Prediction of treatment response at 5-year follow-up in a randomized clinical trial of behaviorally based couple therapies.

Objective: Building on earlier work examining predictors of short- and moderate-term treatment response, demographic, intrapersonal, communication, and interpersonal variables were examined as predictors of clinically significant outcomes 5 years after couples completed 1 of 2 behaviorally based couple therapies.

Method: One hundred and thirty-four couples were randomly assigned to Integrative Behavioral Couple Therapy (IBCT; Jacobson & Christensen, 1998) or Traditional Behavioral Couple Therapy (TBCT; Jacobson & Margolin, 1979) and followed for 5 years after treatment. Outcomes include clinically significant change categories of relationship satisfaction and marital status at 5-year follow-up. Optimal subsets of predictors were selected using an automated, bootstrapped selection procedure based on Bayesian information criterion.

Results: Higher levels of commitment and being married for a longer period of time were associated with decreased likelihood of divorce or separation (odds ratio [OR] = 1.39, p = .004; OR = 0.91, p = .015). Being married for a longer period of time was also associated with increased likelihood of positive, clinically significant change (OR = 1.12, p = .029). Finally, higher levels of wife-desired closeness were associated with increased odds of positive, clinically significant change and decreased odds of divorce for moderately distressed, IBCT couples (OR = 1.16, p = .002; OR = 0.85, p = .007, respectively), whereas the opposite was true for moderately distressed, TBCT couples (OR = 0.77, p < .001; OR = 1.17, p = .002, respectively).

Conclusions: Commitment-related variables are associated with clinically significant outcomes at 5-year follow-up as well as at termination and moderate-term follow-up.