Evaluating the effects of interventions with women’s groups on health outcomes: Resources for programme planners and evaluators

Abstract

There is growing interest in the potential of women’s groups to improve health. Large-scale government programs, such as India’s National Health Mission and National Rural Livelihoods Mission, promote groupbased interventions to improve maternal and child health. The evidence base has also expanded considerably in recent years, with a growing number of impact evaluations that examine the effects of group-based interventions on health outcomes. Unfortunately, recent evidence syntheses have also identified several weaknesses in evaluations (Anderson et al., 2019; Gichuru et al., 2016; Kumar et al., 2018; Orton et al., 2016). Orton and colleagues’ review of evaluations of microfinance-based groups found that 17 of 31 evaluations were of low quality, mostly due to selection bias. A systematic review of 44 studies in India found that one in three experimental and quasiexperimental studies were at high risk of bias due to selection bias, failure to pre-specify primary outcomes, and not accounting for missing data (Desai et al., 2020). Further, inconsistent measurement approaches and insufficient descriptions of group models make evidence syntheses difficult, which limits learning across contexts. Public health researchers have provided guidance for intervention design, evaluation, and reporting, such as the Medical Research Council guidance on developing and evaluating complex interventions, the CONSORT guidelines for reporting on randomized control trials, and the TIDieR guidelines for describing interventions. This document aims to complement these tools. We provide guidance and resources specific to evaluations of the effect on health outcomes of interventions with women’s groups—with a focus on design and reporting.