C-COMPASS: protocol for a quasi-experimental hybrid type I effectiveness-implementation study of community-based compassionate care after stillbirth in India.

Background: India contributes approximately one-quarter of the global stillbirth burden, with an estimated 460,000 stillbirths annually-one of the highest absolute burdens globally. Despite this, women experiencing stillbirth or early neonatal death face a high and largely unaddressed burden of postpartum depression, complicated grief, and functional impairment in the weeks and months following loss. No structured, community-level psychosocial support protocol is currently integrated into India's public health system, leaving Accredited Social Health Activists (ASHAs)-who routinely visit women after birth- without standardised guidance for compassionate care after perinatal loss. Methods: This protocol describes a quasi-experimental hybrid type I effectiveness-implementation study embedded within an ongoing implementation research project, SHRiSTI, aiming to reduce stillbirths in India. The Community Continuum of Care for Compassionate Support After Stillbirth (C-COMPASS) is a co-designed, community health worker-delivered counselling intervention for women experiencing stillbirth and early neonatal death. The intervention comprises five structured contacts: a telephonic follow-up within 72 h of discharge, and on Day 7, in-person visits on Day 14 and Day 21, and a Day 42 postpartum review, supported by a red-flag referral pathway. The study will use a quasi-experimental Difference-in-Differences (DiD) analytical design across four administrative blocks (two intervention, two control), with a pre-intervention observation period (approximately 10-12 weeks) and a post-intervention period (approximately 26 weeks). This design was chosen in place of a cluster-randomised controlled trial because block-level allocation is operationally determined within the existing SHRiSTI programme. The primary effectiveness outcome will be mean Edinburgh Postnatal Depression Scale (EPDS) score at 60 days postpartum, measured by independent assessors and analysed using a linear DiD model with block-level fixed effects. An expected sample of approximately 56 women per arm in the post-intervention period provides 81% power to detect a clinically meaningful difference of 3.5 EPDS points with a total anticipated enrolment of approximately 146 women. Secondary outcomes will include binary EPDS score thresholds, grief symptoms, functional disability, and incremental cost per woman counselled. Early neonatal deaths within 24 h will be included in a pre-specified sensitivity analysis. Implementation outcomes will be assessed using the Proctor et al. taxonomy. Discussion: C-COMPASS is among the first community-based interventions in India designed to address psychosocial care following stillbirth through frontline health workers. If demonstrated to be effective and feasible, this model may provide a scalable, low-cost approach for integration into India's ASHA-based maternal health system.