Yesterday I came across a blog exchange between Dr. Jekyll & Mrs. Hydeand fellow SBer Physioprof about principal investigators (PIs) who still do experiments in the lab. For those not in the science business, a “principal investigator” is in general the faculty member who runs the lab and whose grants fund the salaries of the postdocs, graduate students, and technicians working in the lab. J&H pointed out (correctly) that few PIs who have been faculty more than five years do any actual lab work anymore and described the case of a PI who persists in doing experiments himself, describing him with admiration bordering on awe. Physioprof made a rejoinder that doing such work is a poor use of a PI’s time in that working on grants and papers and mentoring trainees are a far more productive use of his time.
So who’s right? The both are. Of course, neither of them are clinician-scientists, and a whole new range wrinkles mess up the equation of PIs doing experiments in academic surgery departments.
I continued to do actual lab work long past the time Physioprof did. Indeed, it’s only been a year or two since I did my last experiment myself, and I’ve been out of fellowship eight and a half years. Before that, I used to do experiments routinely, albeit with decreasing frequency, over the year, and, believe it or not, several of my papers since I became faculty are mostly data that I generated personally.
Did I do it because I loved bench research so much? Well, to some extent, but not entirely. No, the reason that I did it is because if I didn’t do it necessary experiments wouldn’t get done when I needed them to be done. One thing that’s true of most new clinician-scientists is that we are underfunded and overworked, and because we are underfunded and overworked we often take longer to attain independent funding. If I said what my startup package as a new assistant professor was, both Physioprof and J&H would likely either laugh or show me extreme pity. In retrospect, I can’t believe myself just how paltry my startup package for my lab was. If I hadn’t been fortunate enough to compete successfully for a small one year seed grant of $25,000 and then later a somewhat bigger state grant of $90,000 over two years, my lab would have shut down after two or three years at the most. Indeed, in retrospect, I don’t know if I was really, really good or just really, really lucky. I like to think it was more the former than the latter, but I can’t honestly say that it wasn’t mostly the latter.
Granted, my startup package was probably smaller than most, even in the clinical sciences. I didn’t realize this until years later, when some fellow surgeons and I actually discussed money, something I was never comfortable doing. However, I was young and naive and needed a job. The main consolation was that my clinical responsibilities were light enough that I could spend around three days a week in the lab, and the institution appeared genuinely committed to letting me do my own thing. What this combination of factors meant was that I had to do experiments if there was to be any hope of generating enough preliminary data to be funded, particularly given that it was months before I could hire a technician to help me.
And, as a surgeon-scientist, my story is by no means unique. In many academic departments of surgery in the U.S., there are assistant professors who are not as fortunate as I was in that they have not just an inadequate startup package to contend with but much more onerous clinical responsibilities standing between them and their lab than I had. In that I was fortunate because at least division chief would back me up when I aggressively protected my time. In any case, given the demands of a surgical practice combined with a lack of resources, it’s not surprising that very few surgeons make it through this gauntlet to become true independent investigators. What happens to the vast majority of them is that they end up giving up their labs. Usually, it happens slowly, as increasing clinical responsibilities keep them from paying adequate attention to their labs. Eventually, they find themselves drifting away until one day they find themselves doing close to 100% clinical work.
Why do clinicians tend to have fewer resources thrown at them in their first job? In the old days, clinical departments made lots and lots of money, so much so that it was not particularly onerous to set aside a goodly amount to support the development of new faculty with generous startup packages and protected time. Now, even in academic medical centers, such slush funds no longer exist because reimbursement is usually not any higher than that for private practices anymore. Given that, academic practices have had to become just as efficient and money-oriented as any private practice. That means that high-volume clinicians have to be willing to accept less pay in order to subsidize the researchers if this model is to be viable, or researchers, who because of their research will always have a low clinical volume, have to find a way to cover their expenses with their billing. More than one surgeon (some of whom I know personally) have fallen afoul of this new reality, with their chairmen demanding that they see ever more patients in order to bring in more clinical revenue and then castigating them when their research effort falters and they can no longer bring in grant money.
Actually, I would tend to agree with most of what Physioprof said except that his situation, and those of most basic biomedical researchers, is sufficiently different from that of clinician-researchers that I thought I’d wax “professorial” and give a little advice to budding surgeon-researchers or clinician researchers.
The most critiical thing a newly minted assistant professor of surgery (or other clinical departments) doing research can do is to pick the right position and negotiate for as much as possible In this two things are absolutely key. First is protected time. It is virtually impossible to develop a basic or translational research program without at least 50% protected time, and it’s damned difficult to do it with even that given that the competition is basic scientists. In reality, that’s a bare minimum, and if it isn’t aggressively protected, it will soon shrink, as clinical demands inevitably encroach upon it. Far better is to have between 60-80% protected time. Next is to join partners who understand the importance of research to the academic mission of the department and support it, even if they do not themselves do any translational or basic research. Such partners will be more willing to cover for the clinical load that you do not cover and will not resent your not increasing your clinical load to fill the time available. Finally, an adequate startup package is absolutely essential. At the bare minimum, it should be enough to hire an experienced technician for three years and have enough left over to buy supplies for that same period of time. Much better is to try to negotiate for enough to pay a postdoc or Ph.D.-level researcher for three years. Remember, a surgeon will never have as much time to spend in the lab as a basic researcher; he or she needs a strong, smart, and reliable “right hand man” or woman who is capable of higher-level thinking than a technician to run the lab.
Once the position is yours, then many of Physioprof’s bits of advice apply, particularly the part about spending most time on “high value” activities like writing grants and papers or training and supervising your new minions–but only if you can attract a talented enough postdoc to work in your lab. If you can’t, then you will end up doing experiments in the evenings and on the weekends, as I often did. Still, if I had one bit of advice that is more important than any other for the new assistant professor of surgery who has been hired to do some research, it is this;
PROTECT YOUR RESEARCH TIME. Ruthlessly and without compunction. Do not apologize for it, either.
This is a very difficult order. New surgeons want to operate a lot, and doctors are trained to take care of patients. We want to be good at it; we have to be good at it; and patients deserve nothing less than the best care we can possibly provide. Moreover, the traditional strategy to build a surgical practice generally involves making onself available, no matter what, for surgical consults, whether on call or not. Newly minted surgical researchers must resist this temptation. My cardinal rule was (and is): If I’m not on call, I’m not available to see new consults except in my regularly scheduled clinic time, and I almost never overbook my new patient slots (although I will always overbook to see established patients who are having problems). This may sound harsh, but that’s the way it has to be if I am to successfully balance research and patient care and my research is to succeed. That is not to say that I am not always available for my own patients (I am); it just means that I won’t take any new urgent consults if I’m not on call. Part of the reason I have been fairly successful thus far is that I have been ruthless about protecting my research time, even at the risk of sometimes giving the impression that I’m not interested in clinical work. My response to that is better that impression than a lab with no funding. In actuality, this rule is very much like J&H’s comment about how experimental time should be sacred and not easily interrupted. Of course, in the case of the surgeon-scientist, patient care must come above all else. A surgeon can’t just ignore sick patients, especially if they’re his, although he can and should be very clear about when he can take on new patients–usually only on call.
Being a surgical researcher involves becoming very good at time management. Sadly, that’s a skill that I have not yet mastered. (If I had, I would probably be the PI of a PO1 by now.) Indeed, it is probably the one skill whose lack is most holding me back. This is where Physioprof’s comments become most relevant. Perhaps the hardest thing for me to realize–something that took me years to figure out–is that I’m no longer just a researcher, responsible for no one but myself and nothing other than my one or two projects. I’m now a PI, and I’m responsible not just for the employment of everyone in my lab, but for their mentoring and making sure they make progress. It’s a heavy burden. Particularly hard is the constant knowledge that, if I lose my funding, I can always go back to taking care of patients while consequences for the people working for me could be much worse: They would lose their jobs. There is not a day that goes by when I do not obsess of what could happen to them in a couple of years when my current grants run out if in this very tight funding climate I am not successful in renewing my funding and attracting new funding. Consequently, like Physioprof, I agree that doing a lot of actual lab work is no longer the most profitable use of my time. Writing papers and grants, as well as taking the time to read and come up with the “big picture” (or the theme of my lab’s research) is what I have to devote my precious non-clinical time to, not doing physical techniques that I used to be but am no longer good at.
I guess what it boils down to is the difference between being a “just” a researcher and developing into a leader. Even though I’m not exactly young anymore, I still marvel that I am now a PI and, whether I like to acknowledge the role or not, a leader responsible for motivating my troops to do their best work, helping them advance their careers, and holding their feet to the fire when they do not live up to expectations. I don’t know yet if I’m even that good at it, but I’m learning. Given such a role, unless there is no one else to do a particular task, my time is much better utilized not grinding out assays and doing experiments that might be interrupted by clinical responsibilities but by working to make sure that my lab’s research plan is both relevant to cancer and top-notch in quality.
Anything less, and I could in two years find myself bereft of funding and soon after bereft of my lab–while my lab personnel could find themselves bereft of their jobs.