Gender Differences in a Social Stress Model Predicting Relapse to Substance Use

Background and Purpose:  Previous research has examined the relationship between stress and relapse, repeatedly finding that stress is a strong predictor of relapse following abstinence from drug/alcohol use.  Pearlin's social stress theory indicates that factors affecting placement in the social environment, such as gender, race, and socio-economic status (SES), often affect the experience of stress. There are gender differences in the experience and effects of stress, and women have been found to experience higher levels of both physiological and psychological effects of stress than men and are more vulnerable to a number of life stressors, such as poverty. This study explored whether there were gender differences in the functioning of a social stress theoretical model predicting relapse.

Methods:  An empirical model predicting relapse was created using social stress theory as a heuristic, including social stress factors (race, income, education, employment, relationship status), child custody, perceived stress, and economic hardship, for predicting relapse at one-year follow-up in a statewide sample of Kentucky substance abuse treatment participants (N = 1123).  Path analysis with logistic regression was used to test the model’s ability to predict relapse for men (n = 599) and women (n = 524) separately, examining potential mediators such as efficacy, recovery support, and self-control. 

Results:   The model significantly predicted relapse for both men (χ² = 145.47, p ≤.001, Nagelkerke R² = .31) and women (χ² = 124.44, p ≤.001, Nagelkerke R² = .28).  For men, race and custody of children were significant individual predictors, with Black men more likely to relapse than White men, and men with dependent children less likely to relapse.  For women, higher education levels significantly predicted relapse.  Perceived stress was significantly positively associated with relapse for both men and women, but this relationship was mediated by recovery support and efficacy for men, and by efficacy and self-control for women.     

Conclusions and Implications:  This study filled a gap in the literature by examining gender differences in social stress factors predictive of relapse as well as in mediators between stress and relapse.  The model holds promise for predicting relapse, although alterations in the model should be made for future research and the model should be tested with other samples.  Notably, many of the social stress factors were not significant for women, and education functioned contrary to the social stress theory hypothesis.  Similar to previous research, this study found that perceived stress predicted relapse for both men and women.  However, mediators varied by gender; results suggest that men might benefit from interventions targeting increasing recovery support and efficacy, in particular, while women might benefit from increasing efficacy and self-control.  Although future research needs to further examine gender differences in social stress factors’ abilities to predict relapse, particularly comparing differences by addiction severity or type of substance used, the findings from this study appear to indicate that there may be gender differences in how social stress theory can be used to understand the relapse process.