A systematic review of behaviour change interventions to improve maternal health outcomes in sub-Saharan Africa.

The rate of decline in the global burden of avoidable maternal deaths has stagnated and remains an issue of concern in many sub-Saharan Africa countries. As per the most recent evidence, an average maternal mortality ratio (MMR) of 223 deaths per 100,000 live births has been estimated globally, with sub-Saharan Africa's average MMR at 536 per 100,000 live births-more than twice the global average. Despite the high MMR, there is variation in MMR between and within sub-Saharan Africa countries. Differences in the behaviour of those accessing and/or delivering maternal healthcare may explain variations in outcomes and provide a basis for quality improvement in health systems. There is a gap in describing the landscape of interventions aimed at modifying the behaviours of those accessing and delivering maternal healthcare for improving maternal health outcomes in sub-Saharan Africa. Our objective was to extract and synthesise the target behaviours, component behaviour change strategies and outcomes of behaviour change interventions for improving maternal health outcomes in sub-Saharan Africa. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Our protocol was published a priori on PROSPERO (registration number CRD42022315130). We searched ten electronic databases (PsycINFO, Cochrane Database of Systematic Reviews, International Bibliography of Social Sciences, EMBASE, MEDLINE, Scopus, CINAHL PLUS, African Index Medicus, African Journals Online, and Web of Science) and included randomised trials and quasi-experimental studies. We extracted target behaviours and specified the behavioural interventions using the Action, Actor, Context, Time, and Target (AACTT) framework. We categorised the behaviour change strategies using the intervention functions described in the Behaviour Change Wheel (BCW). We reviewed 52 articles (26 randomized trials and 26 quasi-experimental studies). They had a mixed risk of bias. Out of these, 41 studies (78.8%) targeted behaviour change of those accessing maternal healthcare services, while seven studies (13.5%) focused on those delivering maternal healthcare. Four studies (7.7%) targeted mixed stakeholder groups. The studies employed a range of behaviour change strategies, including education 37 (33.3%), persuasion 20 (18%), training 19 (17.1%), enablement 16 (14.4%), environmental restructuring 8 (7.2%), modelling 6 (5.4%) and incentivisation 5 (4.5%). No studies used restriction or coercion strategies. Education was the most common strategy for changing the behaviour of those accessing maternal healthcare, while training was the most common strategy in studies targeting the behaviour of those delivering maternal healthcare. Of the 52 studies, 40 reported effective interventions, 7 were ineffective, and 5 were equivocal. A meta-analysis was not feasible due to methodological and clinical heterogeneity across the studies. In conclusion, there is evidence of effective behaviour change interventions targeted at those accessing and/or delivering maternal healthcare in sub-Saharan Africa. However, more focus should be placed on behaviour change by those delivering maternal healthcare within the health facilities to fast-track the reduction of the huge burden of avoidable maternal deaths in sub-Saharan Africa.