The Relationship between Substance Abuse Treatment Duration and Discharge Outcomes: The Moderating Role of Homelessness

Background the Purpose:

Treatment outcomes vary widely among individuals with substance use disorders, with a rate of treatment completion estimated to be less than half of individuals who enter treatment. While studies have focused on individual and interpersonal determinants of treatment completion, few have investigated the role of treatment duration, and whether the relationship varies as a function of homelessness. The purpose of this study is to examine the relationship between treatment duration and discharge outcomes as well as the moderating effect of homelessness among a national sample of discharges from long-term residential and outpatient substance abuse treatment settings. Based on theory and prior literature, we hypothesized that individuals with longer treatment duration are more likely to have completed treatment compared to other discharge outcomes, and the magnitude of the relationship differs significantly between homeless and housed individuals.

Methods:

A sample of 8,844,354 discharges from long term residential and outpatient treatment settings was drawn from the 2006-2013 Treatment Episode Data Set Discharges (TEDS-D), a public use database of discharges from substance abuse treatment settings across the United States. First, multinomial logistic regression analyses were conducted to examine the association between treatment duration, defined by number of days between the date of admission and last contact, and six different discharge outcomes: treatment completed, left against professional advice, terminated, transferred, incarcerated, and death. The moderating effect of homelessness was tested by adding the interaction between treatment duration and homelessness. Predictive margins were computed and compared to interpret the significance of main effects and interaction effects. All models adjusted for predisposing, need, and enabling factors.

Results:

Among the sample of discharges from long-term residential and outpatient treatment settings, treatment completion rate is 40%. The largest age group is 25-34 years (31%), with 67% male, and 58% identifying as white. Eight percent of discharges occurred among homeless individuals, and 30% were among individuals with co-occurring mental health disorders. The main effects model showed that individuals with longer treatment duration are significantly less likely to have any of the five non-completed discharge outcomes compared to treatment completion in both long term residential and outpatient treatment settings. In outpatient treatment settings, homeless individuals with longer outpatient treatment duration are significantly more likely to be transferred and incarcerated, and are significantly less likely to complete treatment and leave against professional advice compared to non-homeless individuals. In long-term residential treatment settings, homeless individuals are significantly more likely to leave against professional advice, and are significantly less likely to complete treatment, be transferred, and be incarcerated compared to non-homeless individuals.

Conclusions and Implications:

Longer treatment duration is significantly associated with a less likelihood of adverse discharge outcomes, and the relationship is significantly moderated by homeless status. These findings highlight the importance of understanding treatment retention in predicting positive discharge outcomes, especially among individuals with unstable housing situations. The findings support and reinforce the continued efforts in improving housing stability among homeless individuals with substance abuse problems.