A Claims-Based Method for Identification and Characterization of Practicing Interventional Radiologists.

PURPOSE: Interventional radiologists (IRs) are inconsistently defined in health services research and underrepresented by specialty designation in claims data. This work proposes a research method for identifying 'Practicing IRs' using two national claims datasets.

MATERIALS AND METHODS: 2015-2019 100% Medicare Part B data and 2015-2019 private insurance claims from Optum's Clinformatics® Data Mart (CDM) Database were used to rank-order radiologists' IR-related work as a percent of total-billed work Relative Value Units (wRVUs). Characteristics were analyzed at various threshold percentages. External validation used Medicare self-designated specialty with Society of Interventional Radiology membership records; Youden's Index evaluated sensitivity and specificity. Multivariate logistic regression assessed Practicing IR characteristics.

RESULTS: In Medicare data, above a 10% IR-work threshold, only 23.8% of selected Practicing IRs were designated as IRs in the Medicare data; above 50% and 90% thresholds, this percentage increased to 42.0% and 47.5%, respectively. The average percentage of IR-related work among Practicing IRs was 45% , 84% and 96% of total wRVUs for the 10%, 50% and 90% thresholds, respectively. At these thresholds, the CDM Practicing IRs included 21.2%, 35.2% and 38.4% designated IRs and E/M comprised relatively more total wRVUs. Practicing IRs were more likely to be male, metropolitan, and earlier in their careers than other radiologists, at all thresholds.

CONCLUSION: Most radiologists performing IR-related work are designated in claims data as DRs, indicating insufficiency of specialty designation for IR identification. The proposed method to identify Practicing IRs by percent IR-related work effort could improve generalizability and comparability across claims-based IR studies.