CPT coding by interventional radiologists: accuracy and implications.

Abstract

PURPOSE: To evaluate the accuracy of Current Procedural Terminology (CPT) coding for interventional radiology services when coding is performed by the operating physician. MATERIALS AND METHODS: Coding data for 1,174 interventional radiology encounters in 736 patients were analyzed for appropriate use of CPT codes. Physician operators initially assigned provisional codes. Formal coding for billing purposes was performed at a later date by one of two experienced interventional radiology physician coders. Initial operator coding errors and associated relative value unit (RVU) impact were analyzed. The coding patterns of experienced physician coders were compared with those of the other interventionalists. RESULTS: Only 82% of encounters were initially coded correctly, with a small net tendency toward undercoding. The overall net RVU impact of errors was only -1.2%, with the effects of undercoding outweighing those of overcoding. More complex cases (> or =4 CPT codes) were much more likely to be coded erroneously than less complex cases (24% vs 14%, P <.001). Experienced physician coders committed significantly fewer errors than other physicians (10% vs 25%, P <.001), but there was a similar minimal net RVU impact of errors (-1.1% vs -1.4%, P =.198). CONCLUSION: Although initial physician coding errors for interventional radiology procedures are common, the net RVU impact is minimal. The accuracy of experienced physician coders is significantly higher than that for other interventionalists. Because of the regulatory consequences of coding inaccuracies, practices should establish quality improvement systems to minimize errors and use the skills of experienced individuals in their coding processes.