OBJECTIVES: Studies of the hospital volume-outcome relationship (VOR) have highlighted that greater volume activity improves patient outcome. While this finding has been known for years in health services research, most VOR studies to date have failed to delve into what underlies this relationship. This study aimed to disentangle the VOR by comparing treatment modalities (e.g. chemotherapy protocols) for epithelial ovarian carcinoma (EOC) patients.

METHODS: A comprehensive cohort of 267 EOC patients diagnosed in 2012 in the Rhone-Alps region of France was used. Three outcomes were considered: progression-free survival, re-operation, and tumor resection. Discrete choice and count data models were first applied to underscore different treatment approaches. VOR was then estimated with instrumental variable (IV) analysis while controlling for the different treatment approaches. Hospital volume activity was instrumented by the distance of the patients’ homes to the hospital and the median net incomes in the specific areas where the patients resided.

RESULTS: After controlling for selection biases, being treated in a higher volume hospital significantly improved all three patient outcomes. This can be partially explained by major differences in treatment approaches. On average, patients were more likely to receive neoadjuvant chemotherapy (p<0.001), a higher number of cure (p<0.001), and to be treated with both adjuvant and neoadjuvant chemotherapy (p<0.001) in higher volume hospitals. IV analysis showed that the VOR decreased when we controlled for treatment approaches.

CONCLUSIONS: We have highlighted several factors that characterize the observed part of VOR for EOC patients. Higher volume hospitals appear to more often make the right decisions in regard to how to treat patients, which contributes to the positive impact of the hospital volume activity on patient outcomes. Volume alone is, an imperfect correlate of quality. To build volume-based policies, policy makers need to know what volume is a proxy for.