UTILIZATION OF HEALTH CARE:
THE INTERPLAY BETWEEN HEALTH INSURANCE,
PATIENT KNOWLEDGE and PATIENT SATISFACTION
Stephanie L. Ayers
B.A., University of Florida, 1998
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
2004 by Stephanie L. Ayers
All rights reserved.
This thesis for Master of Arts
Stephanie L. Ayers
has been approved
Ayers, Stephanie L. (M.A, Sociology)
Utilization of Health Care: The Interplay between Health Insurance, Patient
Knowledge, and Patient Satisfaction.
Thesis directed by Associate Professor Candan Duran- Aydintug
In this study, the main goals are to examine the determinants of formal health care
system utilization. Three questions guide this thesis. First, does having health
insurance impact utilization of health services? Second, how does satisfaction
affect utilization, and third, how does gaining knowledge about health affect the
likelihood of using health services? Results suggest that the main determinant
affecting utilization of sick care is type of gaining health knowledge. For
utilization of preventive health care services satisfaction with doctor is the main
determinant. These determinants help to explain entry into the formal health care
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Candan Duran- Aydintug
I dedicate this thesis to Steve for his support, love, and encouragement during this
journey, and to my parents who instilled in me the values of dedication, tenacity, and
I would like to thank my advisor, Candan Duran- Aydintug, who gave me the
direction, support, opportunity, excitement, and laughter I often needed during these
last two years. I would also like to thank Yili Xu and Virginia Fink for giving me
the opportunity to grow. Finally, I would like to thank the entire UCD Sociology
Department for pushing me further than I ever thought I could go.
2. REVIEW OF LITERATURE.......................................4
Type of Health Insurance................................4
3. RESEARCH METHODS..........................................31
2.1 Health Belief Model....................................................28
3.1 Conceptual Model of Analysis...........................................39
2.1 U.S. Health Insurance Coverage Rates by Demographic Characteristics.......5
3.1 Demographic Characteristics of Community Tracking Survey Sample
4.1 Descriptive Statistics of Community Tracking Survey Sample
4.2 Linear Regression for Sick Care...........................................42
4.3 Linear Regression for Preventive Care.....................................43
Examining the determinants of formal health care system utilization is the
main focus of this thesis. More specifically, what makes individuals choose to enter
the formal health care system? Do these factors lie within the health care system
such as having health insurance or being satisfied with the doctor or do they lie
outside the health care system such as gaining health knowledge from family and the
Internet or do demographic factors such as race, income, or gender play a significant
Health, illness, and health care are important issues in America today.
According to the US Census Bureau, in a report issued September 2003, the number
of individuals and families without health insurance rose to 43.6 million people, an
increase of 2.4 million in one year, while the number of individuals having the
opportunity to receive health insurance through an employer fell from 2001 to 2002
(Mills and Bhandari 2003). Meanwhile, the number of individuals receiving
government health insurance, such as Medicaid and Medicare, rose slightly (Mills
and Bhandari 2003).
Health and illness are socially constructed, but disease refers to biological
problems. While disease reflects objective meaning, illness reflects subjective
meaning. Because illness is socially constructed, the same disease will have different
meanings and implications across cultures or groups. Individuals create the meanings
and definitions that each illness has. Illness is a social condition that indicates
goodness or badness, worthiness or unworthiness, of a person (Weitz 2004: 119).
Because illness is socially constructed, individuals who suffer from the same disease
might not utilize the formal health care system the same way.
The utilization of health care services has increased dramatically over the past
70 years. During early 1900s less than 50 percent of Americans utilized health
services on a regular basis. The early use of health services was not solely based on
individuals ability to pay since charities were often the institutions giving these
services. Doctors and nurses viewed giving health services to the poor via charities
as part of their duty (Cockerham 2004). By the 1950s, the number of individuals
visiting their physician on a regular basis had increased to approximately 66 percent
(Andersen 1976). By the early 1980s, approximately 86 percent of individuals saw a
physician (Steiber and Feber 1981). In 1972, 67.8 percent of all Americans used
ambulatory care, outpatient care, compared to 72.9 percent in 1996 (Weinick,
Zuvekas, and Cohen 2000; Weitz 2004).
Why individuals utilize health care services has been widely studied, however
there exists limited research that fully empirically and systematically encompasses all
the variables associated with health care utilization. The major empirical research in
this area is over 20 years old and in an era, of rapid change in policy, budgetary
constraints, or private market developments, it is... important to collect and analyze
more recent data to reflect current conditions and examine the health system features
(Schoen, Davis, DesRoches, Donelan, and Blendon 2000: 68).
In order to evaluate the utilization of health care, I aim to answer three
research questions. First, does having a critical resource such as health insurance
impact utilization of health services? Second, how is being satisfied or unsatisfied
with the health care system affect an individuals utilization of the formal health care
system, and lastly, does gaining outside health knowledge increase or decrease the
likelihood of using health services?
REVIEW OF LITERATURE
Need for health services has been the primary variable to examine health care
utilization and has been widely found to correlate with utilization. Research indicates
that individuals decide to enter the health care system when they perceive their health
problem as severe (Wan and Soifer 1974). This orientation to health care utilization
is based upon individuals becoming unhealthy, but it does not encompass utilization
of preventative services, or how factors such as health insurance, patient knowledge,
and patient satisfaction impact the decisions made once sick.
Preventive care refers to routine physical examinations, immunizations,
prenatal care, dental checkups, screening for heart disease and cancer, and other
services intended to ensure good health and prevent disease (Cockerham 2004: 112).
On the other hand, sick care refers to any physical examinations intended to cure an
illness, which is present in the body (Cockerham 2004).
Type of Health Insurance
According to the 2002 US Census data, Hispanics are more likely than all
other groups to be uninsured (Mills and Bhandari 2003). Table 2.1 shows health
insurance coverage rates by demographic variables. In general, males are more likely
than females to be uninsured. When it comes to age, individuals between
U.S. Health Insurance Coverage Rates by Demographic Characteristics
Male 843 833
Female 86.5 86.1
18 -24 years 7T3 703
25-34 years 76.6 75.1
35-44 years 833 8X3
45-64 years 853 853
65 years and older 00.2 993
Less than $25,000 7577 753
$25,000-$49,999 813 8077
$50,000-$74,999 : 8877 883
$75,000 and more 913 9TT8
Less than high school 714 7X0
High school diploma 8X5 313
Some college 8X5 8X0
College degree 313 793
White 853 8X8
Black 81 793
Hispanic 66.8 67.6
Asian/Facitic/Other oc >* oc 813
Source: Mills and Bhandari 2003
the ages 18-24 are more likely than all other age groups to not have health insurance
(Mills and Bhandari 2003). Education and health insurance coverage are positively
correlated such that the higher the educational attainment the higher the probability of
having health insurance. Similarly, income and health insurance coverage are also
positively correlated. For households earning less than $25,000 annually, 76.5
percent have health insurance. For households earning more than $75,000 annually,
91.8 percent have health insurance coverage (Mills and Bhandari 2003).
In order to have entry into the formal health care system, having health
insurance, whether it is public or private, is important (Leclere, Jensen, and
Biddlecoml994). The type of health insurance, including Medicaid, Medicare, or
private insurance, may be the intervening factor for determining the extent of health
care utilization (Ward 1987; Holly, Gardiol, Domenighetti, and Bisig 1998; Gibbons
and Wilcox-Gok 1998). Health insurance has a positive relationship with health care
utilization (Racine, Kaestner, Joyce, and Colman 2001; Kouzis and Eaton 1998).
Studies report large differences in the utilization of medical care across health care
providers (Daniels and Gatsonis 1999), specifically, those individuals having a
regular physician, which is most often associated with having health insurance, are
twice as likely to have utilized preventive services than those having no regular
physician (Cornelius, Smith, and Simpson 2002). By having a choice of physician,
which is associated with private health insurance, utilization of health care increases
(Ward 1987). Utilization is also affected by the amount of out-of-pocket expenses
associated with health insurance coverage, such that the higher the out-of-pocket
expenses such as co pays, the lower the utilization (Wan and Soifer 1974). When
sex, race, marital status, and education are controlled, type of health insurance is
strongly associated with utilization (Powell-Griner, Bolen, and Bland 1999). The
specific types of medical procedures also impacts health care utilization. For
example, Domenighetti, Bisig, Zaccheo, Gutzwiller, Lecomte, and Mizrahi (1996)
found that surgery rates are much higher for individuals with private insurance.
In 1996, an estimated 44.5 million Americans younger than 65 had no health
insurance (Powell-Griner, Bolen, and Bland 1999). Those individuals who are
uninsured have fewer physician visits, delayed medical care, limited access to this
care, lower use of preventive services, are less likely to have taken prescription drugs,
are less likely to be hospitalized, and have poorer health than individuals who have
health insurance (Franks Clancy, and Gold 1993; Newacheck, Stoddard, Hughes, and
Pearl 1998a; Schoen et al 2000; Stoddard, St. Peter, and Newacheck 1994; Powell-
Griner, Bolen, and Bland 1999; Davis and Reynolds 1976; Manning, Newhouse,
Duan, Keeler, Leibowitz, and Marquis 1987; Sudano and Baker 2003; Angel, Angel,
and Markides 2002; McWilliams, Zaslavsky, Meara, and Ayanian 2003; Baker,
Shapiro, and Schur 2000, 2003). These differences are not explained by differences
in perception of need to seek health care. Rather, lacking health insurance presents a
barrier to utilization of health care, such as an inability to pay for medical services
(Baker, Shapiro, and Schur 2000).
Medicaid is an entitlement program of federal-state matching dollars that
provides health insurance coverage to low-income persons falling below the federal
poverty level (Currie and Thomas 2001). Medicaid provides health insurance
coverage to approximately 40 million individuals (Coughlin, Long and Kendall
2002). The inception of Medicaid in 1965 has been linked to an increase of
hospitalization rates and doctor visits as well as a decrease in infant mortality (Currie
and Thomas 2001). Doctors must choose to participate in the Medicaid program and
most doctors do not. For those doctors that do accept Medicaid, they spend
significantly less time with Medicaid patients than with patients with private health
insurance (Currie and Thomas 2001; Sloan, Mitchell and Cromwell 1978; Decker
Medicare is a federal insurance-based program for those individuals aged 65
years and older and those individuals who are disabled. For Medicare,
socioeconomic factors are the prevailing ones on utilization of health care because
many [low-income] Medicare beneficiaries lack supplemental insurance that defrays
or eliminates cost sharing for visits with the physician and diagnostic tests and may
cover prescription drugs (McWilliams et al 2003: 762). Medicare coverage has been
linked to higher utilization rates for African Americans, the less educated, and those
individuals who were uninsured before receiving Medicare (McWilliams et al 2003).
Gomick, Eggers, Reilly, Mentnech, Fitterman, Kucken, Vladeck (1996) found that
African Americans and lower income Medicare recipients have fewer physician visits
than white and upper income recipients of Medicare.
There is no clear irrefutable empirical evidence between the utilization of
health services of individuals with private insurance and individuals with public
insurance. Some studies have found that recipients of Medicaid and Medicare have
lower levels of physician utilization than those who have private insurance
(Loewenstein 1971; Palmore and Jeffers 1971; Richardson 1971; Newacheck, Pearl,
Hughes, and Halfon 1998b); contradictory, Medicaid coverage has been correlated
with higher health care utilization and is significantly larger than the effect of other
types of insurance (Leclere, Jensen, Biddlecom 1994: 378). Hispanics with
Medicaid are more likely than others [Medicare and private insurance] to have had a
large number of doctors visits (Angel, Angel, and Markides 2002: 1266). In Currie
and Thomas (2001) research, children on Medicaid are more likely than children
who were uninsured or privately insurance to have a checkup within the last six
months. African American children on Medicaid are more likely than white children
on Medicaid to report using preventive services; however for sick care, white
children on Medicaid report a higher utilization than African American children
(Currie and Thomas 2001). Paradoxically, Sudano and Baker (2003) found very
little difference in the use of preventive services between privately insurance and
publicly insured individuals (131).
Health care utilization is impacted by knowledge (Leclere, Jensen, and
Biddlecom 1994). Specifically, health knowledge is predictive of health care
utilization (Green and Pope 1999). Knowledge is a factor that increases patients
participation and the ability to make informed decision about health (Falvo 1985;
Rankin and Stallings 1990). Gaining knowledge of health can be obtained from
family or friends, or the mass media including the Internet, television, radio, books,
or magazines. In general, lower-income individuals have less access to health
information and resources (Cockerham 2004: 99), and research has shown that
women, better educated, and younger individual engage in more information seeking
behaviors and therefore obtain more health knowledge (Green and Pope 1999;
Cockerham 2004; Butow, Brindle, McConell, Boakes, and Tattersall. 1998).
The family plays a valuable role in health knowledge by providing
information and advice to help the individual know how and when to enter into the
formal health care system (Doherty and Campbell 1988; Leclere, Jensen, and
Biddlecom 1994; Litman and Venters 1979; Litwak and Messeri 1989; Schor,
Starfield, Stidley, and Hankin 1987). The family, a significant social group, transmits
societal values to individuals. Having knowledge of disease, via familial information,
appears as the key intervening variable in a persons medical orientation, as
knowledge assists in recognition of symptoms (Cockerham 2004: 122). Families
have subcultural traits, which lead to differing opinions and definitions of illness
(Zola 1966) and affect how the individual perceives his or her illness (Cockerham
2004). While one family may define a fever and congestion as just a cold, another
may define it as an illness, which needs formal medical intervention.
Results from research are contradictory on whether gains in knowledge by
family and friends decrease or increases utilization. Family and friends can serve to
normalize [illness] attributions (e.g. I have that pain too) and reduce [utilization of
health services] (Kouzis and Eaton 1998: 1303). Family and friends can provide
informal medical treatments such as over-the-counter medicines for sick individuals,
which will keep the individual from utilizing formal health care services (Leclere,
Jensen, Biddlecom 1994). On the other had, utilization can increase because of the
failure of alternative medical remedies and perceptions of family and friends about
the problem (Walsh 1995). Eliot Friedsons (1960) research on ethnic groups found
that ethnic families have a strong sense of norms regarding health information, which
may contradict and conflict with utilizing the formal health care system.
Health care professionals also aid in individuals gaining health knowledge.
Patient education became essential when the high prevalence of acute diseases were
replaced with a high prevalence of chronic diseases (Redman 2002). Examples of
acute diseases are those that strike suddenly and disappear rapidly and include
chicken pox, influenza, and colds (Weitz 2004). Chronic illnesses are those illnesses
that strike slowly over time and are more than likely to be with the individual until
death. Examples of chronic illnesses are diabetes, asthma, and muscular sclerosis
(Weitz 2004). The goal of patient education is to help patients understand how
constant monitoring of the disease is important if not critical, as well as, to help
patients adhere to the medical treatment of their disease (Redman 2002).
In Flocke, Stange, and Goodwins (1998) study, doctors recommended and
educated patients about preventive services during an sick care visit 32 percent of the
time; however, the authors did not determine if receiving the preventive education
translated into higher utilization of preventive services. Doctors who educated their
patients about the difference between a common cold and an upper respiratory
infection found that utilization for health services decreased with no increase in
complications from the common cold (Mainous, Zoorob, Oler, and Haynes 1997).
The mass media, specifically the Internet, has allowed individuals to readily
access information about health and potential treatments. This form of knowledge
gain is not only easily accessible, but is also affordable which helps to overcome
financial barriers in assessing health issues (Krishna, Francisco, Balas, Konig, Graff,
and Madsen 2003).
Sixty-two to 80 percent of Americans have used the Internet to retrieve health
information (Conhaim 2003; Baker, Wagner, Singer, and Bundorf 2003). Other
research has estimated that 98 million Americans have used the Internet to get
additional health information, an estimated increase of 44 million since 1998 (Grover
et al 2002). The Internet has been reported to have played a crucial or important
role in healthcare decisions for one-quarter of the families that are dealing with a
serious illness (Conhaim 2003: 30). The most common uses of the Internet are
getting assistance with medical conditions and seeking information (Monnier, Laken,
and Carter 2002).
Individuals over 75 years of age are significantly less likely to use the Internet
to retrieve health information than younger individuals (Baker et al. 2003; Monnier,
Laken, and Carter 2002; Murray, Lo, Pollack, Donelan, Catania, White, Zapert, and
Turner. 2003). There is also a strong correlation with educational level and Internet
use, such that the higher the educational level the higher the Internet use for gaining
health information (Monnier, Laken, and Carter 2002; Murray et al 2003). Like
education, income is also positively related to gaining health knowledge via the
Internet (Murray et al 2003). The phrase, digital divide, has been coined to
describe the low use of the Internet by the lower class. Studies indicate that the lower
class represents only 9.7 percent of the Internet users (Dupuits 2002). Contrary to
this notion of the digital divide and Murray et als (2003) findings, Baker et al
(2003) found that income is not a predictive variable for Internet use. In addition to
socioeconomic variables, debates also exist when gender differences are in question.
Some research has showed no differences between men and women in gaining health
knowledge through the Internet while others have found a higher use of the Internet
for health knowledge by men (Dupuits 2002). When examining racial differences in
use of the Internet, the findings are constant. African Americans are less likely than
whites to use the Internet to gain information regarding health (Murray et al 2003).
There are still debates regarding how the Internet is changing health care
utilization patterns. Some researchers hypothesize that the Internet decreases health
care utilization while others feel that the Internet increases health care utilization
(Baker et al 2003; Hardey 1999).
The Internet could influence health care utilization by making patients more
knowledgeable, comfortable and confident when interacting with their physician.
This increase in knowledge, comfort, and confidence could make patients better able
to care for themselves and reduce the need to consume expensive health problems
that can be managed without additional help (Baker et al 2003: 2405).
This hypothesized reduction could be due to a number of variables. First,
individuals have access to alternative medical care such as herbal treatments and
homeopathy. These treatments often do not require any orthodox medical
intervention and can be obtained through the marketplace (Hardey 1999). In his
study, Hardey (2001) interviewed a respondent called David who echoed this. What
I found most useful [on the Internet] was the information about homeopathy and
allergies.... I had no idea that there were treatments that could get rid of sneezing
and itching with no side effects (392).
Secondly, individuals can use the mass media to get medical questions
answered instead of making doctor appointments or phone calls to answer questions.
Individuals feel that their doctors are too busy to answer the myriad of questions
relating to their health problems and found the Internet to be easier to get answers to
questions (Hardey 2001; A Few Doctors Seeing Patients Online 2004). In Baker et
al.s (2003) study, 67 percent of individuals who used the Internet for health
information did so to advance their knowledge of health issues.
The Internet has also provided a vehicle for selling prescription drugs, which
circumvents the formal health care system. By having the ability to get needed
prescription drugs prescribed, bought, and shipped through the Internet, individuals
have further reduced their need of utilizing their primary care physician. The
following is an example of a prescription drug website.
SafeWeb Medical is proud to bring you prescription
medicine through an easy, secure and confidential
environment.... Receive your order in 24-48 hours after
your prescription is approved by the medical doctors....
No appointment. No embarrassment. No waiting rooms.
Confidential (Hardey 2001: 400).
Conversely, it is also hypothesized that the gaining health information through
the Internet could increase formal health care utilization. The Internet can provide
improper, immaterial or unsound information, and inappropriate health screening
(Hardey 2001; Baker et al. 2003). If individuals use this type of information, they
may become sicker and need treatment from their primary care physician. Just the
knowledge that there is misinformation on the Internet may cause individuals to
utilize their primary care physician more to test the information found on the Internet
(A Few Doctors Seeing Patients Online 2004). Individuals also decide to go see
their primary care physician if they feel the medical condition is something the doctor
needs to look at and evaluate, such as strep throat (A Few Doctors Seeing Patients
Patient satisfaction is an attitude (Tucker 2002). Approximately 80 percent of
individuals are highly satisfied with their health care, (Jacobs, Shapiro, and Schulman
1993; Shapiro and Young 1986; Collins and OCathain 2003; Escarce, Kapur,
Solomon, Mangione, Lee, Adams, Wickstrom, and Quiter 2003); this high level of
satisfaction is only reflected in standardized surveys. Qualitatively, patients report
negative experiences and perceptions about their health care provider, which are not
ascertained through quantitative approaches (Collins and OCathain 2003; Bruster,
Jarmanc, Bosanquent, Weston, Ereus, and Delbanco 1994; Williams, Coyle, and
Healy 1998; Dougal, Russell, Rubin, and Ling 2000).
In Collins and OCathains (2003) study, patients emphasized that having
satisfaction with their physician included:
the importance of receiving a diagnosis, treatment and
cure; minimum waiting time for appointments and
treatment; the need to receive adequate information and
explanation; receiving individualized personal care,
including the need to be taken seriously; and the
importance of practitioner characteristics and good
Satisfaction has also been measured by the quality of the last physician-
patient interaction which included questions if the patient felt that the doctor
listened to everything he or she had to say including nonmedical topics, if the
patient was involved in the decision making, if the patient felt he or she spent
adequate time with the doctor, and if the patient felt the doctor respected
him/her (Saha, Arbelaez, and Cooper 2003; Wilson, Noseworthy, Rowe, and
Holroyd 2002). One variable that does not have an impact on patient
satisfaction is having tests run on the patient. In the study by Wilson et al
(2002), patients who presented in the emergency room with acute ankle
injuries and did not receive radiography were as satisfied as those patients
with the same condition who did receive radiography. The strongest predictor
for satisfaction among whites, African Americans, and Asians is having the
doctor treat the individual with respect; however for Hispanics, the strongest
predictor is spending adequate time with the patient (Saha, Arbelaez, and
Satisfaction is highest in individuals who have higher income and education
(Howard, Konrad, Stevens, and Porter 2001; Schoen et al 2000; McAlpine 2001).
However, Tucker (2002) found education to be negatively associated with
satisfaction, such that the higher the educational level, the lower the satisfaction.
African Americans were less satisfied with the quality of their medical care than
whites (La Veist, Nickerson, Bowie 2000), and Hispanics and Asians reported lower
satisfaction than whites and African Americans (Saha et al 1999; Saha, Arbelaez, and
Cooper 2003). One explanation to the lower levels of satisfaction among Hispanics
and Asians is language barriers between the patient and the physician (Carrasquillo,
Orav, Brennan, and Burstin 1999). In Carrasquillo et als (1999) study only 50
percent of patients who were non-English speakers were satisfied. Those individuals
who were younger are less satisfied with their physician (Carrasquillo et al 1999;
Type of health insurance is also associated with satisfaction such that higher
satisfaction of health care is associated with those individuals having an HMO
(Health Maintenance Organization) than those who did not (Harris and Associates
1986; Luft 1981,1983; Pope 1978; Ward 1987; Wolinsky 1980); however in those
health insurances which required preauthorization for referrals to specialists, patient
satisfaction was the lowest (Escarce et al 2003). Also those individuals who are
treated by a large group practice are less satisfied than those who are treated by a
smaller group practice. This could be due to the amount of time and personal
attention physicians in a large group practice are able to give to their patients
(Escarce et al 2003). Individuals, who have insurance, whether it is public or private,
have greater satisfaction than those individuals who are uninsured (Racine et al
2001). Similarly, the higher out-of-pocket expenses, the lower the patient satisfaction
(Steiber and Feber 1981). Specifically, Forrest and Frosts (1996) study found that
women who paid $20 or more for their gynecological exam were less satisfied than
those women whose insurance company paid the entire fee or who received free care.
Individuals receiving Medicaid and those uninsured are more likely to report low
satisfaction (McAlpine 2001); however between different private insurance plans no
significant difference has been found (Tudor, Riley, and Ingber 1998). Lastly, having
low trust in primary care physician is associated with low satisfaction (McAlpine
2001; La Veist, Nickerson, and Bowie 2000).
Numerous studies have found a relationship between satisfaction and health
care utilization. When patients are more satisfied, they are then more likely to utilize
health services in the future (Cehelyk 1994; Saha et al 1999; La Veist, Nickerson, and
Bowie 2000). Kersnik, Svab and Vegnuti (2001) examined individuals who had eight
contacts with the health care system in one year, which they labeled frequent
attenders. Compared to those who had less than eight contacts per year, the frequent
attenders were more satisfied with their general practioner. Kersnik, Svab, and
Vegnuti (2001) theorized that this high level of satisfaction transforms the doctor as a
friend and explains why these individuals become frequent attenders. Satisfaction
ratings are negatively correlated to utilization, specifically the higher the number of
hospitalizations the lower the satisfaction (Slater, Mitchell, and Cromwell 1981).
Dissatisfaction with primary care physician is strongly associated with either not
receiving or postponing needed care such that dissatisfied individuals are five times
more likely to postpone medical care (McAlpine 2001: 91). When satisfaction is the
dependent variable there is no correlation found between utilization of health care and
satisfaction (Ward 1987).
The relationship between income and health care utilization is positive
(Kouzis and Eaton 1998), specifically the poor receive less health care services than
the rich (National Center for Health Statistics 1974a; Schoen et al 2000; Currie and
Thomas 2001). Economic factors are the strongest explaining variable to why
individuals access care (Klugler, Yeash, and Rumbaugh 1993).
Research has shown that individuals in the lower socioeconomic status are
less likely to use preventive care while the middle and upper class use preventive care
routinely (Cockerham 2004). This may be that low-income individuals do not have a
regular physician and preventive services might not be covered under their health
insurance coverage, if any, therefore increasing the amount of money to poor must
pay out of pocket (Cockerham 2004). This underutilization of preventive care means
that the poor are more likely to use physicians for sick care more often (Cockerham
2004). More specifically, poorer individuals are more likely to report no use of
health care services, including preventative services, for at least 12 months
(Cornelius, Smith, and Simpson 2002; Schoen et al 2000; Sudano and Baker 2003).
The poor cannot afford to pay for health insurance or additional services they might
need because their incomes are low and affordable health insurance is lacking
(Dutton 1978). Higher income allows individuals to use discretion in health care
such as to receive preventative services and to have more flexibility to receive care
not covered by insurance (Bice 1971; Kravits and Schneider 1975; Leclere, Jensen,
and Biddlecom 1994; National Center for Health Statistics 1974b). Consumerism is
one explanation for higher utilization rates among the rich. In Cockerhams (2004)
research he noted that persons with higher socioeconomic status were more
consumer minded and expressed greater personal responsibility for their own health.
The poor were less discriminating in deciding which symptoms warranted a doctors
attention (133). Utilization of preventative services has the strongest positive
correlation to income (Dutton 1978). The middle-income range has the lowest
utilization of health care services (Dutton 1978). Contradictory to these previous
findings, Wan and Soifer (1974) and Green and Pope (1999) found that
socioeconomic status was not predictive of utilization of health care.
Being better educated is significantly associated with higher utilization rates
(Leclere, Jensen, and Biddlecom 1994; Kouzis and Eaton 1998), specifically;
individuals having at least a high school education are twice as likely to have
received preventative care (Cornelius, Smith and Simpson 2002). One reason is that
the less educated a more likely to be uninsured (Andersen et al. 1976; Sudano and
Baker 2003). Contradictory, Green and Pope (1999) found education to not affect
use of health services.
Gender is predictive of health care utilization (Green and Pope 1999),
specifically, women are more likely to utilize health services, including preventive
services, than men (Albizu-Garcia et al 2001; Nathanson 1975, 1977; Verbrugge
1979,1985,1988; Gove and Huges 1979; Briscoe 1987; Gijsbers Van Wijk, Kolk,
Vanden Bosch, Vanden Hoogen 1992; Macintyre, Hunt, and Sweeting 1996; Green
and Pope 1999; Mustard, Kaufert, Kozyrskyj, and Mayer 1998; Randhawa and Riley
1995; Schappert 1993; Bertakis, Azari, Helms, Callahan, and Robbins 2000). In
Tudiver and Talbots (1999) study, physicians reported than women utilize health
services not only for specific problems, such as sick care, but for general items, such
as check-ups. Contrary to previous studies, Bertakis et al (2000) concluded from
their study that for services such as emergency room care and hospitalizations, there
was no difference between use of services between men and women.
Women have higher numbers of physician visits due to the need for
obstetrical and other related care (Leclere, Jensen, and Biddlecoml994; Steiber and
Ferber 1981; Wan and Soifer 1974), but these obstetrical visits only account for less
than 20 percent of all doctor visits by women (Cockerham 2004). Other explanations
for womens higher utilization of health care services include higher rates of
morbidity in women than men, differences in health perceptions and the reporting of
symptoms and illness (Bertakis et al 2000: 147). Why men chose not to enter the
formal health care system may be because they are could tough it out because men
have a higher threshold of tolerance or because of traditional social roles men feel
immortal and do not want to relinquish control (Tudiver and Talbot 1999).
Other studies have shown that women are also more likely to be uninsured
and less likely to use a wide array of preventive services (Sudano and Baker 2003).
Lewis (1976) findings in his article, Women and National Health Insurance: Issues
and Solutions, revealed that women as a group may be identified as one of the
underserved or at least underinsured elements of the U.S. population (1976: 550).
In 1900 the average age expectancy was 47.3 years and has risen dramatically
to 76.9 years in 2000 (Cockerham 2004). Since 1958 the fertility rate has dropped
(Cockerham 2004). These two factors coupled together has translated into a larger
population of individuals ages 65 years and older. In 1900, only four percent of the
population was 65 years and older (Cockerham 2004) and by 2000, that number
reached 13 percent. It is projected that by the year 2020, the number of individuals
older than 64 will reach 18.5 percent of the total population (Weitz 2004). The
demand for utilization of health care will rise with this population (Cockerham 2004).
Utilization of health services is heavily age dependent (Cooper, Smaje, and
Arber 1998) and is predictive of health care utilization (Green and Pope 1999).
Those individuals age 65 and older have higher number of physician visits due to a
higher prevalence of chronic and degenerative diseases (Leclere, Jensen, and
Biddlecom 1994; Steiber and Ferber 1981; Wan and Soifer 1974). Individuals aged
45 or older had more than half of the physician visits in 2001 (Baby Boomers Making
Use of Health Care Services 2003). In a study conducted by Agency for Healthcare
Research and Quality, people aged 65 years and older were more likely to have had a
checkup in the last 12 months than those adults aged 18-64 years old, 89.2 percent to
75.3 percent respectively (National Survey Details Americas Experiences with
Health Care for Chronic Conditions 2002). Females aged 65 years and older have
higher utilization of health care than elderly men (Cockerham 2004). Contrary to all
of these previous findings, Klugler, Yeash, and Rumbaugh (1993) found that age did
not significantly affect utilization rates.
The study by Powell-Griner, Bolen, and Bland (1999) examined individuals
considered the near elderly, ages 55-64. For this age group men are more likely to
have health insurance than women, and African Americans and Hispanics are less
likely than whites to have health insurance. Education and income were inversely
related to having health insurance coverage. This study echoes previous research
findings that the insured were much more likely to report having a regular source of
medical care than those who uninsured.... [and this] was also associated with having
had a routine checkup ... and receiving a variety of clinical preventative services
(Powell-Griner, Bolen, and Bland 1999: 884).
Minority populations use less health care resources because they have less
access to care than the white population (Saha et al 1999). Being white is
significantly associated with higher use of physicians services regardless of income
or health status (Leclere, Jensen, and Biddlecom 1994; Racine et al 2001). Leclere,
Jensen, and Biddlecom (1994) found that being Hispanic does not impact the use of
physicians services. Contradictory, Sudano and Baker (2003) found that African
Americans and Hispanics were more likely to be uninsured and less likely to use
preventive services. Numerous studies have found that Mexican Americans have the
lowest utilization rates for any minority group (Angel and Angel 1996a, 1996b;
Roberts and Lee 1980). In Saha, Arbelaez, and Coopers (2003) study, African
Americans had the highest utilization rates and Hispanics and Asians had the lowest;
however once satisfaction with the physician and patient-physician race concordance
was adjusted, there was no association between race and health care utilization. This
echoes the findings in Klugler, Yeash, and Rumbaughs (1993) study, which found
that race did not significantly affect rates of utilization of health services.
Historically, whites used preventive care significantly more than minority groups;
however since 1970, African Americans utilization of preventive services has
increased (Cockerham 2004).
Whites are more likely than any other group to have health insurance
coverage and a regular source of care (Waidmann and Rajan 2000). For example,
whites use hospital services more than minority groups (Cooper, Smaje, and Arbeer
1998). African Americans are more likely to be insured through the government, but
are less likely than whites to have a regular source of care (Weinick, Zuvekas, and
Cohen 2000). Hispanics are more likely to have either no or only inadequate health
care coverage than non-Hispanics (Angel, Angel, and Markides 2002), as well as
twice as likely to lack a regular source of medical care (Waidmann and Rajan 2000;
Weinick, Zuvekas and Cohen 2000). Both African Americans and Hispanics are less
likely to be employed and in turn are less likely to receive health insurance
(Waidmann and Rajan 2000).
The health belief model explains ways in which individuals seek to avoid
illness. Irwin Rosenstock conceptualized the health belief model in 1966
(Cockerham 2004; Weitz 2004). Illness, in this model, is seen as a negative value in
which people will actively attempt to avoid while at the same time actively pursuing
positive values. Within the health belief model, there are four factors that are closely
interrelated and assist the individual in choosing healthy behaviors. The first factor is
individual perceptions. Individuals will seek preventive measures because the
individual perceives the susceptibility and seriousness or the illness as negative.
Therefore, the seeking of preventive measures would reduce the perceived negative
severity and seriousness of the illness (Cockerham 2004; Weitz 2004).
In addition to individual perceptions, modifying factors also affects healthy
behavior. There are three modifying factors. First are demographic variables, which
include age, gender, race, and income. Second are structural variables, which include
personality and peer pressure, and last are structural variables, including knowledge
about illnesses and prior contact with an illness (Cockerham 2004; Weitz 2004).
The third and fourth factors are cues to action and the likelihood of action.
Cues to action include the media campaigns or doctors advice. In the likelihood of
action, the individual weighs the perceived benefits of preventive measures against
the perceived barriers of preventive measures. Both of these factors act as triggers to
aid the individual to have healthy behaviors (Cockerham 2004; Weitz 2004).
The health belief model can explain both why individuals seek sick care as
well as preventive care. The health belief model was designed to explain why
individuals seek preventive care (Cockerham 2004). For sick care, the health belief
model explains why individual remain compliant with the treatment. If an individual
perceives the illness as having serious consequences if gone untreated, then the
individual will mostly likely comply with the treatment (Weitz 2004). The health
belief model is illustrated in figure 2.1.
Health Belief Model
Cues to Action
Although the health belief model is considered and reflected in this section,
these hypotheses are based on findings from previous research. The aim of this
research is to determine which factors best predicts utilization of formal health
services. In order to determine this objective, the following hypotheses will be
Hi: Individuals who have private health insurance will utilize formal
health care services more often than those individual having Medicaid,
Medicare, or those uninsured.
H2: Individuals who are satisfied with doctors and the health care system
will utilize formal health care services more often than those
individuals who are less satisfied with doctors and the health care
H3: Individuals who gain knowledge about health through family or
friends, the Internet, or health care professionals will utilize formal
health care services more often than those individuals who have not
gained knowledge about health.
H4: Individuals who have a higher annual income will utilize formal health
care services more often than those individuals who have a lower
H5: Individuals who have a higher educational level will utilize formal
health care services more often than those individuals who have a
lower educational level.
He: Women will utilize formal health care services more often than men.
H7: Whites will utilize formal health care services more often than
African Americans, Hispanics, Native Americans, or Asians.
The data were collected in 2000-2001 and comprised the third round of the
Community Tracking Study (CTS) Household Survey, which is sponsored by the
Robert Wood Johnson Foundation. The first and second rounds occurred in 1996-97
and 1998-99 respectively. Sixty areas, fifty-one metropolitan areas and nine
nonmetropolitan, were randomly selected to become the sites in which the research
was collected. These sites were representative of the nation as a whole.
The household surveys were obtained from individuals living in the 60 CTS
sites by using random-digit dialing techniques, as well as area probability sampling of
housing units so that those households without telephones would be represented. The
telephone samples were derived by randomly selecting some samples of Round Two
(1998-1999) telephone numbers and randomly selected some samples of telephone
numbers that were not part of the 1998-1999 samples. For the non-telephone sample,
interviewers attempted to recontact the addresses that were part of the 1998-1999
non-telephone sample and to contact new addresses of households without
telephones. All respondents were civilian, non-institutionalized.
Respondents provided information about and demographic characteristics,
household composition, unmet health care needs, use of health services, health
insurance coverage, out-of-pocket expenses for health care, usual source of care, last
visit to a medical provider, health status, patient and trust satisfaction, and risk
behaviors. There were 59,725 cases in total. The sample was derived from
individuals who answered the survey and was nationally representative. Table 3.1
presents the demographic characteristics of the sample as a whole.
The first dependent variable is utilization of health services for sick care.
This variable measures the number of times in the previous 12 months of the survey
that respondents reported receiving health services when sick. Combining four
questions from the data set created the dependent variable. The first question,
hospital visits, asked How many different times did you stay in any hospital
overnight or longer during the past 12 months? The answers ranged from 0-20
times. The second question, doctors visits, asked, In the past 12 months, about how
many times have you seen a doctor? Do not count doctors seen while an overnight
patient in a hospital or emergency room. The answers ranged from 0-20 times. The
third question, other medical professional visits,
asked, No counting doctors visits, how many times have you seen a nurse
practioners, physician, or midwife during the last 12 months? The answers ranged
from 0-4 times. The last question, surgeries, asked, Altogether, how many different
times have you had surgery during the past 12 months? The answers ranged from 0-
4. In the creation of the new variable, sick care, the answers ranged from 0-39 sick
care visits during the last 12 months. The total number of respondents was 10,719.
The second dependent variable is utilization of preventive services, which
measured the number of times in the previous 12 months that respondents reported
utilizing health care services for preventive care. The responses were yes=l and
no = 0. The other question that was included was, Did you get a flu shot in the
previous 12 months, with responses scored yes = 1 and no = 0. When these two
questions were combined, the answers ranged from 0-2. A0 meant that the
individual did not receive any preventive care and a 2 meant that the individual
received both a routing preventive check-up and a flu shot.
The first independent variable was satisfaction with health care. This
variable measured how satisfied the respondent reported being with various aspects
of the health care system. Combing three questions from the data set created this
independent variable. All of the responses were coded on a five point scale: 1 = very
satisfied, 2= somewhat satisfied, 3= neither satisfied nor dissatisfied, 4=somewhat
dissatisfied, 5= very dissatisfied. The first question was a constructed variable in the
data set that indicated the respondents satisfaction with their health care. The second
question was also a constructed variable in the data set that indicated the respondents
satisfaction with choice of primary care physician. The third question was a
constructed variable in the data set that indicated the respondents satisfaction with
choice of specialist. When these three questions were combined, the independent
variable remained coded on a five-point scale.
The second independent variable was type of health insurance. This
variable measured the type of insurance whether it is private, public, or no insurance
the respondent reported having at the time of the interview. Combining six questions
from the data set resulted in this independent variable. All of the responses were
yes or no. The following are the six questions used to make the variable type of
health insurance: 1. Are you covered by a health insurance plan from current or past
employers or union?; 2. Are you covered by a health insurance plan that you
bought on your own?; 3. Are you covered by a health insurance plan provided by
someone who does not live in this household?; 4. Are you covered by Medicare?;
5. Are you covered by Medicaid? 6. Do you have any health insurance coverage
that I might have missed? When recoded, questions one, two, and three were
recoded as a three, meaning they were covered by private health insurance. Question
four, was recoded as a three, meaning that the respondent had a government
entitlement health insurance. Question five was recoded as a two, meaning that the
respondent had a government health insurance program based on poverty, and
question six was recoded as a one, meaning that the respondent was uninsured.
The third independent variable, gaining health care knowledge, measured if
the respondent reported obtaining health knowledge in the last 12 months. The
following seven questions were combined to obtain this independent variable.
During the past 12 months, did you look for or get information about a personal
health concern: 1. on the Internet, 2. from friends or relatives, 3.from TV or radio, 4.
from books or magazines, 5. from somewhere else other than your doctors, other
health care professionals or health care organization, 6. from health care professional,
and 7. from health care organization. All of the responses were a yes or no.
Other independent variables include family income, age, sex, education,
gender, and race. Income was measured by asking the respondents, What was the
income your family received during 1999/2000 before taxes and deductions? The
answers were coded in the following categories: $0-$4999, $5000-$9999, $10,000-
$19,999, $20,000-$29,999, $30,000-$39,999, $40,000-$49,999, $50,000-$59,999,
$60,000-$69,999, $70,000-$79,999, $80,000-$89,999, $90,000-$90,999, $100,000-
$109,999, $110,000-$ 119,999, $120,000-$ 129,999, $130,000-$139,999, $140,000-
$149,999, $150,000 and above. Age was measured by asking respondents, What is
your age? Answers were categorized as the following: 0-17 years, 18-24 years, 25-
34 years, 35-54 years, 55-64 years, 65-90 years, and 91+ years. Sex was measured
by asking the respondents, What is your sex? Answers were categorized as either
male or female. Education was measured by asking each respondent, What is the
highest grade or year of school you completed? The answers were categorized as
follows: 6 or less years, 7-11 years, 12 years, 13-15 years, 16 years, 17-18 years, 19
or more years. Gender was coded as 1 = male and 2 = female. The independent
variable of race was combined from two questions in the data set. First respondents
were asked, Do you consider yourself to be Hispanic origin such as Mexican, Puerto
Rican, Cuban, or other Spanish backgrounds? Respondents answered yes or no.
The second question was What race do you consider yourself to be? The answers
were categorized as white, African American, or Native/Asian/Pacific/Other. By
combining these two questions, race is measured as what race or ethnic origin to
consider yourself to be? The answers are now, white, Hispanic, African American,
and Native/Asian/Pacific/Other. Because there are four variables, three were recoded
into dummy variables.
The primary aim of this study is to determine which who utilizes formal
health care services. Understanding how the independent variables, health insurance
coverage, patient satisfaction, knowledge, and demographics, influence the two
dependent variables, utilization of health services when sick and utilization of health
services for preventive care will achieve the aim of this study. Figures 2.1 shows the
conceptual model of analysis for this thesis.
Demographic Characteristics of Community Tracking Survey Sample
0 times receiving care 1976 18.4
1-5 times receiving care 6063 56.6
6-10 times receiving care 1621 15.1
11-15 times receiving care 573 5.4
16-20 times receiving care 245 2.3
21-25 times receiving care 203 1.9
26 + times receiving care 38 .3
0 times receiving care 61 .7
1 time receiving care 442 4.1
2 times receiving care 1322 12.3
3 times receiving care 840 7.8
Missing 8054 75.1
Private Insurance 6728 62.8
Medicare 1725 16.1
Medicaid 765 7.1
Uninsured 1233 11.5
Missing 268 2.5
Knowledge of health
Received from 0 sources 6472 60.4
Received from 1 source 1762 16.4
Received from 2 sources 1087 10.1
Received from 3 sources 675 6.3
Received from 4 or more sources 296 2.8
Missing 427 4.0
Table 3.1 Cont Count Percentage
Very Satisfied 2577 24.1
Somewhat Satisfied 903 8.5
Neither Satisfied or Dissatisfied 306 2.8
Somewhat Dissatisfied 111 1.0
Very Dissatisfied 31 .3
Missing 8054 63.3
Less than high school 1127 12.6%
High school diploma 3144 35.3%
Some college 2048 23.0%
College graduate 2596 29.1%
Male 4934 46.0%
Female 5785 54.0%
White 7812 73.4%
Hispanic 1172 11.0%
Black 1222 11.5%
Native/Asian/Other 431 4.1%
0-20 2267 21.1%
11-20 1300 12.1%
31-40 1818 17.0%
41-50 1964 18.3%
51-60 1436 13.4%
61-70 982 9.2%
71-80 733 6.8%
81-90 205 1.9%
$0 $50,000 5579 52.1
$50,001 $75,000 2096 19.6%
$75,001-$100,000 1372 12.7%
$100,001-$125,000 1059 9.9%
$125,001 -$150,000 616 5.7%
Conceptual Model of Analysis
Descriptive statistics were computed for the dependent variables and
independent variables and are presented in table 4.1. Because the kurtosis and
skewness for variables of sick care, satisfaction, and knowledge were above 1.0 or
below -1.0 the log for each was taken in order to ensure a normal distribution.
Linear regression was performed for the dependent variable sick care. The
findings are shown in Table 4.2. For the dependent variable, sick care, type of health
insurance was not statistically significant. Knowledge was statistically significant at
.001 with a positive standardized beta coefficient. The unstandardized beta was .125
which means that a one unit increase in knowledge results in a .125 unit increase in
utilization of the doctor for sick care. Satisfaction with the health care system was
not statistically significant. Age was statistically significant at .000 with a positive
standardized beta coefficient. The unstandardized beta was .006 which means that a
one unit increase in age will result in a .006 unit increase in health care utilization.
Income was statistically significant at .009 with a negative standardized beta
coefficient which means that as income decreases the likelihood of utilizing health
care increases. Gender was significant at .026 which means that women are more
Descriptive Statistics of Community Tracking Survey Sample (2000-2001)
Mean Median Std. Dev. Skewness Kurtosis Range N
Sick Care (log) 1.28 1.39 0.858 0.201 -0.65 4.00 10719
Preventive Care 1.08 1.00 0.656 -0.083 -0.692 2.00 10719
Health Insurance 3.33 4.00 1.038 -1.335 0.334 3.00 10719
Satisfaction (log) 1.44 1.39 0.399 1.041 0.192 2.00 10719
Knoweldge (log) 0.50 0.69 0.504 0.323 -1.337 2.00 10719
Income 58941 50000 40895.18 0.788 -0.251 150000 10719
Education 13.43 13.00 2.679 -0.182 0.247 13.00 10719
Gender 1.54 2.00 0.498 -0.159 -1.975 1.00 10719
Age 40.4 40.00 20.82 0.083 -0.734 91.00 10719
Race 1.46 1 0.849 1.675 1.553 3 10719
likely to utilize health care services. Education and race were not significant.
Linear regression was performed for the dependent variable preventive care.
The findings are shown in Table 4.3. Type of health insurance and knowledge gain
were not significant. Satisfaction was significant at .01 and had a negative
standardized beta coefficient of -.060. The unstandardized beta coefficient was -.096
Linear Regression for Sick Care
B Beta Significance
Insurance -0.017 -0.017 0.521
Satisfaction (log) -0.071 -0.037 0.129
Knowledge (log) 0.125 0.079 0.001
Age 0.006 0.129 0.000
Income -1.33E-06 -0.071 0.009
Education 0.002 0.007 0.791
Gender 0.091 0.054 0.026
White -0.032 -0.015 0.782
Black 0.087 0.03 0.508
Hispanic -0.037 -0.011 0.788
which means that one unit increase in satisfaction results in a .096 unit of decrease in
utilization of health care. Race and income were not significant. Age was significant
at .000 with a standardized beta coefficient of .361. The unstandardized beta
coefficient was .015, which means that as age increases one unit, utilization of health
care services will increase .015 units. Education was significant at .002 with at
standardized beta coefficient of .078. The unstandardized beta was .020, which
means that as education increases one unit, use of health care services increases by
.020 units. Gender was significant at .005 with a positive standardized beta
Linear Regression for Preventive Care
B Beta Significance
Insurance 0.013 0.016 0.538
Satisfaction (log) -0.96 -0.06 0.01
Knowledge (log) 0.046 0.035 0.135
Age 0.015 0.361 0.000
Income -9.60E-09 -0.006 0.812
Education 0.02 0.078 0.002
White -0.031 -0.17 0.737
Black 0.049 0.02 0.638
Hispanic -0.032 -0.012 0.765
coefficient of .065. The unstandardized beta coefficient was .092 which means that
women are more likely to utilize preventive services than men.
There were seven hypotheses tested. The first hypothesis, Individuals who
have private health insurance will utilize formal health care services more often than
those individuals having Medicaid, Medicare, or those uninsured, was not
significant for either utilization of sick or preventive care. Individuals do not need a
specific type of health insurance in order to enter the formal health care system. This
means that perhaps the health care system, including private doctors, are allowing any
individual, regardless of ability to pay, to receive needed sick and preventive health
care services. Another explanation is that individuals are paying critical versus non-
critical expenses out-of-pocket. Future studies should examine the amount of out-of-
pocket expenses to determine if indeed individuals are willing to pay whatever out-
of-pocket expenses are necessary to seek medical treatment.
Individuals who are satisfied with doctors and the health care system will
utilize formal health care services more often than those individuals who are less
satisfied with doctors and the health care system was the second hypothesis tested.
Although, satisfaction was significant for utilization of preventive services, the
direction of the beta did not support the hypothesis. This could be due to how
preventive care was conceptualized. Those individuals who are less satisfied are
more likely to utilize preventive services than those individuals who are satisfied.
Being satisfied with the health care system was not a determinant in utilizing sick
care. Interestingly, being satisfied with the health are system does not determine
utilization of services. It is quite the opposite. Being less satisfied with the health
care system promotes utilization of preventive services. This could bet that those
unsatisfied individuals fear being stricken with a chronic or acute illness because
they will be more entrenched in the health care system. They, therefore, see more of
a need to remain well than those individuals who are satisfied. Future studies need to
examine some of the other variables, such as time spent waiting for the doctor or the
quality of personal attention by the doctor. That affect satisfaction. By asking more
specific questions, perhaps the relationship between satisfaction and utilization will
The third hypothesis, Individuals who gain knowledge about health through
family or friends, the Internet, or health care professionals will utilize formal health
care services more often than those individuals how have not gained knowledge about
health, was supported but only for sick care. Those individuals who gain more
knowledge about health are more likely to utilized sick care services than those
individuals who did not gain knowledge about health. Perhaps this is because most
of the new medical information gained via friends, family, or the Internet are based
around new treatments for chronic and acute conditions. Individuals may have
previous knowledge about preventive measures, such as receiving a flu shot or having
an annual gynecological exam, prior to the twelve months of this study. Future
studies should seek to examine what type of information and knowledge the
individual is gaining in order to better determine the relationship between knowledge
and utilization of health care.
Individuals who have a higher annual income will utilize formal health care
services more often than those individuals who have a lower annual income was the
fourth hypothesis tested. Although income was significant for utilization of sick care
services, the direction of the beta did not support the hypothesis. Individuals with
lower income are more likely to utilize sick care services. This could be due to the
fact that they are sicker due to environment and sociological causes such as stress,
poor nutrition, lack of transportation, or unsafe housing.
The fifth hypothesis, Individuals who have a higher educational level will
utilize formal health care services more often than those individuals who have a
lower educational level, was supported but only for utilization of preventive
services. Individuals with higher educational levels are more likely to utilize
preventive services than individuals with lower educational levels. Education was
not a determinant for sick care. Individuals with higher educations may have access
to more knowledge about the importance of preventive services.
Women will utilize formal health care services more often than men was the
sixth hypothesis tested. This hypothesis was supported for both sick and preventive
care. Women do utilize health care services more than men.
The seventh hypothesis tested was Whites will utilize formal health care
services more often than African Americans, Hispanics, Native Americans, or
Asians. Race was not significant for either utilization of sick care or preventive care
and therefore the hypothesis was not supported.
The strengths of this study are three fold. First, satisfaction was an
independent variable is used significantly less in studies than satisfaction as the
dependent variable. This study goes beyond determining who is satisfied and asks
the question so what? What are the implications for the health care system?
Rethinking patient satisfaction is a tremendous strength in this study. Second,
examining the differences between utilization of sick care versus preventive care is a
major strength. As the results show, there are different variables to explain why
individuals utilize these different health services. Being able to clearly define how
these health services are used does beyond any previous study. Third, including
knowledge gain in this study is innovative. With the creation of the Internet,
individuals now have more medical information at their fingertips than ever before.
The relationship between the Internet and utilization has not been widely studied.
This study draws from the lack of research presently out there and brings the 21st
century into medical sociology.
There are several limitations to this study. First, this study was limited to the
questions asked by the primary data collection. The construction of the dependent
variable, preventive care, was extremely limited to two questions in the data set.
More questions regarding types of preventive care services received, such as have
you had a colonoscopy? or have you had a physical? would have greatly added to
the construction of this variable. Second, the data collected were based on self-
reports and were not validated by another source. For example, it was not validated
how many times an individual saw a doctor in the past 12 months. The data might be
subject to recall bias.
This study sought to examine the determinants of formal health care system
utilization. Three questions guided this thesis. First, does having health insurance
impact utilization of health services? Second, how does satisfaction affect utilization,
and third, how does gaining knowledge about health affect the likelihood of using
health services? Results suggested that the main determinant affecting utilization of
sick care is gaining health knowledge. For utilization of preventive health care
services, satisfaction with doctor was the main determinant.
Albizu-Garcia, Carmen E., Alegria, Margarita, Freeman, Daniel, and Vera, Mildred.
(2001). Gender and Health Services Use for a Mental Health Problem. Social
Science and Medicine, 53, 865-878.
Andersen, R., Lion, J., and Anderson, 0. (1976). Two Decades of Health Services:
Social Survey Trends in Use and Expenditure. Cambridge, MA: Ballinger.
Angel, Ronald J., Angel, Jacqueline L., and Markides, Kyriakos S. (2002). Stability
and Change in Health Insurance among Older Mexican Americans:
Longitudinal Evidence From the Hispanic Established Populations for
Epidemiologic Study of the Elderly. American Journal of Public Health,
Baby Boomers Making Use of Health Care Services. (2003). The Journal of Dental
Hygiene, 77(111), 156,184.
Baker, David W., Shapiro, Martin F., and Schur, Claudia L. (2000). Health Insurance
and Access to Care for Symptomatic Conditions. Arch of Internal Medicine,
Baker, Laurence, Wagner, Todd H., Singer, Sara, and Bundorf, M. Kate. (2003). Use
of the Internet and E-mail for Health Care Information. JAMA, 259(18), 2400-
Bertakis, Klea D., Azari, Rahman, Helms, L. Jay, Callahan, Edward J., and Robbins,
John A. (2000). Gender Differences in the Utilization of Health Care Services.
The Journal of Family Practice, 49(2), 147-152.
Bice, T.W. (1971). Medical Care for the Disadvantaged. Baltimore: Johns Hopkins
University Department of Medical Care and Hospitals.
Briscoe, M. (1987). Why Do People Go To the Doctor? Sex Differences in the
Correlates of GP Consultation. Social Science and Medicine, 25, 507-513.
Bruster, S., Jarmanc, B., Bosanquent, N., Weston, D., Ereus, R., and Delbanco, T.
(1994). National Survey of Hospital Patients. BMJ, 309,1542-1546.
Carrasquillo, Olveen, Orav, E. John, Brennan, Troyen A., and Burstin, Helen R.
(1999). Impact of Language Barriers on Patient Satisfaction in an Emergency
Department. Journal of Gen Intern Med, 14, 82-87.
Cehelyk, Mindy Klaus. (1994). Utilization Review and Its Impact on Patient
Knowledge and Satisfaction. Unpublished doctoral dissertation, New York
University, New York, NY.
Collins, Karen and OCathain, Alicia. (2003). The Continuum of Patient Satisfaction:
From Satisfied to Very Satisfied. Social Science and Medicine, 57,2465-
Conhaim, Wallys W. (2003 February). Is the Doctor Online? Information Today, 29-
Cooper, Helen, Smaje, Chris, and Arber, Sara. (1998). Use of Health Services by
Children and Young People According to Ethnicity and Social Class:
Secondary Analysis of a National Survey. BMJ, 317,1047-1051.
Cornelius, Llewellyn J., Smith, Pamela L., Simpson, Gaynell M. (2002). What
Factors Hinder Women of Color From Obtaining Preventive Health Care?
American Journal of Public Health, 92(4), 535-538.
Coughlin, Theresa A., Long, Sharon K., and Kendall, Stephanie. (2002). Health Care
Access, Use, and Satisfaction among Disabled Medicaid Beneficiaries. Health
Care Financing Review, 24(2), 115-129.
Daniels, Michael J. and Gatsonis, Constantine. (1999). Hierarchical Generalized
Linear Models in the Analysis of Variations in Health Care Utilization.
Journal of the American Statistical Association, 94(445), 29-43.
Davis, Karen, and Reynolds, R. (1976). The Impact of Medicare and Medicaid on
Access to Medical Care. In Richard. N. Rosett (Ed.),772e Role of Health
Insurance in the Health Services Sector (pp. 391-435). New York: National
Bureau of Economic Research.
Decker, Sandra. (1993). The Effect of Physician Reimbursement Levels on the
Primary Care of Medicaid Patients. Cambridge: Harvard University.
Doherty, William and Campbell, Tom. (1988). Families and Health. Beverly Hills,
Domenighetti G., Bisig, B., Zaccheo, A., Gutzwiller, F., Lecomte, T., Mizrahi, A.R.
(1996). Consommation Chirurgicale En Suisse Et Camparaison Avec La
France. Lausanne: Office federal de la statistique, Editions Realites Sociales.
Dougal, A., Russell, A., Rubin, G., and Ling, J. (2000). Rethinking Patient
Satisfaction: Patient Ecperiences of an Open Access Flexible Sigmoidoscopy
Service. Social Science and Medicine, 50, 53-62.
Dupuits, Francois M.H. (2002). The Effects of the Internet on Pharmaceutical
Consumers and Providers. Mismanage Health Outcomes, 10(11), 679-691.
Dutton, Diana. (1978). Explaining the Low Use of Health Services by the Poor:
Costs, Attitudes, or Delivery Systems? American Sociological Review, 43(3),
Escarce, Jose J., Kapur, Kanika, Solomon, Matthew D., Mangione, Carole M., Lee,
Paul P., Adams, John L., Wickstrom, Steven L., and Quiter, Elaine S. (2003).
Practice Characteristics and HMO Enrollee Satisfaction with Specialty Care:
An Analysis of Patients with Glaucoma and Diabetic Retinopathy. Health
Services Research, 38(4), 1135-1155.
Falvo, D.R. (1985). Effective Patient Education. Rockville, MD: Aspen Systems.
A Few Doctors Seeing Patients Online. (2004, January 12). Health and Medicine
Flocke, Susan A., Stange, Kurt C., and Goodwin, Meredith A. (1998). Patient and
Visit Characteristics Associated with Opportunistic Preventive Services
Delivery. Journal of Family Practice, 47(3), 202-212.
Forrest, Jacqueline Darroch, and Frost, Jennifer J. (1996). The Family Planning
Attitudes and Experiences of Low-Income Women. Family Planning
Perspectives, 28(6), 246-255.
Franks, Peter, Clancy, Carolyn M., and Gold, Marthe R. (1993). Health Insurance
and Mortality: Evidence from a National Cohort. JAMA, 270(6), 737-741.
Friedson, Eliot. (1960). Client Control and Medical Practice. American Journal of
Sociology, 65, 374-382.
Gibbons, Robert D. and Wilcox-Gok, Virginia. (1998). Health Service Utilization and
Insurance Coverage: A Multivariate Probit Analysis. Journal of the
American Statistical Association, 93(441), 63-72.
Gijsbers Van Wijk, C.M.T., Kolk, A.M.M., Vanden Bosch, W.J.H.M., and Vanden
Hoogen, H.J.M. (1992). Male and Female Morbidity in General Practice: The
Nature of Sex Differences. Social Science and Medicine, 35, 665-678.
Gomick, Marian E., Eggers, Paul W., Reilly, Thomas W., Mentnech, Renee M.,
Fitterman, Leslye K., Kucken, Lawrence E., and Vladeck, Bruce C. (1996).
Effects of Race and Income on Mortality and Use of Services Among
Medicare Beneficiaries. New England Journal of Medicine, 335(11), 791-799.
Gove, W.R., and Huges, M. (1979). Possible Causes of the Apparent Sex Differences
in Physical Health: An Empirical Investigation. American Sociological
Review, 44, 126-146.
Green, Carla A. and Pope,Clyde R. (1999). Gender, Psychosocial Factors and the
Use of Medical Services: A Longitudinal Analysis. Social Science and
Hardey, Michael. (1999). Doctor in the House: The Internet as a Source of Lay
Health Knowledge and the Challenge to Expertise. Sociology of Health and
Illness, 21(6), 820- 836.
Hardey, Michael. (2001). E-Health: The Internet and the Transformation of
Patients into Consumers and Producers of Health Knowledge. Information,
Communication, and Society, 4(3), 388-405.
Harris, Louis and Associates, Inc. (1986). A Report Card on HMOs: 1980-1984.
Report prepared for the Hendy J. Kaiser Family Foundation, Palo Alto, CA.
Holly, Alberto, Gardiol, Lucien, Domenighetti, Gianfranco, and Bisig, Brigitte.
(1998). An Economic Model of Health Care Utilization and Health Insurance
in Switzerland. European Economic Review, 42, 513-522.
Howard, Daniel L., Konrad, Thomas R., Stevens, Catherine, and Porter, Carol Q.
(2001). Physician-Patient Racial Matching, Effectiveness of Care, Use of
Services, and Patient Satisfaction. Research on Aging, 23(1), 83-108.
Jacobs, Lawrence R., Shaprio, Robert Y., and Schulman, Eli C. (1993). Poll Trends:
Medical Care in the United States- an Update. Public Opinion Quarterly,
Kersnik, Janko, Svab, Igor, and Vegnuti, Miljana. (2001). Frequent Attenders in
General Practice: Quality of Life, Patient Satisfaction, Use of Medical
Services and GP Characteristics. ScandJPrim Health Care, 19, 174-177.
Klugler, John P., Yeash, James, and Rumbaugh, Paul C. (1993). The Impact of
Sociodemographic, Health Care System, and Family Function Variables on
Prenatal Care Utilization in a Military Setting. Journal of Family Practice,
Kouzis, Anthony C. and Eaton, William W. (1998). Absence of Social Networks,
Social Support, and Health Services Utilization. Psychological Medicine, 28,
Kravits, J. and Schneider, J. (1975). Health Care Need and Actual Use by Age, Race,
and Income. In R. Andersen, J. Kravits, and O.W. Andersen (Eds.), Equity in
Health Services (pp. 169-187). Cambridge, MA: Ballinger.
Krishna, Santosh, Francisco, Benjamin D., Balas, E. Andrew, Konig, Peter, Graff,
Gavin R., and Madsen, Richard W. (2003). Internet-Enabled Interactive
Multimedia Asthma Education Program: A Randomized Trial. Pediatrics,
La Veist, Thomas A., Nickerson, Kim J., and Bowie, Janice V. (2000). Attitudes
about Racism, Medical Mistrust, and Satisfaction with Care among African
American and White Cardiac Patients. Medical Care Research and Review,
Leclere, Felicia B., Jensen, Leif, and Biddlecom, Ann E. (1994). Health Care
Utilization, Family Context, and Adaptation Among Immigrants to the United
States. Journal of Health and Social Behavior, 35(4), 370-384.
Lewis, D. (1976). Women and National Health Insurance: Issues and Solutions.
Medical Care, 14, 549-558.
Litman, Theodor and Venters, Maureen. (1979). Research on Health Care and the
Family. Social Science and Medicine, 13A, 379-385.
Litwak, Eugene and Messeri, Peter. (1989). Organizational Theory, Social Supports,
and Mortality Rates: A Theoretical Convergence. Annual Review of
Loewenstein, R. (1971). Early Effects of Medicare on the Health Care of the Aged.
Social Security Bulletin, 34, 3-20.
Lufit, Harold S. (1981). Health Maintenance Organizations: Dimensions of
Performance. New York: Wiley.
Luft, Harold S. (1983). Health-maintenance Organizations. In D. Mechanic (Ed.),
Handbook of Health, Health Care, and the Health Professions (pp. 318-351).
New York: Free Press.
Macintyre, S., Hunt, K., Sweeting, H. (1996). Gender Differences in Health: Are
Things Really as Simple as They Seem? Social Science and Medicine, 42(4),
Mainous, Arch G., Zoorob, Roger J., Oler, Michael J., and Haynes, Delicia M.
(1997). Patient Knowledge of Upper Respiratory Infections: Implications for
Antibiotic Expectations and Unnecessary Utilization. Journal of Family
Practice, 45(1), 75-84.
Manning, Willard, Newhouse, Joseph, Duan, Naihua, Keeler, Emmett, Leibowitz,
Arleen, and Marquis, Susan. (1987). Health Insurance and the Demand for
Medical Care: Evidence from a Randomized Experiment. American
Economic Review, 77(3), 251-277.
McAlpine, Donna. (2001). Trust in Primary Care Physicians: Does Managed Care
Matter? Unpublished doctoral dissertation, Rutgers University, New
McWilliams, J. Michael, Zaslavsky, Alan M., Meara, Ellen, and Ayanian, John Z.
(2003). Impact of Medicare Coverage of Basic Clinical Services for
Previously Uninsured Adults. JAMA, 290(6), 757-764.
Mills, Robert J. and Bhandari, Shailesh. (2003). Health Insurance Coverage in United
States: 2002. www.census/gov/prod/2003pubs/p60-223.pdf.
Monnier, Jeannine, Laken, Marilyn, and Carter, Cindy L. (2002). Patient and
Caregiver Interests in Internet-Based Cancer Services. Cancer Practice,
Murray, Elizabeth, Lo, Bernard, Pollack, Lance, Donelan, Karen, Catania, Joe,
White, Martha, Zapert, Kinga, and Turner, Rachel. (2003). The Impact of
Health Information on the Internet on the Physician-Patient Relationship.
Arch Intern Med, 163,1727-1734.
Mustard, Cameron A., Kaufert, Patricia, Kozyrskyj, Anita, and Mayer, Teresa.
(1998). Sex Differences in the Use of Health Care Services. New England
Journal of Medicine, 335(23), 1678-1683.
Nathanson, C.A. (1975). Illness and the Feminine Role: A Theoretical Review.
Social Science and Medicine, 9, 51-62.
Nathanson, C.A. (1977). Sex, Illness, and Medical Care: A Review of Data, Theory,
and Method. Social Science and Medicine, 11, 13-25.
National Center for Health Statistics, U.S. Department of Health, Education, and
Welfare. (1974a). Disability Days, U.S., 1971. (Vital and Health Statistics,
Ser. 10, No. 90, DHEW Publication. (HSM) 72-1064). Washington, D.C.:
U.S. Government Printing Office.
National Center for Health Statistics, U.S. Department of Health, Education, and
Welfare (1974b). Physician Visits: Volume and Interval Since Last Visit,
U.S., 1971. (Vital and Health Statistics, Ser. 10, No. 97, DHEW Publication
No (HRA) 74-1517). Washington, D.C.: U.S. Government Printing Office.
National Survey Details Americas Experiences with Health Care for Chronic
Conditions. (2002, September 9). Managedhealthcare.Info, 5.
Newacheck, P.W., Stoddard, J.J., Hughes, D.C., and Pearl, M. (1998a). Health
Insurance and Access to Primary Care for Children. New England Journal of
Medicine, 338, 513-519.
Newacheck P.W., Pearl, M., Hughes, D.C., and Halfon, N. (1998b). The Role of
Medicaid in Ensuring Childrens Access to Care. JAMA, 280,1789-1793.
Palmore, E. and Jeffers, F.C. (1971). Health Care in a Longitudinal Panel Before and
After Medicare. Journal of Gerontology, 26,532-536.
Pope, Clyde R. (1978). Consumer Satisfaction in a Health Maintenance Organization.
Journal of Health and Social Behavior, 19,291-303.
Powell-Griner, Eve, Bolen, Julie, and Bland, Shayne. (1999). Health Care Coverage
and Use of Preventive Services Among the Near Elderly in the United States.
American Journal of Public Health, 89(6), 882-886.
Racine, Andrew D., Kaestner, Robert, Joyce, Theodore J., and Colman, Gregory J.
(2001). Differential Impact of Recent Medicaid Expansions by Race and
Ethnicity. Pediatrics, 108(5), 1135-1142.
Randhawa, J. and Riley, R. (1995). Trends in Hospital Utilization, 1982-1983 to
1992-1993. Health Rep, 7,41-19.
Rankin, S.H. and Stallings, K.D. (1990). Patient Education: Issues, Principles and
Practices (2nd ed.). Philadelphia, PA: J.B. Lippincott.
Redman, Barbara K. (2002). Patient Education Services as Oppressor? Health Care
for Women International, 23, 773-780.
Richardson, W.C. (1971). Ambulatory Use of Physicians Services in Response to
Illness Episodes in a Low-income Neighborhood. (Research Series No. 29).
Chicago: Center for Health Administration Studies.
Saha, Somnath, Arbelaez, Jose J., and Cooper, Lisa A. (2003). Patient-Physician
Relationship and Racial Disparities in the Quality of Health Care. American
Journal of Public Health, 03(10), 1713-1719.
Schappert, S.M. (1993). National Ambulatory Care Survey: 1991 Summary.
(Advance Data From Vital and Health Statistics. No. 230). Hyattsville, MD:
National Center for Health Statistics.
Schoen, Cathy, Davis, Karen, DesRoches, Catherine, Donelan, Karen, and Blendon,
Robert. (2000). Health Insurance Markets and Income Inequality: Findings
from an International Health Policy Survey. Health Policy, 51, 67-85.
Schor, Edward, Starfield, Barbara, Stidley, Christine, and Hankin, Janet. (1987).
Family Health: Utilization and Effects of Family Medicine. Medical Care,
Shapiro, Robert Y. and Young, John T. (1986). The Polls: Medical Care in the
United States. Public Opinion Quarterly, 50(3), 418-428.
Slater, V., Linn, M.W., and Harris, R. (1981). Outpatient Evaluation of Mental Health
Care. Southern Medical Journal, 74, 1217-1219.
Sloan, Frank, Mitchell, Janet, and Cromwell, Jerry. (1978). Physician Participation in
State Medicaid Programs. Journal of Human Resources, 13(2), 211-245.
Steiber, Steven R. and Leonard A. Ferber. (1981, Summer). Support for National
Health Insurance: Intercohort Differentials. Public Opinion Quarterly, 179-
Stoddard, J.J., St. Peter, R.F., and Newacheck, P.W. (1994). Health Insurance and
Ambulatory Care for Children. New England Journal of Medicine, 330, 1421 -
Sudano, Joseph J., and Baker, David W. (2003). Intermittent Lack of Health
Insurance Coverage and Use of Preventive Services. American Journal of
Public Health, 95(1), 130-137.
Tucker, Jessie L. (2002). The Moderators of Patient Satisfaction. Journal of
Management in Medicine, 16(1), 48-66.
Tudiver, Fred, and Talbot, Yves. (1999). Why Dont Men Seek Help? Family
Physicians Perspectivers on Help-Seeking Behavior in Men. Journal of
Family Practice, 48(1), 47.
Verbrugge, L.M. (1979). Female Illness Rates and Illness Behavior: Testing
Hypothesis about Sex Differences in Health. Women and Health, 4, 61-79.
Verbrugge, L.M. (1985). Gender and Health: An Update of Hypotheses and
Evidence. Journal of Health and Social Behavior, 26,156-182.
Verbrugge, L.M. (1988). Unveiling Higher Morbidity for Men: The Story. In Riley,
M.W., B.J. Huber, B.B. Hess (Eds), Social Structures and Human Lives.
Newbury Park, CA: Sage Publications, 138-160.
Waidmann, Timothy, and Rajan, Shruti. (2000). Race and Ethnic Disparities in
Health Care Access and Utilization: An Examination of State Variation.
Medical Care Research and Review, 57, 55-84.
Walsh, Mike. (1995). The Health Belief Model and Use of Accident and Emergency
Services by the General Public. Journal of Advanced Nursing, 22, 694-699.
Wan, Thomas T.H. and Soifer, Scott J. (1974). Determinants of Physician
Utilization: A Causal Analysis. Journal of Health and Social Behavior,
Ward, Russell A. (1987). HMO Satisfaction and Understanding Among Recent
Medicare Enrollees. Journal of Health and Social Behavior, 28(4), 401-412.
Weinick, Robin M., Zuvekas, Samuel H., and Cohen, Joel W. (2000). Racial and
Ethnic Differences in Access to and Use of Health Care Services, 1977 to
1996. Medical Care Research and Review, 57(1), 36-54.
Williams, B., Coyle, J., and Healy, D. (1998). The Meaning of Patient Satisfaction:
An Explanation of High Reported Levels. Social Science and Medicine,
Wilson, Deborah E., Noseworthy, Thomas W., Rowe, Brian H., and Holroyd, Brian
R. (2002). Evaluation of Patient Satisfaction and Outcomes After Assessment
for Acute Ankle Injuries. American Journal of Emergency Medicine, 20(1),
Wolinsky, Frederic D. (1980). The Performance of Health Maintenance
Organizations: An Analytic Review. Milbank Memorial Fund Quarterly, 58,
Zola, I.K. (1966). Culture and Symptoms- an Analysis of Patients Presenting
Complaints. American Sociological Review, 31, 615-630.