Background: Many cases of maternal mortality and morbidity are preventable. A delayed response to clinical warning signs contributes to preventability. Therefore, the National Partnership for Maternal Safety devised maternal early warning criteria (MEWC), composed of abnormal vital signs that trigger bedside evaluation by a provider with the capacity to escalate care. The relationship of the MEWC to maternal morbidity has not been studied. We evaluated the correlation between the MEWC and maternal morbidity.
Methods: We retrospectively reviewed the first 400 deliveries at the University of Chicago in 2016. We analyzed the electronic medical record to determine whether vital signs triggered the MEWC during the admission to labor and delivery and whether patients experienced morbidity during their delivery hospitalization. The association between MEWC and morbidity was tested using χ analysis. We calculated the sensitivity, specificity, and positive and negative predictive values of the MEWC.
Results: Two hundred eighty-one (70%) of 400 patients triggered the MEWC at least once, and 198 (50%) of 400 patients had multiple or recurrent triggers. Ninety-nine (25%) of 400 patients experienced morbidity. The most common causes of morbidity were hemorrhage, suspected infection, and preeclampsia with severe features. The relative risk of maternal morbidity with at least a single trigger was 13.55 (95% confidence interval [CI], 4.38-41.91) and with recurrent or multiple triggers was 5.29 (95% CI, 3.22-8.71). The sensitivity of the MEWC in predicting morbidity was 0.97 (95% CI, 0.92-0.99) and the specificity was 0.39 (95% CI, 0.33-0.44) when patients with at least a single trigger were included. When including only patients with multiple or recurrent triggers, the sensitivity was 0.84 (95% CI, 0.75-0.90) and the specificity was 0.62 (95% CI, 0.56-0.67). The positive predictive value of the MEWC in our population was 0.34 (95% CI, 0.29-0.40), and the negative predictive value was 0.97 (95% CI, 0.93-0.99). When considering only patients with multiple or recurrent triggers, the positive predictive value was 0.42 (95% CI, 0.38-0.46) and the negative predictive value was 0.92 (95% CI, 0.88-0.95).
Conclusions: The MEWC are associated with maternal morbidity. As a screening tool, they appropriately prioritize sensitivity and have an excellent negative predictive value. The criteria demonstrate low specificity, which is slightly improved by considering only patients with recurrent or multiple triggers. Additional efforts to improve the specificity of MEWC, with a focus on identifying sustained or recurrent patterns of abnormal vital signs, may be necessary before their widespread implementation.