Citation
Another missed opportunity

Material Information

Title:
Another missed opportunity accuracy of parental knowledge of immunization status during hospitalization
Portion of title:
Accuracy of parental knowledge of immunization status during hospitalization
Creator:
Wheeler, Mary Catherine
Publication Date:
Language:
English
Physical Description:
viii, 188 leaves : illustrations, forms ; 29 cm

Subjects

Subjects / Keywords:
Immunization of children ( lcsh )
Vaccination of children ( lcsh )
Immunization of children ( fast )
Vaccination of children ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 175-188).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Public Administration.
General Note:
School of Public Affairs
Statement of Responsibility:
by Mary Catherine Wheeler.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
36455649 ( OCLC )
ocm36455649
Classification:
LD1190.P86 1996d .W44 ( lcc )

Full Text
ANOTHER MISSED OPPORTUNITY:
ACCURACY OF PARENTAL KNOWLEDGE
OF IMMUNIZATION STATUS
DURING HOSPITALIZATION
by
Mary Catherine Wheeler
B. S. N., University ofNorth Carolina, 1976
M. P. A., Fairleigh Dickinson University, 1982
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirement for the degree of
Doctor of Philosophy
PublicAdministration
1996


This Thesis for the Doctor of Philosophy
degree by
Mary Wheeler
has been approved
by
Peter deLeon
J3 ( f9V<
Date
James K. Todd


Wheeler, Mary Catherine (Ph.D., Public Administration)
Another Missed Opportunity: Accuracy of Parental Knowledge of
Immunization Status During Hospitalization
Thesis directed by Professor Peter deLeon
ABSTRACT
Immunizations have been among the most successful preventive
health interventions in the United States and other countries. Concern
exists in the United States that recent outbreaks of vaccine-preventable
diseases, as well as the number of children under the age of two not
adequately immunized, signal the existence of major underlying problems
in immunization policy. There is a lack of effectiveness in national, state
and local public health programs to administer these preventive services
to children. One reason for under-immunization may the opportunities
missed to vaccinate children during hospitalization. While hospitalization
may not be the most convenient time for parents to recall current
immunization status or for the health care provider to assess and
administer immunizations, this time represents missed opportunities to
vaccinate. This research examines the parental knowledge of
in


immunization status and the functions of the health care provider to
assess and administer immunizations during hospitalization as an
opportunity to improve childhood immunization status. The specific aims
of this research are to: 1) determine the accuracy of parental knowledge
of immunization status; 2) determine the accuracy of the assessment of
immunization statues by the health care provider; and 3) determine the
resulting use of the assessment during hospitalization. The specific aims
and hypotheses are addressed through a survey design using the medical
record, a parent questionnaire, and a follow up verification with the childs
primary care provider. The sample consists of 197 in-patient admissions,
with a birthdate after August 1,1993.
The study results indicate that 59 percent of the parents knew their
childs immunization status. The admission history obtained from the
health care provider was accurate in 64 percent of the records. However,
an immunization was administered prior to patient discharge in only one
case. Therefore, this indicates that the missed opportunity to vaccinate
children during hospitalization was not related to parental knowledge nor
to the assessment by the health care provider. The missed opportunity
was the result of a failure by the health care provider to view


hospitalization as an event that could improve the immunization status of
preschool children.
This abstract accurately represents the content of the candidates thesis.
I recommend its publication.
Signed
v


CONTENTS
CHAPTER
1. INTRODUCTION....................................................I
Research Assumptions.........................................II
Thesis Outline...............................................13
2. REVIEW OF THE LITERATURE.......................................15
Barrier to Immunization......................................21
Health Economic Theory.......................................29
Summary of Health Economics...............................43
Public Choice Theory.........................................45
Summary of Public Choice..................................51
Health Belief Model..........................................52
Other Behavioral Models...................................62
Summary for the Health Belief Model.......................63
General Summary..............................................64
3. STUDY DESIGN...................................................65
Objectives...................................................66
Materials and Methods
68


Study Population..................................................68
Questionnaire Design..............................................70
Study Period and Sample Size.........................................75
Definitions..........................................................79
Research Methodology.................................................79
Statistical Methods..................................................82
4. FINDINGS................................................................85
Introduction.........................................................85
Statistical Analysis.................................................86
Description of the Sample............................................87
Immunization Status..................................................98
Documentation.......................................................100
Admission History in Medical Record..............................100
Discharge Orders in the Medical Record...........................105
Parental Knowledge..................................................106
Factors Associated with Delayed Immunization........................118
Summary.............................................................128
5. ANALYSIS...............................................................130
Introduction........................................................130
Health Care Economic................................................133
vii


Public Choice
135
Health Belief Model.....................................138
Summary.................................................142
Limitations.............................................143
Recommendations for Public Policy.......................144
Missed Opportunity....................................151
Education.............................................153
Summary for Public Policy.............................159
General Summary.........................................160
Action Items............................................161
Public Policy.........................................161
Internal Policy Recommendations.......................162
APPENDIX
A EDUCATIONAL TOOL
B. QUESTIONNAIRE 3.1 MEDICAL RECORD
C. QUESTIONNAIRE 3.2 PARENT
D. QUESTIONNAIRE 3.3 PARENT AFTER INFORMATION
E. QUESTIONNAIRE 3.4 VERIFICATION WITH PCP/CLINIC
F. QUESTIONNAIRE 3.5 POST DISCHARGE
BIBLIOGRAPHY
171


CHAPTER 1
INTRODUCTION
Good health is the bedrock on which social progress is
built. A nation of healthy people can do those things which
make life worthwhile and as the level of health increases
so does the potential for happiness.
B. Watkins, 1978
This quotation underscores the dual values that underpin
intervention by modem governments in the pursuit of health policies. Good
health is seen as a positive benefit to both the individual and to the nation.
The provision of health services has been justified in terms of fulfilling
individual needs and in terms of being necessary for collective national
progress. The development of health services and health policies
contributes, in part, to the growth of modem welfare states.
Beneath the surface of the commitment to improve health lies
uncertainties and ambiguities as to what constitutes appropriate actions for
national or state childhood immunization policies. The reality of a universal
immunization policy for all infants may be far removed from an enforceable
policy, but aiming toward those goals unquestionably leads to improvements
in the nations health status.
1


This research represents an attempt to evaluate the opportunity
that hospitalization offers for educating parents about immunizations, as
well as a potential opportunity for the health care provider to vaccinate
children under the age of 24 months, based on information provided to
them at the time of their childs admission. The disciplinary
expectations of this research will be its contributions to the theories of
public choice, health economics, and the Health Belief Model in terms of
how they relate to the pertinent policymaking decisions. The research
also contributes to the existing knowledge of the problem identification
with immunization practices in the pre-school population. Finally, it
addresses the public policy issues attendant to early vaccination and their
implementation.
Since 1991, national immunization standards have reached the 90 to
95 percent range at the time of school enrollment at age five (CDC, 1994).
However, approximately 40 percent of children under the age of two have not
received appropriate immunizations (CDC, 1994). This is a critical lapse, for
approximately 80 percent of childhood vaccine doses are recommended
during the first 2 years of life.
Immunizations have been among the most successful preventive
health interventions in the United States and other countries. However,
2


concern now exists in the United States that recent outbreaks of vaccine-
preventable diseases, as well as the number of children under the age of two
not adequately immunized, signal the existence of major underlying problems
in immunization policy and a lack of effectiveness in national, state, and local
public health programs to administer these preventive services to children.
The use of vaccines to prevent infectious diseases is one of the
principal defining hallmarks of modem preventive medicine and public
health. During the twentieth century, improvements in immunization
practices in the United States have dramatically reduced the number of
reported cases of potentially serious diseases such as diphtheria,
measles, mumps, pertussis, poliomyelitis, rubella, Haemophilus influenzae
type b, and tetanus. Despite such reductions, unacceptable outbreaks of
vaccine-preventable diseases continue to occur. Recent outbreaks of
measles and pertussis-particularly in under-immunized and un-
immunized children-raise continued concern regarding the effectiveness
of existing childhood immunization practices. Such concerns are
stimulating action from public health professionals, physicians, and
ultimately public policymakers to improve early childhood immunization
programs in this country.
3


Currently, a marked and regrettable disparity exists between the
immunization levels of school-age children and preschool-age children in
the United States. This is due to the strict enforcement of state
immunization laws that assure that school-aged children are
appropriately immunized with the recommended dosage of critical
vaccines. Although nationally the proportion of school-aged children fully
immunized is at least 95%, the story is quite different among preschool
children (Hinman, 1990). Complete immunization coverage for 2-year-old
children has been reported to be as low as 50% in some inner-city
populations (National Vaccine Advisory Committee, 1991).
Current data suggest that the recent increase in the incidence of
measles in various parts of the country has occurred most often in un-
immunized children (Hinman, 1990; Farizo, 1992). Unless efforts are
made to improve the immunization levels of preschool-age children, this
specific population remains susceptible to a number of vaccine
preventable infectious diseases.
It has become clear that reaching these goals requires coordinated
collaboration among public and private health care providers. Thus, the
necessary public health efforts must occur at the national, state, and local
levels. Consensus among these governmental levels are clearly
4


necessary to develop and implement programs for overcoming current
barriers to childhood immunization.
Recent outbreaks of preventable diseases, previously presumed
to have been eradicated, have brought the issue of childhood
immunization back into national prominence. To increase public
awareness of these growing health problems and improve vaccination
rates for American preschoolers, President Clinton proposed a Childhood
Immunization Initiative (CM) soon after taking office.
Begun in early 1993, CM was a multifaceted policy that strove to
develop new and better vaccines in addition to enhancing immunization
delivery. CM seeks to establish a comprehensive national regime to
ensure that all children complete the recommended series of 11 to 15
vaccine doses by their second birthday. Cll establishes immunization of
preschool-age children a national priority. It targets 1996 as the year for
achieving 90% immunization levels for one dose of measles-mumps-
rubella, three doses of oral polio, three doses of diphtheria-tetanus-
pertussis, and three doses of Haemophilus influenzae type b vaccine and
70% coverage for three doses of hepatitis B vaccine (CDC, 1994).
A key element of the vaccination initiative was the Comprehensive
Child Immunization Act of 1993, unveiled by Donna Shalala, Secretary of
5


the Department of Health and Human Services (DHHS). This $1.1 billion
program proposed to increase the rate of immunization and was the first
attempt at health care reform on by the Clinton Administration. By
choosing a seemingly mainstream an issue such as childrens
immunizations, the Clinton Administration anticipated little conflict in
passing the legislation. As Hinman (1991) points out, Immunization
ranks with motherhood and apple pie as a value for most Americans
(p.560). Yet, much discussion surrounded the proposed legislation. In
fact, lack of agreement on the cause of the immunization problem rapidly
resulted in polarization that crossed party lines. Public health experts and
members of the pharmaceutical industry joined the fray, citing parental
issues, inattention of providers, lack of access, and cultural barriers,
among other disputatious factors.
Even when a version of this legislation eventually passed as part of
the Omnibus Reconciliation Act of 1993 (HR 2264, P.L. 103:66),
consensus was not reached on a clear, single definition of the
immunization problem in the United States. Because this complex
process of problem definition can affect the formulation and
implementation of legislation-as well as the potential for successful
health outcomes-it is imperative that providers be precisely aware of the
6


issues involved. Only then can they assert their position in policymaking
decisions and lobby for the constituents they serve.
It is an accepted fact that pre-school children continue to contract
vaccine-preventable diseases in spite of the substantial resources being
allocated to public and private immunization programs. The goal announced
in 1990 by Lewis Sullivan, Secretary of Health and Human Services in the
Bush Administration, for improving disease prevention was outlined in the
Healthy People 2000 report "....increase to at least 90 percent the number of
children who receive the basic immunization series by age two" (U.S.
Department of Health and Human Services, 1990). The plan to accomplish
the goal was less prescriptive, as reflected in the current national average of
approximately 60 percent vaccine coverage for this age population.
The President and Mrs. Clinton have cited the vaccine program as
an example of what governments should do to promote public health.
However, reservations about the Cll program have been reported in a
1994 study conducted by the General Accounting Office (GAO). The
GAO concluded, among other findings, that there was insufficient
evidence that the cost of vaccine to parents was a major barrier to
childrens timely immunization (Pear, 1995). An objective of the present
7


research is to identify other alternative barriers, especially assuming
costs are not a key obstacle to timely immunization.
Reasons for delayed immunization have been linked to a number
of factors, including parental issues, as well as barriers in the health
delivery system (Hinman, 1990). A list of conceivable barriers to early
vaccination include vaccine cost, continuously changing
recommendations for vaccine administration, educational and cultural
differences affecting parental knowledge and attitude about vaccinations,
missed opportunities by health care providers for administering
immunizations, inaccessible health care systems, and misunderstandings
by parents and providers concerning the true contraindications for
administering immunizations (Cutts, 1993; Abbotts, 1993; Petry, 1993).
Overcoming these multiple barriers is vital to improving vaccination levels
in preschool-age children requiring a variety of interventions and
educational programs directed at parents and health care providers
(CDC.1994).
Issues specific to childrens health must be addressed in the
United States through cooperative policy reforms. The United States has
been through a decade of health policy that emphasized market systems
featuring deregulation, individual responsibility, and volunteerism. The
8


impression of most analysts is that these approaches have not worked
(Enthoven, 1988; Fein, 1986; Fuchs, 1988). Important indicators of child
health have not improved significantly. To some, the federal government
is regarded as the solution, not the problem. To others the
implementation of health policies and programs are regarded as a matter
for local communities, not a federal government endeavor. Regulatory
authority, resource distribution, and payment mechanisms are negotiated
at the highest levels of government. Community government implements
a residual obligation to serve as the guarantor provider for essential
services, even when private provider systems are extensively utilized.
The weakness of the federal governments role in formulating
explicit child health policy is manifested in procedures that led to
development of health objectives for the nation for the year 2000.
However, no working group was established to recommend objectives for
children. This omission reflected the mistaken belief that the specific
needs of children would be addressed if broad topical priorities were
considered for entire populations. Health and Human Service planners
finally capitulated to pressures from childrens advocates and aggregated
age-specific objectives that were derived from the topical priorities.
9


The underlying theme that pervades the objective for the year 2000
is that if society takes care of everybody, of course, children will be
included. The immunization medical records for pre-school age children,
do not support that claim. Therefore, the formulation of an explicit,
measurable national health objective across the entire population is
potentially a powerful instrument of public policy. It could identify
priorities, support them with appropriate regulatory and financing
authority, and provide public accountability for provider systems and
expenditures, especially for children less than 24 months of age.
However, prospects are dimmed by indications that federal agencies
regard the objectives as guidelines for action by state and local
communities, not as governmental initiatives at the national level.
All children upon entering school must have documentation of
appropriate immunizations (excluding religious or health exemptions).
However, universal immunization of school age children (5 years and older)
will not prevent the life threatening diseases of infancy (0-2 years) for which
vaccines exist. From an economic standpoint, Colorado hospital charges
exceeded the $18,000,000 mark for diagnoses related to vaccine
preventable diseases from January 1,1988 to June 30, 1994 (Colorado
Health Association, (CHA), 1995). Failure to appropriately immunize
10


affects the well being of children, as well as having an economic impact
on the health care delivery system.
A recurrent theme of Colorado Governor Romers Administration has
been: ...to make Colorado the very best place to raise a child"
(Brown,1996). The January 11,1996 State of the State address, Romer's
message was kids matter." The Democratic Governor asked the lawmakers
to focus on four challenges, the first being to protect young children.
Fourteen million dollars has been set aside to accomplish this health care
goal. Mrs. Bea Romer is dedicated to the mission of childhood immunization
within the State of Colorado. The continuous interest that has been
demonstrated by the Romer Administration towards childhood immunization
and health has made the Cll 90% immunization goal attainable for pre-school
children in 1996 for the State of Colorado. The value of such a thesis is to
provide policy makers with the empirical data needed to move forward with
successful strategies to accomplish this goal.
Research Assumptions
There are two major assumptions of this research.
1. For many children, hospitalization represents a major interaction with
the health care delivery system and represents an opportunity to
educate parents/guardians about immunizations, as well as to update
the needed immunizations for children.
11


2. There is a general expectation on the part of the health care provider that
the parent /guardian is knowledgeable regarding what their child has
received, as well as what is needed for appropriate immunization
schedules.
The normative assumption to this thesis is all children deserve good
health, and freedom from childhood diseases that have been made virtually
obsolete. Moreover, parenting is difficult enough without exposing their
children to these diseases. Based on that consideration, the key variables
in this thesis are parental knowledge of immunizations, the manner in which
the health care provider assesses and acts upon the immunization
information, and the true compliance rate of immunizations for children under
the age of 24 months. This study uses a survey design to collect data from
parents, the medical record, and the primary care provider.
The purpose of the research is to amend the existing body of
knowledge by linking the missed opportunity to access and vaccinate during
hospitalization with the missed opportunity to educate parents about
vaccination at the time of hospitalization. The hospitalized population
considered for this study are in-patient admissions and same-day surgery
admissions at The Childrens Hospital in Denver, Colorado.
12


Therefore, for children under the age of 24 months, the dissertation
asks, does an empirical relationship exist between the parental
knowledge of immunization status, physician accuracy in the
determination of the immunization history during hospitalization, and
vaccination opportunities?
Thesis Outline
This thesis will consist of four additional chapters in addition to this
Introduction. Chapter 2 is dedicated to the review of the literature and
previous research that supports the thesis. Chapter 3 describes the
sample population and methods used for conducting the study. Chapter 4
is the analysis of the collected data. Chapter 5 completes the thesis with
conclusions on how this research can be used to improve pre-school
immunization rates through public policy recommendations.
The Health Care Economics theory will be supported in the
conclusion with data obtained from the Colorado Health Association
(CHA). A five and one-half year retrospective analysis is performed using
the International Classification of Diseases-dth addition (ICD-9) codes for the
nine vaccine preventable diseases. Total charges are calculated for each
disease and can be used as a marker for the estimated cost of hospitalized
13


cases of vaccine preventable diseases. Recommendations are also
supported in the conclusion by using comparative models of successful
immunizations programs. The Kaiser Permanente, South Carolina and
Finland models of immunization practices are used to compare and contrast
supported recommendations.
The Health Belief Model may help explore other barriers in parental
knowledge to understand effective strategies for successful immunization
practices.
14


CHAPTER 2
LITERATURE REVIEW
In the United States (see Table 2.1), children under the age of two
are routinely immunized against nine diseases (vaccine): diphtheria, tetanus,
pertussis (DTP), poliomyelitis (OPV), Haemophilus influenzae type b, (HIB),
hepatitis B (HBV), and measles, mumps, rubella (MMR).
DTP OPV i HIB HBV MMR I
BIRTH X
2 MONTHS - X X X X
4 MONTHS w -#V $ & j > X
6 MONTHS X X X X
15 MONTHS X* X
18 MONTHS A
4-6 YEARS * \/ A X-v ^ W \J" V
TABLE 2.1 immunization schedule for children 6 years and younger
* HIB vaccine (depending on manufacturer, may need additional dose)
15


In 1983, a record low number of cases of measles (1,497) were
reported in the United States. However, in 1990, over 500 cases per week
occurred (CDC, 1994), exceeding the Surgeon General's year-long goal of
500 measles cases nationwide by more than 50-fold (National Vaccine
Advisory Committee, 1991). Moreover, this represented a 52 percent
increase over the 18,193 cases reported in 1989. During the 1989-1990
school year, five school districts in the United States experienced measles
epidemics affecting over 1,200 children in each outbreak (CDC, 1994),
indicating that the dosage schedule for MMR needed to be re-evaluated.
Questions of poor vaccine versus a phenomenon of "waning immunity" (less
protective antibody over time) were addressed. The outcome was a revised
recommendation to administer a second dose of MMR in later years to confer
continued immunity.
By the late 1980s, birth defects from congenital rubella syndrome
almost disappeared in the United States because of the success of rubella
immunization. Two cases were reported in 1989, but that number increased
to 24 in 1991. Concurrently, the number of cases of infectious rubella
increased 500 percent after 1988 to 1,372 cases (Lee, Ewert, Fredrick,
1992).
16


These outbreaks have occurred even though the immunization rates
of school aged children in the United States are among the worlds highest.
A survey in 1990 of kindergarten and first-grade public school students in
eight inner-city areas (Boston, Bronx, Cleveland, Houston, Jersey City,
Phoenix, Pittsburgh, and Seattle) revealed that, although more than 95% had
received measles vaccine by the time of school entry, only 51 % to 81 %
(depending on the city) had received the vaccine before their second birthday
(CDC, 1994). Some had been vaccinated only shortly before entering school
to comply with school immunization requirements and therefore, had been at
a dangerous risk of disease for several years.
The National Health Interview Survey (NHIS), provides quarterly
data to the Centers for Disease Control and Prevention (CDC) that
enables calculations of national vaccine coverage rates (Massey, Moore,
et.al, 1989). It comprises a probability sample of the civilian, non-
institutionalized U.S. population. Quarterly estimates for children aged
19-35 months were based on sample sizes of 483 (third quarter 1993),
490 (fourth quarter 1993), and 608 (first quarter 1994). Table 2.2
demonstrates the vaccination rates as reported by NHIS for children two
years of age.
17


The rates of early childhood immunization in the United States
compare poorly with other Western industrialized nations and even many
developing countries (CDC.1994). In 1988-1989,16 countries had infant
immunization rates higher than the United States, including Bulgaria, Hungry,
Greece, Brazil, China, Mexico, North Korea, Chile, and Romania (Shea,
1991). The World Health Organization (WHO) reported in 1990 that the rate
of measles immunization among one-year olds had reached 87% in India
and 98% in China. It is ironic that immunization programs in developing
countries have achieved these levels of success in light of the situational
difficulties that must be overcome. For example, during the civil war in El
Salvador, two-and three-day cease fires were called to allow childhood
immunization teams safe passage through battle zones (Brown, 1991).
18


TABLE 2.2
Vaccination Levels Among Children Aged 19-35 Months. Bv Selected
Vaccines-United States. Third And Fourth Quarters 1993 And First
Quarter 1994. (CPC. 1995).
: pBrd'Quater' f: <1093 Fpurth Quarter ISRNt (%> First Quarter 1994 (%)
DTP 3 doses 89.9 88.1 87.0
4 doses 74.8 71.6 67.2
Polio 3 doses 80.4 78.5 76.0
H influenzae type b 3 doses 60.3 58.3 70.6
MMR 85.9 86.9 89.6
Hepatitis B 3 doses 15.7 22.5 25.5
Immunization includes not just the vaccine itself, but also the
mechanism or system for their distribution and administration. Many of the
nation's immunization problems originated because the current delivery
system was a conglomerate of complex and uncoordinated efforts in the
public and private sector. Approximately half of the American children
receive their immunization from private physicians. The other half receive
their immunizations in the public sector, usually from the city or county health
19


departments, federally funded community health centers, or public clinics
(Hinman, 1991). Federal funds pay for approximately half of the
immunizations administered via public sector providers through CDC grant
programs, such as Medicaid, Maternal and Child Health Block Grants, with
the remainder funded by state and local governments (Hinman, 1991).
Between 1963-1991, federal funds were not used for the actual
delivery of immunization at the local level. Rather, these funds were used to
purchase vaccines for use in the public sector and to support other
immunization related programs, such as vaccine promotion and surveillance
activities (Orenstein, Atkinson, Mason, 1990). In 1992, additional federal
funds were made available for immunization delivery. The level of funding for
this program has increased dramatically in the past 14 years, from a low of
$5 million in fiscal year 1976 to the fiscal year 1991 appropriation of $185
million (Hinman, 1994). Still, there continues to be limited coordination in the
federal role for vaccine programs, and no mechanism exits to coordinate
efforts between public and private sector organization (Shea, 1991). For
example, no central federal record keeping mechanism tracks a child's
immunization status or monitors immunization rates accurately at either the
national level or in Colorado.
20


Barriers to Immunization
Low immunization coverage has been attributed to difficulties in
reaching certain groups in the population (McConnochie, 1992). In
particular, these groups include the urban poor and racial and ethnic
minorities. Recent evaluations suggest, rather, that the health care delivery
system itself bears much of the responsibility (Orenstein, 1990, Cutts, 1992).
Parents seeking immunization for their children face significant barriers and
obstacles that impede efficient vaccine delivery.
Much of the current knowledge about the risk factors for delayed or
non-immunization comes from research conducted 15 to 25 years ago
(Guyer, 1976), or from more recent studies that have examined the
characteristics of children involved in measles outbreaks (CDC, 1991; CDC,
1990). These studies have variously reported that low levels of parental
education or immunization knowledge, low income, large family size, younger
parental age, using the public health care system, and non-white race are
risk factors or explanation for delayed immunization. One study suggests
that health belief parameters differ between the parents of immunized and
non-immunized children (Markland, 1976); others have concluded that health
21


care system barriers, such as lack of insurance or the need for immunization
appointments, contribute to the immunization problem (Orenstein, 1990).
Many parents and physicians consider immunizations routine. As a
consequence, they fail to discuss this seemingly mundane aspect of health
care. The need for parents to know about immunizations has the following
components:
1. Parents have a general need to know about everything that
happens to their child (Gunby, 1982).
2. They have a need for specific information about each vaccine
and each dose of vaccine (Fulginiti, 1984).
3. They have a need to know the place that vaccines occupy in
their child's total health care and in the health of their community
(Cooper, 1982).
4. There is a need for parents (and subsequently patients) to comply
with recommended schedules and be knowledgeable about their
immunization histories (McCormick, 1981).
Vaccination recommendations have naturally changed with the
improvement and development of vaccine products. Hence, education
pertaining to current recommendations may be a key element identifying
barriers in both parental knowledge, as well as in the health care providers
responsibility to know and verify with the parent It can be argued that failure
to document and treat under-immunization as a medical problem is not in the
22


best interest of children, nor, more indirectly, the parents. In addition, the
lack of health care providers' attention to immunizations may give parents of
hospitalized children the mixed message that immunization is not of major
medical importance for overall child health care.
There are many other barriers to immunization that merit
consideration. The Centers for Disease Control and Prevention (CDC)
surveyed immunization program managers from 54 of the 57 largest
immunization projects in May 1990 to identify barriers leading to low
immunization levels among preschool children (Orenstein, 1991). The major
problems identified in this survey were resource and policy barriers that
limited access to vaccination: immunizations being available by appointment
only; requirements for a physical exam prior to the immunization; requiring a
physician referral; enrollment in a well-baby clinic; and administrative fees.
While there is today no magic formula that could guarantee a perfect
immunization rate, there are opportunities by the health care providers that
represent a "missed opportunity" to vaccinate. A significant proportion of
non-vaccination is the result of the failure of the health system to make use of
every given opportunity for immunization. In looking more specifically at the
"where" and "when" questions of non-immunization, missed opportunities can
be identified at three locations-immunization clinics, health centers, and
23


hospitals (Foster, 1989). The most surprising of these is the non-
immunization rate of eligible children attending immunization clinics. Exit
interviews have shown that many needed vaccinations (as high as 10-20% in
some clinics) are not given. Investigation of these missed vaccinations has
identified five problem areas: the child was not screened; was screened
incorrectly; was screened, but not referred; was referred, but not immunized;
or was given incomplete immunization (Foster, 1989). Equally troubling,
although easier to explain, is the failure to use the opportunities for
immunization at well-child clinics, at sick-child clinics, and at the hospital
outpatient or emergency rooms.
Many poor urban children use emergency departments as their
primary source of health care (Wood, 1990). Indeed, the emergency room
encounter may represent a baby's singular exposure to a health care
provider in early years. Therefore, providing immunizations in emergency
departments may represent the best (maybe the only) opportunity to
increase vaccination coverage in these community. A recent study of
unvaccinated, but vaccine-eligible preschool aged children with measles in
Los Angeles, Dallas, Milwaukee, Chicago, and New York indicated that 38%
had been seen in an emergency department and 24% at a hospital outpatient
department at some time prior to developing measles (Hutchins, 1993).
24


Comparatively, studies in the adult population suggest that emergency
departments can be important sites for pneumococcal immunization (Polis,
1987).
An analogous problem is evident when the medical provider does not
adequately review the child's immunization status and suggests needed
immunizations when the provider is seeing the child for reasons and illness
not related to vaccinations. Hospital visits have characteristically not been
used as an immunization opportunity. In a study conducted at a pediatric
hospital in Tennessee, the admitting physicians failed to document the
immunization status of 22% of the patients (Kum-Nji,1995).
Other factors include excessive interpretation of contraindications to
vaccination and failure to administer all indicated vaccines at a single visit.
For instance, recent publicity of the alleged neurological side-effects of DTP
vaccine has deterred some parents from initiating vaccination, but has also
made physicians uncomfortable administering vaccine to a child with a mild
illness or several at multiple sites. In 1987, The Vaccine Compensation Act
was passed that made available funds for damages to children who were
injured as a result of vaccine administration. The Vaccine Compensation
Act has alleviated some of these physician concerns; however, there
25


continues to be an apparent hesitancy of physicians to vaccinate children
with a minor illness.
The existence and enforcement of laws requiring immunization before
school entry in all 50 states have provided a mechanism for ascertaining with
great confidence the immunization levels of 5-6 year old children. In the
U.S., for the 1989-1990 school year, more than 97% of these children had
documentary evidence of having received a full course each of DTP, MMR,
and OPV, with little variation around the country. Levels nearly as high for
pre-schoolers are reported from Head Start and licensed day-care centers,
where the school immunization requirements also apply (Hinman, 1991). As
mentioned earlier, kindergarten and first-grade public school students in eight
inner-city areas revealed that, although more than 95% had received
measles vaccine by the time of school entry, only 51 % to 81 % (depending on
the city) had received the vaccine before their second birthday (CDC, 1991).
Those who are not immunized on time are likely to be members of
racial and ethnic minorities and to be socio-economically disadvantaged. An
investigation in Chicago demonstrated that only 50% of the students enrolled
in predominantly black and Hispanic schools had been immunized against
measles by their second birthday, compared with 80% of students in
predominantly white schools (CDC, 1989). In predominantly black or
26


Hispanic schools, 25 to 30% of the students had been vaccinated shortly
before entry, presumably to meet school criteria.
The striking disparity between the extraordinary success in assuring
immunization in school age children and unacceptable low levels seen in
inner-city (and in some rural area) preschoolers demonstrates two important
observations. First, the high immunization levels in school children indicate
that parents want to have (or at least do not mind having) their children
immunized. Aggressive enforcement of the school immunization
requirements, which began in the late 1970s, has met little widespread
opposition. In Los Angeles and New York, although thousands of children
were denied entry to school because of a lack of evidence of immunization,
most returned to school within one or two-days, either having found their
immunization records or having been immunized. This would indicate that
access to care was not an obstacle since it took only a few days for the
majority of children to return to school.
Second, since the private and public systems to deliver immunizations
in the United States have the capacity to serve all children (because it
ultimately does so), the financial costs of immunization in either the private or
the public sector also does not pose insurmountable barriers. Thus,
27


appropriate immunization may depend on a highly motivated parent or
provider who views immunization as a priority.
During the 1950s, the Public Health Service was oriented toward
prevention, not the treatment of disease. There was widespread failure of
people to accept disease prevention or screening tests for the early detection
of asymptomatic disease; these included tests or prevention for tuberculosis,
cervical cancer, dental disease, and later polio and influenza. To a large
extent, these factors influenced theories to explain preventive health
behavior. Such theories have to deal with the behavior of individuals who
were not currently suffering disabling disease. While it would not overlook
the potential role of barriers to accepting health services, it would have to
explain the behavior and motivational factors of people who were being
charged with the allocation of health care and those receiving the service.
In American society, it is still the citizens who, through their voice
in the marketplace and voting place, ultimately determine how their
money will be allocated. In 1990, the American public devoted 12.4% of
its gross domestic product (GDP) to the health care industry, thus clearly
indicating the importance it attaches to the industrys output (Marmow,
1985). Health economists have questioned whether the large increases
in total resources deployed in the health arena since the 1960s have
28


resulted In commensurate gains to the consumer through either increased
longevity or reduced morbidity. The economists major contribution is to
suggest how the improved use of resources can result in more desired
outputs.
An advanced technological society, such as the United States, can
no longer rely exclusively on the private sector to provide the goods and
services, including health care, that it needs, wants, and can afford. The
linkage that now ties health care to public policy supports the focus of this
thesis on developing a better understanding of the processes for
appropriated immunization practices.
This section has detailed important research findings regarding
barriers to childhood immunization practices. This research will focus on
the theories of health economics, the Health Belief Model, and Public
Choice Theory to broaden the existing body of knowledge to appropriate
childhood immunization practices.
Health Economic Theory
Economics is concerned with allocation of scarce resources to
alternate, and often unlimited, wants in the most cost-effective manner.
29


The mounting problem of the economics and politics of medical care
today, is basically a problem of resource allocation.
Americans health industry developed primarily in response to
market demand and, as a result, the interplay of supply and demand
resulted in mutually acceptable prices that reflected the actual resource
cost. In the past, individual decisions guided the medical industrys
response in providing needed services so that if the demand for a certain
type of care increased, then more resources would be allocated to
provide that type of care through the market price mechanism. For this
reason, symptomatic care took precedence over preventive care.
Economic demand and medical need are not necessarily the same.
Economic demand represents an intent to pay a given market price. The
market economy presents a formal, rational framework for the problem of
resource allocation defined by economic demand. However, medical
need is not easily defined-its meaning changes from person to person
and varies for the same person over time. Physician and patient
perceptions of an individuals medical need may not be equivalent.
There is no simple method to allocate medical care in accordance with
medical need. The cost of universal immunization for children under the
age of 24 months has an economic impact on health care, both in dollars
30


spent for universal immunization, as well as the health care costs of
vaccine preventable diseases. Health Economic theory will be the
framework for evaluating the cost benefits of immunization practice.
Extensive formal relations between medicine and what is now
economics, existed in the seventeenth and eighteenth centuries than at
any subsequent time until the present. Several physicians made
important contributions to the development of knowledge about the
production, distribution, and consumption of wealth. John Lockes work in
politics and the theory of knowledge was seminal in the history of the
social sciences within the context of his experience as a physician
(Routh, 1975). William Petty in England and Francois Quesnay in
France, both physicians, participated in the development of modern
economic doctrines (Hutchinson, 1964, Mini, 1974). Bernard Mandeville,
a London physician of Dutch origin, has a significant place in the history
of social analysis foreshadowing the elaboration of classical economic
theory (Clark, 1971). These men brought to economics a profound sense
of the value of individual human effort and of the social costs of illness.
From the middle of the eighteenth to the early twentieth century,
however, physicians and economists seem to have been members of
professional networks segregated from one another. Although the
31


classical economists-notably Malthus, Ricardo, and Miil-were deeply
concerned with issues of subsistence, health, and disease, their work
appears to have aroused little interest in the medical profession. Even
when relations should logically have been close-as in movements to
reform sanitation and protection of public health in England and
Germany, for example-there was little connection between medical and
economic ideas (Fox, 1979).
Other early economists, such as Jeremy Bentham and several of
his disciples, especially Edwin Chadwick, went so far as to develop plans
to reorganize public health and medical care and to regulate medical
practice (Cowen, 1969). Benthamite influence was considerable on the
establishment of legislative and administrative standards for public health.
For a century, Benthamite principles dominated arguments for an
expanded public role in the prevention of illness and for compulsory
health insurance in England, and subsequently, in the United States.
Several well-known professional economists in 1886 wrote about
health issues affecting public policy. Among the most prominent were
Richard Ely and John Commons, both from the University of Wisconsin,
and Simon Patten of the University of Pennsylvania (Dunlop, 1965).
Patten advocated the application of theories derived from both the
32


German Historical School and neoclassical welfare economics to
problems of social policy and the quality of life. Two of his students,
William H. Allen and Edward T. Devine, wrote a great deal about what
generations would subsequently call"health economics
Allen, an employee of the Bureau of Municipal Research in New
York City, applied economists concepts of expense to problems of
hospital efficiency and explored the implications of Pattens theories
about potential abundance for health services. Devine, an economist with
an administrative role in professional social work, discussed issues of
public health and entitlement to services in numerous books and papers
(Fox, 1967).
Probably, the best known early health economist was Irving
Fisher of Yale (Fisher, 1966). A pioneer in the application of
econometrics to economic analysis, he actively promoted changes in the
health behavior of citizens and reform in the health policies of
corporations and government agencies. In 1907, he founded the
Committee of One Hundred on National Health to press for the creation of
a department of health in the federal government. Two years later, he
was the principle author of the Report on National Vitality issued by the
National Conservation Commission appointed by President Theodore
33


Roosevelt. He created the Life Extension Institute to persuade insurance
companies that health education and physical examinations for policy
holders would reduce untimely deaths and therefore raise profits (Fisher,
1956). Physicians began to promote his view that health pays' in
medical journals. Fisher was subsequently the president of the American
Association for Labor Legislation during the campaign for compulsory
health insurance from 1912 to 1918 (Fisher, 1956).
The rising cost of medical care as well as an unstable economy
caused people to turn to health insurance as a way to minimize their risk
of financial hardship in case of a serious illness. The health insurance
industry grew from humble beginnings in 1929. By assuring payment for
a hospital stay, health insurance benefited both the hospital and the
patients. Although ostensibly created primarily to assist the patient,
health insurance also was a direct outcome of the financial crises that the
hospital sector experienced during the Great Depression.
Another important development was the establishment of the
Veterans Administration (VA) that offered veterans of World War I
treatment for service-related injuries or diseases at government expense.
This was an important turning point in the history of health policy-for the
34


first time outside of a war situation-total medical care for a designated
group of entitled persons became a public sector responsibility.
During the Depression years, access to health care again became
an issue. In 1927, the Committee on the Costs of Medical Care was
formed to look into what many felt were excessive health care costs and
to explore ways of subsidizing medical care for the poor. The
comprehensiveness and foresight of this committees recommendations
had considerable effect on national health policy. Roosevelts New Deal
was to have a lasting influence on public health care. Title V of the
Social Security Act of 1935 authorized grants to individual states for
maternal and child health programs. Title VI authorized annual
appropriations for the purpose of assisting states, counties, health
districts and other political subdivisions of the states in establishing and
maintaining adequate public health services (House Committee on Ways
and Means, 1935). Working through the states, in collaboration with local
governments, Title V and VI focused on specialized problems such as
communicable disease control and special risks groups such as
women and children.
Following World War II, a major expansion of health insurance
coverage by both Blue Cross-Blue Shield (nonprofit third party
35


reimbursement) and commercial insurance companies (mutual and stock)
took place. In 1950, expenditure for health care in the United States
totaled $12.7 billion or 4.4 % of the gross national product (Gibson,
1982), surely sufficient to attract the attention of the economists. The
Hill-Burton Act of 1946 subsidized numerous hospitals in post-war
construction by providing governmental grants and loans. The health
sector of the general economy expanded rapidly in the 1950s.
Expenditures for new facilities and for hospital services increased sharply
(Wells, Klees, 1980). Physicians income began to rise faster than those
of other professionals (Wells, Klees, 1980). Information collected by
public and private agencies to plan, regulate, and justify new programs
created data resources that could be used for economic analysis.
Events in economics and in the health sector in the 1940s and
1950s seem to account for the rapid expansion of interest among
economists in research on the behavior of the health industry in the
1960s. That interest focused on three features of the American system of
medical care: increases in medical research, hospital construction, and
federal health insurance programs (Marmor, 1970). The federal
government increased its support of medical research primarily through
the National Institutes of Health, the research area of the Public Health
36


Service. One result of the increased in economic research has been
that economists have been consulted with growing frequency by officials
of government and private agencies concerned with health policy.
In 1965, Congress enacted major health care legislation when it
passed the Medicare and Medicaid programs to eliminate economic
barriers to medical care. These sets (both parts A and B) extended
coverage for the elderly and the medically indigent. Medicaid and
Medicare expended billions of dollars on a status quo basis, paying usual,
customary, and reasonably fees to physicians and reimbursing costs for
hospital care. However, by doing so, they surely altered the status quo as
health care costs escalated. These two programs pumped money into an
essentially non-market economy lacking checks, balances, and regulatory
feedback. The Laissez-Faire strategy for financing persisted through the
1960s in to the 1970s. The Nixon administration would attempt wage and
price controls in 1971, reducing the rate of cost escalation for the short
term, but producing a rebound phenomenon that wiped out the savings
and set costs back on a more rapidly rising trajectory.
Events peculiar to economics as a discipline contributed to the
increased interest in health affairs. These events included the growing
prestige of the economics profession, the application of economic
37


analysis to problems of defense and foreign affairs, a growing
professional interest in public finance as a field in which to apply new
theories and methods in welfare economics, and microeconomics (Wells,
Klees, 1980).
The 1960s brought about increasing concern from economists
who applied economics to health. Arrows (1963) Uncertainty and the
Welfare Economics of Medical Care connected the economics of health
to a mathematical economic analysis embedded in welfare economics.
He advocated the paradigm of the market modified by uncertainty that
justified unusual intervention as the basis for studying the difference
between health and other industries.
In the 1970s, the Professional Standards Review Organizations
(PSROs) were established by the federal government to monitor quality of
care. Medicare and Medicaid, once separate organizationally, were
technically joined in what is now known as the Health Care Financing
Administration (HCFA) of the Department of Health and Human Services.
How the extent of health insurance coverage affects the demand for
medical services has been a key issue in the American debate over
medical care financing.
38


The Theory of Demand refers to the quantity of a good or service
that a consumer will purchase at different prices during a given time
period. To help describe these problems, the federal government
sponsored a large -scale social experiment, the Rand Health Insurance
Experiment (HIE) (Newhouse, 1987,1993). The HIE was intended to be
a definitive study of the effects of alternative financing arrangements on
the use of services and on health status. Families participating in the
experiment were assigned to one of 14 different fee-for-service insurance
plans or to a prepaid group practice. As reported by Newhouse (1987),
findings indicated that per capita expenditures by participants who were
subject to a coinsurance rate of 95% were 50 percent lower than those of
families assigned to the plan with no cost sharing. These differences
were attributed to the variations in the quantity of care consumed rather
than to a systematic dependence of the price per visit of the various
options. The participants in the deductible plan used less inpatient care
than their counterparts on the free plan. The costs per hospitalized
participant were found, in general, to be independent of differences in the
various policies. Hence, the results were consistent with expectations
derived from traditional theory of demand. That theory being, individuals
will demand less when the service is provided.
39


Most health experts today agree that diet, exercise, smoking, drug
and alcohol abuse, are behaviors that are major determinants of
morbidity and mortality. Many experts also agree that health education,
promotion, and prevention can result in modification of these behaviors.
Elimination of these risk factors is a key task in the field of public health in
identifying those interventions and behavioral changes that can yield the
greatest payoff. Prevention is cheaper than cure.
For example, for every dollar spent on a MMR shot, there is a
savings of $14.00 on medical and long term care costs. The same dollar
spent on pertussis vaccine saves $11.00 on hospital bills and treatment
(Chapman, 1995). During the first 6 months of 1994, the average
hospital bill for pertussis in the state of Colorado was $14,854 (Colorado
Health Association, (CHA), 1995).
Poliomyelitis is the most remarkable in terms of the benefit of a
successful vaccine strategy. When the analysis is done from the CHA
database, the total hospital charges from January 1, 1988 to the first half
of 1994 indicated 1,561 hospital admissions for a total hospital charge
amount of $15,048,158. When these data are further broken down to
reflect cases admitted with the birth date after 1955 (vaccine
introduction), the total number of hospital admissions decreased to 40
40


compared to 1,561 when birth dates prior to 1955 are not used. The
hospital charges for the latter time period decreased to $445,229. This
analysis is meant to reflect the time period before and after polio vaccine
became available, and the economic impact to the health care system that
an effective vaccine program can accomplish. This is compelling evidence
that preventive care is not only the right" thing to do, but has an impact
on the health economics of medical care.
Caution must be exercised when discussing the CHA database.
These figures can only be used as a benchmark for the effects of vaccine
preventable diseases on the health care system. Because the CHA
database reflects only hospital charges, it does not take into account
other factors such as non-hospital ized cases that probably account for the
majority of these diseases, as well as lost time, and medical costs
associated with treatment of disease that is not severe enough to warrant
hospitalization. These figures are probably a gross under-estimation of
the true impact that vaccine-preventable diseases have on the society as
a whole.
At the more formal level, the problem of setting priorities for health
education and promotion can be viewed as an aspect of the general
problem of allocating scare resources to satisfy human wants. There are
41


always trade-offs. Is a disease-free society worth the increasing costs of
health care? Many families have elected to spend their income on items
other then health. When people become sick, they must decide whether
to get medical care. They consider the severity of the illness, the
perceived value of treatment, and the out-of-pocket and time costs of
going to the doctor. If the costs of treatment seem higher than the
benefits, sick people may decide to give time and home remedies a
chance to work. A survey of episodes of illness among government
workers found that in three-eighths of the episodes with disability days,
the medical care system was not used (Riedel, 1982). Spending on
chronic, dental, or well-care episodes of treatment may not be triggered
by an acute problem but entails a similar decision. In short, it is a matter
of choice. Some choices are made by individuals and families allocating
their income to acquire market-produced goods and services and their
time and energy to produce goods and services outside the market.
Other choices are made by government and are financed by taxes. The
preferred mix between private and public efforts falls within this general
framework. The problem is to choose the preferred sets of interventions.
To do this, identification of the costs and benefits of each intervention is
necessary.
42


Economics does not pretend to think about equity. The United
States is an ethnically pluralistic society with diverse subcultures and a
larger number of underclass." The words inscribed on the Statue of
Liberty Give me your tired, your poor, your huddled masses yearning to
breathe free" connote a preference for opportunity over equity.
Accordingly, the American health care enterprise seems likely to continue
to address the problem of infinite needs and finite resources by pursuing
excellence and efficiency over equity.
Summary of Health Economics
Improvement in health, amelioration of disease, and prevention of
illness remain the fundamental rationale of the health care sector. The
achievement of these objectives with the lowest resources expenditures,
at the best levels of efficiency, remains the major challenge of health
economics, the age-old challenge of efficiency in resource allocation.
Medical care remains a top social priority. Priorities have shifted
from the creation of new programs to the maintenance of existing ones.
Some of the existing programs were actually dismantled. For example,
the Hill-Burton grants were changed to loans in 1970, and the program
terminated entirely in 1974. However, despite these measures, it was still
43


estimated that by 1980 the nation had over 100,00 excess beds in acute
care hospitals(Stoline, 1993). In addition, increases in biomedical
research spending were also tempered in the late 1960s. Until 1968, for
example, Congress had always granted the NIH more dollars than were
requested, but in that year, only the requested amount was granted
(Blainpain, 1985). Today, requested amounts are difficult to obtain,
although they have proven highly resistant to rounds of continuous
budget cuts.
Health economics plays a role in answering the question of, Why
study ways to improve the immunization rate for children under 24 months
of age? Driven by economic necessity, the spiraling costs of health care
sets the stage for health economists for a new movement the era of
cost control and economic efficiency. Health promotion and disease
prevention are areas for anticipatory strategies.
Cost effective health care requires an informed laity and intensive
collaboration of medicine and public management. The following
sections discuss the motivation for the public and private sectors to rein in
expenditures.
44


Public Choice Theory
Economics is basically about scarcity and choice; it is about
opportunity costs; it is about maximizing the benefit to society from the
finite resources available. The stated goal is efficiency, i.e. getting the
most one can from the available labor, land and capital, often with some
pre-defined concern around equity. The viewpoint is that of society at
large and not any single individual or individual group's perspective. The
theory of public choice has been developed by scholars who are mostly
economists and rests on a single decision structure. It involves the
explicit introduction of a democratic model, one in which the rulers are
also the ruled.
Public choice assumes that individuals, although acting in self
interest, are aware that they are making decisions for a public of which
they form a part. Therefore, individuals are cognizant that they are
selecting results that affects others and themselves.
Public choice theory is a direct outgrowth of welfare economics. It
attempts to construct a critically important bridge between the behavior of
individuals acting in the marketplace and the behavior of individuals who
act in the political process. This analysis of behavior in separate
45


capacities recognizes the single decision structure that exists in a
democracy by examining the behavior of individuals as they participate in
formulating public collective choices that must apply to all (Buchanan,
1984).
The origin of public choice can be traced two hundred years to the
voting studies by Bentham (1789). The formal theory of public choice is
fairly recent, initiated by Arrow's Impossibility Theorem (1963). Buchanan
and Tullock (1962) extended Arrows social choice theoretical base into
public choice theory. Both Arrow in 1972 and Buchanan in 1986 were
awarded Nobel prizes in economics for their social and public choice
contributions to welfare economics.
Buchanans (1984) work focused on the distinction between
public and private choice. He assumed that individuals react differently
than might be assumed in a market model when they are making choices
that affect their private and public interest. To test his beliefs, Buchanan
studied the inconsistencies in the British National Health Service (NHS).
He examined the questions: Can services that are privately valued by
individuals be provided free by the government? Are difficulties in
maintaining consistency related to the very structure of the institutions
46


which provide services? Therefore, could those difficulties be removed
only after major reforms in the institutional structure?
In the NHS, Buchanan found that the private individual choice
behavior of users of health services was inconsistent with their public or
collective choice behavior as voters and taxpayer. Thus, the same
person acted in two separate roles either as a demander or supplier. This
behavior is not irrational, but arises exclusively from the institutional
setting for choice in the two sides of the account. Individuals thus make
choices in two separate capacities: as buyers-sellers in ordinary markets
for private goods and services and as buyers-sellers of public goods and
services in the political process (1984, p.34).
When individuals participate in both market decisions and political
decisions, they demand more services privately than they supply publicly.
An individual, in his or her capacity as a participant in collective choice,
must balance costs against benefits. Buchanan concluded that, in order
to have reform, the inconsistency between supply-influenced choice, and
demand-influenced choice must be eliminated and the two types of
decisions must be brought into the same framework.
47


Research in public choice explored self interest and public interest
in public choice. Downs (1957) concluded that the individual acts in his
own best interest when making collective democratic decisions. The
question of whether the motivation behind the self interest or public
interest is based on egotistical private consumption, on altruism, on ideal
principles of ethics, or upon any combination of these factors are studied
by Hylland (1986) and Margolis (1982).
This concept is relevant to immunization practices because parents
make choices that affect a public of which they form a part. Also, many
have the options of receiving public or private vaccinations. This dual
role, as consumer and citizen, fits with Buchanans conclusions that
individuals make choices in two separate capacities as buyers-sellers of
private goods and as buyer-sellers of public goods. As in the NHS, the
private individual choice behavior or negligence to maintain age
appropriate immunizations for their children may be inconsistent with the
public or collective behavior for the demand of a disease free society.
The theory examines the behavior of persons as they participate
in the formation of public or collective choices. In acting or behaving as
a public choice participant, the individual is presumed to be aware that
he/she is, in part, selecting results that affect others than him/herself. A
48


person has no sensation that the behavior modifies the environment of
other persons. This behavior may be more consistent with the Health
Belief Model, which suggests that unless the person believes there is an
actual risk of developing an illness, little importance is placed on the
preventive action. Preventive actions not only protects themselves from
contagious diseases but others in the community.
Lack of immunization not only effects the individual, it also has
economic effects on the society as a whole. Despite the analyst's
recognition that each economic act influences, even if infinitesimally, the
conditions confronted by all participants, the participant may not be
cognizant of this public influence.
Gordon Tullocks theory of public choice might be taken to
reflect economic imperialism, interpreted as efforts by economists to
expand the boundaries of their own discipline so as to make it more
applicable to human behavior (Tullock, 1972). Another approach is
the Kantian principle of enlightened self-interest that is based on a
generalized recognition that there is reciprocity within social
interactions. The interdependence that seems to be inferred from the
results arise because the goods the individual values are more
inclusive for the society.
49


The theory of public choice makes a sharp distinction between the
choice behavior of the individual in the market process and his/her
behavior in the political process. Individuals make decisions in two
separate capacities, as buyers-sellers in ordinary markets for private
goods and services, and as buyers-sellers of public goods and services in
the political process. While this duality has certain utility in health
economics, public choice theory is not the focus of this dissertation.
It has frequently been observed that people acting individually or
collectively are willing to spend much more to treat an illness than to
prevent it. Such behavior is characterized as irrational and inefficient
(Fuchs, 1986). These characterizations would be justified if the goal of
individuals were to maximize life expectancy for any given level of
expenditure. It is more plausible, however, to assume that their goal is to
maximize individual utility, and under that assumption, so-called
irrational behavior could be quite rational, and public choice theories
would be violated. Willingness to pay for a given change in the
probability of survival seemingly varies directly with the individuals
proximity to death. For instance, if an individual were very sick, with a
probability of survival of .05, it may be worth a great deal to raise the
probability to .10 (double) through some form of treatment. On the other
50


hand, if the individuals probability of survival is .90, he may be unlikely
to value as highly a preventive service that will raise it to .95. In short,
when people are healthy, they are not eager to spend money to become
even healthier, but when they are sick, and especially when they are
facing a high rate of morbidity, they may be willing to spend a good deal
for even a small chance of improvement.
Summary of Public Choice Theory
Individuals have to make certain choices related to immunization
practices. Reasons for poor immunization compliance have been
described earlier in this thesis, public choice theory may help discern the
rational related to those reasons. Parents/guardians make choices that
affect a public of which they form a part. This decision process is not the
framework in which parents/guardians operate. Non-immunization is due
to reasons other then collective decisions or even rational self interest.
The rational self interest of parents/guardians should be to protect their
child from preventable diseases. The Health Belief Model suggests other
reasons or non-compliance. Collective public choice decision making
should not be introduced until the explanatory analysis of the choice
behavior of a single, isolated individual can be better understood.
51


Health Belief Model
The Health Belief Model (HBM) relates psychological theories of
decision making to an individuals decision about health behaviors.
Rosenstock (1966) has attributed the origins of behavior motivation
theory underlying the Health Belief Model to Lewinian (1935) theory of
goal setting in the level-of-aspiration situation. Lewin (1944)
hypothesized that behavior depends mainly upon two variables:
1. The value placed by an individual on a particular outcome.
2. The individual's estimate of the likelihood that a given action
will result in that outcome.
The development of the Health Belief Model by Hochbaum (1959),
Leventhal (1966), and Rosenstock (1966) extends the use of socio-
psychological variables to the explanation of preventive health behavior.
The model analyzes an individuals motivation to act as a function of the
expectancy of goal attainment in the area of health behavior. Rosenstock
(1966) states that the HBM is derived from the social-psychological
theory of Lewin (1944), and Becker and others (1974) have suggested
that the Model can be categorized as an "expectancy x value theory,
attempting to describe behavior of decision-making under conditions of
52


uncertainty. The HBM, which is concerned with the subjective world of
acting individuals, proposes the following theoretical conditions and
components:
1. The individuals psychological readiness to take action related
to a particular health condition, determined by both the
persons perceived susceptibility or vulnerability to the particular
condition, and by his perception of the severity of the
consequences of contracting the condition.
2. The individuals evaluation of the advocated health action in
terms of its feasibility and efficacy weighed against his
perception of psychological and other barriers or costs of the
proposed action.
3. A stimulus, either internal (perception of bodily states) or
external (interpersonal interactions, mass media, personal
knowledge of someone affected by the condition) must occur to
trigger the appropriate health behavior.
(Becker, 1974)
The earliest characteristics of the Model were that for an individual
to take action to avoid a disease, he or she would need to believe:
1. That he or she was personally susceptible to the disease.
2. That the occurrence of the disease would have at least
moderate severity on some component of his/her life.
3. That taking a particular action would in fact be beneficial by
reducing his/her susceptibility to the conditions, if the disease
occurred, by reducing its severity, and that it would not entail
53


overcoming important psychological barriers such as cost and
convenience.
(Rosenstock, I960,1966)
Individuals vary widely in their acceptance of personal
susceptibility to a condition. At one extreme might be the individual who
denies any possibility of contracting a given condition. In a more
moderate position is the person who may admit to the statistical
probability of a disease occurrence, but senses a possibility that it is not
likely to happen. Finally, a person may express a feeling that he/she is in
real danger of contracting the condition. In short, personal susceptibility
refers to the subjective rather than objective risks of contracting a
disease.
Convictions concerning the seriousness of a given health problem
vary subjectively from person-to-person. The degree of seriousness may
be judged both by the degree of emotional arousal created by the thought
of a disease as well as by the kinds of difficulties the individual believes a
given health condition will create for him or her (Robbins, 1962).
Persons may see a health problem in terms of its medical or
clinical consequence. They would thus be concerned with such questions
as whether a disease could lead to their death, reduce their physical or
54


mental functioning for long periods of time, or disable them financially.
However, the perceived seriousness of a condition may include such
broader and more complete implications as the effects of the disease on
their job, on their family, and on their social relationships. Thus, a person
may not believe, for example, that pertussis is medically serious, but may
nevertheless believe that its occurrence would be serious if it created
important psychological and economic tensions with in the family.
Perceived susceptibility and severity having a strong cognitive
component are at least partly dependent on knowledge. An individual
may believe that a given action will be effective in reducing the threat of
disease, but at the same time see that action itself as being inconvenient,
expensive, unpleasant, painful or upsetting (e.g., chemotherapy). These
negative aspects of health care action serve as barriers to action and
arouse conflicting motives of avoidance. Several resolutions of the
conflict may be possible.
If the readiness to act is high and the negative aspect is seen as
relatively weak, the action in question was likely to be taken.
The readiness to act is low while the potential negative aspect is
seen as strong, the negative aspects functioned as barriers to
prevent action.
(Becker, 1974)
f
55


When the readiness to act may be great and the barriers to action
are also great, the conflict may be more difficult to resolve. Individuals
will be highly oriented toward acting to reduce the likelihood or impact of
the perceived health danger, but the same individuals will also be
motivated to avoid action since they see it as highly unpleasant or even
painful (Becker, 1974). Figure 3.1 (Becker, 1974) depicts the HBM
schematically.
The expectancy theory (Rosenstock, 1966) approach to health
behavior views the action that an individual takes as related to the
subjective desire to lower susceptibility and severity, and to a subjective
estimation of benefits minus costs, in which medical advice is only one
component of the mental calculus. The attractiveness of incentives value
to the health action or goal of the individual is therefore, its ability to
lower the individuals perceived susceptibility to the particular illness
and/or to reduce consequential severity of the illness. The expectancy
variable may be conceived of as a quantitatively varying belief that some
particular action in a particular situation will lead to a goal. The
perceived likelihood of successfully attaining the goal or the expectancy
of success of the health action is a function of the perceived benefits of
56


taking the health action minus the perceived barriers or costs of that
action.
The Health Belief Model assumes that motivation is a necessary
condition for action and that motives selectively determine an individuals
perceptions of the environment. The concept of motivation is
operationalized in the models dual dimensions, i.e., psychological state
of readiness to take specific action and extent to which a particular course
of action is believed to be beneficial in reducing the threat
(Rosenstock.1966).
57


FIGURE 3.1
THE HEALTH BELIEF MODEL AS PREDICTOR OF PREVENTIVE HEALTH
CARE (Becker, 1974)
Mfvidial Perception* Modifying Factors Likelihood of Action
It has been suggested by Becker and associates (1974) that the
HBM should be expanded to include a separate motivational variable
representing the need or desire for achieving health related goals by
employing the concept of general health motivation (different degrees of
58


readiness to undertake health actions). According to this modification,
motives are viewed as dispositions within the individual to approach
certain classes of positive incentives, and it is postulated that the desire
to attain or maintain a positive state of health and to avoid a state of
illness is (or should be) a dimension of health motivation (Becker, 1974).
Consideration of several studies of how preventive health services
are used and an understanding of why they are used is in order.
Stoeckle, Zola, and Davidson (1963) studied the effects of ethnic values
upon the specific decision to seek medical attention and on the
differential interpretation of objective symptoms. Friedson (1961)
illustrated the different processes through which members of different
social groups move in obtaining diagnosis (lay and professional) and in
seeking care. These are studies of utilization and limit their focus to
illness behavior, that is, behavior in response to symptoms. These
studies are of unknown relevance to the situation confronting the person
who must decide whether to seek preventive or detection services before
the appearance of events that he/she interprets as symptoms.
In research conducted on behavioral beliefs, several studies
conducted offered the study cohorts the opportunity to take preventive
action through directed messages and circumstances that could serve as
59


cues to stimulate action. Hochbaum (1952) studied more than 1,200
adults in three cities in an attempt to identify factors underlying the
decision to obtain a chest X-ray for the detection of tuberculosis (TB). He
tapped beliefs in susceptibility to TB and beliefs in the benefits of early
detection. The findings for the group of persons that exhibited both
beliefs found that 82% of those individuals had at least one voluntary
chest X-ray during a specified period preceding the interview. The group
exhibiting neither of these beliefs, only 21 % had obtained a voluntary X-
ray during the criterion period. Thus, four out of five people who exhibited
both beliefs took the predicted action, while four of five people who
accepted neither of the beliefs had not taken the action. Therefore,
Hochbaum appears to have demonstrated with considerable precision
that a particular action is a function of the two interacting variables-
perceived susceptibility and perceived benefits. Hochbaums study,
used mass media in the three cities to urge the population to obtain chest
X-rays. The limitation of the Hochbaum study, as it relates to
immunization practices, automatically assumes the patient has pre-
existing knowledge about the dangers of TB. The current study is making
an assumption that there may be no pre-existing knowledge of the
vaccine preventable diseases.
60


Kegeles (1963) dealt with the conditions under which members of a
prepaid dental care plan will come in for preventive dental check-ups or
for Prophylaxis in the absence of symptoms. He attempted to measure
the respondents perceived susceptibility to a variety of dental diseases,
the perceived severity of those conditions, his/her beliefs about the
benefits of preventive action and his/her perceptions of barriers to those
actions. Kegeles showed that with successive increases in the number of
beliefs exhibited by respondents from zero to three, their frequency of
making preventive dental visits also increased. In the Kegeles study
(1963), every member of the population was offered free or inexpensive
dental treatment and was urged to use it. This was done through
education in the newspapers, television, and at shopping centers. Again,
pre-existing knowledge of the disease along with the small sample size
are major limitations of these findings as it relates to the immunization
question of lack-of-knowledge."
In the Leventhal, Hochbaum, and Rosenstock study (1960), a
prospective model was designed to test the impact of the threat of
influenza on families. Two hundred randomly selected respondents in
each of two cities in the US were interviewed twice. The first interview
was intended to be made before most people had the opportunity to seek
61


vaccination or to take an/ other preventive action and before much
influenza-like illness had occurred in the communities. The second
interview was to be made after all available evidence indicated that the
epidemic had subsided. Although the samples were very small and
possibly not representative, the differences were statistically significant.
Analyses of the available data suggested that prior beliefs in susceptibility
and severity were instrumental in determining subsequent action. Again,
the population had been alerted by newspapers and by public health
officials to the desirability of obtaining influenza immunizations.
In short, the populations in these studies had been exposed to
information that both indicated the availability of a health procedure and,
in most cases, urged them to avail themselves at relatively low cost to
that procedure.
Other Behavioral Models
Since the seminal work of Lewin, several other predictive and
parallel models have been advanced to describe action in situations
involving risk taking or decision making under uncertainty: 1) Tolmans
(1955) analysis of performance behavior; 2) Atkinsons (1957) view of
62


risk-taking behavior as a theory of achievement motivation; and 3)
Feathers (1959) analysis of decision making under uncertainty.
These models of motivation focus on general attainment and
changes of the whole organism in relation to its environment rather than
on an isolated reaction (such as Pavlovs interest in salivary response).
As in the Health Belief Model, they have two similarities. First, there is a
value placed by the individual on a particular outcome or action. Second,
the subjective probability the individuals expectation or estimate of the
likelihood that a particular action will produce the desired outcome.
Summary for The Health Belief Model
The Health Belief Model suggests that for individuals to act upon a
health action, there must be a perception that the individual is at risk for the
development of that particular disease. The HBM inquiry that this theses
addresses is viewed as a precursor to the perceived susceptibility of an
illness, that is, Does the study population even know of the disease?"
Therefore, individuals lacking knowledge of a specific disease would indicate
that there can be no perceived risk, and hence, that the disease would not
be viewed by the individual as a threat to their well being. The HBM
63


triangulates with health economic theory and public choice theory to better
define immunization strategies.
General Summary
Health economics sets the "rational actor theory reflected by a series
of cost/benefit analysis, even if the underlying costs and benefits are of
questionable accuracy, i.e., they are more subjective then the reportedly
objective calculation would have one believe. Health economics does
provide, however, a series of necessary independent variables, i.e., socio-
economic status, educational level, and parental knowledge.
Public choice exhibits its apparent relevancy to many, nevertheless,
public choice is not a principle contributor to this study because rational self
interest or collective decisions are not the driving force that motivates
parents/guardians to comply with recommendation to immunize their
children.
The Health Belief Model, although a very persuasive component of
this framework, cannot be accepted exclusively to explain the deficiencies
in immunization practices. It is offered only as an explanation to a barrier
within parental knowledge. The thesis is constructed on the framework
that the three theories are interdependent.
64


CHAPTER 3
STUDY DESIGN
There have been many efforts on the part of the Colorado
Department of Health and Community Groups that have focused on reaching
children in this susceptible age group less than 24 months. Hospitalization
offers a potential opportunity for the health care delivery system to influence
immunization status and provides parental edification with respect to
immunizations. The problems addressed in this study are the unanticipated
opportunities that hospitalization presents to health care providers
concerning immunization practice. The principal research questions are:
1. When a child is hospitalized for a non-vaccine related illness,
does the health care system provide sufficient information to the
parent so they can respond accurately to the question, Is your
child up-to-date on their immunizations?
2. When a child is hospitalized for a non-vaccine related illness
and the parent is given correct information regarding the
definition of up-to-date," are they then able to respond accurately?
3. Does parental/guardian knowledge concerning immunizations
have an effect on vaccine administration during hospitalization?
4. Do parents/guardians have knowledge of what the nine vaccine
preventable diseases are?
65


The primary goal of this study is to distinguish the opportunities
that hospitalization presents for childhood immunization of children under
the age of 24 months in the State of Colorado. This dissertation uses a
descriptive study design to explore the missed opportunities of
vaccination in a pediatric tertiary care hospital, in this case, Denvers
Childrens Hospital. An understanding of hospitalization immunization
opportunities may provide important lessons that may be used by the
politician, health care professional, and the public as they continue to
search for ways of strengthening the delivery of health care services to
American children, under 2 years.
As was deduced from the earlier review of the literature, the
research hypothesis is:
During hospitalization, evaluation of immunization status
is a function of (a) parental knowledge and (b) accurate
assessment by the medical care provider.
Objectives
There are four main phases of this thesis:
1. To obtain a representative study population of children
admitted to a tertiary care hospital for non-vaccine related
illness between the ages of two and 24 months of age to
66


determine if hospitalization would offer an opportunity to
improve immunization rates.
2. To have the parent or guardian respond to a questionnaire
to determine the parent/guardian knowledge of immunization
schedules and understanding of inquiries about immunization
schedules at the time of hospitalization.
3. To determine health care provider's utilization of
hospitalization as an opportunity to educate parents/guardians,
as well as vaccinate children.
4. To assess parent/guardian knowledge of the nine vaccine-
preventable diseases.
The independent variables are: parental knowledge of
immunizations, socio-economic status, educational level, and whether the
health care provider used an open ended question, i.e., Tell me the
vaccinations you child has received" ora leading question, i.e., Is your
child up-to-date (UTD) on his vaccinations to assess the childs
immunization status. The dependent variable is binary and is the verified
rate of immunizations (the child is or is not appropriately immunized).
67


Materials and Methods
Study Population
The research study takes place at The Childrens Hospital (TCH)
located in Denver, Colorado. The Childrens Hospital serves as the
principal tertiary pediatric health care center for the entire State of
Colorado and parts of the surrounding fifteen state region, with
approximately 9,000 in-patient hospitalizations per year. TCH is also a
teaching institution with a large pediatric residency training program.
Resident physicians work in the hospital under the direct
supervision of the faculty physicians. An important function of the
resident physician is to check the immunization histories of all children
admitted to the hospital. If found to be incomplete, upon discharge, there
is a standard pre-printed discharge order form that provides the
opportunity for the physician to update needed immunizations.
The first task in designing this research to determine the accuracy
of parental knowledge and immunization practice is to determine which
parents to study. Previous research has concentrated on groups of
children from poor, inner city households (Bates, 1994; Wood, 1994;
68


Guyer, 1994; Atkinson, 1994), ethnic minorities (CDC, 1994), military
clinics (Jones, 1994), middle-upper and lower income populations
(Salsberry, 1994), and differences in public and private settings (Hueston,
1994).
Researchers (Guyer, 1994; Orenstein, 1990; McConnochie, 1992)
who studied dichotomous groups compared immunization rates and
missed opportunities between groups. Using this method, they explored
the risks associated with missed opportunities of immunization. As such,
they overlooked key dimensions addressed by the health belief model
with conceptually distinct dimensions of decision making about health
behaviors. These dimensions include values placed by an individual on a
particular outcome and the individuals estimate of the likelihood that a
given action will result in a projected outcome.
As the research literature suggests, hospitalization has been
identified as a missed opportunity for updating immunization status. What
is lacking in the research literature is the use of hospitalization, not only
as a vehicle for providing immunizations, but also as an opportunity to
educate parents adequately to maintain regular immunization schedules
for their children, with the assumption that a better educated
69


parent/guardian leads to increased compliance with immunization
schedules.
Patients eligible for participation are all in-patient and same-day
surgery admissions, with the birth date falling between August 1,1993
and June 1,1995. Exclusion criteria are defined as:
infants younger than 2 months of age
children admitted to the Pediatric Intensive Care Unit (PICU)
(these families were surveyed after transfer from the PICU)
families that did not understand nor speak English
families refusing to sign the informed consent form
discharged from the hospital prior to enrollment (e.g., short stay
admission)
Questionnaire Design
The study questionnaires were developed specifically for this
research. They were pilot tested at The Childrens Hospital prior to the
study period after peer review. The questionnaire is designed to take 15
minutes to administer, and written at a fifth to sixth grade level to minimize
literacy barriers. Answers are coded numerically and consisted of
70


primarily yes-no options with minimal interpretation by the researcher
necessary. The questionnaire contained:
1. A consent form pursuant to the Colorado Multiple Institutional Review
board (COMIRB) approved protocol, and from the University of
Colorado (Denver) Human Research Committee. The consent form is
read to the parent/guardian with ample opportunity to ask questions.
Both the researcher and the parent are required to sign the consent
form. The parent/guardian is offered a copy of the signed consent.
2. Section 3.1 Questionnaire obtaining parent responses prior to
education
Section 3.2 Questionnaire to obtain information from the medical
record
Section 3.3. Questionnaire obtaining parent information after
appropriate immunization information was provided
Section 3.4 Verification questionnaire obtained from the childs
primary care physician or clinic
Section 3.5 Questionnaire to obtain information from post discharge
medical record review
The content and structure of each section of the questionnaire is
described in detail in the Appendix: Questionnaires 3.1 through 3.5
Section 3.1 Parent Information
The parent questionnaire requested information about the place of
primary care, past immunization experiences, and information requested
71


by the admitting health care provider during the current hospitalization. It
also asks the number of times the child has been seen in the previous
twelve months by a health care provider. This question addresses the
potential bias of seasonality and predispositions of a cohort of disease-
specific children in the analysis.
It also asks questions about family income and educational level.
Because of the sensitive nature of these questions, parents/guardians are
given the option of refusal to answer these questions. The number of
other children in the household is asked for comparisons among the
families previous experiences with children's immunization schedules.
Finally, the years lived in Colorado may be an important variable when
making conclusions about the public health care system in the state.
Section 3.2 Medical Record
The childrens current medical records are reviewed for demographic
information and immunization status as determined by documentation in
the required history and physician (H&P) documentation. The H&P is
completed by the resident physician on duty within four hours of
admission. This immunization history is a standard inquiry that is
recorded. Information is recorded by the researcher on the questionnaire
72


3.2. This information is used to search for similarities in demographic
location, gender, and payor information. It also provides information to
the researcher on the primary care provider for further verification of
immunizations actually given. Parents/guardian name, address, and
telephone numbers are also entered into the database to be used not only
for demographic comparisons, but also to send reminder cards of needed
immunizations and, if necessary, the location that would be free of cost
for parent/guardian to obtain needed immunizations.
Section 3.3 Parent Knowledge After Education
After questionnaire 3.2 is administered to the parent/guardian, the
researcher follows with an interview session that gives the
parent/guardian an interpretation of what immunizations would constitute
the meaning of up-to-date" using the American Academy of Pediatrics
(AAP) recommendations for immunization. A reference card is given to
the parent/guardian as a source of information that can be used for future
immunization schedules. Questionnaire 3.3 is then completed by the
researcher (Appendix 3.3). The information card (Appendix A) is
reviewed with the parent to give them the opportunity to determine
visually their childs current immunization status and then they are asked
73


to respond to the question, In reviewing the current recommendations, is
your child up-to-date with the recommended schedule?
There is also a series of nine questions that are asked of the
parent/guardian related to the knowledge about the nine diseases for
which vaccines are currently recommended. These questions are coded
as a yes or no response. This information leads to potential inferences
associated with the Health Belief Model.
Section 3.4 Verification From Primary Care Phvsician/Clinic
Verification of the child's immunization status is obtained by the
researcher calling or visiting the primary care physician or clinic that is
recorded in the childs hospital medical record. Information related to the
immunizations recorded in the childs medical health record (from the
office or clinic) is documented using questionnaire 3.4. The medical
health record is comprehensive in nature, and provides continuity of
health care, therefore, is considered a vital component in the delivery of
services. The medical health record is a source of documentation that is
considered legally and medically accurate and reliable (vonKoss
Krowchuk, 1995). Information about the mechanism the office or clinic
uses to notify parents of immunization schedules are also obtained.
74


Section 3.5 Post Discharge Review
The medical record is reviewed after discharge and questionnaire 3.5
completed by the researcher that evaluates if needed immunizations are
given prior to discharge. Discharge immunization information is part of
the pre-printed discharge orders.
Study Period and Sample Size
The study period was from October 13,1995 to December 20,
1995 and represented a population size of 197 subjects. Enrollment was
continuous, seven days a week, for the in-patient population and Monday
through Friday (days of operation) for the same-day surgery cohort. (The
exception for days of enrollment were November 4,1995, through
November 15,1995, due to an unexpected family death.) Patients were
typically enrolled between the hours of 7:00 a.m. and 12:00 p.m.
However, provisions were made for investigator availability at other hours
if necessary.
The power of a study is defined as the probability that it will yield a
significant result when the true size of effect is as specified (Clayton,
1994). The power is different for each size of effect considered, being
75


greater for larger effects. It is reasonable to select the number of two
hundred patients from the power analysis, which suggests the following:
With the confidence interval set at 95% and the power set at 80%
(probability of being correct), with a 1 to 1 ratio of expected to
unexpected, and a 50% expected frequency, the total sample size
should be 130.
With the confidence interval set at 95% and the power set at 90%, with
a 1 to 1 ratio of expected to unexpected, and a 50% expected
frequency, the total sample size should be 170.
With the confidence interval set at 95% and the power set at 95%, with
a 1 to 1 ratio of expected to unexpected, and a 50% expected
frequency, total sample size should be 204.
Once a patient is enrolled into the study, subsequent admissions
for that patient are excluded for potential enlistment.
The TCH daily census sheet and the same-day surgery sheets
served as the source of admission information. These sheets were
computer generated from the TCH Information Service Department at
midnight each evening. A daily worksheet was prepared with the eligible
patients categorized by unit, admission date, and enrollment status. If a
76


patient was not enrolled in the study on the admission day, i.e., parent
was not available or the patient was admitted to the Pediatric Intensive
Care Unit (PICU), he or she was listed as incomplete for that day and
carried over to the following days worksheet.
The first 70 patients were enrolled exclusively by the author of this
thesis. Seven interviews were monitored by the Director of Nursing
Research at The Childrens Hospital to validate consistency and
appropriateness of interview techniques. The next ten were conducted by
the author of this thesis aided by a research assistant (RA), a medical
doctor awaiting her license. The next ten were done by the RA being
observed by the author of this thesis. The remaining 110 patients were
enrolled by the RA.
The parent questionnaire consists of primarily yes/no answers with
no interpretation of answers necessary from the RA. Questions that
required answers other than yes/no are written down as articulated by the
parent/guardian and are then coded by the author of this thesis within 24
hours.
77


The interpretative questions are:
1. What was the exact wording the admitting physician used when asking
about immunizations?
Answers to question 1 are categorized as:
0=notasked
1 =leading question i.e. Is your child up-to-date? or Is
your child current?"
2=open ended question i.e. What immunizations has your
child had? or Where is your child with his/her
immunization? (requiring the parent/guardian to
articulate what shots had been received or needed).
2. What was your exact response to the immunization question on
admission?
Answers to question 2 are coded as:
1=yes
2=no
3=...let me tell you...
4=not applicable
3. How does your doctor who usually provides your childs care inform
you about immunizations?
Answers to question 3 are categorized as:
0=not applicable
1 =nothing done
2=scheduled next appointment at current time
3=reminder card mailed
4=phone reminder
78


Definitions
For the purpose of this study, children were considered immunized
if specific immunizations to date were documented on the childs
immunization record, or were verified by the primary care physician or
clinic record. Deficiencies were defined as specific immunizations that
were missing more than two months past the due date for the regular
scheduled immunization according to the American Academy of Pediatrics
criteria (Committee on Infectious Disease, 1991). Children that were
more than two months late receiving immunization, but were caught up by
the time of the survey, were considered immunized.
Research Methodology
The immunization status of 197 pre-school children admitted to a
pediatric hospital were determined by interviewing parents/guardians
regarding their childrens immunization histories. The patients
immunization records were subsequently reviewed for confirmation with
the childs primary care provider. The admitting physicians history was
79


also reviewed to determine whether the patient's immunization status had
been noted. The parents were additionally asked if the physician asked
the immunization question in a manner that allowed the parent/guardian
the opportunity to articulate the immunization status, as opposed to a
"yes or no answer. This was an important aspect of this thesis because
parents/guardians were likely to respond in the affirmative. This line of
questioning was also more convenient for the person asking the question.
The answers were coded as open (....tell me about your vaccinations)
or leading" (...are you up-to-date or current with immunizations).
To become a participant, parents/guardians were requested to sign
consent forms so medical records could be reviewed by the researcher.
(TCH unit staff were informed of the study, but not informed of the
purpose of the study.) Patients meeting the study criteria were identified
by the daily census and day surgery sheets. In day-surgery, the
parents/guardians were not approached for the study until after the child
had returned from surgery and was stable. On the patient care units,
parents/guardians were not approached unless it is obvious to the
researcher that it was an appropriate time to conduct the interview, i.e.,
the child was not inconsolably crying or feeding, or the parent/guardian
was resting.
80


The questionnaire responses were then entered into Microsoft
Excel spreadsheet. When the study period was over, and verification
from both the in-patient and primary care physician or clinic medical
record obtained, a reminder card was mailed to the families that were in
need of immunizations and the location where they could receive them at
no cost (e.g., immunizations are provided free of charge at the local
health departments).
The research methodology was a correlation design utilizing a
survey methodology and included the five survey instruments. The
purpose of the design was to correlate immunization rates that have been
verified with responses from parents concerning the questions of the
meaning of up-to-date" or current" with recommended immunization
schedules. The survey method was selected because it offers an
effective and efficient way to gather information on existing parental
knowledge in regard to their child's immunization status, as well as
measure the effectiveness of the health care provider when assessing
this information. Additional information was determined from the medical
record and primary care providers.
The database consisted of 54 data points derived from the
questionnaires. Each participant enrolled in the study is recorded in
81


Microsoft Excel as a separate record, with each record being given an
identification number corresponding to the participants name. Missing
data for any given variable was left as blank in the corresponding cell in
the database. Statistical analysis was performed using SAS. Frequency
calculations, and chi square analysis was performed first by age
distribution (according to AAP's recommended schedules), then as a
whole population. Demographic distribution is compared with total
populations to verify the study cohort was a good representative sample
of true population.
Statistical Methods
Univariate analysis was used to determine factors associated with
delayed immunization; p<.05 (using the chi-square with correction
whenever appropriate) was considered significant. A multivariate logistic
regression model evaluating the significant variables (p<05 chi-square
statistic) was then constructed to determine the independent effects
among several risk factors while controlling for the other variables in the
models.
Trends were analyzed using the chi-square test. Multivariate
analysis was used to assess the effect of selected variables on the risk of
82


not being age-appropriately vaccinated. Logistic regression modeling was
performed by examining the effects of the independent variables singly
and then in differing combinations to determine the independent effects
among several risk factors while controlling for the other variables in the
model of children not being age-appropriately immunized on or before
their second birthday.
For multiple and logistic regression, the analysis process included
those variables hypothesized to act as confounders or interacted for
appropriate immunization practices. The format of the regression
equation was as follows:
LOG (p/1-p) = f -4.78+.52P1 + 1,30p2 + 1,32pa + ,02p4 + .54ps + ,83ps +
1.32p7 + .0ips
Threats to internal validity considered were instrumentation and
selection bias. Both were addressed by using only two researchers
applying the same procedures for each data collection session and
enrolling 100% of eligible in-patients during the designated study period.
Generalizability was addressed as a potential threat to the external
validity of the study. The effects of selection were analyzed by, socio-
economic status, payor, race, family size, and parental educational levels
that have been shown in prior research to be contributing factors
83


(Marland, 1976; Miller, 1994). The Childrens Hospital, because it is a
tertiary care referral center servicing a 15 state region, represents a
sample population that includes children in the Denver, metropolitan area,
as well as the referring areas. In addition, external validity was examined
by comparing these results to other similar studies.
84


CHAPTER 4
FINDINGS
Introduction
This chapter reports the findings of the research thesis. The
analysis was accomplished by separating the sample population into five
sub-divisions (or age groups). The results of the research are presented
in six parts. The first section describes the characteristics of the sample
population and contains the definition for each age group. The second
section provides an analysis of children appropriately immunized for each
of the age specific groups. The third section details the documentation of
immunizations in the childs medical record, first, as evidenced in the
charting from the history taken on admission, and then by the written
orders when the patient was discharged. The next segment addresses
parentai/guardian knowledge regarding immunization practices, as well as
their perception of the nine vaccine-preventable diseases. The fifth
section is dedicated to the research results that identify factors associated
with delayed immunization practice. The final section summarizes the
results of the research.
85


Statistical Analysis
Data from the questionnaires were entered using a Microsoft Excel
spreadsheet. The data sets were checked for missing data or illogical
data using SAS, and the master files were corrected. Univariate analysis
was used to determine factors associated with delayed immunization.
Chi-square was used for the analysis of categorical data. P values <0.05
were considered statistically significant. A multivariate logistic regression
model evaluating the significant variables was constructed. Other
variables, although not statistically significant, were also included in the
multivariate model because of their perceived importance and from
previous research done on this subject. The multivariate logistic
regression model was created to determine the independent effects
among several risk factors while controlling for the other variables in the
model. The SAS Program was used in this analysis.
The research design consisted of several questionnaires: 1) parent
questionnaires including, questions asked by the health care provider
about his/her child when the admission history was taken, the parents
knowledge of immunizations before and after an educational session, and
their knowledge of the nine-vaccine preventable diseases (see Appendix
86


A for educational card); 2) medical record review of the documented
admission data; 3) medical record review after the patient was
discharged; and 4) PCP/clinic verification of documented immunizations.
Description of the Sample
Approximately 336 children under the age of 24 months were
admitted to The Childrens Hospital in-patient or same-day surgery units
between October 13 and December 20,1995. The study population
consisted of 197 participants, or, 58.6% of the potentially eligible patients
were included in the study. Clarification of children that were not
successfully enrolled into the study included: 1) patients that lacked
parent/guardian availability; 2) patients who were admitted to the critical
care unit; 3) newborns; 4) parent/guardian refuse to participate;
5)patients discharged prior to enrollment; 6) patients admitted during a
necessary 13-day leave of absence by the investigator, during which time
the study was suspended.
Tables 4.1 through 4.5 illustrates the characteristics of the
study population. The majority of the parent questionnaires were
answered by the mother, that is, 180 (91.4%) times. The patient sample
consisted of a larger proportion of Caucasian males with mothers having
87


less than a high school education. Males outnumbered females by a 2:1
ratio. Seventy-three percent had some type of health care benefits, i.e.,
private insurance, Medicaid, or HMO. Most children were from families
with less than three children and with a family income of less than
$30,000.
The children in the study population were sub-divided into five
groups: birth to 2 months; 2 to 4 months; 4 to 6 months; 6 to 12 months;
and 12 to 24 months. Definitions of each age group used in the analysis
were as followed: Birth to 2 months were defined by the childs birth
month, day, and year~to the exact birth day two months later (i.e., 1/10/94
to 3/10/94). Two to 4 months were defined by two months and one day
from the childs birth date-to the exact birth day four months later (i.e.,
3/11/94 to 5/10/94). Four to 6 months were defined by four months and
one day from the childs birth date-to the exact birth day six months later
(i.e., 5/11/94 to 7/10/94). Six to 12 months were defined by six months
and one day from the childs birth date-to the exact birth day one year
later (i.e., 7/11/94 to 1/10/95). Twelve to 24 months were defined by one
year and one day from the childs birth date-to the exact birth day 24
months later (i.e., 1/11/95 to 5/10/96).
88


The five age groups were considered independently of each other
and were used to render precise immunization percentages for that
particular age group. It should be noted that a child in one age group was
not exclusive to that cohort. Infants that were 6 months old, for example,
would also be in the analysis of the birth, two and four month age groups.
When comparing the distribution of the birth to 2 month age group
across each patient characteristic, several significant differences were
found. Seven records were excluded from the analysis in this group with
one (0.5%) family with a non-vaccine" belief, and six (3%) records as a
non-response from the PCP/clinic after three verbal requests by the
investigator. Caucasians were more likely to be deficient in the hepatitis
B vaccine, the single vaccine that is given at birth (p<0.05).
Parents/guardians with insurance coverage were also more likely to have
incomplete immunization records when verification was done with the
PCP/clinic (<0.01). These numbers may merely reflect the incomplete
medical chart at the PCP/clinic. Most hospitals in Colorado administer the
hepatitis B vaccine prior to the newborn being discharged from the
hospital. However, since the PCP/clinic had a missing notation of this
vaccine occurrence in the childs medical chart, it was considered a
missing immunization. This may reflect a lack of continuity-of-care from
89


the birth hospital to the PCP/clinic, rather than a missed immunization.
Table 4.1 describes the characteristics of the birth to 2 month age group.
90


TABLE 4.1
Birth to 2 Month
Immunization Status of Children Birth to 2 Months of Age as Documented
Bv the Primary Care Provider
* j t % w <. t f * + Characteristic %*> As ? V ssv t-* f ** , ' s s s ^ Immunksd M Vs-'-* V ** Appropriately x Immunized? s ni(%> , cm- square*?
Gender male female 119 (62.9) 71 (37.1) 98 (64.4) 54 (35.5) 21 (55.2) 17 (44.7) .29
Ethnicity Caucasian non- Caucasian 148 (77.8) 42 (22.1) 114 (75.0) 38 (25.0) 34 (89.4) 4 (10.5) .05
Health care visits <10 10+ 97 (51.0) 93 (48.9) 77 (50.6) 75 (49.3) 20 (52.6) 18 (47.3) .82
Insurance yes no 176 (92.6) 14 (7.3) 144 (94.7) 8 (5.2) 32 (84.2) 6 (15.7) .02
Maternal education <12 12+ 109 (58.2) 78 (41.7) 90 (59.6) 61 (40.4) 19 (52.7) 17 (47.2) .45
Family income <30 30+ 101 (53.4) 79 (43.8) 78 (54.5) 65 (45.5) 23 (62.1) 14 (37.8) .40
Other siblings <3 3+ 172 (90.5) 18 (9.4) 135 (88.8) 17 (11.8) 37 (97.3) 1 (2.6) .10
Medical care provider private HMO/clinic 106 (55.7) 84 (44.2) 84 (55.2) 68 (44.7) 22 (57.2) 16 (42.1) .77
* Total number of immunized and non-immunized in age group. Maternal education
and Income were optional responses, therefore, reflecting different denominators
ac (UTD) up-to-date as verified by the PCP/clinic
q> Missing, incomplete, or not documented in PCP/clinic medical chart
Â¥ Values test the probability of no difference in immunization status between strata of
characteristics
91


In the 2 to 4 month age group, there were no significant differences
when using the chi-square analysis. The number of children used for the
analysis in this age group were 172/197. Twenty-five children were
excluded from the analysis in this group: 18 (9.1%) of the PCPs/clinics
had no medical chart on the child when the investigator called for
verification of the immunization status (this information is obtained from
the parent/guardian at the time of hospital admission and is part of the
medical record); one (0.5%) family with a non-vaccine belief; and six
(3%) records as a non-response from the PCP/clinic after three verbal
requests by the investigator. Table 4.2 displays the distribution of the 2 to
4 month age group.
Among the children in the 4 to 6 month age group (Table 4.3),
there were no significant differences among the variable and the rate of
appropriate immunizations. The number of children used for the analysis
in this age group were 164/189. Twenty-five children were excluded from
the analysis in this group: 19 (9.6%) of the PCPs/clinics had no medical
chart on the patient; one (0.5%) family with a non-vaccine" belief; and
five (2.5%) records as a non-response from the PCP/clinic after three
verbal requests by the investigator. Eight children were too young to be
considered in this age group.
92