Citation
Conceptual models, meet contextual variables! Reintroducing context to the study of fatalism and the cardiovascular disease screening behavior of Latinos

Material Information

Title:
Conceptual models, meet contextual variables! Reintroducing context to the study of fatalism and the cardiovascular disease screening behavior of Latinos
Creator:
Kapral, Craig G.
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Education and Human Development, CU Denver
Degree Disciplines:
Health and behavioral sciences
Committee Chair:
Spencer, Karen Lutfey
Committee Members:
Scandlyn, Jean
Rooks, Ronica
Gujral, Indira

Notes

Abstract:
Cardiovascular disease (CVD) is a leading cause of adult morbidity and mortality for Latinos living in the United States. Latinos demonstrate disproportionately high prevalence rates for many established CVD risk factors, including high cholesterol and uncontrolled high blood pressure. Screening for high blood pressure and high cholesterol is critical for both the early detection and prevention of CVD. Despite widespread availability, Latinos are less likely than non-Hispanic Whites to utilize blood pressure and cholesterol screening services. In the health literature, the underuse of various screening services by Latinos is often attributed to fatalistic attitudes and beliefs. Fatalism refers to a general belief that the course of fate cannot be changed and that life events are beyond one’s control, including the ability to prevent certain health conditions. Those who endorse fatalistic attitudes and choose to forgo screening and treatment are frequently labeled as “irrational” and in need of education. Unfortunately, extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care (IBHC) Latinos confront in the United States. These barriers include limited access to health insurance, lack of a regular source of health care, and experiences with perceived discrimination in the health care system. Using a sequential mixed methods design, I examine the relationship between specific IBHC (lack of health insurance, no usual source of health care and perceived discrimination in the health care system), fatalism towards CVD and use of blood pressure and cholesterol screening services by Latinos in Denver, Colorado. The results of this study indicate that fatalism occurs as a byproduct of complex social processes and offers little in the way of explaining the suboptimal use of CVD screening services by Latinos. Rather, IBHC in the form of lack of health insurance and no usual source of health care negatively impact the use of blood pressure and cholesterol screening services by this population. Based on these results, efforts to increase use of CVD screening services by Latinos must include the provision of unimpeded access to quality health insurance and the establishment of a usual source of health care.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Craig G. Kapral. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
"CONCEPTUAL MODELS, MEET CONTEXTUAL VARIABLES!"
REINTRODUCING CONTEXT TO THE STUDY OF FATALISM AND THE CARDIOVASCULAR DISEASE SCREENING BEHAVIOR OF LATINOS
by
CRAIG G. KAPRAL
B.A., State University of New York at Potsdam, 1992 M.A., University of Colorado Denver, 2005
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 Program
2019


©2019
CRAIG G. KAPRAL
ALL RIGHTS RESERVED


This thesis for the Doctor of Philosophy degree by
Craig G. Kapral has been approved for the Health and Behavioral Sciences Program
by
Karen Lutfey Spencer (Chair) Jean Scandlyn Ronica Rooks Indira Gujral
Date: August 3, 2019


Kapral, Craig G. (PhD, Health and Behavioral Sciences)
"Conceptual Models, Meet Contextual Variables!” Reintroducing Context to the Study of Fatalism and the Cardiovascular Disease Screening Behavior of Latinos
Thesis directed by Associate Professor Karen Lutfey Spencer
ABSTRACT
Cardiovascular disease (CVD) is a leading cause of adult morbidity and mortality for Latinos living in the United States. Latinos demonstrate disproportionately high prevalence rates for many established CVD risk factors, including high cholesterol and uncontrolled high blood pressure. Screening for high blood pressure and high cholesterol is critical for both the early detection and prevention of CVD. Despite widespread availability, Latinos are less likely than non-Hispanic Whites to utilize blood pressure and cholesterol screening services. In the health literature, the underuse of various screening services by Latinos is often attributed to fatalistic attitudes and beliefs. Fatalism refers to a general belief that the course of fate cannot be changed and that life events are beyond one’s control, including the ability to prevent certain health conditions. Those who endorse fatalistic attitudes and choose to forgo screening and treatment are frequently labeled as "irrational” and in need of education. Unfortunately, extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care (IBHC) Latinos confront in the United States. These barriers include limited access to health insurance, lack of a regular source of health care, and experiences with perceived discrimination in the health care system. Using a sequential mixed methods design, I examine the relationship between specific IBHC (lack of health insurance, no usual source of health care and perceived discrimination in the health care system), fatalism towards CVD and use of blood pressure and cholesterol screening services by Latinos in Denver, Colorado. The results of
IV


this study indicate that fatalism occurs as a byproduct of complex social processes and offers little in the way of explaining the suboptimal use of CVD screening services by Latinos. Rather, IBHC in the form of lack of health insurance and no usual source of health care negatively impact the use of blood pressure and cholesterol screening services by this population. Based on these results, efforts to increase use of CVD screening services by Latinos must include the provision of unimpeded access to quality health insurance and the establishment of a usual source of health care.
The form and content of this abstract are approved. I recommend its publication.
Approved: Karen Lutfey Spencer
v


ACKNOWLEDGEMENTS
This dissertation (COMIRB protocol # 15-0451) was made possible with the support of several key individuals. First and foremost, my deep gratitude goes to Dr. Karen Lutfey Spencer. Your guidance and support were critical to the completion of this research and to my overall academic progress. Thank you for your time, encouragement and detailed feedback. I treasure your mentorship. I would also like to thank my committee members, Drs. Jean Scandlyn, Ronica Rooks and Indira Gujral. Thank you for your commitment and leadership over the past several years. I greatly appreciate your time and assistance with this work.
I would also like to thank my research team, Susana Arreola Ponce de Leon, Blanca Flores and Issamar Pichardo. Susana, I can’t thank you enough for your time and energy. You were an invaluable part of this research and I am forever grateful for your participation. Issamar, I thank you for your dedication and high quality work. Completion of this research would not have been possible without your efforts.
In addition, I would like to thank my wife, Kristine Kapral for her love, support and patience during the course of this research. Thank you for your encouragement and devotion. Your love and friendship mean everything to me. I am forever indebted for the sacrifices you made on my behalf.
Lastly, I would like to thank my late mother, Donna Kapral and father Richard Kapral for their unconditional love and support. I am grateful to have such wonderful parents. Mom, I miss you every day. Thank you for always being there with unending emotional support and love. Your memory will forever live in me and continues to inspire. Dad, I simply can’t thank you enough for your love, guidance and feedback. Thank you for
vi


listening during the countless talks we shared. Thank you for always believing in me and letting me know. You are a true inspiration.
Vll


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION...............................................................1
Research Question..........................................................4
Specific Aims and Hypotheses...............................................5
BACKGROUND.................................................................6
Latinos in Colorado........................................................7
CVD Screening Services.....................................................8
CVD Screening Services: Institutional Barriers............................10
Fatalism..................................................................20
Structure of the Dissertation.............................................30
II. THEORY....................................................................31
A Faulty View: Fatalism as a De-Contextualized Cognitive Trait of the
Individual................................................................32
An Alternative View: Fatalism and its Social Origins......................39
Reintroducing Context: Institutional Barriers and the Expression of Fatalistic Attitudes.................................................................41
III. RESEARCH DESIGN AND METHODS...............................................44
Goal and Objectives.......................................................45
Rationale and Purpose.....................................................46
Sampling Design...........................................................47
Setting...................................................................49
viii


Research Design...........................................................52
Data Collection...........................................................54
Data Collection Procedures................................................57
Data Analysis.............................................................58
Data Validation...........................................................60
Data Integration and Interpretation.......................................63
IV. QUANTITATIVE ANALYSES AND RESULTS..........................................65
Review of Methods.........................................................65
V. QUALITATIVE ANALYSES AND RESULTS..........................................79
Review of Methods.........................................................79
The Contextual Nature of CVD Fatalism.....................................82
The Abatement of Fatalism: The Role of Safety Net and Community-Based
Resources.................................................................94
Dynamic Intersections, Social Context and the Variable Expression of Fatalistic Attitudes................................................................100
VI. MIXED METHODS INTERPRETATION AND CONCLUSION...............................103
Summary of Phase One Results: Quantitative Survey........................105
Summary of Phase Two Findings: Qualitative Interviews....................109
Mixed Methods Interpretations: Summary Rationale and Findings............110
Limitations..............................................................121
Conclusion...............................................................125
REFERENCES......................................................................131
IX


APPENDIX
A. Survey Questionnaire......................................................144
B. Semi-Structured Interview Guide...........................................153
C. Consent Form..............................................................156
x


LIST OF TABLES
TABLE
1. Select Characteristics of Denver County Latino Population........................49
2. Phase Two: Sampling Domains for Qualitative Interviews...........................50
3. Participant Characteristics......................................................65
4. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening when
Participants Have Health Insurance and Have a Usual Source of Health Care (USOC)........................................................................68
5. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screening
when Participants Have Health Insurance and Have a Usual Source of Health Care (USOC)........................................................................69
6. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screening
based upon Level of Fatalism..................................................71
7. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening
based upon Level of Fatalism..................................................73
8. Binary Logistic Regression Results for the Odds of Demonstrating High Level of Fatalism Based Upon Having Health Insurance and Having a Usual Source of Health
Care
75


LIST OF FIGURES
FIGURE
1. Theoretical Model.............................................................42
2. Sequential Sampling Design Using Nested Samples...............................47
3. Partially Mixed Sequential Equal Status Design................................53
4. Phase Two Qualitative Interviews Sample Selection.............................80
xii


LIST OF ABBREVIATIONS
CVD ATP III
LDL HDL JNC 7
ACA
USOC
CAHPS
SES
HBM
TPB
IBM
TRA
COMIRB
CAB
BRFSS
RAs
IBHC
U.S.
Cardiovascular disease
Third Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults Low density lipoprotein High density lipoprotein
Joint National Committee on Prevention, Detection, Evaluation, and
Treatment of High Blood Pressure
The Patient Protection and Affordable Care Act
Usual source of health care
National Consumer Assessment of Health Plans Study
Socioeconomic status
Health belief model
Theory of planned behavior
Integrated behavior model
Theory of reasoned action
The Colorado Multiple Institutional Review Board Community advisory board
Behavioral Risk Factor Surveillance System Survey Research assistants Institutional barriers to health care United States of America
xm


CHAPTER I
INTRODUCTION
Cardiovascular disease (CVD) is a leading cause of adult morbidity and mortality for Latinos living in the United States (AHA, 2010). In 2016, CVD accounted for 27 percent and 28.2 percent of all-cause mortality among Latino males and females respectively (AHA, 2016). Latinos are known to demonstrate disproportionately high prevalence rates for many established CVD risk factors, including high cholesterol and uncontrolled high blood pressure (Mozaffarian etal., 2016; CDC, 2011a; Jurkowski, 2006;). For example, recent results from a national, multi-site study of 16, 415 Latino adults found that 45 percent have high cholesterol and 25 percent have hypertension (Rodriguez et al., 2015). Importantly, the findings from this study demonstrate that of those with high cholesterol, almost half (49.3%) were unaware of their condition. In a separate study drawing upon the same data used by Rogriguez et al. (2015) above, Sorlie and colleagues found that among hypertensives, more than 25 percent were unaware of their condition and 62 percent did not have their hypertension controlled (Sorlie et al., 2014).
Screening for CVD risk factors is important for both the early detection and prevention of CVD (AHA, 2019; Jurkowski & Johnson, 2005). Given the large proportion of Latinos with high cholesterol and those who remain unaware of their hypertensive status, screening is imperative to reduce the overall burden of CVD in this population. However, despite widespread availability of screening tests, Latinos are less likely than non-Hispanic Whites to utilize cholesterol and blood screening services (AHA, 2016; AHA, 2007; Balluz, Okoro, & Strine, 2004; Behbakht, Lynch, Teal, Degeest, & Massad, 2004; Facione, Miaskowski, Dodd, & Paul, 2002; Garcia etal., 2017; Luquis & Villanueva Cruz, 2006). In the
1


health literature, the underuse of various screening services by Latinos is often attributed to fatalistic attitudes and beliefs. (AHA, 2007; Balluz et al., 2004; Behbakht et al., 2004;
Cohn & Esparza Del Villar, 2015; Facione etal., 2002; Luquis & Villanueva Cruz, 2006).
Fatalism refers to a general belief that the course of fate cannot be changed and that life events are beyond one’s control (Abraido-Lanza et al., 2007; Leyva et al., 2014). More specifically, fatalists believe that one has little control over life’s vicissitudes. Those who hold fatalistic attitudes often subscribe to the idea of lack of personal control, and thus believe they have no control over life events, including the ability to prevent certain health conditions (Fairchild, 2015; Ramirez etal., 2002). Fatalism is typically operationalized as negative attitudes toward preventive health practices and disease outcomes and has therefore been identified as a potential barrier to disease detection and prevention efforts (Espinosa de los Monteros & Gallo, 2011).
Unfortunately, extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care Latinos confront in the United States. These barriers include limited access to health insurance (Balluz et al., 2004; OMH, 2019), lack of a regular source of health care (Balluz et al., 2004; Hayes et al., 2015), and experiences with perceived discrimination in the health care system (Ariza, Vimalananda, & Rosenzweig, 2010; Benjamins & Middleton, 2019).
The overemphasis on individual-level fatalistic attitudes carries important implications. First, those that endorse fatalistic attitudes and choose to forgo screening and treatment are unfairly labeled as "irrational” and in need of education (Drew & Schoenberg, 2011; Espinosa de los Monteros & Gallo, 2011; Straughan & Seow, 1998). The decontextualized labeling of those who express fatalistic beliefs serves to misdirect
2


attention away from the very real barriers to health care that Latinos face and offers little in the way of producing meaningful interventions designed to increase use of CVD screening services. Second, these views ignore the social, economic and institutional barriers that may inform the health beliefs and decision-making processes of those who hold fatalistic attitudes and perpetuate a shortsighted public health discourse characterizing individuals as solely responsible for their health and wellbeing (Keeley et al., 2009).
Reintroducing context to the study of fatalism and screening behavior is essential if we are to more fully understand how individuals rationalize about the use of prevention resources when faced with persistent structural obstacles to health care services. However, the research community’s overreliance on solely quantitative approaches has left fundamental questions about how people internalize, reason about, and respond to institutional barriers to health care unanswered. This is not to say that quantitative methods are unimportant to the study of the screening behavior of vulnerable populations, but only that when used in isolation they don’t go far enough in answering questions about how people relate to and make sense of these institutional barriers. Failing to answer these types of "how” and "why” questions leaves us with a limited understanding of how contextual variables impact people’s motivations and attitudes towards CVD prevention and use of screening resources.
To fully understand how institutional barriers impact the health beliefs and screening behaviors of Latinos, research must employ the complementary strengths of qualitative methods. While quantitative methods allow for the testing of hypotheses regarding the contribution of fatalism to screening behavior they remain insufficient to the
3


study of fatalism in context when used in isolation. Employing solely quantitative methods to the study of fatalistic attitudes and screening behavior of Latinos fails to account for how people interpret and respond to institutional barriers to health care and leaves our inquiry into the screening behavior of this population shortsighted and woefully incomplete. Adding qualitative methods permits one to critically explore the meaning of the quantitative results obtained while investigating how institutional barriers may be incorporated into the lived experience, decision-making processes and health behavior of this population. Therefore, this study employs mixed methods to more fully understand the relationship between institutional barriers to health care, fatalism towards CVD, and the use of cholesterol and blood pressure screening services by Latinos in Denver, Colorado. The goal of this study is to answer the following research question.
Research Question
1. How does accounting for the social context in which health care decisions are made improve our understanding of fatalism and its impact on the CVD screening behaviors of Latinos in Denver, Colorado?
Specific Aims and Hypotheses
1. Quantitatively examine the relative contributions of specific institutional barriers to health care (e.g., lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care encounter), and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol) by Latinos in Denver, Colorado.
4


Hypothesis la: Institutional barriers to health care resources will
demonstrate an inverse relationship with use of cholesterol and blood pressure screening services.
Hypothesis lb: Fatalism will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services.
Hypothesis lc: Institutional barriers to health care resources will demonstrate a positive relationship with fatalism.
Hypothesis Id: Institutional barriers to care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism.
2. Qualitatively explore how institutional barriers to health care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latinos in Denver, Colorado.
Background
More than doubling since 1990, the Latino population is the largest and fastest growing ethnic minority in the United States; they currently number nearly 58 million or 18 percent of the total U.S. population (Flores, 2017). Recent estimates gauge that by the year 2050, U.S. Latinos will reach 119 million and comprise 29 percent of the total population. Of the major origin groups, those identifying as Mexican are currently the most numerous (63 percent), followed by Central and South Americans (13.5 percent), Puerto Ricans (9.2 percent) and Cubans (3.5 percent) (US Census, 2015).
5


Although Latinos share a number of cultural assets and protective factors that influence their health behaviors, they are widely considered to be a socio-demographically vulnerable population (Palacio, 2009). Nationally, 42 percent of Latinos aged 18 to 64 lack health insurance (CDC, 2011), 67 percent are poor or near poor (CDC, 2011), 41 percent have less than a high school education (CDC, 2011), and 6 percent report perceived discrimination in the health care system in the past 12 months (Saha et al., 2003; Stepanikova & Oates, 2016). These facts are troubling to say the least. Given the Latino population’s rapid rate of growth and disparate risk profile, there is real concern that an increase in CVD among Latinos in the United States is close at hand. According to Davidson et al. (2007), "Along with its growth in size, this ethnic group is experiencing an increase in the prevalence of cardiovascular disease and related risk factors, which is in stark contrast to the limited healthcare resources, screening, and prevention tools at the disposal of Latinos/Hispanics” (p. 568).
Latinos in Colorado
Latinos have been a significant part of Colorado’s population even prior to Mexican and Spanish settlers establishing Colorado’s first permanent town, San Luis, in 1851 (Palacio et al., 2009). Colorado’s Latino population has grown steadily due to migration from the Southwestern United States, Mexico and, in more recent years, Central America. Similar to national trends, Colorado’s Latino population has more than doubled since 1990 and now numbers nearly one million residents. At 20 percent of the state’s population, Latinos are the largest ethnic minority group in the state (Palacio et al., 2009). While Colorado’s Latino population is diverse, the 2010 Census found a majority identifying as "Mexican” (US Census, 2010). Although widely distributed across the state, the population
6


is more densely concentrated in the Denver, Pueblo and Greeley metropolitan areas, the San Luis Valley and mountain resort communities (Palacio et al., 2009).
CVD is a leading cause of death for Colorado Latinos. In addition, the state’s Latino population experiences similar socio-demographic disadvantage to that of their national counterparts with over 40 percent lacking health insurance, over 23 percent living below the poverty level, and 43 percent without a high school education (Palacio et al., 2009). Moreover, Colorado Latinos shoulder a disproportionate burden of CVD risk. According to recent data from the Colorado Department of Public Health and Environment, the prevalence of high blood pressure in Latinos was greater than that for non-Hispanic Whites (Calanan et al., 2011). In addition, Colorado Latinos are twice as likely as non-Hispanic Whites to have diabetes, a major risk factor for CVD (Palacio et al., 2009). Among all adults who report having diabetes in Colorado, 65 percent report high blood pressure and 63 percent report high cholesterol, compared with 19 percent and 31 percent, respectively, of adults who do not report having diabetes (Calanan et al., 2011). These data demonstrate that Latinos are at disparate risk for CVD and may in fact suffer the consequences in years of life lost, as their life expectancy is lower than that for all Coloradans (Palacio et al. 2009). CVD Screening Services
Risk factor screening is an integral part of a comprehensive approach to CVD prevention. Most importantly, results from screening tests provide the individual and their health care provider with the necessary knowledge to make informed decisions about their health and the prevention of disease. Screening for high blood pressure and high cholesterol is critical to efforts concerned with identifying persons whose risk factors may be undiagnosed and who might benefit from lifestyle modification or pharmacologic
7


therapy. To decrease morbidity and mortality from CVD, public health programs must identify persons with multiple risk factors and focus interventions on those populations disproportionately affected (Calanan etal. 2011).
Although progress against CVD has been made, declines in death rates between 1999 and 2008 have not been equally distributed among Colorado’s racial and ethnic groups (Calanan et al., 2011). For example, the decrease in death rates due to heart disease was only significant for non-Hispanic Whites and Blacks during this time period. This may be due in part to differences in the use of screening and treatment services by Colorado’s major racial and ethnic groups.
Cholesterol Screening
According to the National Cholesterol Education Program and the Third Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (ATP III), all adults over the age of 20 should have a fasting lipoprotein profile once every five years (NIH, 2001). A fasting lipoprotein profile consists of a blood test and provides information on total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides. However, both governing bodies recommend that cholesterol be checked more frequently if (a) total cholesterol is 200 mg/dL or more, (b) you are male and over 45 years of age, (c) you are female and over 50 years of age, (d) your HDL cholesterol is less than 40 mg/dL or (e]you have other risk factors for heart disease and stroke. Due to the prevalence of multiple CVD risk factors common in Latino populations, including high cholesterol, uncontrolled hypertension, type II diabetes, and being overweight (Blackwell, Lucas & Clarke, 2014), it stands to reason that members of this population should have their cholesterol checked more often in order to
8


accurately detect the presence of hyperlipidemia and/or assess whether pharmacologic therapy is working as intended. Unfortunately, recent data show that only 58 percent of Colorado Latinos have had a cholesterol test in the past five years compared to 78 percent of non-Hispanic Whites (Calanan et al., 2011).
Blood Pressure Screening
As with cholesterol, guideline-based recommendations state that blood pressure should be checked at regular intervals. According to The Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7), blood pressure should be checked every two years if less than 120/80 mm/Hg (USPSTF, 2007). However, the American Heart Association (2016) recommends that blood pressure be checked more frequently if above this level.
Data compiled by the Colorado Department of Public Health and Environment indicates that 28 percent of the state’s Latinos have hypertension (Calanan et al., 2011). However, prevalence rates of high blood pressure in Colorado are likely to underestimate actual rates due to self-report and other limitations associated with data collection procedures. For example, anyone who has high blood pressure who has not visited a health care provider, has not been tested for high blood pressure, or does not remember being diagnosed is not included in the prevalence estimate. More specifically, prevalence rates among Latinos in Colorado may be further underestimated due to the fact that this population is more likely to experience a lack of health insurance and access to care than non-Hispanic Whites and thus may not receive blood pressure tests and diagnoses as frequently as their White counterparts (Calanan etal., 2011).
9


CVD Screening Services: Institutional Barriers
Institutional barriers to health care have received a great deal of attention in the literature on health disparities between Latinos and non-Hispanic Whites. Evidence suggests that certain barriers such as lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care system negatively impact Latinos’ use of screening services. In the following sections, I review previous research findings regarding the influence of these specific barriers to health care on Latinos’ use of blood pressure and cholesterol screening services.
Health Insurance
Access to health care services in the United States is highly dependent on availability and access to health insurance. Those who lack insurance must independently pay for health care services or seek care from safety net facilities such as free clinics, hospital emergency rooms, and community health centers (Rodriguez et al., 2014). Research has identified lack of health insurance as an important barrier to the receipt of health care services (Burstin etal., 1998; Jurkowski & Johnson, 2005; KFF, 2018; Solis etal., 1990; Summer, 2002; Ward et al. 2004). When compared to other racial or ethnic groups, Latinos are more likely to be uninsured (Jurkowski, 2006). Employment in jobs that do not offer benefits and do not require citizenship or English language proficiency constitute the primary barriers to health insurance for this population (Angel, Angel, & Lein, 2003).
Having access to care may facilitate CVD screening, which in turn may prevent or delay the development of the disease and prevent morbidity and mortality. For example, having health insurance leads to higher rates of CVD diagnoses (Alcala 2015) and reduces the risk for major cardiac events (Amarenco et al., 2014). This suggests that CVD is more
10


likely to be identified and controlled among the insured. Unfortunately, longer-term uninsured status among Latinos is linked to the decreased likelihood of engaging in behaviors that could reduce CVD risk such as screening for hypertension and high cholesterol (Alcala et al., 2015).
Research on CVD screening practices among Latinos found that those with health care coverage were more likely to have had their blood pressure checked and to have obtained a routine check up than those without coverage (DeVoe et al., 2003; Jurkowski & Johnson, 2005). In similar fashion, recent work conducted by Alcala et al. (2015) found that among 464 Latino adults, those with health insurance were at significantly greater odds of having a cholesterol test when compared to those without insurance (OR 3.76, 95% Cl 1.73-7.76). Likewise, research carried out among Latinos in Texas by Salinas et al. (2015), demonstrated that health insurance was positively associated with having both blood pressure screening and cholesterol screening. According to Burstin and colleagues (1998) Latinos without health insurance or those who change health plans are more likely to delay seeking medical care, thus risking the development of complications that could be avoided with early detection.
Using a nationally representative sample of 8,309 adults, Sudano and Baker (2003) demonstrate that lack of health insurance coverage for any period of time is significantly associated with lower rates of use of most preventive health services including cholesterol screening. Results from this study showed that those who obtained health insurance during the study period had rates of use that were approximately halfway between those without coverage and those who had continuous health insurance coverage. This suggests that
11


obtaining health insurance leads to improved use of CVD screening services for those who were previously uninsured.
Lack of health insurance undermines access to and use of preventive health care services including blood pressure and cholesterol screenings. Given their disproportionate representation among the uninsured, Latinos are vulnerable to the negative CVD-related health outcomes associated with undiagnosed hypertension and hypercholesterolemia. Studies concerned with the screening behavior of Latinos must adequately consider the implications that a persistent lack of health insurance presents in this population. Failure to do so ignores the substantial impact this barrier has on use of CVD screening services by Latinos in the United States.
The Patient Protection and Affordable Care Act
Discussion of access to health insurance and its potential impact on use of screening services must consider how The Patient Protection and Affordable Care Act (ACA) has changed the availability of health insurance to those who have not been able to access health care prior to the law’s passing. The ACA became law in the United States on March 23, 2010. Considered a watershed moment in U.S. public health policy, the ACA established basic legal protections that have until now been absent, including access to affordable health care coverage for most Americans from birth through retirement (Rosenbaum, 2011). Consisting of 10 separate legislative titles, the ACA has five major aims including to: (a) achieve near universal coverage through a shared responsibility among government, individuals, and employers; (b) improve the fairness, quality and affordability of health insurance coverage; (c) improve health care value, quality and affordability while reducing wasteful spending; (d) strengthen primary care access while incorporating longer-term
12


changes in the availability of primary and preventive health care; and (e) make targeted investments in the public’s health through expansion of both clinical preventive care and community resources (Rosenbaum, 2011).
The legal protections set forth by the ACA assume that the pervasive disparities in health insurance coverage between Latinos and non-Hispanic Whites will disappear. However, despite the aim to achieve near universal coverage, the ACA is expected to leave nearly 25 million people without health insurance (Rosenbaum, 2011). Recent analyses of American Community Survey data by Artiga and Orgera (2019) demonstrate that along with American Indians and Alaska Natives, Latinos represent the largest proportion of the uninsured in the United States with 19 percent still lacking coverage. In Colorado, Latinos also represent the largest proportion of the uninsured with 16 percent of the total Latino population lacking health insurance compared to only six percent of non-Hispanic Whites (Artiga & Orgera, 2019). In states that chose not to expand Medicaid coverage under the ACA, nearly 27 percent of Latinos lack coverage compared to only 10 percent of non-Hispanic Whites (Artiga & Orgera, 2019). According to data compiled by The Colorado Health Institute, as of 2017, Latinos in Colorado remain almost twice as likely to be without health insurance as non-Hispanic Whites with 115,000 still lacking coverage (CHI, 2017). Although they comprise only 21 percent of the state’s population, Colorado Latinos accounted for 34 percent of the uninsured in 2017 (CHI, 2017).
Despite millions having gained access to health care coverage under the ACA, many individuals within the larger Latino community continue to experience real barriers to obtaining health care coverage and services. This is particularly true for unauthorized individuals, as the ACA does not include health care coverage options for undocumented
13


immigrants (Ortega, Rodriguez & Vargas Bustamante, 2015). Assuming that Latinos will achieve equitable access to health insurance under the auspices of the ACA is short sighted and fails to account for ongoing barriers that this population faces. Culturally appropriate and targeted outreach and enrollment strategies are critical to link community members with available health coverage and services (Maxwell et al., 2011). For example, targeted outreach efforts by local community-based organizations were an essential part of Massachusetts’ health care reform efforts to successfully enroll Latinos in health insurance programs and provide a model for national efforts (Maxwell et al., 2011). Despite such successes, much of this work is just beginning. The real impact of the ACA on access to affordable health insurance for Latinos will only be known in the coming years.
Usual Source of Health Care
Along with limited access to health insurance, not having a usual source of health care is an important institutional barrier to health care for Latinos. Having a usual source of health care (USOC) has been associated with greater service use and earlier receipt of preventive services (Ettner, 1999; Finney Rutten et al., 2015). This may be explained in part because patients with an ongoing, long-term relationship with their physician may have heightened awareness of certain health behaviors along with the opportunity to discuss these behaviors, gain access to health-related information, and treat problems (Ettner, 1999). Previous research has shown that Latinos are less likely to have a USOC than non-Hispanic Whites (CDC, 2013; Corbie Smith etal., 2002; Guendelman & Wagner, 2000).
According to findings from research using a nationally representative sample of Latinos carried out by Rodriguez, Bustamente and Ang (2009), more than 25 percent of
14


participants lacked a USOC. The study’s authors demonstrate that USOC independently predicts a blood pressure check and results in greater likelihood of having both a blood pressure and cholesterol test compared to no usual source of care. In another nationally representative sample of health care use among adults in the United States, having a USOC was independently associated with receiving multiple preventive services, including blood pressure and cholesterol screening (Corbie Smith et al., 2002). In similar fashion, Torres et al.’s (2007) work among a nationally representative sample of U.S. adults showed that lack of health insurance and a USOC were among the strongest predictors for the failure of Latinos to receive preventive services.
In work examining health services utilization among a nationally representative sample of Latinos and non-Hispanic Whites, Guendelman and Wagner (2000) found that having a USOC significantly reduced the gap between the two populations in preventive care utilization including blood pressure and cholesterol screening tests. Much like Torres et al.’s (2007) work, Guendelman and Wagner’s (2000) findings illustrate that USOC and health insurance coverage are the most significant factors involved in eliminating or reducing the disparities in utilization of preventive health services between Latinos and non-Hispanic Whites.
Having a usual source of care is associated with greater use of preventive care and monitoring services. Preventive care and health care monitoring are both associated with better long term health outcomes, including better control of chronic health conditions such as CVD (Livingston, Minushkin & Cohn, 2008). Due to the fact that Latinos are less likely than non-Hispanic Whites to have a USOC, this population may not equally experience the benefits that having a USOC confers such as heightened awareness of and
15


the opportunity to screen for and treat adverse health conditions. Not having a USOC erodes one’s ability to establish a consistent source from which critical health care information and resources can be obtained. Efforts to understand Latinos’ suboptimal use of CVD screening services must assess the relative contribution of USOC as well as how this barrier may influence the very attitudes of Latinos towards CVD prevention services. Otherwise, we risk leaving the critical influence of this institutional barrier unaccounted for.
Perceived Discrimination in the Health Care System
Over the past 15 years, research has paid increasing attention to the impact perceived discrimination has on racial/ethnic disparities in the use of health care (Burgess et al., 2008). Several studies concerned with the relationship between perceived discrimination and use of health care services show support for the hypothesis that perceived discrimination in the health care system negatively influences use of services offered (Burgess et al., 2008). Increases in reported experiences with perceived discrimination are associated with lower rates of participation in use of preventive health services and the elevated likelihood of delaying care (Burgess et al., 2008).
According to Trivedi and Ayanian (2006) perceived discrimination maybe related to use of preventive services in three ways. First, those who feel discriminated against may be less likely to seek visits for routine care at which preventive services are likely to be offered. Second, people who experience perceived discrimination may lose trust in their providers thus making them less likely to follow recommendations for use of certain prevention or screening tests. Third, patients who perceive discrimination may receive less decisive recommendations for health-related tests and treatments. Findings from Trivedi
16


and Ayanian’s (2006) work show that 21 percent of Latinos cite insurance type as the most frequent reason for discrimination, with 20 percent and 12 percent citing race/ethnicity and language/accent as the next most common reasons respectively. According to their findings, persons who report perceived discrimination in a health care encounter are significantly less likely to receive several preventive health screenings including cholesterol testing.
According to work conducted by van Ryn (2002) and van Ryn and Burke (2000), patient race/ethnicity can influence physician interpretation of patients' illness narratives and decisions to refer patients for particular treatments or procedures (van Ryn, 2002; van Ryn & Burke, 2000). Other research findings indicate that quality of health care provided may differ substantially by racial/ethnic group including receiving needed care and timeliness of care (Benkert, Peters, Morales, Elliott, Weech-Maldonado, Spritzer, & Hays, 2001); quality of communication (Collins, Clark, & Kressin, 2002); perceived discrimination due to insurance status (Trivedi & Ayanian, 2006); provider assessment of patient intelligence, likelihood of risk behavior and adherence with medical advice (van Ryn, 2002; van Ryn & Burke, 2000); and language associated perceptions of staff helpfulness (Weech-Maldonado, Morales, Elliott, Spritzer, Marshall, & Hays, 2003).
This body of work demonstrates that minority groups receive less equitable care and face perceived discrimination in health care settings.
Patient perceptions may affect their health-related decision-making with previous studies revealing the impact of perceived quality of care on adherence to medical advice (Blanchard, 2004), screening recommendations (Trivedi & Ayanian, 2006), and medication regimens (Blanchard & Lurie, 2004; Van Houtven etal., 2005). Sorkin etal. (2010) suggest
17


that perceived discrimination during the health care process may represent an important mediator between race/ethnicity and observed racial/ethnic differences in patient perception of care. Research demonstrates that minority patients are more likely to report being the subject of negative attitudes during the health care process than are Whites (Chen, Fryer, Phillips, Wilson, & Pathman, 2005; Haviland, Morales, Reise, & Hays, 2003; Haviland, Morales, Dial, & Pincus, 2005; Kandula, Hasnain-Wynia, Thompson, Brown, & Baker, 2009; LaVeist, Rolley, & Diala, 2003).
In a random sample of more than 40,000 households in California, Sorkin and colleagues (2010) found that Latinos were more likely to feel discriminated against in health care because of their race and ethnicity than were non-Hispanic Whites. In other research based on the National Consumer Assessment of Health Plans Study (CAHPS), Latinos reported negative experiences in every area assessed, including receiving needed health care, communication with health care providers, and timeliness of care (Weech-Maldonado et al., 2010). Recent work on perceived discrimination and use of health care services among Latinos in North Carolina by Keller, Silberberg, Hartmann, and Michener (2010) found that perceived discrimination, based on both race/ethnicity and insurance status, was associated with an increased likelihood of going without needed medical services.
Perceived discrimination has a negative influence on health by increasing the likelihood that individuals will avoid seeking important health services. Experiences of discrimination and unfair treatment may engender negative expectations by Latinos about how they will be treated within medical institutions or by members of the dominant culture, thereby making them hesitant to enter such encounters (Burgess et al., 2008).
18


Perceptions of discrimination in health care encounters by Latinos inhibits trustful communication between patients and providers and leads to members of this population being less likely to use critical preventive medical procedures and services including blood pressure and cholesterol tests.
Fatalism
In an effort to account for the suboptimal use of screening and treatment services by Latinos, research has begun to increasingly examine the impact of health-related beliefs or attitudes. According to Drew and Schoenberg (2011), in the health sciences one such belief, the fatalism construct, has increasingly been applied to account for both health behaviors and outcomes. When applied to health, fatalism is often operationalized as negative or pessimistic attitudes regarding preventive health practices and disease outcomes (Drew & Schoenberg, 2011).
Fatalism (fatalismo) is often considered to be a dominant cognitive orientation of Latinos, a population that is frequently described in the literature as passive, subjugated, and fatalistic (Cueller, Arnold, & Gonzales, 1995; Espinosa de los Monteros & Gallo, 2013). Early anthropological work among Mexican Americans (Madsen, 1973) suggested that suffering endured by this population is made more acceptable by a strong belief in fatalism. According to Madsen (1973), "It is generally believed that the good or bad fortune of the individual is predestined and every occurrence in human existence comes to pass because it was fated to do so” (p. 18). He goes on to state, "What the Anglo tries to control, the Mexican American tries to accept. Misfortune is something the Anglo tries to overcome and the Latin views as fate” (Madsen, 1973, p. 18).
Unfortunately, this type of ahistoric labeling continues with more recent work on
19


fatalism and its affect on the use of disease screening services by minorities, perpetuating the view that fatalism among Latinos is an inherently cultural trait (Balcazar, Castro, & Crull, 1995; Harmon, Castro, & Coe, 1996; Hubbel, Chavez, Mishra & Burciaga-Valdez, 1996; Hubbel, Chavez, Mishra & Burciaga-Valdez, 19962; Otero-Sabogal, Stewart, Sabogal, Brown, & Perez-Stable, 2003; Perez-Stable et al., 1992; Suarez, Nichols, Roche, & Simpson, 1997). These views ignore the context within which health-related beliefs and behaviors are formed and maintained. Failure to account for how institutional barriers impact Latinos’ perceptions of CVD prevention and use of health care screening services leads to faulty assumptions about rationality and Latino culture. Furthermore, decontextualizing the study of fatalism and health behavior leads to the misattribution of underuse of screening services by Latinos to irrational beliefs rather than institutionalized barriers that impact this populations’ health-related agency. According to Espinosa de los Monteros and Gallo (2011), "One of the major criticisms of research emphasizing the role of cognitive processes in explaining the health behavior of disadvantaged populations is that structural barriers may better account for health disparities than cognitive processes which could merely reflect the structural barriers experienced by such populations (p. 316).” The research community’s failure to adequately account for the structural contexts within which fatalism may occur perpetuates the view that fatalism is a faulty belief system held by Latino culture. Unfortunately, this limited view and singular focus on cognitive traits has hindered efforts to better understand the complex relationship between barriers to health care, fatalistic attitudes and use of CVD screening services by this population.
Previous Research on Fatalism and Latinos' Use of Screening Services
In research examining motivations for use of health screenings among 452 Latinos,
20


Ashida et al. (2010) asked participants how much control they believe they have in preventing heart disease and diabetes. On a scale of 1 (no control) to 4 (total control), a mean score of 2.46 + 0.82 revealed that participants partially attributed the prevention of CVD to sources out of their control. As a result of qualitative work conducted with Latinas, Garces et al. (2006) reported that many of the women believed they did not have control over their own health, attributing this lack of control to "system” barriers such as lack of information, language barriers, and lack of access to health care and medications. In similar fashion, Shive et al. (2006), found that among a nationally representative sample, Latinos were more fatalistic toward their health than were African Americans, non-Hispanic Whites, and Asian Americans. However, despite collecting data on health behaviors, such as use of cholesterol and blood pressure screenings, the authors do not report whether fatalism was associated with the use of these services. More importantly, the authors fail to discuss how fatalism may be related to socioeconomic disadvantage and institutional barriers to health care despite collecting data on these socio-demographic variables. This results in unanswered questions about the potential context-dependent nature of fatalism.
In work designed to better understand participation in community health screenings, including blood pressure and cholesterol, Engebretson, Mahoney, and Walker (2005), found a complex array of facilitators and inhibitors across personal, interpersonal, and environmental dimensions to be responsible for whether individuals utilized health screenings. Importantly, their findings reveal that Latinos may not use screening services, even if offered at no cost, due to concern over finding something might be wrong and not having the economic resources with which to address the problem. The worry and stress among Latino study participants over potential negative screening results and the inability
21


to access and pay for follow up care identified by Engebretson and colleagues reveal how a lack of control many Latinos feel may result in expressions of fatalism toward using screening services such as blood pressure and cholesterol. Despite this finding, Engebretson’s qualitative sample is small and does not explicitly evaluate the relationship between fatalism towards CVD and use of blood pressure and cholesterol screening services after accounting for institutional barriers to care. Combining this type of qualitative work with quantitative assessment will more fully explain how these particular barriers are related and the meaning they hold for members of the Latino community.
Perhaps most widely applied in the cancer literature, fatalism has been shown to act as a deterrent to the use of certain screening and prevention services by Latinos. In an effort to determine whether fatalism predicts participation in cancer screening after accounting for structural barriers to health care, Espinosa de los Monteros and Gallo (2011) conducted a meta-analysis of all relevant studies published before February 25, 2010. Of the 11 studies meeting the inclusion criteria, seven reported a statistically significant inverse relationship between fatalism and use of cancer screening services after accounting for structural barriers such as health care access and socioeconomic status (SES). However, the authors note that mixed findings and inconsistencies in study measures and design preclude definitive conclusions regarding the nature of the relationships found. The authors’ decision to include only quantitative studies in their analysis fails to capture the potential complexities of the relationship between structural constraints, fatalism, and Latinos’ use of screening services. For example, despite finding inverse associations between fatalism and key socio-demographic variables such as SES, limited access to health insurance and lower education and/or income in six of the eleven
22


studies, the authors fail to address how fatalism may be more accurately represented as a mechanism linking structural barriers to screening rather than an independent predictor of screening behavior. Moreover, the sole reliance on quantitative studies falls short in answering crucial questions regarding the motivations that underlie people’s fatalistic attitudes and behavioral practices. To answer these types of questions, the inclusion of qualitative methods is essential.
In work to determine which factors predict use of mammography screening by Latinas, Laws and Mayo (1998) include a modified health locus of control variable as a proxy for fatalism and several social variables, including education level and insurance status, in their model to answer whether the relationship between these social factors and screening are the same or different for Latina versus non-Latina women. Their results show that access to primary care services via health insurance and physician behavior act as key determinants of use of mammography screening among their sample. Alternately, the authors illustrate that fatalistic attitudes towards cancer have little to do with screening behavior in their sample of Latina women.
Research designed to elicit the complexity of fatalism and its relationship to rural women’s decision-making surrounding the use of health care services (Drew& Schoenberg, 2011) shows that numerous factors including inadequate access to health care services, a legacy of self-reliance, and a culturally acceptable idiom of fatalism foster the use of, but not necessarily a rigid belief in, fatalism towards disease. For study participants, health decisions and behavior are embedded in a larger context that includes real life circumstances and constraints such as power and agency and personal and familial expectations that intersect with health care provider interactions. According to Drew and
23


Schoenberg (2011),"Rather than reducing complex health decision-making to those ‘factors’ that remain once other predictors show no statistical significance...” (p. 177), we must also consider the complex set of circumstances and experiences that underlie one’s decision to undergo screening and treatment. This is critically important to efforts designed to increase Latinos’ use of CVD screening services. In order to truly understand the health beliefs and behavioral motivations of Latinos in the U.S., we must move beyond quantitative only approaches to the study of screening behavior by incorporating qualitative methods into our research designs. Doing so will serve to strengthen our inquiries by providing us with the ability to answer essential questions about how the context within which decisions are made influence the very behaviors we hope to encourage.
Anthropological Contributions
Anthropological inquiry1 into fatalism and health behavior is sparse but holds important insight for understanding how fatalism may mask social realities and the variations in the way individuals cultivate meaning regarding use of medical services or treatment. In work among terminally ill dialysis patients in Egypt, Hamdy (2009) argues that rather than viewing fatalism as passive submission to divine authority, it is better understood as a disposition that must be actively cultivated through work on the self.
1 The bulk of work on fatalism and health behavior has taken place in the fields of psychology, public health, speech communication, and the socio-medical sciences. Anthropological research on fatalism is much less common. Anthropological findings on the influence of fatalism on health behavior and use of screening services emphasize the role of both structural and cultural variables in understanding fatalism and behavior. Emphasis on social context has led anthropologists to identify social and material deficits (e.g., poverty, unequal power relations, and inadequate access to culturally appropriate health care resources) common among minority racial and ethnic populations as the organizing factors in the development of fatalism among such groups.
24


Hamdy demonstrates that whether and when to cultivate such a disposition is contingent upon how much control patients feel they have when faced with illness and whether there are readily available options for solutions that are acceptable medically, socially and spiritually. Hamdy’s analysis shows that for her study participants, structural and material conditions intersect with the logics of religion as they face difficult decisions about medical care, and oftentimes, limited options for recovery.
In fieldwork among the Fulani and Humbebe in the Mopti region of Mali, Castle (1994) examines the relationship of mothers’ kin and power relations to the process of diagnosing a child with a fatal illness. For children who die before the age of five, death is often attributed to the folk illnesses foondu (the bird - owl) and heendu (the wind). According to Castle, these diagnoses act as explanatory mechanisms in a high mortality area where the Fulani code of behavior known as pulaaku, mandates acceptance and stoicism when faced with physical hardship and emotional distress. For the Fulani, both foondu and heendu are viewed as very serious and incurable illnesses. The incurable nature of these diagnoses can be taken as proxies for fatalistic appraisals of these particular illness events (Castle, 1994). What Castle illustrates is that the application of foondu or heendu as a cause of death rarely occurs during current illness episodes but rather during retrospective accounts. According to Castle, the concepts of fate or destiny are manipulated such that giving fatal diagnoses to sick or dying children or applying them retrospectively after death, absolves mothers from guilt, blame and responsibility. In addition to transferring responsibility away from the individual (mother) to the community at large, Castle’s analyses illustrate how diagnoses of foondu and heendu serve to reinforce hierarchical power relations among women due to the fact that senior women in a patrilineage only
25


confer diagnoses. Castle suggests that it is possible in a community facing extremely high rates of child mortality that foondu and heendu allow psychological adaptations to the frequent experience of death and contribute to a "sense of coherence” under circumstances that would otherwise seem uncontrollable.
In similar fashion, Kamat’s work (2008) among the Zaramo people of Tanzania shows that for parents of children diagnosed with degedege (the indigenous name for a life threatening illness otherwise viewed by biomedicine as cerebral malaria), treatment seeking and therapy management are shaped by multidimensional influences. While decisions to forego treatment in an effort to save a dying child may be construed by Western observers as the result of "irrational beliefs,” "ignorance,” and "fatalism,” Kamat demonstrates that economic, structural, and micro-level social factors intersect with cultural beliefs to influence views about disease and the decision-making process associated with obtaining a diagnosis and seeking treatment (Kamat, 2008). In this context, structural factors, including poverty, lack of access to dispensary facilities, and unpleasant encounters with the dispensary staff serve to affect how the Zaramo view the options available to them when faced with degedege. Failure to gain timely access to biomedicine that is safe, affordable, and efficacious and unpleasant encounters with health care professionals highlight how structural factors, power relations, and economic circumstances conspire to organize health beliefs and behavior among this particular group.
From an anthropological perspective, Nichter (2002) cautions that the study of therapy management must include not only what people do (and can do) including the reasons for their actions, but also what they are unable to do. He urges us to examine the
26


factors that underlie apparent passivity or fatalism. Failure to do so reinforces simplistic impressions that ignorance is at the root of "irrational health behavior” in contexts where other factors are involved (Nichter, 2002).
Anthropological inquiry into fatalism demonstrates that multiple social, cultural, and structural factors frequently serve to organize one’s attitudes, beliefs, and decisionmaking practices regarding health. Rather than viewing health-related behavior as seemingly irrational, anthropologists emphasize the multiple contexts in which decisions are made. Most importantly, insights gleaned from cross-cultural work carried out by anthropologists show that emic or insider perspectives are critical to understanding the rationale behind how, when, and why people do and do not seek health care services.
The present research attempts to apply these lessons in an effort to account for as of yet unanswered questions about the impact of structurally organized barriers to health care on the CVD-related beliefs and behaviors of Latinos in the U.S. To do this, qualitative methods will serve to complement and extend the quantitative approach employed. Through mixed methods, this project seeks to comprehensively address a critical gap in current knowledge regarding how contextual variables influence the expression of fatalistic attitudes toward CVD and suboptimal use of blood pressure and cholesterol screening services. Research that reintroduces context to the study of fatalism and health behavior is critical to the design and implementation of interventions designed to increase Latinos’ use of CVD screening services. By more fully understanding how people rationalize about the use of disease prevention tools when faced with structural barriers to health care, public health interventionists will be better equipped to design and implement tailored approaches that effectively incorporate responses to the challenges of distributing
27


equitable health care to this population.
Structure of the Dissertation
This study draws upon mixed methods and multidisciplinary perspectives to understand the phenomenon of suboptimal use of CVD prevention services by Latinos. Chapter 1 introduces the research questions and explores previous research in the area of screening practices among Latinos. Chapter 2 examines the application of psychological and sociological theories traditionally used to explain health beliefs and health behavior in the context of screening practices. Specifically, this study critiques the use of individually based theories to understand fatalistic attitudes among Latino populations and draws upon the concept of structural violence to account for the expression of fatalistic attitudes towards CVD and suboptimal use of CVD prevention services. Chapter 3 describes the study design as well as the quantitative and qualitative methods employed. Chapter 4 discusses the quantitative analyses and results. Chapter 5 discusses the qualitative analyses and results. Chapter 6 synthesizes the results and discusses the implications of the findings in relation to both intervention efforts and future research.
28


CHAPTER II
THEORY
Despite the persistent view among the health sciences community that fatalistic attitudes and beliefs represent an irrational cognitive orientation held by ethnic minorities, more recent theories on the development and maintenance of fatalism emphasize the importance of considering how social and material barriers (e.g., poverty, discrimination, limited access to quality health services and treatment), contribute to its development and expression (Abraido-Lanza et al., 2007; Espinosa de los Monteros & Gallo, 2013). This line of thinking also proposes that poor health outcomes experienced by vulnerable and disadvantaged populations may actually reinforce fatalistic beliefs about the utility of disease prevention and treatment behaviors (Abraido-Lanza etah, 2007; Freeman, 1989).
Consideration of the role that pervasive institutional barriers to health care may play in the creation and maintenance of fatalistic attitudes and health behavior is critical when studying Latinos’ use of health care services. Given their overrepresentation among the poor, less educated, and uninsured, Latinos face an array of agency-eroding obstacles to achieving equitable access to quality health care services. In light of their vulnerable status, I argue that, for Latinos faced with seemingly insurmountable institutional barriers to health care, fatalistic beliefs about cardiovascular health and illness are grounded in the realistic appraisal of one’s individual control over heart disease and represent a meaningful recognition that while good health is valued, barriers to equitable health care services make it difficult to achieve through personal effort alone (Espinosa de los Monteros &
Gallo, 2013).
29


A Faulty View: Fatalism as a De-Contextualized Cognitive Trait of the Individual
Within the public health literature, much of the work on fatalism and why people choose to forego health screenings, or a particular course of treatment that is known to be efficacious, has taken a primarily cognitive orientation (Drew & Schoenberg, 2011). Despite acknowledging that other factors (e.g., lack of health insurance) may play a role in whether one utilizes screening resources, researchers studying fatalism often attribute individual level fatalistic beliefs to shaping and directing prevention behavior.
The focus on fatalism and cognitive mechanisms in the explanation of suboptimal health behaviors may be partly understood in terms of the shift in emphasis from acute and infectious diseases to chronic disorders as the major cause of morbidity and mortality in twentieth and twenty first century developed nations (Davison et al., 1992). Accompanying this shift was the increasing importance of disease prophylaxis as a central motivator in the everyday behaviors of the healthy population. According to Davison et al. (1992), the most striking feature of this process has been the development of a strong, officially sponsored discourse, which emphasizes the personal responsibility of the individual in the maintenance of their own health and the avoidance of chronic disease. As part of the growth of the prevention, health education, and health promotion movements, public health focused on educating the public about the risks of certain behaviors and the importance of following a "healthy lifestyle”. However, the apparent failure of large numbers of the populace to follow the advice of a healthy lifestyle constituted a major problem for prevention science and was attributed to two major causes: 1) lack of knowledge and 2) the existence of an attitude that perceives health as being primarily
30


determined by factors outside the control of the individual, thus negating the relevance of personal behavior in the prevention and/or management of disease (Davison et al., 1992).
Previous research and critique demonstrates that the locus of control trend in psychology has served to perpetuate the idea that belief in individual control is "correct,” while belief in causal mechanisms external to the individual requires correction, usually in the form of education. "This type of analysis has led to the production of the idea that health promotion is involved in a battle for the hearts and minds of the population, a struggle between a modern belief in lifestyle and an atavistic culture of ‘fatalism’ (Davison et al., 1992, p. 676)."
The differentiation of those who hold fatalistic attitudes and believe that health is governed by sources external to the individual from those who subscribe to the importance of lifestyle and individual control is replete with moralizing judgments. Individuals demonstrating an external source of control are perceived as ignorant, irrational, and in need of education (Straughan & Seow, 1998), while those with an internal source of control are viewed as having embraced modern values that align closely with good lifestyle choices (Drew and Schoenberg, 2011,). Indeed, the perception of where control resides (internal vs. external) and how this relates to health behavior has served as a defining characteristic of the fatalism construct, including the ethical basis on which people, and most often minorities, are judged.
Locus of Control
The locus of control construct is central to the very definition of fatalism and the way it is used to label people. The varying definitions of fatalism, as related to health, all have at their very core, the concept of lack of control over health status (Davison et al.
31


1992). For example, Neff & Hoppe (1993) reference fatalism as "a lack of perceived individual control,” while Straughan and Seow (1998) conceptualize fatalism as "a belief that some health issues are beyond human control.”
Research linking perceived control to health behavior can be traced to social learning theory and the work of Julian Rotter and Albert Bandura. Social Learning Theory (Rotter, 1954; 1982; Rotter, Chance, & Phares, 1972) claims that the likelihood of a person engaging in a particular behavior is dependent upon two things: the person's expectancy that the behavior will lead to a particular outcome (or reinforcement) in that situation and the value of the reinforcement to the person in that situation (Wallston 1997). According to Social Learning Theory, individuals develop generalized expectancies based upon previous experience and learning across a variety of life circumstances. Locus of control (Rotter, 1966) is understood as one of these expectancies in Rotter’s theory. It refers to the person's belief as to whether control over valued outcomes is internal or external to the person. A person with an internal locus of control believes that consequences in life are the result of either some action (or set of actions) taken by the person, or of some individual characteristic (or set of characteristics) specific to that particular person. Internal locus of control orientation is typically equated with a perception of control over outcomes. Conversely, persons with an external locus of control believe that life events are the result of forces outside of the person. Belief in an external locus of control is most often associated with a perception of lack of control over outcomes in one’s life (Wallston, 1997).
The locus of control trend in psychology initiated the classification of individuals as either internally oriented, believing in personal control over events in his/her life, or externally oriented, and thus fatalistic, believing that one has little control over life events,
32


including health or the onset of disease (Davison et al. 1992). Health behavior researchers focused on expanding the individual’s power to avoid ill health emphasize the individual and the belief in the possibility of control over health by personal action and behavior. In their aim to identify people as either leaning towards internal or external locus of control, the health promotion and prevention movements ultimately saw those who gave weight to external forces, or the "fatalistic,” as targets for intervention designed to encourage belief in personal control (Davison etal., 1992).
Locus of control is a fundamental part of the fatalism construct. Prohibitive social structures that deny socioeconomically vulnerable populations equitable access to health promoting resources may result in a loss of control or sense of powerlessness for those affected. In such circumstances, seeking out and using CVD screening resources may hold little value or benefit, particularly if use of screening services results in finding out that one has health problems that require medication or health resources that remain unaffordable and out of reach. Thus, locus of control is an essential part of the fatalism construct linking institutional barriers to health care to CVD screening behaviors.
Self-Efficacy and Models of Individual Health Behavior
A related construct that has found broad appeal among health behavior researchers is self-efficacy (Bandura, 1977). Self-efficacy refers to a person's subjective estimation that he or she is capable of engaging in a particular action (or set of actions) in a situation (Bandura, 1977). Although similar to locus of control, self-efficacy differs from the former in that it refers to control over a behavior rather an outcome (Wallston, 1997). As originally conceived of by Bandura, self-efficacy is highly behavioral and situation specific. In other words, perceived control over a behavior in a certain situation does not necessarily
33


translate to other behaviors or other situations (Wallston, 1997). For example, heart disease patients who are self-efficacious with regard to participation in a rehabilitative exercise program may not experience the same level of self-efficacy when faced with the responsibility of managing and adhering to medications prescribed by a physician (Katch & Mead, 2010).
Despite the differences outlined above, some researchers have substituted self-efficacy for locus of control as the major generalized expectancy construct (Wallston 1992). For example, in tests of Ajzen’s (1988) Theory of Planned Behavior, perceived behavioral control is operationalized similarly to measures of self-efficacy (Wallston 1997). More recently, Judge et al. (2002) have argued that self-efficacy and locus of control are actually related concepts and may in fact measure the same higher order construct. Alternatively, Straughan and Seow (1998) demonstrate support for Bandura’s argument that perceived self-efficacy is correlated with health promotion behavior by showing that those who are more fatalistic toward their health also have lower perceived self-efficacy. In turn, the authors posit that self-efficacy has an important bearing on whether study participants adopt preventive health measures such as breast and cervical cancer screenings.
Although originally advanced by Bandura (1977) and included as part of his Social Cognitive Theory (1986), self-efficacy has been incorporated into multiple models that attempt to explain human behavior. Among these are the Health Belief Model (HBM), the Theory of Planned Behavior (TPB), and the Integrated Behavior Model (IBM) (Glanz, 2008). Work on the role of fatalism in disease screening behavior has utilized both the Theory of Planned Behavior (Facione, 2002; Lopez McKee, 2008) and the Health Belief Model (Gorin, 2005) to explain the underuse of screening services by minorities.
34


Originally developed in the 1950s by social psychologists attempting to explain the widespread failure of people to participate in disease detection and prevention programs, HBM relies heavily on cognitive theory and the role of subjective expectations held by individuals to explain health-related behavior (Glanz et al., 2008). The core constructs that make up HBM are perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action, and self-efficacy. Underlying HBM’s core constructs are two key assumptions: 1. Individuals value avoiding illness/getting well and 2. Expect that a specific health action may prevent illness (Glanz et ah, 2008).
But, the problem with using the HBM to model screening behaviors is that it minimizes the potential role of structural constraints in the underuse of health promoting services by racial and ethnic minorities. For example, Gorin’s (2005) assessment of Latino women’s compliance with free colorectal cancer screening evaluates key constructs from the HBM to determine their impact on the women’s use of the Fecal Occult Blood Test. In this study, Gorin subsumes fatalism under the HBM barriers measure, which in this case is defined as women’s beliefs in effectiveness of the test, testing time, inconvenience, and embarrassment. In doing this, the author perpetuates the idea that fatalism is simply informed by individual perceptions, cognitions and beliefs and is unrelated to socially and institutionally organized barriers. Analyzing fatalism under the HBM in this fashion limits the potential for uncovering the context specific influences on fatalism and its relationship to the underutilization of screening services by Latino women.
The Theory of Planned Behavior (TPB) has its roots in the Theory of Reasoned Action (TRA), both of which focus on individual motivational factors as the primary determinants of the performance of a specific behavior (Glanz et al., 2008). TRA and TPB both assume
35


that behavioral intention, the key predictor of actual behavior, is informed by an individual’s attitude toward the behavior and the social normative perceptions that surround it (Glanz etal., 2008). While TRA focuses on cognitive factors such as beliefs and values that determine motivation to perform a behavior, TPB extends TRA by adding perceived behavioral control as an additional explanatory construct. Lopez-McKee et al. (2008) used TPB to guide their assessment of the level of fatalism in Latinas, categorized as regular versus infrequent mammography screeners. They were able to demonstrate that significant differences in control beliefs existed between the two groups, and fatalistic beliefs were significantly higher in the group characterized by infrequent use of screenings. Importantly, these differences existed among all low-income study participants, indicating that fatalism towards health may be related to barriers beyond low SES.
As with the HBM, the primary weakness of the TPB is that it fails to adequately account for how structural constraints can limit one’s health promoting agency. TPB’s assertion that behavioral attitudes, subjective norms, and perceived behavioral control lead to behavioral intention, and thus to actual behavior, make no mention of how structurally organized institutional barriers may serve to influence one’s attitudes, intentions, and actions. Leaving these types of variables unaccounted for seriously hinders the model’s explanatory power regarding suboptimal use of CVD screening services by Latinos.
Efforts to understand fatalism and its relationship to screening behavior must move beyond the study of subjective cognitions that individually focused explanatory models give primacy to. Rather, it is imperative that research incorporates and fully accounts for the influence of social and material conditions in an effort to better understand how
36


fatalistic attitudes may be a reflection of the structural barriers to equitable health care
that Latinos face.
An Alternative View: Fatalism and its Social Origins
While there is no unifying theory of fatalism in the social sciences literature, there is a long tradition of thinking about fatalism as a social rather than psychological phenomenon. In particular, recent analyses demonstrate the value of viewing fatalism in light of theories put forth by sociologist Emile Durkheim (Acevedo, 2005). In his classic work Suicide (1968), Durkheim offers two distinct, binary concepts (anomie/fatalism) that help us to better understand how over-regulation by social structures can contribute to fatalistic beliefs. In his analysis of differing suicide rates among Protestants and Catholics, Durkheim argues that greater social control among Catholics results in lower suicide rates. For Durkheim, anomic suicide represents an individual’s moral confusion and lack of social direction, which results from dramatic social and economic upheaval. For example, extreme newfound wealth or economic ruin may result in dramatic changes to social roles and expectations such that the individual struggles to find meaning and where they fit into society. According to Durkheim, this experience of moral disorder and under regulation results in a constant state of disappointment and a heightened risk for suicide.
Durkheim’s binary opposite of anomie is fatalistic suicide. This form occurs when an individual is excessively regulated such that futures are "pitilessly blocked and passions violently choked by oppressive discipline” (Durkheim 1968:276). According to Acevedo (2005), this type of structurally induced fatalism is defined as "a sense of powerlessness that results from a combination of over regulation combined with a lack of exit option into the collective body in which the subject lacks the necessary voice and/or exit option to
37


alter their social position, status, rank, or living conditions” (Acevedo, 2005: 75). In this sense, structurally organized fatalism can result from feeling powerless when faced with prohibitive social structures, like poverty or institutional barriers to health care, which over regulate the lives of people and reduce one’s agency to act on his/her own behalf. According to Lockwood (1992), if anomie means that life’s possibilities are expanded such that human aspiration knows no bounds, fatalism refers to hopes so narrowed and diminished that life is itself a matter of indifference. It is in this sense that fatalism is applied here. I propose that for Latinos in the U.S, a persistent and inequitable lack of access to health care resources and perceived discrimination in the health care system has engendered feelings of fatalism towards CVD and the use of screening services designed to assist with the identification, prevention, and management of this disease.
Reintroducing Social Context:
Institutional Barriers and the Expression of Fatalistic Attitudes
Although it is recognized that social and economic barriers are associated with the presence of fatalistic health beliefs and use of screening services, research on fatalism and health behavior has largely failed to explicitly account for the social and institutional factors specific to the presence of fatalism among Latinos. Social context is fundamental to the fatalism concept and must be taken into account in order to fully understand the reason some populations’ hold to the belief that they have little control over their health. Research that fails to address questions about how people internalize, reason about, and respond to institutional barriers to health care limits our understanding of how and why structural obstacles serve to influence vulnerable communities’ use of prevention resources. Studying fatalistic attitudes towards CVD screening services among Latinos in context requires that
38


we account for the relative contribution of institutional barriers to health care and fatalism
to the suboptimal use of blood pressure and cholesterol screening services. However, doing this alone is insufficient to achieving a comprehensive understanding of the motivations that underlie people’s fatalistic beliefs including how this impacts their health behavior.
To fully understand Latinos’ health-related beliefs and dispositions toward CVD screening services, research must give voice to the lived experience of those that face prohibitive institutional barriers to health care. Doing so helps to shed light on the internal logic and rationale employed by those who [seemingly fail] to meet our expectations while also contributing to a better understanding of the results of quantitative surveys when employed in a mixed methods design. Viewed in this way, fatalism may reflect a rational adaptation to persistent obstacles to health care experienced by Latinos in the United States (Abraido-Lanza et al. 2007). The theoretical model proposed below, as Figure 1, was
created specifically for this study.
Institutional Barriers
Health Locus of Control
SA#2
â–¼ tieaitn tsenavior
Limited Access to Health Ins. 1 1 1 r SA#1
Perceived Discrimination No USOC vl/ Expressions of Fatalism 1 1 1 r
W Loss of Control Sense of Powerlessness Externa! Source of Control Suboptimal Use of CVD
w Screening Services Blood pressure & Cholesterol
A â–º 1
SA#1
Figure 1. Theoretical Model
Figure 1 illustrates the relationship between institutional barriers to care, CVD fatalism, and the suboptimal use of blood pressure and cholesterol screening services by Latinos proposed by this study. Institutional barriers to care are viewed to both directly,
39


and indirectly affect Latinos use of blood pressure and cholesterol screenings. Indirect effects work through fatalism. In addition to proposing that specific barriers to health care are directly and indirectly responsible for Latinos’ suboptimal use of CVD screening services (Specific Aim #1), the relationships set forth in this model hypothesize that these barriers also influence the very beliefs and attitudes about whether one can control the onset and course of CVD (Specific Aim #2) and in turn, the use of the blood pressure and cholesterol screening services.
40


CHAPTER III
RESEARCH DESIGN AND METHODS A Mixed Methods Study
To address the multiple lines of inquiry proposed by this study, a mixed methods approach was employed. As currently defined by Johnson, Onwuegbuzie and Turner (2005), mixed methods research comprises investigation in which the researcher mixes or combines quantitative and qualitative research techniques, methods, approaches, concepts or language in a single study or set of related studies. There are three areas where mixed methods research is proposed to be superior to single quantitative or qualitative designs: (a) simultaneously addressing a range of both confirmatory and exploratory questions; (b) allowing for better inferences; and (c) providing the opportunity for a variety of divergent views to surface in the analytic process (Teddlie & Tashakkori, 2009).
Mixed methods were especially important for this study, as solely quantitative approaches have failed to adequately explain the lower rates of screening in Latino populations. Used in isolation, quantitative methods have left fundamental questions about how people internalize, reason about, and respond to institutional barriers to health care unanswered. To fully understand how institutional barriers impact the health beliefs and screening behaviors of Latinos, this project employed the complementary strengths of qualitative methods. This approach enabled me to critically explore the impact contextual variables have on people’s motivations and attitudes towards CVD prevention and use of screening resources. Moreover, adding qualitative methods permitted me to both critically examine the meaning of quantitative results obtained while emphasizing the importance of
41


accounting for how institutional barriers may be incorporated into the lived experience, decision-making processes and health behavior of this population.
Goal and Objectives
The goal of this study was to reintroduce social context to the study of fatalism and screening behavior in an effort to more fully understand the relationship between institutional barriers to health care, fatalism towards CVD, and the use of cholesterol and blood pressure screening services by Latinos in Denver, Colorado. More specifically, the research objectives were to:
1. Quantitatively examine the relative contributions of specific institutional barriers to health care (e.g., lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care encounter), and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol) by Latinos in Denver, Colorado.
2. Qualitatively explore how institutional barriers to health care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latinos in Denver, Colorado.
Rationale and Purpose
The rationale for employing mixed methods in the proposed study was twofold: participant enrichment and significance enhancement (Collins, Onwuegbuzie, & Sutton, 2006). Participant enrichment refers to the mixing of quantitative and qualitative techniques for the rationale of optimizing the sample (Collins etal. 2006). In this regard, gathering quantitative demographic data, including responses to survey questions about
42


institutional barriers to health care and fatalism, allowed for the selection of an appropriate sample of Latinos with whom to explore qualitatively and in depth the relationship between institutional barriers to care, health beliefs and explanations for the suboptimal use of CVD screening services.
The second rationale for mixing methods in the proposed study was significance enhancement. Significance enhancement refers to mixing quantitative and qualitative methods for the rationale of enhancing researchers’ interpretations of data (Collins et al., 2006). With this in mind, the proposed study employed a sequential mixed analysis in which the qualitative data analysis phase followed the quantitative data analysis phase (Collins et al., 2006). More specifically, the qualitative data was used to enhance the interpretation of findings from the quantitative analyses but also to critically explore the social context within which CVD health-related beliefs, decisions and screening behavior are formed and expressed. Qualitative methods were an essential part of this mixed methods study. In-depth exploration of how social conditions serve to organize the ways in which people perceive and reason about CVD and the use of screening services moved beyond quantitative only approaches to allow for a more detailed understanding of how institutional barriers become a central part of the very process by which one formulates health-related attitudes and behavior.
Sampling Design
According to Onwuegbuzie and Collins (2007), a distinction between sampling design and sampling scheme can be made. Sampling schemes are specific strategies used to select the units for analysis, while sampling design refers to the framework within which the sampling takes place, including the number and types of sampling schemes that will be
43


used (Onwuegbuzie & Collins, 2007). Given this distinction, both sampling design and sampling scheme for the proposed study are described below.
Sampling Design
The proposed study employed a sequential design using nested samples (Onwuegbuzie & Collins, 2007). Sequential designs involve the selection of units of analysis through the sequential use of sampling strategies such that information from the first sample is used to draw the second sample (Teddlie & Tashakkori, 2009). In this study, the qualitative sample was derived from the quantitative sample. This link between the two samples is referred to as a nested relationship (Onwuegbuzie & Collins, 2007). As shown in Figure 2, in this type of relationship the sample selected for one phase of the study represents a subset of the sample chosen for the other phase of the investigation (Onwuegbuzie & Collins, 2007).
Figure 2. Sequential Sampling Design Using Nested Samples
Sampling Scheme
A multi-stage purposeful sampling strategy was utilized for this study (Onwuegbuzie & Collins, 2007). This approach entails "choosing settings, groups and/or individuals representing a sample in two or more stages in which all stages reflect
44


purposive sampling of participants” (Onwuegbuzie & Collins, 2007, p. 287). More specifically, a purposive, non-random criterion sampling scheme was used for the quantitative portion of the study while a purposive maximum variation-sampling scheme was used for the qualitative component (Onwuegbuzie & Collins, 2007). Criterion sampling involves choosing settings, groups, and/or individuals because they represent one or more criteria (Onwuegbuzie & Collins, 2007). For the present study, inclusion and exclusion criteria guided the first stage of sampling. Inclusion criteria consisted of: (a) self-identified Latino/a; (b) aged forty to seventy-five years2, and (c) self-identified resident of Denver County, Colorado. Exclusion criteria included (a) previous or current participation in a heart disease or diabetes prevention or education program and (b) self-reported heart disease. Participation in a heart disease prevention or education program often entails receiving information about the benefits of blood pressure and cholesterol screening or the actual tests. Those who have been diagnosed with heart disease and self-report as such are also likely to have been provided blood pressure and cholesterol counseling. These particular populations were listed under exclusion criteria in an effort to minimize the information bias they may introduce to the sample.
Setting
This research took place in Denver County, Colorado. Currently, 34 percent of Denver County residents identify as Latino/Hispanic (Calanan etal., 2011). Table 1 details some of the characteristics of Denver County Latinos relevant to this study.
2 Age range is based on screening recommendations put forth by ATP III and JNC 7, including the presence of elevated risk factors (e.g., cholesterol, Type II diabetes, obesity) common among Latino youths that predispose the Latino population to early age onset and premature death from CVD.
45


Table 1. Select Characteristics of Denver County Latino Population
Characteristic Percent
Race/Ethnicity Latino/Hispanic 34%
Told by medical professional have high cholesterol 39%
Report having cholesterol checked in past 5 years 53%
Told by medical professional have high blood pressure 15%
Diagnosed hypertensive and taking medication 65%
Report health as fair/poor 21%
Report no health insurance 40%
Calanan etal., 2011
Phase One: Quantitative Sample
The Phase One sample was drawn using a venue-based, time-space sampling approach (Muhib etal., 2001). This approach recruits respondents from places and at times where they would be reasonably expected to gather. Utilizing previously collected formative data detailing both Denver County venue-based geography and usage patterns (Pineda-Ford, 2011), the proposed study systematically sampled from community-based sites at times known to serve or contain high-density Latino populations. These venues included faith-based organizations, housing complexes, recreation centers, service-based agencies, schools, and neighborhood business centers such as grocery outlets and restaurants. Although the sampling scheme remained purposive, a venue-based, time-space sampling approach allowed me to construct a sample with known properties, make statistical inferences to the larger population that attends chosen venues, and theorize about biases that may limit generalization of findings to the target population (Muhib et al.,
46


2001). A priori power analyses indicated that a sample size of 300 would provide adequate power (0.80) at the alpha (0.05) level for the logistic regression statistic (Hsieh, Block, & Larson, 1998).
Phase Two: Qualitative Sample
A purposive maximum variation-sampling scheme was employed to draw the sample for the Phase Two qualitative portion of the study. Maximum variation sampling involves choosing settings, groups, and/or individuals to maximize the range of perspectives investigated (Onwuegbuzie & Collins, 2007). As shown in Table 2, four groups of interest comprised the qualitative sample in an effort to explore both confirmatory and contradictory cases regarding the impact of institutional barriers to health care on endorsement of fatalistic attitudes and beliefs.
Table 2. Phase Two: Sampling Domains for Qualitative Interviews
Group 1 High IBHC - High Fatalism Group 2 High IBHC - Low Fatalism
Group 3 Low IBHC - High Fatalism Group 4 Low IBHC - Low Fatalism
Based upon their answers to Phase One survey questions, individuals were placed into one of the Table 2 groups. Of those willing to participate, 49 percent were allocated to Group One, 19 percent to Group Two, 16 percent to Group Three and 15 percent to Group Four. A random sample was then drawn from each group pool to comprise the final Phase Two sample. The size of the qualitative sample was determined through evaluation of information redundancy (Lincoln & Guba, 1985). That is, sampling was terminated when no new information was identified from sampled units.
47


Groups one and four provided support for my hypothesis regarding institutional barriers to health care and fatalism. These two groups helped to explain how and why the level of exposure to institutional barriers to health care results in the expression of a disposition consistent with fatalism toward CVD. Groups two and three, however, presented the opportunity to discover and explore explanations that run contrary to the hypotheses proposed. Remaining open to identifying and understanding alternate explanations was an important part of the exploratory process and served to aid in achieving a more holistic and contextually dependent view of the relationships of interest.
High levels of institutional barriers to health care were calculated as confirmation of an individual experiencing two or more barriers represented by the independent variables of interest (e.g., lack of health insurance and no usual source of health care). Low levels of institutional barriers to health care were defined as experience with no or only one of the independent variables of interest. Level of fatalism was determined by calculation of overall score on the validated instrument, ranging from 0 - 20. Determination of high and low scores was based on the distribution of the data. Due to the skewed distribution, median scores and above were considered high with scores below the median considered low. Fatalism was dichotomized for three purposes. First, in order to facilitate modeling as an independent, dependent and mediating variable for hypotheses lb, lc and Id respectively, dichotomization enabled direct comparison between those who agreed or strongly agreed with scale items to those who disagreed or strongly disagreed with scale items. Second, dichotomization contributed to the preservation of model power (0.80) at the alpha (0.05) level for the logistic regression statistic. Third, defining the fatalism
48


variable in this manner allowed for the creation of a clear and consistent sampling frame from which participants were selected for Phase Two qualitative interviews.
Research Design
As conceptualized by Leech and Onwuegbuzie (2006), mixed methods research designs can be represented as a function of three dimensions; level of mixing, time orientation, and emphasis of approaches. Level of mixing refers to whether the study is partially or fully mixed. According to Leech and Onwuegbuzie (2006), the major difference between partially mixed methods and fully mixed methods is that fully mixed methods studies involve mixing quantitative and qualitative techniques within one or more stages of the research process. By comparison, partially mixed methods research typically conducts the quantitative and qualitative phases either concurrently or sequentially in their entirety before mixing occurs at the data interpretation stage.
Time orientation indicates whether the quantitative and qualitative phases of the research occur at the same time or concurrently, or whether they occur one after the other as in sequence (Leech & Onwuegbuzie, 2006). Lastly, emphasis of approach relates to whether the quantitative and qualitative phases of the study hold the same weight with respect to answering the research questions (Leech & Onwuegbuzie, 2006). If the two phases are equally emphasized then this aspect of the design is known as equal status. If one component is given higher priority than the other then the design is noted as dominant status.
Following this three-dimensional typology, the proposed study constituted a partially mixed sequential equal status design (Leech & Onwuegbuzie, 2006). Specifically, Phase One quantitative data was collected and analyzed first, followed by qualitative data
49


in Phase Two. As demonstrated in Figure 3., the proposed study is considered partially mixed because the two datasets were only mixed during the interpretation phase. Moreover, the quantitative and qualitative data were equally emphasized thus the designation of equal status.
Figure 3. Partially Mixed Sequential Equal Status Design Data Collection
Prior to data collection, approval of the study protocol was obtained from The Colorado Multiple Institutional Review Board (COMIRB). Data collection procedures were informed by the two-phase sequential design and comprised a questionnaire strategy for collecting quantitative data with an interview strategy for collecting qualitative data. For the proposed study, Phase One used a survey instrument consisting of demographic questions (including institutional barriers to health care) validated multi-item measures of fatalism, and questions pertaining to use of blood pressure and cholesterol testing services. This survey was administered to Latinos meeting the inclusion criteria. Phase Two consisted of semi-structured in-depth interviews with a subsample of those completing the quantitative survey in Phase One.
Phase One Instruments
A validated scale (Shen, Condit & Wright, 2009; see Appendix A.) was used to collect measures of fatalism. The authors of this scale presented all questions contained in the measure to a multi-cultural Community Advisory Board (CAB) containing Latinos for
50


comments. Board members reviewed and approved items for level of clarity (including readability), cultural appropriateness, and cultural inclusion. Scale validation was conducted with a nationally representative sample with two 200 oversamples from Latino respondents. Confirmatory factor analyses demonstrate that a second order single-factor model provides a good fit to the scale items. Factor loadings ranged from 0.62 to 0.74. In addition, reliability data (alpha = 0.88) and correlation analyses used to investigate the relationship between three external variables (genetic determinism, perceived benefits of lifestyle changes, and intention to engage in health behaviors) and the scale dimensions (predetermination, luck, and pessimism) demonstrate good evidence for the scale’s reliability and construct validity (Shen et al., 2009). Level of fatalism is calculated by summing each of the responses to the scale’s twenty questions into a composite score. Higher scores indicate a stronger tendency to endorse fatalistic beliefs.
Validated measures from the Behavioral Risk Factor Surveillance System Survey (BRFSS) were used to assess respondent demographics and CVD screening and prevention behavior (CDC, 2011b). The BRFSS survey measures of interest, including blood pressure and cholesterol screening behavior, demonstrate reliability (kappa=0.82) (Brownson, Jackson-Thompson, Wilkerson, & Kiani, 1994) While the validity for blood pressure and cholesterol were lower (BP sensitivity=40%; cholesterol sensitivity=44%; BP specificity=87; cholesterol specificity=85 for women and 75 for men), the measures in question remain the standard for assessing self-reported hypertension and hypercholesterolemia and continue to be used by The Centers for Disease Control and Prevention (Bowlin etal., 1993). Dependent variables included: (a) blood pressure screening in the past two years (yes/no) and (b) cholesterol screening in the past two
51


years (yes/no). Independent variables included (a) fatalism (dichotomous variable, high/low); (b) health insurance (yes/no); (c) regular source of health care (yes/no), and (d) perceived discrimination in a health care encounter in the past two years (yes/no). Covariates included age, gender, language preference, education, income, place of birth, length of residence in U.S. and time since last doctor visit. Univariate analyses and previous research in the area of interest informed the operationalization of all covariates. Age was defined as a categorical variable with consideration of the inclusion criteria. Age categories included (40yrs - 49yrs; 50yrs - 59yrs; 60yrs and older). Gender was dichotomized as female and male. Due to the inclusion criteria and population of interest, language was dichotomized as language of preference Spanish or English. Due to the distribution of the data, education was dichotomized as less than a high school diploma or high school diploma/GED or more. Family income was also dichotomized based on univariate analyses and distribution of the data. As a result of 75 percent of the sample demonstrating income of $19,999 per year or below, a decision was made to define family income as $19,999 or below annually and $20,000 or above annually. Given the sociodemographic composition of the Latino population in Colorado and distribution of responses, place of birth was operationalized as Mexico or USA. Due to 77 percent of the sample indicating that they have lived in the U.S. more than 10 years, length of residence in the U.S. was dichotomized as 10 years or less and more than 10 years. Lastly, based on distribution of the data, time since last doctor visit was defined as one year ago or less, more than one year ago and never.
According to Flores (2017), Colorado’s undocumented population has been steadily decreasing since 2014 with the share coming from Mexico decreasing since 2007. Recent
52


data from The Colorado Health Institute demonstrate that 76 percent of all Coloradans without health insurance are documented and legal residents. Due to these data and concern by project research assistants with compromising trust and participation from community members, a decision was made to exclude immigration status as part of this research.
Phase Two Instrument
Phase Two data collection employed an interview guide approach (Patton, 2002). This approach relies on using a list of questions or topics that allow the interviewer to explore, probe, and inquire of the interviewee such that information about the topic will be elucidated (Patton, 2002). The interview guide helped make the interviewing process systematic, efficient, and focused by delimiting in advance the issues or topics that were explored (Patton, 2002). Interview questions focused on attitudes and perceptions towards institutional barriers to health care, fatalism towards CVD and use of CVD-related screening services (See Appendix B.).
Data Collection Procedures
Bilingual members of the target population experienced in community-based research were hired and trained as research assistants. Research assistants collected Phase One data and assisted me with the collection of Phase Two data. Following study protocols, research assistants visited venue locations at pre-specified times to conduct Phase One recruitment. They screened potential respondents for eligibility, set appointments for face-to-face survey delivery at times and locations convenient for respondents and obtained informed consent (see Appendix C.). Upon completion of each Phase One survey, research
53


assistants informed respondents about Phase Two interviews and solicited interest and contact information.
Study respondents providing contact information for participation in Phase Two were allocated to one of the four groups outlined in Table 2 according to their responses to Phase One survey questions. A random sample was then drawn from each of the four groups to comprise the interviewees for Phase Two qualitative inquiry. Data collection proceeded until informational redundancy was achieved (Lincoln & Guba, 1985). All interviews were audio recorded with participant permission and transcribed verbatim. Data Analysis
Phase One data were formatted and entered into a Microsoft Excel spreadsheet.
Data were then imported into SPSS v.21 for analysis (IBM Corp., 2012). During initial analyses, the data were checked for coding errors and missing data by running descriptive statistics. Outliers were identified at this time. Outliers were defined as scores on variables greater than 3.5 standard deviations from the mean in either direction. Descriptive data analyses characterizing participant level of institutional barriers to health care, fatalism, and use of CVD screening and treatment services in Phase One informed the composition of varied groups (Table 2) of interest for qualitative data collection and analysis in Phase Two. Binomial logistic regression was used to address Specific Aim 1: Quantitatively examine the relative contributions of specific institutional barriers to health care (e.g., lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care encounter), and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol) by Latinos in Denver, Colorado. More specifically, mediation pathway analysis was used to evaluate Hypothesis Id: Institutional barriers to
54


care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism. Unlike ordinary least squares regression, logistic regression: a) does not assume linearity of relationship between the raw values of the independent variables and raw values of the dependent variable; b) does not require normally distributed variables; b) does not assume homoscedasticity; c) requires that observations be independent, and d) requires that the independent variables be linearly related to the logit of the dependent (Agresti & Finlay, 2009). The nature and strength of relationships between independent and dependent variables were evaluated using odds ratios and 95 percent confidence intervals (Agresti & Finlay, 2009).
Phase Two qualitative data were formatted in Microsoft Word and entered into ATLAS.ti for management and analysis (ATLAS.ti., 2012). Data were analyzed using constant comparison analysis (Glaser & Strauss, 1967). Constant comparison analysis involves identifying themes in the data and was undertaken deductively and inductively (Leech & Onwuegbuzie, 2007). Deductive coding procedures utilized an existing framework consisting of established a priori codes developed from the research questions and content specific questions contained on the questionnaire (Patton, 2002). In this form of analysis, the researcher reads through the data looking for meaningful segments that represent or relate to the a priori themes (Patton, 2002). Alternately, inductive procedures lead to codes and themes that are grounded in the data and not given a priori (Teddlie & Tashakkori, 2009). In this respect, the themes "emerge” from the data rather than being established beforehand. For the proposed study, inductive coding procedures occurred first and were followed by deductive analyses. This strategy allowed for immersion in the details and specifics of the data to "discover” important patterns, themes, and
55


interrelationships first. Then, using deductive analysis, emphasis was placed on theory driven and established concepts related to the research questions and to confirm exploratory findings (Patton, 2002). Ordering the analysis in this manner served to prioritize examination of the data less encumbered by theoretical assumptions and thus left me in a better position to uncover critical details that helped to explain the formation and maintenance of fatalism among Latinos including why they do or do not use CVD screening services.
The qualitative portion of this study used member-checking procedures to increase the rigor and trustworthiness of the findings (Leech & Onwuegbuzie, 2007). Member checking refers to the researcher showing the study findings or interpretations to the study participants, in this case a select number of participants, in order to assess the accuracy of the work (Leech & Onwuegbuzie, 2007).
Data Validation
Onwuegbuzie and Johnson (2006) reference data validation from three perspectives: quantitative research, qualitative research, and mixed methods research. Given that mixed methods research involves combining complementary strengths and nonoverlapping weaknesses of quantitative and qualitative research, assessing the validity of study findings is critical.
Teddlie and Tashakkori (2009) discuss four types of validity in quantitative research: construct validity, internal validity, external validity, and statistical conclusion validity. The proposed study utilized previously validated instruments to collect quantitative data in Phase One. As a result of previous testing, these instruments carry established measures of validity and reliability, and as such, their psychometric properties
56


are known. Thus, some of the potential threats to construct validity were minimized. Onwuegbuzie and Johnson (2006) identify a multitude of threats to both internal and external validity across the research design/data collection, data analysis, and/or data interpretation stages of the quantitative research process. The classification of these threats, termed the Quantitative Legitimation Model, offers a comprehensive conceptualization of risks for researchers to consider (Onwuegbuzie & Johnson, 2006).
This model served as a guide to ensure proper mitigation of potential threats to internal and external validity during Phase One of the study. The use of reliable measures, a priori power analyses, appropriate statistical procedures and careful assessment of statistical test assumptions were employed to enhance statistical conclusion validity.
The proposed study drew upon Onwuegbuzie and Leech’s (2006) Qualitative Legitimation Model to review the threats to internal and external credibility during the design/data collection, data analysis, and data interpretation steps of Phase Two qualitative work. Specifically, triangulation techniques and member checking were used to enhance credibility of the qualitative findings.
Lastly, legitimation of the mixed methods (Onwuegbuzie & Johnson, 2006) proceeded by attending to each of the following:
1. Sample Integration: Although this study employed purposive sampling techniques in both phases, careful consideration was given to both the relationship between the two samples and the data in an effort to yield quality meta-inferences.
2. Inside-Outside: This study sought to maintain a balanced perspective when collecting, analyzing and interpreting the collective data. Care was taken to
57


ensure that the data accurately represented the "insider’s” view by including members of the group under study in the data review and integration process. In addition, peer review of data interpretation and study conclusions were conducted by research assistants (RAs) and used to justify the "outsider’s” or investigator’s perspective.
3. Weakness Minimization: Quantitative data informed the development of the qualitative interview instrument. The qualitative data were used to enhance the interpretation of the quantitative findings and answer questions pertinent to Specific Aim 2.
4. Sequential: This study utilized a sequential design such that the qualitative phase followed the quantitative phase. If the sequence were reversed, qualitative investigation could have potentially biased answers to the quantitative questions due to in-depth discussion about the study phenomenon.
5. Conversion: This study did not convert data from one form to another.
6. Paradigmatic Mixing: This study drew upon a pragmatic approach towards investigation of the phenomenon of interest and therefore considered the qualitative and quantitative paradigms as complementary rather than competitive.
7. Commensurability: A rejection of Khunian incommensurability and appreciation of both quantitative and qualitative thinking guided this study. There were no anticipated difficulties with switching back and forth between the two paradigms.
58


8. Multiple Validities: All relevant quantitative, qualitative, and mixed methods legitimations were addressed so that strong meta-inferences could be made.
9. Political: This research sought practical results that would contribute to understanding the issue of interest and lead to the identification of real world solutions and thus strived to balance competing objectives where necessary.
Data Integration and Interpretation
As previously stated, data integration occurred at the interpretation phase with equal emphasis placed on both quantitative and qualitative data. The intent of this sequential mixed model design (Teddlie & Tashakkori, 2009) was to use the qualitative data to both enhance the interpretation of significant quantitative findings and to develop meta-inferences by combining inferences drawn from the quantitative data with those drawn from the qualitative data. Inferences made from the quantitative data included detailed explanation of both statistical and practical significance (Leech & Onwuegbuzie, 2010). Inferences developed from qualitative findings prioritized the voice and perspectives of study participants and were presented together with those obtained from the quantitative portion of the study (Leech & Onwuegbuzie, 2010). As a result, critical insight was gained and meaningful conclusions drawn regarding the relationship between institutional barriers to health care, fatalism towards CVD, and the suboptimal use of blood pressure and cholesterol screening services by Latinos in Denver, Colorado.
59


CHAPTER IV
QUANTITATIVE ANALYSES AND RESULTS
Despite the recognition that institutional barriers to health care in the form of lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care system negatively impact Latinos’ use of screening services, research continues to emphasize individual level fatalistic attitudes and beliefs in the explanation of Latinos’ underuse of various screening services. In Chapter 2,1 discuss the importance of reintroducing contextual variables to the study of fatalism and health screening behavior among Latinos. Rather than viewing fatalism as simply a cognitive trait of individuals, I argue that fatalism is partly a byproduct of social processes that serve to hinder the health-related agency of Latino populations. Understanding the context in which screening behavior occurs requires accounting for the relative contribution of institutional barriers to health care and fatalism to the suboptimal use of blood pressure and cholesterol screening services. In this chapter, I focus specifically on the results of quantitative analyses used to test each of four hypotheses addressing Specific Aim 1.
Review of Methods
I measured fatalism using a validated instrument developed by Shen, Condit &
Wright, (2009). Validated measures from the Behavioral Risk Factor Surveillance System Survey (BRFSS) were used to assess respondent demographics and CVD screening behavior. Statistical methods used in this chapter include descriptive statistics, binomial logistic regression and mediation pathway analysis using the PROCESS macro in SPSS v. 25 (Hayes, 2018). Receipt of blood pressure screening in the past two years and receipt of cholesterol screening in the past 2 years were the dependent variables of interest for
60


hypotheses la, lb and Id. Fatalism was the dependent variable of interest for hypothesis lc. Perceived discrimination in the health care encounter, originally included as one of three specific institutional barriers to health care in Specific Aim 1, is not included in the logistic regression model results presented. Due to insufficient sample size a decision was made to drop this variable from the analyses in order to preserve model power (0.80) at the alpha (0.05) level for the logistic regression statistic. Only twenty-one individuals (5%) in the analysis cohort (n = 386) reported exposure to perceived discrimination.
Sample Characteristics
Table 3 describes the sample socio-demographic characteristics. Participants consisted primarily of females (81%) in their forties (65%) with less than a high school diploma (66%). A majority of participants have resided in the United States Table 3. Participant Characteristics
Demographic Characteristics Total (N=386) Median (range) or %
Age (years) 47.0 yrs (38.0 yrs)
Female 81.1%
Language preference (Spanish) 94.8%
Less than high school diploma 65.8%
Family income (less than $19,999 yearly) 74.4%
Length of residence in US (more than 10 yrs) 76.9%
Time since last MD visit (one year or less) 75.5%
Have health insurance 26.7%
Have usual source of health care 70.7%
61


Table 3 cont’d
BP Screening in past 2 yrs 55.7%
Cholesterol Screening in past 2 yrs 43.8%
Fatalism 2.3 points (4.0 points)
for more than 10 years, prefer to speak Spanish (95%) and have been to see a health care provider in the past year. Among sample participants, only 44 percent indicate they have had a cholesterol test in the past two years. While most participants (73%) report having no health insurance, 70 percent say they have a usual source of health care. With regard to fatalism, careful examination of the results from univariate analyses was conducted to ensure accurate assessment of measures of central tendency. On a five point Likert scale using strongly disagree (1) to strongly agree (5), indicating level of endorsement of statements indicating fatalistic beliefs, mean fatalism score was 2.37+ 0.97. Median score was 2.3 with a range of 4.0.
In the following section I report results from logistic regression analyses employed to test each of the four hypotheses presented under Specific Aim 1.1 first report results from Hypothesis la: Institutional barriers to health care resources will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services.
Institutional Barriers to Health Care and Cholesterol Screening
A binary logistic regression was performed to ascertain the relationship between health insurance status and having a usual source of health care on the likelihood of participants obtaining cholesterol screening. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic
62


regression model was significant, x2 = 97.14, p < .005. The model explained 31% of the variance (Nagelkerke R2) in screening for cholesterol and correctly classified 71.2% of cases. As shown in Table 4, those with a usual source of health care had greater than three times the odds of obtaining cholesterol screening compared to those without a usual source of health care (OR 3.35, 95% Cl: 1.77 - 6.32). Unexpectedly, having health insurance was not related to obtaining screening for cholesterol when compared to not having health insurance after holding all other variables constant. Those of advanced age had an 89% (50-59 yrs; OR 1.89, 95% Cl: 1.02 - 3.50) and more than two and a half times (60 yrs and over; OR 2.74, 95% Cl: 1.37 - 5.47) greater odds of obtaining a cholesterol test compared to younger adults (40-49 yrs). Compared to males, females had 98% greater odds of obtaining a cholesterol test (OR 1.98, Cl: 1.03 - 3.79). Also, those with a family income of $19,999 or less had greater than three times the odds of obtaining a cholesterol test compared to those with a family income of $20,000 or greater (OR 3.32, Cl: 1.80 - 6.62). As expected, increasing length of time since last visit to a doctor resulted in a significantly lower likelihood of obtaining a cholesterol test in the past two years. The results for usual source of health care and use of cholesterol screening demonstrate support for hypothesis la: Institutional barriers to health care resources will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. However, the results for health insurance do not. Discussion of this anomaly occurs under the section detailing qualitative analyses and results.
Table 4. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening when Participants Have Health Insurance and Have a Usual Source of Healthcare (USOC).
63


Table 4 cont’d
Participant Characteristics OR (95% Cl)
Have health insurance 1.59 (0.90-2.78)
Have USOC 3.35*** (1.77- 3.62)
Age (40 - 49 yrs) Reference group
Age (50 - 59) 1.89* (1.02- 3.50)
Age (60 and over) 2.74** (1.37-5.47)
Time since last doctor visit (one year or less) Reference group
Time since last doctor visit (more than one year) 0.48* (0.25-0.91)
Time since last doctor visit (never) 0.11* (0.01-0.91)
Gender (female) 1.98* (1.03- 3.79)
Length of time residing in US (10 years or less) 1.45 (0.79-2.63)
Education (less than a high school diploma) 1.43 (0.86-2.37)
Family income ($19,999 or less) 3.32*** (1.80-6.12)
* p<0.05 ** p<0.01 ***p<0.001
Institutional Barriers to Health Care and Blood Pressure Screening
A binary logistic regression was performed to ascertain the effects of health insurance status and having a usual source of health care on the likelihood of participants obtaining blood pressure screening. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, y} = 115.75, p < .05. The model explained 36.0% of the variance (Nagelkerke R2) in screening for blood pressure and correctly classified 74% of cases. As shown in Table 5 below, those with health insurance had significantly greater
64


odds of obtaining screening for blood pressure (OR 2.30, 95% Cl: 1.23 - 4.30) when compared to those without health insurance after controlling for all other variables in the model. Those with a usual source of health care were five times more likely to obtain screening for blood pressure (OR 4.97, 95% Cl: 2.71 - 9.10) when compared to those without a usual source of health care after holding all other variables constant. Those with less than a high school diploma had 78% greater odds of obtaining a blood pressure test compared to those with a high school diploma or more (OR 1.78, Cl: 1.05 - 3.01). Also, participants with a family income of $19,999 or less had greater than four times the odds of obtaining a blood pressure test compared to those with a family income of $20,000 or greater (OR 4.73, Cl: 2.32 - 9.63). These results offer support for hypothesis la by demonstrating that a lack of institutional barriers to health care relate to significantly greater odds of using blood pressure screening services.
Table 5. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screening when Participants Have Health Insurance and Have a Usual Source of Healthcare (USOC).
Participant Characteristics OR (95% Cl)
Have health insurance 2.29** (1.23-4.30)
Have USOC 4 gy*** (2.72-9.10)
Age (40 - 49 yrs) Reference group
Age (50 - 59) 1.93 (0.99-3.76)
Age (60 and over) 1.54 (0.76-3.11)
Time since last doctor visit (one year or less) Reference group
Time since last doctor visit (more than one year) 0.89 (0.47- 1.70)
65


Table 5 cont’d
Time since last doctor visit (never) 0.62 (0.18-2.10)
Gender (female) 1.29 (0.67-2.51)
Length of time residing in US (10 years or less) 1.54 (0.85-2.77)
Education (less than a high school diploma) 1.78* (1.05-3.01)
Family income ($19,999 or less) 4 y3*** (2.32-9.63)
* p<0.05 ** p<0.01 ***p<0.001
In the next section I report results from logistic regression analyses employed to test Hypothesis lb: Fatalism will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. I first report results on the relationship between fatalism and blood pressure screening followed by results on the relationship between fatalism and use of cholesterol screening.
Fatalism and Blood Pressure Screening A binary logistic regression was performed to ascertain the relationship between fatalism and the likelihood of participants obtaining blood pressure screening. Model covariates included health insurance, usual source of health care, age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, y} = 115.08, p < .05. The model explained 36% of the variance (Nagelkerke R2] in screening for blood pressure and correctly classified 74% of cases. However, as demonstrated in Table 6, there is no significant difference in the likelihood of obtaining screening for blood pressure between those demonstrating a low level of fatalism and those who demonstrate a high level of fatalism after controlling for all other variables. However, as expected, those with health insurance had greater than twice
66


the odds of obtaining a blood pressure screening compared to those without health
insurance (OR 2.26, Cl: 1.21 - 4.24). When compared to those without a usual source of
health care, those who have a usual source of health care have almost five times greater
odds of obtaining a blood pressure test (OR 4.96, Cl: 2.71 - 9.01). Participants with less
than a high school diploma had 78% greater odds of obtaining a blood pressure test
compared to those with a high school diploma or more (OR 1.78, Cl: 1.05 - 3.01) Lastly,
those with a family income of $19,999 or less had an almost five times greater odds of
obtaining a blood pressure test when compared to those with a family income of $20,000
or more (OR 4.76, Cl: 2.33 - 9.70). These results do not offer support for hypothesis lb.
Table 6. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screening based upon Level of Fatalism.
Participant Characteristics OR (95% Cl)
Have health insurance 2.26* (1.21-4.24)
Have USOC 4.96*** (2.71-9.01)
Age (40 - 49 yrs) Reference group
Age (50 - 59) 1.93 (0.98-3.77)
Age (60 and over) 1.54 (0.76-3.11)
Time since last doctor visit (one year or less) Reference group
Time since last doctor visit (more than one year) 0.89 (0.47- 1.70)
Time since last doctor visit (never) 0.62 (0.18-2.10)
Gender (female) 1.28 (0.66-2.50)
Length of time residing in US (10 years or less) 1.54 (0.85-2.77)
Education (less than a high school diploma) 1.78* (1.05-3.01)
67


Table 6 cont’d
Family income ($19,999 or less) 4.76*** (2.33-9.70)
Fatalism (high) 0.93 (0.56- 1.54)
* p<0.05 ** p<0.01 ***p<0.001
Fatalism and Cholesterol Screening
A binary logistic regression was performed to ascertain the effects of fatalism status on the likelihood of participants obtaining cholesterol screening. Model covariates included health insurance, usual source of health care, age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, %2 = 101.11, p < .05. The model explained 32% of the variance (Nagelkerke R2] in screening for cholesterol and correctly classified 71% of cases. As shown in Table 7, a high level of fatalism is significantly related to a 42% lower odds of obtaining cholesterol screening compared to low level of fatalism after considering all other control variables (OR .58, 95% Cl: 0.35 - 0.95). Having a usual source of health care results in more than three times the odds of obtaining a cholesterol test when compared to not having a usual source of health care (OR 3.43, Cl: 1.81 - 6.50). Those of advanced age had twice (50 - 59 yrs; OR 2.03, Cl: 1.07 - 3.82) and three times (60 yrs and over; OR 2.99, Cl: 1.48 - 6.04) greater odds of obtaining a cholesterol test compared to younger adults (40 - 49 yrs).
When compared to those whose last doctor visit was within one year or less or more than one year ago, those who have never visited a doctor were 90% less likely to have obtained a cholesterol test (OR 0.09, Cl: 0.01 - 0.76). Females were 85% more likely to have obtained a cholesterol test compared to males (OR 1.85, Cl: 0.95 - 3.56). Those with a family income of $19,999 or less had more than three times the odds of obtaining a
68


cholesterol test compared to those with a family income of $20,000 or more (OR 3.41, Cl: 1.83 - 6.35). In contrast to the results obtained with blood pressure, these results offer support for hypothesis lb.
Table 7. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening based upon Level of Fatalism.
Participant Characteristics OR (95% Cl)
Have health insurance 1.51 (0.85-2.69)
Have USOC 3.43*** (1.81-6.50)
Age (40 - 49 yrs) Reference group
Age (50 - 59) 2.03* (1.07- 3.82)
Age (60 and over) 2.99* (1.48-6.04)
Time since last doctor visit (one year or less) Reference group
Time since last doctor visit (more than one year) 0.49 (0.25-0.92)
Time since last doctor visit (never) 0.09* (0.01-0.76)
Gender (female) 1.85* (0.95-3.56)
Length of time residing in US (10 years or less) 1.46 (0.80-2.68)
Education (less than a high school diploma) 1.39 (0.83-2.31)
Family income ($19,999 or less) 3 41*** (1.83-6.35)
Fatalism (high) 0.58* (0.36-0.94)
* p<0.05 ** p<0.01 ***p<0.001
In the following section I report results from logistic regression analyses employed to test Hypothesis lc: Institutional barriers to health care resources will demonstrate a positive relationship with fatalism.
69


A binary logistic regression was performed to ascertain the effects of health insurance status and usual source of health care on level of fatalism. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, y} = 21.21, p < .05. The model explained 7.5% of the variance (Nagelkerke R2) in screening for cholesterol and correctly classified 61% of cases. According to the model, having health insurance is negatively related to high level of fatalism though this relationship is not statistically significant after controlling for all other variables in the model (Table 8). Likewise, having a usual source of health care is not statistically related to high level of fatalism after controlling for all other variables included in the model. Those aged fifty to fifty nine years had 78% greater odds of demonstrating a high level of fatalism compared to those aged forty to forty nine years (OR 1.78, Cl: 1.02 - 3.12). Also, when compared to males, females were 53% less likely to display a high level of fatalism (OR 0.47, Cl: 0.27 - 0.83). These results do not offer support for hypothesis lc.
Table 8. Binary Logistic Regression Results for the Odds of Demonstrating High Level of Fatalism Based Upon Having Health Insurance and Having a Usual Source of Health Care.
Participant Characteristics OR (95% Cl)
Have health insurance 0.85 (0.50 - 1.44)
Have USOC 1.24 (0.73-2.11)
Age (40 - 49 yrs) Reference group
Age (50 - 59) 1.78* (1.02-3.12)
Age (60 and over) 1.63 (0.88-3.03)
70


Table 8 cont’d
Time since last doctor visit (one year or less) Reference group
Time since last doctor visit (more than one year) 1.22 (0.69-2.15)
Time since last doctor visit (never) 0.43 (0.15-1.22)
Gender (female) 0.47* (0.27-0.83)
Length of time residing in US (10 years or less) 1.05 (0.62- 1.77)
Education (less than a high school diploma) 0.73 (0.46-1.15)
Family income ($19,999 or less) 0.87 (0.51-1.50)
* p<0.05 ** p<0.01 ***p<0.001
In this final section, I report results from mediation pathway analyses using the PROCESS macro in SPSS v. 25 (Hayes 2018). Mediation pathway analysis was used to test Hypothesis Id: Institutional barriers to health care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism. To test whether fatalism mediates the relationship between institutional barriers to health care and use of blood pressure and cholesterol screening services, four separate tests were performed. The first test examined the relationship between health insurance and use of blood pressure screening with fatalism included as mediator. The second test looked at the relationship between health insurance and use of cholesterol screening with fatalism included as mediator. The third test examined the relationship between usual source of health care and use of blood pressure screening with fatalism included as mediator. The last procedure examined the relationship between usual source of health care and use of cholesterol screening with fatalism again, included as a mediating variable.
71


Results for each of the four tests revealed that fatalism did not significantly mediate the relationship between the main independent and dependent variables of interest. This was not completely unexpected as the relationships between usual source of health care and use of both blood pressure and cholesterol screening services were strong with statistical significance achieved in both logistic regression analyses under hypothesis la. While health insurance was not statistically related to use of cholesterol screening, results demonstrate that it was strongly related to use of blood pressure screening (p< 0.01). Moreover, despite the statistically significant relationship between fatalism and use of cholesterol screening services demonstrated in hypothesis lb, fatalism did not show a statistical relationship to use of blood pressure screening as an independent variable to either health insurance or usual source of health care when modeled as a dependent variable. Given the relatively strong relationships between both health insurance and usual source of health care and use of both blood pressure and cholesterol screening services, and the overall lack of a statistical relationship between fatalism when modeled as an independent and dependent variable, a lack of fatalism acting as a mediator between health insurance or usual source of health care and use of CVD screening services makes sense.
72


CHAPTER V
QUALITATIVE ANALYSES AND RESULTS
Latinos in the U.S. continue to face an inequitable array of agency reducing barriers to health care including lack of access to health insurance and a usual source of health care. Reintroducing context to the study of fatalism and screening behavior is essential if we are to more fully understand how individuals rationalize the use of prevention resources when faced with persistent structural obstacles to health care services. This project included qualitative methods in an effort to gain insight to how this population internalizes, reasons about, and responds to institutional barriers to health care in deciding to access screening services for CVD. In addition to shedding light on the meaning of the quantitative results obtained in Phase One, the qualitative methods used in this project serve to explain how contextual variables impact people’s motivations and attitudes towards CVD prevention and use of screening resources. In this chapter, I focus specifically on the results of qualitative analyses used to address Specific Aim 2: Qualitatively explore how institutional barriers to health care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latinos in Denver, Colorado.
Review of Methods
Interview participants consisted of a subset of all participants who completed quantitative surveys in Phase One. After the completion of each survey, participants were asked whether they would like to participate in a one-on-one interview to further talk about their experiences with health insurance and having a regular place to go see a health care provider, including how these relate to their ideas of heart disease and using blood
73


pressure and cholesterol testing services. Based upon their answers to Phase One survey questions regarding health insurance, usual source of health care and the validated scale used to measure fatalism, I placed participants into one of four groups representing the sampling domains for Phase Two qualitative interviews. I defined high levels of institutional barriers to health care as confirmation of an individual experiencing both barriers represented by the independent variables of interest (e.g. lack of health insurance and no usual source of health care). I defined low levels of institutional barriers to health care as experience with none or only one of these independent variables of interest. Due to the five point Likert-type questions comprising the validated scale, I determined level of fatalism by calculation of an overall score and the performance of a median split. Median score was 2.3 with a range of 4.0.1 defined those scoring below this number as low fatalism and those scoring above this number as high.
After population of all four of the Phase Two sampling domains was complete, I then drew a series of random samples consisting of four participants at a time from each of the four population groups (Figure 4). Research assistants made up to three phone calls to randomly selected participants until they were able to confirm four interviews from each group that would comprise the final interview sample and analysis cohort. All interviews were recorded with participant permission and transcribed verbatim in Spanish with English translation. I conducted all interviews with the aid of one research assistant who performed translation for interviews conducted in Spanish. Transcripts were analyzed and coded in Atlas Ti v.8.3.0.
74


Phase One Quantitative Surveys Completed N = 386 Did Not Want to Participate n= 276
1 '
Group 1
T IBHC - T Fatalism
Group 2 Group 3
f IBHC - | Fatalism j IBHC - f Fatalism
Group 4
J. IBHC - j, Fatalism
Willing to Participate n = 110
Random Sampling n= 48
Participants n = 16
Figure 4. Phase Two Qualitative Interviews Sample Selection
I used constant comparison analysis (Glaser & Strauss, 1967) to analyze all interview data. For the first pass through the data, I used inductive coding procedures. This involved carefully reading and re-reading through the transcripts with an open mind and creating codes that emerged naturally from the data. At this stage of the analysis, I referenced my research question and Specific Aim 2 as guides but did not look to Specific Aim 1 or my hypotheses to guide my work. After inductive analyses were complete, I reread the transcripts and analyzed the data again using deductive coding procedures.
During this stage of the analysis, I utilized an existing framework consisting of established a priori themes developed from Specific Aim 1 and coded the data by referencing these
75


themes and the relationships of interest in Specific Aim 1. During this stage of analysis, I read through the data looking for meaningful segments that represented or related to the a priori themes and coded the data accordingly. Ordering the analysis in this manner allowed me to prioritize examination of the data less encumbered by theoretical assumptions and put me in a better position to uncover critical details that help to explain the formation and maintenance of fatalism among Latinos including why they do or do not use CVD screening services.
The Contextual Nature of CVD Fatalism
Qualitative analyses reveal that fatalism does not exist as a static cognitive disposition confined to the individual psyche, but rather arises as a result of dynamic social processes that serve to limit one’s health-producing agency. Moreover, the manifestation of this disposition does not occur as one form or type but rather varies in its expression and representation. Fatalism towards CVD and the use of screening resources is highly context dependent and often occurs at two intersections. The first is at the junction of institutional barriers to health care and the perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiovascular health. Fatalism at this juncture comprises an expression of social and economic distress and lack of control over CVD due to the unmistakable recognition that heart disease represents a serious health condition for which medical intervention is often costly and ongoing. The second form of fatalism occurs at the convergence of institutional barriers to health care, the presence of disease and the logics of religion that help to define the amount of control one believes they have over CVD. At this intersection, fatalism embodies the expression of a process whereby one actively defines the boundaries
76


of control over disease through belief systems that cut across religious values related to individual responsibility for health. The intersection of these complex and varied social processes gives life to alternating manifestations of fatalism that belie the notion of fatalism as a static and singular concept. Rather than viewing fatalism as representing a consistent and fixed set of beliefs, I argue that this disposition varies in the form of its expression by way of the social circumstances that contribute to its development.
CVD Fatalism Intersection 1: IBHC, Material Deficits and the Perception of Lack of Control
Fatalism towards CVD and the use of screening services is inextricably linked to institutional barriers to health care such as lack of health insurance and no usual source of health care. These barriers intersect with and oftentimes, exacerbate the perception of lack of control over heart disease due to the recognition that limitations in the availability of personal financial resources exist. For those lacking health insurance, screening for and potentially having to treat disease produces a state of chronic stress and depression related to being unable to afford essential medical services and procedures. These conditions give rise to a fatalistic outlook as one makes a realistic appraisal of the lack of options with which risks to heart health can be addressed.
Fatalistic discourse is not evidence for the presence of irrational thinking that needs to be corrected. Rather, fatalism exists as a rational expression of social and economic-related distress and loss of hope due to the understanding that one has limited options with which to address potentially overwhelming financial liabilities for the receipt of health care. For those forced to endure the hardships of these socially organized circumstances, institutional barriers do not simply exist as external to one’s perception, but rather become
77


subjective realities that are incorporated into the everyday lived experience, decisionmaking and health behavior of those with restricted access to social and material resources. More specifically, structurally organized constraints on human agency, such as lack of health insurance and no usual source of health care, are embodied in the very outlook of those that experience them. It is in this sense that fatalism becomes the product of social processes that etch their negative effects in the very attitudes and dispositions of those impacted. In the following quote, we see just how detrimental this loss of control can be to one’s outlook. For this participant, not having health insurance combined with a lack of financial resources with which to pay medical bills can result in severe depression and may lead to fatalistic suicide in an effort to escape overwhelming financial obligations.
And I said (without health insurance), "Oh wow, how we are going to pay the big bills?" Well one patient might die (right away) and the other might die little by little without money to pay (the medical bills) that you owe you know what I mean? How are you going to pay? Make arrangements? That's what you do. You make arrangements and little by little you get depressed and frustrated and then you go down and down through depression. Some of these people kill themselves because of this (depression). I know some people that killed (themselves) because of this kind of depression. For not having the money... I mean not even having money for food, not having enough money to pay the bills.
Health insurance provides people with options to prevent and treat disease and helps with the financial obligations that come with receiving health care. When a lack of this critical resource intersects with a perceived loss of control over health due to the potential of large medical bills, one’s psyche may be overwhelmed such that a loss of the very will to live results. In these tragic circumstances, social processes render hope and vitality for the future worthless. Fatalism towards CVD in this instance represents the extreme vulnerability and lack of control over disease that so many in the Latino community feel as a result of limitations imposed on access to social and material
78


resources. This type of fatalism is rooted in the socio-political and economic
marginalization of Latinos whereby biomedical resources are reserved only for those who
can afford them. Access to health care resources remains a thoroughly social process that
prioritizes high cost technology and profit over access to affordable health care as a
fundamental right for all. As shared by the next participant, living without health insurance
and a regular place to receive health care results in a complete lack of acceptable options
with which to tackle medical bills. Once again, a preference for the end of life reveals that a
total loss of control over options breeds fatalistic statements. This disposition is not an
inherent part of one’s cognitive characteristics, but rather reflects a realistic appraisal that
acceptable courses of action are severely limited.
It affects me a lot right now. Because the whole thing is (without health insurance)... Yeah, there’s nothing you can do. People say, "I prefer to die than have to pay two hundred thousand dollars for a medical bill" you know what I mean. Because at that point you know, it’s cheaper to die than have to face that many bills. It’s the same thing for all the Hispanics that don’t have insurance. But again it’s the same, we’re in the wrong circle, there is no health insurance for us.
When viewed in context, fatalism towards heart disease becomes a rational response to severely limited options with which to address chronic health conditions that threaten both physical and financial autonomy. The prospects of living life with the burden of crushing medical debt for health care is unbearable for some and leads to a demoralized expression of fatalism.
Emotional distress associated with a lack of options with which to access and afford health care is embodied in ways that affect the subjective appraisal of one’s health. Central to these circumstances is the perception of a loss of control over the ability to act on one’s own behalf. In the following quote, a participant describes how having money may not even lead to treatment because without appropriate coverage, access to health care
79


professionals remains out of reach. This lack of access leads to feelings of extreme pressure
and stress. Ultimately, restricted access to essential health services results in the
internalization of deprivation and to a fatalistic loss of control.
I feel pressure because (1 don't have health insurance). I say to myself "What should I do if they don't help me? I don't have a place to go to get checked or treated. " l feel my health gets worse when I think about that I don't have (health insurance). I tell you again even if I take money (to the clinic) they would not treat me because I don't have an identification that has my (insurance information). I feel way more excited and stressed about it and I just go into my head just wondering what should I do? What could I do? Because the sad part of the system is that even if I had the money they would not see me so it's very stressful and I think I just get sicker and sicker and sicker. I feel that you lose all control (without health insurance).
Health insurance is not merely a possession that people use when necessary. Nor is it an abstract agreement between an individual or family and an insurance provider. Health insurance imbues the holder with a sense of security and provides a means of taking action to ensure that one remains healthy or can treat medical problems as they arise. Sociopolitical processes that render vulnerable members of the Latino community unable to procure this critical resource serve to undermine the sense of control and belief that one can effectively prevent and combat heart disease through access to screening and treatment services. For socio-demographically disadvantaged individuals, a lack of health insurance becomes part of the everyday reality of dealing with restrictions imposed on how, when and where one can access affordable health care resources. As expressed by the following participant, frustration, anger and depression over these circumstances can lead to a sense of subjugation and the loss of hope.
Well that's what I'm telling you, you get mad (without health insurance). You get upset with the kids and you don't do anything. Then, well I let myself fall (get depressed) because I tell you I started thinking, "I don't have health insurance, what am I going to do if I can't get checked?" Then I went and I lay on the couch. I got (very) depressed... Yes, I was without hope.
80


Expressions of fatalism and the loss of hope also occur in the context of lacking a medical home or usual source of health care. Along with limited access to health insurance, not having a usual source of health care is an important institutional barrier to health care for Latinos. Having a usual source of health care has been associated with greater use of health services and the earlier receipt of preventive care (Ettner, 1999). This may be explained in part because patients with an ongoing, long-term relationship with their physician may have heightened awareness of certain health behaviors along with the opportunity to discuss these behaviors, gain access to health-related information, and treat problems (Ettner, 1999).
Not having a usual source of health care erodes one’s ability to establish a consistent
source from which critical health care information and resources can be obtained.
Interview participants remain acutely aware of the importance of having access to a usual
source of health care. A medical home is not simply a place where one receives regular
health screenings and services, for it also serves as a setting in which one develops
relationships of trust with members of the health care community and the confidence that
health can be protected, fostered and potentially restored. In the following excerpt, a
participant relates what not having a usual source of health can do to one’s attitude. For
this participant, not having a clinic at one’s disposal means not being able to have her
chronic condition checked and treated by a competent professional.
Well I think my health will deteriorate because well by not having a clinic to go to... well I can't get checked. It's where one gives up and falls down (gives up without hope). I tell you I had three years of not going to the clinic and the day I went well they told me I had (diabetes). Then I said, "Well / can't afford the services. "And one feels like down, like you say to yourself "What am I going to do?" One starts thinking, Well I can’t do anything."
81


Fatalistic attitudes arise at the intersection of institutional barriers to health care
and the perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiovascular health. Fatalism is not an inherently stable personality trait or culturally derived attitude that Latinos possess but rather arises out of dynamic interaction between socially organized limits on personal health autonomy and the rational appraisal that one’s personal options to address health issues are severely restricted. In these contexts, expressions of fatalism represent rational appraisals of specific circumstances at specific times. Social, economic and institutional forces intersect with health beliefs related to the control of CVD and result in fatalistic interpretations of personal health circumstances. These assessments are coherent expressions of social and economic distress.
When viewed in context, fatalism towards CVD becomes a logical response to a socially and politically organized system of health care that denies resources to those who are unable to afford them or access them because of their legal status. Screening and treatment of heart disease is not an enigma for the Latino community. Rather, a lack of equitable access to and the ability to pay for screenings and the potential treatment of CVD places Latinos in a vulnerable position not of their own choosing. The recognition that a diagnosis of heart disease can result in lifelong and overwhelming financial burden breeds fatalistic attitudes. It is in these dispositions that institutional barriers to health care are embodied and serve to affect the very outlook towards heart disease and its prevention of those who experience them.
82


CVD Fatalism Intersection 2: IBHC, the Logics of Religion and the Organization of Control
Biomedicine and the public health community’s tendency to treat fatalism as a fixed personal trait that threatens prevention science and its aim to reach Latinos fails to account for how fatalism varies in its expression as a result of the social context within which it forms (Espinosa de los Monteros & Gallo, 2013). Fatalism comprises dynamic representations that occur at specific times and in specific places. For those faced with challenges to their health, whether and when to cultivate such a disposition is contingent upon the perception of control individuals feel they have over CVD and the health care resources they have at their disposal. In-depth analyses of interview data reveals how living without health insurance or a usual source of health care leads to the internalization of a sense of helplessness and loss of control. When an individual experiences these barriers and is also confronted with a direct threat to their health such as diabetes or hypertension, the ethics of religion may be employed in an effort to organize and define a meaningful appraisal of the control one has and the courses of action available. Once again, rather than exhibiting fixed characteristics, fatalism towards CVD is in this instance, expressed at the unique intersection of institutional barriers to health care, the direct presence of threats to one’s health and the active cultivation of a disposition (Hamdy 2009) in which shared responsibility with God helps to define the measure of control one has over their condition. Appealing to divine will is not evidence of passivity, but rather represents an active organization of shared responsibility when recognition that restricted access to health care resources and the peril of disease co-exist. The manifestation of fatalism at this particular intersection is distinct from that which occurs at intersection one.
83


Rather than representing the hopelessness that ensues from the culmination of socially and materially induced distress, here it arises out of a working process whereby boundaries between control over the presence and responsibility for disease are negotiated through the dictates of religious interpretation and belief.
Interview participants who experience institutional barriers to health care acutely recognize and understand the value of good health. However, they also understand that limits imposed on access to medical resources and challenges to one’s health impose realistic difficulties for which there are no easy solutions. Individuals that have been diagnosed with CVD related conditions or disease must confront life in which chronic illness threatens both short and long term wellbeing. Not having health insurance or a usual source of health care can severely exacerbate health-related stress and confusion. Cultivating a disposition consistent with the logics of religion helps create a meaningful appraisal of one’s responsibility and options when faced with illness and restricted access to health care. While verbal expressions of fatalism often occur within this context, they are not isolated. Oftentimes, fatalistic statements occur alongside others that speak to the value of taking personal responsibility for one’s health. These apparent contradictions serve as an active form of negotiation whereby individuals wrestle with the options available to them at the intersection of institutional barriers to health care, the presence of a health disorder and the measure of control one has over health conditions. The appeal to divine will and authority is not passive or subjugating, but rather represents the simultaneous recognition that while one must act in partnership with God to protect health, a healthful life free of disease is not guaranteed.
84


In the following passage, a longtime diabetic who lacks health insurance speaks to
her acceptance of God’s plan and interprets her current health condition through the
fatalistic appraisal that diabetes was a predetermined outcome for her. Yet at this
intersection of institutional barriers, the presence of disease and the logics of religion,
fatalism occurs alongside statements that support taking action on one’s own behalf.
Of course I realize that I have to take better care of myself. I also recognize that God has a plan for me and if that's what I'm supposed to be dealing with, like diabetes, I accept it. But I also know I have to take care of myself as I'm also conscious that God has a plan for me and I'm supposed to be dealing with this disease and that's something that's supposed to happen. But at the end of the day I need to take care of myself too.
For this participant, the appeal to God creates meaning and helps to organize and define
how she must move forward under difficult and potentially life threatening circumstances.
Dealing with disease includes both acceptance of its presence and the responsibility to take
care of oneself in the context of barriers to health care.
In the following excerpt, another diabetic without access to health care resources
demonstrates that submission to divine authority is not a passive form of fatalism that
negates human agency but rather speaks to the active cultivation of an appraisal in which
the presence of disease is given meaning and the course of one’s actions are laid forth
according to the religious precepts that inform one’s belief system.
Well when they told me about my disease I said, "What am I going to do?" and so many things came to my head but I said, "What God said, (do) what God said."So I have to control myself and I have to accept what comes. Accept what comes but now take care of myself and cope with myself... well I have to put an effort in for myself.
Blindly accepting fatalistic statements indicating that those who believe in God’s will as proof of complete submission to external forces obscures how these very statements give meaning to the intersection of institutional barriers to health care, the presence of
85


health challenges, and the search for control. In this context, reference to religious subtexts
becomes part of the individual's active definition of personal responsibility and the
expectations that come with sharing a relationship with God. It is the embodiment of this
relationship that guides the individual when answers may not be forthcoming. In the quote
that follows, a participant who has long struggled with hypertension and lack of access to
health care discusses how religion informs her perspective on CVD.
Well I think if you eat healthy, you will help your health a lot. But the first thing is I am catholic. First of all I think that the majority (ofyour health) is in God's hands. Next, well we can mobilize and see what we can do. If I can do it (prevent and treat heart disease), well I will do it. Yes, I work like a team with God. But, if someone is destined to get a disease it does not matter what they do because either way they will get sick. Well because I will repeat again, the first thing is God's will. And if is destined it will be like that. Like other times well my Father God says up to this point (I will help you). Yes, because my Father God says help yourself and I will help you. The medicine I take, I entrust my actions to God.
Despite an apparent submission to divine authority, this participant also accepts the
responsibility of working in partnership with God to protect her heart health. Religion does
not serve to suppress her desire to seek better health, rather it serves as encouragement to
take responsibility for her health and helps to defines why and how she should proceed. In
the context of limited access to health care resources and chronic hypertension, the logics
of religion at once help to define where control resides and how it is shared.
In the following excerpt, a participant who has been without health insurance and is
diabetic speaks about this shared relationship. Working like a team with God contributes to
the active appraisal of how much control one has. Development of this appraisal requires
active cultivation of the relationship she shares with God and further illustrates that fate,
destiny and personal control reside at the dynamic juxtaposition of institutional barriers to
86


health care, challenges to one’s health and the active negotiation of control though the rationale of religious belief.
Yes I am a believer of that (life is already written). But also destiny, you have it in your hands. With religion, one asks God with a lot of faith. One says, "Help me God. Help me."But I understand that the one that has to help is me. Since I was small I have the belief of asking San Juditas (patron saint of lost causes) for help and there are many people like that. We have faith and with disease we say, “If God wants, he will help too." We combine both things. But when we say, "God wants to help"... of course you still have to do your part. My God says, "Help me and I will help you." I say that my faith is what leads my God to help me. But I always work like a team with God. Con Dios, si. Yes, with God. Well for me it's important because when they discover a disease I try to do everything I can on my side for that disease to go away, as much as is medically possible and with what I know. Because of that for me it's important, because if my father God helps me, maybe that disease will leave.
Despite voicing beliefs that would otherwise be consistent with a fatalistic
disposition, this participant also accepts a measure of personal responsibility for her health
as informed by the logics of her religion. She understands that her relationship with God
bestows certain responsibilities on her such as actively working towards taking care of her
health. In partnership with God, control is both organized and defined at the intersection of
institutional barriers to health care and the presence of diabetes. In these circumstances,
fatalistic language is contextualized and co-occurs alongside statements that reference
personal control and the acceptance of responsibility for one’s health. At the intersection of
institutional barriers to health care, the presence of disease and the logics of religion,
isolated expressions of fatalism are at once replaced by a disposition that seeks a shared
measure of control over circumstances that would otherwise appear to discourage all hope
of.
The Abatement of Fatalism: The Role of Safety Net and Community-Based Resources
Institutional barriers to health care are a key component in the development and expression of CVD-related fatalism among Latinos. The manifestation of this disposition is
87


Full Text

PAGE 1

REINTRODUCING CONTEXT TO THE STUDY OF FATALISM AND THE CARDIOVASCULAR DISEASE SCREENING BEHAVIOR OF LATINOS by CRAIG G. KAPRAL B.A., State University of New York at Potsdam , 1992 M.A., University of Colorado Denver , 2005 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 Program 2019

PAGE 2

ii © 2019 CRAIG G. KAPRAL ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Doctor of Philosophy degree by Craig G. Kapral has been approved for the Health and Behavioral Sciences Program by Karen Lutfey Spencer (Chair) Jean Scandlyn Ronica Rooks Indira Gujral Date: August 3 , 2019

PAGE 4

iv Kapral, Craig G. (PhD , Health and Behavioral Sciences) Fatalism and the Cardiovascular Disease Screening Behavior of Latinos Thesis directed by Associate Professor Karen Lutfey Spencer ABSTRACT Cardiovascular disease (CVD) is a leading cause of adult morbidity and mortality for Latinos living in the United States. Latinos dem onstrate disproportionately hi gh prevalence rates for many established CVD risk factors, including high cholesterol and uncontrolled high blood pressure. Screening for high blood pressure and high cholesterol is critical for both the early detection a nd prevention of CVD. D espite widespread availability, Latinos are less likely than non His panic W hites to utilize blood pressure and cholesterol screening services. In the health literature, the underuse of various screening services by Latinos is often attr ibuted to fatalistic attitudes and beliefs. Fatalism refers to a general belief that the , including the ability to prevent certain health conditions . T hose who endorse fatalisti c attitudes and of education. Unfortunately , extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care (IBHC) Latinos confront in the United States. These barriers include limited access to health insurance , lack of a regular source of health care, and experiences with perceived discrimination in the health care system. Using a sequential mix ed methods design, I examine the relationship between specific IBHC (lack of health insurance, no usual source of health care and perceived discrimination in the health care system), fatalism towards CVD and use of blood pressure and cholesterol screening services by Latinos in Denver, Colorado. The results of

PAGE 5

v this study indicate that fatalism occurs as a byproduct of complex s ocial p ro cesses and offers little in the way of e xplaining the s uboptimal use of CVD s creening s ervices by L atinos. Rather , IBHC in the form of lack of health insurance and no usual source of health care negatively impact the use of blood pressure and cholesterol screening services by this population. Based on these results, efforts to increase use of CVD screening services by Lati nos must include the provision of unimpeded access to quality health insurance and the establishment of a usual source of health care. The form and content of this abstract are approved. I recommend its publication. Approved: Karen Lutfey Spencer

PAGE 6

vi ACKNOWL EDGEMENTS This dissertation (COMIRB protocol # 15 0451) was made possible with the support of several key individuals. First and foremost, my deep gratitude goes to Dr. Karen Lutfey Spencer. Your guidance and support were critical to the completion of thi s research and to my overall academic progress. Thank you for your time, encouragement and detailed feedback. I treasure your mentorship. I would also like to thank my committee members, Drs. Jean Scandlyn, Ronica Rooks and Indira Gujral. Thank you for you r commitment and leadership over the past several years. I greatly appreciate your time and assistance with this work. I would also like to thank my research team, Susana Arreola Ponce de Leon, Blanca u enough for your time and energy. You were an invaluable part of this research and I am forever grateful for your participation. Issamar, I thank you for your dedication and high quality work. Completion of this research would not have been possible witho ut your efforts. In addition, I would like to thank my wife, Kristine Kapral for her love, support and patience during the course of this research. Thank you for your encouragement and devotion. Your love and friendship mean everything to me. I am foreve r indebted for the sacrifices you made on my behalf. Lastly, I would like to thank my late mother, Donna Kapral and father Richard Kapral for their unconditional love and support. I am grateful to have such wonderful parents. Mom, I mi ss you every day. Thank you for always being there with unending emotional support and love. Your memory will forever live in me and continues to inspire.

PAGE 7

vii listening during the countless talks we shared. Thank you for always believin g in me and letting me know. You are a true inspiratio n .

PAGE 8

viii TABLE OF CONTENTS CHAPTER I. .. 1 Research Question CVD Screening Services: Institutional II. A Faulty View: Fatalism as a De Contextualized Cogni tive Trait of the 32 Reintroducing Context: Institutional Barriers and the Expression of Fatalistic .41 III. Goal and Objectives

PAGE 9

ix IV. V. The Abateme nt of Fatalism: The Role of Safety Net and Community Based 94 Dynamic Intersections, Social Context and the Variable Expression of Fatalistic .............. ............... .................... ............... 100 VI. 103 Mixed

PAGE 10

x APPENDI X A. B. Semi C.

PAGE 11

xi LIST OF TABLES TABLE 1. Select Characteristics of Denver County Latino P opulation 49 2. 50 3. Participant Characteristics 65 4. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening when Participants Have Health Insurance and Have a Usu al Source of Health Car e ............................................. 68 5. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screen ing when Participants Have Health Insurance and Have a Usu al Source of Health Care ...69 6. Binary Logistic Regression Results for the Odds of Using Blood Pressure Screeni ng 7. Binary Logistic Regression Results for the Odds of Using Cholesterol Screening based upon Level of Fatalism 8. Binary Logistic Regression Results for the Odds of Demonstrating High Level of Fatalism Based Upon Having Health Insurance and Havin g a Usual Source of Health

PAGE 12

xii LIST OF FIGURES FIGURE 1. Theoretical Model 2. Sequential Sampling Design Using Nested Samples ...47 3. Partially Mixed Sequential Equal Status Design 4. Phase Two Qualitative Interviews Sample Selection

PAGE 13

xiii LIST OF ABBREVIATIONS CVD Cardiovascular disease ATP III Third Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults LDL Low density lipoprotein HDL High density lipoprotein JNC 7 Joint National Committee on Prevention, Detection, Evaluati on, and Treatment of High Blood Pressure ACA The Patient Protection and Affordable Care Act USOC Usual source of health care CAHPS National Consumer Assessment o f Health Plans Study SES S ocioeconomic status HBM Health belief model TPB Theory of planned behavior IBM Integrated behavior model TRA Theory of reasoned action COMIRB The Colorado Multiple Institutional Review Board CAB Community a dvisory board BRFSS Behavioral Risk Factor Surveillance System Survey RAs Research assistants IBHC Institutional barriers to health care U.S. United States of America

PAGE 14

1 CHAPTER I INTRODUCTION Cardiovascular disease (CVD) is a leading cause of adult morbidity and mortality for Latinos living in the United States (AHA, 2010 ). In 201 6 , CVD accounted for 27 percent a nd 28.2 percent of all cause mortality among Latino males and females resp ectively ( AHA, 201 6 ). Latinos are known to demonstrate disproporti onately high prevalence rates for many established CVD risk factors, including high cholesterol and uncontrolled high blood pressure ( Mozaf farian et al. , 2016; CDC, 2011a ; Jurkowski, 2006; ). For example, recent results from a national, multi site study of 16, 415 Latino adults found that 45 percent have high cholesterol and 25 percent have hypertension (Rodriguez et al., 2015). Importantly, the findings from this study demonstrate that of those with high c holesterol, almost half (49.3%) were unaware of their condition. In a separate study drawing u pon the same data used by Rogrig uez et al. (2015) above, Sorlie and colleagues found that among hypertensives, more than 25 percent were unaware of their conditio n and 62 percent did not have their hypertension controlled (Sorlie et al., 2014). Screening for CVD risk factors is important for both the early detection and prevention of CVD ( AHA, 2019 ; Jurkowski & Johnson, 2005). Given the large proportion of Latino s with high cholesterol and those who remain unaware of their hypertensive status, screening is imperative to reduce the overall burden of CVD in this population . However, d espite widespread availability of screening tests , Latinos are less likely than non Hispanic W hites to utilize cholesterol and blood screening services ( AHA, 2016 ; AHA, 2007; Balluz, Okoro, & Strine, 2004; Behbakht, Lynch, Teal, Degeest, & Massad, 2004; Facione, Miaskowski, Dodd, & Paul, 2002; Garcia et al., 2017 ; Lu quis & Villanueva Cruz, 2006 ) . In the

PAGE 15

2 hea l th literature , the underuse of various screening services by Latinos is often attributed to fatalistic attitudes and beliefs . ( AHA, 2007; Balluz et al., 2004; Behbakht et al., 2004; Cohn & Esparza Del Villar, 2015 ; Facione et al., 2002; Luquis & Villanueva Cruz, 2006 ). Fatalism refers to a general belief that the course of fate cannot be changed and that Abraido Lanza et al., 2007 ; Leyva et al., 2014 ). More hold fatalistic attitudes often subscribe to the idea of lack of personal control, and thus believe they have no control over life events, incl uding the ability to prevent certain health conditions ( Fairchild, 2015 ; Ramirez et al., 2002 ). Fatalism i s typically operationalized as negative attitudes toward preventive health practices and disease outcomes and has therefore been identified as a poten tial barrier to disease detection and prevention efforts (Espinosa de los Monteros & Gallo, 2011). Unfortunately , extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care Latinos confront in the United States. These barriers include limited access to health insurance ( Balluz et al., 2004 ; OMH, 2019 ), lack of a regular source of health care ( Balluz et al., 2004 ; Hayes et al., 2015 ), and experiences with perceived discrimination in the health care system ( Ariza, Vimalananda, & Rosenzweig, 2010 ; Benjamins & Middleton, 2019 ) . The overemphasis on individual level fatalistic attitudes carries important implications . First, t hose that endorse fatalistic attitudes and c hoose to forgo screening and treatment are unfairly labeled 2011; Espinosa de los Monteros & Gallo, 2011; Straughan & Seow, 1998). The decontextualized labeling of those who express fatalistic be liefs serves to misdirect

PAGE 16

3 attention away from the very real barriers to health care that Latinos face and offers little in the way of producing meaningful interventions designed to increase use of CVD screening se rvices . Second, t hese views ignore the social, economic and institutional barriers that may inform the health beliefs and decision making processes of those who hold fatalistic attitudes and perpetuate a short sighted public health discourse characterizing individuals as so lely responsible for their health and wellbeing (Keeley et al., 2009). Reintroducing context to the study of fatalism and screening behavior is essential if we are to more fully under stand how individuals rationalize about the use of prevention resources when faced with persistent structural obstacles to health care services . However, solely quantitative approaches has left fundamental questions about how people internalize, reason about, and respond to institutiona l barriers to health care unanswered. This is not to say that quantitative methods are unimportant to the study of the screening behavior of vulnerable populations , how people relate to and make sense of these insti tutional barriers . Failing to answer these types of how and why questions leaves us with a limited understanding of how motivations and attitudes towards CVD prevention and use of screening res o urces. To fully understand how institutional barriers impact the health beliefs and screening behaviors of Latinos, research must employ the complementary strengths of qualitative methods. While quantitative methods allow for the testing of hypotheses regarding the contribution of fatalism to screening behav ior they remain insufficient to the

PAGE 17

4 study of fatalism in context when used in isolation. Employing solely quantitative methods to the study of fatalistic attitudes and screening behavior of Latinos fails to account for how people interpret and respond to i nstitutional barriers to health care and leaves our inquiry into the screening behavior of this population shortsighted and woefully incomplete. Adding qualitative methods permits one to critically explore the meaning of the quantitative results obtained w hile investigating how institutional barriers may be incorporated into the lived experience, decision making processes and health behavior of this population . Therefore, this study employs mixed methods to more fully understand the relationship between ins titutional barriers to health care, fatalism towards CVD, and the use of cholesterol and blood pressure screening services by Latinos in Denver, Colorado. T he goal of this study is to answer the following research question . Research Question 1. How does accounting for the social context in which health care decisions are made improve our understanding of fatalism and its impact on the CVD screening behaviors of Latinos in Denver, Colorado ? Specific Aims and Hypotheses 1. Quantitatively examine the rel ative c ontributions of specific institutional barriers to health care ( e.g., lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care encounter) , and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol ) by Latino s in Denver, Colorado.

PAGE 18

5 Hypothesis 1a : Institutional barriers to health care resources will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services . Hypothesis 1b : Fatalism will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. Hypothesis 1c : Institutional barriers to health care resources will demonstrate a positive relationship with fatalism. Hypothe sis 1d : Institutional barriers to care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism. 2. Qualitatively explore how institutional barriers to health care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latino s in Denver, Colorado . B ackground More than doubling since 1990, the Latino population is the largest and fastest growing ethnic minority in the United Stat es ; they currently number nearly 5 8 million or 1 8 percent of the total U.S. population ( Flores , 201 7 ) . Recent estimates gauge that by the year 2050, U.S. Latinos will reach 1 19 million and comprise 29 percent of the total population . Of the major origin groups, those identifying as Mexican are currently the most numerous (63 percent ) , followed by Ce ntral and South Americans (13.5 perc ent), Puerto Ricans (9.2 percent) and Cubans (3.5 percent ) ( US Census, 201 5 ).

PAGE 19

6 Although Latinos share a number of cultural assets and protective factors that influence their health beha viors , they are widely considered to be a socio demographically vulnerable population (Pa lacio, 2009). Nationally, 42 percent of Latinos aged 18 to 64 lack he alth insurance ( CDC , 2011 ), 67 percent are poor or near poor ( CDC , 2011 ), 41 percent have less than a high school education ( CDC , 2 011) , and 6 percent report perceived discrimination in the health care system in the past 12 months (Saha et al., 2003 ; Stepanikova & Oates, 2016 ). These facts are troubling to say the least. Given the Latino risk profile, there is real concern that an increase in CVD among Latinos in the United States is close at hand . According to Davidson et al. (2007), the prevalence of cardio vascular disease and related risk factors, which is in stark contrast to the limited healthcare resources, screening, and prevention tools at the disposal of Latinos/His p. 568). Latinos in Colorado Latinos have been a significant part of (Palacio et al. , from the Southwestern United States, M exico and, in more recent years, Central America. Similar to national trend s Latinos are the largest ethni c minority group in the state (Palacio et al. , 2009). W hile Latino population is diverse, the 20 10 C ensus found a majority identifying as (US Census, 2010). Although wide ly distributed across the state, the population

PAGE 20

7 is more densely c oncentrated in the Denver, Pueblo and Greeley metropolitan areas, the San Luis Valley and mountain resort communities (Palacio et al. , 2009). CVD is a leading cause of death for Colorado Latin os . In addition population experiences simi lar socio d emographic disadvantage to that of their nati onal counterparts with over 40 percent la cking health insurance, over 23 percent living below the poverty level, and 43 percent without a high school education (Palacio et al., 2009). Moreover, Colorado Latinos shoulder a disproportionate burden of CVD risk. According to recent data from the Colorado Department of Public Health and Environment , the prevalence of high blood pressure in Latinos was greater than t hat for non Hispanic W hit es ( Calanan et al., 2011 ). In addition , Colorado Latinos are t wice as likely as non Hispanic W hites to have diabetes, a major risk factor for CVD (Palacio et al . , 2009) . Among all adults who report having diabetes in Colorado , 65 percent report high blood pressure and 63 percent report high cholesterol, compared with 19 percent and 31 percent, respectively, of adults who do not report having diabetes ( Calanan et al., 2011). These data demonstrate that Latinos are at disparate risk for CVD and may in fact suffer the consequences in years of life lost , as their life expectancy is lower than that for all Co loradans (Palacio et al. 2009). CVD Screening Services Risk factor screening is an integral part of a comprehensive approach to C VD prevention. Most importantly, results from screening tests provide the individual and their health care provide r with the necessary knowledge to make informed decisions about their health and the pre vention of disease. Screening for high blood pressure and high cholesterol is critical to efforts concerned with identify ing persons whose risk factors may be undiagnosed and who might benefit from lifestyle modification or pharmacologic

PAGE 21

8 therapy. To decrease morbidity and mortality from CVD , public health programs must identify persons with multiple risk factors and focus interventions on those populations d isproportionately affected ( Calanan et al. 2011 ). Although progress against CVD has been made, declines in death rates between 1999 (Calanan et al., 2011). For example, t he decrease in death rates due to heart disease was onl y significant for non Hispanic W hites and Blacks during this time period. This may be due in part to differences in the use of screening and treatment services by major racial and ethnic gro ups. Cholesterol S creening According to the National Cholesterol Education Program and the Third Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (ATP III), all adults o ver the age of 20 should have a fa sting lipoprotein profile once every five years (NIH, 2001) . A fasti ng lipoprotein profile consists of a blood test and provides information on total cholesterol, high density lipoprotein ( HDL ) cholesterol, low density lip oprotein ( LDL ) cholesterol, and triglycerides. However, both governing bodies recommend that cholesterol be checked more frequently if (a) total cholesterol is 200 mg/dL or more, (b) you are male and over 45 years of age, (c) you are female and over 50 yea rs of age, (d) your HDL cho lesterol is less than 40 mg/dL or (e) you have other risk factors for heart disease and stroke. Due to the prevalence of multiple CVD risk factors common in Latino populations, including high cholesterol, uncontrolled hypertensio n, type II diabetes , and being overweight ( Blackwell, Lucas & Clarke , 2014), it stands to reason that members of this population should have their cholesterol checked more often in order to

PAGE 22

9 acc urately detect the presence of hyperlip idemia and/or assess whether pharmacologic therapy is working as intended. Unfortunately, recent data show that only 58 percent of Colorado Latinos have ha d a cholesterol test in the past five years compared to 78 percent of non Hispanic W hites (Calanan et al . , 2011). Blood Pressure S creening As with cholesterol, guideline based recommendations state that blood pressure should be checked at regular intervals. According to The Joint National Committee on Prevention, Detection, Evaluati on, and Treatment of High Blood Pressure (JNC 7) , blood pressure should be checked every two years if less than 120/80 mm/Hg ( U S PSTF, 2007) . However, the A merican H eart A ssociation (2016) recommends that blood pressure be checked more frequently if above this level. D ata compiled by the Colorado Department of Public Health and Environment i (Calanan et al., 2011) . However, p revalence rates of high bloo d pressure in Colorado are likely to underestimate actual rates due to self report and other limitations associated with data collection procedures. For example, anyone who has high blood pressure who has not visited a health care provider, has not been tested for high blood pr essure , or does not remember being diagnosed is not included in the prevalence estimate. More specifically , prevalence rates among Latinos in Colorado may be further un derestimated due to the fact that this population is more likely to experience a lack of health insurance and access to care than non Hispanic W hites and thus may not receive blood pressure tests and diagnoses as frequently as their W hite counterparts (Calanan et al . , 2011).

PAGE 23

10 CVD Screening Services: Institutional Barriers I nstitutional barriers to health care have received a great deal of attention in the literature on health disparities between Latinos and non Hispanic W hites. E vidence suggest s that certain barriers such as lack of access to health insurance, no usual sourc e of health care, and perceived discrimination in the health care system negatively impact use of screening services . In the following sections, I review previous research findings regarding the influence of these specific barriers to health care on Latinos use of blood pressure and cholesterol screening services . Health Insurance Access to health care services in the United States is highly dependent on availability and access to health insurance. Those who lack insurance must independently pay for health care services or seek care from safety net facilities such as free clinics, h ospital emergency rooms, and community health centers (Rodriguez et al., 2014) . R esearch has identified lack of health insurance as an important barrier to the receipt of health care services (Burstin et al., 1998; Jurkowski & Johnson, 2005 ; KFF, 2018 ; Solis et al., 1990 ; Summer, 2002; Ward et al. 2004) . When compared to other racial or ethnic groups, Latinos are more likely to be uninsure d (Jurkowski, 2006) . Employment in jobs that do not offer benefits and do not require citizenship or English language proficiency constitute the primary barriers to health insurance for th is population (Angel, Angel, & Lein, 2003). H aving access to care may facilitate CVD screening, which in turn may prevent or delay the development of the disease and prevent morbidity and mortality. For example, having health insurance leads to higher rates of CVD diagnoses ( Alcala 2015) and reduces the risk for major cardiac events (Amarenco et al., 2014) . This s uggests that CVD is more

PAGE 24

11 likely to be identified and controlled among the insured. Unfortunately , lon ger term uninsured status among Latinos is linked to the decreased likelihood of engaging in behaviors that could reduce CVD risk such as screening for hyp ert ension and high cholesterol (Alcala et al., 2015 ). R esearch on CVD screening practices am ong Latino s found that those with health care coverage were more likely to have had their blood pressure checked and to have obtained a routine check up than those without coverage ( DeVoe et al., 2003; Jurkowski & Johnson, 2005). In similar fashion, recent work conducted by Alcala et al. (2015) found that among 464 Latino adults, those with health insurance were at significantly greater odds of having a cholesterol test when compared to those without insurance (OR 3.76, 95% CI 1.73 7.76). Likewise, research carried out among Latinos in Texas by Salinas et al. (2015), demonstrated that health insurance was positively associated with having both blood pressure screenin g and cholesterol screening. According to B urstin and colleagues (1998) Latinos without health insurance or thos e who change health plans are more likely to delay seeking medical care, thus risking the development of complications that could be avoided with early detection. Using a nationally representative sample of 8,309 adults , Sudano and Baker (2003) demonstrate that lack of health insurance coverage for any period of time is significantly associated with lower rates of use of most preventive health services including cholesterol screening. R esults from this study showed that those wh o obtained health insurance during the study period had rates of use that were approximately halfway between those without coverage and those who had continuous health insurance coverage. This suggests that

PAGE 25

12 obtaining health insurance leads to improved use of CVD screening services for those who were previously uninsured. Lack of health insurance undermines access to and use of preventive health care services including blood pressure and cholesterol screenings . Given their disproportionate representation am ong the uninsured, Latinos are vulnerable to the negative CVD related health outcomes associated with undiagnosed hypertension and hypercholesterolemia. Studies concerned with the screening behavi or of Latinos must adequately consider the implications that a persistent lack of health insurance presents in this population. Failure to do so ignores the substantial impact this barrier has on use of CVD screening services by Latinos in the United States. The Patient Protection and Affordable Care Act Discussion of access to health insurance and its potential impact on use of screening services must consider how The Patient Protection and Affordable Care Act (ACA) has changed the availability of health insurance to those who have not been able to access The A CA became law in the United States on March 23, 2010. Considered a watershed moment in U.S. public health policy, the A CA established basic legal protections that have until now been absent, including access to affordable health care coverage for most Americans from birth through retirement (Rosenbaum, 2011). Consist ing of 10 separate legislative t itles, the A CA has five major aims including to: ( a) achieve near universal coverage through a shared responsibility among government, individuals, and employers ; ( b) improve the fairness, quality and affordability of health insurance coverage ; ( c) improve health care value, quality and affordability while reducing wasteful spending ; ( d) s trengthen primary care ac cess w hile incorporating longer term

PAGE 26

13 changes in the availability of primary and preventive health care ; and ( e) make targeted community resources (Rosenbaum, 2011). The legal protections set forth by the ACA assume that the pervasive disparities in health insurance coverage between Latinos and non Hispanic W hites will disappear . However, d espite the aim to achieve near universal coverage, the A CA is expected to leave near ly 25 million people without health insurance (Rosenbaum, 2011). Recent analyses of American Community Survey data by Artiga and Orgera (2019) demonstrate that along with American Indians and Alaska Natives, Latinos represent the largest propo rtion of the uninsured in the United S tates with 19 percent still lacking coverage . In Colorado, Latinos also represent the largest p rop ortion of the uninsured with 16 percent of the total Latino population lacking health insurance compared to only six percent of non Hispanic Whites (Artiga & Orgera, 2019). In states that chose not to expand Medicaid coverage under the ACA, nearly 27 percent of Latinos lack coverage compared to on ly 10 percent of non Hispanic Whites (Artiga & Orgera, 2019). According to data compil ed by The Colorado Health Institute, as of 2017, Latinos in Colorado remain almost twice as likely to be without health insurance as non Hispanic Whites with 115,000 stil l lacking coverage (CHI, 2017). Although they comprise only 2 1 population, Colorado Latinos accounted for 3 4 percent of the un insured in 2017 ( CHI , 201 7 ). Despite millions hav ing gained access to health care coverage under the ACA, m any individuals within the larger Latino community continue to experience real barriers to obtaining health care cove rage and services. This is particularly true for unauthorized individuals, as the AC A does not include health care coverage opti ons for undocumented

PAGE 27

14 immigrants ( Ortega, Rodriguez & Vargas Bustamante, 2015). Assuming that Latinos will achieve equitable access to health insurance under the auspices of the ACA is short sighted and fails to a ccount for ongoing barriers that this population faces . Culturally appropriate and targeted outreach and enrollment strategies are critical to link community members with available health coverage and services (Maxwell et al., 2011) . For example, targeted outreach efforts by local community based organizations were an essential part of programs and provide a model for national efforts (Maxwell et al., 2011). Despite such successes, m uch of this work is just beginning. The real impact of the ACA on access to affordable health insurance for Latinos will only be known in the coming years. Usual Source of Health C are Along with limited access to health insurance, not having a usual source of health care is an important institutional barrier to health care for Latinos. H aving a usual source of h ealth care (USOC) has been associated with greater service use and earlier receipt of preventive services ( Ettner, 1999 ; Finney Rutten et al., 2015 ). This may be explained in part because patient s with an ongoing, long term relationship with their physician may have heightened awareness of certain health b ehaviors along with the opportunity to discuss these behaviors, gain access to health related information , and treat problems (Ettner, 1999). Previous research has shown that Latinos are less likely to have a USOC than non Hispanic W hites ( CDC, 2013 ; Corbie Smith et al., 2002 ; Guendelman & Wagner, 2000 ). According to findings from research using a nationally representative sample of Latinos carried out by Rodriguez, Bustamente and Ang (2009) , more than 25 percent of

PAGE 28

15 participants lacked a USOC. The stu pre dicts a blood pressure check and results in greater likelihood of having both a blood pressure and cholesterol test compared to no usual source of care . In a nother nationally representative sample of health care use among adults in the United States, having a USOC was independently associated with receiving multiple preventive services, including blood pressure and cholesterol screening (Corbie Smith et al., 2002). In similar fashion, T orres et al. (2007) work among a nationally representative sample of U . S . adults showed that lack of health insurance and a USOC were among the strongest predictors for the failure of Latinos to receive preventive services. In work examining health services utilization among a nationally representative sample of Latinos and non Hispanic W hites, Guendelman and Wagner (2000) found that having a USOC significantly reduced th e gap between the two populations in preventive care utilization including blood pressure and cholesterol screening tests. Much like Torres health insurance coverage are the most significant factors involved in eliminating or reducing the disparities in utilization of preventive health services between Latinos and non Hispanic W hites. Having a usual source of care is associated with greater use of preventive care and mon itoring services. Preventive care and health care monitoring are both associated with better long term health outcomes, including better control of chronic health conditions such as CVD (Livingston, Minushkin & Cohn, 2008). Due to the fact that Latinos are less likely than non Hispanic W hites to have a USOC, this population may not equally experience the benefits that having a USOC confers such as heightened awareness of and

PAGE 29

16 the opportunity to screen for and treat adverse health conditions. Not having a USO C information and resources can be obtained. Efforts to understand Latinos suboptimal use of CVD screening services must assess the relative contribution of USOC as well as how this barrier may influence the very attitudes of Latinos towards CVD prevention services. Otherwise, we risk leaving the critical influence of this institutional barrier unaccounted for. Perceived Discrimination in the Health Care System Over the past 15 years, research has paid increasing attention to the impact perceived discrimination has on racial/ethnic disparities in the use of health care (Burgess et al., 2008). Several studies concerned with the relationship between perceived discr imination and use of health care services show support for the hypothesis that perceived discrimination in the health care system negatively i nfluence s use of services offered (Burgess et al., 2008). Increases in reported experiences with perceived discrim ination are associated with lower rates of participation in use of preventive health services and the elevated likelihood of delaying care (Burgess et al., 2008). According to Trivedi and Ayanian (2006) perceived discrimination may be related to use of preventive services in three ways. First, those who feel discriminated against may be less likely to seek visits for routine care at which preventive services are likely to be offered. Second, peopl e who experience perceived discrimination may lose trust in their providers thus making them less likely to follow recommendations for use of certain prevention or screening tests. Third, patients who perceive discrimination may receive less decisive recom mendations for health related tests and treatments. Findings from Trivedi

PAGE 30

17 (2006) work show that 21 percent of Latinos cite insurance type as the most frequent r eason for discrimination , with 20 percent and 12 percent citing race/ethnicity and language/accent as the next most common reasons respectively. According to their findings, persons who report perceive d discrimination in a health care encounter are significantly less likely to receive several preventive health screenings including chole sterol testing. According to work conducted by van Ryn (2002) and van Ryn and Burke (2000), patient race/ethnicity can influence physician interpretation of patients illness narratives and decisions to refer patients for particular treatm ents or procedures (van Ryn, 2002; van Ryn & Burke, 2000 ). Other research findings indicate that quality of health care provided may differ substantial ly by racial/ethnic group including receiving needed care and timeliness of care ( Benkert, Peters, Morales, E lliott, Weech Maldonado, Spritzer, & Hays, 2001 ) ; quality of communication ( Collins, Clark, & Kressin, 2002 ) ; perceived discrimination due to insurance status ( Trivedi & Ayanian, 200 6 ) ; provider assessment of patient intelligence, likelihood of risk behavior and adherence with medical advice ( van Ryn, 2002; van Ryn & Burke, 2000 ) ; and language associated perceptions of staff helpfulness ( Weech Maldonado, Morales, Elliott, Spritzer, Marshall, & Hays, 2003 ). This body of work demonstrates that minority groups receive less equitable care and face perceived discrimination in health care settings. Patient perceptions may affect their hea lth related decision maki ng with previous studies revealing the impact of perc eived quality of care on adherence to medical advice (Blanchard, 2004), screening recommendations (Trivedi & Ayanian, 2006), and medication regimens (Blanchard & Lurie , 2004; Van Houtven et al., 2005) . Sorkin et al. (20 10 ) suggest

PAGE 31

18 that p erceived d iscrimination during the health care process may repre sent a n important mediator between race/ethnicity and observed racia l/ ethnic differences in patient perception of care. Research demonstrates that m inority patients are more likely to report being the subject of negative attit udes during the health care process than are W hites ( Chen, Fryer, Phillips, Wilson, & Pathman , 2005; Haviland, Morales, Reise, & Hays , 2003; Haviland, Morales, Dial, & Pincus, 2005; Kandula, Hasnain Wynia, Thompson, Brown, & Baker, 2009 ; LaVeist, Rolley, & Diala , 2003) . In a random sample of more than 40,000 households in California, Sorkin and c olleagues (20 10 ) found that Latinos were more likely to feel discriminated against in health care because of their race and ethnicity than were non Hispan ic W hites. In other research based on the National Consumer Assessment o f Health Plans Study (CAHPS) , Latinos reported negative experiences in every area assessed, including rec e i ving needed health care, communication with health care provider s , and timeliness of care ( Weech Maldonado et al., 20 10 ). R ecent work on perceived discrimination and use of health care services a mong Latinos in North Carolina by Keller, Silberberg, Hartmann, and Michener (2010) fo und that perceived discrimination , based on both race/ethnicity and insurance status , w as associated with an increased likelihood of going without needed medical services . Perceived discrimination has a negative influence on health by increasing the lik elihood that individuals will avoid seeking important health services. Experiences of discrimination and unfair treatment may engender negative expectations by Latinos about how they will be treated within medical institutions or by members of the dominant culture, thereby making them hesitant to enter such encounters (Burgess et al., 2008).

PAGE 32

19 Perceptions of discrimination in health care encounters by Latinos inhibits trust ful communication between patients and providers and leads to membe rs of this population being less likely to use critical preventive medical procedures and services including blood pressure and cholesterol tests . Fatalism In an effort to account for the s uboptimal use of screening and treatment services by Latinos, r esearch has begun to increasingly examine the impact of health related beliefs or attitudes. According to Drew and Schoenberg (2011), i n the health sciences one such belief, the fatalism constr uct , has increasingly been applied to account for both health be havior s and outcomes . When applied to health, fatalism is often operationalized as negative or pessimistic attitudes regarding preventive health practices and disease outcomes (Drew & Schoenberg, 2011). Fatalism ( fatalismo ) is often considered to be a dominant cognitive orient ation of Latino s, a population that is frequently described in the literature as passive, subjugat ed , and fatalistic (Cueller, Arnold, & Gonzales, 1995; Espinosa de los Mon t eros & Gallo, 2013). Early anthropological work among Mexican Americans (Madsen, 1973) suggested that suffe ring endured by this population is made more acceptable by a strong belief in fatalism. of the individual is predestined and every occurrence in human existence comes to pass because Mexican American tries to accept. Misfortune is something the Anglo t ries to overcome and Unfortunately, this type of ahistoric labeling continues with more recent work on

PAGE 33

20 fatalism and its affect on the use of disease screening services by minorities, perpetuating the view th at fatalism among Latinos is an inherently cultural trait ( Balcazar, Castro, & Crull, 1995; Harmon, Castro , & Coe, 1996; Hubbel, Chavez, Mishra & Burciaga Valdez , 1996 ; Hubbel, Chavez, Mishra & Burciaga Valdez, 1996 2 ; Otero Sabogal, Stewart, Sabogal , Brown, & Perez Stable, 2003; Perez Stable et al., 1992; Suarez, Nichols, Roche, & Simpson, 1997 ). These views ignore the context within which health related beliefs and behaviors are formed and maintained. Failure to account for how institutional barrier perceptions of CVD prevention and use of health care screening services leads to faulty assumptions about rationality and Latino culture. Furthermore, decontextualizing the study of fatalis m and health behavior leads to the misattribution of underuse of screening services by Latinos to irrational beliefs rather than institutionalized barriers that impact this populations health related agen cy . According to Espinosa de los Monteros and Gallo emphasizing the role of cognitive processes in explaining the health behavior of disadvantaged populations is that structural barriers may better account for health disparities than cognitive processes which could merely reflect the structural barriers ex The which fatalism may occur perpetuates the view that fatalism is a faulty belief system held by Latino culture. Unfortunately, this limited view and singular focus on cognitive traits has hindered efforts to better understand the complex relationship between barriers to health care, fatalistic attitudes and use of C VD screening services by this population . In research examining motivations for use of health screenings among 452 Latino s,

PAGE 34

21 Ashida et al. (2010) asked participants how much control they believe they ha ve in preventing heart disease and diabetes. On a scale of 1 (no control) to 4 (total control), a mean score of 2.46 + 0.82 revealed that participants partially attribute d the prevention of CVD to sources out of their control. A s a resu lt of qualitative work conducted with L atinas , Garces et al. (2006) report ed that many of the women believed they d id not have control over their own health , information, language barriers, and lack of access to health care and medications. I n similar fashion, Shive et al. (2006), found that among a nationally representative sample, Latinos were more fatalistic toward their health than were African Americans, non Hispanic W hites, and Asian Americans. However, d espit e collecting data on health behaviors, such as use of cholesterol and b lood pressure screenings , the authors do not report whethe r fatalism was associated with the use of these services . More importantly, the authors fail to discuss how fatalism may be related to socioeconomic disadvantage and institutional barriers to health care despite collecting data on these socio demographic v ariables. This results in unanswered questions about the potential context dependent nature of fatalism. In work designed to better understand participation in community health screenings, including blood pressure and cholesterol, Engebretson, Mahoney, an d Walker (2005), found a complex array of facilitators and inhibitors across personal, interpersonal, and environmental dimensions to be responsible for whether individuals utilized health screenings. Importantly, their findings reveal that Latinos may not use screening services, even if offered at no cost, due to concern over finding something might be wrong and not having the economic resources with which to address the problem. The worry and stress among Latino study participants over potential negative screening results and the inability

PAGE 35

22 to access and pay for follow up care identified by Engebretson and colleagues reveal how a lack of control many Latinos feel may result in expressions of fatalism toward using screening services such as blood pressure an d cholesterol . Despite this finding, between fatalism towards CVD and use of blood pressure and cholesterol screening services after accounting for institutional b arriers to care. Combining this type of qualitative work with quantitative assessment will more fully explain how these particular barriers are related and the meaning they hold for members of the Latino community. Perhaps most widely applied in the cance r literature, fatalism has been shown to act as a deterrent to the use of certain screening and prevention services by Latinos. In an effort to determine whether fatalism predicts participation in cancer screening after accounting for structural barriers to health care , Espinosa de los Monteros and Gallo (201 1 ) conducted a meta analysis of all relevant studies published before February 25, 2010. Of the 11 studies meeting the inclusion criteria, seven reported a statistically significant inverse relationsh ip between fatalism and use of cancer screening services after accounting for structural barriers such as health care access and socioeconomic status (SES) . However, the authors note that mixed findings and inconsistencies in study measures and design prec lude definitive conclusions regarding the nature of the relationships found. T quantitative studies in their analysis fails to capture the potential complexities of the relationship between structural constraints, fatali , despite finding inverse associations between fatalism and key socio demographic variables such as SES, limited access to health insurance and lower education and/or income in six of the eleven

PAGE 36

23 studie s, the authors fail to address how fatalism may be mor e accurately represented as a mechanism linking structural barr iers to screening rather than an independent predictor of screening behavior . Moreover, the sole reliance on quan titative studies falls sho rt in answering crucial questions regarding the motivations fatalistic attitudes and behavioral practices. To answer these types of questions, the inclusion of qualitative methods is essential. In work to determine which factors predict use of mammogra phy screening by Latinas, Laws and Mayo (1998) include a modified health locus of control variable as a proxy for fatalism and several social variables, including education level and insurance status, in their model to answer whether the relationship between these social factors and screening are the same or different for Latina versus non Latina women. Their results show that access to primary care services via health insurance and physician behavior act as key determinants of use of mammography screening among their sample. Alternately, the authors illustrate that fatalistic attitudes towards cancer have little to do with screening behavior in their sample of Latina women. Research designed to elicit the complexity of fatalism an d its relationship to rural decision making surrounding the use of health care services (Drew & Schoenberg, 2011) shows that numerous factors including inadequate access to health care services, a legacy of self reliance, and a culturally acceptabl e idiom of fatalism foster the use of, but not necessarily a rigid belief in, fat alism towards disease. For study participants, health decisions and behavior are embedded in a larger context that includes real life circumstances and constraints such as pow er and agency and personal and familial expectations that intersect with health care provider interactions. According to Drew and

PAGE 37

24 Schoenberg (2011 Rather than reducing complex h ealth decision making to those must also decision to undergo screening and treatment. This is critically important to efforts of CVD screening services. In order to truly understand the health beliefs and behavioral motivations of Latinos in the U . S . , we must move beyond quantitative only approaches to the study of screening behavior by incorporating qualitative methods into our research designs. Doing so will serve to strengthen our inquiries by providing us with the ability to answer essential questions about how the context within which decisions are made influence the very behaviors we hope to encourage. Anthropological Cont ributions Anthropological inquir y 1 into fatalism and health behavior is sparse but holds important insight for understanding how fatalism may mask social realities and the variations in the way individuals cultivate meaning regarding use of medical services or treatment. In work among terminally ill dialysis patients in Egypt, Hamdy (2009) argues that rather than viewing fatalism as passive submission to divine authority, it is better understood as a disposition that must be actively cultivated through work on the self. 1 The bulk of work on fatalism and health behavior has taken place in the fields of psychology, public health, speech communication, and the socio medical sciences. Anthropological research on fatalism is much less common. Anthropological findings on the inf luence of fatalism on health behavior and use of screening services emphasize the role of both structural and cultural variables in understanding fatalism and behavior. Emphasis on social context has led anthropologists to identify social and material def icits (e.g., poverty, unequal power relations, and inadequate access to culturally appropriate health care resources) common among minority racial and ethnic populations as the organizing factors in the development of fatalism among such groups.

PAGE 38

25 Hamdy demonstrates that whether and when to cultivate such a disposition is contingent upon how much control patients feel they have when faced with illness and whether there are readily available optio ns for solutions that are acceptable medically, socially and conditions intersect with the logics of religion as they face difficult decisions about medical care, and oftentimes, limited options for recovery. In fieldwork among the Fulani and Humbebe in the Mopti region of Mali, Castle diagnosing a child with a fatal illness. Fo r children who die before the age of five, death is often attributed to the fol k illnesses foondu (the bird owl) and heendu (the wind). According to Castle, these diagnoses act as explanatory mechanisms in a high mortality area where the Fulani code of b ehavior known as pulaaku , mandates ac ceptance and stoicism when faced with physical hardship and emotional distress. For the Fulani, both f oondu and heendu are viewed as very serious and incurable illnesses. The incurable nature of these diagnoses can be t aken as proxies for fatalistic appraisals of these particular illness events (Castle, 1994). What Castle illustrates is that the application of foondu or heendu as a cause of death rarely occurs during current illness episodes but rather during retrospecti ve accounts. According to Castle, the concepts of fate or destiny are manipulated such that giving fatal diagnoses to sick or dying children or applying them retrospectively after death, absolves mothers from guilt, blame and responsibility. In addition to tr ansferring analyses illustrate how diagnoses of foondu and heendu serve to reinforce hierarchical power relations among women due to the fact that senior women in a patril ineage only

PAGE 39

26 confer diagnoses . Castle suggests that i t is possible in a community facing extremely high rates of child mortality that foondu and heendu allow psychological adaptations to the frequent experience under circumstances that would otherwise seem uncontrollable. (2008) among the Zaramo people of Tanzania shows that for parents of children diagnosed with degedege (the indigenous name for a life threatening illness otherwise viewed by biomedicine as cerebral malaria), treatment seeking and therapy management are shaped by multidimensional influences. While decisions to forego treatment in an effort to save a dying child may be construed by Western observers , , , demonstrates that economic, structural, and micro level social factors intersect with cultural beliefs to influence views about disease an d the decision making process associated with obtaining a diagnosis and seeking treatment (Kamat, 2008). In this context, structural factors, including poverty, lack of access to dispensary facilities, and unpleasant encounters with the dispensary staff se rve to affect how the Zaramo view the options available to them when faced with degedege . Failure to gain timely access to biomedicine that is safe, affordable, and efficacious and unpleasant encounters with health care professionals highlight how structur al factors, power relations, and economic circumstances conspire to organize health beliefs and behavior among this particular group. From an anthropological perspective, Nichter (2002) cautions that the study of therapy management must include not only w hat people do (and can do) including the reasons for their actions, but also what they are unable to do. He urges us to examine the

PAGE 40

27 factors that underlie apparent passivity or fatalism . Failure to do so reinforces simplistic impressions that ignorance is a other factors are involved (Nichter, 2002 ). Anthropological inquiry into fatalism demonstrates that multiple social, cultural, and structural factors frequently serve to organize attitudes, beliefs, and decision making practices regarding health. Rather than viewing health related behavior as seemingly irrational, anthropologi sts emphasize the multiple contexts in which decisi ons are made. Most importantly, insights gleaned from cross cultural work carried out by anthropologists show that emic or insider perspectives are critical to understanding the rationale behind how, when, and why people do and do not seek health care services. The present research attempts to apply these lessons in an effort to account for as of yet unanswered questions about the impact of structurally organized barriers to health care on the CVD related beliefs and behaviors of Latinos in the U.S. To do this, q ualitative methods will serve to complement and extend the quantitative approach employed. Through mixed methods, this project seeks to comprehensively address a critical gap in current knowledge regarding how contextual variables influence the expression of fatalistic attitudes toward CVD and suboptimal use of blood pressure and cholesterol screening services. Research that reintroduces context to the study of fatalism and health behavior is critical to the design and implementation of interventions design of CVD screening services. B y more fully understanding how people rationalize about the use of disease prevention tools when faced with structural barriers to health care, public health interventionists will be better equipped to design and implement tailored approaches that effectively incorporate responses to the challenges of distributing

PAGE 41

28 equitable health care to this p opulation. Structure of the Dissertation This study draws upon mixed methods and multidisciplinary perspectives to understand the phenomenon of suboptimal use of CVD prevention services by Latino s. Chapter 1 introduces the research questions and explores previous research in the area of screening practices among Latinos. Chapter 2 examines the application of psychological and sociological theories traditionally used to explain health beliefs and health behavior in the context of screening practices. Speci fically, this study critique s the use of individually based theories to understand fatalistic attitudes among Latino population s and draws upon the concept of structural violence to account for the expression of fatalistic attit udes towards CVD and subopti mal use of CVD prevention services. Chapter 3 describes the study design as well as the quantitative and qualitative methods employed. Chapter 4 discusses the quantitative analys e s and results. Chapter 5 discusses the qualitative analys e s and results. Chap ter 6 synthesizes the results and discusses the implic ations of the findings in relation to both intervention efforts and future research.

PAGE 42

29 CHAPTER II THEORY Despite the persistent view among the health sciences community that fatalistic attitudes and beliefs represent an irrational cognitive orientation held by ethnic minorities, more recent theories on the development and maintenance of fatalism emphasize the importance of considering how social and material barriers (e.g., po verty, discrimination, limited access to quality health services and treatment), contribute to its development and expression (Abraido Lanza et al., 2007; Espinosa de los Monteros & Gallo, 2013). This line of thinking also proposes that poor health outcome s experienced by vulnerable and disadvantaged populations may actually reinforce fatalistic beliefs about the utility of disease prevention and treatment behaviors (Abraido Lanza et al., 2007; Freeman, 1989). Consideration of the role that pervasive institutional barriers to health care may play in the creation and maintenance of fatalistic attitudes and health behavior is critical on among the poor, less educated, and uninsured, Latinos face an array of agency eroding obstacles to achieving equitable access to quality health care services. In light of their vulnerable status, I argue that, for Latinos faced with seemingly insurmount able institutional barriers to health care, fatalistic beliefs about cardiovascular health and illness are grounded in the recognition that while good health is v alued, barriers to equitable health care services make it difficult to achieve through personal effort alone (Espinosa de los Monteros & Gallo, 2013).

PAGE 43

30 A Faulty View: Fatalism as a De Contextualized Cognitive Trait of the Individual Within the public health literature, much of the work on fatalism and why people choose to forego health screenings, or a particular course of treatment that is known to be efficacious, has taken a primarily cognitive orientation (Drew & Schoenberg, 2011). Despite acknowledging that other factors (e.g., lack of health insurance) may play a role in whether one utilizes screening resources, researchers studying fatalism often attribute individual level fatalistic beliefs to shaping and directing prevention be havior. The focus on fatalism and cognitive mechanisms in the explanation of suboptimal health behaviors may be partly understood in terms of the shift in emphasis from acute and infectious diseases to chronic disorders as the major cause of morbidity and mortality in twentieth and twenty first century developed nations (Davison et al., 1992). Accompanying this shift was the increasing importance of disease prophylaxis as a central motivator in the everyday behaviors of the healthy population. According to Davison et al. (1992), t he most striking feature of this process has been the development of a strong, officially sponsored discourse, which emphasiz es the personal responsibility of the individual in the maintenance of their own health and the avoidance of chronic disease . As part of the growth of the prevention, health education, and health promotion movements, public health focused on educating the public about the risks of certain behaviors and the , the apparent failure of large numbers of the populace to follow the advice of a healthy lifestyle constituted a major problem for prevention science and was attributed to two major causes: 1) lack of knowledge and 2) the existence of an attitude that per ceives health as being primarily

PAGE 44

31 determined by factors outside the control of the individual, thus negating the relevance of personal behavior in the prevention and/or management of disease ( Davison et al., 1992). P revious research and critique demonstrat es that the locus of control trend in psychology has served to perpetuate the idea that belief in individual control is correct while belief in causal mechanisms external to the individual requires correction, usually in This typ e of analysis has led to the production of the idea that health promotion is involved in a battle for the hearts and minds of the population, a struggle between a modern belief in lifestyle and an atavistic culture of fatalism (Davison et al., 1992, p. 6 The differentiation of those who hold fatalistic attitudes and believe that health is governed by sources external to the individual from those who subscribe to the importance of lifestyle and individual control is replete with moralizing judgments. Individuals demonstrating an external source of control are perceived as ignorant, irrational, and in need of education (Straughan & Seow, 1998), while those with an internal source of control are viewed as having embraced modern values that align closely with good lifestyle choices (Drew and Schoenberg, 2011,). Indeed, the perception of where control resides (internal vs. external) and how this relates to health behavior has served as a defining characteristic of the fatalism construct, including the ethical basis on which people, and most often minorities, are judged. Locus of Control The locus of control construct is central to the very definition of fatalism and the way it is used to label people. The varying definitions of fatalism , as related to health, all have at their very core, the concept of lack of control over health status (Davison et al.

PAGE 45

32 1992). For example, individual control , a belief that some health issues are beyond human control . Research linking perceived control to health behavior can be traced to social learning theory and the work of Julian Rotter and Albert Bandura. Social Learning Theory ( R otter, 1954; 1982; Rotte r, Chance, & Phares, 1972) claims that the likelihood of a person engaging in a particular behavior is dependent upon two things: the person's expectancy that the behavio r will lead to a particular outcome (or reinforcement) in that situation and the value of the reinforcement to the person in that situation (Wallston 1997). According to Social Learning Theory, individuals develop generalized expectancies based upon previous experience and learning across a vari ety of life circumstances. Locus of control (Rotter, It refers to the person's belief as to whether control over valued outcomes is internal or external to the person . A person with an in ternal locus of control believes that consequences in life are the result of either some action (or set of actions) taken by the person, or of some individual characteristic (or set of characteristics) specific to that particular person . I nternal locus of control orientation is typically equated with a perception of control over outcomes . Conversely, perso ns with an external locus of con trol believe that life events are the result of forces outside of the person. Belief in an e xternal locus of control is mo st often associated with a perception of lack of control over The locus of control trend in psychology initiated the classification of individuals as either internally oriented, believing in personal control over ev ents in his/her life, or externally oriented, and thus fatalistic, believing that one has little control over life events,

PAGE 46

33 including health or the onset of disease (Davison et al. 1992). Health behavior researchers er to avoid ill health emphasize the individual and the belief in the possibility of control over health by personal action and behavior. In their aim to identify people as either leaning towards internal or external locus of control, the health promotion and prevention movements ultimately saw those who gave weight to personal control (Davison et al., 1992). Locus of control is a fundamental part of the fatal ism construct. Prohibitive social structures that deny socioeconomically vulnerable populations equitable access to health promoting resources may result in a loss of control or sense of powerlessness for those affected. In such circumstances, seeking out and using CVD screening resources may hold little value or benefit, particularly if use of screening services results in finding out that one has health problems that require medication or health resources that remain unaffordable and out of reach. Thus, l ocus of control is an essential part of the fatalism construct linking institutional barriers to health care to CVD screening behaviors. Self Efficacy and Models of Individual Health Behavior A related construct that has found broad appeal among health behavior researchers is self efficacy (Bandura, 1977 ). Self efficacy refers to a person's subjective estimation that he or she is capable of engaging in a particular action (or set of actions) in a situ ation (Bandura, 1977). Although similar to locus of control, self efficacy differs from the former in that it refers to control over a behavior rather an outcome (Wallston, 1997). A s originally conceived of by Bandura, self efficacy is highly behavio r al and situation specific. In other words, perc eived control over a behavio r in a certain situation does not necessar ily

PAGE 47

34 translate to other behavio rs or other situations (Wallston, 1997). For example, heart disease patients who are self efficacious with regard to participation in a rehabilitative exerc ise program may not experience the same level of self efficacy when faced with the responsibility of managing and adhering to medications prescribed by a physician (Katch & Mead, 2010). Despite the differences outlined above, some researchers have substituted self efficacy for locus of control as the major generalized expectancy construct (Wallston 1992). control is operationalized similarly to measures of self efficacy (Wallston 1997). More recently, Judge et al. (2002) have argued that self efficacy and locus of control are actually related concepts and may in fact measure the same higher order construct. Alternatively, Straughan and Seow (1998) demonstrate sup self efficacy is correlated with health promotion behavior by showing that those who are more fatalistic toward their health also have lower perceived self efficacy. In turn, the authors posit that self efficacy h as an important bearing on whether study participants adopt preventive health measures such as breast and cervical cancer screenings. Although originally advanced by Bandura (1977) and included as part of his Social Cognitive Theory (1986), self efficac y has been incorporated into multiple models that attempt to explain human behavior. Among these are the Health Belief Model (HBM), the Theory of Planned Behavior (TPB), and the Integrated Behavior Model (IBM) (Glanz, 2008). Work on the role of fatalism i n disease screening behavior has utilized both the Theory of Planned Behavior (Facione, 2002; Lopez McKee, 2008) and the Health Belief Model (Gorin, 2005) to explain the underuse of screening services by minorities.

PAGE 48

35 Originally developed in the 1950s by so cial psychologists attempting to explain the widespread failure of people to participate in disease detection and prevention programs, HBM relies heavily on cognitive theory and the role of subjective expectations held by individuals to explain health rela ted behavior (Glanz et al., 2008). The core constructs that make up HBM are perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action, and self two key assumptions : 1 . Individuals value avoiding illness/getting well and 2. E xpect that a specific health action may prevent illness (Glanz et al., 2008). But, the problem with using the HBM to model screening behaviors is that it minimizes the potential role of structural constraints in the underuse of health promoting o this study, Gorin subsumes fatalism under the HBM barriers measure, which in th is case is embarrassment. In doing this, the author perpetuates the idea that fatalism is simply informed by individual perceptions, cognitions and beliefs and is unr elated to socially and institutionally organized barriers. Analyzing fatalism under the HBM in this fashion limits the potential for uncovering the context specific influences on fatalism and its relationship to the underutilization of screening services b y Latino women. The Theory of Planned Behavior (TPB) has its roots in the Theory of Reasoned Action (TRA), both of which focus on individual motivational factors as the primary determinants of the performance of a specific behavior (Glanz et al., 2008). TRA and TPB both assume

PAGE 49

36 that behavioral intention, the key predictor of actual behavior, is informed by an surround it (Glanz et al., 2008). While TRA focuses on cognitive factors such as beliefs and values that determine motivation to perform a behavior, TPB extends TRA by adding perceived behavioral control as an additional explanatory construct. Lopez McKee et al. (2008) used TPB to guide their assessment of the level of fatalism in Latinas, categorized as regular versus infrequent mammography screeners. They were able to demonstrate that significant differences in control beliefs existed between the two groups , and fatalistic beliefs were significantly higher in the group characterized by infrequent use of screenings. Importantly, these differences existed among all low income study participants, indicating that fatalism towards health may be related to barriers beyond low SES . As with the HBM, the primary weakness of th e TPB is that it fails to adequately s assertion that behavioral attitudes, subjective norms, and perceived behavioral control lead to behavioral intention, and thus to act ual behavior, make no mention of how intentions, and actions. Leaving these types of variables unaccounted for seriously hinders uboptimal use of CVD screen ing services by Latinos. E ffort s to understand fatalism and its relationship to screening behavior must move beyond the study of subjective cognitions that individually focused explanatory models g ive p rimacy to . Rather, it is imperative that research incorporate s and fully account s for the influence of social and material conditions in an effort to better understand how

PAGE 50

37 fatalistic attitudes may be a reflection of the structural barriers to equitable health care that Latinos fac e. An Alternative View: Fatalism and its Social Origins While there is no unifying theory of fatalism in the social sciences literature , there is a long tradition of thinking about fatalism as a social rather than psychological phenomenon. In particular, recent analyses demonstrate the value of viewing fatalism in light of theories put forth by sociologist Emile Durkheim (Acevedo, 2005). In his classic work Suicide (1968), Durkheim offers two distinct, binary concepts (anomie/fatalism) that help us to better understand how over regulation by social structures can contribute to fatalistic beliefs. In his analysis of differing suicide rates among Protestants and Catholics, Durkheim argues that greater social control among Catholics results in lower suicide rates. direction, which results from dramatic social and economic upheaval. For example, extreme newfound wealth or economic ruin may result in dramatic changes to social roles and expectations such that the individual struggles to find meaning and where they fit into society. According to Durkheim, this experience of moral disorder and under regulation results in a constant state of disappointment and a heightened risk for suicide. Durk h individual is excessively regulate d such that futures are pitilessly blocked and passions violently choked by oppressive discipline (Durkheim 1968:276). According to Acevedo that results from a com bination of over regulation combined with a lack of exit option into the collective body in which the subject lacks the necessary voice and/or exit option to

PAGE 51

38 alter their social position, status , (Acevedo, 2005: 75). In this sense, structurally organized fatalism can result from feeling powerless when faced with prohibitive social structures, like poverty or institutional barriers to health care, which over regulate the lives of people and reduce o his/her own behalf. that human aspiration knows no bounds, fatalism refers to hopes so narrowed and diminished that life is itself a matter of i ndifference. It is in this sense that fatalism is applied here. I propose that for Latinos in the U.S, a persistent and inequitable lack of access to health care resources and perceived discrimination in the health care system has engendered feelings of fa talism towards CVD and the use of screening services designed to assist with the identification, prevention, and management of this disease . Reintroducing Social Context: Institutional Barriers and the Expression of Fatalistic Attitudes Although it is recogni zed that social and economic barriers are associated with the presence of fatalistic health beliefs and use of screening services, research on fatalism and health behavior has largely failed to explicitly account for the social and institutional factors speci fic to the presence of fatalism among Latinos. Social context is fundamental to the fatalism concept and must be taken into account in order to fully understand the reason some hold to the belief that they have little control over their health . Research that fails to address questions about how people internalize, reason about, and respond to institutional barriers to health care limits our understanding of how and why structural obstacles serve to influence ntion resources. Studying fatalistic attitudes towards CVD screening services among Latinos in context requires that

PAGE 52

39 we account for the relative contribution of institutional barriers to health care and fatalism to the suboptimal use of blood pressure and cholesterol screening services. However, doing this alone is insufficient to achieving a comprehensive understanding of the motivations To fully understand Latino related beliefs and dispositions toward CVD screening services, research must give voice to the lived experience of those that face prohibitive institutional barriers to health care. Doing so helps to shed light on the internal logic and rationale employed by those who [seemingly fail] to meet our expectations while also contributing to a better understand ing of the results of quantitative survey s when employed in a mixed methods design. Viewed in this way, f atalism may reflect a rational adaptation to persistent obstacles to health care experienced by Latinos in the United States (Abraido Lanza et al. 2007). The theo retical model proposed below , as Figure 1, was created specifically for this study. Figure 1. Theoretical Model Figure 1 illustrates the relationship between institutional barriers to care, CVD fatalism, and the suboptimal use of blood pressure and chol esterol screening services by Latinos proposed by this study. Institutional barriers to care are viewed to both directly,

PAGE 53

40 and indirectly affect Latinos use of blood pressure and cholesterol screenings. Indirect effects work through fatalism. In addition to proposing that specific barri ers to health care are directly and indirectly services (Specific Aim #1) , the relationships set forth in this model hypothesize that these barriers also infl uence the very beliefs and attitudes about whether one can control the onset and course of CVD (Specific Aim #2) and in turn, the use of the blood pressure and cholesterol screening services.

PAGE 54

41 CHAPTER III RESEARCH DESIGN AND METHODS A Mixed Methods Study To address the multiple lines of inquiry proposed by this study, a mixed methods approach was employed. As currently defined by Johnson, Onwuegbuzie and Turner (2005), m ixed methods research comprises investigation in which the researcher mixes or combines quantitative and qualitative research techniques, methods, approaches, concepts or language in a single study or set of related studies. There are three areas where mixed methods research is proposed to be superior to sing le quantitative or qualitative designs: (a) simultaneously address ing a range of both confirmatory and exploratory questions; (b) allow ing for better inferences ; and (c) provid ing the opportunity for a variety of divergent views to surface in the analytic process (Teddlie & Tashakkori, 2009). Mixed methods were especially important for this study, as solely quantitative approaches have failed to adequately explain the lower rates of screening in Latino populations. Used in isolation, quantitative methods have left fundamental questions about how people internalize, reason about, and respond to institutional barriers to health care unanswered. To f ully understand how institutional barriers impact the health beliefs and screening beh aviors of Latinos, this project employ ed the complementary strengths of qualitative methods . This approach enable d me to critically explore the impact contextual variable screening resources. Moreover, a dding qualitative methods p ermit ted me to both critically examine the meaning of quantitative results obtained while emphasizing the importance o f

PAGE 55

42 accounting for how institutional barriers may be incorporated into the lived experience, decision making processes and health behavior of this population . Goal and Objectives T he goal of this study was to reintroduce social context to the study of fatalism and screening behavior in an effort to more fully understand the relationship between institutional barriers to health care, fatalism towards CVD, and the use of cholesterol and blood pressure screening s ervices by Latinos in Denver, Colorado. More specifically, the research objectives were to: 1. Quantitatively examine the relative contributions of specific institutional barriers to health care ( e.g., lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care encounter), and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol) by Latino s in Denver, Colorado. 2. Qualitatively explore how institutional barriers to healt h care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latino s in Denver, Colorado . Rationale and Purpose The rational e for employing mix ed methods in the proposed study was twofold: participant enrichment and significance enhancement ( Collins, Onwuegbuzie, & Sutton, 2006) . Participant enrichment refers to the mixing of quantitative and qualitative techniques for the ratio nale of optimizing the sample ( Collins et al. 2006 ). In this regard, gathering quantitative demographic data , including responses to survey questions about

PAGE 56

43 institutional barriers to health care and fatalism , allow ed for the selection of an appropriate sample of Latino s with whom to explore qualitatively and in depth the relationship between institutional barriers to care , health beliefs and explanations for the suboptimal use of CVD screening services. The s econd rationale for mixing methods in the proposed study was significance enhancement. Significance enhancement refers to mixing quantitative and qualitative 2006) . With this in mind, the proposed study employ ed a sequential mixed analysis in which the qualitative data analysis phase follow ed the quantitative data analysis phase (Collins et al., 2006). More specifically, the qualitative data w as used to enhance the interpretation of findings from the quantitative analyses but also to critically explore the social context within which CVD health related beliefs, decisions and screening behavior are formed and expressed. Qualitative methods were an essential part of this mixed methods study . In depth exploration of how social condition s serve to organize the ways in which people perceive and reason about CVD and the use of screening services move d beyond quantitative only approaches to allow for a more detailed understanding of how institutional barriers become a central part of the very process by which one formulates health related attitudes and behavior. Sampling Design According to Onwuegbuzie and Collins (2007), a distinction betwee n sampling design and sampling scheme can be made. Sampling schemes are specific strategies used to select the units for analysis , while sampling design refers to the framework within which the sampling takes place, including the number and types of sampli ng schemes that will be

PAGE 57

44 used (Onwuegbuzie & Collins, 2007). Given this distinction, both sampling design and sampling scheme for the proposed study are described below. Sampling Design The proposed study employ ed a sequential design using neste d samples (Onwuegbuzie & Collins, 2007). Sequential designs involve the selection of units of analysis through the sequential use of sampling strategies such that information from the first sample is used to draw the second sample (Teddlie & Tashakkori, 20 09). In this study, the qualitative sample w as derived from the quantitative sample. This link between the two samples is referred to as a nested relationship ( Onwuegbuzie & Collins, 2007) . As shown in Figure 2 , in this type of relationship the sample selected for one phase of the study represents a s ubset of the sample chosen for the other phase of the investigation ( Onwuegbuzie & Collins, 2007). Figure 2. Sequential Sampling Design Using Nested Samples Sampling Scheme A multi stage purposeful sampling strategy w as utilized for this stu dy individuals representing a sample in two or more stages in which all stages reflect

PAGE 58

45 sp ecifically, a purposive, non random criterion sampling scheme w as used for the quantitative portion of the study while a purposive maximum variation sampling scheme w as used for the qualitative component (Onwuegbuzie & Collins, 2007). Criterion sampling in volves choosing settings, groups, and/or individuals because they represent one or more criteria (Onwuegbuzie & Collins, 2007). For the present study, inclusion and exclusion criteria guide d the first stage of sampling. Inclusion criteria consist ed of: (a) self identified Latino/a ; (b) aged forty to seventy five years 2 , and (c) self identified resident of Denver County, Colorado. Exclusion criteria include d (a) previous or current participation in a heart disease or diabetes prevention or education program and (b) self reported heart disease. Participation in a heart disease prevention or education program often entails receiving information about the benefi ts of blood pressure and cholesterol screening or the actual tests. Those who have been diagno sed with heart disease and self report as such are also likely to have been provided blood pressure and cholesterol counseling. These particular populations were listed under exclusion criteria in an effort to minimize the information bias they may introduce to the sample . Setting This research took place in Denver County, Colorado. Currently, 34 percent of Denver County residents identify as Latino/Hispanic (Calanan et al., 2011). Table 1 details some of the characteristics of Denver County Latinos relevant to this study . 2 Age r ange is based on screening recommendations put forth by ATP III and JNC 7, including the presence of elevated risk factors (e.g., cholesterol, Type II diabetes, obesity) common among Latino youths that predispose the Latino population to early age onset an d premature death from CVD.

PAGE 59

46 Table 1. Select Characteristics of Denver County Latino Population Characteristic Percent Race/Ethnicity Latino/Hispanic 34% Told by medical professional have high cholesterol 39% Report having cholesterol checked in past 5 years 53% Told by medical professional have high blood pressure 15% Diagnosed hypertensive and taking medication 65% Report health as fair/poor 27% Report no health insurance 40% Calanan et al., 2011 Phase One: Quantitative Sample The P hase O ne sample w as drawn using a venue based, time space sampling approach (Muhib et al., 2001). This approach recruits respondents from places and at times where they would be reasonably expected to gather. Utilizing previously collected formative data detailing both Denve r County venue based geography and usage patterns (Pineda Ford, 2011) , the proposed study systematically sample d from community based sites at times known to serve or contain high density Latino populations. These venues include d faith based organizations, housing complexes, recreation centers, service based agencies, schools, and neighborhood business centers such as grocery outlets and restaurants. Although the sampling scheme remain ed purposive, a venue based, time space sampling approach allow ed me to construct a sample with known properties, make statistical inferences to the larger population that attends chosen venues, and theorize about biases that may limit generalization of find ings to the target population (Muhib et al.,

PAGE 60

47 2001). A priori power analyses indicate d that a sample size of 300 w ould provide adequate power (0.80) at the alpha (0.05) level for the logistic regression statistic (Hsieh, Block, & Larson, 1998). Phase Two: Qualitative Sample A purposive maximum variation sampling scheme w as employed to draw the sample for the P hase T wo qualitative portion of the study . Maximum variation sampling involves choosing settings, groups, and/or individuals to maximize the r ange of perspectives investigated (Onwuegbuzie & Collins, 2007). As shown in Table 2 , four groups of interest comprise d the qualitative sample in an effort to explore both confirmatory and contradictory cases regarding the impact of institutional barriers to health care on endorsement of fatalistic attitudes and beliefs. Table 2. Phase Two: Sampling Domains for Qualitative Interviews Group 1 High IBHC High Fatalism Group 2 High IBHC Low Fatalism Group 3 Low IBHC High Fatalism Group 4 Low IBHC Low Fatalism Based upon their answers to P hase O ne survey q u estions, individuals w ere placed into one of the Table 2 groups. Of those willing to participate, 49 percent were allocated to Group One, 19 percent to Group Two, 16 percent to Group Th ree and 15 percent to Group Four. A random sample w as then drawn from each group pool to comprise the final P hase T wo sample. The size of the qualitative sample w as determined through evaluation of information redundancy (Lincoln & Guba, 1985) . That is, sa mpling w as terminated when no new information was identified from sampled units.

PAGE 61

48 Group s one and four provide d support for my hypothesis regarding institutional barriers to health care and fatalism. These two groups help ed to explain how and why the level of exposure to institutional barriers to health care results in the expression of a disposition consistent wit h fatalism toward CVD. Groups two and three, however, present ed the opportunity to discover and explore explanations that run contrary to the hypotheses proposed. Remaining open to identifying and understanding alternate explanations was an important part of the exploratory process and served to aid in achieving a more holistic and contextually dep endent view of the relationships of interest. High levels of institutional barriers to health care w ere calculated as confirmation of an individual experiencing two or more barriers represented by the independent variables of interest ( e . g. , lack of health insurance and no usual source of health care). Low levels of institutional barriers to health care w ere defined as experience with no or only one of the indepe ndent variables of interest. Level of fatalism w as determined by calculation of overall score on the validated instrument , ranging from 0 20 . Determination of high and low score s w as based on the distribution of the data . Due to the skewed distribution, median scores and above were considered high with scores below the median considered low. Fatalism was dichotomize d for three purposes. First, in order to facilitate modeling as an independent, dependent and mediating variab le for hypotheses 1b, 1c and 1d respectively, dichotomization enabled direct comparison between those who agreed or strongly agreed with scale items to those who disagreed or strongly disagreed with scale items . Second, dichotomization contributed to the p reserv ation of model power (0.80) at the alpha (0.05) level for the logistic regression statistic. Third, defining the fatalism

PAGE 62

49 variable in this manner allowed for the creation of a clear and consistent sampling frame from which participants were selected for Phase Two qualitative interviews. Research Design As conceptualized by Leech and Onwuegbuzie (2006), mixed methods research designs can be represented as a function of three dimensions; level of mixing, time orientation, and emphasis of approaches. Lev el of mixing refers to whether the study is partially or fully mixed. According to Leech and Onwuegbuzie (2006) , t he major difference between partially mixed methods and fully mixed methods is that fully mixed methods studies involv e mixing quantitative and qualitative techniques within one or more stages of the research process . By comparison, partially mixed methods research typically conducts the quantitative and qualitative phases either concurrently or sequentially in their entirety befo re mix ing occurs at the data interpretation stage. Time orientation indicates whether the quantitative and qualitative phases of the research occur at the same time or concurrently, or whether they occur one after the other as in sequen ce (Leech & Onwuegb uzie , 2006). Lastly, emphasis of approach relates to whether the quantitative and qualitative phases of the study hold the same weight with respect to answering the research questions (Leech & Onwuegbuzie , 2006). If the two phases are equally emphasized th en this aspect of the design is known as equal status. If one component is given higher priority than the other then the design is noted as dominant status. Following this three dimensional typology, the proposed study constitute d a partially mixed sequen tial equal status design (Leech & Onwuegbuzie, 2006). Specifically, Phase O ne quantitative data w as collected and analyzed first, followed by qualitative data

PAGE 63

50 in P hase T wo. As demonstrated in Figure 3 ., the proposed study is considered partially mixed beca use the two datasets w ere only mixed during the interpretation phase. Moreover, the quantitative and qualitative data w ere equally emphasized thus the designation of equal status. Figure 3. Partially Mixed Sequential Equal Status Design Data Collection Prior to data collection, approval of the study protocol w as obtained from The Colorado Multiple Institutional Review Board (COMIRB). Data collection procedures w ere informed by the two phase sequential design and comprise d a questionna ire strategy for collecting quantitative data with an interview strategy for collecting qualitative data. For the proposed study, P hase O ne use d a survey instrument consisting of demographic questions (including institutional barriers to health care) valid ated multi item measures of fatalism , and questions pertaining to use of blood pressure and cholesterol testing services. This survey w as administered to Latino s meeting the inclusion criteria. Phase T wo consist ed of semi structured in depth interviews wit h a subsample of those completing the quantitative survey in P hase O ne. Phase One Instruments A validated scale (Shen, Condit & Wright, 2009; see Appendix A .) w as used to collect measures of fatalism. The authors of this s cale presented all questions contained in the measure to a multi cultural Community Advisory Board (CAB) containing Latinos for

PAGE 64

51 comments. Board members reviewed and approved items for level of clarity (including readability), cultural appropriateness, and cultural inclus ion. Scale validation was conducted with a nationally representative sample with two 200 oversamples from Latino respondents. Confirmatory factor analyses demonstrate that a second order single factor model provides a good fit to the scale items. Factor lo adings ranged from 0.62 to 0.74. In addition, reliability data (alpha = 0.88) and correlation analyses used to investigate the relationship between three external variables (genetic determinism, perceived benefits of lifestyle changes, and intention to eng age in health behaviors) and the scale dimensions reliability and construct validity (Shen et al., 2009). Level of f atalism is calculated by summing each of the responses to the scale s twenty questions into a composite score. Higher scores indicate a stronger tendency to endorse fatalistic beliefs. Validated measures from t he Behavioral Risk Factor Surveillance System Survey (BRFSS) w ere used to assess respondent demographic s and CVD screening and prevention behavior (CDC, 2011 b ). The BRFSS survey measures of interest, including blood pressure and cholesterol screening behavior, demonstrate reliability ( kappa=0.82) ( Brownson, Jackson Thompson, Wilkerson, & Kiani, 1994) While the validity for blood pressure and cholesterol were lower (BP sensitivity=40%; cholesterol sensitivity=44% ; BP specificity =87; cholesterol specificity =85 for women and 75 for men), the measures in question remain the standard for assessing self report ed hypertension and hypercholesterolemia and continue to be used by The Centers for Disease Control and Prevention (Bowlin et al ., 1993). Dependent variables include d : (a) blood pressure screening in the past two years (yes/no) and (b) cholesterol sc reening in the past two

PAGE 65

52 years (yes/no). Independent variables include d (a) fatalism ( dichotomous variable , high/low ); (b) health insurance (yes/no); (c) regular source of health care (yes/no), and (d) perceived discrimination in a health care encounter in the past two years (yes/no). Covariates include d age, gender, language preference, education, income, place of birth, length of residence in U.S. and time since last doctor visit. Univariate analyses and previous research in the area of interest informed t he operationalization of all covariates. Age was defined as a categorical variable with consideration of the inclusion criteria. Age categories included (40yrs 49yrs; 50yrs 59yrs; 60yrs and older). Gender was dichotomized as female and male. Due to the inclusion criteria and population of interest, language was dichotomized as language of preference Spanish or English. Due to the distribution of the data, e ducation was dichotomized as less than a hig h school diploma or high school diploma/GED or more. Family income was also dichotomized based on univariate analyses and distribution of the data. As a result of 75 percent of the sample demonstrating income of $19,999 per year or below , a decision was ma de to define family income as $19,999 or below annually and $20,000 or above annually. Given the sociodemographic composition of the Latino population in Colorado and distribution of responses, place of birth was operationalized as Mexico or USA. Due to 77 percent of the sample indicating that they have lived in the U.S. more than 10 years, l ength of residence in the U.S. was dichotomized as 10 years or less and more than 10 years. Lastly, based on distribution of the data, time since last doctor visit was defined as one year ago or less, more than one year ago and never. According to Flores decreasing since 2014 with the share coming from Mexico decreasing since 2007. Recent

PAGE 66

53 data from The Colorado Health Institute demonstrate that 76 percent of all Coloradans without health insurance are documented and legal residents. Due to these data and concern by project research assistants with compromising trust and participation from c ommunity members, a decision was made to exclude immigration status as part of this research. Phase Two Instrument Phase T wo data collection employ ed an interview guide approach (Patton, 2002). This approach relie s on using a list of questions or topics t hat allow the interviewer to explore, probe, and inquire of the interviewee such that information about the topic will be elucidated (Patton , 2002). The interview guide help ed make the interviewing process systematic, efficient, and focused by delimiting i n advance the issues or topics that w ere explored (Patton, 2002). Interview questions focus ed on attitudes and perceptions towards institutional barriers to health care , fatalism towards CVD and use of CVD related screening services (See Appendix B.) . Dat a Collection Procedures Bilingual members of the target population experienced in community based research w ere hired and trained as research assistants . Research assistants collect ed P hase O ne data and assist ed me with the collection of P hase T wo data. Following study protocols, research assistants visit ed venue locations at pre specified times to conduct P hase O ne recruitment. They screen ed potential respondents for eligibility, set appointments for face to face survey delivery at time s and locati on s convenient for respondent s and obtain ed informed consen t (see Appendix C .) . Upon completion of each P hase O ne survey, research

PAGE 67

54 assistants inform ed respondents about P hase T wo interviews and solicit ed interest and contact information. Study res pondents providing contact information for participation in P hase T wo w ere allocated to one of the four groups outlined in Table 2 according to their responses to P hase O ne survey questions. A random sample w as then drawn from each of the four groups to comprise the interviewees for P hase T wo qualitative inquiry. Data collection proceed ed until informational redundancy was achieved (Lincoln & Guba, 1985). All interviews w ere audio recorded with participant permi ssion and transcribed verbatim . Data Analysis Phase O ne data w ere formatted and entered into a Microsoft Excel spreadsheet. Data w ere then imported into S PSS v. 21 for analysis ( IBM Corp. , 201 2 ). During initial analyses, the data w ere checked for coding errors and missing data by running descriptive statistics. Outliers w ere identified at this time. Outliers w ere defined as scores on variables greater than 3.5 standard deviations from the mean in either direction. D escriptive data anal yses characterizing participant level of institutional barriers to health care, fatalism, and use of CVD screening and treatment services in P hase O ne inform ed the composition of varied groups (Table 2 ) of interest for qualitative data collection and analy sis in P hase T wo. Binomial logistic regression was used to address Specific Aim 1: Quantitatively examine the relative contributions of specific institutional barriers to health care ( e.g., lack of access to health insurance, no usual source of health care , and perceived discrimination in the health care encounter), and fatalism towards the use of CVD screening services (e.g., blood pressure and cholesterol) by Latino s in Denver, Colorado. More specifically, mediation pathway analysis was used to evaluate H ypothesis 1d: Institutional barriers to

PAGE 68

55 care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism . Unlike ordinary least squares regression, logistic regression: a) does not assume linearity of relationship between the raw values of the independent variables and raw values of the dependent variable; b) does not require normally distributed variables; b) does not assume homoscedasticity; c) requires that observations be independent, and d) requires that the independent variables be linearly related to the logit of the dependent (Agresti & Finlay, 2009) . The nature and strength of relationship s between independent and dependent variables w ere evaluated usi ng odds ratios and 95 percent confidence intervals (Agresti & Finlay, 2009). Phase T wo qualitative data w ere formatted in Microsoft Word and entered into ATLAS.ti for management and analysis (ATLAS.ti., 2012). Data w ere analyzed using constant comparison analysis (Glaser & Strauss, 1967) . Constant comparison analysis involves identifying themes in the data and was undertaken deductively and inductively (Leech & Onwuegbuzie, 2007). Deductive coding procedures utilize d an existing framework consisting of established a priori codes developed from the research questions and content specific questions contained on the questionnaire (Patton, 2002). In this form of analysis, the researcher reads through the data looking for mean ingful segments that represent or relate to the a priori themes (Patton, 2002). Alternately, inductive procedures lead to codes and themes that are grounded in the data and not given a priori (Teddlie & established beforehand. For the proposed study, inductive coding procedures occur red first and were followed by deductive analyses. This strategy allow ed for immersion in the

PAGE 69

56 interrelationships first. Then, using deductive analysis, emphasi s was placed on theory driven and established concepts related to the research questions and to confirm exploratory findings (Patton, 2002). Ordering the an alysis in this manner serve d to prioritize examination of the data less encumbered by theoretical assumptions and thus l eft me in a better position to uncover critical details that help ed to explain the formation and maintenance of fatalism among Latinos i ncluding why they do or do not use CVD screening services. T he qualitative portion of this study used member checking procedures to increase the rigor and trustworthiness of the findings (Leech & Onwuegbuzie , 2007). Member checking refers to the research er showing the study findings or interpretations to the study participants , in this case a select number of participants, in order to assess the accuracy of the work (Leech & Onwuegbuzie, 2007). Data Validation Onwuegbuzie and Johnson (2006) reference da ta validation from three perspectives: quantitative research, qualitative research, and mixed methods research. Given that mixed methods research involves combining complementary strengths and non overlapping weaknesses of quantitative and qualitative rese arch, assessing the validity of study findings is critical. Teddlie and Tashakkori (2009) discuss four types of validity in quantitative research: construct validity, internal validity, external validity, and statistical conclusion validity. The proposed study utilize d previously validated instruments to collect quantitative data in P ha se O ne. As a result of previous testing, these instruments carry established measures of validity and reliability, and as such, their psychometric properties

PAGE 70

57 are known. Thus, some of the potential threats to construct validity w ere minimized. Onwuegbuzie and Johnson (2006) identify a multitude of threats to both internal and external validity across the research design/data collection, data analysis, and/or data interpretation stages of the quantitative research process. The classification of these threats , termed the Quantitative Legitimation Model, offers a comprehensive conceptualization of risks for researchers to consider (Onwuegbuzie & Johnson, 2006). This model serve d as a guide to ensure proper mitigation of potential threats to internal and externa l validity during P hase O ne of the study. The use of reliable measures, a priori power analyses, appropriate statistical procedures and careful assessment of statistical test assumptions w ere employed to enhance statistical conclusion validity. The propos ed study dr ew Legitimation Model to review the threats to internal and external credibility during the design/data collection, data analysis, and data interpretation steps of P hase T wo qualitative work. Speci fically, triangulation techniques and member checking w ere used to enhance credibility of the qualitative findings. Lastly, legitimation of the mixed methods (Onwuegbuzie & Johnson, 2006) proceed ed by attending to each of the following: 1. Sample Integration: Although this study employ ed purposive sampling techniques in both phases, careful consideration w as given to both the relationship between the two samples and the data in an effort to yield quali ty meta inferences. 2. Inside Outside: This study s ought to maintain a balanced perspective when collecting, analyzing and interpreting the collective data. Care w as taken to

PAGE 71

58 ensure that the data accurately represent ed member s of the group under study in the data review and integration process. In addition, peer review of data interpretation and study conclusions w ere conducted by research assistants ( RAs ) and perspective. 3. Weakness Minimization: Quantitative data inform ed the development of the qualitative interview instrument. The qualitative data w ere used to enhance the interpretation of the quantitative findings and answer questions pertinent to Specific Aim 2 . 4. Sequential: This study utilize d a sequential design such that the qualitative phase follow ed the quantitative phase. If the sequence were reversed, qualitative investigation could have potentially bias ed answers to the quantitative questions due to in depth discussion about the study phenomenon. 5. Conversion: This study did not convert data from one form to another. 6. Paradigmatic Mixing: This study dr ew upon a pragmatic approach towards investigation of the phenomenon of interest and therefore consider ed t he qualitative and quantitative paradigms as complementary rather than competitive. 7. Commensurability: A rejection of Khunian incommensurability and appreciation of both quantitative and qualitative thinking guide d this study. There were no anticipated diff iculties with switching back and forth between the two paradigms.

PAGE 72

59 8. Multiple Validities: All relevant quantitative, qualitative, and mixed methods legitimations w ere addressed so that strong meta inferences could be made. 9. Political: This research s ought pra ctical results that w ould contribute to understanding the issue of interest and lead to the identification of real world solutions and thus strive d to balance competing objectives where necessary. Data Integration and Interpretation As previously stated, data integration occur red at the interpretation phase with equal emphasis placed on both quantitative and qualitative data. The intent of this sequential mixed model design (Teddlie & Tashakkori, 2009) was to use the qualitative data to both enhance the interpretation of significant quantitative findings and to develop meta inferences by combining inferences drawn from the quantitative data with those drawn from the qualitative data . Inferences made from the quantitative data include d det ailed explanation of both statistical and practical significance (Leech & Onwuegbuzie, 2010). Inferences developed from qu alitative findings prioritize d the voice and perspectives of study participants and w ere presented together with those obtained from t he quantitative portion of the study (Leech & Onwuegbuzie, 2010). As a result, critical insight w as gained and meaningful conclusions drawn regarding the relationship between institutional barriers to health care, fatalism towards CVD, and the suboptimal u se of blood pressure and cholesterol screening services by Latino s in Denver, Colorado.

PAGE 73

60 CHAPTER IV QUANTITATIVE ANALYSES AND RESULTS Despite the recognition that institutional barriers to health care in the form of lack of access to health insurance, no usual source of health care, and perceived discrimination in the health care system use of screening service s , research continues to emphasize individual level fatalistic attitudes and beliefs in the explanation of In Chapter 2, I discuss the importance of reintroducing contextual variables to the study of fatalis m and health screening behavior among Latinos . R ather than viewing fatalism as simply a cognitive trait of individuals , I argue that fatalism is partly a byproduct of social processes that serve to hinder the health related agency of Latino populations. U nderstand ing the context in which screening behavior occurs requires account ing f or the relative contribution of instituti onal barriers to health care and fatalism to the suboptimal use of blood pressure and cholesterol screening services. In this chapter, I focus specifically on the results of quantitative analyses used to test each of four hypotheses addressing Specific Aim 1. Review of Methods I measur ed fatalism using a validated instrument developed by Shen, Condit & Wright, ( 2009 ). Validated measures from t he Behavioral Risk Factor Surveillan ce System Survey (BRFSS) were used to assess respondent demographics and CVD s creening behavior. Statistical methods used in this chapter include descriptive statistics, binomial logistic regression and mediation pathway analysis using the PROCESS macro in SPSS v. 25 (Hayes, 2018 ) . Receipt of blood pressure screening in the past two years and receipt of cholester ol screening in the past 2 years were the dependent variables of interest for

PAGE 74

61 hypotheses 1a, 1b and 1d. Fatalism was the dependent variable of interest for hypothesis 1c. Perceived discrimination in the health care encounter, originally included as one of three specific institutional barriers to health care in Specific Aim 1, is not included in the logistic regression model results presented. Due to insufficient sample size a decision was made to drop this variable from the analyses in order to preserve mod el power (0.80) at the alpha (0.05) level for the logistic regression statistic . Only twenty one individuals (5%) in the analysis cohort (n = 386) reported exposure to perceived discrimination . Sample Characteristics Table 3 describes the sample socio demographic characteristics. Participants consisted primarily of female s (81%) in their forties (65%) with less than a high school diploma (66%). A majority of participants have resided in the United States Table 3 . Participant Characteristics Demographic Characteristics Total ( N =386) Median (range) or % Age (years) 47.0 yrs (38.0 yrs ) Female 81.1 % Language preference (Spanish) 94.8 % Less than high school diploma 65.8 % Family income (less than $19,999 yearly) 74.4 % Length of residence in US (more than 10 yrs) 76.9 % Time since last MD visit (one year or less) 75.5 % Have health insurance 26.7 % Have usual source of health care 70.7 %

PAGE 75

62 BP Screening in past 2 yrs 55.7 % Cholesterol Screening in past 2 yrs 43.8 % Fatalism 2.3 points (4.0 points ) for more than 10 years, prefer to speak Spanish (95%) and have been to see a health care provider in the past year. Among sample participants, only 44 percent indicate they have had a cholesterol test in the past two years. While most participants (73%) report having no health insurance, 70 percent say they have a usual source of health care. With regard to fatalism, careful examination of the results from univariate an alyses was conducted to ensure accurate assessment of measures of central tendency. O n a five point Likert sc ale using strongly d isagree (1) to strongly agree (5), indicating level of endorsement of statements indicating fatalistic beliefs , mean fatalism score was 2.37 + 0 .97. Median score was 2.3 with a range of 4.0. In the following section I report results from logist ic regression analys es employed to test each of the four hypotheses presented under Specific Aim 1 . I first report results from Hypothesis 1a : Institutional barriers to health care resources will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. Institutional Barriers to Health Care and Cholesterol Screening A binary logistic regression was performed to ascertain the relationship between health insurance status and having a usual source of health care on the likelihoo d of participants obtaining cholesterol screening. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic

PAGE 76

63 regression model was significant, 2 = 97.14, p < .005. The mod el explained 31% of the variance (Nagelkerke R 2 ) in screening for cholesterol and correctly classified 71.2% of cases. As shown in Table 4 , those with a usual source of health care had greater than three times the odds of obtaining cholesterol screening co mpared to those without a usual source of health care (OR 3.35, 95% CI : 1.77 6.32) . Unexpectedly, h aving health insurance was not related to obtaining screening for cholesterol when compared to not having health insurance after holding all other variables constant. Those of advanced age had an 89% (50 59 yrs; OR 1.89, 95% CI : 1.02 3.50) and more than two and a half times (60 yrs and over; OR 2.74, 95% CI : 1.37 5.47) greater odds of obtaining a cholesterol test compared to younger adults (40 4 9 yrs). Compared to males, females had 98% greater odds of obtaining a cholesterol test (OR 1.98, CI : 1.03 3.79). Also, those with a family income of $19,999 or less had greater than three times the odds of obtaining a cholesterol test compared to those with a family income of $20,000 or greater (OR 3.32, CI : 1.80 6.62). As expected, increasing l ength of time s ince last visit to a doctor resulted in a significantly l ower likelihood of obtain ing a cholesterol test in the past two years. The results for u sual source of health care and use of cholesterol screening demonstrate support for hypothesis 1a : Institutional barriers to health care resources will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. Howev er, the results for health insurance do not. Discussion of this anomaly occurs under the section detailing qualitative analyses and results. Table 4 . Binary Logistic Regression Results for the Odds of Using Cholesterol Screening when Participants Have Hea lth Insurance and Have a Usual Source of Health Care (USOC).

PAGE 77

64 Participant Characteristics OR (95% CI) Have health insurance 1.59 (0.90 2.78) Have USOC 3.35 *** (1.77 3.62) Age (40 49 yrs) Reference group Age (50 59) 1.89 * (1.02 3.50) Age (60 and over) 2.74 ** (1.37 5.47) Time since last doctor visit (one year or less) Reference group Time since las t doctor visit (more than one year ) 0.48 * (0.25 0.91) Time since las t doctor visit (never ) 0.11 * (0.01 0.91) Gender (female) 1.98 * (1.03 3.79) Length of time residing in US (10 years or less) 1.45 (0.79 2.63) Education (less than a high school diploma) 1.43 (0.86 2.37) Family income ($19,999 or less) 3.32 *** (1.80 6.12) * p<0.05 ** p< 0 .01 ***p< 0 .001 Institutional Barriers to Health Care and Blood Pressure Screening A binary logistic regression was performed to ascertain the effects of health insurance status and having a usual source of health care on the likelihood of participants obtaining blood pressure screening. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, 2 = 115.75, p < .05. The model explained 36 .0% of the variance (Nagelkerke R 2 ) in screening for blood pressure and correctly classified 74% of cases. As shown in Table 5 below, t hose with health insurance had significantly greater

PAGE 78

65 odds of obtaining screening for blood pressure (OR 2.30, 95% CI : 1.2 3 4.30) when compared to those without health insurance after controlling for all other variables in the model . Those with a usual source of health care were five times more likely to obtain screening for blood pressure (OR 4.97, 95% CI : 2.71 9.10) whe n compared to those without a usual source of health care after holding all other variables constant. Those with less than a high school diploma had 78% greater odds of obtaining a blood pressure test compared to those with a high school diploma or more (O R 1.78, CI : 1.05 3.01). Also, participants with a family income of $19,999 or less had greater than four times the odds of obtaining a blood pressure test compared to those with a family income of $20,000 or greater (OR 4.73, CI : 2.32 9.63). These resu lts offer support for hypothesis 1a by demonstrating that a lack of institutional barriers to health care r elate to significantly greater odds of using blood pressure screening services. Table 5 . Binary Logistic Regression Results for the Odds of Using Blo od Pressure Screening when Participants Have Health Insurance and Have a Usual Source of Health Care (USOC). Participant Characteristics OR (95% CI) Have health insurance 2.29** (1.23 4 . 30 ) Have USOC 4.97 *** (2.72 9.10 ) Age (40 49 yrs) Reference group Age (50 59) 1.93 (0.99 3.76 ) Age (60 and over) 1.54 (0.76 3.11 ) Time since last doctor visit (one year or less) Reference group Time since las t doctor visit (more than one year ) 0.89 (0.47 1.70 )

PAGE 79

66 Time since las t doctor visit (never ) 0.62 (0.18 2.10 ) Gender (female) 1.29 (0.67 2.51 ) Length of time residing in US (10 years or less) 1.54 (0.85 2.77 ) Education (less than a high school diploma) 1.78 * (1.05 3.01 ) Family income ($19,999 or less) 4.73 *** (2.32 9.63 ) * p<0.05 ** p< 0 .01 ***p< 0 .001 In the next section I report results from logistic regression analyses employed to test Hypothesis 1b : Fatalism will demonstrate an inverse relationship with use of cholesterol and blood pressure screening services. I first report results on the relationship between fatalism and blood pressure screening followed by results on the relationship between fatalism and use of cholesterol screening. Fatalism and Blood Pressu re Screening A binary logistic regression was performed to ascertain the relationship between fatalism and the likelihood of participants obtaining blood pressure screening. Model covariates included health insurance, usual source of health care, age, tim e since last doctor visit, gender, length of residence in the US, education and family income. The logistic regression model was significant, 2 = 115.08, p < .05. The model explained 36% of the variance (Nagelkerke R 2 ) in screening for blood pressure and correctly classified 74% of cases. However, a s demonstrated in Table 6 , there is no significant difference in the likelihood of obtaining screening for blood pressure between those demonstrating a low level o f fatalism and those who demonstrate a high level of fatalism after controlling for all other variables. However, as expected, those with health insurance had greater than twice

PAGE 80

67 the odds of obtaining a blood pressure screening compared to those without health insurance (OR 2.26, CI : 1.21 4 .24). When compared to those without a usual source of health care, those who have a usual source of health care have almost five times greater odds of obtaining a blood pressure test (OR 4.96, CI : 2.71 9.01). Participants with less than a high school di ploma had 78% greater odds of obtaining a blood pressure test compared to those with a high school diploma or more (OR 1.78, CI : 1.05 3.01) Lastly, those with a family income of $19,999 or less had an almost five times greater odds of obtaining a blood p ressure test when compared to those with a family income of $20,000 or more (OR 4.76, CI : 2.33 9.70). These results do not offer support for hypothesis 1b. Table 6 . Binary Logistic Regression Results for the Odds of Using Blood Pressure Screening based u pon Level of Fatalism. Participant Characteristics OR (95% CI) Have health insurance 2.26 * (1.21 4.24) Have USOC 4.96 *** (2.71 9.01) Age (40 49 yrs) Reference group Age (50 59) 1.93 (0.98 3.77) Age (60 and over) 1.54 (0.76 3.11) Time since last doctor visit (one year or less) Reference group Time since las t doctor visit (more than one year ) 0.89 (0.47 1.70) Time since las t doctor visit (never ) 0.62 (0.18 2.10) Gender (female) 1.28 (0.66 2.50 ) Length of time residing in US (10 years or less) 1.54 (0.85 2.77) Education (less than a high school diploma) 1.78 * (1.05 3.01)

PAGE 81

68 Family income ($19,999 or less) 4.76 *** (2.33 9.70 ) Fatalism (high) 0.93 (0.56 1.54) * p<0.05 ** p< 0 .01 ***p< 0 .001 Fatalism and Cholesterol Screening A binary logistic regression was performed to ascertain the effects of fatalism status on the likelihood of participants obtaining cholesterol screening. Model covariates included health insurance, usual source of health care, age, time since last doctor visit, ge nder, length of residence in the US, education and family income. The logistic regression model was significant, 2 = 101.11, p < .05. The model explained 32% of the variance (Nagelkerke R 2 ) in screening for cholesterol and correctly classified 71% of case s. As shown in Table 7, a h igh level of fatalism is significantly related to a 42% lower odds of obtaining cholesterol screening compared to low level of fatalism after considering all other control variables (OR .58, 95% CI : 0.35 0.95) . Having a usual source of health care results in more than three times the odds of obtaining a cholesterol tes t when compared to not having a usual source of health care (OR 3.43, CI : 1.81 6.50). Those of advanced age had twice (50 59 yrs; OR 2.03, CI : 1.07 3.82) and three times (60 yrs and over; OR 2.99, CI : 1.48 6.04) greater odds of obtaining a choleste rol test compared to younger adults (40 49 yrs). When compared to those whose last doctor visit was within one year or less or more than one year ago, those who have never visited a doctor were 90% less likely to have obtained a cholesterol test (OR 0.09 , CI : 0.01 0.76). Females were 85% more likely to have obtained a cholesterol test compared to males (OR 1.85, CI : 0.95 3.56). Those with a family income of $19,999 or less had more than three times the odds of obtaining a

PAGE 82

69 cholesterol test compared to those with a family income of $20,000 or more (OR 3.41, CI : 1.83 6.35). In contrast to the results obtained with blood pressure, these results offer support for hypothesis 1b. Table 7 . Binary Logistic Regression Results for the Odds of Using Cholesterol Screening based upon Level of Fatalism. Participant Characteristics OR (95% CI) Have health insurance 1.51 (0.85 2.69) Have USOC 3.43 *** (1.81 6.50) Age (40 49 yrs) Reference group Age (50 59) 2.03 * (1.07 3.82) Age (60 and over) 2.99 * (1.48 6.04) Time since last doctor visit (one year or less) Reference group Time since last doctor visit ( more than one year ) 0.49 (0.25 0.92 ) Time since last doctor visit ( never ) 0.09 * (0.01 0.76 ) Gender (female) 1.85 * (0.95 3.56 ) Length of time residing in US (10 years or less) 1.46 (0.80 2.68 ) Education (less than a high school diploma) 1.39 (0.83 2.31 ) Family income ($19,999 or less) 3.41 *** (1.83 6.35 ) Fatalism (high) 0.58 * (0.36 0.94 ) * p<0.05 ** p<0.01 ***p<0.001 In the following section I report results from logistic regression analyses employed to test Hypothesis 1 c : Institutional barriers to health care resources will demonstrate a posi tive relationship with fatalism .

PAGE 83

70 A binary logistic regression was performed to ascertain the effects of health insurance status and usual source of health care on level of fatalism. Model covariates included age, time since last doctor visit, gender, length of residence in the US, education and family income. The log istic regression model was significant, 2 = 21.21, p < .05. The model explained 7.5% of the variance (Nagelkerke R 2 ) in screening for cholesterol and correctly classified 61% of cases. According to the model, having health insurance is negatively related to high level of fatalism though this relationship is not statistically significant after controlling for all other variables in the model (Table 8) . Likewise, h aving a usual source of health care is not statistically related to high lev el of fatalism after controlling for all other variables included in the model . Those aged fifty to fifty nine years had 78% greater odds of demonstrating a high level of fatalism compared to those aged forty to forty nine years (OR 1.78, CI : 1.02 3.12). Also, w hen compared to males, females were 53% less likely to display a high level of fatalism (OR 0.47, CI : 0.27 0.83). These results do not offer support for hypothesis 1c. Table 8 . Binary Logistic Regression Results for the Odds of Demonstrating Hig h Level of Fatalism Based Upon Having Health Insurance and Having a Usual Source of Health Care. Participant Characteristics OR (95% CI) Have health insurance 0.85 (0.50 1.44 ) Have USOC 1.24 (0.73 2.11 ) Age (40 49 yrs) Reference group Age (50 59) 1.78 * (1.02 3.12 ) Age (60 and over) 1.63 (0.88 3.03 )

PAGE 84

71 Time since last doctor visit (one year or less) Reference group Time since las t doctor visit (more than one year ) 1.22 (0.69 2.15 ) Time since l ast doctor visit (never ) 0.43 (0.15 1.22 ) Gender (female) 0.47 * (0.27 0.83 ) Length of time residing in US (10 years or less) 1.05 (0.62 1.77 ) Education (less than a high school diploma) 0.73 (0.46 1.15 ) Family income ($19,999 or less) 0.87 (0.51 1.50 ) * p<0.05 ** p<0.01 ***p<0.001 In this fin al section, I report results from mediation pathway analyse s using the PROCESS macro in SPSS v. 25 (Hayes 2018) . Mediation pathway analysis was used to test Hypothesis 1d : Institutional barriers to health care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Indirect effects will work through fatalism. To test whether fatalism mediates the relationship between institutional barriers to health c are and use of blood pressure and cholesterol screening services, four separate tests were performed. The first test examined the relationship between health insurance and use of blood pressure screening with fatalism included as mediator. The second test looked at the relationship between health insurance and use of cholesterol screening with fatalism included as mediator. The third test examined the relationship between usual source of health care and use of blood pressure screening with fatalism included as mediator. The last procedure examined the relationship between usual source of health care and use of cholesterol screening with fatalism again, included as a mediating variable.

PAGE 85

72 Results for each of the four tests revealed that fatalism did not signif icantly mediate the relationship between the main independent and dependent variables of interest. This was not completely unexpected as the relationship s between usual source of health care and use of both blood pressure and cholesterol screening services w ere strong with statistical significance achieved in both logistic regression analyses under hypothesis 1a. While health insurance was not statistically related to use of cholesterol screening, results demonstrate that it was strongly related to use of b lood pressure screening ( p< 0.01). Moreover, despite the statistically significant relationship between fatalism and use of cholesterol screening services demonstrated in hypothesis 1b, fatalism did not show a statistical relationship to use of blood press ure screening as an independent variable to either health insurance or usual source of health care when modeled as a dependent variable. Given the relatively strong relationships between both health insurance and usual source of health care and use of both blood pressure and cholesterol screening services, and the overall lack of a statistical relationship between fatalism when modeled as an independent and dependent variable, a lack of fatalism acting as a mediator between health insurance or usual source of health care and use of CVD screening services makes sense.

PAGE 86

73 CHAPTER V QUALITATIVE ANALYSES AND RESULTS Latinos in the U.S. continue to face an inequitable array of agency reducing barriers to health care including lack of access to health insurance and a usual source of health care. Reintroducing context to the study of fatalism and screening behavior is essential if we are to more fully under stand how individuals rationalize the use of prevention resources when faced with persistent struct ural obstacles to health care services . This project include d qualitative methods i n an effort to gain insight to how this population internalize s , reason s about, and respond s to institu tional barriers to health care in deciding to access screening services for CVD. In addition to shedding light on the meaning of the quantitative results obtained in P hase O ne, the qualitative methods used in this project serve to explain how motivations and attitudes tow ards CVD prevention and use of screening res o urces. In this chapter, I focus specifically on the results of qualitative analyses used to address Specific Aim 2: Qualitatively explore how institutional barriers to health care may inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latinos in Denver, Colorado. Review of Methods Interview part icipants consist ed of a subset of all participants who completed quantitative surveys in P hase O ne. After the completion of each survey, participants were asked whether they would like to participate in a one on one interview to further talk about their exper iences with health insurance and having a regular place to go see a health care provider, including how these relate to their ideas of heart disease and using blood

PAGE 87

74 pressure and cholesterol testing services. Based upon their answers to P hase O ne survey questions regarding h ealth insurance , usual source of health care and the validated scale used to measure fatalism, I placed participants into one of four groups representing the sampling domains for P hase T wo qualitative interviews. I defined h igh levels of institutional barr iers to health care as confirmation of an individual experiencing both barriers represented by the independent variables of interest (e.g. lack of health insurance and no usual source of health care). I defined l ow levels of institutional barriers to health care as experience with no ne or only one of the se in dependent variables of intere st . Due to the five point Likert type questions comprising the validated scale, I determined level of fatalism by calculation of an overall score and the performance of a median split. Median score was 2.3 with a range of 4.0. I defined t hose scoring below this number as low fatalism and those scoring above this number as high. After population of all four of the P hase T wo sampling do mains was complete, I then drew a series of random sample s consisting of four participants at a time from each of the four population groups (F igure 4) . Research assistants made up to t hree phone calls to randomly selected participants until the y were able to confirm four interview s from each group that would comprise the final interview sample and analysis cohort. All interviews were recorded with participant permission and transcribed verbatim in Sp anish with English tran slation. I conducted all interviews with the aid of one research assistant who performed translation for interviews conducted in Spanish. Transcripts were analyzed and coded in Atlas Ti v.8.3.0.

PAGE 88

75 Figure 4. Phase Two Qualitative Interviews Sam ple Selection I used constant c omparison analysis ( Glaser & Strauss, 1967) to analyze all interview data . For the first pass through the data , I used i nductive coding procedures . This involved carefully reading and re reading through the transcripts with an open mind and creating codes that emerged naturally from the data. At this stage of the analysis, I referenced my research question and Specific Aim 2 as guides but did not look to Specific Aim 1 or my hypotheses to guide my work. A fter inductive analyses were complete, I re read the transcripts and analyzed the data again using deductive coding procedures. During this stage of the analysis, I utilize d an existing framework cons isting of established a priori theme s developed from Sp ecific Aim 1 and coded the data by ref erencing these

PAGE 89

76 themes and the relationships of interest in Specific Aim 1 . During this stage of analysis, I read through the data looking for meaningful segments that represent ed or relate d to th e a priori themes and c oded the data accordingly . Ordering the analysis in this manner allowed me to prioritize examination of the data less encumbered by theoretical assumptions a nd put me in a better position to u ncover critical details that help to explain the formation and m aintenance of fatalism among Latinos including why they do or do not use CVD screening services. The Contextual Nature of CVD Fatalism Qualitative analyses reveal that fatalism does not exist as a static cognitive disposition confined to the individual psyche, but rather arises as a result of dynamic social producing agency. Moreover, the manifestation of this disposition does not occur as one form or ty pe but rather varies in its expression and representation . F atalism towards CVD and the use of screening resources is highly context dependent and often occurs at two intersections. The first is at the junction of institutional barriers to health care and th e perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiovascular health . Fatalism at this juncture comprises an expression of social and economic distress and lack of control over CVD due to the unmistakable recognition that heart disease represents a serious health condition for which medical intervention is often costly and ongoing . The second form of fatalism occur s at the convergence of institutional barriers to health care , the presence of disease and the logics of religion that help to define the am ount of control one believes they have over CVD. At this intersection , fatalism embodies the expression of a process whereby one actively defines the boundaries

PAGE 90

77 of control over disease through belief systems that cut across religious values related to individual responsibility for health. The intersection of these complex and varied social processes gives life to altern ating manifestations of fatalism that belie the notion of fatalism as a static and singular concept. Rather than viewing fatalism as representing a consistent and fixed set of beliefs, I argue that this disposition varies in the form of its expression by w ay of the social circumstances that contribute to its development . CVD Fatalism Intersection 1 : IBHC , Material Deficits and the Perception of Lack of Control Fatalism towards CVD and the use of screening services is inextricably li nked to institutional barriers to health care s uch as lack of health insurance and no usual source of health care . These barriers intersect with and oftentimes, exacerbate the perception of lack of control over heart disease due to the recognition that limitations in the availability of personal financial resources exist . For those lacking health insurance, screening for and potentially having to treat disease produces a state of chronic stress and depression related to being unable to afford essential medical services and procedures . These conditions give rise to a fatalistic o utlook as one makes a realistic appraisal of the lack of options with which risks to heart health can b e address ed . Fatalistic discourse is not evidence for the presence of irra tional thinking that needs to be cor rected. Rather, fatalism exists as a rational expr ession of social and economic related distress and loss of hope due to the understanding that one has limited options with which to address potentially ov erwhelming finan cial liabilities for the receipt of health care. For those forced to endure the hardships of these socially organized circumstances , i nstitutional barriers do not simply exist as external to but rather become

PAGE 91

78 subjective realities that are incorporated into the everyday lived e xperience, decision making and health behavior of those with restricted access to social and material resources . More specifically, s tructurally organized constraints on human agency, such as lack of health insurance and no usual source of health care, are embodied in the very outlook of those that experience them. It is in this sense that fatalism becomes the product of s ocial process es that etch their negative effects in the very attitudes and dispo sition s of those impacted . In t he following quote, we see just how detrimental this loss of control can t having health insurance combined with a lack of financial resources with which to pay medical bills can result in severe depression and may lead to fatalistic suicide in an effort to escape overwhelming financial obligations. And I said (without health insurance h wow , how we are going to pay the big bill Well one patient might die (right away) and the other might di e little by little without money to pay (the medical bills) that you owe you know what I me an? How are you going to pay? Make arrangements? T ha . You m ake arra ngements and little by little you get d epress ed and frustrated and then you go down and down through depression . Some of these people kill themselves because of this (depression) . I know some people that kill ed (themselves) be cause of this kind of depre s sion. For not having I mean not even having money for food, not having enough money to pay the bills. Health insurance provides people with options to prevent and treat disease and helps with the financial obligations that come with receiving health care. When a lack of this critical resource intersects with a perceived loss of control over health due to the poten psyche may be overwhelmed such t hat a loss of the very will to live results . In these tragic circumstances, social processes render hope and vitality for the future worthless. F atalism towards CVD in this instance represents the extreme vulnerability and lack of control over disease tha t so many in the Latino community feel as a result of limitations imposed on access to social and material

PAGE 92

79 resource s. This type of fatalism is rooted in the socio political and econom ic marginalization of Latinos whereby biomedical resources are reserved o nly for those who can afford them. Access to health care resources remains a thoroughly social process that prioritizes high cost technology and profit over access to affordable health care as a fundamental right for all . As shared by the next participant , living without health insurance and a regular place to receive health care results in a complete lack of acceptable options with which to tackle medical bills. Once again, a preference for the end of life reveals that a total loss of control over options breeds fatalistic statements. This disposition is not an s a realistic appraisal that acceptable courses of action are severely limited. It affect s me a lot right now. Becau s e the whole thing is (without health insurance) Yeah, to pay two hundr ed thousand dollars for a medical bill you know what I mean. Because at that cheap er to die than have to face t thing for all the Hispanic s the wrong circle, there is no health insurance for us . When viewed in context, fatalism towards heart d isease becomes a rational response to severely limited options with which to address chronic health cond itions that threaten both physical and financial autonomy. The prospects of living life with the burden of crushing medical debt for health care is unbearable for some and leads to a demoralized expression of fatalism. Emotional distress associated with a lack of options with whi ch to access and afford health care own behalf. In the following quote, a parti cipant describes how having money may not even lead to treatment because without appropriate coverage, access to health care

PAGE 93

80 professionals remains out of reach. This lack of access leads to feeling s of extreme pressure and stress . Ultimately, restricted ac cess to essential health services results in the internalization of deprivation and to a fatalistic loss of control. What shou ld I ve a place to g o to get check ed or treated . I feel my health gets worse when I think about that I don t have (health insurance). I tell you again even if I take money (to the clinic) an identification that ha s my (insurance i nformation). I feel way more excited and stress ed about it and I just go into my head just wondering what should I do? What could I do? Because the sad part of the system is that even if I had the money they ver y stressful and I th ink I just get sicker and sicker and sicker. I feel that you lose all control (without health insurance) . Health insurance is not merely a possession that people use when necessary. Nor is it an abstract agreement between an individual or family and an insurance provider. Health insurance imbues the holder with a sense of security and provides a means of takin g action to ensure that one remains healthy or can treat medical problems as they arise. Socio political processes that render vulnerable members of the Latino community unable to procure this critical resource serve to undermine the sense of contro l and b elief that one can effectively prevent and combat heart disease through access to screening and treatment services. For socio demographically disadvantaged individuals, a lack of health insurance becomes part of the everyday reality of dealing with restri ctions imposed on how, when and where one can access affordable health care resources . As expressed by the following participant, frustration, anger and depression over these circumstances can lead to a sense of subjugation and the loss of hope. Well that m tellin g you, you get mad (without health insurance). Y ou get , well I let myself fall (get depressed) because I tell you I started thinking, hat couch. I got (very) depress ed ... Yes, I was without hope.

PAGE 94

81 E xpressions of fatalism and the loss of hope also occur in the co ntext of lacking a medical home or usual source of health care . Along with limited access to health insurance, not having a usual source of health care is an important institutional barrier to health care for Latinos. H aving a usual source of health care has been associated with gr eater use of health service s and the earlie r receipt of preventive care (Ettner, 1999). This may be explained in part because patients with an ongoing, long term relationship with their physician may have heightened awareness of certain health behaviors along with the opportunity to discuss these behaviors, gain access to health related information , and treat problems (Ettner, 1999). Not having a usual source of health care source from which critic al health care information and resources can be obtained. Interview participants remain acutely aware of the importance of having access to a usual source of he alth care. A medical home is not simply a place where one receives regular health screenings an d services , for it also serves as a setting in which one develops relationships of trust with members of the health care community and the confidence that health can be protected , fostered and potentially restored . In the f ollowing excerpt , a participant relates what not having a usual source of health can this participant , e to have her chronic condition checked and treated by a competent professional. Well I think my health will deteriorate because well by not having a clinic to go to ... ed . I s where one gives up and falls down (gives up without hope) . I tell you I had three years of not going to the clinic and the day I w ent wel l afford And one feels like down, hat am I going to do? Well I

PAGE 95

82 Fatalistic attitudes arise at the intersection of institutional barriers to health care and the perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiovascular health. F ata lism is not an inherently stable personality trait or culturally derived attitude that Latinos possess but rather arises out of dynamic interaction between personal options to address health issues are severely restricted. In these contexts, expressions of fatalism represent rational appraisals of specific circumstances at specific times. Social, economic and institutional forces intersect with health beliefs related to the control of CVD and result in fatalistic interpretations of personal health circumstances. These assessments are coherent expressions of social and economic distress . When viewed in context , fatalism towards CVD becomes a logical response to a socia lly and poli tically organized system of health care that denies resources to those who are unable to afford them or access them because of their legal status . Screening and treatment of heart disease is not an enigma for the Latino community. Rather, a lac k of equitable access to and the ability to pay for screenings and the potential treatment of CVD places Latinos in a vulnerable position not of their own choosing . The recognition that a diagnosis of heart disease can result in lifelong and overwhelming financial burden breeds fatalistic attitudes. It is in these dispositions that institutional barriers to health care are embodied and serve to affect the very outlook to wards heart disease and its prevention of those who experience them.

PAGE 96

83 CVD Fatalism Intersection 2 : IBHC, the Logics of Religion and the Organization of Control personal t rait that threatens prevention science and its aim to reach Latinos fails to account for how fatalism varies in its expression as a result of the social context withi n which it forms (Espinosa de los Monteros & Gallo, 2013) . Fatalism comprises dynamic repr esentations that occur at specific times and in specific places. For those faced with challenges to their health, whether and when to cultivate such a disposition is contingent upon the perception of control individuals feel they have over CVD and the heal th care resources they have at their disposal. In depth analyses of interview data reveals how living without health insurance or a usual source of health care leads to the internalization of a sense of helplessness and loss of control. When an individual experiences these barriers and is also confronted with a direct threat to their health such as diabetes or hypertension, the ethics of religion may be employed in an effort to organize and define a meaning ful appraisal of the control one has and the course s of action available . Once again, r ather than exhibiting fixed characteristics, fatalism towards CVD is in this instance, expressed at the unique intersection of institutional barriers to health care , the direct pres the active cultivation of a disposition (Hamdy 2009) in which shared responsibility with God helps to define the measure of control one has ove r their condition . Appealing to divine will is not evidence of passivity, but rather represents an active org anization of shared responsibility when recognition that restricted access to health care resources and the peril of disease co exist. The manifestation of fatalism at this particular intersection is distinct from that which occurs at intersection one.

PAGE 97

84 Rat her than representing the hopelessness that ensues from the culmination of socially and material ly induced distress, here it arises out of a working process whereby boundaries between control over the presence and responsibility for disease are negotiated through the dictates of religious interpretation and belief. Interview participants who experience institutional barriers to health care acutely recognize and understand the value of good health. However, they also understand that limits imposed on access realistic difficulties for which there are no easy solutions. Individuals that have been diagnosed with CVD related conditions or disease must confront life in which chronic illness threatens both short and long term wellb eing. Not having health insurance or a usual source of heal th care can severely exacerbate heal th related stress and confusion . Cultivating a disposition consistent with the logics o f religion help s create a meaningful when faced with illness and re strict ed access to health care . While verbal expressions of fatalism often occur within this context, they are not isolated . Oftentimes, fatalistic statements occur alongside others that speak to the value These apparent contradictions serve as a n active form of negotiation whereby individuals wrestle with the options available to them at the inter section of institutional barriers to health care , t he presence of a health disorder and the measure of control one has over health conditions. The appeal to divine will and authority is not passive or subjugating, but rather represents the simultaneous recognition that while one must act in partnership with God to protect health, a healthful life free of disease is not guaranteed.

PAGE 98

85 In the followi ng passage , a longtime diabetic who lacks health insurance speaks to her and interprets her current health condition through the fatalistic appraisal that diabetes was a predetermined outcome for her. Yet at this intersection of in stitutional barriers, the presence of disease and the logics of religion, Of course I realize that I have to take better care of myself. I also recognize that God has a pl something t supposed to happen. But at the end of the day I need to take care of myself too. For this participant, the appeal to God creates meaning and helps to organize and define how she must move forward under difficult and potentially life threatening circ umstances . Dealing with disease includes both acceptance of its presence and the responsibility to take care of oneself in the context of barriers to health care. In the following excerpt , another diabetic without access to health care resources demonstrat es that submission to divine authority is not a passive form of fatalism that negates human agency but rather speaks to the active cultivation of an appraisal in which according to control myself a nd I have to accept what comes. Accept what comes but now take care of myself and cope with myself... well I have to put an effort in for myself. Blindly accepting f atalistic statements indicat ing that those who believe in as proof of complete submission to external forces obscure s how the se very statements give meaning to the intersection of institutional barriers to health care , the presence of

PAGE 99

86 health challenges, and the search for control . In this context, reference to religious subt exts s active definition of personal responsibility and the expectations that come with sharing a relationship with God. It is the embodiment of this relationship that guides the individual when answers may not be forthcoming . In the quote th at follows, a participant who has long struggled with hypertension and lack of access to health care discusses how religion informs her perspective on CVD. Well I think if you eat healthy, you will help your health a lot. But the first thing is I Next, well we can mobilize and see what we can do. If I can do it (prevent and treat heart disease), well I will do it. Yes, I work lik e a team with God. But, if som eone is destined to get a dise ase it does not matter what they do because either way they will get sick. Well because I will repeat again, the it will be like that. Like other time s well my Father God says up to this point (I will help you). Yes, because my Father God says help yourself and I will help you. The me dicine I take, I entrust my actions to God. Despite an apparent submission to divine authority, this participant also accepts the responsibility of working in partnership with God to protect her heart health. Religion does not serve to suppress her desire to seek better health , rather it serves as e ncouragement to take responsibility for her health and helps to defines why and how she should proceed. In the context of limited access to health care resources and chronic hypertension , the logics of religion at once help to define where control resides and how it is shared. In the following excerpt, a participant who has been without health insurance and is diabetic speaks about this shared relationship. Working lik e a team with God contributes to the active appraisal of how m uch control one has. Develo pment of this appraisal requires active cultivation of the relationship she shares with God and further illustrates that fate, destiny and personal control reside at the dynamic juxtaposition of institutional barriers to

PAGE 100

87 health care , challenges to active negotiation of control though the rationale of religious belief . Yes I am a believer of that (life is already written). But also destiny, you have it in lp me God. have the belief of asking San Juditas (patron saint of lost causes) for help and there wants, he that my faith is what leads my God to help me. But I always work like a te am with God. Con D a disease I try to do everything I can on my side for that disease to go away, as much s important, b ecause if my father God helps me, maybe that disease will leave. Despite voicing beliefs that would otherwise be consistent with a fatalistic disposition , this participant also accepts a measure of personal responsibility for her health as i nformed by the logics of her religion. She understands that her relationship with God bestows certain responsibilities on her such as actively work ing towards taking care of her health. In partnership with God, control is both organized and defined at the intersection of institutional barriers to health care and the presence of diabetes. In these circumstances, fatalistic language is contextualized and co occurs alongside statements that reference personal control and the acceptance of responsibility for on institutional barriers to health care, the presence of disease and the logics of religion, isolated expressions of fatalism are at once replaced by a disposition that seeks a shared measure of control over circumstances t hat would otherwise appear to discourage all hope of. The Abatement of Fatalism: The Role of Safety Net and Community Based Resources Institutional barriers to health care are a key component in the development and expression of CVD related fatalism among Latinos . The manifestation of this disposition is

PAGE 101

88 variable and occurs at t he dynamic intersection of social processes related to acces s to health care resources and perceptions of control over heart disease. Given the importance of institutional barriers to health care to the development and expression of CVD fatalism, it i s imperative to consider why certain cases that exp erience lack o f health insurance or no usual source of health care demonstrate low rather than high l evels of fatalism. These other factors that serve to influence fatalism towards CVD. Intervie ws with study participants demonstrating high levels o f institutional barriers to health care and low levels of fatalism reveal that safety net resources in the Denver area may play an important role in diminishing fatalism . As part of the broader public health system, including non profit, community based and other health related organizations and services to the metropolitan areas most vulnerable populations ( Krager & McCann 2016). Safety net clinics in Denver range from small clinics to large health care facilities with multiple sites across the city. These clinics serve diverse populations including those who lack health insurance or a usual source of health care and also provide access to pr imary care services for undocumented persons. The range of services provided includes both standard physical and mental health care as well as educational classes across a variety of health topics (Krager & McCann 2016). As part of their broader mission, m any of the clinics provide free access to cholesterol, blood pressure and hemoglobin A1C screening and classes or information sessions on how to read food labels, eat healthier and engage in physical activity. A variety of these c lasses focus on CVD and diabetes prevention and management and reach participants through both in facility contact and public health outreach initiatives.

PAGE 102

89 Interview participants demonstrating high levels of institutional barriers to health care and low levels of fatalism consist ently mentioned gaining a sense of control over their health by accessing prevention and educational services at clinic locations throughout Denver County. Despite not having health insurance and a usual source of health care, participants are identifying and utilizing services that may help them to feel more in control over their CVD health. One potential explanation for this is that safety net resources may help to assuage fatalism towards CVD that arise s at the intersection of institutional barriers to h ealth care and the perception of lack of control due to the lack of social and financial means with which to address threats to c ardiovascular health. Safety net clinics that offer free or sliding scale screening, education and treatment services to uninsu red populations help those who would otherwise perceive a lack of control over their CVD to feel more confident and optimistic about the future . As the following uninsured participant relates, receiving care at a local clinic is the best way for her to pre vent CVD. Well for control (of CVD), I think everybody knows that we have to eat well , you know eat as healthy as possible and in the most natural way right. But well there is nothing like prevention in the clinic. Like getting care, regular care. Well I try to do it right because I prefer to prevent and go the clinic. Well I like to be healthy (with my diet) but also do my check ups, at the clinic righ t. Interview p articipants recognize that local clinics offering culturally and l inguistically appropriate care to the Latino community are available and represent an important part of CVD prevent ion and treatment . Control is tied to the idea that in part, health professionals are access ible and affordable even when one insurance or a regular medical home. In the following quote, a n uninsured interview participant states that controlling CVD is a matter of going to the clinic whether you have

PAGE 103

90 insurance or not. The recognition that clinic resources are available to her is readily apparent. Well to control heart disease I have to go to get care at the clinic. W hen there is a problem or diseas you have insurance or not. Despite lacking health insurance or a usual source of health care, interview participants acknowledge that gaining control over heart disease requires accessing resources from health care professionals at formal health care settings. Knowing that there are clinics that will admit and care for those without health insurance offers individuals facing institutional barriers to health care a sense of security and a measure of confidence that help is available and something can be done to take of their health. W hen the perception that professional help and resource s are readily available, fatalism towards CVD is tempered, an in many cases, dissolves completely. Community h ealth e ducation is to serving vulnera ble populations in the Denver metropolitan area. Health education classes support the greater safety net mission to provide culturally competent care and preventive health services to the medically underserved. Much of the health information provided by sa fety net clinics would otherwise be inaccessible to many in the disadvantaged Latino community. In the following passage, an interview participant talks As the participant relates, obtaining information from the health care team at Clinica allows her to maintain the belief t hat the course and ou tcomes of health can, in part be influenced through persona l agency. Well I have listen ed to people (at the clinic) speak on the topic, talks that they have given on how to prevent diabetes, pre diabetes and other things like that. They talk a bout diabetes, type one and type two, all of that. The doctor, the nurses , all of them

PAGE 104

91 give you information on how to take care of yourself . You do it with what you know and what you hear to take care of yourself. B y going to the doctor well you get inform ation about your diet and all of the information you need to maintain and control the disease. In addition to clinic esidents. Recent data from The Colorado Department of Public Health and Environment (2018) shows ten public health programs targeting Denver County received funding for 2019 through 2021 from the Cancer, Cardiovascular and Pulmonary Disease Competitive Gra nts Program alone. Including programs that receive funding from other sources such as local foundations and trusts adds to this number. Health education c lasses offered by local public health programs targeting healthy eating and active living and CVD and diabetes prevention may give individuals without health insurance and regular access to health care resources the hope and confidence that they hold the power to influence their health. As shared by th e following participant, ed ucational classes are readily available in the community and provide the means for people to learn how to affect their health. taken nutrition classes in the community. had a lot of nutrition cl asses on how to balance f oods and how to read classes or information about how to eat better. Now there are more programs out there , how to take care of your health. The presence of community based prevention programs may help individuals without health insurance or a usual source of health care feel empowered to take control of their health. Fatalistic expressio ns are notably absent among participants who access these health education classes while also e xperiencing institutional barriers to health care and a lack of financial resources. This may be due to in part, to the recognition that health resources are available outside of traditional health care facilities that continue to deny so

PAGE 105

92 many in the Latino community because of a lack of health insurance. According to the following participant, community programs have empowered and prepared her to prevent CVD. Well I feel now thanks to the programs in the community I have enough informatio n about how to eat healthy, make the right portions, measure quantities, and read the labels for information about sugar and sodium . I feel to prevent disease in me and in my f amily. Community based health programs reach members of the Latino community that may otherwise fall through the cracks of a system that prioritizes access to health care through health insurance and the establishment of a usual source of care. For the uninsured an d those that lack a usual source of care and t he financial resources with which to pay for health care, educational programs offered by local community based organizations provide a means to access resources and develop the knowledge that staves off the perception of a lack of control over CVD and dia betes. In the following passage, an uninsured participant that lacks a usual source of health care speaks to the availability of belief in prevention and management of disease runs counter to a fatalistic outlook. The availability of health education classes tailored to local Latino communities may help to temper the sense that there is nothing one can do to prevent CVD or other health related conditions. I have been to classes with people in the neighborhood -as I would like but they give us classes for nutrition and how to have a healthy diet. They show us how diabetes can be prevented and managed. Now there is help for us. There is advice for how to take control of your body. Fatalism resides at the intersection of institutional barriers to health care and the perception of a lack of control over heart disease due to the recognition that one does not

PAGE 106

93 have the social and fi nancial means with which to address threats to c ardiovascular health. It also lives at the crossroads where these barriers, the presence of disease and the logics of religion meet. Safety net resources provide vulnerable members of the Latino community wit h access to health care resources and a usual source of health care despite their uninsured status. Many of the services offered by both clinics and community based public health programs are free of charge and exist a s a means to care for the medically un derserved and marginalized. Addressing existing instit utional barriers to health care, safety net resources offer access to critical health services and information for those who would otherwise lack hope. Perhaps most important, safety net resources help to restore the dignity of the most vulnerable community members by in stilling a sense of optimism about health. Dynamic Intersections, Social Context and the Variable Expression of Fatalistic Attitudes Fatalism is not a singular and fixed concept, rather the form of its expression varies by the social circu mstances that contribute to its development and representation. Fatalism borne at the intersection of institutional barriers to health care and the pe rception of lack of control over heart disease due to limits in the availability of financial resources comprises an expression of social and economic distress. These expressions represent the rational appraisal that practical options with which to access and pay for CVD screening and treatment are severely restricted. In these pa rticular circumstances, fatalistic expression represents the outcome of unjust social processes that render an already vulnerable community uncertain and with diminished hope of mi tigating the burden of heart disease.

PAGE 107

94 For those confront ing the prospects of living with heart related illness or diabetes and without health insurance or a usual source of health care, the presence of disease and boundaries of responsibility are at once defined through the active cultivation of a disposition in which shared responsi bility with God helps to define the measure of control one has over their condition. In these circumstances, expressions of fatalism often co occur apparent contradictions serve as an active negotiation over control at the intersection of institut ional barriers to health care , the presence of a health disorder and the logics of religious belief. Fatalistic statements at this juncture represent the organization of respons ibility and control over CVD and are highly contextualized. While reference to divine authority may appear as passive and subjugati ng, fatalistic statements at this intersec tion demonstrate how the definition of control is socially derived and actively produced . based public health resources m ay be serving to diminish the presence of CVD fatalism among Latinos, they do not replace the social, economic and health securities that are derived from procuring health insurance and a usu al source of health care . According to Mile High Health Alliance (2016) s ubstantia l ability to provide or secure complete care for their patie nts, particularly in the domains of fully integrated care on site and access to specialty care on or off site. Moreover, c urrent payment mode ls do not support the sustainability of existing programs focused on integrated care (Krager & McCann 2016) . Relying solely on safety net resources to provide care for vulnerable members o f the Latino community is

PAGE 108

95 short sighted and fails to account fo r the social injustices perpetrated on minority communities as a result of the inequitable distribution of health care resources . CVD fatalism is not a faulty personality trait that needs correction via educational initiatives. Rather, it is an inherently nuanced disposition produced at the intersection of dynamic social processes. If we hope to account for the presence of fatalism towards CVD screening and treatment, we must first account for the way institutional barriers to health care are emb odied in the very outlook of those that must endure them. Removing these barriers and establishing more equitable access to culturally co mpetent health care is the first step to instilling a sense of control over CVD and hopefulness for the future .

PAGE 109

96 CHAPT ER VI MIXED METHODS INTERPRETATION AND CONCLUSION For the past several decades, health science researchers have attempted to explain the apparent failure of large segments of the US population to engage in health promoting behavior such as use of preventi ve health screenings . Much of this work ha s led to inadequate understanding and unsuccessful attempts to increase pre vention minded actions among Latinos, the nation . While several potential reasons for a lack of behavio r change have been explored, a renewed focus on health beliefs and attitudes fate, destiny, luck or divine intervention , has gained popularity among investigators concerned with explaining suboptima l screening rates among Latino popula tions. Classified under the general conceptual category of fatalism, health researchers often view this disposition as a barrier to increasing screening rates . Thi s is due to the understanding that fatalism encompasses a perception of health as being primarily determined by factors outside the control of the individual, thus ne gating the relevance of individual behavior in the prevention and/or management of disease . Regrettably , extant research has failed to adequately contextualize this disposition in view of the pervasive institutional barriers to health care Latinos confront in the United States. Foremost among these barriers are limited access to health insura nce and the lack of a regular source of health care. The overemphasis on individual level fatalistic attitudes carries important implications. First, t hose that endorse fatalistic attitudes and choose to forgo screening and treatment are unfairly labeled Second, t he de contextualized labeling of thos e who express fatalistic attitudes serves to

PAGE 110

97 misdirect attention away from the very real barriers to health care tha t Latinos face by ignoring the social, eco nomic and institutional forces that shape the health beliefs and decision ma kin g processes of this population . Reintroducing context to the study of fatalism and screening behavior is essential if we are to more fully under stand how indivi duals rationalize the use of prevention resources when faced with persistent structural obstacles to health care services . Unfortunately, much of the work on fatalism and screening behavior among Latinos has utilized quantitative only research designs , thu s leaving fundamental questions about how people internalize, reason about, and respond to institutional barriers to health care unanswered. When used in isolation, q uantitative methods are inadequate to improving our understanding of health related fatali sm and ultimately leave us with a limited motivations and attitudes towards prevention behaviors and use of screening res o urces. To fully understand how institutional barriers to health care impact the CVD related health beliefs and scree ning behaviors of Latinos, this project employed a mixed methods design. Adding qualitative methods allowed for the critical exploration of qua ntit ative results while also permitting investigation into how institutional barriers are incorporated into the lived experience, decision making processes and health behavior of this population . In an effort to more fully understand the relationship between i nstitutional barriers to health care, fatalism towards cardiovascular disease and the suboptimal use of cholesterol and blood pressure screening services b y Latinos , this study sought to answer the following research question : How does accounting for the s ocial context in which health

PAGE 111

98 care decisions are made improve our understanding of fatalism and its impact on the CVD screening behaviors of Latinos in Denver, Colorado ? Summary of Phase One Results: Quantitative Survey The summary data referenced in this section come from a series of binary logistic regression analyses that I conducted to test the hypotheses laid forth in this project . Quantitative analyses reveal that i nstitutional barriers to health care play a key role in the suboptim al use of CVD screening services among Latinos. Those with a usual source of health care had greater than three times the odds of obtaining cholesterol screening compared to those without a usual source of health care (OR 3.35, 95% CI : 1.77 6.32) . Likewi se, t hose with a usual source of health care were five times more likely to obtain screening for blood pressure (OR 4.97, 95% CI : 2.71 9.10) when compared to those without a usual source of health care after holding all other variables constant. T hose wi th health insurance had significantly greater odds of obtaining screening for blood pressure (OR 2.30, 95% CI : 1.23 4.30) when compared to those without health insurance after controlling for all other variables in the model . Unexpectedly, h aving health insurance was not related to obtaining screening for cholesterol when compared to not ha ving health insurance. Discussion of this apparent anomaly occurs below under Mixed Methods Interpretation . It should be noted that perceived discrimination in the health care encounter, originally included as one of three institutional barriers to health care under Specific Aim 1, fell out of the model due to low exposure to this particular variable. Only twent y one individuals or five percent of the total sample, reported exposure to perceived discri mination. This number is less than what other studies have reported. For example, among a population based sample of adults responding to the California Health Inte rview

PAGE 112

99 Survey, Sorkin et al. (2010) re port 13.4 percent of Latinos experiencing perceived discrimination in a health encounter due to race /ethnicity compared to 2.6 percent of non Hispanic Whites. In work among another population based sample responding to three waves of the California Quality of Life Survey, Mays and colleagues (2017) found 31 percent of Lat inos reported being subjected to discrimination by a health care worker or provider. This was more than twice the rate reported by non Hispanic Whites . Alternately, among a nationally representative sample of U.S. adults, Stepanikova & Oates (2016) found that only six percent of Latinos reported experience with perceived discrimination in the health care setting within the past 12 months. Lastly, Shepherd et al. (2018) report 12 percent of Latinos in their non random sample from a Mid Western city have experienced discrimination by a health care professional because of their cultural background. One potential explanation for the low rate of perceived disc rimin ation found by this research may have availability of culturally and linguistically appropriate health care resources for Latino populations. Denver is home to many clinics and health care facilities that focus on providing health care to low income minority populations , Westside Family Health Center, Clinica Campesina Family Health Services, Inner City Health Center and Caritas Community Clinic. Health care institutions that incorporate culturally competent and patient centered framework s in their models of care may help t o lessen the likelihood of patients experiencing perceived discrimination during interactions with clinic staff. According to Livingston, Minushkin and Cohn (2008), 80 percent of Latinos who have a usual source of health care rate their health care as excellent or good. Moreover, these authors report that Spanish speaking patients who are able to participate

PAGE 113

100 in healt h care appointments conducted in Spanish are less likely to rate their care as fair or poor. The results contained in Table 3 of this dissertation research show that 70 percent of participants claim to have a medical home and that 95 percent prefer to spea k Spanish. Given the rather common presence of Latino specific and minority serving health care institutions across the Denver metro area, Latinos in this city may be less likely to experience discriminatory practices in the health care encounter than loca tions where culturally competent care is less accessible. demonstrated by usual source of health and health insurance. While h igh level of fatalism was significantly rel ated to a 42% lower odds of obtaining cholesterol screening compared to low level of fatal ism (OR .58, 95% CI : 0.35 0.95) , there was no significant difference in the likelihood of obtaining screening for blood pressure between those demonstrating a low l evel o f fatalism and those who demonstrate a high level of fatalism after controlling for all other variables. Consideration of these results is offered under Mixed Methods Interpretation. In an effort to understand if there is a relationship between institutional barriers to health care and fatalism, a model was created to test the effects of both health insurance status and usual source of health care on level of fatalism. Despite demonstrating that having health in surance is negatively related to high level of fatalism , this relationship was not statistically significant after controlling for all other variables in the model . However , the results do show a directional trend such that those with health insurance have 15% lower odds of demonstrating high fatalism than those without health insurance after

PAGE 114

101 controlling for all other variable in the model (OR .85 , CI: 0.50 1.44) . H aving a usual source of health care was not statistically related to h igh level of fatalism . As discussed in Chapter 4, mediation pathway analysis was used to test Hypothesis 1d : Institutional barriers to health care will demonstrate an indirect relationship with use of cholesterol and blood pressure screening services. Ind irect effects will work through fatalism. Results for each of the four tests revealed that fatalism did not significantly mediate the relationship between the main independent and dependent variables of interest. This was not completely unexpected as the r elationships between usual source of health care and use of both blood pressure and cholesterol screening services were strong with statistical significance achieved in both logistic regression analyses . Given the relatively strong relationships between bo th health insurance and usual source of health care and use of both blood pressure and cholesterol screening services, and the overall lack of a statistical relationship between fatalism when modeled as an independen t and dependent variable, the failure of fatalism to act as mediator between health insurance or usual source of health care and use of CVD screening services makes sense. Summary of Phase Two Findings: Qualitative Interviews Qualitative analyses reveal ed three main findings: 1. Fatalism is not a static, cognitive trait of the individual but rather arises at the dynamic intersection of context specific social processes that serve to Fatalism forms where institutional barriers to health care, material deficits and a perception of lack of control over heart disease meet. It also occurs at the crossroads of institutional barriers to health care, illness and the logics of religion. 2. Fatalism is not a singular conceptual category but rather varies in its expression and representation as a result of the context

PAGE 115

102 specific circumstances that produce it. Fatalism may represent the rational expression of social and economic distress as a result of severely li mited options with which to access health care resources. It may also take the form of an active cultivation or work ing process whereby one defines the boundaries of control over disease through belief systems that cut across religious values related to in dividual responsibility for health. 3. Safety net and community based resources may help to assuage fatalism towards CVD as a result of making educational and prevention based resources available to members of the Latino community. t and community based public health resources may be serving to diminish the presence of CVD fatalism among Latinos, they do not replace the social, economic and health securities that are derived from procuring health insurance and a usual source of healt h care. In many respects, these resources may be more accurately viewed as n ecessary but insufficient to obviating CVD fatalism among Latinos and ensuring increased use of blood pressure and cholesterol screening among this population. Mixed Methods Inter pretations: Summary Rationale and Findings Mixed methods were especially important for this study, as solely quantitative approaches have failed to adequately explain the lower rates of screening in Latino populations. Used in isolation, quantitative metho ds have left fundamental questions about how people internalize, reason about, and respond to institutional barriers to health care unanswered. To fully understand how institutional barriers impact the health beliefs and screening beh aviors of Latinos, thi s project employ ed the complementary strengths of qualitative methods .

PAGE 116

103 The rational e for employing mix ed methods in the proposed study was twofold: participant enrichment and significance enhanc ement . Participant enrichment refers to the mixing of quantitative and qualitative techniques for the rati onale of optimizing the sample. In this regard, gathering quantitative demographic data, including responses to survey questions about institutional barriers to health care and fatalism , al low ed for the selection of an appropriate sample of Latino s with whom to explore qualitatively and in depth the relationship between institutional barriers to care , health beliefs and explanations for the suboptimal use of CVD screening services. The second rationale for mixing methods in the proposed study was significance enhancement. Significance enhancement refers to mixing quantitative and qualitative of data. More specifical l y, the qualitative data was used to enhance the interpretation of findings from the quantitative analyses but also to critically explore the social context within which CVD health related beliefs, decisions and screening behavior are formed and expressed. Qualitative methods proved to be an essential part of this mixed methods study. In depth exploration of how social conditions serve to organize the ways in which people perceive and reason about CVD and the use of screening services moved beyond quantita tive only approaches to allow for a more detailed understanding of how institutional barriers become a central part of the very process by which one formulates health related attitudes and behavior. In the following sections, I discuss three main conclusio ns derived from a mixed methods analysis of the quantitative and qualitative findings produced by this project .

PAGE 117

104 Inference 1. Traditional Interpretations of the Origins and Nature of Health Related Fatalism Among Latinos are Flawed . Much of the literature focused on the relationship between health beliefs and health behavior cites fatalism as a dominant belief system inherent to Latino culture . Fatalism is perceived as a negative characteristic that deters members of this population from engagin g in various preventive activities including utilizing health screenings and other disease detection and prevention resources (Abraido Lanza et al., 2007). These views present f atalism as a static, cognitive trait of the individual such that it s very mention presupposes the existence of faulty intellectual architecture native to members of the Latino community. The oft mentioned fix is to design public health interventions that serve to re educate members of the Latino community in the name of cha nging fatalistic beliefs and increasing use of prevention services. The m ixed methods analyses undertaken as part of this project demonstrate that this view of fatalism among Latinos is faulty. Fatalism does not exist as a static naturally occurring trai t embedded in the cognitive apparatus of Latinos. Rather this disposition arises as heal th agency. Quantitative results demonstrate that having health insurance is negatively related to high level of fatalism. Although this relationship was not sta tistically significant, it does show that those with health insurance have 15% lower odds of demonstrating high fatalism than those without health insu rance . While traditional quantit ative analytics wo uld have us dismiss this result as insignificant, we fall short in our understanding of complex health decision making if we reduce our findings to those factors that remain once other predictors show no statistical significance (Drew & S choenberg, 2011).

PAGE 118

105 Qualitative findings expand on this particular lead by reveal ing that a lack of both health insurance and a usual source of health are critically linked to the expression of fatalism among Latinos. Produced at the interchange of highly flexible social conditions , fatalism arises at specific times and under specific cir cumstances. T he first is at the junction of institutional barriers to health care and the perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiov ascular health. Fatalism at this juncture comprises an expression of social and economic distress and lack of control over CVD due to the unmistakable recognition that heart disease represents a serious health condition for which medical intervention is of ten costly and ongoing. The second form of fatalism o ccurs at the convergence of institutional barriers to health care , the presence of disease and the logics of religion that help to define the amount of control one believes they have over CVD. At this i ntersection, fatalism embodies the expression of a process whereby one actively defines the boundaries of control over disease through belief systems that cut across religious values related to individual responsibility for health. These insights into the format ional complexity of fatalism would not be possible with close ended quantitative only research designs. T he addition of qualitative meth ods to this project enabled an analytical framework that moved bey ond the rendering of fatalism as a categorical v ariable to a multidimensional view of the contextual underpinnings of this disposition including how and why it occurs . In the absence of these qualitatively generated findings, quantitative results may lead to the faulty conclusion that fatalism is u nrelated to the presence of institutional barriers to health care and thus occur s as an intrinsic trait of the individual.

PAGE 119

106 Research on fatalism and health behavior typically treat this disposition as a singular conceptu al category that encompasses a fixed set of beliefs or attitudes. Lack of control, pessimi sm, predetermination, luck, and destiny are some of the more common dimensions incorporated under the conceptual category of fatalism. Despite more recent attempts to imbue the fatalism concept with multidimensional measures, the overall conceptual rendering of fatalism remains a largely singular proposition that defines this negative, irrational and static. These conventional views por tray fatalism as singularly problematic due to the assumption that its presence indicates a rigid belief in ext ernal sources of control over health. Q ualitative findings generated by this project show that fatalism does not occur as a fixed form or singular conceptual category but rather varies in its expression and representation per the social circumstances that give it life. Fatalism produc ed at the intersection of institutional barriers to heal th care and the perception of lack of control over heart disease due to the recognition that one does not have the social and financial means with which to address threats to cardiovascular health takes the form of an idiom of frustration and hopelessness. Under these particular circumstances, social and economic distress and lack of control over CVD a re expressed through fatalistic discourse. These expressions are not inherentl y negative but rather represent a rational assessment that heart disease is a se rious health condition for which medical interventio n is often costly, ongoing and financially out of reach . F atalism manifests differently at the convergence of institutional barriers to health care , the presence of disease and the logics of religion that help to define the amount of control one believes they have over CVD. At this intersection, fatalism embodies the expression of a process whereby one actively defines the boundaries of control over disease

PAGE 120

107 through belief systems that cut across religious values related to individual responsibility for health. Expressions of fatalism at this crossroads exemplify the active cultivation of a disposition in which shared responsibility with God helps to d efines the measure of control one has over their condition. Appealing to divine will in this instance represents an active organization of shared responsibility due to the recognition that restricted access to health care resources and the peril of dis ease co exist. The appearance of fatalism at this particular intersection is distinct from that which occurs at intersection one. Rather than representing the hopelessness that ensues from the culmination of socially and materially induced distress, here it ar ises out of a working process whereby boundaries between control over the presence and responsibility for disease are negotiated through the dictates of religious interpretation and belief. The intersection of these complex and varied social processes give s life to alternating manifestations of fatalism that belie the notion of fatalism as a static and singular concept. Rather than viewing fatalism as representing a consistent and fix ed conceptual category, the qualitative data from this project show this d isposition varies in the form of its expression by way of the social circumstances that contribute to its development. The in depth interviews show that expressions typically understood to indicat e fatalism often co occur alongside statements that presuppose personal responsibility for heart health. In particular, fatalism produced at the intersection of institutional barriers to health care , the presence of disease and the logics of religion is bi furcated such that the process of actively defining boundaries of control over disease results in the allocation of responsibility to both internal and external sources. This finding is critical to expanding our understanding of CVD fatalism

PAGE 121

108 and its relationship to prevention behavior and use of screening resources. Blindly accepting the conventional view that the presence of fatalism indicates belief in the absolution of personal responsibility misses the key point that fatalism is variable a nd under certain conditions may actually serve to partition responsibility for health such that taking personal action is viewed as an integral part of a larger belief system. Fully appreciating the origins and nature of CVD fatalism negates the view that the presence of this disposition indicates a stringent opposition to engaging in prevention activities. Under certain circumstances, its presence may actually signal an opportunity to better understand havior that protects cardiovascular health. Inference 2. Fatalism is a Byproduct of Complex Social Processes and Offers Little in the Way of Explaining the Suboptimal Use of CVD Screening Services by Latinos. Uncovering the root origins of CVD f atalism ma kes clear that this disposition does not exist in a cognitive vacuum independent of s ocial conditions and processes. More specifically, f atalism does not constitute a preexisting belief system native to Latino culture and patterns of thought. Rather, it is a byproduct of inequitable social conditions that restrict the options available for members of the larger Latino community to protect and promote their own cardiovascular wellbeing. Structurally organized political, soc ial and economic processes limit ing acc ess to health insurance and medical homes serve as the foundation of critical intersections at which control over CVD health is compromised. Control is tied not only to what an individual believes they can do to promote health, but also to the perception t hat there are institutionalized resources both available and accessible to support and augment these efforts . Institutional barriers to health care

PAGE 122

109 exacerbate the perception of lack of control over heart disease by reinforcing grave uncertainties related t o how one can access and pay for services that protect wellbeing. For the uninsured and those without a usual source of health care, limited access to cardiovascular disease screening and prohibitive costs associated with treatment protocols produce a stat e of chronic s tress and depression . These conditions give rise to fatalistic expression as one makes a realistic appraisal of the lack of options with which risks to heart health can be addressed. Fatalism exists as a byproduct of these social processes th at serve to marginalize and oppress members of the Latino community. Quantitative analyses demonstrate that there is no significant difference in the likelihood of obtaining screening for blood pressure between those demonstrating a low level o f fatalism and those who demonstrate a high level of fatalism after controlling for all other variables. However, h igh level of fatalism was significantly related to lower odds of obtaining cholesterol screening co mpared to low level of fatalism . In contrast to b lood pressure measurement , which is usually offe red at no additional cost and performe d during in office appointments, lipid measurement typically requires additional fees and a separate visit to a lab at which blood is drawn and analyzed. A lack of health ins urance with which to offset costs related to cholesterol measurement coupled with the ration al assessment that one can not personally afford these tests or any treatment that is recommended may induce the kind of fatalism witnessed at Intersection One. Give n that blood p ressure is offered in house, at no cos t and is a fairly quick and easy procedure, the lack of a relationship between obtaining scre ening for blood pressure and fatalism makes sense.

PAGE 123

110 Fatalism that aris es at Intersection Two is also the byproduct of social processes that occur under specific circumstances and at specific times . For those faced with CVD related illness, the cultivation of this disposition is contingent up on a perception of limited access to health care resources and the pr outlook and actions. In depth an alyses of interview data reveal how living without health insurance or a usual source of health care leads to the internalization of a sense of helplessness and loss of control. When an individual experiences these barriers and is also confronted with a direct threat to their health such as diabetes or hypertension, the ethics of religion may be employed in an effort to organize and define a meaningful appraisal of the control one has over available courses of action . Once again, rather than exhibiting fixed characteristics, fatalism towards CVD is in this instance, produced at the unique intersection of institutional barriers to health care health and the active cultivation of a disposition (Hamdy 2009) in which shared responsibility with God helps to define the measure of control one has over their condition. Fatalism surfaces as a derivative of social circumstances tied to restricted ac cess to health care resources, the presence of illness and the active appraisal of where control over CVD resides. In other words, fatalism does not occur as a principal component that contributes to the intersection of social processes but rather arises o ut of these intersectional circumstances. When c ompared to the foundational importance of institutional barriers to health care, fatalism by itself offers little to the explanation of suboptimal use of CVD screening resources by Latino populations.

PAGE 124

111 Infe rence 3 . IBHC are a Critical Part of the Explanation for the Suboptimal Use of CVD Screening Services by Latinos . The preoccu pation with examining the direct connection between fatalistic beliefs and the adopt ion of prevention beha viors has resulted in the failure to adequately consider how institutional barriers to health care may be related to the suboptimal use of screening reso urces by this population . Many of the shortcomings related to the investigation of fatalism can be attributed to the overreliance on quantitative only research designs that decontextualize fatalism by assuming it exist s independent ly of other contextual va riables. This approach results in the labeling of those who espouse fatalistic talk as irrational and in need of redirection through education. More importantly , use of quantitative only research designs to study fatalism and the careless labeling of the Latino community serve to indiscriminately direct attention away from the very real barriers to health care that Latinos face . Mixed methods analyses of this pro insurance and no usual source of health care play prominent roles in the suboptimal use of CVD screening resources by Latinos. Quantitative results show that those who have a usual source of health care had great er than three times the odds of obtaining cholesterol screening compared to those without a usual source of health care and were five times more likely to obtain screening for blood pressure when compared to those without a usual s ource of health care. In similar fashion, t hose with health insurance had significantly greater odds of obtaining screening for blood pressur e when compared to thos e without health insurance. Qu alitative findi ngs support and extend these results by illustrating how

PAGE 125

112 important healt h insurance and a usual source of health care are toward CVD and the use of screening resources. Despite the assistance offered by safety net clinics and public health programs, Latino community members who lack health insurance and a medical home view traditional health practice based CVD screening and treatment as simply inaccessible and unaffordable. Institutional barriers to health care lie at the root of this perception and are integral to the intersectional formation of CVD related fatalism. When viewed in context , fatalism towards CVD becomes a logical response to a socially and politically organized system of health care that denies resources to those who are unable to afford them. It is a lack of equitable access and ability to pay for screenings an d the potential treatment of cardiovascular disease that results in the suboptimal use of CVD screening resources . R esults from mediatio n pathway analyses support this inference further by revealing that fatalism did not significantly mediate the relationship s between institutional barriers to health care and use of blood pressure or cholesterol screening services. This was not completely unexpected as the relationships between usual source of health care and use of both blood pressure and cholesterol screening services were strong with statistical significance achieved in b oth logistic regressions . Likewise, having health insurance was sig nificantly associated with increased odds of obtaining blood pressure screening. Given the se strong quantitative relationships and the qualitatively generated insights into how lack of health insurance and a usual source of health care lie at the root of C VD related beliefs and screening behavior, the central role that institutional barriers to health care play in the suboptimal use of blood pressure and cholesterol screening services by Latinos is evident .

PAGE 126

113 Quantitative findings from this project show that those with a family income of $19,999 or less had greater than three times the odds of obtaining a cholesterol test compared to those with a family income of $20,000 or greater (OR 3.32, CI: 1.80 6.62). Likewise, participants with a family income of $19 ,999 or less had greater than four times the odds of obtaining a blood pressure test compared to those with a family income of $20,000 or greater (OR 4.73, CI: 2.32 9.63). These results seemingly run contrary to what is expected . However, it is not uncommon to witness J shaped curves as exposure and risk assessment are evaluated from public health perspectives (Jani et al., 2014) . Much of the funding in contemporary public health initiatives are allocated toward interventions that target those who are most vulnerable and at highest risk for negative health outcomes. Frequently, this means programs will specifically target the most socioeconomically disa dvantaged portions of a given population. In the case of blood pressure and chol esterol screening, outreach efforts that focus on offering these services to the poor may increase exposure to screening among this group while simultaneously leaving those whose at risk of less exposure to the benefits of screening . As income increases, access to health care resources and information regarding the benefits of using CVD screening services may lead to higher exposure to screening services thus account ing for the tail of the J shaped curve. Limitations As with all research, there were limitations associated with this project. Perhaps foremost was my status as an outsider to the Latino community. Working among Latinos as a non Hispanic W hite person presents several issues that must be acknowledged and m ade transparent . First, m y position as university student, lead investigator, provide r of funding

PAGE 127

114 and member of the dominant culture afford a measure of privilege that I recognize . Second, as a result of the roles that I occupy, perceptions related to differences in power are likely to occur between research staf f, participants and me as cultural outsider . I recognize that a ssumptions related to my expectations and access to information , personnel, or resources could have remained implicit and affected the ways in which the research was explained to the community and how subject s were recruited and data was collected. In an effort to diminish these concerns, I chose to work with members of the Latino community who have been professionally conducting research and public health dat a collection protocols for the past ten years. Duri ng my time working in public health practice, I was fortunate to work closely alongside these individuals over the course of six years. As a result of their rich experience in both cultural brokerage and cultural competence, some of the se concerns may have been lessened. As a team, we held several meetings prior to beginning work to discuss the project , review scopes of work, discuss timelines and negotiate budget items. Just as important, these meetings included ample time to share both work and personal e xperiences, talk about family and establish rapport. In addition, I maintained close communication with staff throughout each phase of the data collection process to ensure members of the team felt comfortable with the protocol and were ab le to freely addr ess concerns or ask questions. I have found that regularly scheduling time to discuss both challenges and successes in an honest and open way is valuable to everyone involved. Third , d espite my time working and interacting with a multitude of organizations , representatives and individual members of the Latino community, I am not fully fluent in Spanish so remain limited in my ability to read materials presented in this language or fully engage with community members that are monolingual Spanish speakers. Wo rking

PAGE 128

115 alongside bilingual members of the community and establishing clea r protocols for how communication, including both tran slation and transcription, would be handled was essential in this regard. Another li mitation that became apparent during the course of this project was the time that elapsed be tween collection of P hase O ne and P hase T wo data . Due to the partially mixed sequential research design employed , P hase O ne quantitative data w ere collected and ana lyzed in their entirety prior to contacting those who were interested in participating in P hase T a P hase T wo qualitative interview until several months after they completed the P hase O ne questionnair e. Consequently, the potential for recall bias was enhanced. Some participants could not remember how they answered questions on the quantitative survey or what the survey was about. I n an effort to minimize these concerns , we rev iewed answers to survey q u estions with participants prior to beginning in depth interviews when necessary . Nevertheless, it is possible that participant recollections of survey questions or topics impacted the way they engaged in conversations that took place as part of the in dept h qualitative interviews. Concurrent designs in which quantitative and qualitative data collection and analysis take place at the same time may serve to assuage this concern. However, given the need to establish population parameters related to institution al barriers to health care , fatalism and use of screening services, this type of research design was not possible for this project. D ue to the non random purposive sampling design employed , the generalizability of the results to the larger Latino population may be limited . However, employing criterion sampling during Phase One, with clear conditions for participant inclusion and exclusion

PAGE 129

116 serve d to strengthen the appropriateness of the sample . Moreover, the use of purposive maximum variation sampli ng guided by the theoretical framework for Phase Two enabled me to capture a wide range of perspectives relating to fatalism, institutional barriers to health care and use of CVD screening services. More specifically, this particular sampling approach pres ented the opportunity to explore both confirmatory and contradictory cases regarding the phenomena of interest. Remaining open to exploring explanations that r a n contrary to the hypotheses proposed was an important part of the exploratory process and serve d to aid in achieving a more holistic and contextually dependent view of the relationships of interest. Despite the rationale employed for dichotomizing fatalism, choosing not to define this measure as a continuous variable may have resulted in a loss of variability for this particular measure. As a result, variation in outcome measur es in terms of use of cholesterol or blood pressure screenings or potential relationships with institutional barriers to health care may have been subsumed under the two dicho tomized groups. Future work employing a multi item measure of fatalism with the aim of evaluating the relationship between fatalism and health behavior should carefully consider necessary sample size through a priori power an alyses such that variability wi ll be preserved by operationalizing fatalism as a continuous variable. Lastly, due to 276 participants out of the total sample of 386 choosing not to be contacted for the qualitative interviews, the potential for selection bias in Phase Two was enhanced. Previous research suggests that participants who choose to take part in health research rate more highly in the domains of altruism, intellectual and health related motivations. Financial motivation may also be a factor in participation (Soule et al., 2016).

PAGE 130

117 T he qualitative sampling schemes employed in this project were guided by theor y driven concepts and p urposive maximum variation sampling techniques designed to maximize the range of perspectives investigated. The four groups of interest comprising the qualitative sample were specifically created in an effort to explore both confirmatory and contradictory cases regarding the impact of institutional barriers to health care on endorsement of fatalistic attitudes and beliefs. As a result, the impact of selection bias on qualitative interviews may have been minimized. Despite these limit ations, the mixed meth ods data generated from this research provide highly contextualized findings related to the relationship between institutional barriers to health care , CVD fatalism and use of blood pressure and chol esterol screening services by Latinos in the greater Denv er metropolitan area. The use of validated and reliable quantitative measures in P hase O ne and the rigorous approach to P hase T wo sampling, data collection and data analysis increase the trustworthiness of the findings presented herein and provide a strong foundation for future research on this topic. Conclusion Focusing on fatalism rather than the pervasive institutional barriers to health care and social processes that produce it pardons the presence of unjust social, economic and institutional barriers t hat inform the health beliefs and decision making processes of Latinos. More importantly, it perpetuates a false public health narrative that perception of control over CVD as external to the individual is part of a cognitive irrationality that plagues Lat ino culture to the detriment of individual responsibility for health. For study participants, health beliefs and behavior are embedded in a larger social context in which real life circumstances and constraints such as limited access to health care resources, the

PAGE 131

118 presence of disease, lack of monetary resources, and systems o f faith serve to organize how one perceives control over CVD and interacts with the medical establishment. The failure to account for these c omplexities by isolating fatalism as a problematic and faulty trait of Latinos only serves to further marginalize t his community and does little to increase use of screening services . Moreover, recklessly labeling the Latino population without giving voice to their experience does a terrible disservice to the tenets of public health and the mission to serve the needs o r our most vulnerable members of society. The decontextualization of fatalism leads to the characterization of Latinos as possessing health beliefs and attitudes that stand in the way of effor ts to increase adoption of prevention and treatment protocols b y members of this population. As a result , durable perceptions of Latinos as fatalistic among the bio medical and pub lic health communities may serve to further harm this population as health care providers may be less likely to refer members of this popula tion for screening tests or engage in the exchange of health related information and treatment options. More broadly, the uncritical view that Latino culture is plagued by irrational dispositions towards the prevention and treatment of disease may lead to a lack of investment in soci al, economic and health institution resources for this population; reserves of which are sorely needed. CVD fatalism is not a faulty personality trait that needs correction via educational initiatives. Rather, it is an inherently nuanced disposition rationally produced at the intersection of dynamic social processes that limit options to preserve and protect cardiovascular health. If we hope to account for the presence of fatalism towards CVD screening and treatment, we must first account for the way institutional barriers to health care are embodied in the very outlook of those that must endure them. Removin g these

PAGE 132

119 barriers and establishing more equitable access to culturally competent health care resources for Latinos is the first step to instilling a sense of control over CVD and hopefulness for the future. Recommendations Interdisciplinary health researchers intere sted in examining fatalism among Latinos must account for contextual variables in the ir conceptual models . Failure to do so perpe tuates lines of inquiry that ignore the social processes that critically inform the formation and organization of fatalistic expression. Mixed methods and qualitative research designs that incorporate avenues for exploring how and why fatalism arises, including how it impacts the health decision making and behavior of minority populations , are essential to uncover ing the social foundations and variability present in this disposition. Future research on fatalism and the use of screening resources by Latinos should consider how the removal of institutional barriers to health care impact s the intersectional formation of this disposition and use of screening services across various disease categories. Do material deficits or the presence of dis ease and the logics of religion continue to intersect with perception of control such that fatalism is produced independently of institutional barriers to health care? If so, in what forms does fatalism appear? Research on fatalism and health behavior should also consider whether interruptions to health insurance and payment services serve to initiate forms of fatalistic discourse and whether these forms impact use of health care resources over time. Lastly, futu re research would make valuable contributions to our overall understanding of fatalism and health behavior by investigating whether generational differences in the presence or forms of fatalism exist among Latinos, including how contextual variables such

PAGE 133

120 a s socioeconomic instability, personal or family experience with disease, access to healthcare and quality of life measures may intersect with perceptions of control to produce fatalistic expression or affect health behavior. Institutional barriers to hea lth care play a critical role in the explanation of suboptimal use of CVD screening services by Latinos. Public health efforts designed to sustainably increase use of blood pressure and cholesterol screening services by Latinos must fully account for the i mportance of unimpeded access to quality health insurance and the establishment of a usual source of health care or medical home. Attempts to incorporate educational initiative s designed to change or alter fatalistic beliefs or attitudes among Latinos m a y be counterproductive and harmful. Fatalistic discourse represents rational expression s of limited agency and the search for control over disease based upon religious dictates. Efforts to treat fatalism as a uni dimensional cognitive trait and obstacle to screening requiring change through education is ill advised and may only serve to further marginalize vulnerable members of the Latino community. Public health interventions that seek to increase use of screening services by Latinos must include h onest and transparent conversations with community members about how they view using such services. These conversations should proceed in a culturally competent manner and with leaders from the Latino community acting as lead liaisons. Offering blood pr essure and c holesterol tests is encouraged and should be viewed as part of the larger effort to assist members of the Latino community with access to these critical services. Data collected as part of this research shows that only 27 percent of participants have hea lth insurance. Moreover, fully 30 percent of all participants lack a usual source of health care. Despite progress in reducing uninsured rates among Latinos as a result of the

PAGE 134

121 Affordable Care Act, recent data from the Ka i ser Family Foundation (2018) show that Latinos continue to have the highest uninsured rates in the country. In 2017, the number of uninsured people increased by nearly 700,000 people, the first increase since implementation of the A ffordable C are A ct (KFF, 2018) . Ongoing efforts to alter the Affordable Care Act or to make receipt of Medicaid contingent on work status may further ero de coverage gains seen under this legislation . In an effort to reduce expenditures related to the burden of heart disease carri ed by Latinos and increase use of preventive screening services , p olicymakers must continue to support current measures that seek to expand Medicaid offerings in all fifty states. Ideally, policy that implements a version of universal , single payer healthc are while retaining no cost preventive care, including screening for blood pressure, cholesterol and diabetes, protections for preexisting conditions, and Medicare based cost controls for pr escription drugs would make large segments of the Latino populatio n going without health insurance or a usual source of health care a thing of the past. are evidence of the highly disputed and contested space that is American healthcare. T he trans formation of the US healthcare system from one in which access to health care resources remains limited to those who can afford them to one in which all residents are able to access quality health care regardless of ability to pay would render institutional barriers to health care such as lack of health insurance and no usual source of health care among Latinos inconsequential. Most importantly, it would serve to fulfill the call of The Universal Declaration of Human Rights and the wealthiest and most diverse nation on

PAGE 135

122 earth: sanction health care as a right for all rather than a privilege enjoyed by the most fortunate.

PAGE 136

123 R EFERENCES Abraido Lanza, A. F., Viladrich, A., Florez, K.R., Cespedes, A., Aguirre, A. N., & De L a Cruz, A. A. (2007). Commentary: fatalismo reconsidered: A cautionary note for health related research and practice with Latino populations. Ethnicity & Disease 17 , 153 158. Acevedo, G.A. (2005). multidimensional alienation theory. Sociological Theory 23(1): 75 85. Agresti, A. & Finlay, B. (2009). Statistical methods for the social sciences . Upper Saddle River, New Jersey: Pearson Prentice Hall. Ajzen, I. (1988). Attitudes, personality and behavior. Milton Keynes: Open University Press. Alcala´, H. E., Albert, S.L., Roby, D.H., Beckerman, J., Champagne, P., Brookmeyer, R., Prelip, M. L., Glik, D.C., Inkelas, M., Garcia, R.E. & Ortega, A.N. (2015). Access to care and cardiovascular disease p revention : A cross s ection al study in 2 Latino c ommunities . Medicine 94( 34 ). Amarenco P , Abboud H, & Labreuche J, (2014). Impact of living conditions and socioeconomic characteristics on cardiovascular risk in ischemic stroke patients. Int J ournal of Stroke 9:1065 1072. American Heart Association (2019) Health Topics: Heart Health Screenings. heart.org. Accessed on June 9, 2019. American Heart Associatio n (2016). Conditions: Heart Health Screenings. heart.org. Accessed on February 12, 2017. American Heart Association (2010). Heart Disease and Stroke Statistics. At a Glance, 1 36. American Heart Association (2007). FACTS bridging the gap: CVD health disparities partnership for prevention. P reventive Care: A National Profile on Use, Disparities, and Health Benefits 5 : 1 4. Angel, R.J., Angel, J.L. & Lein, L. (2003). The Health Care Safety Net for Mexican Americans, Center for Health and Social Policy Working Paper Series , No. 03 1. Austin, TX: Lyndon B. Johnson School of Public Affairs, The University of Texas at Austin. Ariza, M. A., Vimalananda V. G., & Rosenzweig , J.L. (2010). The economic consequences of diabetes and cardiovascular disease in the United States. Reviews in Endocrine and Metabolic Disorders 11 :1 10.

PAGE 137

124 Artiga S., Orgera , K. & Damico, A. (2019). Disparities Policy: Changes in health coverage by race and ethnicity since implementation of the ACA 2013 2017. Henry J. Kaiser Family Foundation, Accessed on June 8, 2019. Ashida, S., Wilkinson, A.V. & Koehly, L.M. (2010). Mo tivation for health screening: e valuation of social influence among Mexican American adults. American Journal of Preventive Medicine . 38 (4):396 402. ATLAS.ti. Version 7.0 [Computer sofware] (2012) Berlin, Scientific Software Development Balcazar , H., Castro, F.G., & Krull, J.L. (1995). Cancer risk reduction in Mexican American women: the role of acculturation, education, and health risk factors. Health Education Quarterly 22:61 84. Balluz, L.S., Okoro, C.A., & Strine, T.W. (2004). Access to hea lth care and preventive services among Hispanics and Non Hispanics: United States, 2001 2002. National Center for Chronic Disease Prevention and Health Promotion, CDC Morbidity and Mortality Weekly Report; 53 (40):937 941. Bandura, A. (1977). Self efficacy: toward a unifying theory of behavior change. Psycholo gi cal Review 84, 191 2 15. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory . Englewood Cliffs, NJ: Prentice Hall . Behbakht , K., Lynch, A., Teal, S., Degeest, K., & Massad, S. (2004). Social and cultural barriers to papanicolaou test screening in an urban population. Obstetrics and Gynecology 104 :1355 1361. Be Healthy Denver (2014). Health of Denver report: Community health assessment. BeHealthyDenver.org. Accessed on April 08, 2016. Pgs. 1 35. Benjamins, M.R. & Middleton, M. (2019). Perceived discrimination in medical settings and perceived quality of care: a population based study in Chicago. PLoS O NE 14(4): e0215976. https://doi.org/10.1371/journal.pone.0215976 . Benkert, R., Peters, R.M., Clark, R., & Keves Foster, K. (2006). Effects of perceived racism, cultural mistrust and trust in providers on satisfaction with care. Journal of the National Medical Association 98 (9):1532 40 . Blackwell ,D.L., Lucas, J.W., & Clarke, T.C. (2014). Summary health statistics for U.S. adults: National Health Interview Survey, 2012. National Center for Health Statistics . Vital Health Stat istics 10(260). Blanchard J., & Lurie N. (2004). R E S P E C T: patient reports of disrespect in the health care setting and its impact on care. Journal of Family Practice 53 (9):721 30.

PAGE 138

125 Bowlin, S.J., Morrill, B.D., Nafziger, A.N., Jenkins, P.L., Lewis, C. & Pearson, T.A. (1993). Validity of cardiovascular disease risk factors assessed by telephone survey: The behavioral risk factor survey. Journal of Clinical Epidemiology 46 (6):561 571. Brownson, R.C., Jackson Thompson, J., Wilkerson, J.C., & Kiana, F. (1994). Reliability of i nformation on chronic disease risk factors collected in the Missouri behavioral risk factor surveillance system. Epidemiology 5 (5):545 549. Burgess, D.J., Ding, Y., Hargreaves, M., van Ryn, M., & Phelan, S. (2008). The association between perceived discrimination and underutilization of needed medical and mental health care in a multi ethnic community sample. Journal of H ealth Care for the Poor and U nderserved 1 9:894 911. of health insurance on access to care. Inquiry 35 :389 397. Calanan, R., Gujral, I., Jensen, L., Juhl , A., Kuhn, S., & Wilde, M. (2011) At the heart of the matter: 2011 report on heart disease and stroke in Colorado. Colorado Department of Public Health and Environment. 1 98. Castle, S.E. (1994). The (re)negotiation of illness diagnoses and responsibility for child death in rural Mali. Medical Anthropology Quarterly 8(3):314 335. Centers for Disease Control and Prevention (2013). No usual source of health care among adult s aged 18 64, by selected characteristics: United States, average annual, selected years 1993 1994 through 2011 2012. Trend Tables, Table 73. Centers for Disease Control and Prevention. (2011 a ). CDC health disparities and inequalities report United States . Morbidity and Mortality Weekly Report 60 :1 114. Centers for Disease Control and Prevention (2011 b ). Behavioral Risk Factor Surveillance System Survey Questionnaire . Atlanta, Georgia: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention . Chen, F.M., Fryer, G.E., Phillips, R.L., Wilson E., & Pathman, D.E. (2005). Patients beliefs about racism, preferences for physician race, and satisfaction with care. Annals of Family Medicine 3 (2):138 43. Cohn, L.D. & Esparza Del Villar, O.A. (2015). Fatalism and health behavior: A meta analytic review. Colección Reportes Técnicos de Investigac ión ISBN: 978 607 7953 80 7 Serie s ICSA, Vol. 26. ISBN: 978 607 520 145 0 . Universidad Autonoma de Ciudad Juarez.

PAGE 139

126 Collins, T.C., Clark, J.A., Petersen L.A., & Kressin, N.R. (2002). Racial differences in how patients perceive physician communication regarding cardiac testing. Medical Care. 40 (1):I27 34. Collins, K. M. T., Onwuegbuzie, A. J., & Sutton, I. L. (2006). A model incorporating the rationale and purpose for conducting mixed methods research in s pecial education and beyond. Learning Disabilities: A Contemporary Journal 4 ; 67 100. Colorado Health Institute (2017). insurance gains. Findings from the 2017 Colorado health access survey. coloradohealthinstitute.org. September . 1 30. Corbie Smith, G., Flagg, E.W., Doyle, J.P., & O'Brien, M.A. (2002). Influence of Usual Source of Care on Differences by Race/Ethnicity in Receipt of Preventive Services. Journal of General Internal Medicine 1 7 :458 464. Cuéllar, I. , Arnold, B., & González G. (1995). Cognitive referents of acculturation: Assessment of cultural constructs in Mexican Americans. Journal of Community Psychology . 23:339 355. Davidson, J.A., Kannel, W.B., Lopez Candales, A., Morales, L., Moreno, P.R., Ovalle, F., Rodriguez, C.J., Rodbard, H.W., Rosenson, R.S., & Stern, M. (2007). Avoiding the looming Latino/Hispanic cardiovascular health crisis: A Call to Action. Ethnicity and Disease 17 :568 573. Davison, C., S. Frankel & Davey Smith , G. (1992). The l imits of l ifestyle: Re a ssessing f atalism in the p opular c ulture of i llness p revention. Social Science and Medicine 34(6):675 685. DeVoe, J.E., Fryer, G.E., Phillips, R. & Green L. (2003). Receipt of preventive care among American Journal of Public Health, 93(5):786 791. Drew, E.M. & Schoenberg, N.E. (2011). Deconstructing fatalism: Ethnographic perspectives Medical Anthropology Quarterly 25 (2):164 182 Durkheim, E. (1968). Suicide, a Study in Sociology . New York: Free Press. Engebretson, J., Mahoney, J.S., & Walker, G. (2005). Participation in community health screenings: a qualitative evaluation. Journal of Community Health Nursing 22(2):77 92. Espinosa de los Monteros, K. & Gallo, L.C. (2013). Fatalism and cardio metabolic dysfunction in Mexican American women. International Journal of Behavioral Medicine . 20(4):487 94. doi: 10.1007/s12529 012 9256 z.

PAGE 140

127 Espinosa de los Monteros, K. & Gallo, L.C. (2011). The relevance of fatalism in the study of I nternational Journal of Behavioral Medicine 18 :310 318. Ettner, S.L. (1999). The relationship between continuity of care and the health behaviors of patients: Does having a usual physician make a difference? Medical Care 37 (6):547 555 . Facione, N.C., Miaskowski , C., Dodd, M.J., & Paul, S.M. (2002). The self reported likelihood of patient delay in breast cancer: New thoughts for early detection. Preventive Medicine 34 :397 407. Fairchild, R.M. (2015). Fatalism and health behaviors: exploring the context for clinician patient interactions. Annals of Nursing and Practice 2(4):1032. Finegold, K. & Gunja, M.Z. (2014). Survey data on health insurance coverage for 2013 and 2014: ASPE issue brief. Department of Health and Human Services . 1 15. Finney Rutten, L .J., Agunwamba, A.A., Beckjord, E., Hesse, B.W., Moser, R.P. & Arora, N.K. (2015). The relation between having a usual source of care and ratings of care quality: Does patient centered communication play a role? Journal of Health Communication: Internatio nal Perspectives 20(7):759 765. Flores, A. (2017). How the US Hispanic population is changing. Pew Research Center . Accessed on June 7, 2019 Freeman, H. (1989). Cancer and the socioeconomically disadvantaged. CA Cancer Journal for Clinicians. 39:266 88 . Garce s, I.C., Scarinci, I.C., & Harrison, L. (2006). An examination of socio cultural factors associated with health and health care seeking among Latina immigrants. Journal of Immigrant Health 8 :377 385. Garcia, M.L., Talavera, G.A. Keir, K.B., Falcon, A.P. & Castaneda, S.F. (2017). Hispanic community engagement (HCE): pilot study to reduce hypertension among low income Mexican Americans. SM Journal of Community Medicine 3:1024. Glanz, K., Rimer , B.K., & Viswanath, K. (2008). Health Behavior and Health Education: Theory, Research, and Practice (fourth edition). San Francisco, CA:Jossey Bass. Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitativ e research . Chicago: Aldine. Gorin S.S. (2005). Correlates of colorectal cancer screening compliance among urban Hispanics. Journal of Behavioral Medicine 28:125 137.

PAGE 141

128 Guendelman, S., & Wagner, T. H. (2000) Health services utilization among Latinos and white non Latinos: Results from a national survey. Journal of Health Care for the Poor and Underserved 11 (2);179 194. Hamdy, S.F. (2009). Islam, fatalism, and medical intervention: Lessons from Egypt on the cultivation of forbearance (sabr) and reliance on God (tawakkul). Anthropological Quarterly 82(1):173 196. Harmon, M.P., Castro, F.G., & Coe, K. (1996). Acculturation and cervical cancer: knowledge, beliefs, and behaviors of Hispanic women. Women Health 24(3):37 57. Haviland, M.G., Morales, L.S., Dial, T.H., & Pincus, H.A. (2005) Race/ethnicity, socioeconomic status, and satisfaction with health care. American Journal of Medical Quality 20 (4):195 203. Haviland, M.G., Morales, L.S., Reise, S.P. & Hays, R.D. (2003). D o health care ratings differ by race or ethnicity? Joint Commission Journal on Quality and Patient Safety 29 (3):134 45. Hayes, A.F. & Little, T.D. (2018). Introduction to mediation, moderation and conditional process analysis: a regression based approach. The Guilford Press, New York. Hayes, S.L., Riley, P., Radley, D.C. & McCarthy, D. (2015). Closing the gap: past perf ormance health insurance in reducing racial and ethnic disparities in access to care could be an indication of future results. The Commonwealth Fund. Issue Brief 1805(5)1 12. Henry J. Kaiser Family Foundation (2018). Key facts about the uninsured populat ion. Electronic Document. https://www.kff.org/uninsured/fact sheet/key facts about the uninsured population/ . Accessed June 9, 2019. Henry J. Kaiser Famil y Foundation (2018). Key facts about the uninsured population. Electronic Document. https://www.kff.org/uninsured/fact sheet/key facts about the uninsured p opulation/ . Accessed April 26, 2019. Hsieh, F.Y., Bloch, D.A., & Larsen, M.D. (1998). A simple method of sample size calculation for linear and logistic regression. Statistics in Medicine 17 : 1623 1634. GmbH. Hubbell FA, Chavez LR, Mishra SL, & Burciaga Valdez R. (1996). Differing beliefs about breast cancer among Latinas and Anglo women. Western Journal of Medicine . 164:405 409. IBM Corp. Released 2012. IBM SPSS Statistics for Macintosh, Version 21.0. Armonk, NY: IBM Corp.

PAGE 142

129 Jani, B.D., Cavanagh, J., Berry, S.J., Der, G., Sattar, N. & Mair, S.F. (2014). Revisiting the j shaped curve, exploring the association between cardiova scular risk factors and concurrent depressive symptoms in patients with cardiometabolic disease: Findings from a large cross sectional study. BMC Cardiovascular Disorders 14(139):2 8. Johnson, R. B., Onwuegbuzie, A. J., & Turner, L. A. (2005, April). Mixed methods research: Is there a criterion of demarcation? Paper presented at the annual meeting of the American Educational Research Association , Montreal, Canada. Judge, T.A., Erez, A., Bono, J.E., & Thoreson, C.J. (2002). Are measures of self esteem, neuroticism, locus of control, and generalized self efficacy indicators of a common core construct? Journal of Personality and Social Psychology 83(3):693 710 Jurkowski, J.M. (2006). Nativity and cardiovascular disease screening practices. Journal of Immigrant Health 8 :339 346 Jurkowski, J.M. & Johnson, T.P. (2005). Acculturation and cardiovascular disease screening practices among Mexican Americans living in Chicago. Ethnicity & Disease. 15 : 411 417. and child survival among the Zaramo of Tanzania. Medical Anthropology Quarterly 22(1):67 93. Kandula, N.R., Hasnain Wynia, R., Thompson, J.A., Brown, E.R., & Baker, D.W. (2009). Association between prior experiences of discrimination and patients attitudes towards health care providers collecting information about race and ethnicity. Journal of General Interna l Medicine 24 (7):789 94. Katch, H. and Mead, H. (2010). The role of self efficacy in cardiovascular disease self management: A review of effective programs. Patient Intelligence (2);33 44. Keeley, B. Wright, L. and C.M. Condit. (2009). Functions of Health Fatalism: Fatalistic t alk as f ace s aving, u ncertainty m anagement, s tress r elief and s ense m aking. Sociology of Health & Illness 31(5):734 747. Keller, S.C., Silberberg, M., Hartmann, K.E., & Michener, J.L. (2010). Perceived discrimination and use of health care services in a North Carolina population of Latino immigrants. Hispanic Health Care International 8( 1): 4 13. LaVeist T., Rolley , N.C., & Diala, C. (2003). Prevalence and patterns of discrimination among US health care consumers. International Journal of Health Services 33 (2):331 44. Laws, M.B. & Mayo, S.J. (1998). The Latina breast cancer control study, year one: factors pred icting screening mammography utilization by urban Latina women in Massachusetts (1998). Journal of Community Health 23(4):251 267.

PAGE 143

130 Leech, N. L., & Onwuegbuzie, A. J. (2007). An array of qualitative data analysis tools: A call for qualitative data analysis triangulation. S chool Psychology Quarterly, 22 , 557 584. DOI:10.1037/1045 3830/22/4/557 Leech, N. L., & Onwuegbuzie, A. J. (2010). Guidelines for conducting and reporting mixed research in the field of counseling and beyond. Journal of Counseling and Develop m ent , 88 , 61 69. doi:10.1002/j.1556 6678.2010.tb00151.x Levya, B., Allen, J.D., Tom, L.S., Ospino, H., Torres, M.I. & Abraido Lanza, A.F. (2014). Religion, fatalism and cancer control: A qualitative study among Hispanic Catholics. American Journal of Health Behavior 38(6):839 849. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage. Livingston, G., Minushkin, S., & Cohn, D. (2008). Hispanics and health care in the United States: Access, information, and knowledge. Pew Hispanic Center and Robert Wood Johnson Foundation Joint Report , 1 78. Lockwood, D. (1992). Solidarity and Schism: The Problem of Disorder in Durkheimian and Marxist Sociology. Oxford, UK: Clarendon/Oxford University Press. Lopez McKee G, McNeill JA, Badder J, Morales P. (2008). Comparisons of factors affecting repeat mammography screening of low income Mexican American women. Oncology Nursing Forum 35:941 7. Luquis, R.R., & Villanueva Cruz, I.J. (2006). Knowledge, attitudes, and perceptions about breast cancer and breast cancer screening among Hispanic women residing in south central Pennsylvania. J ournal of Community Health 31 (1):25 42. Madsen, W. (1973). Mexican Americans of south T exas (Case studies in cultural anthropology). New York. Holt, Reinhardt, and Winston. Mays, V.M., Jones, A., Delany Brumsey, A., Coles, C. & Cochran, S.D. (2017). Perceived discrimination in healthcare and mental health/substance abuse treatment among Blacks, Latinos and Whites. Medical Care 55(2): 173 181. Maxwell, J., Cort s, D.E., Schneider, K.L., Graves , A. & Rosman , B. (2011). Massachusetts' health care reform increased access to care f or Hispanics, but disparities r emain . Health Affairs 30(8):1451 1460 doi: 10.1377/hlthaff.2011.0347

PAGE 144

131 Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, Das SR, de Ferranti S, Despr s J P, Fullerton HJ, Howard VJ, Huffman MD, Isasi CR, Jim nez MC, Judd SE, Kissela BM, Lichtman JH, Lisabeth LD, Liu S, Mack ey RH, Magid DJ, McGuire DK, Mohler ER III, M oy CS, Muntner P, Mussolino ME, Nasir K, Neumar RW, Nichol G, Palaniappan L, Pandey DK, Reeves MJ, Rodriguez CJ, Rosamond W, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Woo D, Yeh RW, Turner MB; on beha lf of the American Heart Associati on Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics 2016 update: a report from the American Heart Association. Circulation. 2016; 133(4):e38 e360. Muhib, F.B., Lin, L.S., Stueve , A., Ford, W.L., Miller, R.L., Johnson, W.D., & Smith, P.J. (2001). Community intervention trial for youth (CITY) study team: A venue based method for sampling hard to reach populations. Public Health Reports, Volume 116 (1): 216 222. National Cholesterol Education Program (2001). Detection, evaluation and treatment of high blood cholesterol in adults (adult treatment panel III): executive summary. National Heart Lung and Blood Institute, National Institutes of Health NIH Publication No. 01 3670. 1 30. Neff, J. A., and Hoppe, S. K. (1993). Race/ethnicity, acculturatio n, and psychological distress: Fatalism and religiosity as cultural resources. Journal of Community Psychology 21:3 20. Nichter, M. (2002). The social relations of therapy management. In New Horizons in Medical Anthropology. Mark Nichter and Margaret Lock, eds. Pp. 81 110. London: Routledge. Office of Minority Health (2019). Policy and data; minority population profiles: Hispanic/Latino Americans. US Department of Health and Human Services . Accessed on June 8, 2019. Onwuegbuzie, A. J., & Collins, K. M. T. (2007). A typology of mixed methods sampling designs in social science research. The Qualitative Report , 12 (2). Retrieved January 13, 2009, from http://www.nova.edu/ssss/QR/QR12 2/onwuegbuzie2.pdf Onwuegbuzie, A. J., & Johnson, R. B. (2006). The validity issue in mixed research. Research in the S chools, 13 (1), 48 63. Onwuegbuzie, A. J., & Leech, N. L. (2006). Linking research questions to mixed methods data analysis procedures. The Qualitative Report, 11 , 474 498. Ortega, A.N., Rodriguez, H.P. & Vargas Bustamante, A. (2015). Policy dilemmas in Latino health care and implementation of the affordable care act. Annual Review of Public Health , 36:525 544.

PAGE 145

132 Otero Sabogal R, Stewart R, Sabogal F, Brown BA, Perez Stable EJ. (2003). Access and attitudinal factors related to breast and cervical cancer screeni ng: why are Latinas still underscreened? Health Education Behavior , 30(3):337 359. Palacio M., Reynolds, R., Drisko, J., Lucero, C., Hunt, C., & Phi, K. (2009) Racial and ethnic health disparities in Colorado 2009. Colorado Department of Public Health an d Environment . Patton, M.Q. (2002). Qualitative research & evaluation methods . Thousand Oaks, CA: Sage. Perez Stable EJ, Sabogal F, Otero Sabogal R, Hiatt RA, McPhee SJ. (1992). Misconceptions about cancer among Latinos and Anglos. J ournal of the A merican M edical A ssociation . 268:3219 3223. Pineda Ford, D. (2011). [Community Heart Health Actions for Latinos at Risk]. CREA Results. Unpublished raw data. Ramirez, J. R., Crano, W.D., Quist, R. , Burgoon, M., Alvaro, E. M., & Grandpre, J. (2002). Effects of fatalism and family communication on HIV/AIDS awareness variations in Native American and Anglo parents and children. AIDS Education and Prevention , 1 4(1): 29 40. Rodríguez, M.A.,Vargas Bustamante, A., & Ang, A. (2009). Perceived Quality of Car e, Receip t of Preventive Care, and Usual Source of Health Care Among Undocumented and Other Latinos J ournal of Gen eral Intern al Med icine 24(Suppl 3):508 13 Rodriguez, C.J., Allison, M., Daviglus, M.L., Isasi, C.R., Keller, C., Leira, E.C., Palaniappan, L. , Piña, I.L., Ramirez, S.M., Rodriguez, B., & Sims, M. (2014) Status of Cardiovascular Disease and Stroke in Hispanics/Latinos in the United States : A Science Advisory From the American Heart Association Circulation . 2014;130:593 625.) Rodriguez, C.J., Cai, J., Swett, K., Gonzalez, H.M., Talavera, G.A.,Wruck, L.M., Wassertheil Smoller, S., Lloyd Jones, , D., Kaplan, R. & Daviglus, M.L. (2015). High Cholesterol Awareness, Treatment, and Control Among Hispanic/Latinos: Results From the Hispanic Community Health Study/Study of Latinos . J ournal of the Am erican Heart Assoc iation 2015;4: e001867 doi: 10.1161/JAHA.115.001867 Public Health Reports. (126), 130 135. Rotter, J.B. (1954). S ocial learning and clinical psychology . Englewood Cliffs, NJ: Prentice Hall. Rotter, J.B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80 (whole No. 609. 1).

PAGE 146

133 Rotter, J.B. (1982). The development and applications of social learning theory: selected papers. Brattleboro, VT: Praeger. Rotter, J.B., Chance, J. & Phares, E . J. (1972) . Applications of a social learning theory of personality . New York: Holt, Rhinehart & Winston. Saha, S., Arbelaez, J. J. & Cooper, L.A. (2003). Patient physician relationships and racial disparities in the quality of health care. American Journal of Public Health 93 (10):1713 1719. Salinas, J.J., de Heer, H.D., Lapeyrouse, L.M., Heyman, J.M. and Balcazar, H.G. (2015). Insurance status is a greater barrier than income or acculturation to chronic disease screening in the Mexican origin population in El Paso, Texas. Hispanic Health Care International 13(4):197 208. Shen, L., Condit, C.M. & Wright, L. (2009). The psychometric property and validation of a fatalism scale. Psychology and Health 24 (5), 597 613. Shepherd, S.M., Willis Esqueda, C., Paradies, Y., Sivasubramanium , D., Sherwood, J. & Brockie, T. (2018). Racial and cultural minority experiences and perceptions of health care provision in a mid western region. International Journal for Equity in Health 17(33):1 10. Shive, S.E., Ma, G., Tan, Y., Toubbeh, J.I., Para meswaran, L., & Halowich, J., (2006) Racial differences in preventive and complementary health behaviors and attitudes. Journal of Health Disparities Research and Practice 1 (1):75 92. Solis, J.M., Marks, G., Garcia, M., & Shelton, D. (1990). Acculturation, access to care, and use of preventive services by Hispanics: findings from HHANES 1982 84. American Journal of Public Health 80 :11 19. Sorkin, D.H., Ngo Metzger, Q., & De Alba, I. (2010). Racial/ethnic discrimination in health care: Impact on perceived quality of care. Journal of General Internal Medicine 25 (5):390 39 6. Sorlie , P.D., Allison, M.A., Aviles Santa, M.L., Cai, J., Daviglus, M.L., Howard, A.G., Kaplan, R., Lavange, L.M., Raij, L., Schneiderman, M., Wassertheil Smoller, S. & Talavera, G.A. (2014). Prevalence of hypertension, awareness, treatment and control in the H ispanic community health study/study of Latinos. American Journa l of Hypertension 27(6):793 800. Soule, M.C., Beale, E.E., Suarez, L., Beach, S.R., Mastromauro, C.A., Celano, C.M., Moore, S.V. & Huffman, J.C. (2016). Understanding motivations to particip ate in an observational research study: Why do patients enroll? Social Work and Health Care 55(3):231 246.

PAGE 147

134 Straughan, P.T. and A. Seow (1998). Fatalism reconceptualized: A concept to predict health screening behavior. Journal of Gender, Culture and Health 3:85 100. Stepanikova , I. & Oates, G.R. (2016). Dimensions of racial identity and perceived discrimination in health care. Ethnicity and Disease 26(4):501 512. Suarez, L., Nichols, D., Roche, R.A., & Simpson, D.M. (1997). Knowledge, behaviors and fears concerning breast a nd cervical cancer among older low income Mexican American Women. American Journal of Preventive Medicine 13:137 142. Sudano, J.J. & Baker, D.W. (2003). Intermittent Lack of Health Insurance Coverage and Use of Preventive Services . American Journal of Public Health 93 ( 1 ). Summer, L. (2002). Screening for chronic conditions. Health Policy Institute, Georgetown University. Data Profile 1: Retrieved February 01, 2015 from, https://hpi.georgetown.edu/agingsociety/pubhtml/screening/screening.html Teddlie, C., & Tashakkori, A. (2009). Foundations of m ixed m ethods research: Integrating quantitative and qualitative approaches in the social and behavioral sciences. Thousand Oaks, CA: Sage. Torres, M.E., Bellinger, J.D., Probst, J.C., Harun, N., & Johnson, A.O. (2007). Use of Preventive Services Among Hispanic Sub Groups: Does One Size Fit All? South Carolina Rural Health Research Center: Report , 1 50. Triedman, N., Silvernale, R., Bontrage r, J., Clark, B., Downs, A., Foster, C., Goeken, D., Hanel, J., Keeney, T., Lueck, M., Schmitt, S., & Wear, H. (2015) Colorado health access Survey: A new day in Colorado. The Colorado Health Institute , 1 35. T rivedi, A.N., & Ayanian, J.Z. (2006). Perceived discrimination and use of preventive health services. Journal of General Internal Medicine 21 (6):553 8. U.S. Census Bureau (2015). Selected Population Profile in the United States: 2015 American Community Survey 1 Year Estimates. Population Tables S0201. Accessed on June 9, 2019. U.S. Census Bureau, Population Division. Table 3. Annual Estimates of the Resident Population by Sex, Race, and Hispanic Origin for the United States: April 1, 2000 to July 1, 2010 (NC EST2009 03). U.S. Preventive Services Task Force. Screening for High Blood Pressure: Clinical Summary of U.S. Preventive Services Task Force Recommendation . December 2007. http://www.uspreventiveservicestaskforce.org/uspstf07/hbp/hbpsum.htm ).

PAGE 148

135 Van Houtven, C.H., Voils, C.I., Oddone, E.Z., Weinfurt, K.P., Friedman, J.Y., Schulman, K.A., & Bosworth, H.B. (2005). Perceived discrimination and reported delay of pharmacy prescriptions and medical tests. Journal of General Internal Medicine 20 (7):578 83. van Ryn, M. (2002). Research on the provider contribution to race/ethnicity disparities in medical care. Medical Care . 1140 51. van Ryn, M., & Burke, J. (2000). The effect of patient race and socio economic status on physicians perceptions of patients. Social Science and Medicine 50 (6):813 28. Ward, E., Jemal, A., Cokkinides, V., Singh, G., Cardinez , C., Ghafoor, A., & Thun, M. (2004). Cancer disparities by race, ethnicity, and socioeconomic status. CA A Cancer J ournal for Clin icians (54):78 93. Weech Maldonado, R., Morales, L.S., Elliott, M., Spritzer, K., Marshall. G., & Hays, R.D. (2003). Race/ethnicity, language, and patients assessments of care in Medicaid managed care. Health Services Research 38 (3):789 808. Wallston, K.A. (1997). Perceived co ntrol and health behavior. In A. Baum, C. McManus, S. Newman, J. Weinman, and R. West (Eds). Cambridge Handbook of Psychology, Health and M edicine (Vol. 1). Cambridge, England: Cambridge University Press

PAGE 149

136 APPENDIX A: SURVEY QUESTIONNAIRE Project Title: R eintroducing Context to the Study of Fatalism and the CVD Screening Behavior of Latinos PI: Craig G. Kapral Protocol #: 15 0451 Kapral: Appendix A . University of Colorado Denver: Health and Behavioral Sciences Phase One Survey Location: PID#: Interviewer: Eligibility 1. Are you Latino, Hispanic or of Spanish origin? Yes No Stop Survey 2. What is your age? Interviewer Note: (Eligible= 40 75 yrs: date of birth= 1942 1977) Enter 3. Are you a resident of Denver County? Yes No Stop Survey 4. Have you EVER or are you now taking a course , class or program about heart disease or diabetes? Yes Stop Survey No Now I would like to ask you some questions about general health conditions. Has a doctor, nurse, or other health professional EVER told you that you had any of the 5. Have you EVER been told by a doctor, nurse or other health professional that you had a heart attack also called a myocardial infarction? Yes Stop Survey No

PAGE 150

137 Read only if necessary: professional. Not Sure 6. Have you EVER been told by a doctor, nurse or other health professional that you ha ve coronary heart disease? Yes Stop Survey No Not Sure 7. Have you EVER been told by a doctor, nurse, or other health professional that you had a ngina ? Interviewer Note: By angina, we mean chest pain and/or discomfort confirmed by a doctor. Yes Stop Survey No Not Sure 8. Have you EVER been told by a doctor, nurse, or other health professional that you have high blood pressure? Yes No Not Sure 9. Have you EVER been told by a doctor, nurse or other health professional that your blood cholesterol is high? Yes No Not Sure 10. Have you EVER been told by a doctor, nurse, or other health professional that you have diabetes? Yes No Not Sure Screening Now I will ask you two questions about whether you have had certain screening services given by health care professionals . 11 . Have you had your blood pressure checked by a doctor, nurse or health care professional in the last two years ? Yes No

PAGE 151

138 12 . Have you had your blood cholesterol checked by a doctor, nurse or health care professional in the last two years ? Yes No Institutional Barriers Now I am going to ask you some questions about your health care. 13 . Do you have any kind of health care coverage, including health insurance, prepaid plans such as HMOs, or government plans such as Medicare? * If yes, answer 1 3 a . Yes No 13 a. Type of health insurance? I am going to read you a list of different types of health insurance coverage. Please tell me if you currently have any of the following types of insurance. Do not include any health insurance plans that cover only ONE type of service, like plans for dental car e or prescription drugs. Do you have : Interviewer note: If necessary: Health insurance through your work or through a union: This could be through COBRA, through a former employer, or a retiree benefit. Medicare: Medicare is health insurance for persons 65 years old and older or persons with disabilities. This is a red, white, and blue card and includes Medicare parts A, B, or C. Medicaid: This is a program for low income families with children, seniors, and people with disabilities. CHP+: This is a Colorado program for low and moderate income children under age 19, and pregnant women who live in families that earn more than is allowed to be in Medicaid. Private health insurance: Health insurance bought directly from Anthem, Kaiser, United, or another company, or bought through an insurance broker. Health Insurance through Your work or Someone through belonging to a Union Y e s N o Medicare Y e s N o Affairs, Military Health Care or TRICARE Y e s N o Indian Health Service Y e s N o

PAGE 152

139 Medicaid Y e s N o Any Other Type of Private Health Insurance Y e s N o 14. Was there any time in the past 12 months that you did not have insurance coverage? Yes No 15. About how long has it been since you last visited a doctor for a routine checkup? Interviewer Note: A routine checkup is a general physical exam, not an exam for a specific injury, illness, condition mammography Within the Past year Between One and Two Years Ago Between Two and Five Years Ago More than Five Years Ago Sure Never 16 . Is there a place that you USUALLY go to when you are sick or need advice about your health? Yes There is NO Place There is MORE THAN ONE place Sure

PAGE 153

140 17. Do you have one person you think of as your personal doctor or health care provider? *If No, answer 17 a. Yes No 17a. Is there more than one person who you think of as your personal doctor or health care provider? Yes No 18 . Have you been treated unfairly (or discriminated against), at a health care facility in the past two years? *If yes, answer 18 a. Yes No 18 a. Which of the following best describes the reason you felt you were treated unfairly (or discriminated against) ? Interviewer Note: check all that apply. Race/Ethnicity Language Gender Insurance Other 19. Was there a time in the past 12 months when you needed to see a doctor but could not? *If yes, answer 19 a. Yes No 19a. Why you were unable to see the doctor when you needed to? Interviewer note: Read each and m ark all that apply. Cost Too Much No Insurance Insurance Problem (coverage ended, unable to verify insurance) Provider who Accepts Insurance Transportation Problems

PAGE 154

141 Could Not Get an Appointment Child Care Issues Could Not Time Off of Work Other Reason Sure Fatalism ( Shen , Condit , and Wright ( 2009 ) Now I am going to read you some statements about beliefs and attitudes towards disease. Think about the next several statements in terms of heart disease and tell 20 . If someone is meant to get a serious disease, it that disease anyway. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 21 . If someone is meant to get a serious disease, they will get it no matter what they do. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 22 way they were meant to die. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 23 . If someone is meant to have a serious disease, they will get that disease. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree

PAGE 155

142 24 . If someone has a serious disease and gets treatment for it, they will probably still die from it. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 25 . If someone was meant to have a serious tell them to do, they will get the disease anyway. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree Now I will read some statements about beliefs and attitudes towards health and life. For each 26 . How long I live is predetermined. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 27 . I will die when I am fated to die. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 28 . My health is determined by fate. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 29 . My health is determined by something greater than myself. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 30 . I will get diseases if I am unlucky. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree

PAGE 156

143 31 . My health is a matter of luck. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 32 . How long I live is a matter of luck. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 33 . I will stay healthy if I am lucky. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 34 . Everything that can go wrong for me does. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 35 . I will have a lot of pain from illness. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 36 . I will suffer a lot from bad health. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 37 . I often feel helpless in dealing with the problems of life. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree 38 in life. Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree

PAGE 157

144 39 . There is really no way that I can solve some of the problems I have Strongly Disagre e Disa gree Neit her Agre e Strong ly Agree Demographics The next few questions will ask about your personal characteristics . No information you provide will be used to identify you. 40 . What is your gender? Male Female Other 41 . What language do you prefer to speak? Spanish English Other 43 . What country were you born in? * If outside of US, answer 42 a. US Mexico Other 43 a. How long have you lived in the US? 2 yrs or less 3 to 5 yrs 6 to 10 yrs 10 yrs or more 44 . What level of education do you have? Less than HS Diploma HS Diploma/GED

PAGE 158

145 Some College College Diploma More than College Diploma 45 . Which of the following best describes your household income last year? Annu ally Month ly Weekly A Under 14,99 9 Under 1,250 Under 288 B 15,00 0 to 19,99 9 1,251 to 1,667 288 to 385 C 20,00 0 to 29,99 9 1,668 to 2,499 386 to 576 D 30,00 0 to 59,99 9 2,450 to 4,999 577 to 1,153 E 60,00 0 or more 5,000 or more 1,154 or more

PAGE 159

146 APPENDIX B: SEMI STRUCTURED INTERVIEW GUIDE Project Title: Reintroducing Context to the Study of Fatalism and the CVD Screening Behavior of Latinos Principal Investigator: Craig G. Kapral Protocol #: 15 0451 Research Question : H ow may institutional barriers to health care inform the development and expression of fatalism or other attitudes towards CVD and the subsequent use of cholesterol and blood pressure screening services among Latinos in Denver, Colorado ? BEFORE: o Set up recorder and test o on the recording o Begin Script: Welcome and thank you for your participation today. My name is Craig Kapral and I am a graduate student at The University of Colorado Denver conducting a study in partial fulfillment of the requirements for the degree of Do ctor of Philosophy in Health and Behavioral Sciences. Thank you for completing the survey questions with _________________, and this follow up interview. The interview today will take about 60 minutes and will include questions about your experiences with health insurance, having a regular place to go see a health care provider and how these relate to your ideas of heart disease and using blood pressure and cholesterol testing services. At the end of the interview we have $_________ to give you to thank yo u for your time. I will ask you to sign a Receipt of Payment Form before you leave. I would like your permission to tape record this interview so I may accurately document the information you share. If at any time during the interview you wish to discont inue the use of the recorder or the interview itself, please feel free to let me know. All of your responses are confidential. Your responses will remain confidential and will be used to develop a better understanding of how you and other members of the Latino community view access to health insurance and having a place to go see a health care provider and how these may relate to using blood pressure and cholesterol screening services. The purpose of this study is to increase our understanding of these is sues so we can do a better job of promoting heart health among Latinos. Do you have any questions or concerns before we begin? Then with your permission we will begin the interview. Introductory Questions (Brief) Tell me a bit about yourself. How long have you lived in Denver? What do you like about living here? Tell me a little about your family? Children? Grandchildren?

PAGE 160

147 Tell me how you feel about your health? CVD Fatalism/Control A lot of people have heart attacks these days. Why do you think that happens? o Prompt: Have you ever considered it might be fate or that it was meant to happen? Tell me how you feel about being able to prevent heart disease or heart attacks? Do you think people have control over getting heart disease or having h eart attacks? Why? Why not? If someone gets heart disease or has a heart attack, do you think it was their fate to have this happen? Why? Why not? Why do you think people in the Latino community use blood pressure and cholesterol screening tests? Why do use blood pressure and cholesterol screening tests? Health Insurance Tell me about your history with health insurance/coverage. What have your experiences been with having or not having health insurance? How has y our experience with health care insurance made you feel about being able to take care of your [heart] health? How might not having health insurance impact your beliefs or attitudes towards being able to prevent heart disease or having a heart attack? o Prob e: C ontrol? Probe: How does not having health insurance relate to feeling that you can control or prevent a heart attack or heart disease? How might not having health insurance impact your beliefs or attitudes towards getting your blood pressure and chole sterol tested by a health care provider? o Probe: C ontrol? Probe: How does not having health insurance relate to feeling that you can/want to have your blood pressure and cholesterol tested? Usual Source of Health Care Tell me about where you usually go to get health care when you or your family needs it.

PAGE 161

148 How would you describe the relationship you have with the health care provider(s) at this location? o Do you see the same person when you go there? How does it make you feel to see the same health care provider vs. a different one each time you go to a health care facility? How has your experience with the providers at this location made you feel about being able to take care of your health? your beliefs about being able to prevent he art disease? o Probe: S eeing the same provider? o Probe: C ontrol? How does not having a regular place to get health care relate to feeling that you can control or prevent a heart attack or heart disease? ow might that impact your beliefs or attitudes towards getting your blood pressure and cholesterol tested by a health care provider? o Probe: S eeing the same provider o Probe: C ontrol? How does not having a regular place to get health care relate to feeling that you can control or prevent a heart attack or heart disease? Thanks very much for your time today. Is there anything else you would like to add? We appreciate your input.

PAGE 162

149 APPENDIX C: CONSENT FORM Project Title: Reintroducing Context to the Study of Fatalism and the CVD Screening Behavior of Latinos Principal Investigator : Craig G. Kapral Protocol #: 15 0451 Consent Form You are being asked to be in a research study. This form provides you with information about the study. A member of the research team will describe this study to you and answer all of your questions. Please read the informatio n below and ask questions about anything you do not understand before deciding whether or not to take part. Why is this study being done? This research study plans to learn more about how barriers to health care and attitudes towards heart disease affect the way people use heart disease screening services. Barriers to health care include not having health insurance or a regular health care provider and feeling like you have been treated unfairly in a health care setting. Screening services include t ests for blood pressure and cholesterol. This research will be used to address: the health service needs of the Latino community, health care providers, and p olicymakers so people of Latino background can have better access to and use of heart disease scre ening services. You are being asked to be in this research study because you were identified as an adult aged 40 to seventy five years old, self identified as someone of Latino background, and live in Denver County. You are also being asked to be in this study because you are not currently or have not in the past participated in a heart disease or diabetes prevention program and do not have heart disease. What happens if I join this study? If you join this study you will answer several survey questions about health care resources, your attitudes towards heart disease, and use of heart disease screening services. The survey will take about 10 to 15 minutes. At the end of the surv ey, the interviewer will ask if you are interested in participating in a longer interview about these topics at a different time. If you choose to participate in the interview, you will provide your first name and phone number so that the interviewer can r each you to set up a time and place to talk. You will also give permission for your interview information (e.g., your audio recorded and/or written transcripts of your discussions) to be used in our study. If you participate in the interview you will be gi ven $15.00 for your time. You will NOT be asked for details about your health. We will make every effort to protect your privacy and keep all information you give confidential. We will make audio recordings but will not keep any information that can be used to identify you.

PAGE 163

150 What are the possible discomforts or risks? It is possible that talking about your health care resources use of health care may make you uncomfortable or upset. If you become distressed and would like further help we will connect yo u to appropriate resources. What are the possible benefits of the study? This study is designed for the researcher to learn more about how barriers to health care and attitudes towards heart disease affects the use of heart disease screenin g services by pe rsons of Latino background. Heart disease screening services includes tests for high blood pressure and high cholesterol. The information from this study will be used to inform public health workers, health care providers, health researchers, and policyma kers of ways to help persons of Latino background receive heart disease prevention services so that they may stay healthy. Besides providing an opportunity for you to answer questions about these topics and discuss these issues openly, there is no direct benefit to the recruits. Indirectly, participants may benefit by connecting with researchers from the University of Colorado Denver, who may be able to increase awareness of barriers to health care and attitudes towards heart disease in the Denver Latino community. Who is paying for this study? This research is being paid for by the Department of Health and Behavioral Sciences at the University of Colorado Denver. Will I be paid for being in the study? Will I have to pay for anything? You will not be paid to answer survey questions. However, If you choose to participate in the interviews you will be paid $15.00 for your time. It will not cost you anything to be in the study. Is my participation voluntary? Your participation in this research study is VOLUNTARY. You have the right to choose not to take part in this study. If you choose to take part in this study, you have the right to stop at any time. If you refuse or decide to withdraw from this s tudy later, you will not lose any benefits or rights to which you are entitled. Who do I call if I have questions? The researcher or Principal Investigator carrying out this study is Craig G. Kapral, a graduate student in the Department of Health and Behavioral Sciences at the University of Colorado Denver. You may ask any questions you have of the interviewer now. If you have questions later, you may contact Craig G. Kapral (303 880 3384 or (craig.kapral@ucdenver.edu)

PAGE 164

151 You may have questions about your rights as someone in this study. You can call or email Craig G. Kapral with questions. You can also call the Color ado Multiple Institutional Review Board (COMIRB), who reviews and approves the research conducted at the University of Colorado Denver (303 315 2732). Who will see my research information? We will do everything we can to keep your records a secret. It c annot be guaranteed. Others may look at the records that identify you and the consent form signed by you. They are : Federal agencies that monitor human subject research, The Colorado Multiple Institutional Review Board, The research team doing this stud y Regulatory officials from the U niversity of Colorado Denver who want to make sure the research is safe. The results from this research may be shared at a meeting and/or published in articles. Your name will be kept private when information is presented. The interviews will be audio recorded and written down. Audio recordings will be protected under lock and key in a locked file cabinet. When the interviews are finished and the audio recordings are written down, the audio recordings will be d eleted or destroyed. Agreement to be in this study I have read this paper about the study or it was read to me. I understand the possible risks and benefits of this study. I know that being in this study is voluntary. I choose to be in this study: I w ill get a copy of this consent form. Signature: Date: Print Name: Consent form explained by: Date: Print Name: