Smartphone-Assisted Single-Session Psychotherapy Intervention for Postpartum Depression (Motherly Plus)

Background. Postpartum depression (PPD) affects about 17% of women worldwide and carries risks of chronicity, suicidality and impaired child development. Psychotherapy is effective, but access is limited in low- and middle-income countries by workforce shortages and unequal distribution of services. Tele-mental health, smartphone apps and single-session psychotherapy are scalable alternatives. Evidence for single-session psychotherapy in PPD is limited to two trials of multi-hour group workshops, and no trial has combined an individual single session with digital support.

Design. Parallel-group, two-arm randomized controlled trial with masked outcome assessment, conducted remotely in Brazil. A total of 104 women aged 18-40 years, up to 12 months postpartum, with a DSM-5-TR diagnosis of major depressive disorder will be randomized to (1) smartphone app-supported single-session psychotherapy or (2) waitlist control. Assessments occur at T1 (week 1, baseline), T2 (week 2, mid-treatment), T3 (week 4, post-treatment) and T4 (week 8, follow-up). A pilot study to test procedures was completed before the start of recruitment.

Recruitment and screening. Participants are recruited through social media and the project website. Stage 1 is an online eligibility form. Stage 2 is a telephone or video interview to confirm criteria and a diagnostic assessment with the Mini-International Neuropsychiatric Interview (MINI). Eligible women sign an electronic informed consent form before randomization.

Intervention. One online single-session psychotherapy of about 90 minutes, based primarily on behavioral activation (BA). Components: pre-session questionnaire; rapport and assessment of daily routine, reinforcers and support networks; psychoeducation on the behavioral model of depression; collaborative scheduling of achievable, value-aligned activities and problem-solving of barriers and avoidance; training in the Motherly app; session closing and commitment. Optional components (stress management, emotion regulation, cognitive restructuring, sleep hygiene) are used according to clinical judgment. Participants then use the Motherly app independently for 4 weeks. The app includes modules on behavioral activation, emotion regulation, stress management, cognitive restructuring, sleep hygiene and relaxation, and mindfulness, plus activity scheduling, symptom tracking and an educational library. Therapists are psychologists who receive 10 hours of training and are supervised by senior researchers.

Control. Waitlist with no intervention for 4 weeks after randomization, after which participants receive the same intervention. Women with current suicidal ideation with intent or plan, a recent suicide attempt, or treatment-resistant depression are excluded because a waitlist is not appropriate for them.

Randomization and masking. The randomization list is generated by an independent statistician in R and allocation is automated through the REDCap randomization module, ensuring concealment. Outcome assessors are masked and have no role in recruitment, allocation notification, intervention delivery or fidelity assessment. Measures that could reveal allocation are collected online after the clinician-rated assessment. Access to allocation data in REDCap is restricted by role. At post-treatment, assessors guess each participant's allocation and rate their confidence, and masking success is reported with 95% confidence intervals. The main analysis is run with a pre-specified R script on a dataset with masked group labels.

Sample size. A meta-analysis of eight RCTs of single-session interventions for adult depression (Metapsy database, version 24.0.2) gave g = 0.60. With two-sided alpha of 5%, 80% power and 15% attrition, 104 participants are required (G*Power 3.1.9.4).

Statistical analysis. Primary and secondary outcomes will be analyzed with generalized estimating equations (GEE) with an AR(1) working correlation structure and robust sandwich variance estimator, in the intention-to-treat sample. Each model includes the outcome as dependent variable and group, time (categorical) and the group-by-time interaction as predictors. Marginal means will be compared with pairwise comparisons. Missing data will be handled with multiple imputation by chained equations under a missing-at-random assumption. Standardized effect sizes (Cohen's d) will be computed from marginal means and observed standard deviations. Sensitivity analyses will compare results with completers only, and baseline predictors of dropout will be examined. Significance level is p < 0.05 with 95% confidence intervals.

Moderators (exploratory). Interaction models will test maternal age, skin color/ethnicity, education, family income, number of previous children, relationship with the father, perceived social support (MSPSS), baby's age, time spent with the child, number of pediatric and prenatal consultations, baseline depression (QIDS-SR16, QIDS-C16), baseline severity (CGI-S) and functioning (SDS). The trial is not powered for these analyses.

Mediators (exploratory). Behavioral activation and avoidance (modified BADS-SF, two-week recall) and environmental reward (Reward Probability Index) are assessed at T1 and T2. Parallel mediation models will test whether they mediate the effect on depressive symptoms at T3, adjusting for baseline levels of symptoms and mediators. Depressive symptoms at T2 will also be considered to account for early change.

Treatment-specific processes (intervention arm). Therapeutic alliance (WAI-SR, client and therapist versions, at T2) and seven app engagement indicators will be related to clinical outcomes with regression models.

Fidelity. An independent rater will assess adherence from audio recordings of the sessions, using a checklist adapted from a previous behavioral activation study.

Safety. Adverse events are recorded at post-treatment and follow-up. Clinical worsening is defined as an increase of 30% or more in the QIDS-16 total score (clinician-rated or self-report) relative to baseline. Serious adverse events include death, life-threatening events, hospitalization, significant disability and suicide risk (score of 2 or more on QIDS-16 item 12). REDCap sends automatic alerts to the investigators. Participants with emerging suicidal ideation receive a written safety plan and are referred for psychiatric evaluation; domestic violence is addressed with referrals to specialized services. Participants without remission at the end of the study are referred to specialized care.

Ethics. Approved by CAPPesq, HC-FMUSP (CAAE 97532226.8.0000.0068; approval 8.574.338). Conducted in accordance with the Declaration of Helsinki.

Registration note. The protocol was submitted to the Brazilian Clinical Trials Registry (ReBEC; UTN U1111-1345-9666) on [PREENCHER data], before the start of recruitment, but was not published by that registry. This ClinicalTrials.gov record was created after the first participants were enrolled. No changes to outcomes or analyses were made.