Severe maternal morbidity and comorbid risk in hospitals performing <1000 deliveries per year

BACKGROUND: While research has demonstrated increasing risk for severe maternal morbidity in the United States, risk at lower volume hospitals remains poorly characterized. More than half of all obstetric units in the United States perform< 1000 deliveries per year and improving care at these hospitals may be critical to reducing risk nationwide. OBJECTIVE: We sought to characterize maternal risk profiles and severe maternal morbidity at low-volume hospitals in the United States. STUDY DESIGN: We used data from the Nationwide Inpatient Sample to evaluate trends in severe maternal morbidity and comorbid risk during delivery hospitalizations in the United States from 1998 through 2011. Comorbid maternal risk was estimated using a comorbidity index validated for obstetric patients. Severe maternal morbidity was defined as the presence of any 1 of 15 diagnoses representative of acute organ injury and critical illness. RESULTS: A total of 2,300,279 deliveries occurred at hospitals with annual delivery volume <1000, representing 20% of delivery hospitalizations overall. There were 7849 cases (0.34%) of severe morbidity in low-volume hospitals and this risk increased over the course of the study from 0.25% in 1998 through 1999 to 0.49% in 2010 through 2011 (P<.01). The risk in hospitals with >= 1000 deliveries increased from 0.35-0.62% during the same time periods. The proportion of patients with the lowest comorbidity decreased, while the proportion of patients with highest comorbidity increased the most. The risk of severe morbidity increased across all women including those with low comorbidity scores. Risk for severe morbidity associated with obstetric hemorrhage, infection, hypertensive diseases of pregnancy, and medical conditions all increased during the study period. CONCLUSION: Our findings demonstrate increasing maternal risk at hospitals performing <1000 deliveries per year broadly distributed over the patient population. Rates of morbidity in centers with >= 1000 deliveries have also increased. These findings suggest that maternal safety improvements are necessary at all centers regardless of volume.