This article was downloaded by: [University of Newcastle, Australia] On: 09 January 2015, At: 18:13 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Women & Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/wwah20
Primary Care Arrangements and Access to Care Among AfricanAmerican Women in Three Chicago Communities a
b
Michele A. Kelley MSW, ScD , Janet D. Perloff PhD , c
Naomi M. Morris MD, MPH & Wangyue Liu MS
d
a
Assistant Professor of Community Health Sciences, School of Public Health, University of Illinois at Chicago b
Associate Professor, School of Social Welfare and the School of Public Health, The State University of New York at Albany c
Professor and Director of Community Health Sciences, School of Public Health, University of Illinois at Chicago d
Research Specialist, School of Public Health, University of Illinois at Chicago Published online: 05 Nov 2010.
To cite this article: Michele A. Kelley MSW, ScD , Janet D. Perloff PhD , Naomi M. Morris MD, MPH & Wangyue Liu MS (1992) Primary Care Arrangements and Access to Care Among African-American Women in Three Chicago Communities, Women & Health, 18:4, 91-106, DOI: 10.1300/J013v18n04_06 To link to this article: http://dx.doi.org/10.1300/J013v18n04_06
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
& Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/ page/terms-and-conditions
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Primary Care Arrangements and Access to Care Among African-American Women in Three Chicago Communities A. Kelley, MSW, ScD Janet D. Perloff, PhD Naomi M. Morris, MD, MPH Wangyue Liu, MS
Michele
ABSTRACT. African-American women of child-bearing age residing in Uuee high-risk communities in Chicago were surveyed regard-
ing their primary care arrangements and access to care (n = 552). This study examined factors which differentialed women who used office-based ractices from Umse who used inslilutional settings (community cinics, health deparlment clinics, hospital-based clinics) for primary care. Results of multivariate analysis indicate that women who used office-based practices were more likely than those who --
-
--
Michele A. Kelley is Assistnnt Professor of Community Health Sciences, Naomi M. Monis is Professor and Director of Community Health Sciences and Wangyue Liu is Research Specialist, all at the School of Public Health, The University of Illinois at Chicago. Janet D. Perloff is Associate Professor at the School of Social Welfare and the School of Public HealQ The State University of New York at Albany. This paper was presented at the Public Health Association. Department of Preventive nnd Social Medicine. The University of Olago. Dunedin, New Zealand, March, 1991, and at the Annual Meeting of the American Public Health Association, Atlanta, November, 1991. This research was supported in part by a grant from the Chicago Community Trust, Grant # 8703H. The authors gratefully acknowledge the helpful comments of Uuee anonymous reviewers and the commitment of the women interviewed in h i s study. Women & Health, Vol. 18(4) 1992 Haworlh Press, Inc. All rights resewed.
Q 1992 by The
91
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
92
WOMEN & IIEALTIf
used instilulionnl settings to see the same provider, to walk to heir provider, to have less travel time and to walk in without an appointment. They were less likely to be hospitalized in the past year and less likely to report the availability of family planning at their usual source of care. Salisfaction with care, insurance status and sociodemographic characteristics were not associated with use of a parlicular facility type. Implications for organizing comprehensive health services for this population are discussed. INTHODUCTION
National data continue to reveal racial disparities in health status and access to medical care. For black women, the life expectancy at birth in 1988 was 73.8 years compared with 78.9 years for white women (National Center for Health Statistics [NCHS],1990). Black women of child-bearing age are significantly more likely to give birth to a low birth weight infant or to an infant who does not survive the first year of life (U.S.Public Health Service, 1991). A major goal of the U.S. Public Health Service as stated in Healthy People 2000 is to reduce inequities in health status through prevention. Specific objectives targeting blacks refer to decreasing risk factors (e.g., smoking, overweight), reducing preventable deaths (e.g., heart disease, stroke) and increasing access to primary health care (e.g.. screening, health education) including preconceptional and prenatal care (U.S. Public Health Service, 1991). Much of the literature on access to urimarv care for blacks focuses on racial differences in the amount of care received, accessibility of care and satisfaction with care. Studies consistently find that compared to whites, blacks have less access to care, are less likely to use office-based physicians, are less likely to be satisfied with their care, and have fewer physician visits relative to their health status (Aday, Anderson & Fleming 1980; Blendon, Aiken, Freeman & Corey, 1989: Kleinman, Gold & Makuc. 1981; NCHS. 1990). Pregnant black women are less likely to receive early and sufficient prenatal care (US. Public Health Service, 1991). African-Americans and other minority groups clearly appear to have benefitted from policies designed to increase access to care as evidenced by increases in health care utilization and improved health status (NCHS, 1990). However, there is evidence that progress has diminished in the 1980's with regard to access to care. While the average number
.
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Kelley
et
al.
93
of physician contacts for whites increased from 5.2 to 5.5 between 1983 and 1988, the average number of visits for blacks decreased from 4.9 to 4.8 (NCHS,1990). For both blacks and whites, changes over the past decade were observed in which use of office-based physicians decreased and use of non-profit community clinics increased. However, blacks were more likely to change from use of hospital outpatient departments to office-based physicians during this period so that in 1988, almost one-half (49.3%) identified this facility type as the usual source of care (NCHS,1990). There is evidence that institutional sources of care such as government sponsored clinics and hospital clinics have more structural barriers to care, e.g., increased office and appointment waiting time (Dutton, 1986). However, there is a paucity of research on the comparative experiences of low income black women using private office-based physicians relative to other care settings. Having a regular source of care is in$ortant for prevention, especially for women of child-bearing age. While it is estimated that only twenty percent of African-Americans are without a regular source of care (Robert Wood Johnson Foundation, 1987), having a regular source of care does not ensure that preventive care is received. A study by Makuc and colleagues (1989) found that most women without a cancer screening test had a recent physician contact. While health reform is needed to ensure that all women have access to preventive health services, insights into the current care arrangements of black women residing in high-risk communities can be useful for planning health promotion interventions directed toward achievement of the health objectives for the year 2000 (Healthy People 2000). Toward this end, two overall questions guided our analyses for this study:
,
.I. What factors differentiate women who use office-based practices versus institutional sources for primary care? 2. Controlling for socio-demographiccharacteristics, what is the relative importance of health status and selected characteristics of the usual source of care in explaining use of officebased versus institutional sources for care? The conceptual basis for the study is taken from the work of Aday and Anderson (1974). This behavioral model classifies factors
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
94
WOMEN & HEALTH
potentially associated with health care utilization into predisposing variables (socio-demographic factors, perceptions of efficacy-of care), enabling variables (insurance coverage, cost, convenience factors), and need (health status). Based on general population studies and given the geographic proximity of office-based physicians to our study population (Chicago Department of Health, 1990), it was hypothesized that women who used private office-based physicians would report better access to care compared to those who used institutional settings.
METHODS Subjects were black women aged 18-44, residing in three low income communities in Chicago with high infant mortality rates: the Near West Side, West Garfield Park and Austin. These community areas, together with two additional communities make up a larger area referred to by local health officials as the West Side Corridor (Figure 1). The West Side Corridor is 89.7% African-American, with 59.3% of the population under 200% of the federal poverty level. Each of the five community areas comprising the West Side is at least 70% African-American and each is designated as a "high risk" area for maternal and child health (Chicago Department of Health, 1990). with an overall infant mortality raie of 20.9(per 1000 live births) in 1988, nearly twice the national target rate of 11 for black infants (U.S. Public Health Service, 1990). Subjects were part of a study designed to examine trends in health care use among black women of child-bearing age, given recent state and local maternal and child health policy initiatives. The authors conducted cross-sectional surveys in the fall of 1987 and the fall of 1989. Technical expertise for the sampling design, instrument development and data collection was provided to the authors' specifications under a contract with the University of Illinois Survey Research Laboratory (SRL). Interviews were conducted by black females trained and supervised by SRL. A two stage sampling method was used. First, census tracts were randomly selected from among those in the lowest two quartiles of household income as indicated in the 1980 census. Then blocks were randomly selected from within the selected tracts. Subjects
Kelley el al.
95
PIGORE 1
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
City of Chicago Community Areas Featuring the West Side Corridor
-
camunities included in st*
senpic.
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
96
WOMEN & HEALTH
were eligible to be interviewed in their homes if they were black, between the ages of 18 and 44 and were either pregnant at the time of the interview or had a child under six years of age in the home. Subjects were identified by door-to-door screening. Total cooperation, i.e., with both screening and completed interview, was 89.1 percent in Wave I and 87.7 percent in Wave 2. Cases were pooled from both time periods to maximize statistical power as no significant differences were found in the distributions of variables of interest in each time period (n = 607). (No subject was included in both surveys.) Women who reported that they had no regular source of care (n = 49). or whose source of care could not be identified or classified into a facility type (n = 6) were excluded, yielding a final sample size of 552. Measures
The questionnaire measured socio-demographic characteristics, self-reported health status, pregnancy history and health care access and use. Items were adapted from the widely used National Health Interview Survey core questionnaire (NCHS, 1989). To reduce the possibility of misclassification error in the dependent variable, subjects were asked both the name and address of the place that they usually used for ambulatory (non-emergent) care. This information was then verified using local medical provider listings and site visits to assign the appropriate facility type: office-based physician, hospital EWOPD, non-profit community clinic or health department clinic. As preliminary analysis indicated no statistically significant differences in the distributions of independent variables across "institutional" facility types (hospital EWOPD, non-profit community clinic or health department clinic), a dichotomous outcome variable was created indicating whether or not the medical provider was institutionally based or a private, office-based physician. RESULTS
The sample was almost evenly divided between the two major facility types. Approximately 51.8% of the women identified an office-based practice as the usual source of care.The remaining
Kelley el of.
97
48.2% used institutional facilities. Most women (68.4%) used the usual source of care exclusively in the past year. The majority of women (71.4%) were receiving benefits from the Aid to Families with Dependent Children (AFDC) program, indicating a low income population. The propoaion of women completing high school was 61.8%, with older women more likely to complete high school = 4.85,g = 1; p < .05). As many socio-demographic indicators were intercorrelated, selected variables and their association with the health care facility type are shown in Table 1. Only two factors were associated with primary care facility type: women with three or more children and with a working telephone were more likely to use an institutional source for care. In this rather homogeneous population, socio-demographicvariables (predisposing factors), generally failed to distinguish users and non-users of office-based physicians. Table 2 describes the association of health status and access indicators with primary care facility type. Three measures of self-reported health status including hospitalization in past year, were not associated with facility type. Neither an overall measure of satisfaction with care, nor having insurance (including Medicaid-77'.7% of the sample) had an association with type of facility used. In the bivariate analysis, indicators of access to care (enabling factors) and access to family planning services differentiated women who used private office-based physicians from those who used institutional care. With the exception of office waiting time, results consistently indicated better accessibility with use of office-based physicians. Other factors failed to discriminate between users and non-users of office-based physicians: receipt of a pap smear test, parity, current or recent pregnancy, adequacy of prenatal care received (among women delivering in the past year), volume or likelihood of visits in the past year. However, a measure of comprehensiveness of care, availability of family planning services, was associated with use of institutional providers. In order to determine the relative importance of access indicators and other characteristics of the usual source of care in explaining use of facility type, multivariate analysis was performed using the SAS logistic regression procedure (SAS, 1985). To develop a more parsimonious model, a stepwise variable selection procedure was used, with four "background" socio-demographicvariables included in the model prior to the fist step: community area, age, number
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
(x2
WOMEN & HEALTH
98
Table 1. Soclodemographlc Factors by Primary Care Facllity Type
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Office -Based In = 286) n (%I
Age 18-28 27-44 Marital status now marrled not married
Education < 12th grade > 12th grade AFDC~ yes
no
Family structure woman alone with other adult No. children In household less than 3 3 or more Labor force employed unemployed Have telephone Yes no a uslng the x2 test of slgnlllcance
'
b
pc.05 -p 15 minutes
Office wait c 30 minutes > 30 minutes
-
Provider continuity yes no
Family planning avai7able yes
no Satisfaction with care e YE no a Using the
2 test of signfieanfa.
72
i252j
55
Pe.05 -p < .O1 -p
Women & Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/wwah20
Primary Care Arrangements and Access to Care Among AfricanAmerican Women in Three Chicago Communities a
b
Michele A. Kelley MSW, ScD , Janet D. Perloff PhD , c
Naomi M. Morris MD, MPH & Wangyue Liu MS
d
a
Assistant Professor of Community Health Sciences, School of Public Health, University of Illinois at Chicago b
Associate Professor, School of Social Welfare and the School of Public Health, The State University of New York at Albany c
Professor and Director of Community Health Sciences, School of Public Health, University of Illinois at Chicago d
Research Specialist, School of Public Health, University of Illinois at Chicago Published online: 05 Nov 2010.
To cite this article: Michele A. Kelley MSW, ScD , Janet D. Perloff PhD , Naomi M. Morris MD, MPH & Wangyue Liu MS (1992) Primary Care Arrangements and Access to Care Among African-American Women in Three Chicago Communities, Women & Health, 18:4, 91-106, DOI: 10.1300/J013v18n04_06 To link to this article: http://dx.doi.org/10.1300/J013v18n04_06
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
& Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/ page/terms-and-conditions
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Primary Care Arrangements and Access to Care Among African-American Women in Three Chicago Communities A. Kelley, MSW, ScD Janet D. Perloff, PhD Naomi M. Morris, MD, MPH Wangyue Liu, MS
Michele
ABSTRACT. African-American women of child-bearing age residing in Uuee high-risk communities in Chicago were surveyed regard-
ing their primary care arrangements and access to care (n = 552). This study examined factors which differentialed women who used office-based ractices from Umse who used inslilutional settings (community cinics, health deparlment clinics, hospital-based clinics) for primary care. Results of multivariate analysis indicate that women who used office-based practices were more likely than those who --
-
--
Michele A. Kelley is Assistnnt Professor of Community Health Sciences, Naomi M. Monis is Professor and Director of Community Health Sciences and Wangyue Liu is Research Specialist, all at the School of Public Health, The University of Illinois at Chicago. Janet D. Perloff is Associate Professor at the School of Social Welfare and the School of Public HealQ The State University of New York at Albany. This paper was presented at the Public Health Association. Department of Preventive nnd Social Medicine. The University of Olago. Dunedin, New Zealand, March, 1991, and at the Annual Meeting of the American Public Health Association, Atlanta, November, 1991. This research was supported in part by a grant from the Chicago Community Trust, Grant # 8703H. The authors gratefully acknowledge the helpful comments of Uuee anonymous reviewers and the commitment of the women interviewed in h i s study. Women & Health, Vol. 18(4) 1992 Haworlh Press, Inc. All rights resewed.
Q 1992 by The
91
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
92
WOMEN & IIEALTIf
used instilulionnl settings to see the same provider, to walk to heir provider, to have less travel time and to walk in without an appointment. They were less likely to be hospitalized in the past year and less likely to report the availability of family planning at their usual source of care. Salisfaction with care, insurance status and sociodemographic characteristics were not associated with use of a parlicular facility type. Implications for organizing comprehensive health services for this population are discussed. INTHODUCTION
National data continue to reveal racial disparities in health status and access to medical care. For black women, the life expectancy at birth in 1988 was 73.8 years compared with 78.9 years for white women (National Center for Health Statistics [NCHS],1990). Black women of child-bearing age are significantly more likely to give birth to a low birth weight infant or to an infant who does not survive the first year of life (U.S.Public Health Service, 1991). A major goal of the U.S. Public Health Service as stated in Healthy People 2000 is to reduce inequities in health status through prevention. Specific objectives targeting blacks refer to decreasing risk factors (e.g., smoking, overweight), reducing preventable deaths (e.g., heart disease, stroke) and increasing access to primary health care (e.g.. screening, health education) including preconceptional and prenatal care (U.S. Public Health Service, 1991). Much of the literature on access to urimarv care for blacks focuses on racial differences in the amount of care received, accessibility of care and satisfaction with care. Studies consistently find that compared to whites, blacks have less access to care, are less likely to use office-based physicians, are less likely to be satisfied with their care, and have fewer physician visits relative to their health status (Aday, Anderson & Fleming 1980; Blendon, Aiken, Freeman & Corey, 1989: Kleinman, Gold & Makuc. 1981; NCHS. 1990). Pregnant black women are less likely to receive early and sufficient prenatal care (US. Public Health Service, 1991). African-Americans and other minority groups clearly appear to have benefitted from policies designed to increase access to care as evidenced by increases in health care utilization and improved health status (NCHS, 1990). However, there is evidence that progress has diminished in the 1980's with regard to access to care. While the average number
.
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Kelley
et
al.
93
of physician contacts for whites increased from 5.2 to 5.5 between 1983 and 1988, the average number of visits for blacks decreased from 4.9 to 4.8 (NCHS,1990). For both blacks and whites, changes over the past decade were observed in which use of office-based physicians decreased and use of non-profit community clinics increased. However, blacks were more likely to change from use of hospital outpatient departments to office-based physicians during this period so that in 1988, almost one-half (49.3%) identified this facility type as the usual source of care (NCHS,1990). There is evidence that institutional sources of care such as government sponsored clinics and hospital clinics have more structural barriers to care, e.g., increased office and appointment waiting time (Dutton, 1986). However, there is a paucity of research on the comparative experiences of low income black women using private office-based physicians relative to other care settings. Having a regular source of care is in$ortant for prevention, especially for women of child-bearing age. While it is estimated that only twenty percent of African-Americans are without a regular source of care (Robert Wood Johnson Foundation, 1987), having a regular source of care does not ensure that preventive care is received. A study by Makuc and colleagues (1989) found that most women without a cancer screening test had a recent physician contact. While health reform is needed to ensure that all women have access to preventive health services, insights into the current care arrangements of black women residing in high-risk communities can be useful for planning health promotion interventions directed toward achievement of the health objectives for the year 2000 (Healthy People 2000). Toward this end, two overall questions guided our analyses for this study:
,
.I. What factors differentiate women who use office-based practices versus institutional sources for primary care? 2. Controlling for socio-demographiccharacteristics, what is the relative importance of health status and selected characteristics of the usual source of care in explaining use of officebased versus institutional sources for care? The conceptual basis for the study is taken from the work of Aday and Anderson (1974). This behavioral model classifies factors
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
94
WOMEN & HEALTH
potentially associated with health care utilization into predisposing variables (socio-demographic factors, perceptions of efficacy-of care), enabling variables (insurance coverage, cost, convenience factors), and need (health status). Based on general population studies and given the geographic proximity of office-based physicians to our study population (Chicago Department of Health, 1990), it was hypothesized that women who used private office-based physicians would report better access to care compared to those who used institutional settings.
METHODS Subjects were black women aged 18-44, residing in three low income communities in Chicago with high infant mortality rates: the Near West Side, West Garfield Park and Austin. These community areas, together with two additional communities make up a larger area referred to by local health officials as the West Side Corridor (Figure 1). The West Side Corridor is 89.7% African-American, with 59.3% of the population under 200% of the federal poverty level. Each of the five community areas comprising the West Side is at least 70% African-American and each is designated as a "high risk" area for maternal and child health (Chicago Department of Health, 1990). with an overall infant mortality raie of 20.9(per 1000 live births) in 1988, nearly twice the national target rate of 11 for black infants (U.S. Public Health Service, 1990). Subjects were part of a study designed to examine trends in health care use among black women of child-bearing age, given recent state and local maternal and child health policy initiatives. The authors conducted cross-sectional surveys in the fall of 1987 and the fall of 1989. Technical expertise for the sampling design, instrument development and data collection was provided to the authors' specifications under a contract with the University of Illinois Survey Research Laboratory (SRL). Interviews were conducted by black females trained and supervised by SRL. A two stage sampling method was used. First, census tracts were randomly selected from among those in the lowest two quartiles of household income as indicated in the 1980 census. Then blocks were randomly selected from within the selected tracts. Subjects
Kelley el al.
95
PIGORE 1
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
City of Chicago Community Areas Featuring the West Side Corridor
-
camunities included in st*
senpic.
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
96
WOMEN & HEALTH
were eligible to be interviewed in their homes if they were black, between the ages of 18 and 44 and were either pregnant at the time of the interview or had a child under six years of age in the home. Subjects were identified by door-to-door screening. Total cooperation, i.e., with both screening and completed interview, was 89.1 percent in Wave I and 87.7 percent in Wave 2. Cases were pooled from both time periods to maximize statistical power as no significant differences were found in the distributions of variables of interest in each time period (n = 607). (No subject was included in both surveys.) Women who reported that they had no regular source of care (n = 49). or whose source of care could not be identified or classified into a facility type (n = 6) were excluded, yielding a final sample size of 552. Measures
The questionnaire measured socio-demographic characteristics, self-reported health status, pregnancy history and health care access and use. Items were adapted from the widely used National Health Interview Survey core questionnaire (NCHS, 1989). To reduce the possibility of misclassification error in the dependent variable, subjects were asked both the name and address of the place that they usually used for ambulatory (non-emergent) care. This information was then verified using local medical provider listings and site visits to assign the appropriate facility type: office-based physician, hospital EWOPD, non-profit community clinic or health department clinic. As preliminary analysis indicated no statistically significant differences in the distributions of independent variables across "institutional" facility types (hospital EWOPD, non-profit community clinic or health department clinic), a dichotomous outcome variable was created indicating whether or not the medical provider was institutionally based or a private, office-based physician. RESULTS
The sample was almost evenly divided between the two major facility types. Approximately 51.8% of the women identified an office-based practice as the usual source of care.The remaining
Kelley el of.
97
48.2% used institutional facilities. Most women (68.4%) used the usual source of care exclusively in the past year. The majority of women (71.4%) were receiving benefits from the Aid to Families with Dependent Children (AFDC) program, indicating a low income population. The propoaion of women completing high school was 61.8%, with older women more likely to complete high school = 4.85,g = 1; p < .05). As many socio-demographic indicators were intercorrelated, selected variables and their association with the health care facility type are shown in Table 1. Only two factors were associated with primary care facility type: women with three or more children and with a working telephone were more likely to use an institutional source for care. In this rather homogeneous population, socio-demographicvariables (predisposing factors), generally failed to distinguish users and non-users of office-based physicians. Table 2 describes the association of health status and access indicators with primary care facility type. Three measures of self-reported health status including hospitalization in past year, were not associated with facility type. Neither an overall measure of satisfaction with care, nor having insurance (including Medicaid-77'.7% of the sample) had an association with type of facility used. In the bivariate analysis, indicators of access to care (enabling factors) and access to family planning services differentiated women who used private office-based physicians from those who used institutional care. With the exception of office waiting time, results consistently indicated better accessibility with use of office-based physicians. Other factors failed to discriminate between users and non-users of office-based physicians: receipt of a pap smear test, parity, current or recent pregnancy, adequacy of prenatal care received (among women delivering in the past year), volume or likelihood of visits in the past year. However, a measure of comprehensiveness of care, availability of family planning services, was associated with use of institutional providers. In order to determine the relative importance of access indicators and other characteristics of the usual source of care in explaining use of facility type, multivariate analysis was performed using the SAS logistic regression procedure (SAS, 1985). To develop a more parsimonious model, a stepwise variable selection procedure was used, with four "background" socio-demographicvariables included in the model prior to the fist step: community area, age, number
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
(x2
WOMEN & HEALTH
98
Table 1. Soclodemographlc Factors by Primary Care Facllity Type
Downloaded by [University of Newcastle, Australia] at 18:13 09 January 2015
Office -Based In = 286) n (%I
Age 18-28 27-44 Marital status now marrled not married
Education < 12th grade > 12th grade AFDC~ yes
no
Family structure woman alone with other adult No. children In household less than 3 3 or more Labor force employed unemployed Have telephone Yes no a uslng the x2 test of slgnlllcance
'
b
pc.05 -p 15 minutes
Office wait c 30 minutes > 30 minutes
-
Provider continuity yes no
Family planning avai7able yes
no Satisfaction with care e YE no a Using the
2 test of signfieanfa.
72
i252j
55
Pe.05 -p < .O1 -p
