Typologies of Women's Substance Use Recovery Networks and 12-Month Sobriety Outcomes

Background and Purpose:

Women’s personal social networks (PSN) often contain members (alters) who simultaneously support and endanger their recovery from substance use disorders. Trauma affects the ability to use PSNs to support recovery. This study uses latent profile analysis (LPA) to identify typologies of PSN characteristics that support sobriety over the first year of recovery following treatment entry, which are regressed on sobriety outcomes. It is theorized that having a recovery network that contains recovery-supportive alters and has isolates (unconnected alters) that represent ties to recovery resources would be associated with increased ability to maintain sobriety. Network density, a structural variable, may also play a key role in reinforcing recovery. Finally, trauma is examined as a potential explanation for differences in women’s recovery networks.

Methods:

LPA was conducted with 6 recovery-specific PSN indicators (sobriety, history of use, sobriety support, treatment-related alters, isolates, and density). Participants included women with substance use disorders at one week after treatment entry (N = 377). Typology membership was regressed on sobriety over the first 6 and 12 months after treatment entry, controlling for demographic (age, race, income source, education) and clinical characteristics (treatment type, traumatic symptomatology). T-tests compared mean Trauma Symptom Checklist-40 (TSC-40) scores.

Results:

Sample characteristics included mean age of 36.5 years; 59.9% Black; 69.2% on public assistance; 58.9% with at least a high school degree; 31.6% in residential treatment. The mean TSC-40 score was 44.70, significantly higher than means reported in most previous studies. 56.0% and 47.2% had maintained sobriety over 6 months and 12 months, respectively. Three social network types were identified: Women in Class 1 (14.32%) had highly dense networks with few isolates and few treatment-related alters. Women in Class 2 (49.34%) had low density networks, and high numbers of isolates and treatment alters. Women in Class 3 (36.34%) also had networks with low density and high numbers of isolates, but had less sobriety support, fewer sober or treatment-related alters, and higher numbers of alters with whom they had used. Women in Class 3 were significantly less likely to maintain sobriety over both 6 months (B=-0.75, Wald χ² =5.48, p < .05; OR=0.47, 95% CI [0.25, 0.89]) and 12 months (B=-0.72, Wald χ² =6.36, p < .01; OR=0.49, 95% CI [0.28, 0.85]) following treatment entry than women in Class 2, after controlling for demographic and clinical characteristics. TSC-40 scores for Class 3 (m=48.58, SD=21.55) were significantly higher than for Classes 1 (m=38.98, SD=21.87; t=2.76, p<.01) and 2 (m=43.48, SD=20.72; t=2.14, p<.05), with no significant difference in between Classes 1 and 2.

Conclusions and Implications:

Network isolates may represent key structural elements for women in early recovery. Treatment-related isolates may link to new resources, while isolates who endanger recovery may represent the remnants of the woman’s using network. The connections that a woman makes in early recovery may give her a way to build a stronger, more recovery-supportive social network. However, the ability to forge these new connections may be affected by the level of trauma the woman has experienced.