Morgan P, Strand J, Albanese M. A national comparison of characteristics of patients and patient visits attended by physician assistants or physicians in office-based care (preliminary report). Presented at: Physician Assistant Education Association Education Forum; October 28, 2006; Quebec City, Quebec.
INTRODUCTION: Rapid growth of the PA profession has been accompanied by an expanding scope of practice, and analysts have noted that PA activities increasingly overlap with those of physicians. One means of examining this trend is to determine whether patients attended by PAs are becoming more similar to those attended by physicians. This project addressed the following research questions: 1. How do patients seen by PAs and physicians compare with regard to demographic, geographic, socioeconomic, insurance, and health status? 2. What are the trends in these characteristics between 1996 and 2003? METHODS: This project uses data from the Medical Expenditure Panel Survey (MEPS) to compute weighted national estimates of patient characteristics and examine trends in the case mix of patients seen by PAs and physicians between 1996 and 2003. RESULTS: Patients who saw PAs were similar to those who saw physicians except that patients of PAs were more likely to be rural. Some small differences found between the groups disappeared over the 8-year period. Analysis of data is preliminary and ongoing. DISCUSSION AND CONCLUSIONS: Findings from this study may alter prevailing assumptions regarding the types of patients attended by PAs.
ON THE WEB Please see the online version of this month's Research Corner at www.jaapa.com for the accompanying Web-only article “Interpreting the methodology.”
Morgan P. Do physician assistants still care for the underserved? (Preliminary report). Presented at: Physician Assistant Education Association Education Forum; October 28, 2006; Quebec City, Quebec.
INTRODUCTION: The PA profession has been touted for extending care to underserved populations. Declining portions of PAs choosing rural and primary care practice have prompted observers to question whether PA practice patterns now mirror the urban and specialty predominance typical of physician practice. METHODS: This is a national comparison of the tendency of PAs and physicians to care for underserved populations. National data from the MEPS and the Area Resource File are used to compare patients seen by physicians or PAs in office-based settings on a variety of measures of underservice, including rurality, race/ethnicity, individual and county socioeconomic status, individual insurance status, county Health Professional Shortage Area designation, and county density of physicians, PAs, and nurse practitioners per unit population. RESULTS: Larger portions of PA than physician patients live in rural and remote areas and in counties with lower practitioner densities. Patients of PAs have a slightly higher socioeconomic status. Analysis of data is preliminary and ongoing. DISCUSSION: This comparison of PA and physician office-based practice indicates that PAs are more likely than physicians to care for underserved patients. On measures that emphasize socioeconomic status, physicians were more likely than PAs to care for underserved patients. This could be explained by a tendency of physicians to care for the urban poor.
Dehn RW. The distribution of physicians, advanced practice nurses, and physician assistants in Iowa.Journal of Physician Assistant Education.2006; 17(1):36-38.
INTRODUCTION: The geographic distribution of PAs, compared to physicians and advanced practice nurses (APNs), is not well understood other than in terms of the overall statewide or national population of providers. This study set out to answer the question of whether Iowa nonphysician clinicians are providing care in rural areas compared to physicians and APNs. METHODS: In 1995 the Iowa legislature mandated the establishment of databases to study and track medical provider distribution. The Iowa data were obtained in November 2005 for physicians, APNs, and PAs. The county-by-county datasets were stratified by population from greatest to least and then grouped into quartiles of equal population. RESULTS: The data demonstrate that a greater proportion of Iowa's physicians practice in the most populous quartiles of counties, while PAs and APNs are more prevalent in the least dense quartiles, consisting of the most medically underserved counties. APNs fall somewhere between physicians and PAs, ie, practicing in the most populous counties at a lesser rate than physicians but at a greater rate than PAs, and practicing in the least populous counties at a greater rate than physicians but at a lesser rate than PAs. CONCLUSION: This data currently reveal that in Iowa, a greater proportion of PAs practice in more sparsely populated rural geographic areas when compared to physicians and APNs.
Orcutt VL. A supply and demand model for the physician assistant profession. Presented at: Physician Assistant Education Association Education Forum; October 28, 2006; Quebec City, Quebec.
INTRODUCTION: Interest concerning the contribution of the PA profession in the delivery of health care in the United States is increasing as debates on physician workforce supply and demand intensify. Knowledge of current PA scope of practice and other factors affecting the supply of and demand for PAs will be necessary to assist in the accurate prediction of 21st-century health care workforce demands. METHODS: Review of state PA practice regulations were analyzed for credential, prescriptive authority, and potential limitations as compared to 1992 data. A supply model was developed that includes current PAs eligible for practice as a baseline measure, and then adjusted for retirement and retention rates. PAEA database new graduate rates were added to the pool and adjusted. The demand model reflects US Census Bureau population estimates, specialty distributions, and GDP estimates. Supply and demand equations were analyzed using ordinary least squares estimation techniques for trend analyses. RESULTS AND CONCLUSIONS: Demand models were constructed utilizing no growth in supply, 10% growth in supply, and 25% growth in supply. In all three models, demand for PAs exceeded supply; therefore utilizing this model, even a 25% growth in PA supply would be inadequate to address future workforce demand for PAs.
Morgan and colleagues reported on a preliminary analysis of data from the Medical Expenditure Panel Survey (MEPS), a massive database that surveys patient experience with their medical encounters. Although this database likely undercounts nonphysician care to the extent that it may not be quantitatively nationally representative, it does provide data on more than 6,500 office-based visits that are attributed to PAs. These visits are likely qualitatively representative of national PA patient care because they are diverse with regard to patient age, race, socioeconomic status, region of residence, rural-urban status, and insurance status. Morgan compares these visits to those attributed to physicians to show that PAs generally have patient populations, diagnoses, and visit encounters that are similar to those of physicians and that, over time, the patient encounters of PAs have gradually become even more similar to those of physicians. Morgan and colleagues have documented the undercounting of PAs in the MEPS; thus if this otherwise valuable data source is used to produce national estimates of numbers of office visits to PAs, the results may underrepresent PA patient care activities.1
The second Morgan presentation reported on a preliminary analysis of the same data set to show that PAs are more likely to see patients in rural and underserved geographical regions than are physicians, which addresses the historical question of whether PAs fill in gaps in the health care system.
Dehn reported medical practitioner data for physicians, APNs, and PAs from a statewide Iowa census and analyzed it for geographic practice location. The data demonstrated that in 2005, Iowa PAs were more likely than both physicians and APNs to practice in the most rural counties.
The presentation by Orcutt described a demand model for PAs based on the physician supply model designed by Cooper and colleagues,2 where future practitioner demand was projected by estimated economic and population growth. Using this model, Orcutt concluded that the demand for PAs will outpace supply through 2025, even if growth in the PA profession increases 25% above current levels (growth in the PA profession is projected to increase 10.9% over the next 5 years3).
These abstracts present evidence that the PA profession has evolved to provide patient services similar to those provided by physicians in outpatient clinical settings. Data provide evidence that PAs are more likely to locate in rural and underserved areas than are physicians and APNs. Finally, an economic growth model previously used to predict physician supply and demand suggests that the demand for PAs will likely exceed supply in the foreseeable future.
1. Morgan PA, Strand J, Østbye T, Albanese MA. Missing in action: care by physician assistants and nurse practitioners in national health surveys [published online ahead of print February 16, 2007]. Health Serv Res. doi:10.1111/j.1475-6773.2007.00700.x
2. Cooper RA, Getzen TE, McKee HJ, Laud P. Economic and demographic trends signal an impending physician shortage. Health Aff (Millwood). 2002;21(1):140-154.
3. Physician Assistant Education Association. PA program enrollment trends, preliminary data. Presented at: Physician Assistant Education Association Educational Forum; October 26, 2006; Quebec City, Quebec.