Previous studies of cancer care have demonstrated improved long-term patient outcomes for those treated at high-volume centers. The influence of treatment center characteristics on outcomes for primary nonmetastatic melanoma is not currently established.
We aimed to investigate the association of cancer treatment center case volume and academic affiliation with long-term patient survival for cases of primary nonmetastatic melanoma.
Cases of melanoma diagnosed in US adults from 2004 to 2014 and included in the National Cancer Database were identified. Hospitals were grouped by yearly case-volume quartile: bottom quartile, 2 middle quartiles, and top quartile.
Facility case volume was significantly associated with long-term patient survival (P < .0001). The 5-year survival rates were 76.8%, 81.9%, and 86.4% for patients treated at institutions in the bottom, middle, and top quartiles of case volume, respectively. On multivariate analysis, treatment at centers in both middle quartiles (hazard ratio, 0.834; 95% confidence interval, 0.778-0.895) and in the top quartile (hazard ratio, 0.691; 95% confidence interval, 0.644-0.741) of case volume was associated with improved survival relative to that of patients treated at hospitals in the bottom quartile of case volume. Academic affiliation was associated with improved outcomes for top-quartile- but not middle-quartile-volume facilities.
Disease-specific survival was not available.
Treatment at a high-volume facility is associated with improved long-term patient survival for melanoma. High-volume academic centers have improved patient outcomes compared with other high-volume centers.Le texte complet de cet article est disponible en PDF.
Key words : case volume, melanoma, NCDB, outcomes, survival
Abbreviations used : BQV, CI, HR, IL-2, MQV, NCDB, SE, TQV
| Funding sources: Supported by the Leon Rosenberg, MD, Medical Student Research Fund in Genetics and the Jane Danowski Weiss Family Foundation Fellowship. The funders had no role in the present study.
| Conflicts of interest: None disclosed.
| Reprints not available from the authors.