Correlates of drug abuse among homeless and transient people in the Washington, DC, metropolitan area in 1991.

Sociodemographic correlates of drug use among the general household population may have less accuracy and utility for describing risk factors for drug use among non household populations like the homeless and transient. This analysis examines correlates of past year use of marijuana, cocaine, and needles among homeless and transient people in the Washington, DC, metropolitan statistical area (DC MSA) and discusses them vis-a-vis traditional indicators of drug use among the general household population. Data are from a study conducted in the DC MSA in 1991 that used a multistage sampling design and surveyed a random sample of 908 homeless and transient people ages 12 years and older. The analysis uses multiple logistic regression to assess the independent effect of demographic and other predictors on selected drug use measures among this population. Three key socioeconomic correlates of drug use among the general household population (educational attainment, employment status, and marital status) were nonsignificant predictors of drug use among the homeless. However, other factors were significant, including past year institutionalization, location within the DC MSA, and stage of homelessness. The age group at greatest risk for use of marijuana and cocaine in the past year were the homeless ages 26 to 34, but the oldest group (35 years and older) had the highest risk of needle use. Although men were more likely to have used marijuana and cocaine in the past year, there were no sex differences in the use of needles in the past year. Only past year use of cocaine differed significantly by race or ethnicity, with a greater likelihood among homeless blacks than among homeless whites. Measures of social achievement and socioeconomic status related to the prevalence and risks of drug use among the general household population have but limited applicability in predicting drug use among people who are homeless and transient. Along with other indicators of behavioral and health risks,history of institutionalization, urbanization, and chronicity of homelessness should be considered to improve the epidemiologic assessment of this population.