Hospital volume and physician volume in association with survival in patients with nasopharyngeal cancer after radiation therapy

Background: Cancer patients treated at higher-volume hospitals or by more experienced physicians have better outcomes. However, little is known about the effect of these provider volumes on the prognosis of patients receiving definitive radiotherapy (RT). This study aims to examine the independent association between hospital volume and physician volume in relation to the overall survival (OS) of nasopharyngeal cancer (NPC) patients after RT. Methods: Patients with newly diagnosed NPC receiving definitive RT between 2001 and 2017 were identified from Taiwan's National Health Insurance Research Database. We collected demographic characteristics of patients, physicians and hospitals, as well as cancer treatment and comorbidities. Patients were categorized into quartiles according to cumulative hospital and physician volumes of their treatment providers. The effects of hospital and physician volumes on OS was examined by the frailty Cox regression model. Results: A total of 16,315 NPC patients treated by 258 physicians in 92 hospitals were identified. When the effects of hospital volume and physician volume on survival were separately examined, both of them were positively associated with OS. In a fully adjusted model considering patient and provider characteristics, hospital volume, physician volume, and clustering effects, it showed that hospital volume significantly predicted OS, while physician volume did not. In patients treated with advanced technique RT, hospital volume was significantly associated with OS. However, volume effect was not observed in 2-D RT subgroup. Conclusions: NPC patients receiving RT at higher-volume hospitals saw better survival. Treatment for NPC should be centralized to high-volume hospitals rather than high-volume physicians.