Risk factors for maternal death revisited.

Abstract

For many years the UK Confidential Enquiry into Maternal Deaths (CEMD) has reported increased risks of pregnancy-associated death in certain groups of women, such as older mothers and substance abusers. The CEMD, however, lacks a control group, so these conclusions have had to be based on comparisons with data from other sources, e.g. the Office for National Statistics or Hospital Episode Statistics. Comparison between deaths and ‘near miss’ events became possible after the UK Obstetric Surveillance System (UKOSS) was established to study uncommon disorders of pregnancy. Conditions such as eclampsia and pulmonary embolism feature in reports from both CEMD and UKOSS, allowing researchers to compare women who died and those who survived. Kayem et al. (PLoS ONE 2011;6:e29077) found that the risk of death was increased by older age, black ethnicity, obesity and unemployment. In a similar, larger study, Nair et al. (BJOG. 2015;122:653-62) identified six risk factors, such as inadequate use of antenatal care, which together accounted for 70% of the increased risk of death. UKOSS also collects details on controls—one or more women delivering in the same hospital immediately before the near-miss case was delivered. These now provide an opportunity for the long-awaited comparison between women who died and women with uncomplicated pregnancies. In this issue, Nair et al. compare 383 maternal deaths reported to MBRRACE-UK in 2009–2013 and 1516 controls from UKOSS studies between 2010 and 2012. The two groups, drawn from the same national population over similar time periods, are as closely comparable as possible. This study is ground-breaking in several ways. First, it combines data from two well-established and accurate surveillance programmes. Their reliability contrasts with much of our global information on maternal mortality, which is derived either from estimates by international bodies or from national registers—notoriously subject to under-reporting. Secondly, the study includes maternal deaths from both direct and indirect causes. This is important because in the UK indirect deaths have outnumbered direct deaths since 1997. Worldwide, however, direct causes still account for 73% of maternal deaths (Say et al. Lancet Global Health. 2014;2:e323-33). This difference is one example of why lessons for reducing maternal mortality must be learned at a national level. Thirdly, the study challenges our ideas about risk factors. Of the 13 factors examined, only seven were found to be associated with maternal mortality. They include inadequate utilisation of antenatal care, medical comorbidities and, surprisingly for the UK, anaemia. Missing from the list are ethnicity, parity, body mass index and cigarette smoking, all of which were significant on initial analysis but not after adjustment for other factors. The clinical implications of this study are important. It confirms the risks associated with unemployment and increasing maternal age. The authors point out that women from black African backgrounds should still be considered high risk, but their analysis strongly suggests that the excess mortality in this and other at risk groups can be substantially reduced by targeted interventions, such as attention to medical co-morbidities and improved access to antenatal care. None declared. Completed disclosure of interests form available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.