Provider-Level Variation in Treatment Planning of Radiation Oncology Procedures in the United States

Abstract

PURPOSE: Variation in the use of radiation oncology procedures and technologies is poorly characterized. We sought to identify associations between the treatment planning codes used to bill for radiotherapy procedures and the demographic characteristics of the radiation oncologists submitting them. METHODS: The Physician and Other Supplier Public Use File was linked to the Physician Compare database by using the physician National Provider Identifier for the year 2016. Analysis was stratified by practice setting, considering both the freestanding non–facility-based (NFB) setting and the facility-based (FB) setting. Multivariable logistic regression was used to determine provider characteristics (gender, practice rurality, and years since graduation) that predicted for the use of 3D-conformal RT (3DCRT) planning, intensity-modulated RT (IMRT) planning, and brachytherapy planning in the Medicare population. RESULTS: Three thousand twenty-nine physicians were linked for analysis. In both the FB and NFB settings together, male gender predicted for decreased likelihood of 3DCRT planning (OR, 0.70, 95% CI, 0.62 to 0.80, P < .001) and increased likelihood of IMRT planning (OR, 1.35, 95% CI, 1.19 to 1.54, P < .001). Brachytherapy planning was also more likely with increasing years since medical school graduation (OR, 1.03, 95% CI, 1.01 to 1.04, P < .001) in the combined FB and NFB settings. These significant associations persisted when examining the NFB and FB settings individually. In both settings overall, brachytherapy planning was more likely in male providers (OR, 1.75, 95% CI, 1.10 to 2.76, P = .02) and also more likely for providers practicing in metropolitan regions compared with those practicing in rural areas (OR, 3.01, 95% CI, 1.23 to 7.39, P = .02). CONCLUSION: Male gender predicts for utilization of IMRT planning, whereas female gender predicts for utilization of 3DCRT planning. Future research is warranted to better understand the role that provider gender and rurality play in the selection of radiation planning techniques for Medicare patients.

National variation in the delivery of radiation oncology procedures in the non‐facility‐based setting

“…With regard to physician gender as a predictor of RT procedure use, an earlier study highlighted significant gender differences in the use of radiotherapy treatment planning codes, 13 whereas our analysis of treatment delivery codes did not reveal any significant gender differences other than with regard to use of LDR brachytherapy. A likely explanation for these disparate findings lies in the distinction between RT planning and delivery codes.…”

“…The latter may consequentially abrogate any gender distinctions in the utilization of specific procedures. We also note that unlike the previously published RT planning analysis, 13 the present study excluded providers in facility-based practices where technical billing is not allowed. Thus, when considering the larger national population of radiation oncologists, there may in fact exist gender differences in the delivery of RT procedures that our study was unable to capture.…”

“…37 While the elimination of facility-based radiation oncologists from our study limits the generalizability of our findings, we still believe that in a small field such as radiation oncology, the practice patterns of over 1,800 physicians are still worth exploring because these data sharpen our understanding of billing practices in "real-world" freestanding radiation oncology practices where technical billing is allowed. Prior work examining procedural planning codes in the larger radiation oncology population (including facility-based providers) similarly uncovered significant variation in multiple practice settings, 13 so it is clear that significant practice variation prevails irrespective of the practice setting studied. Moreover, this study is timely insofar as a shrinking proportion of practitioners in the NFB setting 19 and a shift towards episodic billing 5 may hinder future national practice pattern studies of this kind.…”

The Association of Rural Residence With Surgery and Adjuvant Radiation in Medicare Beneficiaries With Rectal Cancer

“…Although overall receipt and timing of radiation therapy did not differ across the rural-urban continuum, we found the use of IMRT varied across geography, as reported in other settings. 26 , 27 Nonetheless, the routine use of IMRT for rectal cancer remains a source of ongoing controversy in the radiation oncology community 28 , 29 and recent guidelines do not recommend it given mixed efficacy data 30 , 31 , 32 despite dosimetric studies showing reduced dose to small bowel, bladder, and bone marrow. 33 , 34 , 35 , 36 Similarly, we found that surgeon training and surgical modality varied across geography, with rural and small town patients least likely to receive care from subspecialty-trained surgeons or minimally invasive surgical approaches, trends observed in other disease settings.…”