An exploration of sub-national variability in institutional maternal mortality ratios (iMMRs) in Kenya: A meta-analysis of the 2021 health facility data

Francis Githae Muriithi ^1* Christina Easter ² Alfred O Osoti ³ Zahida Qureshi ³ Adam Devall ¹ Arri Coomarasamy ¹

• ¹ WHO Collaborating Centre for Global Women’s Health, Institute of Metabolism and Systems Research, College of Medical and Dental Sciences, University of Birmingham, Birmingham, United Kingdom
• ² Institute of Applied Health Research, College of Medical and Dental Sciences, Public Health Building, University of Birmingham, Birmingham, United Kingdom
• ³ Department of Obstetrics and Gynaecology, School of Medicine, College of Health Sciences, University of Nairobi, Nairobi, Kenya

Background: In many countries in sub-Saharan Africa, the burden of preventable maternal deaths is still unacceptably high. Most Maternal Mortality Ratio (MMR) estimates are national, rarely sub-national. This study explores Kenya’s 2021 national health facility dataset on maternal deaths and live births for sub-national variability to describe the pattern and trend in variation in institutional maternal mortality ratios (iMMRs).Methods: Country-wide health facility data on live births and maternal deaths for 2021 were requested from the District Health Information System (DHIS-2). A descriptive comparison of sub-national (Regional and County) iMMRs to national iMMR was carried out. Against a national average iMMR for Kenya of about 100 per 100,000 live births, those regions and counties with an iMMR <75 per 100,000 live births were defined as positive outliers, and those with an iMMR >125 were defined as negative outliers. Results: In 2021, 1162 maternal deaths and 1,174,774 live births occurred within Kenya’s health facilities. The annual national average iMMR was 99 per 100,000 live births (95% confidence interval (CI): 93.3, 104.8). There was sub-national variability in iMMR at both regional and county levels. Central, Western and Rift Valley regions were positive outliers; North-Eastern Coast and Nairobi regions were negative outliers, while Nyanza and Eastern regions had an iMMR consistent with the national average. Seventeen counties were positive outliers, namely Baringo, Siaya, Nyamira, Elgeyo-Marakwet, West Pokot, Nandi, Kiambu, Laikipia, Nyeri, Samburu, Marsabit, Vihiga, Bungoma, Nyandarua, Kajiado, Murang’a and Trans-Nzoia. Ten counties were negative outliers: Tana River, Mandera, Machakos, Kilifi, Taita–Taveta, Kisumu, Nairobi, Garissa, and Mombasa and Isiolo. The iMMR in the remaining twenty counties was consistent with the national average. The effect sizes of the observed health facility variation were zero and there was no evidence of month-to-month variation. Conclusion: There is evidence of sub-national variability in Kenya's iMMRs. Understanding these reasons for the variability is crucial for developing strategies for improving maternal health outcomes. If positively deviant behaviours and practices are identified, they could form the basis for adopting asset-based approaches such as the positive deviance approach to improve maternal healthcare delivery processes and outcomes and reduce preventable maternal deaths.