Every so often, real life intrudes on blogging. This is one of those times. So enjoy this bit of Classic Insolence from back in April 2007 and be assured that I'll be back tomorrow. Remember, if you've been reading less than three years, it's new to you, and, even if you have been reading more than three years, it's fun to see how posts like this have aged.
From fellow ScienceBlogger Abel, I'm made aware of an excellent post on the Health Care Renewal Blog about the financial reality of being an academic physician in a modern U.S. medical school. It's an excellent overview of how medical schools view clinical faculty as, in essence, cash cows that have to bring in the cash and pay for themselves. The same thing actually applies to basic scientists as well, and I do have to quibble a bit with the internal medicine-centered view of the various ways that faculty are expected to bring in money, which do not necessarily apply to academic surgery departments (more about both quibbles below), but overall it's a great primer about what it is like to do research and teach in medical schools circa 2007.
First, let's set the stage, with the definitions of four types of faculty:
Dr Goldman is known for saying it like it is. In the interview, he explains with brutal honesty how things now work in academic medicine, and thus explains the apparent paradox of medical schools that do not pay their teachers to teach.
[Dr Goldman was asked to talk about academic careers....] I'd add something about the cold, hard facts of Academic Economics 101. There are four categories of faculty: 1) 'Taxpayers' who generate more than they cost and help fuel the academic mission; 2) 'Hired workers' who get paid to do a job that many people might like to do; 3) 'Loss leaders' who get short-term investments in the expectation that they will become successful 'taxpayers;' and 4) 'Welfare recipients' - faculty with more tenuous status.
Bottom line, you should strive to be a 'taxpayer.' If you're a 'hired worker,' you should strive to be better than the others who would like your job.
Dr Goldman reveals that in the typical medical school, the most important criterion for faculty success if generation of external funding, that is, generation of fees for clinical work, or of grants from external sponsors. Whether a faculty member is good at patient care, teaching, or research, or whether he or she upholds the highest professional standards, is secondary.
At first glance, to someone outside of academic medicine, this seems nonsensical. The incentive system described by Dr Goldman seems to be like the commission system used to reward some automobile sales people (at the smarmier dealerships). The system seems utterly different from that used in other parts of "higher" education, in which faculty are usually paid straight salaries based on rank and seniority.
The system used in academic health care may be nonsensical. It is too bad those of us in or recently departed from this arena have gotten so used to it that it seems normal.
Dr. Goldman and Dr. Poses (who wrote the article) are mostly correct. Before I explain why, let me explain to you the reality of what is expected of faculty in academic medical centers. First, I'll start with basic science faculty. One of the biggest advantages of being basic science faculty in a medical school is that usually the teaching load is considerably less than it is for faculty in basic science departments, allowing for more time for research, but this advantage comes at a high price: near total dependence on external grant funding, particularly NIH or other government funding. In essence, faculty doing biomedical research at such universities are indeed treated like free lance used car salesmen. The university provides them with a "startup package" of salary, lab space, and a certain amount of startup funds to get the lab started, and in return expect that the faculty will, within three to four years, successfully compete for external grant funding to keep his lab going and pay his expenses. Most grants pay salary support commensurate with the amount of effort devoted to the project (previously a percentage of the researcher's time, now in more recent grant applications expressed as the number of person-months in a typical year devoted to the project). The more grant support a researcher achieves, the less a university has to pay of that researcher's salary. Indeed, most departments expect researchers to have grant support sufficient to pay at least 50% of their salaries, some 75% or even higher. Free-lancers, indeed, selling their research ideas to granting agencies. If a young researcher does not pay a high proportion of his salary through grants and show the potential of being able to continue to do so indefinitely, come tenure decision time he will be in a world of hurt.
But universities have it even better than that. For federal grants, the Holy Grail for which we all strive, the government sweetens the pot. Universities negotiate with the federal government for what is called an "indirect cost" agreement. Indirect costs are costs associated with keeping a university going; i.e., building maintenance, electricity, secretarial support, new construction, etc. My university, for instance, has a negotiated indirect rate of 55.5%. This means that, for every $100,000 in grant support that I bring in, the government gives the university an additional $55,500 (for a total of direct + indirect costs of $155,500). I can spend the $100,000 any way I see fit to achieve the aims proposed in my project. The university can spend the $55,500 any way it sees fit, supposedly to maintain the facilities, recruit faculties, give startup packages, hire support staff, etc. Consequently, researchers in essence are expected to pay for themselves almost completely, and the most successful, best-funded researchers with millions of dolllars of grant support bring in more money for the university than they cost. Just as an example, I'll use my R01 grant, which is funded for around $175,000 a year for five years. That means the university gets around $97,000 a year on top of that, to do with whatever it pleases, in addition to the 25% of my base salary that the direct costs of the grant pay for. Also, believe it or not, some universities have even better deals than this; my university's indirect cost agreement is rather modest, compared to some, which can be over 60% in some institutions.
The bottom line for universities is that the system is bit of a gravy train based on, in essence, the indentured servitude of young researchers, and it is not surprising that the doubling of the NIH budget from FY 1998 to 2003 resulted in an orgy of new construction and recruitment of new faculty to fill this new lab space. Indeed, the Director of the NIH, Dr. John Neiderhuber, in a speech that I attended yesterday at the Annual Meeting of the American Association of Cancer Research, stated that the NIH estimated that there was around $14-16 billion worth of new lab construction started in that period. (At some point, I may blog the rather distressing hard facts he presented; I took notes.) When the NIH budget flattened out from FY 2004 to 2008, resulting in a 12% drop in purchasing power in real dollar terms over the most recent period, all this new construction and new recruiting of researchers, many of whom were chasing the now constrained NIH dollars, led to the current crisis in NIH grant funding. I've been meaning to research how the current system evolved, in order to make a more comprehensive post on this matter, but in essence, it makes biomedical researchers into free-lancers to whom universities rent out lab space and support functions, all funded by the researchers' grant dollars. Woe indeed to the researcher who loses his grants in the current bleak climate. If not yet tenured, he'll be out on his tail. If tenured and without other funding, he'll lose his lab and be relegated to an uncomfortable position of being a "welfare case." I've seen it happen. I was in graduate school in the early 1990's, the last time the NIH funding situation was this bleak, and I saw more than one tenured faculty in our department lose their laboratories and see them given to researchers with funding.
That's the reality for basic researchers, but what about clinicians? For clinicians, the situation is more complex. We have a triple function: To see patients (which can generate considerable revenue for the medical school), to teach (which does not); and to do research (which can also generate revenue). The complication, of course, is the balance between the first and third function; i.e., which can generate more revenue. Clinical faculty are in general expected to pay for themselves through clinical income. This includes paying for office space, secretarial support, plus a dean's tax that all departments pay to the university. Part time faculty who spend the other part of their time doing research are at a disadvantage here in that they can't possibly generate sufficent clinical income to cover their costs. Moreover, they are often at a disadvantage in obtaining NIH funding, because they are also in essence "part time scientists" competing against full time scientists for the same pool of money. Of course, they also have the advantage of being able to do research on human subjects, an advantage basic scientists lack unless they team up with a clinician, but on balance they are mostly at a disadvantage.
Dr. Goldman is almost certainly correct that, to be a "taxpayer" in a specialty that is not driven by procedures, like internal medicine, a clinical researcher needs to bring in NIH grant money. This is mainly because of the low reimbursement for the "cognitive" specialties compared to procedure-driven specialties. However, he is almost certainly incorrect about the best way to become a "taxpayer" in a department of surgery. In surgery departments, the busiest clinicians can bill $1,000,000 a year or even more (busy cardiac surgeons can bill up to several million dollars a year), which, when you subtract expenses (typically maybe $250,000 or more a year per faculty member) still leaves the department with a pretty chunk of change. Indeed, in medical schools, departments of surgery often end up subsidizing medicine and pediatrics departments because of this. For a surgeon, it's pretty darned difficult, if not impossible, to generate that much money in grant support, given the usually high clinical loads compared to academic clinicians in other specialties that surgeons routinely carry. Basically, unless a surgeon is pulling in several hundred thousand dollars of grants per year, something that's only possible if one has multiple large R01 grants, runs several large multi-institutional clinical trials (often funded by big pharma, which will pay as much as several thousand dollars per patient enrolled, depending on the nature of the trial), or is the principal investigator of a P01 program project grant or a SPORE, it's basically impossible. Thus, unlike a lab-based academic physician in, for example, internal medicine, where the reimbursement is not as high and thus it's much more difficult to bring in nearly as much clinical income, the surest route to being a "taxpayer" for a surgeon is to operate. A lot. (Unless that surgeon is a breast surgeon, a specialty for which the reimbursement is almost as bad as it is for an internist, but that's a story for another post.) For an academic internist, in contrast, the best strategy is usually to minimize patient contact and do enough high quality research to be funded.
Because of the intricacies of how medical schools do their accounting, of course, the true situation isn't quite as simple as that (no doubt someone will accuse me of "framing" again), but it's pretty darned close to the real situation. Basically, there are no incentives to teach, and academic physicians aren't paid for teaching, the only common exception that I can think of being residency directors, who often receive a small additional payment for the very large additional aggravation that comes with running a residency program. It's perverse. Worse, medical schools receive funding that could be used to promote teaching:
But these accounting methods are often very artificial. These institutions actually receive quite a lot of money that is meant to support patient care and the academic mission. They collect tuition fees to support education. Academic medical centers receive millions from Medicare for graduate medical education, that is, education of interns, residents and fellows, divided into direct graduate medical education and indirect medical education funds (look here). State medical schools receive state support funds. Private, not-for-profit universities receive contributions that are tax-deductible for the donors, and collect interest and dividends from their endowments that are also not taxed. Yet rarely do faculty get "credit" for generating any of these sorts of funds. Why they don't is unclear, and, on its face, absurd.
Academic surgery is a great undertaking, but no other endeavor can I think of has as many perversely negative incentives and unrealistic demands, all placed on dedicated people who stay in academia even though most of them could double their income by going into private practice. Upon our shoulders are placed the responsibilities of advancing basic science and patient care, taking care of patients, and teaching the next generation of physicians, but there are no incentives for teaching other than negative incentives in the form of a possibility that young faculty won't get tenure if they do not have an "adequate" teaching portfolio to go along with their research and clinical portfolios. Not surprisingly, aside from faculty who truly love teaching, most will do the minimum to get by.
Recognizing this problem, some institutions, ours included, have come up with the concept of the "master clinician," whose job it is to see a lot of patients and to use his or her expertise at taking care of patients to teach residents and medical students. In essence, a "master clinician" is a "taxpayer" upon whom has been placed the additional responsibility of teaching, and it is how well such physicians do at these two tasks that will determine whether they get tenure. Other tenure tracks proposed are the physician-scientist (a lab-heavy physician whose success will depend upon the traditional ability to do basic and translational research, publish, and obtain external funding) and the clinician-scientist (whose success will depend upon the ability to generate clinical income and do clinical research and trials). I do not know if these proposed new tenure tracks are the answer, given how they risk entrenching what is already a de facto "class" system for the different types of tenure without adding any real incentives to teach to anyone other than the master clinicians, upon whom will likely fall nearly entire teaching load. I do know, however, that our current system is broken, and something needs to be done.
There are a lot of overly simplistic misconceptions in that first quoted block, but I will start with this part:
The system seems utterly different from that used in other parts of "higher" education, in which faculty are usually paid straight salaries based on rank and seniority.
Huh? This comment seems almost as misinformed as the public's views of med school that the article laments.
There are certainly differences in our approach from med schools, but it certainly is not as drastic as this.
Indeed. Tenure & promotion are heavily dependent on one's publishing record & external grants. Unless they love teaching, faculty put in as little effort as possible so they can actually make time for research during the academic year. Aside from that, the basic salary is rather deficient, and external or internal grants are necessary to make up the difference. Depending on the institution, 40-60% overhead is common.
My PhD advisor used to tell a story to his students when they fled the nest: he was originally denied tenure, but a week later received notification of a large NSF grant and almost immediately his tenure packet went under reconsideration by the committee, and tenure was granted.
It's not uncommon in my non-medical field for professors to have to pay their summer salary (even after tenure) as well as part of their academic year salary. They also have to pay for all of their post-docs, technicians, and grad students benefits/salary/tuition - unless the department has grad student tuition scholarships or the post-doc comes with their own money. A successful lab generally needs at least three grad students, a post-doc, and a technician or two just to get the research for 3 or 4 grants done.
It's not uncommon in my non-medical field for professors to have to pay their summer salary (even after tenure) as well as part of their academic year salary. They also have to pay for all of their post-docs, technicians, and grad students benefits/salary/tuition - unless the department has grad student tuition scholarships or the post-doc comes with their own money. A successful lab generally needs at least three grad students, a post-doc, and a technician or two just to get the research for 3 or 4 grants done