Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Characteristics of Primary Care Office Visits to Physician Assistants, Nurse Practitioners, and Physicians in the VA , 2005-2010 Perri Morgan1 •David H. Abbott2• Rebecca McNeill2 • Deborah A. Fisher2,3 1Duke Department of Community and Family Medicine 2 Durham VA HSRD • 3 Duke Department of Medicine We examine whether primary care provider types are occupying patient care niches within the VA. We also examine the assumption that PAs and NPs see patients who are less medically complex than those seen by physicians. METHODS RESEARCH QUESTIONS 2. How do patient and patient encounter characteristics vary among provider types? RESULTS Primary Care Visits by Provider Type, 2005-2010 80 20 45 18 40 16 35 30 25 PAs 20 NPs 15 PAs + NPs 10 5 0 Variable PA NP Mean age (years) 62.8 61.1 61.7 Male (%) 93.4 93.3 90.0 21.0 20.9 20.4 African American 5.8 5.0 6.0 Hispanic 2/7 1.1 1.8 Other 0.3 0.2 0.3 70.1 72.8 71.7 92.4 85.7 90.3 Unscheduled visit 4.2 5.3 4.5 Physical exam to determine compensation and pension eligibility 3.4 9.0 5.2 50 Physicians Physician assistants Nurse practitioners Missing Purpose of visit (%) 10 Scheduled clinic visit 0 2005 2006 2007 2008 2009 2010 PAs and NPs attend 28-30% of VA primary care visits, with NPs attending twice as many visits as PAs. 10 8 Physicians 6 PAs NPs 4 0 Physician 60 30 12 Use of PAs and NPs in varies widely by region. Inter-facility variation may also be large. White 40 14 2 1 2 3 4 5 6 7 8 9 10 11 12 15 16 17 18 19 20 21 22 23 Veterans Integrated Service Network (VISN) number Race/ethnicity (%) 70 Comparison of primary encounter diagnoses, 2010 50 Comparison of VA primary care encounter characteristics, 2010 20 This project uses national data from the Veterans Administration to compare characteristics of primary care patients and patient encounters attended by PAs, NPs or physicians . % of Primary Care Visits Variation in Number of VA Primary Care Visits to PAs and NPs by Region, 2010 1. What are the trends in use of PAs and NPs in VA primary care between 2005 and 2010? % of primary care visits The U.S. Department of Veterans Affairs has been a frontrunner in utilization of physician assistants (PAs) and nurse practitioners (NPs). Evaluation of the roles and impact of PAs and NPs in the VA is critical to ensuring optimal care for American veterans and may inform best practices for use of PAs and NPs in other settings. % of primary visit diagnoses BACKGROUND Comparison of primary visit procedure codes, 2010 Diagnosis categories were created from the encounter primary diagnosis ICD-9 codes. Diagnoses present in >.02% of visits are categorized above. Nearly 30% of visits have ICD-9 diagnoses present in <.02% of visits. These will be categorized in future work, and could change to proportions above slightly.. *Symptoms without diagnoses include headache, dizziness, chest pain, edema, abdominal pain, convulsions, and shortness of breath. FINDINGS 1. A substantial proportion (28-30%) of VA primary care encounters are attended by PAs and NPs. This proportion was stable from 2005-2010. 2. Nurse practitioners attend approximately twice as many VA primary care encounters as PAs. 3. Most VA primary care encounters record only one provider type. 40 Physician assistant Nurse practitioner % of visits with physician listed as a second provider 0.75 0.52 30 25 5. Characteristics of patients and encounters are similar for physicians, PAs, and NPs in VA primary care clinics. 20 15 10 PAs 5 NPs Established patient E/M* codes New patient E/M* codes Other Preventive medicine Lifestyle counselling Disability evaluations Immunizations Comprehesive mod-hi severity Mod-hi severity problem Moderate severity problem Low-moderate severity problem Comprehesive mod-hi severity Mod-hi severity problem Low-moderate severity problem 0 Minor problem Very few (<1%) visits to NPs and PAs list a physician as a second provider. It is possible that physicians were involved in some of these visits, but not coded as providers. Physicians Minimal problem • Veterans Administration electronic medical record data from 2005-2010 • 11 million primary care encounters in 2010 • Physicians: 7,700,000 • NPs: 2,000,000 • PAs: 890,000 Provider type % of visits DATA SOURCE 4. Use of PAs and NPs in VA primary care varies widely by region. 35 Other Categories were created from encounter CPT codes. *E/M= Evaluation and management. IMPLICATIONS 1. VA data is a promising source for research on care by PAs and NPs. 2. Other sources show recent growth in VA employment of PAs and NPs. Since we did not find an increase in numbers of primary care PA and NP visits, these new employees may be working in specialty care. 3. PAs , NPs, and physicians seem to fill similar roles in VA primary care clinics. Additional research is needed to examine whether this is the most effective and efficient division of labor. FUTURE WORK • Person-level analysis comparing patient characteristics by assigned primary care provider type. • Analysis of primary care team composition and division of labor. • Comparison of patient-level continuity of care, resource use, and health outcomes by assigned primary care provider type. 6. PAs perform more general medical exams for purposes of determining disability than do physicians or NPs. Contact information: [email protected]