Individual characteristics of the literally homeless, marginally housed, and impoverished in a US substance abuse treatment-seeking sample

Background Many researchers and clinicians believe that understanding substance use problems is key to understanding homelessness. This study’s purpose was to test, in a national sample of urban substance abuse treatment seekers, whether (1) income was related to amount of money spent on substances and (2) homeless chronic substance users had more severe psychosocial problems or histories than housed chronic substance users. Method Questions assessing homelessness were inserted into the Drug Evaluation Network System—a computerassisted intake interview (including the Addiction Severity Index) implemented in addiction treatment programs across the U.S. Based on these data, clients were divided into four residential groups: literally homeless (n = 654), marginally housed (n = 1138), housed poor (n = 3119), and housed not poor (n = 718). Income, human capital (education level and acquisition of a trade/skill), substance use, mental health, and social support were examined. Results The literally homeless was not the poorest group, although these clients did spend the most money on substances. All four groups’ incomes were positively related to amount of money spent on drugs, but only the marginally housed’s income was related to money spent on alcohol. The literally homeless had the most severe alcohol, mental health, and social support problems. The literally homeless and marginally housed had similar incomes and human capital and the most severe cocaine problems. In general the housed poor and housed not poor fared better than the literally homeless and marginally housed groups. Discussion Practitioners should continue to intervene with the homeless and consider working with the marginally housed’s social support systems. Future research should examine the marginally housed as an at-risk group for homelessness.