Individual, health facility and wider health system factors contributing to maternal deaths in Africa: A scoping review

Abstract

The number of women dying during pregnancy and after childbirth remains unacceptably high, with African countries showing the slowest decline. The leading causes of maternal deaths in Africa are preventable direct obstetric causes such as haemorrhage, infection, hypertension, unsafe abortion, and obstructed labour. There is an information gap on factors contributing to maternal deaths in Africa. Our objective was to identify these contributing factors and assess the frequency of their reporting in published literature. We followed the Arksey and O'Malley methodological framework for scoping reviews. We searched six electronic bibliographic databases: MEDLINE, SCOPUS, African Index Medicus, African Journals Online (AJOL), French humanities and social sciences databases, and Web of Science. We included articles published between 1987 and 2021 without language restriction. Our conceptual framework was informed by a combination of the socio-ecological model, the three delays conceptual framework for analysing the determinants of maternal mortality and the signal functions of emergency obstetric care. We included 104 articles from 27 African countries. The most frequently reported contributory factors by level were: (1) Individual-level: Delay in deciding to seek help and in recognition of danger signs (37.5% of articles), (2) Health facility-level: Suboptimal service delivery relating to triage, monitoring, and referral (80.8% of articles) and (3) Wider health system-level: Transport to and between health facilities (84.6% of articles). Our findings indicate that health facility-level factors were the most frequently reported contributing factors to maternal deaths in Africa. There is a lack of data from some African countries, especially those countries with armed conflict currently or in the recent past. Information gaps exist in the following areas: Statistical significance of each contributing factor and whether contributing factors alone adequately explain the variations in maternal mortality ratios (MMR) seen between countries and at sub-national levels.

Fig 1. Conceptual framework.

A modification of the social-ecological model into three levels of contributing factors: Individual, health facility and wider health system levels and their linkage to the three delays framework.

Fig 2. Prisma flow chart of the scoping review process.

The flow chart illustrates the stages of article processing from identification from the various databases, screening using the inclusion and exclusion criteria, and inclusion in the review. In the final screening, primary articles reporting on maternal deaths (population), factors contributing to maternal deaths (concept) and within an African country (setting) were included.

Fig 3. Individual-level factors contributing to maternal deaths in Africa.

They are ranked from the most frequently reported (delay in care-seeking) to the least frequently reported (husband’s level of education). The number and proportion n (%) of articles reporting a specific factor is on the X-axis. The total number of included articles (N = 104) is the denominator. The factors and their explanatory descriptions are on the Y-axis.

Fig 4. Health-Facility level factors contributing to maternal deaths in Africa.

Fig 5. Wider health system-level factors contributing to maternal deaths in Africa.

They are ranked from the most frequently reported (transport to and between health facilities) to the least frequently reported (hospital type, management, and mode of payment for health services). The number and proportion n (%) of articles reporting a specific factor is on the X-axis. The total number of included articles (N = 104) is the denominator. The factors and their explanatory descriptions are on the Y-axis.

Fig 6. A comparison of the various factors contributing to maternal deaths in Africa, emphasising health facility-level factors.

The total counts were 822 (Individual level 276 (33.6%), health facility level 403 (49.0%), and wider health system level 143 (17.4%). The health facility-level factors were service delivery 226 (56.1%), staffing 65 (16.1%), infrastructure 42 (10.4%), medications 40 (9.9%), staff training and skills 15 (3.7%) and essential commodities and equipment 15 (3.7%). See S2 Table for further details.