If you prefer to view this article in PDF form, click here. If you prefer to view this article in PDF form, click here.
We write to correct misleading impressions appearing in the December 2006 issue of JAAPA. The news brief “Three years after resident work hour rules, hospital PA employment remains steady” in From the AAPA (page 3) conveys the impression that, based on the AAPA Census figure of 22% of PAs who indicate that their primary employer was a hospital, the number of PAs who may have been hired to substitute for physician residents has “remained virtually the same.” We looked at this issue using a more precise set of measurements also taken from the AAPA Census (Cawley JF, Hooker RS. The effects of resident work hour restrictions on physician assistant hospital utilization. Journal of Physician Assistant Education. 2006;17:41-43). In our analysis, we considered the five Census subsets of PAs employed in the hospital setting (inpatient, ICU/CCU, surgery, outpatient, and emergency department) and tracked trends in the three subsets we believe to best reflect PA substitution for physician residents. These subsets were inpatient, ICU/CCU, and surgery. We felt that outpatient and emergency department settings were unlikely to be affected by the 80-hour work restriction. As seen in the exhibit below, the number and percentages of PAs working in inpatient units and in ICU/CCU settings went up from 2003 to 2006; inversely, the number and percentages of PAs working in surgery went down over the same period.
That also explains why the 22% overall figure remains the same during this period. We interpret these changes to mean that more PAs may indeed have been hired for roles in inpatient medical units and ICU/CCUs, perhaps in response to resident work hour limitations, and that the numbers in surgery went down due to the trend of more surgical procedures being performed in outpatient settings. Simply looking at the overall 22% figure seems to us to be an imprecise measure of whether or not a greater number of PAs are indeed assuming positions in hospital settings as a result of resident work hour limitations. As the anecdotal stories cited in the article suggest, the numbers of PAs in specific hospital units have changed, possibly in response to the GME restrictions. In tracking and analyzing this possibility, we should avoid broad generalizations and examine the data that most accurately reflect these workplace patterns.
James F. Cawley, MPH, PA-C
School of Public Health and Health Services
The George Washington University
Washington, DC
Roderick S, Hooker, PhD, PA-C
Department of Rheumatology
VA Medical Center
Dallas, Texas
In his letter, Mr. Cawley fails to recognize the distinction between “work setting” and “employer type” as collected on the AAPA Census Survey. The article “Three years after resident work hour rules, hospital PA employment remains steady” correctly asserts that the percentage of PAs employed by hospitals, according to the annual AAPA Census, remained steady at about 22%. Mr. Cawley refers to “work setting” only, as he fails to recognize that a substantial number of PAs whose primary work setting is a hospital unit are in fact employed by private practices. As employees of private practices, these PAs are not likely to be filling roles vacated by residents due to the reduction in resident work hours. When we more precisely restrict the analysis to those PAs who are employed by hospitals, the shifts described by Mr. Cawley all but disappear.
For example, the percentage of PAs whose primary employer is a hospital and whose primary work setting is a hospital intensive care or critical care unit (ICU/CCU) was 1.4% in 1999. This percentage stood at 1.5% in 2002 (the year prior to the rule change) and 1.1% in 2003 (the year the resident hour restriction went into effect). In the years after 2003, the percentage of PAs whose primary employer is a hospital and whose primary work setting is an ICU/CCU remained relatively stable at 1.3%, 1.5%, and 1.5% in 2004, 2005, and 2006, respectively.
A similar trend is observed in the inpatient unit. The percentage of PAs whose primary employer is a hospital and whose primary work setting is a hospital inpatient unit (not ICU/CCU) was 5.2% in 1999. Between 1999 and 2003, that percentage fluctuated somewhat from year to year; it stood at 5.0% in 2002 and dipped to 4.4% in 2003 (the year the resident hour restriction went into effect). In the years after 2003, the percentage of PAs whose primary employer is a hospital and whose primary work setting is a hospital inpatient unit remained relatively stable at 5.0%, 5.0%, and 5.3% in 2004, 2005, and 2006, respectively. Thus, I submit that data from the AAPA Census, when used correctly, do not support the assertion that an increasing proportion of PAs are being employed by hospitals either in inpatient units or ICU/CCUs as a result of the resident hour restriction imposed during 2003.
Kevin Kraditor
Director, Data Services and Statistics
American Academy of Physician Assistants
703/836-2272, ext. 3809
kkraditor@aapa.org