From Feasibility to Outcomes: the Case of a Postpartum Depression Prevention Intervention in Shelter

Abstract Background Evidence-based interventions (EBIs) are often seen as the gold standard but it is unclear if they are effective in different settings or populations. We developed a model to adapt and test the implementation of a postpartum depression prevention intervention (Reach Out and Stay Strong, Essentials for new Mothers -- ROSE) in shelter. Our model entailed: (1) Intervention Adaptation and Planning (2) Feasibility Testing, (3) Monitoring Outcomes. The adapted intervention was called Strong in Shelter (SIS). Methods Intervention adaptation and planning entailed provider engagement, shelter resident input, and responsive adaptations. Feasibility testing began in pilot sites, and we made modifications based on feedback. Fidelity to the model was tracked through session checklists, audio recordings and weekly consultations. Additional modifications were made due to COVID-19. Outcome measures were collected pre- and post-intervention, and included measures of depression (PHQ-9), perceived social support (MSPSS) and perceived stress (PSS). We also collected data on trauma exposure. Paired Samples T-Test analyses were conducted to examine changes in PHQ-9, MSPSS and PSS scores over time. These repeated measures were stratified by interpersonal trauma, mode, and number of sessions completed using general linear model analyses. Results Involving the community being served and the intervention facilitators early on was an important factor in the success of SIS. Barriers to participant engagement were seen across each phase of SIS, but once residents did engage 70% (40 of 57 women who were consented) attended at least half the sessions. Satisfaction was high, with nearly 100% of women participating in SIS and receiving a consistent questionnaire reporting the group was helpful (N=22). The PHQ-9 decreased by 2.68 points after SIS (p=0.002); this change was stronger for those who experienced interpersonal trauma, and findings were consistent in the virtual implementation during COVID-19 (N=46). Conclusions Providing evidence-based preventive interventions to underserved populations is essential to ensure that effective programs reach those in need. Our data suggest that by engaging staff and the population early and continually, and utilizing an iterative and flexible process, EBIs such as SIS can effectively be implemented in institutional settings, and may reduce depressive symptoms among women in shelter.