Secondary Logo

Journal Logo

Are We Paying Our Housestaff Fairly?

Morris, Jon B. MD*; Burke, Sean V. JD; Berns, Jeffrey S. MD*

doi: 10.1097/SLA.0000000000000703
Surgical Perspectives

Trainees in US graduate medical education training programs are compensated on the basis of postgraduate year of training, not hours worked. This creates an inherently unfair compensation structure, with large discrepancies in hourly pay rates. We propose restructuring of salaries to promote fundamental fairness in how housestaff are compensated.

*Perelman School of Medicine

University of Pennsylvania, Philadelphia, PA.

Reprints: Jon B. Morris, MD, Department of Surgery, Perelman School of Medicine, University of Pennsylvania, 3400 Spruce St, 4 Maloney, Philadelphia, PA 19104. E-mail:

Disclosure: The authors declare there are no conflicts of interest and that no funding was received in support of this work.

Imagine 2 recent college graduates starting their first jobs in 2 different departments for the same company. One works Monday through Friday from 8:30 AM to about 5 PM, with little or no night or weekend responsibilities. The other works much longer hours, typically from 6:30 AM until 7 or 8 PM. On some nights, she may be called back to work from home. She works 1 to 2 weekends per month. For both, responsibility and work hours will remain substantially unchanged over the subsequent few years of employment. The company's policy specifies that both be paid the exact same amount for each year of employment, with only a small (and identical) annual pay increase. Neither is eligible for a bonus or incentive pay.

Does this compensation policy sound fair? Of course not. But this is precisely how interns, residents, and fellows in US graduate medical education (GME) training programs are compensated on the basis of postgraduate year (PGY) of training, not hours worked, regardless of specialty or subspecialty.1 We believe fairness compels us to consider changing this approach to housestaff compensation. Tables 1 and 2 provide data available from the American Medical Association from 2012 regarding average hours on duty per week for trainees in their first year of training in selected core residency and subspecialty fellowship training programs.2 It is clear that there are substantial discrepancies in work hours across different programs, which, given our current practice of paying all trainees at any given PGY level the same annual salary, results in similarly substantial discrepancies in hourly pay rates. Assuming a PGY-1 salary of $52,000 per year ($1083 per week, assuming 4 weeks of vacation), the hourly pay rate for the listed residency programs ranges from $23.85 for Dermatology residents to $14.54 for General Surgery residents (Table 1). Likewise, for the fellowship programs shown in Table 2, assuming an annual salary of $58,000 ($1208 per week), hourly pay ranges from $25.76 for Dermatopathology and Sleep Medicine fellows to $17.85 for Vascular Surgery and Surgical Critical Care fellows. Similar discrepancies were reported 25 years ago.3





How might we remedy these discrepancies? One approach would be to pay each intern, resident, and fellow a PGY-based hourly rate for actual hours worked (up to 80 hours per week, the ACGME duty-hour regulation maximum). Such an approach would of course require that house staff be “on the clock” like many other employees, clocking in at the start of a duty period and out at the end. Although this approach might help with enforcement of duty-hour regulations, a primary obstacle to this proposal is that if housestaff are paid on an hourly basis, they would not be exempt from federal overtime law. Any hours worked in excess of 40 in a given week would have to be paid at 1.5 times the regular hourly rate. Furthermore, we have found that housestaff and program directors are very resistant to the use of a time clock for monitoring resident work hours.

One alternative would be to use regional or national benchmarking data on usual hours worked per week in each different training program to set weekly salaries based on an effective hourly rate adjusted to each subsequent PGY level. This model might entail, for instance, a PGY-1 effective hourly rate of $20 such that a PGY-1 Dermatology resident working 48 weeks would be paid an annual salary of $43,584 whereas a PGY-1 General Surgery resident would be paid an annual salary of $71,520. A simpler approach, which we favor, would be for institutions to compensate trainees at 1 of 3 or more different levels based on benchmark data for approximate hours worked in each specialty or subspecialty area. This could be accomplished by setting a specific salary for interns and residents in programs with a typical average workweek of less than 50 hours, 50 to 65 hours, and greater than 65 hours, with stepwise salary increases for each successive PGY level within the program. Alternatively, an institution might examine work hours of its own trainees, identify those programs whose trainee work hours below some threshold, perhaps the bottom tertile, and use these dollars to increase compensation for those trainees in programs whose duty hours worked are at the upper tertile. These need not be equivalent transfers, in that salary from the bottom 20 percentile programs, for instance, could be used to increase salaries of trainees in only the programs in the top 10 percentile depending on individual institutional preferences. We would also favor providing an additional hourly stipend to those who take home call, which is currently generally uncompensated and is allowable under federal labor law without creating an overtime issue.

Implementing a new salary structure for housestaff would not be simple and could be costly unless some salaries were reduced to offset other increases. The financial impact of a salary structure such as we describe for a training institution would depend on the hourly rate paid for each level of training, actual or assumed hours worked, number of housestaff at each PGY training level, and the mix of residency and fellowship programs that each institution supports. Given increasing pressures to reduce GME support from federal sources, new and innovative revenue sources must be considered such as private insurers. Barring this, any increased costs of a new salary structure would need to be borne by the hospitals and/or clinical programs that benefit from the work efforts of the housestaff. As discussed earlier, however, cost-neutral solutions to these salary disparities are possible. Implementation of any new salary structure for residents and fellows would probably need to be broadly implemented across all training programs to avoid huge discrepancies among programs using different compensation formulas. A benefit of a new compensation system such as we propose is that it would likely reduce the desire of residents and fellows in programs with longer working hours to seek moonlighting opportunities to supplement their GME-based salaries.

Although housestaff are in some respect students, it no longer is the case that their compensation can be considered a student's stipend, intended only to cover living expenses rather than serve as compensation for work. In 2005, the Internal Revenue Service published new regulations that made clear that housestaff compensation is taxable as wages.4 Since that time, under federal law, housestaff pay has been no different from the pay provided to any other hospital employee and thus ought to be subject to the same fairness considerations noted in determining other employees’ salaries.

We can anticipate several objections to our proposals, none of which negates the need to remedy an underlying fundamentally unfair compensation structure. First, one might argue that housestaff choosing a training program that requires long work hours “knew what they were getting themselves in for.” Although almost certainly true, this sort of consideration is not one that we allow to override considerations of fundamental fairness. The second objection one might make is that our fairness concerns are addressed by disparities in future earning potential. In our view, this too misses the point, as it would hardly be considered just to grossly underpay an orthopedic surgery resident, for example, simply because we believed that he or she would eventually earn more after completion of training than a pediatrician. A third objection might be that a tiered salary structure could be perceived as creating a hierarchy of value among housestaff, undermining a culture of teamwork and morale. Explaining to housestaff that the purpose is to create equitable hourly rates, regardless of specialty or subspecialty, should address this concern. Finally, one might object that remedying inequity in housestaff compensation could further diminish the appeal of certain specialties experiencing shortages. It is unlikely, however, that anyone would select a training program based on his or her housestaff salary.

In summary, current resident and fellow salaries, although arguably too low across the board given the years of training and long hours worked by many housestaff, are structured in a way that is inherently unfair. Some residents and fellows are paid a much higher effective hourly rate than others, with PGY level rather than hours worked being the primary driver of salary levels. This leads to advanced residents and fellows in some programs being paid less on an hourly basis than interns and junior residents in others. Given that it is unlikely that the differences in hours worked per week will diminish, we propose restructuring of salaries to promote fundamental fairness, as we are left with the fundamental concern that our current housestaff compensation approach is grossly unfair and needs to be changed.

Back to Top | Article Outline


1. Association of American Medical Colleges. AAMC Survey of Resident/Fellow Stipends and Benefits. October 2012. Available at Accessed March 5, 2014.
2. American Medical Association. Available at Acce-ssed November 16, 2013.
3. Silbderger AB, Thran SL, Marder WD. The changing environment of resident physicians. Health Aff (Millwood). 1988;7:121–133.
4. Internal Revenue Service. Student FICA Exception, 69 FR 76404, Example 4. Available at Accessed March 5, 2014.
© 2014 by Lippincott Williams & Wilkins.