Defining severe maternal morbidity—When is it time to stop?

This special issue of Paediatric and Perinatal Epidemiology contains three papers devoted to defining severe maternal morbidity. Himes and colleagues 1 used criteria developed by the Centers for Disease Control to interrogate their routine hospital data, and then validated the identified cases using American College Of Obstetricians and Gynecologists criteria as their ‘gold standard’. They identified that about half the cases based on the CDC criteria were false positives according to ACOG criteria, and that there was a likelihood of substantial numbers of false negative cases. As part of the Europeristat collaboration, Chantry and colleagues 2 investigated potential indicators of severe maternal morbidity that could be consistently used across hospital discharge databases in Europe. They compared their findings to population-based mortality estimates, and concluded that data for only two indicators, hysterectomy and red cell transfusion, when associated with an obstetric haemorrhage code, were of sufficient quality for international comparisons. Dzakpasu and colleagues 3 developed another consensus set of criteria amongst a multi-disciplinary group of experts in Canada, on the basis of their observation of the lack of agreement internationally, and perceived lack of comparability of other studies.