Travel distance and time to adjuvant chemotherapy in veterans with colorectal cancer.
84 Background: Patients from rural areas have well described disparities in quality cancer care. We hypothesized that longer travel distance is associated with less chemotherapy acceptance and less timely treatment. Methods: We reviewed 705 electronic medical records of patients with colorectal surgeries from January 1, 2000 to December 31, 2015 at the Veterans Health Administration Tennessee Valley Healthcare System. Two trained abstractors reviewed standard elements (k = 0.79 – 0.92). The study sample included patients with pathological stage high risk II or III CRC and excluded those with metastatic disease or documented National Comprehensive Cancer Network (NCCN) defined medical exclusions from chemotherapy. Primary exposure was distance to care calculated from central zip code of residence to Nashville infusion center. Primary outcomes were receipt of any chemotherapy, and days from surgery to first treatment (truncated at 120 days). We analyzed 2 populations; chemotherapy received and a second sensitivity population who were eligible for, but did not receive, chemotherapy (no documentation of NCCN ineligibility or declined). Results: Of 705 colorectal resections, we excluded 262 for non-cancer, 220 for stage I or low risk stage II, and 46 for NCCN exclusion criteria, yielding 177 cases: 120 colon and 57 rectal. Most patients were male (98%) and white (85%); median age was 64 [Interquartile Range 60, 70]. Distribution by travel distance was 60/177 [33.9%] < 50 miles, 61/177 [34.5%] 50-99 miles, and 56/177 [31.6%] > 100 miles. Of all eligible patients, 123/177 [69.5%] patients received chemotherapy and 54/177 [30.5%] did not receive chemotherapy. Among receivers, median times to treatment were 52 days [40, 61] < 50 miles; 48.5 [40,61] 50-99 miles; and 54 [43,77] > 100 miles, p = 0.3. Patients not receiving chemotherapy varied by distance: 15/60 (25%) < 50 miles; 18/61 (30%) 50-99 miles; 21/56 (38%) > 100 miles, p = 0.3. Including non-receivers, median times to treatment were 58 days [43, 120] < 50 miles; 58.5 [46.5, 120] 50-99 miles; and 80 [48.5, 120] > 100 miles, p = 0.1. Conclusions: Distance to care may influence acceptability of chemotherapy. Understanding patient/provider reasons for omission merits exploration.