Co-designed and co-delivered place-based community interventions to reduce inequity in early initiation of antenatal care: Findings from the cluster randomised controlled Community REACH trial

BackgroundEarly initiation of antenatal care provides timely screening, advice and support. Inequities in early care initiation exist in high income countries, but there is scant evidence on effective interventions. The Community REACH trial aimed to assess the effectiveness of co-produced place-based interventions to strengthen community support for early care initiation. MethodsMatched-pair cluster randomised trial in socially disadvantaged and ethnically diverse areas in England. Electoral wards with low rates of early care initiation were matched and randomly allocated to intervention or control (usual care) (n=10 pairs). Following a three month co-design phase, community organisations and volunteers in intervention sites conducted targeted outreach activities over six months. The primary outcome was initiation of antenatal care by the 12th completed week of pregnancy. ResultsThere was no evidence of a difference in the primary outcome (OR 1.07(0.89; 1.28)). There were also no statistically significant differences in rates of emergency caesarean, pre-term birth, low birth weight, smoking, or breastfeeding. There was a higher rate of care initiation by 10 weeks and fewer antenatal admissions in the intervention arm during the intervention period although differences were not sustained after it finished. ConclusionThis rigorous evaluation found limited impact of short-term place-based interventions to strengthen community support for early initiation of antenatal care. Future initiatives may benefit from embedding in integrated health and care structures to ensure sufficient time and resources for mobilisation of community assets and focusing on smaller hyper-local neighbourhoods. Actions to tackle wider structural and organisational barriers are also needed. Trial registrationISRCTN registry: registration number 63066975. Registered on 18 August 2015. O_TEXTBOXWHAT IS ALREADY KNOWN ON THIS TOPICO_LIPrevious research in high income countries has identified inequalities in access to antenatal care, yet there is little evidence on interventions to improve early initiation of antenatal care. C_LIO_LICo-produced place-based interventions which develop and strengthen community support offer a promising approach to tackle health inequalities but this type of approach has not yet been rigorously tested in the context of early initiation of antenatal care in high income countries. C_LI WHAT THIS STUDY ADDSO_LIIt was possible to develop and implement a co-designed and co-delivered place-based six month intervention to increase early access to antenatal care in ethnically and linguistically diverse inner-city neighbourhoods. C_LIO_LIThere is no evidence that this short-term intervention increased the proportion of women accessing antenatal care before 12 completed weeks of pregnancy; initiation of care by 10 weeks increased significantly whilst the intervention was running, but this was not maintained after it ended. C_LIO_LIIntervention implementation varied across sites in the extent to which it reflected the underpinning intervention logic of co-production and community development; there was, however, no evidence of an association between implementation and intervention effect estimates. C_LI HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICYO_LIGiven the results of this study, funders and intervention developers of future initiatives to strengthen community support should ensure a longer lead in time so that interventions can be embedded within integrated health and care structures and allow sufficient time, resources and capability for co-production, the development of community relationships and asset-based ways of working. C_LIO_LIEfforts to tackle inequalities in early initiation of antenatal care should also consider service-level actions such as geographical location, availability of interpreters and anti-racist practice training. C_LI C_TEXTBOX