Heterogeneity in the Effects of Interventions to Prevent Depression in Couples Facing Job Loss: Studying Baseline Target Moderation of Impact

Introduction: Couples’ communication styles are associated with onset of depression following a major stressor such as job loss. Combined with consistent findings that job loss by itself increases risk for depression, we hypothesized that preventive interventions targeting job seekers would be enhanced by adding intervention components designed to improve couple communication. We developed the Couples Employment Program (CEP) by integrating components from the JOBS program (for individual job seekers) with modules from a couple’s communication skills program. Within a BTM framework we hypothesized compensatory effects, such that those job seekers who began the program with fewer job search skills and lower motivation would benefit more, as would couples who began the program with less positive communication.

Method: We conducted a trial of CEP with 1477 heterosexual couples where one member had recently lost a job, randomizing couples to either CEP or a control condition, collecting baseline data on depressive symptoms, job search behavior, job search motivation, and couple’s communication. Those couples assigned to the intervention participated in couples groups that focused on communication, and both partners were invited to participate in five group sessions on job search skills and ways to maintain motivation. Outcomes were assessed after the intervention and 6 and 12 months later.

Results: Baseline measures of job seeker behavior and motivation were strongly correlated, as were measures of couples communication. We created latent variables to assess baseline variation in each, estimated growth models on depressive symptoms, and tested baseline target moderation (BTM) models that specified interactions between treatment condition and baseline target levels. We found that baseline levels of job seeker behavior and motivation did moderate the impact of intervention on job seeker depression slopes, but not partner slopes. On the other hand, baseline levels of couples communication moderated the impact of intervention on partner slopes, but not job seeker slopes. However, rather than conforming to a compensatory effect, in both cases we found evidence of a crossover effect. The intervention reduced risk for depression in job seekers having lower baseline levels of motivation, but increased risk for depression in those with higher baseline levels. The intervention also appeared to reduce risk for partners in couples with better communication, but increased risk for depression in those who came into the intervention with poorer communication. These crossover effects also appeared to wash out any main effects of the intervention, given that some subgroups benefited while others were negatively impacted.

Conclusion: These results may help to illuminate findings from other recent studies of couples preventive interventions, which have shown inconsistent or even negative effects. They also point to the importance of using baseline target information to evaluate differential impact of prevention programs and illustrate that this may operate in different ways for different baseline targets.