Fragmented maternal care linked to higher maternal deaths across Mexico

A fifteen-year nationwide analysis of Mexico has found that greater discontinuity along the maternal-care pathway is associated with higher maternal mortality, and that part of this association extends across territorial boundaries into neighbouring areas. The study, published in the International Journal for Equity in Health, analysed administrative data from 2008 to 2022 and measured how often births failed to complete a connected sequence of maternal care, from antenatal visits through skilled delivery. Its central conclusion is that expanding formal health coverage alone may not protect women when care remains fragmented across institutions, territories and referral networks.

The research team, led by Carlos Pineda-Antúnez and Edson Serván-Mori, addressed a long-standing gap in the evidence on segmented health systems. Mexico’s health system has long been divided into separate financing and delivery arrangements: women with employment-based social security are served through one set of institutions, while those without such entitlements rely on public ministries and other providers. Previous studies linking fragmentation to maternal deaths have generally been cross-sectional, used partial measures of discontinuity, and rarely examined whether the relationship holds over time or spills across space. To address this, the researchers defined health-system fragmentation as the proportion of live births not completing a full maternal-care trajectory comprising any antenatal care, at least five antenatal contacts, initiation in the first trimester, institutional delivery and skilled birth attendance.

Because maternal deaths are rare events, municipal-level mortality ratios are statistically unstable. The authors therefore grouped Mexico’s 2,457 municipalities into 244 spatially contiguous clusters using a method known as Spatial ‘K’luster Analysis by Tree Edge Removal, which merges neighbouring territories until stable analytical units emerge. The unit of analysis was the cluster, social-security group and year, preserving the distinction between women with and without employment-based social security. The team estimated non-spatial fixed-effects panel models alongside spatial lag and spatial Durbin models, the latter allowing outcomes and exposures in one territory to depend on conditions in adjacent ones. A formal impact decomposition then separated the average direct association within a cluster from the indirect association transmitted through neighbouring territories.

The results showed that maternal mortality was strongly clustered in space and concentrated in marginalised regions of the country. In 2022, the direct maternal mortality ratio stood at 53 deaths per 100,000 live births among women without employment-based social security, compared with 46 among those with such security. Across the models, a 10-percentage-point higher fragmentation score was associated with 7.4 additional total maternal deaths per 100,000 live births and 6.6 additional direct maternal deaths per 100,000 live births, both statistically significant. The association with indirect maternal mortality was smaller and less precise, at 0.8 additional deaths per 100,000 live births. Indirect spatial associations accounted for 19.2 percent of the total estimated association with maternal mortality, and spatial dependence was substantial across models, with a spatial autocorrelation coefficient of 0.16.

The authors conclude that these findings carry implications beyond Mexico. Formal entitlement, they argue, is unlikely to translate into protection where maternal care remains discontinuous across providers and jurisdictions, so progress towards equitable universal health coverage requires continuous maternal-care pathways, interoperable information systems and coordination across providers, institutions and territories. They also propose that measures of care-process discontinuity could complement conventional coverage indicators, helping identify underserved populations and prioritise territories for investment in referral systems, emergency transport and access to specialised obstetric care. As an ecological analysis covering grouped municipalities rather than individual women, the study quantifies associations at the population level rather than individual risk, and the authors note that most maternal deaths are preventable through timely access to high-quality care, making the coherence of the care pathway a measurable target for policy.

Subject of Research: The association between health-system fragmentation and maternal mortality across Mexican municipalities from 2008 to 2022

Article Title: Health-system fragmentation and maternal mortality in Mexico, 2008–2022: a spatio-temporal spillover analysis

Article References: Pineda-Antúnez, C., Mendoza-González, M. Á., Cerecero-García, D., Hone, T., Castro, A., Contreras-Loya, D., Gómez-Dantés, O., & Serván-Mori, E. (2026). Health-system fragmentation and maternal mortality in Mexico, 2008–2022: a spatio-temporal spillover analysis. International Journal for Equity in Health. https://doi.org/10.1186/s12939-026-03081-w

Image Credits: AI Generated

DOI: 10.1186/s12939-026-03081-w

Keywords: maternal mortality, health-system fragmentation, Mexico, spatial analysis, spillover effects, universal health coverage, antenatal care, skilled birth attendance, health equity, spatial econometrics, social security, obstetric care