Background: Increased facility surgical treatment volume is sometimes associated with improved survival in patients with cancer; however, published studies evaluating volume are heterogeneous and disparate in their patient inclusion and definition of volume. The purpose of this work was to evaluate uniformly the impact of surgical facility volume on survival in patients with cancer.
Methods: The National Cancer Database was searched for patients diagnosed in 2004 through 2013 with the 12 cancers most commonly treated surgically. Facilities were stratified by 4 categories using the overall population (low, intermediate, high, and very high), each including 25% of patients, and then stratified by each individual disease site. Five-year postsurgery survival was estimated using both the Kaplan-Meier method and corresponding log-rank tests for group comparisons. Cox proportional hazard models were used to evaluate the effects of facility volume on 5-year postsurgery survival further, adjusted for multiple covariates.
Results: A total of 3,923,618 patients who underwent surgery were included from 1,139 facilities. Of these, 40.4% had breast cancer, 12.8% prostate cancer, and 10.0% colon cancer. Most patients were female (65.0%), White (86.4%), and privately insured (51.6%) with stage 0-III disease (64.8%). For all cancers, the risk of death for patients undergoing surgery at very high-volume facilities was 88% of that for those treated at low-volume facilities. Hazard ratios (HRs) were greatest (very high vs low volume) for cancer of the prostate (HR, 0.66; 95% CI, 0.63-0.69), pancreas (HR, 0.75; 95% CI, 0.71-0.78), and esophagus (HR, 0.78; 95% CI, 0.73-0.83), and for melanoma (HR, 0.81; 95% CI, 0.78-0.84); differences were smallest for uterine and non-small cell lung cancers. Overall survival differences were greatest for cancers of the brain, pancreas, and esophagus.
Conclusions: Patients treated surgically at higher-volume facilities consistently had improved overall survival compared with those treated at low-volume centers, although the magnitude of difference was cancer-specific.