Improving utilisation of maternal health related services: the impact of a community health worker pilot programme in Neno Malawi

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Malawi has one of the highest maternal mortality ratio (MMR) in sub-Saharan Africa (SSA). Despite investments in family planning and emergency obstetric care (EmOC), Malawi’s Millennium Development Goal (MDG) target of reducing maternal deaths to 155 deaths per 100,000 live births was not met by the end of 2015. Between 2010 and 2015, Malawi was only able to reduce the MMR from 675 to 439 per 100,000 live births. Inadequate utilisation of perinatal services is the contributing factor to the MMR target not being achieved. One approach for improving the utilisation of perinatal services is to invest in community health workers (CHWs). CHWs can be trained to: identify women of child bearing age (WCBA) who need perinatal services; provide community education; encourage timely referral of clients to the nearest health facility; and undertake community follow up for WCBA who are pregnant and/or have recently given birth. We evaluated changes in utilisation of antenatal care (ANC), facility based births, and postnatal care (PNC) after CHW deployment to conduct monthly home visits to WCBA for pregnancy identification and escorting women to ANC, labour and facility birth and PNC clinics in Neno district, Malawi. The CHW programme was implemented in two catchment areas from March 2015 to June 2016. Methodology: We employed a retrospective quasi-experimental study design to evaluate the impact of CHWs on changes in the utilisation of ANC, facility based births, and PNC in Neno district, Malawi between March 2014 and June 2016 (pre-intervention period: March 2014 to February 2015, and post-intervention period: March 2015 to June 2016). Monthly outcomes were compared between a combined CHW intervention area and its synthetic control area using the synthetic control method. The synthetic control area (or synthetic counterfactual of the CHW) 14 was the control area that was created from multiple available control sites where the CHW programme was not implemented to allow the comparison of outcomes between the sites where CHWs were implemented and the sites where CHWs was not implemented. Two hundred and eleven CHWs (128 existing CHWs plus 83 new CHWs from the community) were trained in maternal health and deployed to cover an estimated 5,132 WCBA living in a catchment area of about 20,530 people. The primary focus of the CHWs was to conduct monthly household visits to identify pregnant women, and then escort pregnant women to their initial and subsequent ANC appointments, facility births, and to PNC check-ups. As part of package of care, community mobilisation and improvements in services to achieve a minimum package of services at the local health centres were also added. Using the synthetic control method, as developed by Abadie and Gardeazabal (2003) and Abadie, Diamond and Hainmueller (2010) and a Bayesian approach of synthetic control developed by Brodersen (2015), a synthetic counterfactual of the CHW intervention was created based on six available public control facilities. The synthetic counterfactual trend was created to have similar pre-intervention characteristics as the CHW intervention trend. The impact of the CHW intervention was the difference between the CHW intervention site and its synthetic counterfactual Results: CHWs in the intervention areas visited an average of 3,147 (range 3,036 – 3,218) of WCBA monthly, covering 61.0% of WCBA. During these visit 3.6% (97 women per month) of WCBA were suspected to be pregnant every month. Of those women suspected to be pregnant, 67.8% (66 women per month) were escorted to health facilities immediately every month. CHWs 15 visited an average of 254 pregnant women enrolled in ANC and 64 women in postpartum period monthly. ANC and facility births utilisation in the CHW intervention site increased in comparison to the control site. Firstly, the number of new pregnant women enrolled in ANC per month increased by 18.0 % (95% Credible Interval (CrI) 8.0%, 28.0%), from 83 to 98 per pregnant women. Secondly, the proportion of women starting ANC in first trimester increased by 200.0% (95% CrI 162.0%, 234.0%), from 9.5% to 29.0% per month. Thirdly, the number of women attending four or more ANC visits increased by 37.0% (95% CrI 31.0%, 43.0%), from to 28.0% to 39.0%. Lastly, the number of facility births increases by 20% (CrI 13.0%, 28.0%), from 85 women to 102 per month. However, there was no net difference on PNC visits between the CHW intervention site and its counterfactual unit (-37.0%, 95% CrI -224.0%, 170.0%). Conclusions: CHW intervention significantly increased the utilisation of ANC and facility based births in Neno, Malawi. However, CHWs had no net difference on PNC utilisation.