What Factors Hinder Women of Color From Obtaining Preventive Health Care?

Abstract

This commentary examines how women of color fare on the use of preventive care. Logistic regression models of women's use of preventive care were computed with data from the 1994 Commonwealth Fund Minority Health Survey. It was found that having a regular doctor was the most consistent predictor of the use of preventive care, irrespective of the women's racial/ethnic background, socioeconomic circumstances, or place of residence.

These findings reinforce the importance of physicians in the delivery of preventive care. Suggestions for improving the use of preventive services by women of color are provided.

SINCE THE 1950S, researchers have consistently provided evidence that health disparities exist between Whites and other Americans in infant mortality and in mortality due to diabetes, heart disease, cerebrovascular disease, and malignant neoplasms.1,2 However, it was the landmark 1985 Report of the Secretary's Task Force on Black and Minority Health3 that highlighted the array of factors contributing to disparities in health care among people of color. This report suggested that such disparities resulted from a lack of access to health care providers, differences in healthy behaviors (e.g. diet, smoking, exercise), socioeconomic differences, and other environmental factors. Moreover, the report suggested that interventions to reduce disparities among persons of color should emphasize the full continuum of screening, detection, treatment, and follow-up care.

Within the arena of preventive health care, recent studies have highlighted the importance of maintaining healthy eating habits, stopping smoking, and drinking sufficient water; adhering to a rigorous regular exercise regimen; obtaining regular prevention screenings or examinations; and detecting risks for disease as ways to reduce the risk of death due to breast cancer, lung cancer, cervical cancer, cardiovascular problems, asthma, diabetes, pneumonia, and influenza.4–13 In spite of the known merits of these preventive measures, women of color have historically been less likely than Whites to receive preventive health services and to engage in healthy behaviors.14–19 In this commentary we examine the barriers to receipt of preventive care for African American, Asian American, and Latina women.

METHODS

The data for this study come from the 1994 Commonwealth Fund Minority Health Survey (CMHS).20 The 1994 CMHS was a national probability sample of 3789 adults 18 years and older. The CMHS sample was weighted to reflect the minorities' proportionate representation in the US population as reflected in the Census Bureau's March 1994 Current Population Survey.

The variables used in the analyses for this article were as follows:

1. Race and ethnicity. Classification by ethnic/racial background was based on information reported for each household member. Respondents were asked if their racial background was best described as African American, Black, Asian or Pacific Islander, Native American or Alaskan Native, White, or another race. All respondents were also asked whether their main national origin or ancestry was among the following Hispanic American subpopulations, regardless of racial background: Puerto Rican, Cuban, Mexican, Dominican, Costa Rican, or other Hispanic American.

Respondents who described themselves as Asian or Pacific Islander were asked if they were of Chinese, Vietnamese, Korean, or other Asian heritage. The categories of White, Asian/Pacific Islander, Vietnamese, Korean, Chinese, and African American excluded those with Hispanic ancestry.

2. Location (urban, rural, suburban) and region of residence.

3. Insurance status. Respondents were asked whether they were covered by health insurance provided through the respondent's work or union, health insurance through someone else's work or union, health insurance purchased directly by the respondent or the respondent's family, health insurance obtained through some other group insurance, Medicare, Medicaid, or public aid. Respondents who did not have any of these types of insurance were defined as being uninsured.

4. Income level. Income level was based on the respondent's total 1993 household income.

5. Education level.

6. Perceived health status. Perceived health status was based on self-reported responses to the question “In general, how would you describe your own health—excellent, good, fair or poor?”

8. Perceived discrimination. Perceived discrimination was defined according to the respondent's answer of yes or no to the question of whether she had felt uncomfortable or felt that she had been treated badly because of her race or ethnicity, sex, age, health or disability, income level, or any other characteristic.

7. Regular doctor. Regular doctor was defined according to the respondent's answer of yes or no to the question of whether she had a doctor she would usually go to when sick or in need of health care.

9. Receipt of preventive care. Receipt of preventive care was defined as a response of yes to the question of whether the respondent had “received preventive care (such as blood pressure tests, pap smears, or cholesterol level readings).”

Eight logistic regression models (one for each racial/ethnic group: African Americans; all Latinas; Mexican Americans; Puerto Ricans; all Asians/Pacific Islanders; and Vietnamese, Korean, and Chinese Americans) were computed. The dependent variable for each model was whether or not the respondent received preventive care in 1994. Each model included demographic characteristics (region and location of residence, education level, income), perceived health status, insurance status, and perceived discrimination. Odds ratios (ORs) were reported for each of the logistic regression models. The Student t statistic was used to report the statistical significance of the coefficients in these models. The Stata statistical package21 was used for the analyses because it adjusts for the multistage sampling strategy used in CMHS.

RESULTS

Table 1 ▶ shows a logistic regression model of correlates of receipt of any preventive care for women of color. By far the most consistent predictors of the decision to seek preventive care were having a regular doctor and having a high school education. Women who had a regular doctor were at least twice as likely as those who did not have a regular doctor to have received preventive care in 1994 (P < .05). (For Puerto Rican women, this relationship was close to being statistically significant [P = .08].) Women who had a high school education were more than twice as likely as other women to have received preventive care in 1994 (P < .05). Neither perceived discrimination nor location of residence (urban, rural, or suburban) made a difference in whether the women received preventive care.

TABLE 1.

—Odds Ratios for Correlates of Receipt of Preventive Health Care by Women of Color 18 Years and Older, by Race/Ethnicity: United States, 1994

	African Americans (n = 444)	All Latinas (n = 494)	Mexican Americans (n = 261)	Puerto Ricans (n = 80)	All Asian/Pacific Islander Americans (n = 313)	Vietnamese Americans (n = 94)	Korean Americans (n = 108)	Chinese Americans (n = 103)
Location
Urban	1.07	0.89	0.77	0.31	2.70	. . .a	. . .a	. . .a
Rural	0.55	0.77	0.62	0.31	. . .	. . .a	. . .a	. . .a
Suburban	1.00	1.00	1.00	1.00	1.00	. . .a	. . .a	. . .a
Region
Northeast	0.71	0.66	0.13*	0.55	1.06	1.86	0.49	1.32
Midwest	1.40	0.57	0.82	2.32	1.22	2.16	0.42	2.96
South	1.04	0.93	0.94	0.43	0.85	1.18	0.31	1.31
West	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Insurance status
Uninsured	0.57	0.56*	0.58	0.04*	0.81	1.11	0.76	0.27
Insured	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
1993 Household income, $
<15 000	1.26	0.55†	0.89	1.44	0.54	0.49	0.24	0.84
15 000–25 000	1.35	0.78	1.01	0.79	0.82	0.56	0.63	1.05
25 000–50 000	0.77	0.91	1.02	0.67	0.91	0.96	1.20	0.61
>50 000	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Education
<High school	0.41*	0.39***	0.30**	0.06**	0.91	0.86	4.04	0.47
High school	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
>High school	0.44**	0.38***	0.32**	0.32	0.46*	0.39	0.76	0.24*
Perceived health
Fair/poor	2.25*	2.05**	1.62	2.70	1.35	1.36	1.35	2.88
Good/excellent	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Regular doctor
Yes	2.58**	2.74***	3.41***	3.90†	2.76***	3.13*	3.39*	3.08*
No	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Perceived discrimination
Yes	0.69	0.89	0.71	0.69	1.40	1.74	1.89	0.76
No	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
χ2	42.24	76.26	51.83	26.80	33.98	11.25	21.49	19.75
df	14	14	14	14	13	12	12	12
P	<.001	<.001	<.001	<.05	<.01	.5078	<.05	.072

Source. 1994 Commonwealth Fund Minority Health Survey.20

^aCoefficients could not be computed for these variables because of small cell sizes.

*P < .05; **P < .01; ***P < .001; †P = .08.

African American women with a high school education were more than twice as likely as other African American women to have received preventive care. Among Latinas taken as a group, having a regular doctor, having insurance, having a high school education, and perceived fair or poor health were correlated with receipt of preventive care (P < .05). Living in a low-income household was a nearly statistically significant (P = .08) barrier to the receipt of preventive care for Latinas. There were, however, differences between Latina subpopulations. Having insurance was correlated with receipt of preventive care for Puerto Ricans, but not for Mexican Americans. Living in the Northeast was a barrier to the receipt of preventive care for Mexican Americans, but not for Puerto Ricans.

For Asian/Pacific Islander Americans taken as a group, having a high school education and having a regular doctor were correlated with receipt of preventive care. However, educational level was a significant correlate for Chinese Americans but not for Vietnamese or Korean Americans.

DISCUSSION

This study found that regardless of race and ethnicity, women of color who had a regular doctor were at least twice as likely as those who did not to receive preventive care. None of these women saw perceived discrimination as a barrier to receiving preventive care. Living in a rural location did not, per se, pose a barrier to receipt of preventive care. Being uninsured was a barrier to receipt of preventive care for Puerto Rican women, while being poor was a barrier to the receipt of preventive care for all Latinas.

These findings suggest that ensuring access to a regular physician is an important part of addressing the preventive health needs of all women of color. At the same time, we may need to develop educational interventions to help these women meet their preventive health needs. The findings of this study regarding education level, income level, and having a regular doctor support other studies that have examined the relationship between these factors and the receipt of preventive care.15,21–26

This study also suggests that women of color may have needs that vary within and among ethnic populations. For example, insurance status was significant for Puerto Ricans but not for Mexican Americans. Perceived health status was a significant correlate of receipt of preventive care for African Americans and Latinas, but not for Asian/Pacific Islanders. These factors suggest that we may need different interventions to target different racial or ethnic groups.

There was one finding that contradicted previous literature. This study did not find differences in receipt of preventive care between urban, suburban, and rural women of color. Other studies have reported that rural and inner-city residents are typically less likely than others to have access to health care providers and that they have to travel further to seek medical care.27–30 It may be that the barrier to receipt of preventive care by rural women of color is the unavailability of health care providers. One study reported that there was almost a 2-fold difference in the overall population–physician ratio between nonmetropolitan areas (1117 persons per physician) and metropolitan areas (579 persons per physician) in the 399 counties that constitute Applachia. The same study found that there was almost a 3-fold difference in the availability of registered dieticians between metropolitan areas (1009 persons per dietician) and nonmetropolitan areas (2923 persons per dietician).31

While it is encouraging to report that having a regular doctor facilitated the receipt of preventive care among women of color, other studies suggest that having a regular doctor is a necessary but not sufficient criterion for ensuring access to preventive care, because some physicians do not provide preventive screenings during office visits. In one study, “physicians reported offering counseling about physical activity during 19.1 percent of the office visits, diet during 22.8 percent, and weight reduction during 10.4 percent.”32^(p741) A second study reported that fewer than half of the patients surveyed received screening for hypertension and diabetes.18 A third study reported that only 18% of office visits focused on dietary counseling and that internists were consistently more likely than general practitioners and family physicians to check blood pressure.33

These studies show that we need to do more than just ensure that women of color have access to a regular doctor. There are a multitude of things that can be done in the areas of policy, practice, and research to meet the preventive health needs of women of color.

Policy

• Target physicians through graduate medical education programs showing the importance of health prevention activities during office visits as a way to reach the Healthy People 201034 goals.

• Address differences in coverage for preventive exams by insurance status (private and public insurance).

• Evaluate the time spent during office visits for preventive care relative to other activities to assess costs and benefits.

• Decrease the number of women of color who are uninsured.

• Examine the rate of underinsurance for preventive health care.

• Infuse funding to increase the supply of health care providers in inner-city and rural communities through mechanisms such as section 330 of the US Public Health Act (community health centers), and assure funding to the Health Research and Services Administration to increase the number of providers in medically underserved areas.

• Increase funding for community-based health promotion projects and community involvement in local health decisionmaking activities.

• Provide funding for dietary outreach efforts in communities.

Practice

• Develop and implement cultural competency training for health care providers.

• Work with professional associations to set and implement standards for routine preventive screening activities.

• Develop curriculums for medical and allied health education that are gender sensitive and that factor psychosocial issues into the treatment process.

Research

• Evaluate the effectiveness of information dissemination activities relating to Healthy People 2010 objectives for preventive health.

• Evaluate the effectiveness of knowledge, attitude, and behavior interventions as they relate to Healthy People 2010 objectives for preventive health.

• Advocate the funding of large-scale, quantitative, localized studies of ethnic populations and of rural populations to examine within-group differences in preventive health care practices.

• Advocate the funding of qualitative evaluations to complement quantitative studies examining the context of patient–provider interactions and health decisionmaking activities.

Like all studies, this study has limitations that constrain our ability to generalize its findings. First, the data obtained from the CMHS are based on self-report and may under- or overreport health behaviors. This is important from a cultural point of view, since people of color tend to vary in their reporting of the severity of their health problems. Second, the small sample size limited our ability to examine certain subpopulations or to detect statistically significant findings for some of the racial and ethnic groups studied.