INTRODUCTION: There has been significant interest in the relationship between hospital volume and outcomes across fields. However, the relationship between hospital volume, established quality metrics, and survival has not been established for head and neck cancer. We investigated the relationship between hospital volume and outcomes for head and neck cancer. METHODS: Using the National Cancer Database (2004-2016) data for upper aerodigestive tract squamous cell carcinoma, hospital volume deciles were determined. Univariate log-rank analysis was used to determine volume thresholds with significantly different overall survival (OS). The primary outcome measure was overall survival; other outcomes included complete neck dissection (≥18 lymph nodes removed), negative margins, and time to adjuvant radiation <6 weeks. Statistical analysis was performed using multivariable logistic regression and Cox proportional hazards models. RESULTS: Data from 270,047 patients and 1325 facilities were analyzed. Volume thresholds were: fewer than 22 cases per year (lowest volume), 22-38 cases per year (low-moderate volume), 38-122 cases per year (moderate volume), 122-160 cases per year (high-moderate volume), and 160 or more cases per year (highest volume). There was a positive trend in the adjusted odds ratio of complete neck dissection compared to moderate volume (OR range 0.53-1.52 from lowest to highest volume, p-for-trend<0.0001). There was also a positive trend in the odds of negative margins with increasing volume (OR range 0.85-1.45 lowest-highest, p-for-trend<0.0001). All centers had similar odds for time to post-operative radiation less than six weeks. Unadjusted five year OS was 52.3% in the lowest volume, 52.9% in the low-moderate volume, 54.4% in the moderate volume, 56.0% in the high-moderate volume, and 57.6% in the highest volume group. Cox proportional hazards models showed decreased survival in the lowest (HR 1.10, 99% CI 1.05-1.15, p<0.001) and low-moderate volume groups (HR 1.07, 99% CI 1.02-1.13, p<0.001) compared to moderate volume, and a protective effect in the higher volume groups (high-moderate volume HR 0.95, 99% CI 0.86-1.05, p=0.23; highest volume HR 0.94, 99% CI 0.87-1.01, p=0.04). The same trend was present when data were stratified by anatomical site. Among those who initially received non-surgical treatment, the lowest and low-moderate volume groups had significantly lower survival compared to moderate volume. However, the non-surgical high-moderate and highest volume groups were not significantly different from moderate volume. No groups significantly differed among patients who had surgery first. CONCLUSION: This study of a nationally representative database confirms the volume-outcome relationship for all head and neck cancer treatment and shows that surgical quality metrics are associated with hospital volume. When stratified by treatment type, volume was associated with overall survival for patients who had non-surgical treatment as their initial course.

Citation Format: Ari D Schuman, Andrew J Rosko, Scott E Regenbogen, Steven B Chinn. Volume and quality in head and neck cancer [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr A139.