Effect of patient risk on the volume-outcome relationship in obstetric delivery services

Health care organizations that offer more delivery services are assumed to provide better quality of care, and a higher rate of cesarean section (CS) is generally assumed to be an indicator of poor quality of care. This study analyzed whether the volume-outcome relationship in delivery services, measured by the rate of CS, differed depending on the risk status of delivery patients. Delivery claims were identified in the National Patient Sample (NPS) for 2009. The study hospitals were categorized into low and high delivery-volume groups, and patients were categorized into three risk groups (below average, medium, and high) based on their risk status. Risk factors were included in the adjustment model to identify differences among patients and produce risk-adjusted CS rates. Risk-adjusted CS rates did not differ significantly between patients in low- and high-volume hospitals when the sample was not divided according to risk status. However, when the sample was divided according to patient risk status, significant differences in risk-adjusted CS rates in the below-average- and medium-risk groups were revealed between low- and high-volume hospitals. No such significant difference was observed for the high-risk group. The largest difference in CS rates between low- and high-volume hospitals was observed in the medium-risk group, and the high-risk group showed the smallest difference between the two volume groups. The high-risk group had the highest CS rates, and the below-average-risk group had the lowest CS rates. Although we found the traditional volume-outcome relationship in delivery patients, the data also revealed that patient risk status influenced this relationship. Policies and interventions based on volume-outcome theory should differ according to patient risk status.