Private Practice Radiologist Subspecialty Classification Using Medicare Claims.

Area of Science:

• Radiology
• Health Services Research
• Medical Economics

Background:

• Accurate subspecialty classification of radiologists is essential for performance measurement and reimbursement.
• Existing Medicare provider codes have limitations in reflecting current radiology practice patterns.
• Private practice radiologists often have diverse practice profiles not captured by traditional coding.

Purpose of the Study:

• To evaluate the accuracy of current Medicare provider codes for subspecialty classification of private practice radiologists.
• To assess a novel claims-based system using relative value units (wRVUs) for subspecialty identification.

Main Methods:

• Analysis of self-reported subspecialties from 1,476 radiologists across 100 large US private practices.
• Comparison of existing Medicare codes for nuclear medicine and interventional radiology with self-reported data.
• Application of the Neiman Imaging Types of Service (NITOS) classification system using Medicare claims data (2012-2014) to determine wRVU distribution by subspecialty.

Main Results:

• Existing Medicare codes correctly identified only 8.0% (nuclear medicine) to 10.7% (interventional radiology) of radiologists.
• The NITOS-based system assigned a median of 51.9% of wRVUs to self-identified subspecialties.
• A 50% wRVU threshold accurately categorized 48.8% of radiologists, but 48.3% had varied practice patterns unsuitable for single subspecialty assignment.

Conclusions:

• Current Medicare provider codes are inadequate for accurately reflecting radiologists' subspecialties.
• Claims-based data, particularly wRVUs, provide a more granular and accurate method for subspecialty identification.
• This claims-based approach shows potential for transparent and reproducible matching of radiologists to practice-relevant metrics in evolving payment models.