Does Distance to Treatment Affect Mortality Rate for Surgical Oncology Patients?

Abstract

Introduction Using the National Cancer Database (NCDB), we seek to analyze the relationship of patient distance to hospital of treatment on mortality trends after surgery, since patients often travel large distances to referral centers. Methods A retrospective cohort study of the NCDB from 2004 to 2013 was performed, and patients with gastrointestinal, melanoma, and head and neck primary site tumors who underwent surgery were included. We excluded cases with no recorded mortality status or distance from the hospital. A multivariable logistic regression was conducted with adjustments for population density, treating facility location, age, race, gender, education, income, insurance, comorbidities (Charlson-Deyo score), days from diagnosis to treatment, positive margin, tumor stage and grade, and lymph or vascular invasion. Results A total of 1 424 482 patients were included. Overall median distance to hospital was 9.7 miles (range 4.2-23.7 miles); 696 647 (48.91%) of the sample traveled a distance greater than 10 miles to the institution where the procedure was performed. The multivariable regression analysis demonstrated overall lower mortality for those patients travelling a longer distance to care for multiple tumor types, including: liver (OR .87, .77-.99, P = .032), pancreas (OR .82, .76-.89, P < .001), colon (OR .92, .89-.95, P < .001), rectum (OR .90, .83-.96, P = .003), melanoma (OR .83, .79-.88, P < .001), and tumors of the larynx (OR .80, .69-.94, P = .005). Discussion Increased distance traveled for surgical treatment has a significant correlation with decreased odds of mortality for multiple cancers, highlighting the importance of centralized referral patterns for oncology care.