The specialty distribution of physician assistants (PAs) has varied over time. Initially, most PAs worked in primary care specialties, but over time, PAs have adapted to roles in virtually every practice specialty.1-3 One factor that has affected specialty choices for PAs has been the availability of jobs.
The practice specialty distribution in the PA job market is of interest to a variety of stakeholders. PA educators are responsible for ensuring that graduates are prepared to work in the fields that await them. Policy makers may want to influence PA employment opportunities to meet policy goals, such as increasing the number of clinicians practicing in primary care.4 And new graduates as well as practicing PAs considering a change in employment are eager for information about the types of jobs available.
In this project, we provide an overview of 2014 job postings at the national level by practice specialty.
National data on 2014 job postings for PAs were obtained from Burning Glass Technologies, a leading labor analytics firm. Burning Glass Technologies collects information from online job postings by monitoring more than 38,000 websites. Data for specific variables, such as job title, location, and employer, are extracted from each posting. Using these variables, Burning Glass Technologies eliminates duplicate postings (e-mail communication with Bledi Taska, Chief Economist, Burning Glass Technologies, August 26, 2015).
We examined 42,768 job postings that advertised PA jobs and excluded postings that advertised nonclinical jobs (N = 336), postings for jobs in Guam and Puerto Rico (5), and postings for which we could not identify a specialty (N = 8,290), resulting in a final sample of 34,137 PA job postings. Trained coders in our offices read each posting and assigned specific practice specialties to each posting. Additional details concerning Burning Glass Technologies' data collection and processing procedures, data validity issues, and our work preparing the data have been published elsewhere.5
For this project, we combined specialties into meaningful categories as shown in Figure 1 (primary care, adult surgical specialties, adult medical subspecialties, emergency/urgent care, and other) and tabulated the number and percentage of job postings in each category. In the primary care category, we included family medicine, general internal medicine, general pediatrics, general geriatrics, women's health, and primary care postings. We included emergency medicine, urgent care, and positions that include emergency medicine and urgent care in the postings in the emergency category. The other category included a number of setting-specific specialties (occupational medicine, retail clinics, correctional medicine, military medicine, long-term care facilities, palliative care, and student health) as well as the pediatric medical and pediatric surgical subspecialties and pathology and public health. Within each category, we identified the specialties with the most numerous job postings.
The largest proportions of job postings were in the surgical (28%) and medical subspecialty (23%) categories followed by primary care (19%), other (18%), and emergency/urgent care (12%) (Figure 1). Among the surgical specialties, almost one-third (2,999 postings, 31% of surgical postings) were for orthopedic positions, followed by general surgery (1,796 postings, 19%) and cardiothoracic surgery (1,501 postings, 16%) (Figure 2). In the medical subspecialties, the most numerous postings were for PAs to work in cardiology (1,108 postings, or 14% of medical subspecialty postings), followed by hospital medicine and psychiatry (1,064 and 1,051, respectively, or 13% each), and pain management and hematology/oncology (884 and 848, or 11% each) (Figure 3). In the emergency category, there were slightly more postings (2,152 postings, or 54% of emergency medicine/urgent care postings) for emergency medicine than for urgent care (1,721 postings, or 44%), and only 2% that listed both emergency medicine and urgent care (90 postings).
Labor market demand for PAs, as measured by job postings, was strong in 2014, with about one job posting for every three certified PAs. With an estimated 7,556 graduates in 2014, there were about five jobs per graduate.6 However, PAs already in practice competed with new graduates for these jobs. A precise estimate of the number of PAs who change employment positions is not available, but the National Commission on Certification of Physician Assistants reports that 7.4% of the 101,977 PAs in practice in 2014 (about 7,546 PAs) planned to leave their jobs in the next year.7 About 36% of postings were described in the advertisement as available to either an NP or a PA, and NPs are more plentiful nationally than PAs. Geographic variation in the number of jobs available also could pose barriers to employment in some areas.5
The specialty distribution of job postings differs somewhat from that of PAs in clinical practice. Comparing the proportion of PAs in a specialty with the proportion of job postings by specialty can give insight into which specialties might be growth areas for PAs. Many job postings compared with existing jobs may indicate that the job market is strong in that specialty. As we reported previously, medical and surgical specialties have proportionately more postings compared with existing filled positions, unlike primary care, suggesting that subspecialty practice has higher labor market demand for PAs than does primary care.5 Our current analysis allows similar comparisons of the relative labor market demand for PAs in specific specialties. For example, perhaps the smaller proportion of job postings in dermatology (1.4%) compared with existing jobs reported by NCCPA (4.4%) suggests that the labor demand for PAs in dermatology might be proportionately less than in psychiatry or occupational medicine, where the proportion of job postings is higher than the proportion of existing filled positions (for psychiatry, 3.1% of Burning Glass Technologies' job postings versus 1.2% filled positions per NCCPA; occupational medicine, 3.7% of postings versus 1.2% of filled positions).
Our findings will be useful to several audiences. Practicing and graduating PAs will be interested in the availability of jobs in specialties in which they might want to work. Trends in the labor market might influence PA educators to add or emphasize certain curriculum topics or skills. And prevalence of PA jobs might influence policies regarding potential expansion or contraction of training programs for PAs.
Because researchers have relatively little experience with analyses of job postings by medical specialty, we are unable to gauge the precision and stability of these annual estimates. This contributes to the difficulty of assessing the importance of small differences between job postings and actual positions. Other limitations of using real-time labor market data have been reported elsewhere, but could include inaccuracies due to jobs that are not posted online (estimated at less than 10%), imperfections in procedures to reduce duplication procedures, and inaccuracies in our inhouse coding of medical specialties.5 Independent analyses of Burning Glass Technologies' data suggest that the first two flaws are minimal.8-10 We sought to minimize specialty coding errors by training coders until 95% agreement was reached between the coder and our trainer.
Our analysis of national job postings data for PAs can help to identify labor market demand for PAs by practice specialty and will interest educators, policy makers, PA students, and practicing PAs. Future research should examine trends in these data. Trend analyses will be useful for multiple purposes, including tracking labor market demand for the profession as a whole in order to address concerns about overproduction of PAs.11
1. Hooker RS, Cawley JF, Asprey DP. Physician Assistants: Policy and Practice
. 3rd ed. Philadelphia, PA: F.A. Davis; 2010.
2. Morgan P, Everett CM, Humeniuk KM, Valentin VL. Physician assistant specialty
choice: distribution, salaries, and comparison with physicians. JAAPA
3. Morgan PA, Hooker RS. Choice of specialties among physician assistants in the United States. Health Aff (Millwood)
4. Physician Assistant
Education Association. Physician Assistant
Education Association 2013 Strategy. http://paeaonline.org/wp-content/uploads/2015/07/PAEA-Strategy.pdf. Accessed October 17, 2017.
5. Morgan P, Himmerick KA, Leach B, et al Scarcity of primary care
positions may divert physician assistants into specialty
practice. Med Care Res Rev
. [e-pub Feb. 4, 2016.]
6. Physician Assistant
Education Association. 30th Annual Report on Physician Assistant
Educational Programs. http://paeaonline.org/wp-content/uploads/2016/12/2015-by-the-numbers-program-report-30.pdf. Accessed October 5, 2017.
7. National Commission on Certification of Physician Assistants. 2014 Statistical Profile of Certified Physician Assistants. http://www.nccpa.net
/news/2014StatisticalReport. Accessed October 5, 2017.
8. Jobs for the Future, New York City Labor Market Information Service. Real-time labor market information: an environmental scan of vendors and workforce
development users. http://www.jff.org
-development. Accessed October 5, 2017.
9. US Department of Labor, Employment and Training Administration, Office of Workforce
Investment. Promising practices in leveraging traditional and real-time labor market information. http://www.gc.cuny.edu
/CUNY_GC/media/365-Images/Uploads%20for%20LMIS/Reports%20and%20Briefs/ETA_LMI_Promising_Practices_Report_20150623.pdf. Accessed October 5, 2017.
10. Carnevale A, Jayasundera T, Repnikov D. Georgetown University Center on Education and the Workforce
. Understanding online jobs data: a technical report. https://cew.georgetown.edu/wp-content/uploads/2014/11/OCLM.Tech_.Web_.pdf. Accessed October 5, 2017.
11. Salsberg E, Quigley L. Are we facing a physician assistant
Keywords:Copyright © 2018 American Academy of Physician Assistants
physician assistant; workforce; job postings; primary care; specialty; subspecialty