I really shouldn't get into this discussion.
Ah well, here goes:
A. What are the costs of training residents?
Hospitals have 3 costs of residents:
1) Health insurance
2) Malpractice coverage
3) Salary
This is not complete. Additional direct costs include:
1. Portions of the PD and APD's salaries/benefits.
2. Program administrator salaries/benefits.
3. ACGME fees, Match fees, ERAS fees, ITE fees, etc.
4. Recruiting costs, including all those dinners.
5. GME office costs (shared between all the residency programs)
6. Infrastructure costs -- computers for residents, pagers, scrubs, etc.
However, there are many INDIRECT cost savings associated with a residency program, noted below.
B. How much money do programs gain from training residents?
Programs get resident income from two sources -- directly from Medicare as DME and IME, and by billing patient services directly to insurers.
The amount of money programs get from Medicare is quite variable. An old but informative article about this is
here. As this is an NEJM article, you may need a subscription to see it. I have attached the relevant table to this message. Ignore the first column -- this is the amount of money a residency program "would" get if other payers paid like Medicare does. They don't, so what programs actually get per resident is in the last column. The study is obviously dated, but overall the amount of money from GME has decreased since the passage of the BBA of 1997, not increased.
Yet, this is not the whole story. This study only looks at DME. DME flows directly to the sponsor of the residency program (could be a medical school, hospital, or other health care organization). IME flows directly to hospitals where residents deliver care. There is a very complex and insane formula for how much IME a hospital gets. In general, the total IME spending is double the DME spending, so you can multiply the number in the last column by 3 and get a reasonable estimate of how much money hospitals get for training residents. You can see that a small number of programs, many in NYC, make a tremendous amount. Most other programs do not.
Hospitals also obtain income by billing for resident services. This is not true additional income -- inpatient services are all billed by faculty, and residents who see patients in the outpatient setting are supervised by a faculty member who usually sees less patients -- at our institution, our calculations show that a faculty member seeing patients on their own generates the same $$ as one who works with 3.5 residents (mixture of PGY levels). I should add that this is not true for fellows, who often see patients semi-autonomously and hence can add to billing.
So, for most programs, the amount of DIRECT money received for teaching residents is about equal to the DIRECT costs of training residents. Over time, the DIRECT costs of training residents have increased, and the DME/IME have decreased, such that now it is usually a break even or small loss for programs.
C. What about indirect costs?
Although the direct costs of resident training is usually equal to or more than the direct income obtained, there is a tremendous costs savings to the system in general. This is simply because many hospitals rely on residents for their night coverage. Replacing residents with other resources is extremely expensive -- many NP/PA's are unwilling to work at night, or do not have the clinical skills to do so. Faculty coverage at night, either with moonlighters or hospitalists, is enormously expensive and does not generate any additional income for the system in general. Hence, in any discussion of the financial impact of residents, the cost of replacing the service that those residents provide needs to be considered. However, when hosiptals look at their residency programs, these "cost savings" are invisible.
D. So will residents get paid more in the future?
This is a difficult question. I have stated, multiple times in the forum, that I think that residents deserve a higher salary than is currently offered (or perhaps loan forgiveness, which is essentially the same thing). Residency is not a free market, and probably shouldn't be one. There are many people out there willing to be a resident for free -- a true free market could actually decrease salaries.
In addition, virtually all programs have started some sort of uncovered service to address increased census and workload, duty hour restrictions, team caps, etc. As programs start these uncovered services, which by the IM-RRC rules must be uncovered 24/7, the cost savings associated with residency programs is disappearing. As programs start other overnight coverage plans, residency programs are potentially going to become true money losers. It will not surprise me at all if we see many programs start to shrink their residency programs in the next few years, especially when further GME funding cuts are necessary to pay for the exploding costs of Medicare Part D. Hospitals are now "capped" for the total number of residents in training -- if you go over your cap, you get no additional funding. In addition, CMS is looking closely at what residents do, and some payors are refusing to reimburse for resident time spent in conference, research projects, etc. Hospitals will be tempted to decrease the size of "loser" programs (like IM) and increase the size of money "winner" programs (such as ortho) -- more ortho residents = more active OR time (in general) = more joint replacements = more $$ (at least for now, although the rules will change).