Interventions targeting cardiometabolic non-communicable diseases (NCDs) in pregnancy in low- and middle-income countries (LMICs); a systematic review with focus on complex interventions

Abstract Background Non-communicable diseases (NCDs) such as diabetes, hypertension, and obesity are increasingly recognised important causes of adverse pregnancy outcomes in low- and middle-income countries (LMICs). Despite this growing burden, few large-scale programmes exist to tackle NCDs in pregnancy in these settings. Identification of complex interventions that may be transferable to similar contexts is an important component of future intervention development. Methods We conducted a systematic review to characterise interventions addressing NCDs in pregnancy across LMICs. We included experimental, quasi-experimental, and pre-experimental studies targeting diabetes, hypertension, obesity and chronic kidney disease, or related gestational syndromes among pregnant women or those in the pre-conception or postpartum periods. Searches were conducted in Scopus, MEDLINE, CINAHL, PubMed and Cochrane CENTRAL from 1 January 1990 to 14 June 2024. Two reviewers independently screened studies and extracted using COVIDENCE software. Intervention complexity was assessed using the Cochrane iCAT_SR tool. Complex interventions underwent risk of bias assessment, and all studies underwent credibility checks using open-source databases. Review conducted in accordance with PRISMA guidelines. Results Of 3,477 records screened, 135 studies were included. Most were small, single site trials conducted in tertiary care facilities. Leading contributors were Iran (27%), China (16%) and India (13%), whilst only ten studies (7%) originated from sub-Saharan Africa. Less than one quarter (31/135, 23%) targeted women with underlying NCDs; most focussed on gestational syndromes such as gestational diabetes or pre-eclampsia. Interventions primarily targeted individual participants rather than wider health systems, commonly using educational or pharmacological approaches. Approximately one quarter (37/135, 27%) of studies reported formal intervention design processes or stakeholder involvement. Seventeen studies were classified as complex interventions (iCAT_SR ≥ 12/18), of which 59% (10/17) were judged at high risk of bias. Compared with simple interventions, complex interventions more frequently reported implementation outcomes (41% vs 8%), stakeholder involvement (65% vs 14%), and intervention design processes (82% vs 19%). Conclusions We demonstrate a marked shortage of complex, system-level interventions addressing NCDs in pregnancy across LMICs. Future intervention design should prioritise co-designed, multi-level interventions with embedded implementation outcomes, particularly in under-represented regions such as sub-Saharan Africa. This review provides a practical resource for researchers and policy makers seeking adaptable and scalable interventions for use in comparable settings. Trial registration PROSPERO CRD42025643691