Evaluating Quality of Maternal Health Care in Low- and Middle-Income Countries: Evidence to Inform Policy and Practice

Abstract

Ending preventable maternal and newborn deaths and realizing the right to health will require expanding access to high-quality care across the pregnancy-postpartum continuum. Progress in driving down maternal mortality has slowed over the past decade of the Sustainable Development Goal (SDG) era, and maternal deaths remain heavily concentrated in low- and middle-income countries (LMICs), particularly in sub-Saharan Africa.1 Access to maternal and newborn care has improved dramatically, but remains uneven across geography and income, and extensive problems with quality undermine the value of care. Maternal health services in LMICs were further impacted by the unprecedented strain of the COVID-19 pandemic on health systems. This dissertation aims to contribute to a better understanding of the quality of care provided during pregnancy, childbirth, and the postpartum period in LMICs, considering the delivery of evidence-based clinical care and women’s experiences of care. The three studies presented assess gaps in the provision of basic postnatal care for women in Kenya, potential spillover effects of continuous quality improvement for antenatal HIV testing on comprehensive antenatal care (ANC) in South Africa, and women’s decision-making around antenatal and childbirth care-seeking and perceptions of quality of care in Côte d’Ivoire In Chapter 1, I evaluated the coverage and quality of postnatal care and examined predictors of receiving basic maternal postnatal care among 2,676 postpartum women in eight counties in Kenya. Using longitudinal survey data from the control group of a randomized controlled trial, I found that despite near-universal postnatal care-seeking and coverage of infant-focused measures of care, fewer than one-third of women received basic maternal postnatal care. These findings suggest that postnatal visits in Kenya are often treated primarily as infant check-ups, with many women not receiving a basic standard of care. Mixed effects logistic regression models showed that women who were married and reported always being treated with respect during delivery had significantly higher odds of receiving basic maternal postnatal care, while there was no significant difference seen among women with high-risk pregnancies. Exceptionally high care-seeking in this sample highlights a substantial gap between crude and quality-adjusted coverage, suggesting the need to measure quality of care for both women and infants and ensure that women are provided with a basic standard of postnatal care. In Chapter 2, I evaluated the potential spillover effect of a nurse-led continuous quality improvement initiative focused on improving two HIV-related components of ANC—viral load monitoring and repeat HIV testing—on the provision of other national guideline-based services for the first ANC visit in a rural and HIV hyperendemic area in Kwazulu-Natal, South Africa. Using data from a pragmatic, stepped-wedge cluster-randomized controlled trial and applying mixed effects Poisson regression, I found no spillover effects—either beneficial or detrimental—of the intervention on the provision of 18 components of care recommended by South Africa’s 2015 national guidelines for maternity care. Regardless of intervention exposure, there were notable gaps in coverage of essential antenatal care services such as symphysis fundal height measurement and iron supplementation. These results suggest that looking beyond vertical quality improvement initiatives aimed at a limited number of services towards more comprehensive and system-level changes may be needed to strengthen the clinical quality of antenatal care more broadly. In Chapter 3, I explored women’s decision-making around antenatal and delivery care-seeking, and their perceptions and experiences of quality of care within the Taabo Health and Demographic Surveillance System (HDSS) in Côte d’Ivoire. Taking a phenomenological qualitative approach, our team conducted focus group discussions (FGDs) and in-depth interviews (IDIs) with pregnant and postpartum mothers to inform the design and evaluation of future interventions to improve access to high-quality maternal health care in the Taabo HDSS. Despite a national free maternity care policy, unpredictable and rising out-of-pocket (OOP) costs were a major cause of missed and delayed ANC, and further entrenched women’s lack of autonomy in health care decision-making. Lack of information on facility capacity was prevalent, and experience of care—particularly being treated with warmth and respect by providers—predominated women’s decision-making on where to seek care. Our findings indicate the need to design and evaluate interventions addressing the effect of costs on women’s access to care and autonomy, lack of information on facility capacity, and widespread disrespect and abuse of pregnant and postpartum women within the health system. Considered as a whole, these findings contribute to a growing body of evidence on the quality of maternal health care in LMICs, revealing substantial gaps in the delivery of clinically effective and person-centered care across three geographically and epidemiologically diverse settings and underscoring the need for fundamental health system reform to deliver high-quality care for all women and newborns.