The purpose of this study is to evaluate the effects of income, insurance, race, ethnicity, and family situation on access to ambulatory medical care in order to expand the base of information available to policy-makers attempting to monitor and improve access for the indigent, using data from the 1987 National Medical Expenditure Survey (NMES). The specific objectives are, in persons under 65: (1) to evaluate how access for Medicaid recipients and uninsured persons differs from that of privately insured persons and explore whether Medicaid recipients are more likely to be denied care than the even the indigent uninsured; (2) to study the effects of income on access to care for the uninsured and for those with private insurance; (3) to determine whether poorer, older persons (but still under 65) have greater access problems than others, even if no overall income effect is found; (4) to assess what independent effects race and ethnicity have on access, after controlling for income, insurance, age, sex, education and health status; and (5) to evaluate whether uninsured adults or children in single parent households have worse access than other uninsured persons. RATIONALE: There have been few studies of access based on national samples which have properly controlled for insurance, income, race/ethnicity, age, family situation and other factors, and very little work has been done analyzing interaction effects to identify particularly disadvantaged subgroups. Analysis of access for indigent and other disadvantaged persons, separating the effects of these factors, is important for guiding future alterations in the health insurance and health care delivery systems. The 1987 NMES was a national, randomized survey of approximately 13,000 households with about 30,000 individuals. It contains extensive information about income, insurance coverage, medical care utilization, health status and access to care. The survey design oversampled blacks, Hispanics and the poor. A number of access measures available from this survey, including barriers to care, diagnostic and therapeutic services obtained, and preventive care received, will be used as dependent variables in multiple logistic and linear regression analyses to study the effects of income, insurance status, race, ethnicity and single parenthood, while controlling for age, sex, education, residence, and health status. First-order interactions will be evaluated to determine whether specific subgroups have greater access problems. The study will provide information about the independent effects of the studied factors on access to care, particularly for the indigent, helping policy-makers to target changes in both our health insurance system and the delivery system. It will assess whether, nationally, Medicaid recipients are even more likely to be denied care than the indigent uninsured.
