Behavioral Activation for Perinatal Suicide Ideation

Importance Suicide is a leading cause of maternal postpartum death. Evidence-based interventions are needed. Objective To determine whether the likelihood of endorsing suicide ideation (SI) changed during a brief behavioral activation (BA) psychotherapy for perinatal depression irrespective of clinician or delivery types. Design, Setting, and Participants This is a secondary analysis of a multisite, noninferiority, 4-arm randomized clinical trial called SUMMIT(Scaling Up Maternal Mental Health Care by Increasing Access to Treatment). SUMMIT compared clinicians (nonspecialist vs specialist) and modalities (telemedicine vs in person) in delivering BA. Participants were enrolled from January 2020 to October 2023. The study was conducted at university-affiliated networks in Chicago, Illinois; Chapel Hill, North Carolina; and Toronto, Canada. Pregnant (≤36 weeks) and postpartum (4-30 weeks) adults with depressive symptoms (Edinburgh Postnatal Depression Scale [EPDS] score ≥10) were enrolled. Secondary data analyses were conducted in November 2024. Intervention A manualized 6- to 8-session perinatal BA intervention delivered weekly. Main Outcomes and Measures The primary outcome was SI, as measured by the EPDS item 10, assessed weekly and at 3 months postrandomization. SI endorsements were followed by the Columbia Suicide Severity Rating Scale (C-SSRS) for further safety assessment. Results A total of 1230 pregnant and postpartum adults were enrolled, among whom 1117 completed 1 or more treatment session and provided 1 or more week of EPDS data and thus were included in the current study. Of 1117 included participants, 264 (23.6%) endorsed SI during treatment, and mean (SD) age was 33.4 (4.9) years. Cumulative link mixed models indicated that the odds of endorsing SI decreased by 25% with each additional treatment session (odds ratio [OR], 0.75; 95% CI, 0.58-0.96; P = .03). The odds of endorsing SI decreased by 80% at 3 months postrandomization compared with any time during treatment (OR, 0.20; 95% CI, 0.14-0.27; P < .001). The odds of endorsing SI did not differ between clinician type (nonspecialist vs specialist) or modality (telemedicine vs in person). Goodness-of-fit χ 2 tests further indicated that participants were significantly more likely to endorse suicide thoughts on the C-SSRS at treatment onset compared with any other time point. Conclusions and Relevance In this secondary analysis of the SUMMIT noninferiority randomized clinical trial, the likelihood of endorsing SI decreased over the course of a brief BA psychotherapy for perinatal depression and most significantly at 3 months postrandomization, irrespective of clinician or delivery types. BA may be one evidence-based, scalable approach for perinatal suicide prevention. Trial Registration ClinicalTrials.gov Identifier: NCT04153864