Classifying oncologist rurality by practice location and patient population.

Abstract

136 Background: Geographic variation in the oncology workforce has contributed to reduced access to cancer specialists in rural areas. Workforce studies most often classify provider rurality by their practice location. We hypothesized this approach, given its high specificity, would miss the true extent of providers involved in rural cancer care. In this study, we aim to identify a new method for classifying oncologist rurality. Methods: We created a cohort of Medicare beneficiaries diagnosed with breast, colorectal, or lung cancer in 2018. This informed a cohort of all medical, radiation, and surgical oncologists who treated these patients. Two methods were used to classify oncologists’ rurality using Rural Urban Commuting Area (RUCA) codes: 1) Based on the ZIP Code where they provided the plurality of their care; and 2) Based on the percentage of their patients who reside in a rural ZIP Code. We used sensitivity, specificity, and summary statistics to compare these methods and identify a threshold of rural patients to classify provider rurality. Results: We identified 815 rural oncologists by their practice location and 5,450 by their patient population. We found high concordance between these two definitions (AUC: 0.87), with providers practicing in a rural location treating the highest proportion of rural patients (n=658 [80.7%]). Oncologists practicing in a rural area had a lower patient volume (6.4 [SD: 7.8] vs. 12.2 [SD: 14.2]) and were more likely to be a surgeon (73.1% vs. 51.8%) than their metropolitan counterparts. By categorizing oncologist rurality by their patient panel, we have expanded who is considered a rural oncologist (n=5,450 vs. n=815) and created a continuous measure of oncologist rurality. Oncologists treating a high proportion of rural patients had a higher patient volume than those treating no rural patients (10.9 [SD: 11.6] vs. 8.7 [SD: 10.7]). Conclusions: Special consideration should be made when defining oncologists’ rurality status. Basing rurality on their patient panel results in a broader definition of a rural physician, and could be useful for interventions aiming to improve quality of rural cancer care.[Table: see text]