Sociodemographic correlates of major depression, crack use, and dual diagnosis.

Despite the seemingly extended history of psychiatric illness and drug use research, relatively little is known about the relationship among sociodemographics, major depression, crack use, and dual diagnosis. Using data from the 1997 National Household Survey on Drug Abuse, regression analysis was conducted on 11,766 black and white adults. This investigation explored stress and coping and socioeconomic inequity theories as they relate to depression and crack use. Self-medication theory was proposed to explain variances in the prevalence of dual diagnosis among respondents. Race and gender were central issues to this investigation. In Chapters I and II, the theoretical framing of the study, overview of research goals, and study methodology are presented. Figures illustrating the conceptual relationship between background variables, personal and financial resources, and health outcomes are also depicted. Chapters III, IV, and V were devoted to each of the three health outcomes (i.e., depression, crack use, and dual diagnosis). In Chapter III, analysis indicated a significant race-by-gender interaction, such that white women were more likely than white men to be depressed. Gender differences were not significant among blacks. In Chapter IV, a significant race-by-gender-by age interaction emerged. For black men and women, as age increased, the risk of crack use also increased. Blacks aged 26 years and older exhibited higher rates of past-year and lifetime crack use than whites aged 26 years and older. Blacks aged 35 years and older were more likely than whites aged 35 years and older, to have used crack within their lifetime. The final empirical investigation, Chapter V, revealed men to be more likely to be dually diagnosed than women, both within the past-year and within their lifetime. This study extended previous research on depression, crack use, and dual diagnosis, by examining the intersecting roles of race, gender, and other sociodemographic variables. Regardless of the health outcome (i.e., depression or crack use) race and gender were consistently predictive of the occurrence of depression, crack use, and dual diagnosis. It is suggested that sociodemographic variables be included in future analysis concerning psychiatric disorders and drug use, as well as theory-driven epidemiologic research.Ph.D.Behavioral psychologyClinical psychologyHealth and Environmental SciencesPhysiological psychologyPsychologyPublic healthUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/124890/2/3016811.pd