Personal Networks and Substance Use At 12 Month Post Treatment Among Dually Diagnosed Women
Women with substance use disorders (SUD) tend to enter treatment with limited network resources and reduced recovery support. The purpose of this NIDA funded study was to examine personal networks as predictors of substance use at 12 months post treatment for women with SUD. Both network structure and composition were examined in relation to presence of substance use post treatment intake.
Methods:
Data were collected from 280 women who received treatment at three county-funded substance abuse treatment programs. At intake, trauma symptomatology (Trauma Symptom Checklist 40), co-occurring mental disorders, duration of SUD (Computerized Diagnostic Interview Schedule) and Abstinence Self Efficacy were assessed. At six month follow up, characteristics and structure of personal networks were examined (EgoNet software) and abstinent support was assessed (Social Support for Recovery Scale and Friends Support for Abstinence Scale). Substance use at 12 months post treatment was measured by whether or not women reported any use of alcohol or drugs over 12 months. Covariates correlated to the outcome at p<.10 were controlled in a multivariate logistic regression to examine the contribution of personal network characteristics on substance use at 12 months post treatment intake.
Results:
Participants were 60.7% African American, with a mean age of 36.7 (SD=10.39). 44.6% had not graduated from high school. More than half (60.3%) reported cocaine dependence, 46.2% reported alcohol dependence, and 57.4% were dependent on more than one drug. Mean of substance use duration of participants was 6.35 (SD=7.77) years. Most (73.9%) had co-occurring mental disorders. On average 28% of the network members were substance users. At 12 month follow-up, 44% women reported substance use. Logistic regression analyses indicated that a greater number of substance using network members (OR=1.10, 95% CI=1.04-1.15) and more disconnected subgroups (components)(OR=1.50, 95% CI=1.01-2.23) at 6 month post intake were related to increased likelihood of using substances at 12 months post intake, while abstinence support (OR=0.91, 95% CI=0.85-0.97) and number of people disconnected to the network (network isolates) at 6 month post intake were related to a lower likelihood of substance use at 12 months post intake (OR=0.94, 95% CI=0.91-0.98). Dual disorder (OR=0.95, 95% CI=0.91-0.98) was related to increased likelihood of using substances, and shorter duration of SUD (OR=0.95, 95% CI=0.91-0.98) and self-efficacy (OR=0.98, 95% CI=0.96-0.99) were related to a lower likelihood of substance use.
Conclusion and Implications:
Findings highlight personal network characteristics which both increase and decrease risk of substance use post treatment. In terms of network composition, substance using network members increased the likelihood of substance use. Network structure (how the network was arranged around the respondent) was also a factor. A more compartmentalized network, as indicated by the number of components, increased the likelihood of substance use; more isolates indicating less connected networks decreased likelihood of substance use. This study indicates that in addition to network composition and abstinence support, how networks are arranged may be relevant to interventions to promote sustained sobriety post treatment.