Effectiveness of Long‐Term Residential Substance Abuse Treatment for Women: Findings from Three National Studies

Abstract

The effectiveness of residential substance abuse treatment for women was examined using data from the Center for Substance Abuse Treatment's Residential Women and Children/Pregnant and Postpartum Women (RWC/PPW) Cross‐Site Study and two other recent national studies. Treatment success was defined as posttreatment abstinence from further drug or alcohol use, measured through in‐person follow‐up interviews conducted 6–12 months after each client's discharge. Despite differences in treatment programs, client profiles, follow‐up intervals, data collection methods, and other factors, all three studies found high treatment success rates—ranging narrowly from 68% to 71% abstinent—among women who spent six months or more in treatment. Success rates were lower, and between‐study differences were larger, for clients with shorter stays in treatment. Controlling for salient client and treatment project characteristics, strong associations between length of stay in treatment and posttreatment abstinence rate were found in all three studies, suggesting that women's length of stay in residential treatment is a major determinant of treatment effectiveness. In further analysis of RWC/PPW data, treatment completion was also found to be an important outcome factor. Among clients who remained in treatment for at least three months, those who achieved their treatment goals in three to five months abstinence outcomes were as good as those for clients who took more than six months to complete their treatment (76%–78% abstinent) and substantially better than those for clients who did not complete treatment (51%–52% abstinent). Notably, however, most of the RWC/PPW clients who successfully completed treatment (71%) required six months or more to do so.

• Residential substance abuse treatment
• Posttreatment abstinence

Notes

^aThese adjustments, while statistically desirable, actually had little effect. A second, unweighted series of analyses produced results nearly identical to those reported here.