 | The 131st Annual Meeting (November 15-19, 2003) of APHA |
|
| The 131st Annual Meeting (November 15-19, 2003) of APHA |
Joseph J. Sudano, PhD, School of Medicine, Case Western Reserve University, MetroHealth Medical Center, Rammelkamp 236, 2500 MetroHealth Drive, Cleveland, OH 44109, 216-778-1399, jsudano@metrohealth.org and David W. Baker, MD, MPH, The Feinberg School of Medicine, Northwestern University, 676 N. St. Clair Street, Room 255, Chicago, IL 60611.
Objective: To determine the contributions of socioeconomic status (SES), health behaviors, and health insurance in explaining racial/ethnic disparities in health among adults in late middle age.
Design: We used data from the 1992 and 1998 Health and Retirement Study (HRS), a nationally representative sample of U.S. adults age 51-61 in 1992. We defined 2 outcomes: (1) major decline in self-reported overall health; and (2) death. A combined outcome of major decline/death was used in logistic regression to determine (1) the degree to which SES, health behaviors, and insurance independently explained racial/ethnic disparities, and (2) how much of the SES effect was mediated by adverse health behaviors, and lack of insurance.
Population: 6286 non-Hispanic whites (W), 1391 non-Hispanic Blacks (B), 405 English-speaking Hispanics (E/H), and 318 Spanish-speaking Hispanics (S/H).
Findings: Major decline/death occurred in 29.0 percent of B, 25.3 percent of E/H, and 36.4 percent of S/H compared to 18.9 percent of W. After adjusting for demographics and baseline health, adjusted relative risks for major decline/death were 1.24 for B (95 percent CI 1.09-1.40), 1.23 for E/H (1.01-1.49), and 1.68 for S/H (1.36-2.04) compared to W. After adding SES to the model, there were no longer any significant differences in risk of major decline/death, suggesting that differences between racial/ethnic groups were fully explained by differences in SES. In contrast, adding health behaviors to the baseline model did not change the RR for B and actually slightly increased the RR for E/H (1.27) and S/H (1.77). Adding insurance to the baseline model reduced RR for B to 1.21, 1.18 for E/H, 1.50 for S/H. Adjusting for all independent variables, adjusted RR's were 1.12 for B (0.98-1.27), 1.12 for E/H (0.90-1.37), and 1.28 for S/H (1.00-1.61). Health behaviors moderated 21 percent and insurance status 11 percent of the effect of SES.
Conclusions: Most of the higher rate of morbidity and mortality among blacks and Hispanics results from lower SES. The effect of SES on health was mostly direct, with only 30 percent of the SES effect explained by higher rates of adverse health behaviors and lower rates of insurance coverage among those with lower SES.
Implications for Policy, Delivery, or Practice: Public health initiatives that promote changing individual health behaviors and increasing rates of insurance coverage will not eliminate racial/ethnic health disparities. To eliminate disparities, we must increase efforts to understand the social-structural and institutional mechanisms mediating the relationship between SES and health.
Learning Objectives:
Keywords: Health Insurance, Health Disparities
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.