Socioeconomic marginality and health services utilization among Central Harlem substance users.

The article examines whether decrements in socioeconomic measures in a poor, substance using population predict changes in health services utilization. The sample consisted of 658 "hard drug" (crack, powder cocaine, and heroin) users drawn from Central Harlem in New York City during 1998 and 1999. Chain referral and social networking were used in order to gain access to hidden users. The sample was stratified according to operational measures indicating socioeconomic marginality, one calculated using indices of income, education, and employment and another designed to measure lived homelessness. Rates of self-reported utilization of 10 health services were compared across strata. In this sample socioeconomic marginality reflected by low levels of income, education, and employment sometimes predicts greater rates of health services utilization and, in other cases, it predicts lower rates. When the sample is stratified according to an operational measure of homelessness, the gradient of greater utilization and self-reported morbidity for the homeless is more marked and consistent. Results are supportive of a public health model of drug user treatment that recommends that it occur as part of an integrated strategy addressing poverty, homelessness, violence, and related social problems.