340P
Longitudinal Associations Among Sobriety Support, Substance Using Members, and Substance Use at 6 and 12 Months Post Treatment
Substance using network members are associated with decreased support for abstinence, poor treatment outcomes, and relapse. Sobriety support may play a role in reducing the number of substance users. Little is known about the reciprocal relationships between sobriety support, substance using network members, and substance use post treatment over time. This study examined the bidirectional relationship among sobriety support, substance using network members, and substance use at 6 and 12 months.
Methods:
Data were collected from 258 women at two intensive outpatient treatment program interviewed one week post treatment intake (T1) and at 1 (T2), 6 (T3), and 12 months (T4) (84% retention). Personal network characteristics were measured by the number of non-substance users providing sobriety support and the number of drug/alcohol users. Substance use at T3 and T4 was measured by any substance use in the past 30 days. Trauma symptoms, abstinence self-efficacy, and current treatment status were included as covariates.
Autoregressive cross-lagged model was employed, using Mplus 6.12, to examine longitudinal associations between network sobriety support, substance using network members, and substance use at T3 and T4. Maximum likelihood estimation was used to handle missing values. The comparative fit index (CFI), Tucker Lewis Index (TLI), and the root mean square error of approximation (RMSEA) were used to test the fit of the autoregressive cross-lagged model.
Results:
The majority were African American (62.6%) with a mean age of 36.4 (SD=10.5), had co-occurring mental and substance use disorders (72.7%); 38.1% of participants used alcohol and/or drugs by the 12 month post treatment follow up period; 72.3% of women had previous treatment history. Mean years from the first substance use treatment was 6.21 (SD=7.5) years.
The autoregressive, cross-lagged model was tested and showed an acceptable fit with data, x2(66) = .033, RMSEA=041, CFI=.942, TLI=.917. Based on MacCallum, Browne, & Sugawara (1996)’s criteria, this model had sufficient power (0.83) given RMSEA, df, and n=258. Results showed that greater number of substance using members at T2 was associated with less network sobriety support at T3 (b=-.14, p<.01), which in turn was associated with increased odds of substance use at T4 (b=-.33, p<.01). Also, we found that greater network sobriety support at T2 was associated with fewer substance using members at T3 (b=-.28, p<.01). In addition, substance use at T3 was associated with decreased network sobriety support at T4 (b=-.20, p<.01). In terms of covariates, greater trauma symptoms were associated with more substance using members (b=.26, p<.01) and fewer people providing sobriety support (b=-.18, p<.01) at T1. Higher self-efficacy was associated with fewer substance users (b=-.18, p<.01) and greater network sobriety support at T1 (b=-.18, p<.01).
Conclusion and Implications:
Findings highlight a possible path by which substance users in women’s networks may negatively impact on recovery via undermining network sobriety support. We found a reciprocal relationship between network sobriety support and substance using networks over time. Findings of this study highlight the importance of building and maintaining recovery oriented networks.