping with previous resear c h in Tanzania ( Kidanto et al., 2002 ; Lyimo & Beran, 2012 ; Peters et al., 2010 ) the women we spoke with did not have deep or specific knowledge of cervical cancer as a disease, but many understood that it could be sexually transmitted and that certain sexual behaviors and childbirth played a role in risk for the disease. However, they struggled to name specific tactics to prevent cervical cancer, perhaps because the primary risk factors they perceived sexual activity and childbirth were such a common part of daily life Shany, a young woman in the GD, for exam ple, knew some of the risk factors, but really believed all women were at risk: Any woman can get the disease because they have many sexual partners or give birth to many children, maybe more than five times. On a quick look, every woman would like to have children in their lifetime, as well as to have a sexual partner, and so women are prone to get the disease. Gregg found such attitudes in the Brazilian favela as well where women felt that sexual could actually assume (2003, p. 96). In that case, women had limited agency in their sexual relationships and so chose instead to view the Pap smear, or more specifically, the lack of one, as the factor that increased their risk for cervica l cancer. At St. Francis, the women we spoke with could be divided into two types be importance of the education she does in schools as a Ward Attendant, acknowledged that

PAGE 101

85 there was only so much they could do without also educating men and changing their the HPV vaccines already in existence, a fact that may change with increasing media cover following the success of similar programs in Rwanda 11 T he second group, the majority in fact, thought that something could be done, but their advice essentially boiled down to the belief that medical doctors and facilities had the resources to address difficult illnesses rather than actionable preventive measures. Four participants among the GD patients suggested that women should generally avoid risky behaviors such as having many sexual partners and giving birth at a young age, while two others focused on hygiene and cleanliness, but most put their faith in the information, guidance, and treatment available from doctors and health centers: Yes, if you go to the hospital and ex plain what you feel to the doctors, they will give you medication because they understand the diseases very well and you will be able to prevent it. [Leta] I think there is nothing that can be done other than being close to the doctors and following what you are being instructed. [Zuwena] How will you prevent it while at home you have no medications which are suitable for you to take?...if she comes then the doctors know how to discover the problems, and when given the dose, everything would be so smooth. [ Zuhura ] We did not specify in our question whether we were asking about primary prevention or secondary prevention, but 11 Two women from the GD clinic did propose the development of a vaccine, but they were ignorant of the fact that several already exist ed.

PAGE 102

86 it is interesting that women were largely focused on secondary (or even tertiary) prevention. Among women who felt cervical cancer could be prevented six (evenly split between the GD and CECAP groups) emphasized the importance of early detection and treatment in preventing cervical cancer specifically. The others gave general responses similar to the answers they gave to questions about maintaining health more broadly It may be that in the absence of specific knowledge about prevention, they fell back to the default action of seeking medical advice. While the women in this st udy more typically associated cervical cancer with sexual and reproductive conditions when it came to risk and symptomology, this was not entirely true of prognosis. One of the challenges cervical cancer screening programs have to overcome is the widesprea cultures and nations, cancer is often thought of as incurable or nearly always fatal ( Bingham, Drake, & La Montagne, 2009 ; Dzuba et al., 2005 ; Royse & Dignan, 2011 ; Sontag, 1977 ) Of co urse cervical cancer i s unusual in that early detection is extremely effective at preventing mortality, but it can be difficult to emphasize that cancer of the particularly when dealing with populations who may not make a practice of distinguishing between different types ( Dzuba et al., 2005 ) The concept of fatalism and its influence on health care seeking behavior is therefore important to consider ( Drew & Schoenberg, 2011 ; Hanson, 2000 ) Fatalism, in this arena, refers to the belief that an individual does not possess power or control over their own health; rather it is up to fate or luck.

PAGE 103

87 Straughan and Seow (1998 ) willingness to undergo certain cancer screening tests (including the Pap s mear; their research preceded VIA) through the mechanism of self efficacy. There was certainly a thread of fatalism running through the discussions we had with women at St. Francis, however, it did not manifest itself verbally in a desire to avoid screening. Women largely expressed to us that screening was a good idea, even if no symptoms were currently present. It may be that some of the women in the GD who had never been screened were avoiding screening in part due to this type of fear, but if they felt this way, they did not say so. There were some women ( n= 12) who viewed cervical canc er as a death sentence. A woman might undergo treatment, it might even extend her life, but eventually she would die. For those like Rebekah, this pronouncement was based on their own experience watching a relative or community member diagnosed with breast or cervical cancer: The prognosis is poor. This is due to the fact that the disease is incurable. I am saying this as I had an experience with my sister in law, who was diagnosed with breast cancer. They told us if she had an operation to remove the affec ted breast than that would help her, but as I am talking now, it is one year since she has passed away. So in our family, we have built the belief that if a person gets such a kind of disease then they end up dying. Although different diseases that primar ily affect different parts of the body, breast cancer was the one type of cancer that was occasionally compared to cervical cancer because they both affected women. Combination screening programs in some areas (including CECAP) may have also supported this c on n ection However, in discussions of

PAGE 104

88 prognosis, this was unfortunate because screening and treatment for the two diseases in their early stages can follow very different trajectories. The possibility of death hung like a specter over a cervical cancer d iagnosis, but most women did not see death as inevitable. Early treatment could help prevent it and instructions carefully. This advice ranged from if they come uired funds, access to transportation, and often family support, because such procedures would almost certainly be done in Morogoro or Dar es Salaam. If a woman could not access treatment, the chances of her surviving were viewed as slim. In these discussi ons, the previously remarked upon symptoms of cervical cancer took on growing significance. Cancer, particularly in the West, has traditionally been laden with powerful symbolism I ts very unpredictability and high mortality feels threatening to a society ( Kleinman, 1988 p. 20 ) In particular, it taps into fears of pollution and contamination that can lead to the stigmatization of cancer patients ( K leinman, 1988 ; Lwy, 2011 ; Sontag, 1977 ) This can be literally true for cervical cancer patients, whose symptoms in advanced stages may include the u ncontrollable discharge of bodily fluids, some of which may be foul smelling, and who may experience isolation and familial abandonment as a result ( Lwy, 2011 )

PAGE 105

89 When asked to describe how a woman might suspect she has cervical cancer, participants had responded with a laundry list of potential symptoms. Now, when turning to progno sis, there were fewer responses but the images they created of cervical cancer patie nts were vivid, focused around bodies weakening and decaying over time, from the inside out. Salmah shared a heart breaking story with us about her own experience watching a family member suffer: Why am I saying this? Because I have my sister in law who h ad an operation to remove the baby and after the operation, she had abdominal pains, and she had not gone for cervical cancer testing. As opposed to this time, in the past it was believed that when you came for screening, you would get hurt by the tools us ed for the test and one would end up was told that the cervical cancer had spread to her various body parts, the whole kidney, so they were supposed to go to MNH 12 The doctor at SFR H [ St. Francis ] had told them that if only half of the kidney was affected, then they would have got the nurse to accompany them to Muhimbili. Since they knew that she was in her last days, they were not given any nurse. On the third day she passed away befor e reaching MNH. that she is gone. All the inner things [ organs would have done the tests a long while ago, the story would have been different. I thin k she would not have died. On the 8 th of October, it will be the 40 th day since she had died. This experience had be en influential in motivating Salmah to come in for screening. Her sister in cancer was creeping through her body. Cancer caused internal decay, it was a hungry Ummi called in the opening passage of this chapter, could cause non stop bleedi ng or the passing of blood clots. The only way to stop this process once it had advanced and the uterus was affected, was to remove the offending o rgan; otherwise it would spread. 12 Muhimbili National Hospital, located in Dar es Salaam.

PAGE 106

90 s better if my r elative had HIV, Cervical Cancer and Ugonjwa wa Kisasa Tanzania has been hit hard by the HIV pandemic, with an HIV prevalence that now stands at 5.7% 13 though it varies quite a bit depending on the region ( ICF International, 2004 2012 ) It was one of the first nations in which the epidemic proportions of the disease were recognized, with cases reported in all regions by 1987, just four years after its first reported case ( Centers for Disease Control, 2001 ; Garbus, 2004 ) That year also marked the beginning of the first AIDS control program, and health educa tion, primarily taking the form of national mass media campaigns, has been a part of the program for some time ( Hanson, 2000 ) The HIV pandemi c plays an important role in determining risk of cervical cancer more difficult for women with HPV to clear the infection. HIV positive women are also more likely to become infected with H P V in the first place ( Belhadj et al., 2013 ) In th at issues indicative of economic development in sub Saharan Africa, such as urbanization, migratory work, tourism, and the privatization of public health services, all allow HIV to spread more easily through the population (1998, p. 20). Given that HPV, li ke HIV, is sexually transmitted, it is likely social and economic factors that increase HPV. 13 The prevalence among women is higher, at 6.6% ( ICF International, 2004 2012 )

PAGE 107

91 During our interviews, HIV and cervical cancer were frequently associated wi th one another by participants, either through a direct link (i.e. HIV makes one susceptible to cervical cancer), or through comparison (HIV was like cervical cancer in some way). Although this issue has not been studied extensively in sub Saharan Africa, in Brazil, Gregg ( 2003 ) also noted frequent comparisons between the two diseases by the women she interviewed. There had been a change, some of our participants implied, in the diseases that posed a danger today versus in the past ; several participants used phrases like ugonjwa wa kisasa 14 to refer to HIV. But more interestingly, they then used this same type of language to describe cervical ca ncer and push for screening: person who suffered from it. Just like in the past we could hardly see a person who was HIV positive but nowadays we see a lot of them and we are even living with them in our homes. [Sarah] problem, I will advise my fellow women that, let us go to the hospital. Nowadays there are other diseases that cannot be easily identified. [Faust a] Notions of risk were clearly influenced by HIV health campaigns, with their heavy emphasis on safe sexual activity, monogamy, and waiting to initiate sexual activity. Apart from the similarity in sexual risk factors described earlier, women tied the sy mptoms of weakness and weight loss to both diseases, and the importance o f screening for cervical cancer even when an individual did not have symptoms was compared to screening for HIV. Furthermore, the notion of a weak immune system placing those who were HIV positive at risk of opportunistic infections was present among participants, and not only those who were HIV positive themselves. 14

PAGE 108

92 Women also drew comparisons between HIV and cervical cancer in terms of how the diseases were viewed by community member s, and in how community members would view an individual who had the disease. These views were not monolithic, but rather reflected a diversity of opinion regarding whether or not an individual would be stigmatized or which disease was worse. W hat was inte resting though was that women made the comparison to begin with rather than say, comparing cancer of the cervix to other types of cancer For example, Salmah, the talkative 44 year old divorced mother of five told us a story about a woman in her communit y who had been thought to be HIV positive, but was later found to have cervical cancer She then compared it to her own treatment, as an individual who was in fact living with HIV: At the place where I stay, there is a girl who has been diagnosed with cerv ical cancer. They only said that she had tested for cervical cancer and then she was taken by the nurse, and then on a certain date she was supposed to be at MNH for radiotherapy. Without radiotherapy she would have died. At times I would hear some people Two among the relatives helped her and they escorted her to MNH where she was treated with radiotherapy. So what about me? People would just think I am preten ding, while the truth is that you have the disease [ HIV ] but it has not spread so much in the body. Salmah told us how she had been denied assistance by her own relatives and stigmatized for having HIV, implying that it would have been different if they t hought it was cervical cancer. though w omen differed in their accounts of the degree of stigmatization or support a woman with HIV or cervical cancer could expect Some painted their own these disease s as different, while other felt they would be similar. When they compared treatment of an individual with cervical canc er to an individual with a different disease, however, it was nearly always HIV rather than another cancer or disease condition.

PAGE 109

93 This perspec tive, taken together with the earlier connections made between STIs and cancer of the cervix, has important implications for health education around cervical cancer and for creating linkages between healthcare campaigns and programs that focus on HIV and c ervical cancer. Like Jhpiego, others have explored the integration of HIV and cervical cancer prevention services in sub Saharan Africa with some success though there hav e been challenges as well that will b e discussed further in Chapter VIII ( Belhadj et al., 2013 ; Khozaim et al., 2014 ; Mbulaiteye, Bhatia, Adebamowo, & Sasco, 2011 ) In Tanzania, however, at least one study has shown high acceptance of this idea by Tanzanian women ( Plotkin et al., 2014 ) Local conceptions of disease are not simply a micro level phenomenon, but rather influence health seeking behaviors and the use of health systems ( Kleinman et al., 2006 ; Nichter, 20 08 ) That women identify similar risk factors and make comparisons between HIV and cervical cancer points to potential strateg ies for cervical cancer screening campaigns and prevention programs to take advantage of Clinics could create referral loops to ensure that women coming in for HIV or other STIs are screened for cervical cancer as well and vi ce versa In Ifakara, for example, I was told by the head of the CTC that 70% of their patients were women, women who are at heightened risk for cervical can cer yet when CECAP w as initially started in Ifakara there was no formal referral system put into place T he relationship that now exists between CECAP and CTC is a somewhat informal one created by the staff of the two departments though a CECAP referral question was added to the CTC intake software shortly before I left the field And while women in the GD with specific symptoms are referred for cervical cancer

PAGE 110

94 screening, the CECAP program is not necessarily advertised to the larger hospital patient base in a targeted way. Furthermore, even those vertical h ealth education programs focused solely on HIV and STIs may wish to incorporate discussions of cervical cancer as it relates to HPV linking it to similar sexual behavior risk factors. Women may in fact be more receptive to hearing about cervical cancer screening if it is contextualized within larger discussions about risks posed by sexual activity and childbirth, risks they are already familiar with. At the same time, care must be taken to avoid transpl anting the negative connotations or stigma associated with one onto the other. This will be discussed further in Chapter VII

PAGE 111

95 CHAPTER VI UTILIZING SCREENING SERVICES Salmah Salmah was talkative. She had stories to tell and seemed happy to have a captiv e audience; Rachel listened intently, supplying the necessary anecdote with pole (sorry). A divorced 44 year old farmer and mother of five, Salmah was HIV positive and had a long hist ory of interacting with the biomedical system, including a previous experience with cervical cancer screening. She knew she should be screened annuall y for cervical cancer in law to cervical cancer, and knew that early detec tion was important to survival. After experiencing some troublesome symptoms, Salmah attempted to seek care, but even with her knowledge and past experiences she still faced financial barriers: ist and also I would bleed some clots of blood for 3 5 months. This has become a big health problem to me. When I was at home, sometimes I would have a feeling of a full stomach as if I was pregnant, so this month I was advised by the doctor to go to Sonog raphy to check for pregnancy, but it was not there. I was also told to check for the liver enlargement 15 that money so I had to ask for assistance from peop le until it had reached 15,000TS h 16 which was more than enough for the liver checkup and the results showed I had liver enlargement so I went back to the CTC. I went back home to find some money so that I could get the drugs for the liver enlargement. As I attended the HIV clinic the day before yesterday, the doctor told me desp ite the fact my liver had enlarged, I was faring well. Salmah was then referred on for cervical cancer screening al though she initially had to postpone the trip because of her irregular bleeding S he eventually returned to K ibaoni, the other health center near St. Francis where her difficulties continued: The first time, I went to Kibaoni health center several times and could not succeed until today, when they told me to come to SFRH. Each time I had gone there, at Kibaoni health center, they would tell m e to come the day that followed, as the nurses claimed to have other work to do On arriving here 15 Here she is referring to hepatitis. 16 Approx imately US$8.52

PAGE 112

96 at SFRH, I was received nicely by the same nurse I had seen at Dr. M In the end, Salmah was able to undergo screening and was pleased with the actu al screening experience, but the overall journey had not been without challenges. H er physical condition made the thought of traveling to the hospital worrisome and bicycling (one of the most common forms of transportation in the area) was a difficult pro spect due to the bleeding Although she had been advised to avoid strenuous work or farm cultivation past she had swayed while walking the roads (she did not mention the availability of other options, such as t he bus) Later, we asked her what challenges a mother in general would fac e if she came in for screening and she explained how she herself would handle it if sh e were coming from farther away: As for me, if I want to come for screening, even if I have chil dren, as they have their dad I would leave them with him since screening is free of charge. Maybe the drugs. So I would rather walk on foot if I am coming from a far place. If it is Mtimbira, I woul d tell my children to sell one plastic bucket of maize so that I would get bus fare to come to the would be diagnosed with cervical cancer, then among my three relatives, one would come over an d help me so that I would get treated. coming to St. Francis are illuminated: cost, distance, transportation, physical ill health, childcare, and the potential necessity for repea t visits. Some challenges are specific to the cervical cancer prevention program, but others are true of many medical concerns in low resource settin gs like Tanzania That these barriers exist is not novel ; most have been part of larger global health discu ssions taking place over the past two decades around cervical cancer scr eening in developing countries However, these issues have been understudied in this particular geographic and cultural context, where up until very recently, research

PAGE 113

97 primarily focuse d on the limitations of the health care infrastructure to provide cervical cancer screening ( Chirenje et al., 2001 ; Ngwalle et al., 2001 ) the prevalence of specific HPV genotypes and the viability of a vaccination campaign ( Bosch et al., 1995 ; Mayaud et al., 2003 ; Quentin et al., 2012 ; Ter Meulen et al., 1992 ) and the intersection of cervical cancer with other p revalent diseases ( Kahesa et al., 2008 ; Mayaud et al., 2001 ; Moubayed, Ziehe, Peters, Mwakyoma, & Schmidt, 1995 ; Petry, Scholz, Hollwitz, von Wasielewski, & Meijer, 2003 ) T he previous chapter examined how cervical cancer itself is understood within the local cultural context and how that might impact screening behavior. In this chapter, I discuss the practical barriers that may discourage or outright prevent women from seeking cervical cancer screening in a rural Tanzanian setting. Cervical Cancer: A Disease o f Poverty Early qualitative research conducted on cervical cancer screening in Central and South America has been criticized for its over reliance on the health belief model, or HBM ( Agurto et al., 2004 ) This frequently used model applies specific concepts, including perceived susceptibility, perceived severity, perceived benefits, perceived barriers, and behaviors of interest ( Strecher & Rosenstock, 1997 ) S individual, HBM argues that individuals are rational actors. Health education on personal susceptibility to a specific disease is emphasized, a s are the benefits gained through a particular intervention or action taken by the individual It follows, there fore, that once individual actors understand that the perceived benefits to care are greater than any perceived barriers, these actors will take appropriate action to address health issues ( Agurto et al., 2004 ; Strecher & Rosenstock, 1997 ) This model has since been influential

PAGE 114

98 in research conducted in sub Saharan Africa as well, despite mixed results in predict ing screening behavior ( Abotchie & Shokar, 2009 ; Fort, Makin, Siegler, Ault, & Rochat, 2011 ; Hoque, Ibekwe, & Ntuli Ngcobo, 2009 ; McFarland, 2013 ) While understanding what women know and how they perceive cervical cancer is valuable, this approach tends to gl oss over factors external to the individual that shape beliefs and the ability to take action, including the influence of friends, family, and providers; the availability of necessary resources; and the social or cultural acceptability of particular action s ( Agurto et al., 2004 ; Chan, Yang, Gu, Wang, & Tao, 2014 ) The HBM has also been criticized by medical anthropol ogists and other scholars who argue that while health education is important, consideration must also be given to larger forces that marginalize specific sub populations and increase disease risk at the group level ( Doyal, 1995 ; Farmer, 1999 ; Trostle & Sommerfeld, 1996 ) These larger forces including racism, sexi sm, and poverty for m what Fa r r estrict individual agency (1997, p. 263). Health researchers must examine the context in which health decisions are negotiated and the cultural knowledge and beliefs that impact these decisi ons. Barriers to cervical cancer screening may look different in different environments but in fact many of these challenges arise within the context of poverty and scarcity that is only too common in developing nations. Denny, echoing a sentiment express ed by t would be acc urate to state that cervical cancer is a disease of poor women, representing inequity of access to health care resources, su ch as cervical cancer screening (2005, p. 1204). Scarcity and poverty restrict the options accessible to women and the resources av ailable to take advantage of the options they do

PAGE 115

99 have. In such an environment, simply understanding their susceptibility to cervical cancer and the benefits offered by screening are unlikely to be enough. The Benefits of Screening The women we interviewed at St. Francis, both those at CECAP and those in the GD who were not there for screening, certainly did believe there to be benefits to being screened, there was no doubt about that. Nearly all agreed t he main benefit of screening was that it allowed them to know their own health status. Screening was a tool to Similar attitudes were expressed at the start o f each of ou r interviews when we asked participants wh at individuals should do when they were ill. That this was a clinic based population is again important to note here. By their presence at St. Francis, these women indicated a willingness to seek out biomedical car e, and in response to the general health questions, most expressed confidence in the power of doctor visits and diagnostic test s in general to determine the cause of illness. C ervical cancer screening was no different. Of course, ideally cervical cancer s creening is a process undertaken before a woman actually experiences symptoms; it is meant to be preventive medicine. In the United States and other developed nations, routine screening at annual exams has drastically reduced cervical cancer incidence. Unf ortunately in many areas of the developing world preventive medicine is not common practice, or is not a priority given competing needs ( Abotchie & Shokar, 2009 ; Agurto et al., 2004 ; Fort et al., 2011 ) Furthermore as discussed in the previous chapter, in Tanzania STIs and RTIs are typically addressed through the use of a syndromic approach in response to s pecific

PAGE 116

100 symptoms rather than through routine testing and preventive exams. R outine gynecological exams are not standard practice nor are they feasible in the resource strapped health care system. Despite this, the women we interviewed overwhelmingly felt it was important to be tested as a preventive measure, that is, even if a woman was not currently experiencing symptoms. Women from both the GD and CECAP pointed out an individual cannot always know what is goi ng on internally in her own body. Salama told cancers ] have no symptoms. When you see the symptoms, then the disease will have already become chronic. Mean while Fausta, Naomi, and Salmah drew parallels to HIV testing (the latter two were HIV positive). Naomi remarked that people could appear to be fat, a sign of healt h, but in fact be sick inside. Ano ther major benefit they believed was in the opportunity for guidance from f th e sc reening result was negative, women thought they would be provided with advice as to how to prevent or avoid the disease, and if it was positive, they would be counseled on the next steps to take. Bradley et al. (2006 ) found in their review of studies on cervical cancer screening in developing countries, the decision to be screened was the more difficult decision; once women had been screened the method of concern over what type of trea tment would be administered (though the potential difficulties involved in traveling to where certain treatments were offered will be discussed in more detail in a later section ). The important thing was being treated versus not being treated. Nearly half of the women, evenly represented by both the GD and

PAGE 117

101 CECAP groups felt that gaining access to treatment was an important benefit of screening, particularly early in the course of the disease As Ummi and others suggested in the previous chapter, if cancer is a disease that spreads from the inside out, causing the internal organs to rot or decay, early treatment was one way to forestall this process and prevent the removal of the womb and the potential loss of fertility. For women who were already experienc ing symptoms, screening was a way to determine if these were due to cancer or another condition. This was more commonly discussed by women from CECAP, some of whom like Salmah, had been referred there by other physicians at St. Francis when they had come in for various health troubles She wanted to be screened but was turned away until her bleeding subsided, as it would interfere with the screening process. The main be nefit of screening she f elt, would be to finally have an explanation for her symptoms The results of screening, it was thought, would be a source of peace or calm for women. Knowing was preferable, from this standpoint, to remaining at home and worrying A positive result would of course be cause for c oncern and perhaps fear, but a woman would also be advised by her provider on what to do next. A negative result though meant one was on upande salama particularly Knowing their own status too, several told us, made them feel they could go tell other women in the ir communities about the screening services and encourage them to come in.

PAGE 118

102 Although the women we interviewed clearly saw the benefits of screening, they acknowledged that not all of their fellow women felt the same Answers were often qualified, couched in language that indicated a specific individual screening or treatment would depend on her personal beliefs. Fausta, for example, said At the end, the benefit and have had several checkups in the health center; even when the health workers came to the streets, blood was drawn from me. Now this will depend on how an individual understands it, because you cannot force them. Individual perceptions, the participants themselves felt, were therefore important to consider and, they told us, were not homogenous in their communities. W omen who did not p erceive a benefit from cervical cancer screening had poor perceptions or fears around screening, or were reluctant to go for checkups in general, we re unl ikely to seek out those services. However, we should not assume that these are the only reason s for low utilization ; simply perceiving benefits to screening does not mean women can act on those beliefs within the context of their daily lives. Previous research seems to indicate that when it comes to cervical cancer, the picture is rather more complex. As sociations of perceived risk, susceptibility, and benefits with screening behavior in previous research have been inconsistent when examined as individual variables ( Chan et al., 2014 ; Lyimo & Beran, 2012 ) Women who do not utilize CECAP may be constrained by sociocultural and structural forces that make it more difficult or impossible to access services. E ven among the members of both interview groups many were quick to discuss obstacles that made it

PAGE 119

103 difficult for them to follow through on their belief that the clinic offered the best opportunity for care and treatment. Barriers to Screening There are several step s involved in seeking to utilize the services of a cervical cancer prevention program like CECAP, particularly in rural Tanzania. But before women even leave home, there are important things to consider. First, there is a general lack of coverage when it c omes to screening programs, particularly in rural areas ( Kidanto et al., 2002 ; Peters et al., 2010 ) Tanzan ia is a very poor country with an insufficient health care system, and cervical cancer has not traditionally been a priority ( Kahesa et al., 2012 ; Kamat, 2008 ) Most women do not live near a screening program; utilization of services often requires access to transportation ( Ellis et al., 2007 ; Lyimo & Beran, 2012 ) Secondly, women are charged with most household activities, activities that Ellis intensive, and energy consuming as with caregiving a significant role in a country suffering a heavy burden of HIV (p. 28). Either of these issues, lack of proximity or a burden of household tasks, might prevent a woman from utilizing screening services simply due to time alon e. These household responsibilities may also decrease the time women can spend on income generating activities or work outside the home, and women in the formal economy in Tanzania typically earn less than men as well ( Ellis et al., 2007 ) Furthermore, as Doyal (1995 ) through which most receive their allotted share of both material an d non material 29). Family members often combine their earnings within a household

PAGE 120

104 ( Ellis et al., 2007 ) Given that me n tend to be dominant in the gender hierarchy in Tanzania, a male partner may be very influential in determining whether or not utilizing screening services appears to be a viable option. This has been supported by previous research in Tanzania ( Lyimo & Beran, 2012 ) and will be discussed in more detail in the next chapter. Barriers Coming to the Hospital Researchers and policy makers have long called attention to issues such as transporta tion accessibility, medical costs, and other related expenses as major structural considerations in the design of health screening programs ( Agurto et al., 2004 ; Denny, 2005 ; Lyimo & Beran, 2012 ) A t a rural hospital in Tanzania like St. Francis, it is truly difficult to speak of monetary expenses without also cons idering transportation costs due to poor infrastructure, the distance many women must travel to reach the hospital and the economic realities of most of the rural population ( Ellis et al., 2007 ) In o ther words, the primary financial concern for most women in coming to the hospital for screening is transportation. Because cost is a known barrier, Jhpiego set up CECAP with this in mind. There is no cost for cervical cancer screening though clients stil l must pay a small registration fee when they check in 17 However, the issue of money crept into discussions in numerous ways, most often in relation to the cost of transportation. St. Francis, in its previous incarnation as a Designated District Hospital, already had a large catchment 17 This fee is nominal. While I was in the field, the fee was 500Tsh This was approximately the local cost (or sometimes less) of a Coca Cola. No participant mentioned this specific fee as a barrier for themselves.

PAGE 121

105 Hospital status one of nine in the country, its resp onsibilities have only grown. As a result, patients come from all over the Kilombero District and, in some cases, other neighboring district s The CECAP program itself drew approximately 93% of its clients from Kilombero, however, nine other districts were also represented. In Ifakara a nd the surrounding area, bicycling and walking were the most common modes of transportation. Other forms existed, including piki piki (motorcycles) and dala dala (buses), but were more expensive. Coming to St. Francis from vi llages or the countryside outside of Ifakara was therefore a challenge, and one we heard mentioned over and over during our interviews from women in both the CECAP and GD groups. If a woman lived in Ifakara, we were told, it was no problem. She could just walk. Even a nearby town might not present any issue other than time. But a substantial distance could alone be enough to prevent women from coming for screening. To travel the 30 km from the village of Idete to Ifakara, for example, required a payment of 5,000Tsh (US$3) each way, a significant amount of money in a nation wh ere the GDP per capita is US$694.8 0 ( World Bank, 2010 2014 ) nd potential over night stay if needed Shazia, the CHW I f a woman is staying far from here, she is supposed to pay bus fare, say from Mlimba to here. And suppose if a woman does not have fare? She will die, as there a re a number of women dying due to a lack of her work.

PAGE 122

106 If a woman did not have the money for transportation, she had several options. She could wait saving the money she would need for the trip and possibly suffering at home if she had symptoms. She might even potentially miss the window for early detection, thereby increasing her costs and/or threatening her health further She could ask family members or friends for financial assistance which often presented an additional set of barriers and difficulties, as discussed in more detail in Chapter VII Finally, she could walk or bicycle options that clearly depended on her being physically able to make the trip and havi ng the option of taking that time away from her family, home, and employment This latter point, the ability to take time away to seek screening or medical care, cannot be dismissed or taken lightly. First, childcare was an issue. Only seven of the forty nine women we interviewed were childless (all in the GD group). In the overall CECAP client population during the three year period spanning October 2010 to October 2013 ( n = 1140), only 5.7% were childless, and 77.6% had more than one child (Figure 4 ). This meant that for the majority of women coming in for screening, arrangements had to be made for the care o f the ir children while they were gone if those children were too young to care for themselves for an extended period of time. As mentioned before, pari ty is a risk factor for cervical cancer, thus potentially making the issue of childcare a larger children, particularly many children, yet these very children may make it difficult to leave home for screening. That the majority of women coming into CECAP had children indicates that this was a barrier that could potentially be overcome, but women in both interview groups felt this was a challenge in seeking health care se rvices.

PAGE 123

107 family also likely meant that household activities such as cooking, cleaning, and caregiving for older or sick family members stopped while women were away given that these are all responsibilities that fall primarily to women u nless of course a woman had an older daughter or family member who could help out ( Doyal, 1995 ; Ellis et al., 2007 ) Figure 4 : Parity of CECAP Clients, October 2010 October 2013 For women in Tanzania, time is also, quite literally, money given their high participation in the labor force (including the informal economy and agriculture) ( Ellis et al., 2007 ) If women participated in any form of income generation (whether inside or outside of the home), taking the time for screening or treatment also translated to lost money for the family. Th is was the case for all but three of the women we spoke with; fully 31 of the women we interviewed were farmers, while the others were employed as

PAGE 124

108 tailors, teachers, health care workers, and in business (Table 4 ) Foregoing work for the day was not always an easy decision, if indeed it was an option at all. Ironically, o ne of the health care workers (at a different facility than St. Francis) told us that while she and her fellow employees knew about cervical cancer and wanted to be screened, it was difficul t for them to get time off from their employ er for preventive care! Even when a woma n was self employed, such as Bestina Table 4 : Interview Participants' Occupations CECAP GD 19 Farmers 1 Business woman 1 Ward Attendant 1 Non profit employee 2 Do not work outside the home 9 Farmers 3 Business women 2 Nurses 5 Teachers 5 Tailors 1 Unemployed Many of the women we interviewed clearly felt an internal strain between the belief that health was more important than money and the reality of living in low socioeconomic conditions. Some women were quick to say that money was not an obstacle, given that the program was free person is. The best thing to observe is the life of a person, because money can be found In fact, certain pa rticipants expressed a degree of judgment regarding women who did not follow through on their health, that they were negligent or lazy,

PAGE 125

109 which echoes similar findings in Latin and South America ( Agurto et al., 2004 ; Gregg, 2003 ) The reality, however, was that these sentiments were typically followed by ack nowledgements that there were existing difficulties. Sometimes within a sentence or two in fact, a woman would alternately state that there were no barriers if a woman wanted to be screened, and then go on to say that transportation, distance, or cost was in fact a problem In th ese case s there were tensions between the value women placed on good health, the control they felt they should have over their own health, and the benefit of d the realities of the socioeconomic environment they occupied on the other Barriers at the Hospital The St. Francis screening program is run two days each week, Monday and Wednesday. Although screening begins in the morni ng, typically around 9 or 10am, s tart time is not exact. Furthermore, there is no s et end time, and although staff screen women until none remain in the waiting area, their extensive duties in other parts of the hospital mean that once they have screened all the women who are present, the y must return to their work in other departments. Therefore, women who arrive later in the afternoon may not be seen. While rare, screening days may also be canceled if both staff members have to leave town at the same time. Because of these restrictions, one challenge faced by women seeking cervical cancer screening is the limited availability of services. This is not as great an issue for women who have heard about the services provided by CECAP and come to the clinic specifically to be screened; they are more likely to be aware of which days and times the

PAGE 126

110 services are offered though those days may not be convenient for everyone. However, patients referred to CECAP from other areas of the hospital, like Salmah, or those who see posters advertising the pro gram in the main building that pique their interest generally must make a return trip if they are at the hospital on a day when CECAP is closed. I witnessed this on a fairly frequent basis during my fieldwork. In the course of my interviews and in my daily observations of the clinic, I encountered one woma n who came for screening during her vacation without knowing the schedule and had to immediately return to work the next day, women who were turned away from screening due to heavy blood flow, women who ha d the screening days at Kibaoni confused with those at St. Francis, women who arrived just after AL or MN had been called back to the hospital, and women who arrived on days when family illness, a death, or a conference had called staff members away. In al l of these cases, th e women had to return on another screening day. Since the two open days were non consecutive, it was not a simple matter of staying overnight and returning early the next morning (if having to pay for lodging and potentially missing one More likely the woman re turned home and then had to wait for another opportunity to come back. To some exte nt, given the long queues and wait times involved nearly anytime one needed medical care, and in fact often faced at each individual stage of care, this was an accepted part of life. The benches lining the hallways of St. Francis were often packed shoulder to shoulder I not only observed this during my fieldwork, when ill I experienced it myself. I sat in the queue to register, to see the doctor, to pay for the tests, to have the tests done, to receive my results, etc. just like any other patient. It was not unusual to wait. Marjani a CECAP client, told me,

PAGE 127

111 met the nun [ MN ]. She told us she was going for a mass for a deceased for a while and annoyance or inconvenience for most and for those with responsibilities they had to n eglect while they waited, it was a deterrent to care. Furthermore, clients often lack ed privacy while waiting At St. F rancis, the queue typically formed on benches lining either side of the open air hallways that snake through the hospital. These benches are not only with in sight of anyone else waiting, but also those walking through or around the hospital, which may include other patients, medical personnel, or friends and family members accompanying a patient or bringing in a meal and the situation that everybody would be told us. Lilian and Mercy, both from women from the GD who had never come in for screening shared that people in their communit y might be concerned that doctors would at large. When I asked one of the RAs about this, she told me that of course doctors were not supposed to do that but t hat it d id occasionally happen. St. Francis is a teaching hospital and as such, medical and nursing students are a frequent sight in its halls The medical school itself is located on the same campus. Although rarely mentioned in interviews t his was cause for con cern for two of the participants we interviewed. Salmah, whose story opened this chapter, was one of them. people to see your

PAGE 128

112 Both Salmah and Roseline implied that this feeling was present among other members of their community. For Salmah, it was particularly the issue of being unclothed in front of students. However, Roseline drew attention to the issue of inexperience and treatment of patient s : Sometimes people are afraid of the [ students ] Most of the times they investigate pregnant women, they press them with h ard things, while some may be having severe pain or twins. When they come from there they [ pregnant women [ students ]. Otherwise you could have a lot of people now. If you are taken to a side room with the [ students ] 18 you will be investigated by all of them T his was not a commonly cited concern among the women we interviewed, but I find it worthy of mention because again this was a clinic based sample While it bothered Salmah and Roseline, and it had not stopped them from coming to the clinic, this may not be true of other women. Modesty is important in Tanzania and age is an important marker of respect. It is not unreasonable to think that some individ uals may be reluctant to come in for screening based on these concerns. Certainly the gender, class, and behavior of the provider doing the screening have been important factors in previous studies examining how screening is perceived in developing countri es ( Bingham et al., 2003 ) These idea s were also posited in casual conversations I h ad with informants. Cultural anxieties regarding who is doing the screening, their social role in relation to the client, and how these may in tersect to determine client treatment are important factors to consider in planning screening programs. Overall, these hospital based anxieties were largely voiced by wo men from the GD rather than CECAP, providing a window into some of the concerns that m ay be 18 She referred to the st

PAGE 129

113 preventing othe r women from coming in to be screened Concern over treatment by medical personnel (though not specific to students) and privacy issues have been identified as barriers in other geographic contexts, including Central and South America and sub Saharan Africa ( Abotchie & Shokar, 2009 ; Agurto et al., 2004 ; Mutyaba et al., 2007 ) as well as with other gynecological issues in sub Saharan Africa ( Lyimo & Beran, 2012 ) s for poor uptake have been studied, the most significant ones that can directly inform public policy are not 23 ). However, there have been some previous findings in support of the idea that women may be reluctant to come for screening if they perceive privacy or poor treatment by medical person nel to be an issue. For example, among rural Tanzanian women in the north potential embarrassment during scr eening was a deterrent, and the sex of the provider was also of concern ( Lyimo & Beran, 2012 ) Overcoming Barriers Despite these very large challenges, St. Francis has to date attempted to mitigate at least three of the commonly cited barriers to screening in the literat ure : cost, loss to follow up, and transportation First, the cost of the screening for a CECAP client is free and the program occasionally (depending on funding) can provide financial assistance for transportation when clients require further care in Morog oro or Dar es Salaam. Furthermore, in utilizing VIA screening techniques, clients typically learn their results immediately This one visit approach reduces the risk of loss to follow up that often occurs when patients have to return to the hospital for th eir results There may be exceptions some women for example might require a biopsy for confirmation of cervical cancer but they are rare. This mitigates the need for return trips and the refore

PAGE 130

114 more transportation cost and time. Finally the two staff membe rs who run CECAP not only attempt to keep regular screening hours despite competing duties in other departments (often going as far as to return to CECAP from the main building if they receive a mobi le phone call from a client or R CH staffer about a newly arrived client), but also routinely give up their weekends to travel to outlying villages to conduct outreach when resources permit so that women do not have to travel to St. Francis These are important but small changes; major challenges remain if access to cervical cancer screening is to be made more wid ely available and utilization to increase Transportation, in terms of time, will be an issue as long as the road infrastructure in Tanzania is und erd eveloped (though access to St. Francis specifically is improving; the main road into town from the north is currently being paved). Long wait times at the hospital are symptomatic of system wide shortages in health care personnel and resources in Tanzania. And poverty is an entrenched problem which requires large scale change. Biomedicine tends to prioritize the idea of the individual at risk, placing the individual in a position of responsibility for their own health: O ne can be healthy if one adopts a healthy lifestyle which minimizes personal risk factors ( Trostle & Sommerfeld, 1996 ) The HBM likewise focuses on individual perceptions, and how these shape health behavior. But the idea of the rational actor simply weighing benefits and barriers and, finding the balance tipping towards the former, adopts the proper behavior is often not predictive of actual preventive behavior when it comes to cervical cancer screening in developing countries ( Agurto et al., 2004 ; Chan et al., 2014 ) While the majority of women in this study told us they considered themselves responsible for their own health

PAGE 131

11 5 and believed that knowing their health status was a definite benefit of screening, they did not or could not always translate this into action. larger structural systems and institutions that regulate the inequitable distribution of resources and reproduce social forces such as racism, sexism, and poverty ( Farmer, 1999 ; Galtung, 1969 ) In te rms of health, these shape who is susceptible to disease and who, ( Galtung, 1969, p. 171 ) This structural violence is often so integrated into social and political life, it can feel invisible both to those living within the system, and to health care providers and researchers who may attribute to the individual or to culture differences that are actually a result of socioeconomic inequality or discriminati on ( Farmer, 1999 ; Scheper Hughes & Bourgois, 2004 ) Cervical cancer is a disease of poverty and it is a disease o cervical cancer (as discussed in C hapter IV ), as well as their options to be screened for the disease. poverty, combined with poor infrastructure and an inadequate health care system result in a situation where women who may know of cervical cancer, who are at increased risk for the disease, who may even want to be screened, cannot easily access screening s ervices or may find it problematic to do so. Even the women we interviewed, who did actually arrive at St. Francis, had much to share about the difficulties inherent in the process. The long distances to travel to health centers, the cost of transportation the wait times the overcrowded halls with limited

PAGE 132

116 privacy, the relatively few hours CECAP was staffed; all of these were issues to contend with made more difficult by the financial struggles of both the women themselves, and the institution with which t hey were interacting. In addition, while it was only brought up by two women in this study, the literature also seems to suggest that the provider patient relationship has the potential to be a barrier, due to the influences of gender and differing socioec onomic status on the interaction. Women also have access to fewer resources within a society where they are already suffering the effects of poverty. Within Tanzania, gender is structured to place women in a subordinate position to men, with access to fewe r economic, legal, and political resources. This is partly due to cultural factors, however, Kamat (2008 ) and many others have also pointed to t he influence of neoliberal reforms in Tanzania pushed by global financial institutions such as the International Monetary Fund and the World Bank, particularly their structural adjustment policies, in further marginalizing women especially rural women, ec onomically ( Campbell, Mwami, & Ntukula, 1995 ; Koda, 1998 ; Lugalla, 1995 ) Despite the fact that Tanzanian women participate in the labor force to a slightly higher degree than men, lower education levels and responsibilities at home often restrict their activities to the informal sector ( Ellis et al., 2007 ) All of this makes it more difficult for women to leave their homes, their businesses, and th eir children to seek screening.

PAGE 133

117 CHAPTER VII STIGMA, SHARING, AND SUPPORT Sakina Sakina had an air about her as though being there was tiring. Not necessarily there in the interview room specifically, but rather just having to be upright anywhere. My greetings, Shikamoo! Habari yako mama?, had been received with a big smile when I came into the room, but she had quickly settled back into a slumped position, rubbing her hand over her short come to the GD that day because of abdominal pain; this pain had been with her for three years now. A lthough she had been to St. Francis before for this problem and was given medication, Sakina did not know what her actual diagnosis was and the pain, once seemingly helped by the medication, had returned. Despite being local to the area, having sought trea tment at St. Francis in the past, and having delivered at least one of her five children there, the first time Sakina heard about the cervical cancer screening program was on this particular visit. She had, however, heard on the radio dominal pains, they are supposed to go to the hospital that Sakina was no t feeling well. But one question drew a longer response: If a woman is diagnosed with cervical cancer, do you think she will share aking her head, continuing: She would want to keep the problems to herself. Suppose a disease. It would be a secret between a doctor and the patient, by which they [ the doctor and the patient ] wo uld figure out how they [ the patient ] would get cured. You might tell your At the same time, Sakina told us she herself would consider being screened for cervical cancer so that she could find out if there was a wo uld tell her sister and neighbor about her decision to be screened so that they too could reap the benefits of having the procedure done themselves. But she knew that not everyone would welcome discussion about cervical cancer or support her decision. Whil e some might offer to accompany her

PAGE 134

118 When it came to cervi cal cancer and cervical cancer screening, Sakina made a versus others. The decision to be screened was a personal one and she was not alone in making these types of d istinctions. Previous research has shown that simply choosing to be screened by itself may be stigmatizing or imply guilt, depending on cultural values. These may include objecting to sexual intercourse outside of marriage, considering it shameful to expos e the genital area when not currently experiencing symptoms (particularly to a male provider), or conflating cervical cancer screening with HIV/STI testing ( Belhad j et al., 2013 ; Bingham et al., 2003 ; Buskens & Bradley, 2002 ; Gregg, 2003 ; Sankaranarayanan et al., 2004 ) Given this previous research, I was interested to learn if women had discussed screening with others in their lives, if so whom, and whether or not these individuals would be supporti ve of a decision to follow through. In speaking about perceptions of cervical cancer in the community, the question of whether or not a woman diagnosed with cervical cancer could be open about that diagnosis also arose. For the women in this study, the de cision to share such a diagnosis was largely a hypothetical question that they themselves had not faced (only one had been diagnosed with cervical cancer, though nine had in fact been diagnosed with HIV, another potentially stigmatizing condition). Yet, in discussing cervical cancer screening, it was natural to consider what the eventual outcome of such a test could be and to share stories of others they had known with the disease and how open they had chosen or not chosen to be. Although I was originally i nterested in the topic of stigma around cervical

PAGE 135

119 cancer screening, what came through in the interviews was that although women did believe they had to be careful about whom they might discuss a diagnosis of cervical cancer with, they were much more open ab out screening itself. Cervical Cancer and Stigma Cervical cancer is unique in that it is a cancer, but one that is caused by infection with a sexually transmitted virus. As such, it is burdened by the potentially stigmatizing cultural constructions associ ated with both cancer and sexually transmitted diseases ( Gregg, 2003 ) This has been born e out in a number of different ways in different set tings. For example, in the case of the former, the fatalism associated with cancer as a category of disease may be transposed onto cervical cancer, despite its high survival rates when caught early ( Agurto, 2001 ; Bingham et al., 2003 ) Several studies in Latin and South America, as well as sub Saharan Africa have also pointed to the often negative visceral images ascribed to cancer and how these may be combined with concerns about the impact of necessary gynecological treatments for cervical cancer on fertility and sexuality ( Agurto, 2001 ; Bingham et al., 2003 ; Gregg, 2003 ; PATH, 2004 ) On the other hand, many of these same studies have also shown that women often confuse cervical cancer screening with HIV or other STI tests, and may be concerned about the implications for their rep utation. As I wrote in Chapter V regarding starehe relationships, sexuality and fidelity in Ta nzanian society tend to have very gendered norms. Setel (1999 ) has written of the importance of secrecy in sexual and reproductive matters among women, and often men as well, in northern Tanzania under the kihamba regime, particularly regarding infidelity. The women he spoke to claimed it was the open revelation of a

PAGE 136

120 than the actual act of infidelity, a relat ively accepted behavior in men: relationship remained private and secret, it naturally remained outside the purview of Setel, 1999, p. 103). This emphasis on secrecy and privacy around relationships and sexual activity was prevalent in the daily cultural discourse which I experienced during my fieldwork, as was this double standard in the behavior of men and women. When it comes to cervical cancer screening, this has real implications. In many parts of sub Saharan Africa, women have shared concerns that their husbands or other communit y members will view a positive diagnosis (or even any non pregnancy related pelvic exam) as evidence they were unfaithful, promiscuous, or in ( Belhadj et al., 2013 ; Bingham et al., 2003 ; Buskens & Bradley, 2002 ) The decision to disclose a cervical cancer diagnosis then is one that has the potential to transform what Goffman (1974 ) an individual with a yet to be recognized stigmatizing condition (p. 4). Cervical cancer is, at least in its early stages, typi cally asymptomatic, allowing women with the disease to present themselves as healthy in dividuals. As noted in Chapter V when symptoms do occur in earlier stages, unusual vaginal discharge or blood spotting are the most common, processes of the female body even healthy women may experience. Although, as Sommer ( 2010 ) points out, even normal menstruation may be more difficult to hide in Tanzania where socioeconomic realities mean that inadequate sanitation and access to sanitary materials are common for many women. Avoidance of stigm a then usually depends on controlling information about the diagnosis, choosing

PAGE 137

121 condition ( Goffman, 1974 ; Kleinman, 1988 ) Yet women who come to St. Francis for screening services (as we ll as other health issues) face numerous structural and socioeconomic challenges, which they cannot always overcome on their own. Participants recognize d that these challenges were far greater for women with a positive result who needed treatment for the disease, and that in many cases, these challenges would be insurmountable without the support of others in their lives. Securing support from those in th eir social networks was often critical, and women were very clear in drawing a link between the availability of social support and health. This link has been highlighted across a wide range of health issues in the literature, including cervical cancer ( Seow, Huang, & Straughan, 2003 ) HIV ( Badahdah & Pedersen, 2011 ) breast cancer ( Bloom, Stewart, Johnston, Banks, & Fobair, 2001 ; Peters Golden, 1982 ; Roberts, Cox, Shannon, & Wells, 1994 ) stroke (Lopez Espuela et al. 2014), and mental health ( Berkman & Kawa chi, 2000 ) Social networks, or what Heaney and Israel (1997 ) looked to by individuals to provide a number of different types of support, including emotional, instrumental, informational or appraisal support (p. 181). socioeconomic context are therefore important to consider. The goal of this study was not to conduct a social network analysis, nor was it equipped to do so. However, women themselves brought up and explored the give and take between being open about their health and the need for support during our interviews and in so doing, pointed to important potential areas for intervention and future research. Just as gender norms and

PAGE 138

122 expectations may intersect with poverty to place women at greater risk of certain diseases, including cervical cancer, these social forces likewise influence their access to the resources and support necessary to seek care ( Doyal, 1995 ; Farmer, 1997 ; Wingood & DiClemente, 2002 ) Given t his context, there were often powerful reasons why a woman would choose to discuss her cervical cancer diagnosis with others, particularly her husband and family members, despite the potential stigma she might face. Fear of stigma was less of an issue for the screening process itself among the women we interviewed, though not completely absent. Sharing the Screening Decision Despite evidence from other developing nations, currently there is limited data available on whether or not cervical cancer screening is viewed as stigmatizing in Tanzania. What research does exist on factors that affect cervical cancer screening utilization that relate to cultural acceptability and privacy. Similar concerns have demonstrated importance in health care utilization across a number of sexual and reproductive health related issues in Tanzania ( M finanga et al., 2008 ; Mrisho et al., 2007 ; Pool & Washija, 2001 ) As a case in point, Lyimo and Beran (2012) found that both husband approval of scree ning and preferences regard significant factors in uptake of screening services in a northern rural area of the country. Among the women we interviewed, few felt that the decision to be screened should be kept a secret, whether this was a hypothetical question or one they had already dealt with. For those few, however, privacy and stigma were definite concerns. Khadija, for example, a talkative outgoing CECAP client missing her top front teeth, was

PAGE 139

123 unequivocal about the fact that she would not speak with other women about the screening program after she left the hospital. She was eager to share her aches and pains with us before her interview, and with her animated voice and facial expressions, was engaging to listen to. She seeme d to be enjoying the conversation, and would often look lively and outgoing nature, I was surprised at her response when we asked her if she planned to discuss the screening firmly. She, as well as another CECAP woman, Karima, both indicated a discomfort in speaking about the disease, and were concerned about how others would view them. Women who said they would not share this information felt that health was a personal Another issue cited by two women as a reason for keeping screening plans to themselves was that of distance from family or a support system. This was the case for F uraha, a woman in the GD who had actually been screened in the past at a nearby health center but had not told anyone because she was currently living alone. Cecilia, on the other hand, had told her husband with whom she needed to leave her child, but said that other people in her life thought she was seeking care for repeated fetal loss (which was indeed why she was eventually referred for screening) and that she had not been more specific since St. Francis was far from her home. For these two, cervical ca ncer screening However, it was far more common for women in both groups to say they would talk about screening with family members, friends, even neighbors and other community

PAGE 140

124 members, either before, while making their decision, or afterwards. For example, four of the six women we interviewed in the CECAP clinic that had not discussed their decision to b e screened with anyone prior to utilization planned to share information about the program once they returned home. In two of these cases, the women had been referred that same day from other St. Francis departments and so simply had not had a chance to do so yet This is not uncommon among CECAP clients. Women spoke about cervical cancer screening to those in their lives with multiple goals in mind: to access support that would enable them to go for screening, to share their health status with those import ant to them, and to spread information about the program. The Role of Husbands On both previous visits to Tanzania and during the initial months of my fieldwork, various informants, both within the hospital and in t he course of daily conversation would t ell me one barrier to screening was that women often needed their supports the idea that husband approval was associated with an uptake of screening services both in Tanzania and elsewhere (Bingham et al. 2003; Lyimo & Beran 2012). The women we spoke with they outright rejecte d the notion that their husband s resistance would prevent them from coming in. Leta, for example, acknowledged that some husbands might not allow their wives to come in for screening, but did not view lack of husbandly support as a reason to stay home if it was possible for a woman to go without hi s support. If for example, she went for screening and was diagnosed with cervical cancer, she pointed out pragmatically, her husband would have to come around:

PAGE 141

125 have go t the message that you have a certain problem, so for him, getting angry, will be Women did most commonly cite their husbands as the person they would or did discuss their decision to be screened with, with at least half in both the CECAP ( n= 12) and GD ( n= 18) groups responding so. They largely felt their husbands would be supportive of this decision, provided that they understood the problem at hand This issue of understanding was of more concern to the GD women than those from CECAP, perh aps because the CECAP group had already come in for screening and so this i ssue was behind them. Hellena, the CECAP client employed as a Ward Attendant, told us that, understands it because some may allow their wives and others may not allow them to do pote ntial lack of support to issues of inadequate education or knowledge about the disease. On the other hand, there were several important reasons why women felt their husbands would be supportive if they shared with them and why it was at times in fact neces sary to do so. The first was emotional support, which Bingham et al. (2003) have support). Several women made the point that they simply had close relationships with thei r husbands, with whom they could share concerns and advice. Dinah illustrated this when she said

PAGE 142

126 concer but said that she would discuss it with him if Women who had actually experienced symptoms of illness told us that their husbands expressed concern or fear about their he alth. This in turn led to emotional support for seeking care. Zuhura, a small older woman of 65 with a serious face, shared ave pain whenever she tried to drink or eat. He encouraged her to come in for screening, thoug h he could not assist her financia lly. This was also the case for women whos e symptoms interfered with the sexual relationship they had with their husband s such as Fausta, who came for screening at the urging of her husband and a friend because she had pa in and bloody discharge, or Shazia, whose husband accompanied her to CECAP after she began to have pain and a burning feeling in her abdomen. B oth experienced their symptoms during intercourse. There were also more practical concerns. Husbands could potent ially be sources of instrumental, or tangible, support in the form of financial assistance or transportation. Aisha had been having bad stomach pains when she heard that cervical cancer screening services were being offered. She and her husband farmed and money was tight. Yes, he supported me, as I had a problem with stomach pains and I had been to several hospital s for the same problem and yet the problem was

PAGE 143

127 still there. And so when these white doctors came, he insisted that I should go to the hospital and he carried me on his bicycle all the way from [ there ] to here. Screening services at St. Francis were free; therefore the largest cost was transportation rather than care. For that r eason, as discussed in Chapter VI women often walked or rode bicycles to get to the hospital. A husband might accompany his wife, though Leta pointed out that he may simply give her important for screening, this idea in fact was more prevalent in conjunction with an actual cervical cancer diagnosis, when more travel and hi gher expenses would be an issue. Relatives, Neighbors, and Fellows Perhaps the best indication that most women in this study did not view cervical cancer screening as stigmatizing was their openness with other women in their lives. Other researchers have p ointed out the positive association between strong social support relationships and uptake of cancer screening ( Bloom et al., 2001 ; Buskens & Bradley, 2002 ; Seow et al., 2003 ) Specifically in the context of cervical cancer screening, Buskens and Bradley (2002) found that women in South Africa often brought others with them to sessions to hear information about cervical cancer, and that these women might visit the clinic multiple times to find out more before consenting to be screened themselves. I observed this phenomenon myself multiple times, where a woman would come to the clinic with a friend, who then might or might not consent to be screened that day. Marjani, a CECAP client, for example, nodded her head towards the waiting area ho was here dressed in yellow about a week ago and she agreed [ to come

PAGE 144

128 Aside from husbands, women in this study named a wide range of other individuals they would discuss their decision to be screened with, most commonly family memb ers, but also occasionally neighbors, community members, coworkers, and friends, gendered familial terms like child, children, parents, or family, those who named specific familial relation ships all spoke about female family members which is notable. And in fact, for Edna, who only had sons, this was an important factor in whom she felt she could of my Outside of family members, women also spoke of discussing their decision with neighbors, discussions and the reasons for sharing their decision made it clear that again these were typically relationships with other women. Doyal (1995 ) notes that women are frequently suffer from a lack of emotional support from the male fa mily members in their lives. When this is compounded with cultural taboos around discussing sexual or reproductive health issues with the opposite sex, as in Tanzania, women turn to other women. Husbands were clearly a source of support in this study, but outside of this dyad, women were more likely to turn to a female member of their social network rather than a male. Unlike their husbands who were looked to more for emotional or instrumental support, women attributed their desire to share their decision to be screened with other women in their lives largely to the reciprocal nature of these relationships. Just as many

PAGE 145

129 had heard about cervical cancer or the screening program from other women, most women had a desire to in turn provide informational or appr aisal support if it might encourage other women to consider screening. For example, a neighbor who had been screened herself told Violeth about it and suggested she also utilize the services. When Violeth began to have a burning sensation in her chest, she decided to come to the doctor Yes, as she had come for the screening some years back and she was also diagnosed as having intestinal cancer, so she had advised me to come e arly. She told me that if I would have any problem, then I would be Shakes head ] Violeth had already spread the word about the screening program hers elf, and said that if a woman i s diagnosed wit [ other people Several women felt so strongly about this that they wanted to serve as a model for others, saying that in addition to sh aring information, they would answer questions about the screening procedure if their fellows had them. Sakina told us that she would not only discuss the procedure with her neighbor or sister, but that if they were supportive, she would tell them to come together with her so that they could observe the screening process if they chose. Fausta shared the following story when we asked her if she had discussed her decision to come into the cervical cancer screening clinic with anyone else: Yes, I told my neigh bor because she also has stomach problems. We are of the same age but she has only one child and stomach problems started after giving birth to her child. When I told her, she said I should come come too. Her stomach problems increase more when she is in her menses. I had a chance to observe this phenomenon in action during one of the CECAP mass screening days. For these events, extra screening personnel are typically brought in

PAGE 146

130 and screening is done in multiple rooms at the same time all day long. Sometimes these occur at St. Francis, while on other days they go out to rural village dispensaries. Although my research was focuse d on the clinic population, AL accompany the team on one of the rural outreach days just to observe. The village dispensary was crowded all morning, and in fact women were waiting when we arrived after hearing loudspeaker announcements the day before and that morning. Eventually the numbers dwindled and came nearly to a halt. After a long morning of ensuring women knew where to go and when, I was sitting out on the porch of the dispensary, playing peek a boo with a small chubby toddler whose mother was in one of the screening rooms, much to the amuse ment of the few women l eft on the bench beside me. AL I asked her. She laughed and gesturing with her chin out to the path leading around the r was leading two other women back to the clinic. Some women are afraid of the screening procedure or do not know much about cervical cancer, she told me. So they stay home and let the others go first. Then, later, after their fellows have been screened an d come to tell them about it, they may be persuaded to come in themselves. Sure enough, over the next couple of hours, stragglers would come in groups of two or three accompanied by a previous client. Despite research from other developing nations that uti lizing cervical cancer among the women we interviewed. As noted in other places in this dissertation, this was a hospital based sample and women who were concerned ab out any potential stigma associated with the procedure would be less likely to seek out such services. This is

PAGE 147

131 certainly the case with HIV testing, another stigmatized disease in Tanzania. Kominami, Kawata, Ali, Meena, and Ushijima (2007 ) found that Tanzanian women feared the potential repercussions from their male partners, such as violence, divorce, or abandonment, which could result from a positive HIV diagnosis, leading them to avoid testing, particularly if permission from a male partner was needed. Other researchers have reported similar findings ( Maman et al., 2003 ; Mbago, 2004 ; Njau, Watt, Ostermann, Manongi, & Sikkema, 2012 ; Wringe, Isi ngo, Urassa, Griter, & Mayalla, 2008 ) Privacy and confidentiality were still a concern, even in the case of home based couples HIV participation in testing and bring couples closer together ( Njau et al., 2012 ) These concerns were mitigated for some by the home base d nature of the testing, which they considered more private than the clinic, though others felt that even this method could not provide enough privacy from other members of their family or household. Given both the perceived and actual relationship betwee n HIV and cervical cancer in Tanzania, similar concerns regarding cervical cancer screening may be present among Tanzanian women not represented here. Howeve r, it is worthwhile to note these views were not prevalent even in the GD group who had not come fo r screening, and that all of the mass s creening and outreach days CECAP had held, in which women were sitting out in the open together while they waited to be screened, often in f ront of other community members were well attended. From these, it seems pos sible that cervical cancer screening is not considered to be as stigmatizing in this context as HIV testing might be.

PAGE 148

132 Sharing a Cervical Cancer Diagnosis While few women felt cervical cancer screening needed to be a secretive affair, cervical cancer itself an actual diagnosis was more complicated. Farmer (1997 ) d the large social matrix in which it is embedded in order to see how various large scale social forces come 261). He and others have written extensively on the ways in which social forces, including sexism and poverty not only shape health risk, but also constrain the agency of individuals to seek care for their disease s and mitigate suffering ( Doyal, 1995 ; Farmer, 1997 ; Wingood & DiClemente, 2000 2002 ) These are highly visible phenomena when it comes to cervical cancer in rural Tanzania and are entangled with cultural notions relating not only to appropriate gender roles and relationships, but also expectations of the roles that families should play in assisting sick individuals. P articipants told us that whether or not a woman would share a cervical cancer diagnosis with others depended not only on her perception of the disease itself and how she anticipated her family and community would react, but also her expectations of the types of support needed to access treatment. Fear of Stigma or Rejection Some women may share the news and some may choose not to share the news with anyone. Some may be afraid that they may be laughed at because of having the disease or will be discriminated against; it is acc her husband, mother, or anyone who they are close to. You know, when you have a certain problem, you have to abandon your fears. All the time, fears can bring about a lot of diseases. At f irst you may get stressed, which may be suppressive. Whenever you are having problems, try to be as open as possible about the disease. Even if the people who surround you are bad. Especially if they are illiterate. Although I have not stayed for long in t he village area, in most circumstances it is usually tough for women. [Sandra]

PAGE 149

133 When asked if she thought a woman diagnosed with cervical cancer would share this information with other people, Sandra gave a complex response that demonstrates there are no simple answers. Like many participants, she reasoned the decision would be different for each woman, depending on her own personal fears and perception of cervical c ancer, but she herself felt fear and stress could themselves actually bring on disease. She went on to recognize that the course of action she recommends, openness, may actually be very difficult for women living in small rural villages. In his landmark work on stigma, Goffman (1974 ) fe, whether in town or city, there will be few who merely know of him; those that know 68). In such an environment, potentially harmful information that became public could spread quickly and be discussed wi dely. Those who brought up the potential for stigma referenced the feelings of fear and shame that could come from a cervical cancer diagnosis being made public knowledge, and in some cases drew parallels wit h HIV, as recounted in Chapter V Two women me ntioned witchcraft; one told us that women in her community sometimes claimed to be bewitched because they did not want to believe they had cervical cancer, while a second said that she would tell her son about her cervical cancer so that no one else could claim her poor health was due to witchcraft. In either case, one condition was used to deflect the potential stigma of the other. Shazia known as being diagnosed with having cervical cancer, as they may be ash recounted to us how women in the communities she works in will sometimes approach

PAGE 150

134 the CHWs privately as they go house to house to discuss sensitive topics, rathe r than bring them up in front of other people. While most of the women we interviewed felt that they themselves would tell others about their diagnosis, many, like Sakina in the opening to this chapter, drew lines between those it would be appropriate to tell, i.e. husbands, family, friends and it being public knowledge. Rose that you have got this disease. Relatives that, they are born together and family, but you many women expressing the desire to share the information with other women in their lives who could th en go for screening, caution had to be exercised because not everyone discussing her experi ence with cervical cancer screening with others in her life, took a dim view of being open and doubted that it would result in support of any type: If I do that, it will be like making a public announcement. I better stay with my disease. You are the ones who know what to do so that I get better. Even if I tell them, they will not help me. Why, are the people from your place not good? appearance you are also beautiful but I cannot know how y our soul is. Women in general were more circumspect about the prospect of sharing a cervical cancer diagnosis, but in reality it was not universally considered to be a stigmatizing condition. Those who had known someone with the disease or who discussed c ervical cancer in their communities gave mixed reports. When speaking generally, they might

PAGE 151

135 stigmatize a woman with the disease. And yet when recounting specific individual s they knew or had seen with cervical cancer, these stories typically included family members or woman with cervical cancer and we consider her as a normal woman, like our fellow the actual likelihood of rejection. Marshaling Resources members was viewed as a necessity in most cases in order to gain access to the types of support that would make accessing treatment a possibility, particularly if that course of treatment was a long one. Treatment itself is supposed to be free for women at Ocean Road, but there are many incidentals involved. While in Dar es Salaam, a patient must pay for lodgings or stay with a relative or friend. They may also need someone to help care for them when they are not at the hospital or to cook and bring meals to them while they are undergoing care. Once a woman returns home, she may not be in any state to take care of the household duties and caregiving that largely fall to women. In this case, 46) Shazia shared with us a lengthy story about her sister in law which serves as a case study of the many issues that face a woman and her family going through this experience. Her sister in law had been married to a man who had left her for another woman at one point. Wh en he came back and they resumed sexual intercourse, she began to experience continuous bleeding and was eventually diagnosed with cervical cancer at St. Francis before undergoing treatment at Ocean Road. At first, she had to go to Dar es

PAGE 152

136 Salaam once a mon th, but now it was only once every six months. Even so, the resources and support needed by someone in this position simply to take care of day to day activities were many: To be honest, that mama does not do any work, and whenever she is told to go anywhe re for treatment we [ the relatives ] contribute for her safari and medications, and because for now she is staying away from us, her husband is the one who helps with the daily chores. Worse still, her mother in law had only one daughter from her womb and s he is married step in from time to time and assist her sister in law. Only once in a while we would pay her a visit and help with some of the chores, say fetching water, or when her dates [ for treatment ] were close we would contribute some money. in law had both her husband and her nephew to stay with her while she sought treatment, but her husband made it clear to the extended family that they would need assistance. My sister in law faced a number of challenges as she went back to her health. My brother is not well off. He cuts wood and sells it for a living. There are times when wood is so sc arce, and for now he has stopped productive as he may harvest only two sacks, and so it was upon us to contribute for her hospital expenses. nomic support to her sister in could change. Family members were expected to contribute what they could, but in addition to her own children, left in her care when her other brother died Justification for sharing a diagnosis was therefore framed by many women as necessary to marshal the resources necessary to seek treatment. Neighbors and

PAGE 153

137 community members might provide assistance, but this was secondary to the family, for whom it was a responsibility or duty. The family as a support system or safety net is important in Tanzania and has been examined in light of other health a nd social conditions, as well as in the uptake of certain health behaviors (e.g., Foster, 2000 ; Geneau, Lewallen Bronsard, Paul, & Courtright, 2005 ; Kamat, 2008 ; Kominami et al., 2007 ) family members should contribute support in some fashion if a woman was diagnosed with cervical cancer was made clear during our interviews: How will you get money to go there? There must be a family meeting. Those with love will contribute. [Rahma] Rel atives, they are born together. [Ro seline] I will share it a lot with my relatives so that they can help me because now because it is not easy to get help from neighbors [Karima] Resources could take the form of chi ldcare, transportation, assistance with household duties, a place to stay while seeking treatment or recuperating, accompaniment to appointments, or most commonly, money. But of course, the support n eeded from and provided by the family varied by the econo mic status of the woman and her family. While Susannah pointed out that women who had house help or who could afford to send their children to boarding school were less likely to be affected by the rigo rs of undergoing treatment, these circumstances were n ot common among the women we interviewed. Instead, Mwajuma summarized a common refrain after explaining to us that one could ask someone close to them for help if they were well off: These types of resources

PAGE 154

138 were often needed even for outpatient visits, but were particularly important if long distance t ravel or inpatient treatment were required. Help was not always forthcoming, either for economic reasons, or because as Rahma put it, woman we interviewed who had actually been diagnosed with cervical cancer and who had undergone treatment for it at MNH and Ocean Road told us that although she shared it with peo circumstances had been affected by fluctuations in the prices of cash crops and so eventually she had stopped following up at Ocean Road. Unfortunately, we were seeing her on the day o f her interview because she had returned to St. Francis with a reoccurrence of stomach pain. However, an individual in poor health was seen as a potential burden for a household, which was an added incentive for husbands or relatives to contribute resource s. A sick woman, a woman in pain, would require help with childcare and household duties and may have to stop any income generating activities. Although in law highlights the assumptions made regarding the gendered nature of housework W omen without daughters or close female family members, or who have those but who live at great distance from them could find it difficult to obtain the help they require. Though in regard to a different illness, Karima was in fact experiencing this very issue the day we spoke to her. She had come to the hospital due to continuous bleeding and was referred to the CECAP clinic. At first, she told us, the doctor had wanted to

PAGE 155

139 thing. Although she told us that in the event of a cervical cancer diagnosis she would tell [ today ] because I will just b e suffering, and also there is no one to take care of my children. They are young role of female family members as caretakers is highlighted here her children were male and young, there was no older female child to care for them. But while women felt th at turning to relatives for instrumental support was an n an environment of scarcity being sick for too long or not conforming to cultural notions of what a sick person should look like providing support. C ervical cancer being largely invisible in its earlier stages led on e woman to argue that neighbors and relatives might think a woman was pretending to be ill in order to avoid work In a society where thinness i s often associated with illness, not meeting these expectations could be problematic. For an HIV positive woman like Salmah, as an example a relatively youth ful face at forty four and a heavyset body did not match how she felt internally and she felt conflicted about how others expected a person with her condition to look. This, together with her long illness, she felt, interfered with her ability to access th e resources she needed from a wider social network, until she had only her children to rely on: As for me, I have nobody for sure! Apart from my child, but those who are coming f rom the poor families like mine are supposed to be helped. For a person like m e, would always regret it if you lack the help. This fat body would swing many times while walking on the roads and only my children have been there to help me move on. All of my rel atives are tired of me, but my child has never been tired of me. He has been of help, to an extent

PAGE 156

140 thing he could have got for the whole day. Given the structural barriers to treatmen t, instrumental support from family members in the form of financial assistance, transportation, and housing, was a critical resource that could make the difference between a woman being able to make the journey to Ocean Road or not. However, even advice a nd counsel from a trusted friend or family member was considered a resource by some women, such as Mariamu: As you know, in our families, you can go to certain relatives to seek help yone to go to for seeking help, but I can sit down with someone an d talk. I get advice that way. And of course, women saw themselves as sources of information or appraisal support in return. By sharing their experiences, they could perhaps serve as a res ource for other women they were close to by informing them about the disease. The consequences of not sharing, of cho osing not to of being open but having no family support) could be dire if it meant a woman could not access the resources that would allow her to go for treatment. The majority of women we spoke to who were at least somewhat knowledgeable about cervical cancer, and even worse ou choose to struggle by themselves until they reach the hospital, or otherwise they would A Heart of Sharing Earlie r in this chapter I recounted how women were often open about their cervical cancer screening experience because they wanted to be a source of information

PAGE 157

141 for those around them and to serve as a role model. Cervical cancer screening was not seen as particu larly stigmatizing by most of the women who made these statements; one might suppose given how much more guarded women were about the prospect of openly sharing a cervical c ancer diagnosis that this same role might not be expected of those who were actuall y suffering from the disease. But in fact, the majority of women felt that those diagnosed had a responsibility to share this with other women in their lives so that they too could also be screened. Shazia called this passing on of information or appraisal possibility that other women in their families or communities were harboring the disease without knowing it. By sharing their own diagnosis, they know However, it wa s not just about sharing information, as women who had come for screening did. It was also about the idea of increasing openness and shining a light on a disease than was not typically visible. Take for example, these two quotes: I repeat that the privacy part of most Africans is what is killing us. Just because one is diagnosed with cervical cancer and then keeps the news for themselves, and comes silently for treatment, gets healed, and remains silent, while the rest continue to suffer. That is why I sugg ested that the be sharing the news. [ Dinah ] Yes, she should tell them. Why not? It is because if you have been diagnosed with the disease, you should tell someone so that she would also come for the screening in case she has the emerging symptoms, as hiding about the disease would do no good to both of you. By hiding about that, most of the people would sink in a pool of problems. [ Grace ]

PAGE 158

142 And again, for some m embers of both the CEC AP and the GD group s this was reciprocal. Other women had been a source of information for themselves. A neighbor perhaps, or family member, had spoken with them about their own diagnosis or that of a hare if she herself were in that position and then followed up by telling us how she had learned about cervical cancer: for cervical cancer screening, and also it happened that her rela tive had also been screened and diagnosed with the disease. Since then she had an operation and so far sh e is doing fine. might not agree with the importance they placed on shari ng, nor that there was no potential for stigma if one was diagnosed. But personally they felt that they would rather share so that other women they knew would be aware of the disease and the opportunities me, I will share it with others so as to the disease while I did not expect i pointed out one way around this dilemma without making herself vulnerable to discrimination by her community: If she wants to rescue others, yes, she might surely share with them. But if she feels that onc e she shares the news they would spread them, then she might not opt to share. But if I love my friends, then I would tell them as if other people. Negotiating Stigma and Support Many T anzanian women live within a parti cular socio economic context that often puts economic security out of reach, and shapes and constrains their ability to seek health

PAGE 159

143 care services in numerous ways (Kamat 2013; Setel 1999). As discussed in previous chapters, overrepresentation in agriculture and the informal economy, make it difficult for women to leave home for any length of time; both also contribute to their unequal access to financial resources (Ellis et al. 2007). Within the home, gendered power relations typically place men in the dominant posit ion. W omen may have little to no control over the income they do bring into the household and may find it difficult to exercise agency over their own he alth decisions, particula rly if they are required to obtain permission to seek care (Doyal 1995; Ellis et al. 2007; Lyimo & Beran 2012). These are important issues, heavily embedded within cultural norms, as well as larger structural issues related to poverty (Farmer 1997; Wingood & DiClemente 1997). Given these constraints, it is perhaps not surprising that while many of the women we interviewed acknowledged the fear and shame that could accompany cervical cancer related stigma in their co mmunities, the majority also acknowledged that disclosure to at least some family members or close friends was necessary to obtain the support required to seek treatment. However, the women we interviewed found ways to work within or at times push back a ga inst these larger forces at the same time they were constrained by them in order to utilize the services they needed. For example, t hey explained cervical cancer to their husbands and advised them to gain their support for screening. If they had the resour ces to go on their own, they might choose to do so even without his support. being. S ome men even accompanied their wives to the clinic to show their support.

PAGE 160

144 Participants advised other women and advocated spreading knowledge and information about the disease, even serving as models to other women who were hesitant about screening ( though screening itself was not necessarily stigmatized ) Within social networks, women relied on other female family members or neighbors for assistance with childcare and household tasks. If diagnosed with cervical cancer, they called on these familial networks for money, transportation, and other types of instrumental support and were clear that even if these resources were not forthcoming, they felt they should be. These have important implications for increasing utilization of screening and treatment program s. If women are reliant on the support of their husbands or other famil y members, and if in particular they require the permission of their husbands to seek care, then public health campaigns need to make more effort to increase awareness and knowledge of the disease among both women and men, and to provide women with the language and resources to discuss the issue with their husbands, while at the same time being careful not to further perceptions that could increase stigma for women diagnosed with the dis ease. The need to involve men has been recognized and advocated within the larger sexual and reproductive health literature on sub Saharan Arica in general but relatively little has been done around cervical cancer ( Behrman, Kohler, Watkins, & Kenan Jr, 2002 ; Bingham et al., 2003 ; Kapiga, Shao, Lwihula, & Hunter, 1994 ; McCloskey, William, & Larsen, 2005 ; Mutyaba et al., 2007 ) Furthermore, the women we interviewed felt great responsibility to pass on inf ormation to other women they knew about the opportunity to be screened, and may themselves have been informed by other women in their lives. These social networks are important opportunities to tap into pre existing routes of information, to spread awarene ss

PAGE 161

145 and knowledge of the disease and potentially reduce the cultural associations of shame or fear that may keep women from seeking screening or disclosing their diagnosis. Several women demonstrated a willingness to go so far as to physically accompany the ir fellows to the clinic, either for observation or support. Cervical cancer screening programs may be able to make use of pre existing traditi ons, like Community Health W orkers and Final ly, CECAP is free for women, removing one socioeconomic barrier, but many others still remain for both screening and treatment. Unless larger gains are made in promoting economic security and independence for women in Tanzanian society, women will continue to have to call on the support provided by social networks to access care. While a strong social network has many benefits and can provide important resources to individuals facing health problems, the current situation leaves women reliant on cultural ex pectations of fami lial relationships and the good will of husbands and relatives who may not even have the resources to offer in the first place

PAGE 162

146 CHAPTER VIII COMPETING PRIORITIES Bright and early I made my way once more to the weekly clinical meeting held in the lecture hall at St. Francis. About 25 doctors and interns filled the large room, including four Europe an physicians curre ntly working at St. Francis and myself. To start, two young men presented a case they had recently encountered at St. Francis a nd were critiqued on their treatment plan. The patient had presented with the herpes simplex virus (HSV) and Steven Johnson Syndrome, along with a very low CD4 count, the combination of which would seem to indicate a hi gh likelihood of HIV infection. Unfor tunately, the CTC was still unable to test for the disease. This had been the hot topic at the August 23 meeting, when I lear ned that no HIV testing reagent had been available district wide for two weeks. At that meeting, we were told the supplies were exp ected to arrive by truck the following Monday. Now, two weeks later, no supplies had materialized and no HIV testing had taken place at St. Francis for nearly a month single test i s available in al one doctor declared, emph atically. A discussion about the best treatment course for the patient broke out, i nterspersed with speculation about the missing supplies. Dr. M mentioned seeing a truck the day before yesterday, but a young doctor contradicted him, saying that was for so supplies. Finally, Dr. F, from the CTC called for an end to the discussion, saying that given the low CD4 and Steven Johnson, the patient should be treated for HIV even without testing available. I was still thinking ab out the meeting as I headed back to the GD to meet Paula and Rachel for interviews, and the implication that none of HIV testing in recent weeks. Dr. F and AL had recently been working to build r eferrals between CTC an d CECAP ; how would this lapse in the relationship, at least on one side, affect that process? For the clients themselves, this also had potential ramifications, as any who were HIV positive but unaware of that fact would be instructed to come for their nex t routine screening in three years, rather than the one year typically recommended. Furthermore, the frustrations expressed by those in attendance about the unpredictability of the system, the sporadic nature of services offered and the shortage of resour ces in rural parts of the health care infrastructure these were issues that resonated with what I have been hearing and observing at CECAP T hese latter frustrations followed me through the afternoon interviews in the Gynecology Department. Sarah, a 36 yea r old farmer suffering from abdominal discomfort, ended her interview by stressing that her village needed to be put on the list for cervical cancer outreach, because she told us are really suffering out there, and it was difficult for them to come to St. Francis. Rahma who actually worked as a nurse i n a dispensary quite a ways from Ifakara told

PAGE 163

147 preventive screening due to a staff shortage. She was o nly able to come to St. Francis because she was experiencing actual symptoms. It was widely available, both personally and professionally, particularly because she knew their importance in detecting and identifying disease: at the dispensaries and health center. Sometimes people come with problems from home and when they get at a service center they miss a certain service. So wh ere are we going to put this person? Adapted from field notes and interview transcripts (September 6, 2013) Many of the issues restricting services for cervical cancer screening and treatment in Tanzania related to the health care system itself, includi ng non existent policy guidelines, an inadequate supply of materials, few trained screening or laboratory staff, slow turn around time for results, and even the mismanagement of resources when available, have been noted by researchers in the past, particul arly in the situation analyses conducted by Chirenje et al. (2001 ) and Ngwalle et al. (2001 ) Capacity bu ilding initiatives, like the one that brought CECAP to St. Francis, are actively working to address problems of shortage and training, and VIA has made a significant difference in reducing result wait times and loss to follow up. Unfortunately, uptake of p rogram services once they are in place is still poorly understood, though in the Tanzanian context, there has been a small flurry of rese arch interest in this subject over the past five years ( Kahesa et al., 2012 ; Lyimo & Beran, 2012 ; Perng et al., 2013 ; Peters et al., 2010 ) My primary research goal at the start of this project was focused on the second might infl uence attitudes towards screening. Despite my primary focus, while observing and en gaging with the program day after day at St. Francis I could not escape conversations and

PAGE 164

148 encounters that highlighted concerns related to these more system level issues, concerns about the position ing and sustainability of the screening program itself with in the context of the hospital and the overall health care system. For CECAP staff, these issues were not necessarily separate from that of utilization, but r ather intertwined. As the link between screening clients on the one hand, and the hospital and pro gram administration on the other the staff was often pulled in different directions by the priorities of these other actors as well as by their own These priorities were shaped by the limited health care infrastructure and the challenges inherent in see king care in a low resource setting While this was not the focus of my research, it has important implications not only for how cervical cancer prevention programs are implemented, but also for client utilization. Low Utilization At th e time that I conducted my research at St. Francis, CECAP was facin g two important challenges related to larger program goals The first issue was low client service utilization which I could see myself in the charts on the wall tracking monthly progress and in the program history provided by the Screening Map Forms. Of course, CECAP was not alone in this. Low utilization has been a persistent issue for screening services in developing nations, which served as one of the rationales for this project ( Agurto et al., 2006 ; Bessler, Aung, & Jolly, 2007 ; Bingham et al., 2 003 ; Tsu et al., 2013 ) But they wer e struggling to bring in clients more than quite other sites Jhpiego, who set the overall program goal to screen 100 women per month, had also expre ssed concerns with their low numbers and this in itself was a source of tension for the staff.

PAGE 165

149 The numbers had been promising at first for a brand new program; when launched in May 2010 they drew between 44 53 people each of the first three months. The st aff advertised the program through the use of posters and loudspeakers, and conducted a mass screening day in July that drew an additional 125 people from the surrounding area. But the numbers did not grow from there ; in fact, they did not even remain cons istent from month to month From 32 women in August to 75 in November, they then dropped back to 30 in December. The next year, 2011, was even more discouraging, with numbers sinking down to nine one month, five another. With the exception of two off site mass screening s in March 2011 (what they refer lients came for screening in any one month; most months were substantially lower. Though there had been no single digit months since 2011, the number of clients screened mon thly averaged only 33.8 in 2012 ( range 14 45). While I was at St. Francis in 2013 the average was similar (32.9) Six of the months from January to October 2013 saw 40 clients screened or more, but outreach events were held during three of these months (A pril, August, and September), inflating the numbers high above what was actually seen at the clinic itself In Figure 5 the trajectory of program utilization since its inception can be seen, though I have removed outreach and mass screening clients to bet ter reflect the normal utilization pattern. In Figure 6 I overlay this data with a dotted line that shows both clinic clients and those from the nine outreach events conducted in 2010, 2011, and 2013 (no outreach was done in 2012). Two multi day events to ok place in March 2011, resulting in particularly high numbers that month. Overall, CECAP has never in its history reached the progra m goal of 100 clients per month, unless it has also conducted exceeded 50

PAGE 166

150 clients in an individual month three times over four years all of which were in the first seven months of implementation. Figure 5 : Number of Clients Screened Per Month in Clinic, May 2010 October 2013 0 10 20 30 40 50 60 70 80 90 100 May 2010 Sep Jan 2011 May Sep Jan 2012 May Sep Jan 2013 May Sep Number of Clients Screened

PAGE 167

151 Figure 6 : Total Number of Clients Screened Pe r Month with Outreach May 2010 October 2013 Furthermore, CECAP was not experiencing the n umber of return clients expected heading into their fourth year The standard recommendation for women unde rgoing an initial cervical cancer screening is to return every three years for routine screening. W omen initially screened in 2010 would therefore be expected back in 2013 HIV positive women are one e xception to this recommendation; they are instructed to return every year. The other is those who are VIA positive during their initial screening. If a woman is positive for a cervical lesion, she is instructed to return for a one year follow up appointment. In reality, very few women were actually returning, whether for one year foll ow ups or for routine appointments For example, in four years, only 55 clients returned to CECAP for routine VIA screening, 85 .5 % ( n= 47) of whom were HIV positive. Granted, the year I conducted this fieldwork was the first year th e program 0 100 200 300 400 500 600 May 2010 Sep Jan 2011 May Sep Jan 2012 May Sep Jan 2013 May Sep Number of Clients Screened Clinic Only Total with Outreach

PAGE 168

152 expected VIA negative clients to return. However, even given that, the numbers were quite low particularly considering my research concluded with only two months left in the year. Table 5 illustrates the number of c lients per visit type per year. Table 5 : Clients Per Visit Type Per Year Year Visit Type Total Initial VIA Completed Postponed Cryotherapy One Year Follow Up Routine VIA Screening 2010 356 4 0 0 360 2011 223 0 8 13 244 2012 381 0 4 21 406 2013 308 0 0 21 3 29 Total 1268 4 12 55 1339* *(Visit type data not available for one client) A Program in Transition The second challenge facing CECAP was about support and sustainability. co ntext. As recounted in Chapter IV the cervical cancer prevention program arose as one of several program components Jhpiego implemented at St. Francis through the MAISHA initiative. Although it was part of this larger initiative, CECAP at St. Francis o perated largely as a vertical intervention, meaning it was a disease specific intervention. It was not integrated with these other Jhpiego program components, nor was it truly integrated with other reproductive health services. While representatives met wi th hospital administration when on site, Jhpiego maintained its supervisory role throughout the firs t four years of implementation. They provided training and supplies, they administered funding, and they made supervisory visits each quarter. Each month t he staff would submit the SMFs along with a record of the cryotherapy performed and a Monthly

PAGE 169

153 Summary Form to the Jhpiego office in Dar es Salaam for their records; these were eventually returned to St. Francis for storage once Jhpiego had collected the d ata. In 2014, this funding and oversight was scheduled to largely end. At that point, according to my informants, oversight and funding for CECAP would transition to the district go vernment. St. Francis Hospital would, according to the plan, incorporate CE CAP into the budget it submitted to the district each year. The district would dispense the funds to the hospital administration, which would in turn provide CECAP with the appropriate resources. The overall objective was to develop a sustainable program t hat 2). The transition had not yet taken place i n 2013 while I conducted research activities but it was on the horizon. In terms of carrying out the day to day activities of the program, this si tuation created challenges that influenced utilization of the program and at times made it more difficult for staff to do their work. On the Ground As the individuals who interac ted with clients each week, CECAP staff felt a responsibility to the women the y served. Both understood the importance of the program and their mission MN e screen we AL was particularly proud of what she had been able to accomplish with referrals and in obtaining a p unch biopsy device for the program: Yeah, what can I say? I am happy that I am able to transfer my cancer patients, especially to Ocean Road, because I made a channel for those who are cervical cancer positive. I am able to take a biopsy, and I am able to take it to Muhimbili to examine, and I am getting back the results, and I am sure that my clients are going for treatment.

PAGE 170

154 But they were constrained by their limited number. CECAP began at St. Francis with three individuals, all trained by Jhpiego: one p hysician and two nurses. A fourth joined and was trained the fol lowing year. T he staff had remained largely unchanged since that point in the sens e that all four members were, as of 2013, still involved though two in only a very peripheral sense One of t he original nurses had retired, though she still assisted on outreach days. Dr. M no longer conducted VIA exams, but as the head of the GD, he was likewise considered to be the head of CECAP. For all intents and purposes though, by 2013 CECAP was down to t wo staff members This continuity in staff was valuable because these individuals had seen the pro gram through from its inception and were intimately a cquainted with its processes. Nonetheless, their dwindling numbers were cause for concern not only am ongs t themselves but also for Jhpiego. linkages with other departments within the hospital, and promote the program more widely. CECAP w as simply one part of their job; both had many other demands on their the gynecology clinic, I am supposed to operate, so we have remarked when I asked her about her role. Having screeni ng concentrated to two days per week helped in some ways, and both women carried cell phones so they could be reached while they were at the main building. N onetheless, time spent at the CECAP clinic was time lost elsewhere, particularly if few women showe d up to be screened on a particular day. These other responsibilities also made it more difficult to promote the program, though the staff had certainly made efforts to do so. In the past, few media sources

PAGE 171

155 publicized information about cervical cancer to w omen at the local level in Tanzania ( Ngwalle et al., 2001 ) However, that has begun to change and radio in particular appears to be an important source of this information ( Perng et al., 2 013 ) Among the women we spoke with, this was certainly true: They [ the community ] surely know about cervical cancer as it was broadcasted on the radio several times. [Sandra] I used to hear about it in the radio, that there is breast cancer and cervica l cancer. [Subira] [Lilian] disease called cervical cancer. [Sarah] Dr. M told me that a t the start of the p rogram they had notices and posters up hiring a loudspeaker car to drive thro ugh town making announcement s T hat seemed to draw a larger number at least temporarily but it required funding. Since that point, however, both he and AL had gone on the local radio station several times to discuss cervical cancer and the services available at St. Francis, after which they saw an increase in clients. Still, Dr. M noted that the catchment area served by St. Francis was large and know whether we have such success unless we go there. Or unless somebody goes and P erhaps the most passionate topic for the CECAP staff was that of outreach in other words, traveling to villages to provide screening services. They felt deeply that

PAGE 172

156 outreach was an important part of their work at CECAP, despite not always feeling they had instit utional or program support : have money, and we need the most is means of transport. After we get transport, we can serve those people whether you pay us or not. We just n eed to serve those should take over. Then we decided to go for outreach as a sacrifice in order to serve women, because when we screen them, we find cancers. We just have to help the m. God will pay us. that they were willing to do it without being paid for their time. Staff members were also willing to give up their days off to do outreach, if they could marshal the resources to carry it out, and also recru ited me to join them twice while I was in the field so that I could see in person the importance of the work. Having worked in health care their entire careers, and being women in Tanzania themselves, the CECAP staff were well aware of the challenges their target population faced when it came to accessing screening or treatment, not only in terms of transportation and financial resources, but also things like long wait times, confusion with the health care system, and lack of relatives living near health ce nters with whom they could stay if necessary. They heard about these issues in the course of their daily work, just as I had from the participant in the opening of this chapter, and they often tried to find ways to overcome these issues, whether by returni ng from the hospital to CECAP to assistance for clients who had to travel to Morogoro or Dar es Salaam for treatment, or procuring a punch biopsy so that a tissue sample could be sent by bus to Dar es Salaam

PAGE 173

157 for analysis rather than the woman herself. Outreach was another way to bring services to women, to overcome some of these barrie rs, and they believed it was importa nt Hospital Context CECAP was in a challenging position b oth contextually within the organization of the hospital and physically, though the fact that it had been given its own room was viewed as evidence of administrative support for the program. Located at the very back of the RCH building across the road from the main hospital campus, it was not something that patients would see on their way to or from o ther services and it was not near to the Gynecology Department (though there were a couple of posters about the program posted in and near the GD, see Figure 7 ) The decision to have CECAP screen only two days a week was, according to my informants, an institutional one based on the limited availability of the staff. Although practical in that it meant the staff did no t have to constantly run back and forth, thi s likely contributed to low utilization. Cervical cancer screening was something that took place during certain hours at a certain location, rather taken a different approach with their program, providing screening every day as a routine service and tying it more closely to other reproductive health services ; according to Jhpiego, they had higher utilization. At the same time, one of my St. Francis informants noted, i f CECAP did find a way to increase utilization, they were like ly to create a demand they would not be able to meet given the current staff level

PAGE 174

158 Figure 7 : Cervical Cancer Posters Althoug h Jh pi e with linkages to other health services and departments, at present the program still operated largely as a vertical inter vention independent from the departments around it CECAP did rece ive referrals from the GD if a woman came in with suspicious symptom s, but women were not counseled on cervical cancer screening or informed about the program as a matter of course. AL had made presentations to various departments within the hospital, ye t there was no formal system in place and few referrals were made. The most promising development, and one that Jhpie go specifically encouraged in all of its CECAP sites, was a relationship with the CTC. Integrating HIV and cervical cancer screening has been called for in the literature more recently as well ( Belhadj et al., 2013 ; Plotkin et al., 2014 ) Give n that women with HIV were at higher risk for cervical cancer, there was support from both sides to establish a mutual referr al system. CECAP already provided PITC during intake, and frequently referred women to CTC for HIV testing. As noted in the opening to this chapter, this was sometimes made more difficu lt by shortages in supplies. T here was a significant period while I was in the field when no one at St.

PAGE 175

159 Francis c ould be tested for HIV ; this had occurred in the past as well But AL and the head of the CTC had had several discussions about the relationship between the two c linics and he had spent some time at CECAP getting to know the program. Just before leaving the field, I myself spent a day in the CTC and was taken through their new intake software which specifically asked if each patient ha d been referred for cervical cancer screening. The head of CTC excitedly pointed out that this question would now be a Based on observations and conversations with informants, w ith only a few months left until CECAP was turned over to the district and St. Francis, it was still viewed as a Jhpiego program by hospital administration. I had hoped to arrange an interview with a member of the administration to discuss the program and the upcoming transition but was unable to do so. I did hear from multiple sources that the administration was very supportive in terms of meeting with Jhpiego about the program and in supporting m other areas of the hospital. B ut when it came to department [ CECAP somewh that they di d not know what to expect in the near meeting, for example, in which she requested the use of a hospital vehicle for a one day outreach, the request was turned down. The veh icle could not be used for the program, we were told, although the staff had raised money on their own to pay for the gas and actual outreach activities. They were advised to ask Jhpiego for support instead Staff

PAGE 176

160 were not sure how the program operations m ight change or whether adequate funding would continue, and had already experienced difficulties with the administration earmarking program funding for certain uses and not others. The staff hoped to do more outreach, but were told that the funding could n ot be used for that purpose. They also expressed frustration that the supplies they received through an initial bout of district funding were not the same as those provided by Jhpiego. Certainly it was early in the process and the hospital administration h ad their own concerns and priorities in running a recently promoted referral hospital that was undergoing expansion. But at the same time, Larger Program Goals In creating a tiered system of CECAP sit es in Morogoro Region and elsewhere for capacity in the health care system. Although it began as a vertical intervention, when I met with a member of Jhpiego she stressed to me that the goal was for cervical cancer screening to become a routine service, linked with other health care delivery. As such, their priorities were focused on the health care system itself rather than outreach which was not viewed as sustainable. T hey provided some funding for this purpose, partic ularly at the beginning, but she pointed out that it was not a co st effective way to provide care, and rather than make screening routine, it took it out of the health facility completely. This put CECAP st aff a little at odds with Jhpi ego. For the staff members, outreach was a way to address low utilization and the many barriers women faced in coming to the hospital. They heard about these difficulties from clients themselves, who, as in our interviews, wou ld suggest or even request outreach. They felt a responsibility to

PAGE 177

161 help. Their efforts to increase utilization at St. Francis had not worked; it made sense to take the services to women who needed them. While Jhpiego attempted to be sensitive to this, thei r priority was the long term building of a system of services During our interview, my informant had many positive things to say about the efforts of the CECAP staff at St. Francis, remarking that they had had difficulties to contend with regarding their limited staff and screen ing hours. S he also noted that many of the screening providers they worked with at their various CECAP site s were very overworked. It was a systemic problem, she noted. But she maintained that outreach was not necessary to reach pro gram goals to screen 100 women per month and felt that the staff at St. Francis could n and in the surrounding area. S he pointed out that CECAP was located within Reproduct ive and Child Health, a building that primarily drew a population of target demogr aphic group that was largely untapped. She had stressed to both the hospital administration and the CECAP staff that they could and should train another s creening provider to help with any increase in clients, and to decrease pressure on the current staff. CECAP staff had already told me of plans to train one of the CTC nurses, but for whatever reason, this had not taken place. The Jhpiego informant was als o well aware of the potential difficulties involved in transitioning the program to district and hospital management, but again, this was part of the initiative to create sustainable programs. Despite regularly meeting with the relevant actors, she said that ensuring CECAP was included in the hospital budget was a challenge really aking on a more advisory role meant their recommendations might not always be followed. She herself felt that the

PAGE 178

162 decision to restric t screening to two days a week, for example, had hindered the program somewhat from integrating more fully into the overall hospital structure. But Jhpiego was she said, committed to continuing to provide some support in the form of limited outreach fundi ng and in making sure that the p rogram had supplies. For example, she told me that cryotherapy tanks, like the HIV testing reagent in the opening of this chapter, came from Dar es Salaam and distribution had been problematic. The transportation system was not really developed enough to ensure delivery and Jhpiego was working with programs to find a way to make sure they had ways to obtain the tanks within the health care infrastructure when Jhpiego stepped away. Shared Goal, Different Priorities Frost and Reich (2008 ) 8). The diffe rent stakeholders disc ussed in this chapter share the goal and ultimately utilization of cervical cancer screening However, they came to the table with different priorities and these priorities sometimes c onflict ed Furthermore, the different priorities of the staff, administration, and Jhpiego had an ongoing influence on where and when actual clients had access to screening services. In reflecti ng on some of the challenges to implementing CECAP and managing its i ntegration into the overall hospital structure, this chapter touches on larger conversations in the global health literature around the role of NGOs working within health care systems in developing countries in partnership with public entities and how ver tical interventions can contribute to (or in some cases detract from ) the overall health care

PAGE 179

163 system. The conversation on the relative merits of vertical approaches versus those that take a more horizontal approach in building primary health care has been ongoing for decades, go ing back to Alma Ata and before (e.g., Crofton, 2000 ; Fendall, 1987 ; Maher et al., 2010 ; Mills, 1983 ) In many developing countries, the structural adjustment programs that accompanied neoliberal reforms in the latter half of the Twentie th Century had grave effects on health care systems and access to health care ( Kamat, 2008 ; Pfeiffer & Nichte r, 2008 ) In Tanzania specifically where socialist policies enacted by the government following independence sought to provide widespread access to primary health care, particularly in rural areas, the health care system had rapidly expanded in a short period of time ( Hanson, 2000 ) The subsequent shift from the public to the private sector cau sed by neoliberal health reform and the impleme nta tion of fees for service were not ac companied by initially proposed social safety nets, nor the resources needed to run the expanded system ( Hanson, 2000 ; Kamat, 2008 ; Schraeder, 2000 ; Turshen, 1998 ) There are existing insurance and social security scheme s I saw posters and stickers for such programs at most of the health care centers I visited but they are either not widespread or poorly implemented. T he health care system, meanwhile, continues to experience deficiencies in staff, supplies, and infrastruc ture ( Kamat, 2008 ; Kwesigabo et al., 2012 ) Subsequent years have seen a rise in the role played by NGOs, and in health interventions planned around the funding priorities of donor countries and global health organization s These often focus on one or a limited number of diseases or health issues; Pfeiffer and Nichter call the funneling of aid to vertical int

PAGE 180

164 411). Vertical programs have been criticized for asking even more of already overstr etched health care workers, who are asked to take on additional activities or t o focus on program activities at the expense of other primary care duties ( Doherty e t al., 2010 ) Furthermore, in countries where attempts are being made to develop primary health care, vertical interventions may work outside of or in conflict with primary health care, thus weakening attempts at system strengthening ( Doherty et al., 2010 ; Fendall, 1987 ) Vertical interventions are also more likely to be subject to donor whims in terms of fun ding priorities, and therefore may not focus on neglected diseases ( Molyneux, Hotez, & Fenwick, 2005 ) On the other hand, vertical programs have been shown to be very successful in address ing certain diseases (e.g. Molyneux et al., 2005 ; Rao, Ramani, Hazarika, & George, 2014 ) They can play an important role in bringing specialized knowledge or skills to health care systems and personnel (Fendall 1987; Rao et al. 2014). If they work in conjunction with the primary health care system to strengthen that system, they can be mo ving forward vertical and horizontal programs ( Ooms, Damme, & Temmerman, 2007 ; Samb et al., 2009 ) However, as Mills pointed out as far back as 1983, the distinction between vert ical and horizontal programs is not organisational structures and the potential gap between structural arrangements and what to the incorporation of vertical intervention priorities and activities into the health care

PAGE 181

165 system itself but also into broader public health or public sector programs In the current case, although CECAP took the form of a vertical intervention intended to eventually be integrated into the overall St. Francis hospital structure, it looked very different fro m above. Jhpiego had partnered with the Tanzanian government and Ministry of Health and Social Welfare to offer technical assistance in developing national policy around the disease. They had also from the very start of the initiative implemented regional tiered programs based within health care centers capable of addressing advancing disease in order to build capacity for the treatment of those with cervical cancer From their perspective, they were engaged in system building. Inherent then in some of the challenges outlined in this chapter, were struggles between these more verti cal and horizontal perspectives; sometimes these struggles occurred even within the same individual The transition from Jhpiego to district and hospital oversight was a source of anxiety precisely because it was an attempt at integration, but one whose success was not necessarily assured. Given that the program had continually struggled with low utilization, there was also a concern that the d by the hospital administration moving forward. The efforts to increase referrals not only from departments within St. Francis but also other community programs an d the partnership with the CTC we re therefore important steps to address this issue, parti cularly given the reciprocal interest on the part of the CTC. I n the case of St. Francis and other similar prevention programs i t may be that integration 341).

PAGE 182

166 CHAPTE R IX CONCLUSION Dawa haitolewi bure Medicine is never given freely. Swahili proverb In choosing to explore how rural Tanzanian women understand cervical cancer and how these perceptions shape their utilization of screening services I was motivated in part by the lack of research with an anthropological and ethnographic perspective on the disease, particularly in sub Saharan and East Africa Cervical cancer in developing countries has until recently been largely viewed through a biomedical and public he alth e hand, and the health care system meant to serve them on the other (Kleinman et al., 2006, p. 140). However, the way women approach cervical cancer, screening, and treatment is not purely a phenomenon limited to the microcosm of the individual, community and clinic. National and international health priorities are felt and experienced at the mid and micro levels, yet the voices of the individuals who occupy these spheres often go unheard ing the local back into 463). This should be the case, I believe, with the current push to implement cervical cancer prevention programs in low resource settings as well. This dissertation therefore represents an attempt, in demonstrate how top

PAGE 183

167 cultural beliefs and previous experiences with the disease, b ut also through interactions with the biomedical system and larger discourse on the risks posed by sexual activity and reproduction shaped, at least in part, by the HIV pandemic in sub Saharan Africa. Furthermore, I have shown that their ability to access screening services is restrained by structural constraints that limit their agency. Yet at the same time, women are aware of the benefits of health services and find ways to work within these constraints (to a degree), leveraging social relationships and s upport to access needed health services when possible. Finally, I argue that up stream factors acting on the prevention program itself, including its structural and economic context, further conspire to either mitigate or ng services. Much of the expanding conversation around cervical cancer over the past fifteen years has arisen out of recent innovations in medicine and medical technology. One visit screen and treat approaches, like VIA, have broken down many of the barri ers to cervical cancer screening in developing nations attributable to structural and capacity issues within the health care system. They do not require laboratories or specialized medical personnel; they are low cost and low tech; they can be done quickly and reduce loss to follow up. In countries like Tanzania, where cervical cancer incidence and mortality is high and interest in addressing the issue is growing, these types of approaches have real potential to impact the burden caused by the disease if th ey are accessible, culturally acceptable, and utilized. Likewise, the HPV vaccine has profound potential to reduce the future incidence of cervical cancer if widespread vaccination campaigns are executed in regions with a high burden of disease.

PAGE 184

168 These glob al conversations have largely resulted in programs that are implemented in a top down manner and often involve rather vertical interventions. One of my goals in expanding the scope of this project beyond a focus on rural Tanzanian women to encompass the co ntext of the screening program itself has been to provide a place for discussion around how such approaches can come into conflict with, and at times ignore, local priorities and knowledge regarding how to best reach the intended targets of health interven tions. In the case of CECAP, for example, the metric that determined whether or not a program was being adequately utilized 100 women per month was set by an international NGO. This metric was further defined as only applying to women who were screened wit hin the clinic setting. This reflected the priorities of both the funder and the national government in terms of building the capacity of the health care system and reducing cost. Yet it must be asked, how do such metrics obscure alternative priorities re lated to social justice and health equity, such as to screen women most at risk and with the least access to health services? These alternative priorities were held by CECAP staff, as well as by the women they (and we) encountered both in the screening cli nic and gynecology department. While health care resources are finite, the focus on cost effectiveness can be pernicious when it comes at the expense of those who can least afford to access care, and Farmer (2005), for one, questions its validity: Certainl y, distributing these developments equitably would be expensive. Certainly, excess costs must be curbed. But how can we glibly use terms parlance? You want to help the poor? Then your

PAGE 185

169 agency of the people who are systematic ally marginalized and are subjected to structural enough to overcome the many barriers Tanzanian women, especially rural Tanzania women, face in utilizing the health care system. But even they were quick to note the difficulties involved both for themselves and for others who were not so fortunate. Increasing the capacity of the health care system would not benefit these women if they could not make it to the health care ce nter. CECAP staff, acutely aware of the challenges women faced, believed they could serve more women in need of screening by conducting outreach in addition to in clinic services, than through in clinic services alone. mean to suggest that the work of building up health care systems is unimportant and financial concerns inconsequential. Rather, I instead argue that assumptions of cost effectiveness should be closely examined and considerations of social justice and heal th equity given equal weight. In the present example, while it is true that outreach screening services had associated costs, including at various times vehicle rental, petrol, and staff lodgings and salary, these were balanced to a degree by the sheer num ber of women who could be served in a single day by taking screening services to the place where they lived. In addition, offering in clinic screening only passed on the costs of transportation and time to the clients, who were already burdened by poverty. uniquely positioned, by virtue of our global expertise and yet local focus to advocate such reforms and, at minimum, challenge the dominance of economics and efficiency based rationality i ).

PAGE 186

170 These medical innovations therefore come with complex concerns regarding how programs are locally perceived and implemented, and who is the recipient of their offerings. These should be examined critically to determine how access to such programs are shaped and constrained by poverty, gender inequality, and other social structural issues that impact health care behavior and access to resources. Both of these innovations are relatively recent and cannot have been expected to solve the problem of cervical cancer overnight. Nonetheless, the struggle to increase utilization of each highlights L improving disparities in the global cervical can cer burden are not simply due to Limitations This study is not without its limitations. The primary limitation is its cl inic based population, which was not representative of Tanzanian women in general and has several implications. The women who came to St. Francis were able to overcome many of the potential barriers to care that exist for Tanzanian women, particularly rura l Tanzanian women. That does not necessarily mean women did not encounter these barriers, only that they did not act as brick walls barring any means of seeking care. These women expressed strong support for biomedicine, its practitioners and treatments, a nd believed it to be superior to traditional medicine. Their belief in the effectiveness of biomedicine would seem to be supported by the fact that they sought care at St. Francis; it is more difficult to ascertain whether their expressed beliefs about tra ditional medicine were true, or were simply what they thought would be more acceptable to our ears. Certainly both

PAGE 187

171 my RAs and the informants I worked with believed that more participants were using traditional medicine than would admit to it, and remarked on occasions where women would come into the clinic with physical evidence of having done so, such as leaves inserted in to the vaginal canal This is relevant for this study both because it could perhaps mean that the use of traditional medicine or traditi onal medical practitioners for cervical cancer is occurring among our interview population but we were not told about it, or that women who use traditional medicine were not represented among the population of women we interviewed. Taking this a step furth er, women who do not have similar faith in biomedicine may not seek out the services of clinics at all, in which case they are unlikely to make use of cervical cancer screening services. Language barriers were also a limitation. Although I studied Kiswahil i for several months prior to leaving for Tanzania, and intensively for two months upon my arrival, my language abilities were limited. My Kiswahili certainly helped me to establish rapport with many of the interview participants and with CECAP clients, an d I was able to follow along during the interviews to a degree, but details during in depth conversations or fast paced discussion often escaped me. In order to counteract my ignorance, whenever possible, I checked in with my RAs and the CECAP staff to det ermine if I had understood things correctly. I also asked a lot of questions. I went over each interview with my RAs and discussed nuances in language and meaning, to ensure correct translation. Any documents collected during my review of secondary data wi th Kiswahili terms were likewise reviewed with native speakers. Finally, this study had to be condensed due to a limited time frame. Originally planned for an eight to nine month period of time, data collection instead took place

PAGE 188

172 over six months, with int erviews largely compressed into three months of that period. As a result, I could only conduct a small number of specialized informant interviews, and as discussed in Chapter III, we had to adopt different methods of reviewing transcripts than anticipated. The majority of the time in the field was spent on data collection with an eye towards the shortened period, and while I conducted some limited analysis in the form of field notes and early exploratory coding, most of the analysis occurred back in the Uni ted States. During this time I remained in contact with my RAs however, and we discussed issues with the data as they arose. Future Recommendations While the focus of my research was on a specific screening program and population of women, I believe there are broader practical insights to consider moving forward in designing cervical cancer prevention programs that better balance the priorities of those at the top in charge of implementation with those the programs are meant to serve. Primary among these in sights are those that address the key issue facing CECAP low utilization. Different informants had their own theories as to why this was an issue in the context of St. Francis, but what seems clear from this research is that from a systemic perspective, CE organization was problematic to say the least. First then, is to note that an internal referral system, particularly linking key departments targeting women at risk, such as the HIV clinic and Obstetrics /Gynecology, is essential to capture existing hospital patients who have already overcome barriers in access to health centers. Furthermore, given that an important risk factor for cervical cancer in many developing nations is the often high birth rate, sc reening programs should

PAGE 189

173 take advantage of potential links to maternal and child health clinics. This should be a key part of program planning from the start, rather than a later deployment in response to recruitment issues, with efforts to involve not only the institutional administration, but also mid level staff in relevant departments. Certainly there is support for this in the literature, with calls to integrate HIV and cervical cancer screening specifically gaining traction in recent years (Belhadj et al., 2013; Kahesa et al., 2008; Plotkin et al., 2014). Furthermore, Mbulaiteye et al. (2011) have highlighted the need for collaboration and the sharing of data between HIV and, not only cervical cancer, but other types of cancer programs as well in recog nition of the dual burden of communicable and non communicable diseases in developing nations. The results of this study would seem to suggest the strength of this approach, and going forward dual screening programs should be explored. Two considerations s hould be taken into account with this type of program however. First, that the transfer of a potentially stigmatizing condition, such as HIV, is not transferred onto or does not increase the stigma of cervical cancer. And second, that the importance of scr eening for all women, not just high risk women, is still relayed to potential screening clients. Incorporating referrals for cervical cancer screening into HIV and reproductive health clinic protocols, or even into the overall procedures of a hospital, may also serve to routinize screening in high risk populations. This brings us to a second recommendation: to offer screening as a routine reproductive health service, when possible. This does not necessarily mean as part of an annual gynecological exam; in T anzania and many other developing nations, such exams are not likely to occur. Rather, that screening should be available during normal health center business hours. When screening is offered as a

PAGE 190

174 routine service, rather than as something only available at certain times on certain days, screening programs are more likely to catch women at risk who come through the hospital doors. Meanwhile, patients themselves can rely on services being available when they need or are referred to them. Third, it is importan t to gain buy in from the institution and the program staff as well as ongoing input. This should go beyond consent to implement the program if the program staff b ased on their own experiences working with clients. Additionally, this feedback should also take into account the often over stretched nature of health care employment in an environment of scarcity and the limited time staff may have available for extracur ricular activities outside their normal job duties, such as client recruitment and new staff training. Resources to assist in these activities should be built into program execution. Next, cervical cancer prevention programs should build on existing knowle dge and culturally appropriate methods of knowledge transfer. This research highlighted the difference, in a population of rural Tanzanian women, between knowledge and simple awareness that a disease exists. However, even where the former is low, educators can use the latter to build on local understandings of biomedical models of disease transmission and risk behavior. In addition, when it comes to screening services, trusted lay health educators, such as the Community Health Workers mentioned in this stud y, or of mouth advertising.

PAGE 191

175 Furthermore, a key piece missing in many areas of sexual and reproductive health intervention is the role of men, despite research demonstrating that they have an important part to play (Rosser, Zakaras, Hamisi, & Huchko, 2014). Women we spoke with clearly felt their husbands were sources of support, as well as important partners in the process of seeking screening or treatment for cervical cancer. However, male partners can also act as a barrier. Finding ways to educate men and encouraging them be advocates for their wives to undergo screening could have a beneficial effect on utilization. Finally we come to another significant tool, outreach. It i s important to recognize the huge barriers that transportation cost and time represent in many developing nations, and the role that outreach can play in leveling the field, so to speak. Screening staff interaction, and moreover may in fact be members of the same communities. In this way, they may be acutely aware personally of the barriers women face in seeking health care services in low resource settings, in a way that funders or NGOs may not be. Top d own approaches that do not provide funding for outreach activities when they are valued by screening providers risk frustrating staff members and devaluing their emic knowledge, resulting in push back or resistance. They may also gloss over the potentially important publicity outreach can provide for screening programs. Moving Forward While initiatives like that resulting in CECAP indicate a promising start in a global health focus on cervical cancer, much yet remains to be done. The last decade and a half have seen a series of important developments in cervical cancer screening and

PAGE 192

176 treatment. Since the verification of HPV as a necessary cause of the disease at the turn of the century, not only have one visit approaches made screening more practical in low r esource settings, but HPV testing and the HPV vaccine have been incorporated into discussions around cancer control in developing countries as well. Already the Tanzanian government has begun implementing vaccination among schoolgirls on a broader scale, f ollowing in the footsteps of neighboring Rwanda. However useful each of these technologies will prove to be in reducing the burden of cervical cancer in developing regions depends on our understanding of many of the issues examined in this study. While loc al perceptions around cervical cancer are important to understand, as Farmer (1999) has argued, these issues must be examined with a concern for larger inequalities and power structures or they risk ascribing health inequalities to cultural or individual d ifference In the case of cervical cancer, these larger policies and structures not only mold access to screening and treatment, they influence risk for the disease itself. Important avenues of future research remain if we are to design more effective meth ods to reduce the burden of cervical cancer in developing nations, including investigations into more representative populations and programs beyond the outreach, urban, and semi urban settings, the efficacy of outreach versus routine screening, and the ro le of government in partnering with non governmental actors to implement these programs. In particular, it remains to be seen how well these programs will thrive once turned over to government support and oversight. Again, these are complex issues that cal l for a nuanced approach The political economy of health perspective employed in this study shines a light on multilevel interactions within the context of larger global processes that have shaped the current

PAGE 193

177 outlines of the global cervical cancer burden and contributed to it becoming a national health priority in Tanzania. Approaches that center only on the interaction between patients and providers for example, or to the opposite end, take a top down view focused on existing resources in the health care system, may provide valuable insight but lack context and miss the interactions among these phenomena. Local knowledge and perceptions of cervical cancer shape health seeking behavior; this in turn influences how programs are structured and, depending on u tilization, how successful these programs are seen to be. At the same time, the priorities of governments, funders, and mid level administration also exert their own influence on cervical cancer prevention programs, shaping the contours of the services the y offer and by extension, access to those services. In closing, Dressl studying health, we have stumbled into an arena that demands that we understand how structure and cultural construction i ntersect, because that intersection leaves its mark on 457).

PAGE 194

178 REFERENCES Abotchie, P. N., & Shokar, N. K. (2009). Cervical cancer screening among college s t udents in Ghana: Knowledge and health b eliefs. Int ernational Journal of Gynecological Cancer, 19 (3), 412 416. African Cancer Registry Network. (2015). Afr ican Cancer Registry Network. Retrieved March 15, 2015, from http://afcrn.org/ Agurto, I. (2001). Bridging distance s: Preventive services and women's concerns. Washington DC: Pan American Health Organization. Agurto, I., Bishop, A., Sanchez, G., Betancourt, Z., & Robles, S. (2004). Perceived barriers and benefits to cervical cancer screening in Latin America. Preventi ve Medicine, 39 91 98. Agurto, I., Sandoval, J., Rosa, M. d. l., & Guardado, M. E. (2006). Improving cervical cancer prevention in a developing country. In ternational Journal for Quality in Health Care, 18 (2), 81 86. Anasel, M. G., & Mlinga, U. J. (20 14). Determinants of contraceptive use among married women in Tanzania: Policy implication. African Population Studies, 28 (2 Supplement), 978 988. Anderson, G. F ., & Chu, E. (2007). Expanding priorit ies C onfronting chronic disease in countries with low i ncome. The New England Journal of Medicine, 356 (3), 209 211. Anorlu, R. I. (2008). Cervical cancer: T he sub Saharan African perspective. Reproductive Health Matters, 16 (32), 41 49. Arbyn, M., Castellsagu, X., de Sanjose, S., Bruni, L., Saraiya, M., Br ay, F., & Ferley, J. (2011). Worldwide burden of cervical cancer in 2008. Annals of Oncology, 22 2675 2686. Awodele, O., Adeyomoye, A. A., Awodele, D. F., Kwashi, V., Awodele, I. O., & Dolapo, D. C. (2011). A study on c ervi cal cancer screening amongst n urses in Lagos University Teaching Hospital, Lagos, Nigeria. Journal of Cancer Education, 26 (3), 497 504. Badahdah, A. M. & Pedersen, D. E. (2011). "I want to stand on my own l egs": A qualitative study of antiretroviral therapy adherence among HIV posit ive women in Egypt. AIDS Care, 23 (6), 700 704. Baer, H., Singer, M., & Susser, I. (2003). Medical Anthropology and the World System: A Critical Perspective (2 nd ed.). Westport, Connecticut: Praeger Publishers.

PAGE 195

179 Baer, H. A., Singer, M., & J ohnsen, J. H. (1 986). Toward a critical medical a nthropology. Social Science & Medicine, 23 (2), 95 98. Behrman, J. R., Kohler, H. P., Watkins, S. C., & Kenan Jr, W. R. (2002). Social networks and changes in contraceptive use over time: Evidence from a longitudinal s tudy in r ural Kenya. Demography, 39 (4), 713 738. Belhadj, H., Rasanathan, J. J. K., Denny, L., & Broutet, N. (2013). Sexual and reproductive health and HIV services: Integrating HIV/AIDS and cervical cancer prevention and control. International Journal of Gy naecology and Obstetrics, 121 (2013), S29 S34. Berkman, L. F., & Kawachi, I. (2000). Social Epidemiolgy (L. F. Berkm an & I. Kawachi Eds. 1st ed.). Oxford: Oxford University Press. Bernard, H. R. (2006). Research Methods in Anthropology: Qualitative and Quantitative Approaches (4th ed.). Oxford: AltaMira Press. Bessler, P., Aung, M ., & Jolly, P. (2007). Factors affecting uptake of cervical cancer screening among clinic a ttendees in Trelawny, Jamaica. Cancer, Culture and Literacy, 14 (4), 396 404. Bhatla N., Classen, M., Denny, L., Fagan, J., Franceschi, S., Gueye, S. M., . zur Hausen, H. (2010). Protection against cancer causing infections (pp. 1 56). Geneva: International Union Against Cancer. Bingham, A., Bishop, A., Coffey, P., Winkler, J., Brad ley, J., Dzuba, I., & Agurto, I. (2003). Factors affecting utilization of cervical cancer prevention services in low resource settings. Salud Pbica de Mxico, 45 (S3), S408 S416. Bingham, A., Drake, J. K., & LaMont agne, S. (2009). Sociocultural i ss ues in the introduction of human papillomavirus vaccine in low resource s ettings. Archives of Pediatric and Adolescent Medicine, 163 (5), 455 461. Bloom, J. R., Stewart, S. L., Johnston, M., Banks, P., & Fobair, P. (2001). Sources of support and the physical an d mental well being of young women with breast cancer. Social Science & Medicine, 53 1513 1524. Bomfim, S., Santana Franco, E., & Bahamondes, L. (2005). Visual inspection with acetic acid for cervical cancer detection. International Journal for Gynecolo gy and Obstetrics, 88 (1), 65 66. Bosch, F. X., Ma nos, M. M., Munoz, N., Sherman, M., Jansen, A.M., Peto, J., . Shah, K.V. (1995). Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. Journal of the National Cancer Institute 87 (11), 796 802.

PAGE 196

180 Bourdieu, P. (1980). The Logic of Practice Stanford, California: Standford University Press. Bradley, J., Coffey, P., Arrossi, S., Agurto, I., Bingham, A., Dzuba, I., . White, S. C. (2006). Women's perspectives on cervical screen ing and treatment in developing c ountr ies: Experiences with new technologies and service delivery s trategies. Women & Health, 43 (3), 103 121. Buskens, I., & Bradley, J. (2002). Women's perspectives on cervical cancer prevention procedures New York: Enge nderHealth. Cagney, H. (2013). GAVI to fund HPV vaccines in low income countries. The Lancet Oncology, 14 (3), e92. Campbell, J., Mwami, J., & Ntukula, M (1995). Urban social o rganiza tion: An exploration of kinship, s ocia l n etwo rks, gender relations and household and c ommunity in Dar es Salaam. In C. Creighton & C. K. Omary (Eds.), Gender, Family and Household in Tanzania (pp. 221 252). Aldershot: Avebury. Castellsagu, X. (2008). Natural history and epidemiology of HPV infection and cervical cancer. Gy necologic Oncology, 110 S4 S7. Castellsagu, X., Bruni, L., Alemany, L., Diaz, M., de Sanjose, S., & Bosch, F. X. (2012). The e pidemiology of cervical c ancer. In F. Borruto & M. D. Ridder (Eds.), HPV and Cervical Cancer: Achievements in Prevention and F uture Prospects (pp. 63 84). New York: Springer Castellsagu, X., & Muoz, N. ( 2003). Chapter 3: Cofactors in human papillomavirus c arcinog enesis Role of parity, oral contraceptives, and tobacco s moking. Journal of the National Cancer Institute Monograph s (31), 20 28. Castle, S. (2003). Factors influencing young Malians' reluctance to use hormonal c ontraceptives. Studies in Family Planning, 34.3 (Sep 2003). Centers for Disease Control. (2001). The global HIV and AIDS e pidemic, 2001. Morbidity and Morta lity Weekly Report, 50 (21), 434 439. Centers for Disease Control. (2004, 2004). Genital H PV Infection CDC Fact Sheet. Retrieved September 27, 2006, from http://www.cdc.gov/std/HPV/STDFact HPV.htm Centers for Disease Control. (2007). HPV and Men CDC Fact Sheet. Retrieved September 27, 2006, from http://www.cdc.gov/std/HPV/STDFact HPV.htm

PAGE 197

181 Chan, C. W. H., Yang, S. B., Gu, C., Wang, X., & Tao, L. (2014). Perception of cervical cancer risk and screening behavior: A literature r eview. International journal of Nursing Knowledge, 26 (1), 2 18. Chirenje, Z. M., Rusakaniko, S., Kirumbi, L., Ngwalle, E. W., Makuta Tlebere, P., Kaggwa, S., . Makoae, L. (2001). Situation analysis for cervical cancer diagnosis and treatment in East, Central and Southern African countries. Bulletin of the World Health Organization, 79 (2), 127 132. Chirwa, S., Mwanahamuntu, M., Kapambwe, S., Mkumba, G., Stringer J., Sahasrabud dhe, V., . Parham, G. (2010 ). Myths and misconceptions about cervic al cancer among Zambian women: R apid assessment by peer educators. Global Health Promotion, 17 (Supplement), 47 50. Clifford, G. M., Smith, J. S., Plummer, M. L., Muoz N., & Franceschi, S. (2003). Human papillomavirus types in invasive cervical cancer worldwide: A meta analysis. British Journal of Cancer, 88 (1), 63 73. Crofton, J. (2000). Reforms to the health sector must retain vertical programmes like those for tu berculosis. BMJ, 320 (7251), 1726. Crum, C. P. C., Abbott, D. W., & Quade, B. J. (2003 ). Cervical cancer s c reening: From the Papanicolaou smear to the vaccine e ra. Journal of Clinical Oncology, 21 (10s), 224s 230s. Dailard, C. (2003) HPV in the United States and developing nations: A problem of public h e alth or p olicy? The Guttmacher Report, August 4 5. Davies, S. E. (2010). Global Politics of Health Cambridge: Polity Press. Denny, L. (2005). The prevention of cervical cancer in developing countrie s. BJOG: A n International Journal of Obstetrics and Gynaecology, 112 1204 1212. DeWalt, K. M., & DeWalt, B. R. (2010). Participant Observation: A Guide for Fieldworkers (2 nd ed.): AltaMira Press. Dilger, H. (2003). Sexuality, AIDS, and the lures of mod ernity: Reflexivity and morality among young people in rural Tanzania. Medical Anthropology: Cross Cultural Studies in Health and Illness, 22 (1), 23 52. Doh, A. S., Nkele, N. N., Achu, P., Essimbi, F., Essame, O., & Nkegoum, B. (2005). Visual inspection with acetic acid and cytology as screening methods for cervical lesions in Cameroon. International Journal of Gynaecology and Obstetrics, 89 (2), 167 173.

PAGE 198

182 Doherty, T., Chopra, M., Tomlinson, M., Oliphant, N., Nsibande, D., & Mason, J. (2010). Moving from vertical to integ rated child health programmes: E xperiences from a multi country assessment of the Child Health Days approach in Africa. Tropical Medicine & International Health, 15 (3), 296 305. Douglas, M. (1970). Natural Symbols: Exploratoins in Cosmol ogy London: Routledge. Douglas, M., & Wildavsky, A. (1982). Risk and Culture Berkeley: University of California Press. Doyal, L. (1995). What Makes Women Sick: Gender and the Political Economy of Health New Brunswick, New Jersey: Rutgers University Pr ess. Drain, P. K., Holmes, K. K., Hughes, J. P., & Koutsky, L. A. (2002). Determinants of cervical cancer rates in developing countries. International Journal of Cancer, 100 (2), 199 205. D ressler, W. W. (2001). Medical anthropology: Toward a third momen t in social s cience? Medical Anthropology Quarterly, 15 (4), 455 465. Drew, E. M., & Schoenberg N. E. (2011). Deconstructing f atalism: Ethnographic perspectives on women's decision making about cancer prevention and t reatment. Medical Anthropology Quarte rly, 25 (2), 164 182. Dunlop, D. W., Godiksen, L. H., Micka, M. A., Muhondwa, E., Mujinja, P., & Stinson, W. (1990). The Sustainability of U.S. Supported Health, Population, and Nutrition Programs in Tanzania: 1971 1988 U.S. Agency for International Dev elopment. Dzuba, I. G., Caldern, R., Bliesner, S., Luciani, S., Amado, F., & Jacob, M. (2005). A participatory assessment to identify strategies for improved cervical cancer prevention and treatment in Bolivia. Pan American Journal of Public Health, 18 (1 ), 53 63. Ellis, A., Blackden, M., Cutura, J., MacCulloch, F., & Seebens, H. (2007). Gender and Economic Growth in Tanzania Washington DC: The International Bank for Reconstruction and Development/ The World Bank. Farmer, P. (1997). On suffering and st ructural violence: A view from b elow. In A. Kleinman, V. Das & M. Lock (Eds.), Social Suffering (pp. 261 284). Berkeley: University of California Press. Farmer, P. (1999). Infections and Inequalities: The Modern Plagues Berkeley: University of California Press. Farmer, P. (2005). Pathologies of Power Berkeley: University of California Press.

PAGE 199

183 Farmer, P., & Kleinman, A. (1989). AIDS as human s uffering. Daedalus, 118 (2), 135 160. Feldman Savelsberg, P., Ndonko, F. T., & Schmid t Ehry, B. (2000). Sterilizi ng vaccines or the politics of the w omb: Retrospective study of a r umor in Cameroon. Medical Anthropology Quarterly, 14 (2), 159 179. Fendall, R. (1987). The integration of vertical programmes into primary health c are. Third World Planning Review, 9 (3), 2 75 284. Ferlay, J., Soerjomataram, I., Ervik, M., Dikshit, R., Eser, S., Mathers, C., . Bray, F. (2013). GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: I ARC CancerBase No. 11 [Internet] Lyon, France: International Agency for Research on Cancer. Retrieved from http://globocan.iarc.fr accessed on 16 March 2015. Fort, V. K., Makin, M. S., Siegler, A. J., Ault, K., & Rochat, R. (2011). Barriers to cervical cancer screening in Mulanje, Malawi: A qu alitative study. Patient Preference and Adherence, 5 125 131. Foster, G. (2000). The capacity of the extended family safety net for orphans in Africa. Psychology, Health & Medicine, 5 (1), 55 62. Franco, E. L., Duarte Franco, E., & Ferenczy, A. (2001). Cervical cancer: Epidemiology, prevention and the role of human p apillomavirus i nfection. CMAJ, 164 (7), 1017 1025. Franco, E. L., & Harper, D. M. (2005). Vaccination against h uman papillomavirus infection: A new paradigm in cervical cancer control. Vacc ine, 23 (17 18), 2388 2394. Franco, E. L., Schlecht, N F., & Saslow, D. (2003). The epidemiology of cervical c ancer. The Cancer Journal, 9 (5), 348 359. Frenk, J., Bobadilla, J. L., Stern, C., Frejka, T., & Lozano, R. (1991). Elements for a theory of th e health transition. Health Transition Review, 1 (1), 21 38. Frost, L. J., & Reich, M. R. (2008). Access: How do good health technologies get to poor people in poor countries? Cambridge: Harvard University Press. Fuster, V., & Vote, J. (2005). MDGs: C hr onic diseases are not on the agenda. The Lancet, 366 (9496), 1512 1514. Galtung, J. (1969). Violence, peace, and peace r esearch. Journal of Peace Research, 6 (3), 167 191.

PAGE 200

184 Garbus, L. (2004). HIV/AIDS in Tanzani a. San Francisco: AIDS Research Institute, AI DS Policy Research Center. GAVI. (2013). GAVI funds vaccines to protect g irls against cervical cancer. Retrieved from http:// www.gavi.org/library/news/press releases/2013/gavi funds vaccines to protect girls against cervical cancer/ Geertz, C. (1985). Local Knowledge: Further Essays in Interpretive Anthropology (2nd ed.). New York: Basic Books. Geneau, R., Lewallen, S., Brons ard, A., Paul, I., & Courtright, P. (2005). The social and family dynamics behind t he uptake of cataract surgery: F indings from Kilimanjaro Region, Tanzania. British Journal of Opthalmology, 89 1399 1402. Ghosh, P., Gandhi, G., Kochhar, P. K., Zutshi, V ., & Batra, S. (2012). Visual inspection of cervix with Lugol's iodine for early detection of premalignant & malignant lesions of cervix. Indian Journal of Medical Research, 136 (2), 265 271. Goel, A., Gandhi, G., Batra, S., Bhambhani, S., Zutshi, V., & S achdeva, P. (2005). Visual inspection of the cervix with acetic acid for cervical intraepithelial lesions. International Journal of Gynaecology and Obstetrics, 88 25 30. Goffman, E. (1974). Stigma: Notes on the Management of Spoiled Identity (1st ed.). Lanham, MD: Jason Aronson. Gregg, J. (2003). Virtually Virgins Stanford, California: Stanford University Press. Gustafsson, L., Pontn, J., Zack, M., & Adami, H. O. (1997). International incidence rates of invasive cervical cancer after introduction of cytological screening. Cancer Causes Control, 8 (5), 755 763. Gusterson, H. (1997). Studying Up Revisited. Political and Legal Anthropology Review, 20 (1), 114 119. Hanson, S. (2000). Health sector r eform and STD/AIDS control in resource poor settings T he c ase of Tanzania. International Journal of Health Planning and Management, 15 341 360. Haram, L. (2005). AIDS and risk: The handling of uncertainty in northern Tanzania. Culture, Health & Sexuality, 7 (1), 1 11. Heaney, C. A., & Israel, B. A. (1997) Social Networks and Social Support. In K. Glanz, B. K. Rimer & F. M. Lewis (Eds.), Health Behavior and Health Education (2nd ed.) San Francisco, CA : Jossey Bass.

PAGE 201

185 Hetzel, M. W., Alba, S., Fankhauser, M., Mayumana, I., Lengeler, C., Obrist, B., . Msh inda, H. (2008). Malaria risk and access to prevention and treatment in the paddies of the Kilombero Valley, Tanzania. Malaria Journal, 7 (7). Hollos, M., & Larsen, U. (2008). Motherhood in sub Saharan Africa: The social consequences of infertility in an urban population in northern Tanzania. Culture, Health & Sexuality, 10 (2), 159 173. Hoque, M., Ibekwe, C. M., & Ntuli Ng cobo, B. (2009). Screening and perceived severity of c e rvical c ancer among women attending Mahalapye District Hospital, Botswana. Asian Pacific Journal of Cancer Prevention, 10 1095 1100 Hutubessy, R., Levin, A., Wang, S., Morgan, W., Ally, M., John, T., & Broutet, N. (2012). A case study using the United Re public of Tanzania: C osting nationwide HPV vaccine delivery using the WHO Cervic al Cancer Prevention and Control Costing Tool. BMC Medicine, 10 136 145. ICF International. (2004 2012). Demogr aphic and Health Surveys (Version 1.5.2 ) [Dataset ] Retrieved from http://www.statcompiler .com Ifakara Health Institute. (2014). Who We Are. R etrieved from http://www.ihi.or.tz/about/who we are Ishikawa, K. (1999). Nation Building and Development Assistance in Africa: Different but Equal London: MacMillan Press Ltd. Janes, C.R. (2004). Goi ng global in Century XXI: Medical anthropology and the new primary health care. Human Organization, 63 (4), 457 471. Jeronimo, J., & Schiffman, M. (2006). Colposcopy at a crossroads. American Journal of Obstetrics & Gynecology, 195 (2), 349 353. Jhpiego. ( 2015). Jhpiego Hands Over Cervical Cancer Prevention Equipment to Benefit Women in Tanzania. Retrieved from http://www.jhpiego.org /content/jhpiego hands over cervical cancer prevention equipment benefit women tanzania Jones, S. B. (1999). C ancer in the developing world: A call to action. BMJ, 319 (7208), 505 508. Jordan, B. (1983). Birth in Four Cultures Montral: Eden Press. Ka hesa, C., Kjaer, S., Mwaiselage, J., Ngoma, T., Tersbol, B., Dartell, M., & Rasch, V. (2012). Determinants of acceptance of cervical cancer screening in Dar es Salaam, Tanzania. BMC Public Health, 12 1093 1100.

PAGE 202

186 Kahesa, C., Mwaiselage, J., Wabinga, H. R. Ngoma, T., Kalyango, J. N., & Karamagi, C. A. S. (2008). Association between invasive cancer of the cervix and HIV 1 infection in Tanzania: the need for dual screening. BMC Public Health, 8 262 Kamat, V. R. (2008). Dying under the bird's shadow: Narr ative r epresentations of d egedege and child s urvival among the Zaramo of Tanzania. Medical Anthropology Quarterly, 22 (1), 67 93. Kamat, V.R. (2013). Silent Violence: Global Health, Malaria, and Child Survival in Tanzania Tuscan: University of Arizona Pr ess. Kanavos, P. (2006). The rising burden of cancer in the developing world. Annals of Oncology, 17 (Supplement 8), viii15 viii23. Kangalawe, R. Y. M., & Liwenga, E. T. (2005). Livelihoods in the wetlands of Kilombero Valley in Tanzania: Opportunities a nd challenges to integrated water resource management. Physics and Chemistry of the Earth, Parts A/B/C, 30 (11 16), 968 975. Kapiga, S. H., Shao, J. F., Lwihula, G. K., & Hunter, D. J. (1994). Risk factors for HIV infection among w omen in Dar es Salaam, T anzania. Journal of Acquired Immune Deficiency Syndromes, 7 301 309. Kato, F. (2007). Development of a major rice cultivation area in the Kilombero Valley, Tanzania. African Study Monographs, Supplement, 36 3 18. Khozaim, K., Orang'o, E., Christoffer sen Deb, A., Itsura, P., Oguda, J., Muliro, H., . Washington, S. (2014). Successes and challenges of establishing a cervical cancer screening and treatment program in western Kenya. International Journal of Gynaecology and Obstetrics, 124 12 18. Kid anto, H. L., Kilewo, C. D., & Mos hiro, C. (2002). Cancer of the cervix: Knowledge and attitudes of female patients a dmitted at Muhimbili National Hospital, Dar es Salaam. East African Medical Journal, 79 (9), 467 475. Kielman n (1998). Baren ground: Conte sting identities of infertile women in Pemba, Tanzania. In M. Lock & P. A. Kaufert (Eds.), Pragmatic Women and Body Politi cs (pp. 127 163 ). Cambridge, UK: Cambridge University Press. Kleinman, A. (1988). The Illness Narratives : BasicBooks, Inc. Kleinman, A., Eisenberg, L., & Good, B. (200 6). Culture, illness, and care: Clinical lessons from anthropologic and cross cultural r es earch Focus: The Journal of Lifelong Learning in Psychiatry, IV (1), 140 149.

PAGE 203

187 Knodel, J., Watkins, S., & VanL andingham, M. (2003) AIDS and older persons: An i nternation al p erspective. Journal of Acquired Immune Deficiency Syndromes, 33 S153 S165. Koda, B. (1998). Changing land tenure systems in the contemporary matrilineal social system: The gendered d imension. In P. Seppala & B Koda (Eds.), The Making of a Periphery: Economic Development and Cultural Encounters in Southern Tanzania (pp. 195 221 ). Uppsala, Sweden: Nordiska Afrikainstitutet. Kominami, M., Kawata, K., Ali, M., Meena, H., & Ushijima, H. (2007). Factors determining prenatal HIV testing for prevention of mother to child transmission in Dar Es Salaam, Tanzania. Pediatrics International, 49 (2), 286 292. Krishnan, S. S. (2008). The HPV Vaccine Controversy Westport, Connecticut: Praeger Publishers. Kwesigabo, G., Mwa ngu, M. A., Kakoko, D. C., Warriner, I., Mkony, C. A., Killewo, J., . Freeman, P. (2012). Tanzania's health system and workforce crisis. Journal of Public Health Policy, 33 (S1), S35 S44. Lakoff, G. (1987). Women, Fire, and Dangerous Things: What Cate gories Reveal About the Mind Chicago: University of Chicago Press. LeCompte, M. D., & Schensul, J. J. (2010). Designing and Conducting Ethnographic Research (2nd ed.). Lanham: AltaMira Press. Lock, M. M., & Scheper Hughes, N. (1990). A Critical Interpre tive Approach in Medical Anthropology: Rituals and Routines of Discipline and Dissent. In T. M. Johnson & C. Sargent (Eds.), Medical Anthropology: Contemporary Theory and Method New York: Praeger. Louie, K. S., de Sanjose, S., & Mayaud, P. (2009). Epidem iology and prevention of human papillomavirus and cervical cancer in sub Saharan Africa: a comprehensive review. Tropical Medicine & International Health, 14 (10), 1287 1302. Lwy, I. (2011). A Woman's Disease Oxford: Oxford University Press. Lugalla, J L. ( 1995). The impact of structural adjustment policies on women's and children's h ealth in Tanzania. Review of African Political Economy, 63 43 53. Lyimo, F. S., & Beran, T. N. (2012). Demographic, knowledge, attitudinal, and accessability factors as sociated with uptake of cervical cancer screening among women in a rural district of Tanzania: Three public policy implications. BMC Public Health, 12 22.

PAGE 204

188 Maher, D., Smeeth, L., & Sekajugo, J. (2010) Health transition in Africa: P ractical policy propos als for primary care. Bulletin of the World Health Organization, 88 943 948 Maman, S., Mbwambo, J. K., Hogan, N. M., Weiss, E., Kilonzo, G. P., & Sweat, M. D. (2003). High rates and positive outcomes of HIV serostatus disclosure to sexual p artners: Rea s ons for cautious optimism from a voluntary counseling and testing c linic in Dar es Salaam, Tanzania. AIDS and Behavior, 7 (4), 373 382. Marcus, G. E., & Fischer, M. M. J. (1986). Anthropology as Cultural Critique Chicago: University of Chicago Press. M ar kowitz, L. (2001). Finding the field: Notes on the e thnography of NGOs. Human Organization, 61 (1), 40 46. Mayaud, P., Gill, D. K., Weiss, H. A., Uledi, E., Kopwe, L., Todd, J., . Lacey, C. J. N. (2001). The interrelation of HIV, cervical human papi llomavirus, and neoplasia among antenatal clinic attenders in Tanzania. Sex ually Transm itted Inf ections 77 248 254. Mayaud, P., Weiss, H. A., Lacey, C. J. N., Gill, D. K., & M abey, D. C. W. (2003). Genital human papillomavirus g enotypes in Northwestern Tanzania. Journal of Clinical Microbiology, 41 (9), 4451 4453. Mbago, M. C. Y. (2004). Socio demographic correlates of desire for HIV testing in Tanzania. Sexual Health, 1 13 21. Mbulaiteye, S. M., Bhatia, K., Adebamowo, C., & Sasco, A. J. (20 11). HIV and cancer in Africa: M utual collaboration between HIV and cancer programs may provide timely research and public health data. Infectious Agents and Cancer, 6 16. McCloskey, L. A., William, C ., & Larsen, U. (2005). Gender inequality and intimate partne r violence among w omen in Moshi, Tanzania. International Family Planing Perspectives, 31 (3), 124 130. McFarland, D. M. (2013). Associations of demographic variables and the health belief model constructs with pap smear screening among urban women in Bots wana. International Journal of Women's Health, 5 (Annual 2013), 709. Megevand, E., Van Wyk, W., Knight, B., & Bloch, B. (1996). Can cervical cancer be prevented by a see, screen, and treat program? A pilot study. American Journal of Obstetrics & Gynecolo gy, 174 (3), 923 928.

PAGE 205

189 Mfinanga, S. G., Kahwa, A., Kimaro, G., Kilale, A., Kivuyo, S., Senkoro, M., . Mashoto, K. (2008). Patient's dissatisfaction with the public and private laboratory services in conducting HIV related testing in Tanzania. BMC Hea lth Services Research, 8 167 Mills, A. (1983). Vertical vs. horizontal health programmes in Africa: Idealism, pragmatism, resources and efficiency. Social Science & Medicine, 17 (24), 1971 1981. Ministry of Health and Social Welfare. (2013). Standard Treatment Guidelines (STG) and The National Essential Medicines List (NEMLIT) for Ma inland Tanzania (4th ed.). Ministry of Health and Social Welfare. Ministry of Planning, Economy and Empowerment. (2007). Morogoro Regional Socio Economic Profile 2007. Dar es Salaam: Ministry of Planning, Economy and Empowerment. Molyneux, D. H., Hotez, P. J. & Fenwick, A. (2005). "Rapid impact interventions": How a policy of integrated control for Africa's neglected tropical diseases could benefit the p oor. PLoS Medicine 2 (11), e336. Morhason Bello, I. O., Odedina, F., Rebbeck, T. R., Harford, J., Dangou, J. M., Denny, L., & Adewole, I. F. (2013). Challenges and opportunitie s in cancer control in Africa: A perspective from the African Organisation for Research and Trai ning in Cancer. The Lancet Oncology, 14 (4), e142 e151. Mosha, D., Mahande, M., Ahaz, J., Mosha, M., Njau, B., Kitali, B., & Obure, J. (2009). Factors associated with management of cervical cancer patients at KCMC Hospital, Tanzania: A retrospective cross sectional study. Tanzania Journal of Health Research, 11 (2), 70 74. Moubayed, P., Ziehe, A., Peters, J., Mwakyoma, H., & Schmidt, D. (1995). Carcinoma of the uterine cervix associated with schistosomiasis and induced by human papillomavirus. Internation al Journal of Gynecology & Obstetrics, 49 (2), 175 179. Mrisho, M., Schellenberg, J. A., Mushi, A. K., Obrist, B., Mshinda, H., Tanner, M., & Schellenberg, D. (2007). Factors affecting home delivery in rural Tanzania. Tropical Medicine & International Hea lth, 12 (7), 862 872. Mselle, L. T., Moland, K. M., Mvungi, A., Evjen Olsen, B., & Kohi, T. W. (2013). Why birth care in Tanzania. BMC Health Services Research, 13 174 185.

PAGE 206

190 Muoz, N., Bosch, F. X., Castellsagu, X., Sanjose, S. d., Hammouda, D., Shah, K. V., & Meije r, C. J. L. M. (2004). Against which human papillomavirus types shall w e vaccinate and screen? The international p erspective. International Journal of Cancer, 111 (2), 278 285. Mutyaba, T., Faxelid, E., Mirembe, F., & Weiderpass, E. (2007). Influences on uptake of reproductive health services in Nsangi community of Uganda and their implications for cervical cancer screening. Reproductive Health, 4 (4). Nation al Bureau of Statistics. (2006). Morogoro Regional and District Projections (Vol. XII). Dar es Salaam: Ministry of Planning, Economy and Empowerment. National Bureau of Statistics. (2011). Tanzania Demographic and Health Survey 2010. Dar es Salaam: Nation al Bureau of Statistics. National Bureau of Statistics. (2013). 2012 Population and Housing Census. Dar es Salaam: National Bureau of Statistics. National Institute for Medical Research. (2012). MASCOT: Country Report for Tanzania. Dar es Salaam: Nationa l Institute for Medical Research. Neuman, W. L. (2003). Social Research Methods (5t h ed.). Boston: Allyn and Bacon. Ngwalle, E. W., Mgaya, H. N., Mpanju Shumbusho, W., Chirenje, Z. M., Kirumbi, L., Lebelle, T., & Kaggwa, S. (2001). Situational analysis f or diagnosis and treatment of cervical cancer in m ainland Tanzania. East African Medical Journal, 78 (2), 60 64. Nichter, M. (2008). Global Health Tuscan: University of Arizona Press. Njau, B., Watt, M. H., Ostermann, J., Manongi, R., & Sikkema, K. J. ( 2012). Perceived acceptability of home based couples voluntary HIV counseling and testing in Northern Tanzania. AIDS Care, 24 (4), 413 419. Ocean Road Cancer Institute. (2014). Facts About Ocean Road Cancer Institute. Dar es Salaam: Ocean Road Cancer Inst itute. Retrieved from http://digitallibrary.ihi.or.tz/2747/1/ORCI.pdf Okonofua, F. (2007). HPV Vaccine and Prevention of Cervical Cancer in Africa. African Journal of Reproductive Health, 11 (2), 7 9. Ombelet, W., Cooke, I., Dyer, S., Serour, G., & Devro ey, P. (2008). Infertility and the provision of infertility medical serices in developing countries. Human Reproduction Update, 14 (6), 605 621.

PAGE 207

191 Omran, A. R. (1971). T he Epidemiologic Transition: A theory of the epidemiology of population c hange. The Milb ank Memorial Fund Quarterly, 49 (1), 509 538. Ooms, G., Damme, W. V., & Temmerman M. (2007). Medicines without doctors: Why the Global Fund must fund salaries of health workers to expand AIDS t reatment. PLoS Medicine, 4 (4), 0605 0608. Outwater, A., Nky a, L., Lyamuya, E., Lwihula, G., Green, E. C., Hogle, J., . Dalla betta, G. (2001). Health care seeking behavior for sexually transmitted diseases among commercial sex w orkers in Morogoro, Tanzania. Culture, Health & Sexuality, 3 (1), 19 33. PATH. (200 4). Western Kenya Cervical Cancer Prevention Project (WKCCPP): Final Report. Seattle: PATH. Patton, M. Q. (1990). Qualitative Evaluation and Research Methods (2 nd ed.). Newbury Park: Sage Publications. Perng, P., Perng, W., Ngoma, T., Kahesa, C., Mwaisel age, J., Merajver, S. D., & Soliman, A. S. (2013). Promoters of and barriers to cervical cancer screening in a rural setting in Tanzania. International Journal of Gynaecology and Obstetrics, 123 221 225. Peters Golden, H. (1982). Breast cancer: Varied p erceptions of social support in the illness experience. Social Science & Medicine, 16 (4), 483 491. Peters, L. M., Soliman, A. S., Bukori, P., Mkuchu, J., & Ngom a, T. (2010). Evidence for the need of educational programs for cervical screening in r ural Ta nzania. Journal of Cancer Education, 25 153 159. Petry, K. U., Scholz, U., Hollwitz, B., von Wasielewski, R., & Meijer, C. (2003). Human papillomavirus, coinfection with Schistosoma hematobium, and cervical neoplasia in rural Tanzania. International Jou rnal of Gynecological Cancer, 13 (4), 505 509. Pfeiffer, J., & Ni chter, M. (2008). What can critical medical anthropology contribute to global h ealth? Medical Anthropology Quarterly, 22 (4), 410 415. Plotkin, M., Besana, G. V. R., Yuma, S., Kim, Y. M., K ulindwa, Y., Kabole, F., . Giattas, M. R. (2014). Integrating HIV testing into cervical cancer s creening in Tanzania: A n analysis of routine service delivery statistics. BMC Women's Health, 14 120 ?

PAGE 208

192 Plummer, M. L., Wamoyi, J., Shigongo, Z. S., Msh ana, G., Obasi, A. I. N., Ross, D. A., & Wight, D. (2011). "Seek any means, and keep it your secret": Young women's attempts to control their reproduction through contraceptive and fertility practices in rural Tanzania. Tanzania Journal of Health Research, 12 (3), 1 17. Pool, R., & Washija, N. R. (2001). Traditional healers, STDs and infertility in northwest Tanzania. In J. T. Boerma & Z. Mgalla (Eds.), Women and Infertility in sub Saharan Africa: a multi disciplinary perspective (pp. 241 255). Amsterdam, Netherlands, Royal Tropical Institute: KIT Publishers. Population Reference Bureau. (2012). 2012 World Population Data Sheet. Retrieved from http://www.prb.org/pdf12/2012 population data sheet_eng.pdf Ports, K. A., Reddy, D. M., & Rames hbabu, A. (2013). Barriers and facilitators to HPV v accinati on: Perspectives from Malawian w omen. Women & Health, 53 (6), 630 645. Quentin, W., Terris Prestholt, F., Changalucha, J., Soteli, S., Edmunds, W. J., Hutubessy, R., . Watson Jones, D. (2012). Costs of deliver ing human papillomavirus vaccination to schoolgirls in Mwanza Region, Tanzania. BMC Medicine, 10 137 147. Rao, K. D., Ramani, S., Hazarika, I., & George, S. (2014). When do vertical programmes strengthen health systems? A comparative assessment of diseas e specific interventions in India. Health Policy and Planning, 29 (4), 495 505. Renne, E. (2006). Perspectives on polio and immunization in Northern Nigeria. Social Science & Medicine, 63 (2006), 1857 1869. Roberts, C. S., Cox, C. E., Shannon, V J., & W ells, N. L. (1994). A closer look at social support as a moderator of stress in breast c ancer. Health & Social Work, 19 (3), 157 164. Rosser, J.I., Zakaras, J.M., Hamisi, S., and Huchko, M.J. (2014). Men's knowledge and attitudes about cervical cancer scr eening in Kenya. BMC Women's Health, 14, 138. Royse, D., & D ignan, M. (2011). Fatalism and cancer s creening in Appalachian Kentucky. Family & Community Health, 34 (2), 126 133. Samb, B., Evans, T., Dybul, M., Atun, R., Moatti, J. P., Nishtar, S., . a l., e. (2009). An assessment of interactions between global health initiatives and country health systems. The Lancet, 373 (9681), 2137 2169. Sankaranarayanan, R., Budukh, A. M., & Rajkumar, R. (2001). Effective screening programmes for cervical cancer in low and middle income developing countries. Bulletin of the World Health Organization, 79 954 962.

PAGE 209

193 Sankaranarayanan, R., Gaffikin, L., Jacob, M., Sellors, J., & Robles, S. (2005). A critical assessment of screening methods for cervical neoplasia. Inter national Journal for Gynecology and Obstetrics, 89 54 512. Sankaranarayanan, R., Rajkumar, R., Theresa, R., Esmy, P. O., Mahe, C., Bagyalakshmi, K. R., . Cherian, J. (2004). Initial results from a randomized trial of cervical visual screening in rur al south India. International Journal of Cancer 109 (3), 461 467. their households in rural South Africa. Scandanavian Journal of Public Health Supplement, 69 147 154. Schensul, J. J., & LeCompte, M. D. (2012). Essential Ethnographic Methods: A Mixed Methods Approach (2nd ed.): AltaMira Press. Scheper Hughes, N., & Bourgois, P. (2004). Introduction: Making sense of violence. In N. Scheper Hughes & P. Bourgois (Eds.), V iolence in War and Peace (1st ed., pp. 512). Oxford: Blackwell Publishing. Scheper Hughes, N., & Lock, M. M. (1987). The mindful body: A prolegomenon to future work in medical a nthropology. Medical Anthropology Quarterly, 1 (1), 6 41. Schraeder, P. J. (2 000). African Politics and Society Boston: Bedford/St. Martin's. Seffrin, J. R., Hill, D., Burkart, W., Magrath, I., Badwe, R. A., Ngoma, T., . Grey, N. (2009). It is t ime to i nclu de cancer and other noncommunicable d iseases in the Millennium Develop ment Goals. CA: A Cancer Journal for Clinicians 59 (5), 1 3. Seow, A., Huang, J., & Straughan, P. T. (2003). Effects of social support, regular physician and health related attitudes on cervical cancer screening in an Asian population. Cancer Causes & Control, 11 (3), 223 230. Setel, P., W. (1999). A Plague of Paradoxes: AIDS, Culture, and Demography in Northern Tanzania Chicago: The University of Chicago Press. Shastri, S. S., Dinshaw, K., Amin, G., Goswami, S., Patil, S., Chinoy, R., . Sankaranarayanan, R. (2005). Concurrent evaluation of visual, cytological and HPV testing as screening methods for the early detection of cervical neoplasia in Mumbai, India. Bulletin of the World Health Organization, 83 (3) 186 194. Singer, M. (1989). The coming of age of critical medical a nthropology. Social Science & Medicine, 28 (11), 1193 1203.

PAGE 210

194 Singer, M. (Ed.). (1998). The Political Economy of AIDS Amityville, New York: Baywood Publishing Company, Inc. Sommer, M. Tanzania. Journal of Adolescence, 33 521 529. Sontag, S. (1977). Illness as Metaphor New York: Vintage Bo oks. St. Francis Hospital. (2014). St. Francis Referral Hospital Ifakara Retrieved from http://www.ifakara.org/en/st francis hospital/hospital.php?catid=55 Straughan, P. T ., & Seow, A. (1998). Fatalism reconceptualized: A concept to predict h ealt h scr eening b ehavior. Journal of Gender, Culture and Health, 3 (2), 85 100. Strauss, A., & Corbin, J. M. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques (2nd ed.). Thousand Oaks, CA: SAGE Publications, Inc. Strecher, V. J., & Rosenstock, I. M. (1997). The Health Belief Model. In K. Glanz, F. M. Lewis & B. K. Rimer (Eds.), Health Behavior and Health Education (2nd ed., pp. 41 59). San Francisco: Jossey Bass. Sussman, S., Pokhrel, P., Black, D., Kohrman, M., Hamann, S., Vateesa tokit, P., & Nsimba, S. E. D. (2007). Tobacco control in developing countries: Tanzania, Nepal, China, and Thailand as examples. Nicotine & Tobacco Research, 9 (1), S447 S457. Ter Meulen, J., Eberhardt, H. C., Luande, J., Mgaya, H. N., Changclaude, J., Mt iro, H., . Pawlita, M. (1992). Human pap illomavirus (HPV) infection, HIV infection and cervical cancer in Tanzania, East Africa. International Journal of Cancer, 51 (4), 515 521. Trostle, J. A., & Sommerfeld, J. (1996). Medical a nthrop ology and e pidem iology. Annual Review of Anthropology, 25 253 274. Trottier, H., & Franco, E. L. (2006). The epidemiology of genital human papillomavirus infection. Vaccine, 24 (Supplement 1), S1/4 S1/15. Tsu, V. D., Jeronimo, J., & Anderson, B. O. (2013). Why the tim e is right to tackle breast and cervical cancer in low resource settings. Bulletin of the World Health Organization, 91 683 690. Tsu, V. D., & Levin, C. E. (2008). Making the case f or cervical cancer prevention: W hat about equity? Reproductive Health Ma tters, 16 (32), 104 112.

PAGE 211

195 Turshen, M. (1998). The Political Ecology of AIDS in Africa. In M. Singer (Ed.), The Political Economy of AIDS (pp. 169 184). Amityville, New York: Baywood Publishing Company, Inc. Upton. (2001). 'Infertility m ake s you invisible' : Gender, health and the negotiation of fertility in n orthern Botswana. Journal of Southern African Studies 27 (2, Special Issue on Fertility in Southern Africa), pp. 349 362. Urasa, M., & Darj, E. (2011). Knowledge of cervical cancer and screening practic es of nurses at a regional hospital in Tanzania. African Health Sciences, 11 (1), 48 57. Urassa, W., Moshiro, C., Chalamilla, G., Mhalu, F., & Sandstrom, E. (2008). Risky sexual practices among youth attending a sexually transmitted infection clinic in Da r es Salaam, Tanzania. BMC Infectious Diseases, 8 159 U.S. Preventive Services Task Force. (2012). Recommendation Summary. Retrieved from http://www.uspreventiveservicestaskforce.org/Page/Topic/recommendation summary/cervical cancer screening van Schalkwyk, S. L., Maree, J. E., & Wright, S. C. (2008). Cervical cancer: T he route from signs and symptoms to treatment in South Africa. R eproductive Health Matters, 16 (32), 9 17. Vilos, G. A. (1998). After hours: The history of the Papanicolaou smear and the o dyssey of George and Andromache Papanicolaou. Obstetrics & Gynecology, 91 (3), 479 483. Walboomers, J. M. M., Jacobs, M. V., Manos M. M., Bosch, F. X., Kummer, J. A., Shah, K. V., . Muoz, N. (1999). Human papillomavirus is a necessary cause of invasive cervical c a ncer w orldwide. Journal of Pathology, 189 12 19. Wamoyi, J., Wight, D., Plummer, M., Mshana, G. H., & Ross, D. (2 010). Transactional sex amongst young peop le in rural northern Tanzania: A n ethnography of young women's motivations and negotiation. Reproductive Health, 7 (2). White, H. L., Mulambia, C., Sinkala, M., Mwanahamuntu, M., Parham, G. P., Moneyham, L., . Chamot, E. (2012). 'Worse than HIV' or 'not as serious as other diseases'? Conceptualization of cervical cancer among newly screened women in Zambia. Social Science & Medicine, 74 (10), 1486 1493. WHO/ICO Information Centre on HPV and Cervical Cancer (H PV Information Centre). (2010). Human P apillomavirus and Related Cancers in Tanzani a: Summary Report 2010. World Health Organization.

PAGE 212

196 Wingood, G. M., & DiClemente, R. J. (2000). Application of the theory of gender and power to examine HIV related exposure s, risk factors, and effective interventions for w omen. Health Education & Behavior, 27 (5), 539 565. Wingood, G. M., & DiClemente, R. J. (2002). The Theory of Gender and Power. In R. J. DiClemente, R. A. Crosby & M. C. Kegler (Eds.), Emerging Theories in Health Promotion Practice and Research (pp. 313 346). San Francisco: Jossey Bass. World Bank (2010 2014). Data: GDP per capita (current US$). Retrieved March 15, 2015, from http://data.wor ldbank.org/indicator/NY.GDP.PCAP.CD Wringe, A., Isingo, R., Urassa, M., Griter, M., & Mayalla, R. (2008). Uptake of HIV voluntary counselling and testi ng services in rural Tanzania: I mplications for effective HIV prevention and equitable access to treatm ent. International Health, 13.3 (Mar 2008), 319 327. Yang, B. H., Bray, F. I., Parkin, D. M., Sellors, J. W., & Zhang, Z. F. (2004). Cervical cancer as a priority for prevention in different world regions: An evaluation using years of life lost. Internati onal Journal of Cancer, 109 418 424.

PAGE 213

197 APPENDIX A LIST OF PARTICIPANTS Cervical Cancer Prevention Program Sandra A ma rried 27 year old woman, she had given birth once and did not know her HIV status. She lived in Kilombero and did not work. Shazia Sh e was a married farmer from Kilombero with two children. Currently 46, she had initiated sexual activity at 18 but did not know her HIV status. Debora Thirty one years old, she was married, did not know her HIV status, and had given birth twice. She was a farmer from Kilombero. Hellena A ward attendant from Kilombero, she was HIV negative. She had initiated sexual activity at 15 and was divorced but living with a partner. She was 55 and had five children. Karima She was 38 years old and had had three s exual partners since she initiated sexual activity at 16 Currently married, she had given birth four times, was HIV positive, and worked as a farmer. Flora A 54 year old mother of nine, she had only ever been with her husband since initiating sexual acti vity at 14 and did not know her HIV status. Sh e farmed and lived in Kilombero. Fausta Forty one years old, she was married with six children and farmed in Kilombero. She had had five sexual partners since the age of 18 and was HIV negative. Naomi She was a divorced 36 year old farmer in Kilombero who had given birth three times. She had initiated sexua l activity at 18, had three partners since that time and was HIV positive. Salmah Living in Kilombero, she was 44 years old HIV positive, and divorced, w ith five children. Since initiating sexual activity at 18 she had had three partners. Mwajuma Though unmarried at 54 years old, she had four children whom she sup ported as a farmer in Kilombero. She had had four partners during her lifetime, and was HIV p ositive. Anna Fifty two years old, she had been married but was now a widow with five children. She was HIV positive and had had three partners since initiating sexua l activity at 17. Salama She had begun sexual activity at 16 and had four partners by th e age of 28. Now married, she had given birth three times and farmed in Kilombero. Her HIV status was unknown. Salome A married 52 year old farmer from Kilombero, she had given birth five times and was HIV negative. She had had three sexual partners in he r lifetime, and began sexual activity at the age of 23. Edna She was a divorced mother of three at the age of 51, and had only had two partners since she initiated sexual activity at 18. She was HIV positive and worked with an NGO near her community that focused on HIV issues.

PAGE 214

198 Zuhura At 65 years old, she was the oldest participant. She farmed in Kilombero and was married to her only lifetime sexual partner. She had begun sexual activity at the age of 14, had given birth ten times, and was HIV negative. V ioleth Forty years old, she was unmarried but lived with a man in Kilombero where she was a farmer. She initiated sexual activity at 16 and had had four partners since that time; she had also given birth five times and was HIV negative. Damisi She did not work outside the home in which she lived in Kilombero, and was currently living with a man, though they were not married. Thirty years old, she had begun sexual activity at 18 and had two partners since She was HIV negative and had two children. Khadija Another senior member of the group at 61, she was unmarried and was HIV positive. She supported herself by farming, and had only one child. Since initiating sexual activity at 20, she had had two partners. Marjani At 33 years old, she had had 3 sexual pa rtners since she began sexual activity at 19. She now co habitated with a man, had three children and was HIV negative. She was a farmer in Kilombero. Janeth From Ulanga, she was a 51 year old farmer and married mother of eight. She had initiated sexual activity at 15 and had had five partners since. She was HIV negative. Mariamu She was an unmarried 35 year old woman and had given birth four times. HIV negative, she had had ten sexual partners since the age of 20 when she first began sexual activity. Sh e supported herself through farming and lived in Kilombero. Hadija A farmer from Kilombero, she was a 39 year old married mother of five. She had had four sexual partners since the age of 20 when she initiated sexual activity and was HIV negative. Subir a Twenty five years old, she began sexual activity at 18 and had had two partners in the years since. She was now married with two children; she was HIV negative. She was a farmer in Kilombero. Rose From Kilombero, she was employed as a business woman. At 33 years old, she was married with three children and was HIV positive. She had had four sexual partners since initiating sexual activity at 20. All places of residence given refer to the district in which women lived. Gynecology Department Ummi Forty two years old, she was a married mother of two who worked as a nursery school teacher. She did not know her HIV status, and lived in Kilombero. Susannah She was currently married with one child at 28 years old and lived and worked as a teacher in Morogor o region. She had had four partners in the six years since she initiated sexual activity ; her HIV status was unknown

PAGE 215

199 Rebekah A teacher in Morogoro, she was married and had given birth once. She had had five partners since she began sexual activity at 20. She was 33 now and did not know her HIV status. Neema She was 28 years old and had only had one sexual partner: her husband. They had not had any children yet. She was HIV negative and worked as a tailor in Kilombero. Mercy Living in the immediate area of St. Francis she worked as a secondary school teacher. HIV negative, her only lifetime sexual partner was her husband, and at 27, she had given birth once. Leta A 27 year old tailor living in Iringa, she was married but did not have children. She initi ated sexual activity at 18, after which she had two partners, and she did not know her HIV status. Gladys Twenty six years old, she supported herself as a tailor in Kilombero. She was married but had not given birth; since the age of 15 she had had two se xual partners and was HIV negative. Sakina She was a widowed mother of five who farmed in Kilombero. At 45 years old, she had only been with her husband since initiating sexual activity at 15; she did not know her HIV status. Zuwena A farmer from Kilombe ro, she was married but had had three other sexual partners since she initiated sexual activity at 17. She was currently 29 years old and did not know her HIV status. Lilian From Kilombero, she was a married farmer and mother of five. At 33, she had only been with her husband since beginning sexual activity at 17 and was HIV negative. Sarah A married 36 year old, she had one child and was HIV negative. She had had four partners since age 16, and was a farmer in Kilombero. Aisha Thirty seven years old, sh e had been with three partners since she initiated sexual activity at 20. She was now married with five children and was HIV negative. She farmed in Kilombero. Rahma An unmarried nurse from Iringa, she had no children at the age of 39 and was HIV negative She had begun sexual activity fairly late at 25 and had had three partners since that time. Sophy She was a 38 year old primary school teacher in Kilombero, was married, and had two children. Her husband, who m she married when she was twenty, was the on ly se xual partner she had ever had. S he was HIV negative. Furaha A business woman in Kilombero, she was unmarried but had one child. Now 38, she initiated sexual activity at 17 and had had 6 partners since then. Her HIV status was negative. Shany She wa s 30 years old and married, but had no children. She was HIV negative, and had been with two sexual partners since the age of 18 S he worked as a business woman in Kilombero. Flaviana Thirty two years old, she was currently engaged to be married. She had no children, was HIV positive, and had had four sexual partners since initiating sexual activity at 20. She lived in Kilombero and was unemployed.

PAGE 216

200 Hasina Forty four years old, she was married and farmed in Kilombero. She had begun sexual activity at 23 bu t her husband was the onl y partner she had ever had. They had four children and she was HIV negative. Grace She had a home business cookin g food and lived in Kilombero. At 36 years old, she lived with a man and had given birth three times. She was HIV neg ative, and had had three sexual partners since initiating sexual activity at 14. Jesca A 30 year old married tailor from Kilombero, she had given birth three times. She had two lifetime sexual partners after beginning sexual activity at 19 and was HIV neg ative. Irene An HIV positive woma n from Kilomber o she worked as a farmer and was widowed at 42. She had four children, and had had three sexual partners since initiation of sexual activity at 20. Roseline Married with seven children, she was 48 years ol d and farmed in Kilombero. She was HIV negative and had two lifetime sexual partners since initiating sexual activity at 20. Hakika A nurse from Ulanga, she was HIV negative an d had only one sexual partner since initiating sexual activity at 25 her husba nd. No w 38, she was married with two children. Eliza Employed as a tailor in Kilombero, she was 32 years old and unmarried, though she had one child. She did not know her HIV status and had had three sexual partners since initiating sexual activity at 25 Dina h Twenty five years old, she had had seven sexual partners since the age of 17 when she began sexual activity. She was now married but did not have any children yet; she was also HIV negative. She farmed in Kilombero.

PAGE 217

201 APPENDIX B PARTICIPANT INTE RVIEW GUIDES Note : The participant interviews conducted in the course of this study were semi structured qualitative interviews and therefore this question guide served as a i nterviews depending on participant answers. PARTICIPANT QUESTION GUIDE [ Asked of All Participants ] Thank you for agreeing to speak with us today about health, cervical cancer, and the cervical cancer screening program. First, I would like to ask you a fe w questions about health and illness here in Tanzania. Are you ready to begin? GENERAL QUESTIONS ABOUT HEALTH 1. a. [ Prompt s: For example, do you think that a person has control over their own health? Are th What are they?] 2. If a person wants to improve their health, what should they do? 3. If a person is healthy right now, how can they stay healthy? 4. If a person does feel ill, what should they do? a. Are there thing s they can do at home? b. Where should they go for treatment? i. [ Prompts : A traditional healer? A hospital? A clinic or dispensary?] 5. If you feel ill, how do you decide where to go or who to see for treatment? a. Are there certain symptoms or illnesses that trad itional medicine is better at healing? b. Are there certain symptoms or illnesses that a doctor or clinic is better at healing? PERCEPTIONS/KNOWLEDGE OF [CERVICAL] CANCER Thank you for your responses. The next set of questions is will ask about your knowled ge of cancer. 1. What can you tell me about cancer? a. Have you ever heard about a specific type of cancer called cervical cancer? [ If yes ] What can you tell me about cervical cancer? [Note to Interviews: If the woman is familiar with cervical cancer, ask the fo llowing questions specifically about cervical cancer. If she is not, ask her about cancer in general.]

PAGE 218

202 2. What causes [cervical] cancer? 3. Who is likely to get [cervical] cancer? a. [ Prompts : Are there certain people/women who are more likely to get cancer than others? Who?] 4. How does a person suspect they might have [cervical] cancer? a. [Prompts : Are there certain symptom that will appear? What symptoms would make a person think they might have cancer?] 5. If a person believes they have [cervical] cancer, what shoul d they do? a. Who should they see? b. How should they treat it? c. Where should they go or who should they see for treatment? d. Is there anything they can or should do at home? 6. Do you think that [cervical] cancer can be prevented? [ If yes ] What can be done? 7. What do you think is the prognosis for someone who is diagnosed with [cervical] cancer? Now I would like to ask you about [cervical] cancer in your community. CERVICAL CANCER IN THE COMMUNITY 1. How do you think people in your community view [cervical] cancer? a. [ Pr ompts : Do you think they are aware of the disease? How do people in your community view a woman who has been diagnosed with [cervical] cancer?] 2. How do you think people in your community view cervical cancer screening? a. [ Prompts : Have you heard anyone talk about screening? What do they say?] 3. Should a woman be screened if she currently has no symptoms? Or only if she has symptoms? 4. If a woman wants to be screened for cervical cancer, are there any difficulties she might face? a. [ Prompts : For example, will she have difficulties with t ransportation, cost, childcare, h er h usband or family etc ? Is it difficult to get screened for cervical cancer?] 5. If a woman is diagnosed with cervical cancer, do you think she will share this information with other people? a. [ Pr ompt s : Why or why not? Who will she share it with? Who is she likely to go to for support? ] 6. I have asked you about difficulties a woman might face in getting screened for cervical cancer. If she is diagnosed with the disease, what difficulties might she face in seeking treatment for cervical cancer? a. [ Prompts : For example, will she have difficulties with t ransportation, cost, childcare, h er h usband or family etc. ? Is it difficult to get treatment for cervical cancer?

PAGE 219

203 [ The Following Two Question Sets Va ry Depending on Group ] CERVICAL CANCER SCREENING [CECAP Clients] Thank you. The next set of questions is about cervical cancer screening, the procedure you are here for today. 1. Is today the first time you have ever been screened for cervical cancer or hav e you been screened before? [ ] a. Where was the screening done? b. Were you satisfied with the experience? Why or why not? c. Did you have to wait for the results or did they tell you immediately that day? d. What w as the outcome of the test? 2. How did you find out about the cervical cancer screening program here at St. Francis? a. [ Prompts : Radio, Sign, Brochure, Poster, Told by a nother person, Referred ] DECISION MAKING [CECAP Clients] We are almost done! This is the last set of questions. 1. Why did you decide to come in to the clinic for cervical cancer screening? a. [ Prompts : What influenced your decision to come in? What factors did you consider? Were you having any symptoms?] 2. Did you discuss your decision with anyone else in your life? a. [ If No ] Why not? b. [ If Yes ] With whom did you discuss it? c. Were they supportive? d. Did they have concerns? 3. Who made the decision that you should come in for screening? 4. What fears or concerns, if any, did you have about screening when you c ame in today? 5. What benefits did you hope to gain by getting screened? 6. Did you encounter any difficulties in coming in for screening? a. [ Prompts : For example, with transportation, work, childcare, money, opposition from family or friends, etc.] 7. Have you told any other women you know about the screening program? a. Do you plan to tell other women you know about the screening program after you leave today? Thank you so much for your time. Do you have any questions for me before you leave? [ OR ]

PAGE 220

204 CERVICAL C ANCER SCREENING [GD Clinic] Now I would like to ask you a this test, a health care provider checks a patient for cervical cancer by looking her cervix. 1. To your knowledge, have you ever been screene d for cervical cancer in the past? [ If the answer is NO, move on to question 2 ] a. Where was the screening done? b. Were you satisfied with the experience? Why or why not? c. Did you have to wait for the results or did they tell you immediately that day? d. What was the outcome of the test? 2. Were you aware that there is a cervical cancer screening program here at St. Francis? a. [ If Yes ] How did you find out? i. [ Prompts : Radio, Sign, Brochure, Poster, Told by a person, Referred here] DECISION MAKING [GD Clinic] We are almost done! This is the last set of questions. 1. Have you considered or would you consider being screened for cervical cancer? a. [ Prompts : Why or why not?] 2. What fears or concerns, if any, do you have about screening? 3. Are there any benefits in your opinion to being screened? 4. When making this decision, would you discuss this it with anyone else in your life? a. [ If No ] Why not? b. [ If Yes ] With whom would you discuss it? c. Do you think they would be supportive? d. Do you think they would have concerns? 5. Would there be an y factors that would make it difficult for you to come in for screening? a. [ Prompts : For example, with transportation, work, childcare, money, opposition from family or friends, etc.] Thank you so much for your time. Do you have any questions for me before you leave?

PAGE 221

205 APPENDIX C PARTICIPANT INFORMATION FORM Participant Study ID No. _________________________________________ Name of Data Collector: __________________________________________ Participant Information Question Participant Response 1. Age 2. Ag e at Initiation of Sexual Activity 3. Number of Sexual Partners 4. Marital Status 5. Parity 6. District of Residence 7. Occupation Visit Information 1. Which program/department do you plan/have you planned to visit today? ___ Cervical Cancer Screeni ng Program ___ Other Obstetrics/Gynaecological Services ___ Other Department (list):_________________________________ 2. Is this your first visit to St. Francis Hospital? ___ Yes ___ No 3. Were you referred to St. Francis Hospital by another medical provider, clinic, or dispensary? ___ Yes (list): ____________________________________________ ___ No

PAGE 222

206 APPENDIX D SCREENING MAP FORM