Comparison of nurse practitioner and family physician relative work values

With the enactment of the Balanced Budget Act of 1997, American nurse practitioners were granted direct Medicare reimbursement for Part B services. Payment structures in fee-for-service and managed care systems are physician-based, leading to difficulties in constructing payments for other health care professionals. The purpose of this pilot study was to examine the feasibility of using nurse practitioner data for specifying relative work values in the Medicare Fee Schedule for three office-visit codes.

An exploratory survey was designed to establish relative work values using magnitude-estimation scaling. Nurse practitioners (N = 43) responded to a structured questionnaire in a national mail survey. Physician data (N = 46) were obtained from a computerized database from the American Academy of Family Physicians.

The methods used in this study were the same as the process used by the American Medical Association and the Health Care Financing Administration to establish relative work values in the Medicare Fee Schedule. Respondents established relative work values for three Current Procedural Terminology (CPT) codes for office visits (99203, 99213, 99215) commonly billed in primary care practice. Each CPT code descriptor and associated vignette were compared with reference services germane to the practice of nurse practitioners and family physicians, using magnitude-estimation scaling. To establish relative work values for each code, respondents were asked to consider the time to provide the service and intensity of the work involved for each CPT code.

No significant differences between nurse practitioners and family physicians were found in the three CPT codes for relative work values and intensity. Nurse practitioners estimated significantly (p < .01) higher intraservice (face to face) time with patients than did family physicians, and family physicians estimated significantly (p < .05) higher pre-service time for two codes and significantly (p < .05) higher postservice times for three codes.

Nurse practitioner relative work values did not differ significantly from family physician relative work values. Although the sample sizes were small, the significance of the findings support the need for further research with large data sets and additional CPT codes. Such studies could then be used as a basis for decisions about Medicare payment and public policy.