National hospital care survey demonstration projects: Severe maternal morbidity in inpatient and emergency departments

Abstract

Objective—This report demonstrates the ability of the National Hospital Care Survey (NHCS) to examine delivery hospitalizations involving severe maternal morbidity (SMM). These data are unweighted and not nationally representative, so the results are intended to illustrate the unique capability of NHCS to track patients across hospitalizations and emergency department (ED) visits rather than provide nationally representative estimates of these outcomes. Methods—Administrative claims data and electronic health records data from the 2016 NHCS were used to examine SMM in the inpatient and ED settings. The 21 indicators of SMM classified by the Centers for Disease Control and Prevention were used to identify delivery hospitalizations involving SMM. The collection of patient identifiers (such as patient name, Social Security number, and date of birth) allowed for patients with SMM at delivery to be tracked to predelivery hospitalizations and postpartum ED visits within 2016. The data are unweighted and not nationally representative. Results—In the 2016 NHCS, 117,336 patients delivered between April 1, 2016, and September 30, 2016 (97.0% of these had a live birth, 1.0% had a stillbirth, and 2.0% were unspecified). Among this study population, 2,335 (2.0%) patients had SMM at delivery. These patients tended to be older, had a longer hospital stay, had a higher proportion of preterm births, and were more likely to have either a predelivery hospitalization or a postpartum ED visit. Nearly one-half of patients (45.5%) with SMM delivered at 37 weeks or earlier, compared with 19.9% of patients without SMM. Most patients with SMM who had a postpartum ED visit went to the ED within 14 days of their delivery discharge. This study highlights the unique analytical capabilities of NHCS.