Search | Search by Center | Search by Source | Keywords in Title
Huang PY, Yano EM, Lee ML, Chang BL, Rubenstein LV. The Impact of Provider Mix on the Quality of Primary Care. Paper presented at: VA HSR&D National Meeting; 2003 Feb 13; Washington, DC.
Objectives: Since 1996, VA policy has promoted the increased utilization of nurse practitioners (NPs) and physician assistants (PAs) in primary care (PC), and quality indicators of chronic disease care and preventive care and patients' ratings have steadily improved. No studies have evaluated whether practice outcomes specifically improved in those VA PC practices that employed more nonphysician clinicians (NP/PAs) in their provider mix. This study examines the impact of increasing the proportion of NP/PAs-per-physician on the quality of PC. Methods: We surveyed the PC practice leaders of all 170 VA medical centers (VAMCs) in 1999 (94% response rate) to assess VAMC and PC practice characteristics. Survey data for each VAMC was linked to 1999 PC practice outcomes as measured by: 1) Prevention Index (PI) and Chronic Disease Index (CDI), from the External Peer Review Program, and 2) Patients' ratings from the VHA National Ambulatory Care Survey. The proportion of NP/PAs-per-physician for each facility was calculated by dividing the number of PC NP/PAs by the number of PC physicians (MDs). We performed multivariate regression to predict the influence of the NP/PA-per-MD proportion on practice outcomes, adjusting for facility and PC practice characteristics. Results: On average, PC practices employed 1 NP/PA per 2 MDs. Academic VAMCs, practices with NP/PA trainees, or with greater reliance on managed care-oriented provider education programs employed higher proportions of NP/PAs-per-MD. After controlling for facility and practice characteristics, a higher proportion of NP/PAs-per-MD remained independently associated with worse PI and CDI scores (p < 0.05). Conclusions: Practices with higher proportions of NP/PAs-per-MD performed worse than practices with lower proportions on measures of preventive and chronic disease care. Future research should investigate whether these worse outcomes in practices with higher proportions of NP/PAs reflect performance specifically by NP/PAs, MDs, or by both types of clinicians. Impact: Research on NP/PAs has focused on care of individual patients in self-selected practices, rather than the impacts of PC staffing as it occurs naturalistically across the country. Our results suggest that PC practices cannot be assured of improved preventive care merely by employing more NP/PAs. Without increased understanding of the optimal roles, relationships, and responsibilities of different types of PC clinicians, the potential benefit of NP/PAs may not be realized.