Saturday, March 7, 2009

How long is an MD/PhD program anyway?

The meeting about the return

This past week I had an important meeting. It’s a meeting I’ve been looking forward to for years now--with both anticipation and anxiety. It’s the turning point; the light at the end of the tunnel for the mudphud. It’s the meeting for all the MD/PhDs returning to the clinics in the next few months. The room swelled with over 25 people, all in the same excited and anxious mood. The excitement comes from finally being done with graduate school.

Not everyone will come out with the same opinion of graduate school. My once very optimistic and outgoing MSTP friend calls graduate school “soul crushing”, for example. Our once weekly or more frequent meetings have slowed to about once every couple months as he struggles almost without sleep to get enough data for a single abstract after his project was scooped by someone in his own lab. While not every story is as extreme as his, almost nobody wants to stay in the lab for yet more time once they’ve been in for a few years. If anyone ever tells you that a MD/PhD wanted to stay in graduate school longer, point your finger in their face and call them a liar for me. Tell them Neuronix told you to do it.


Finally! Once you were never sure when you would be allowed to defend, when that last experiment would work, or whether that paper central to your thesis would ever get accepted. But that’s over now. There’s a direct timeline to graduation. Of course, things are never so simple for me. Before the meeting the program director tapped me on the shoulder and asked me to speak with him in private after the meeting. Would I really be finishing this year? How could I not know so close to the deadline? What do I mean my committee wouldn't come to a consensus about it? Didn't I already have many papers, presentations, a grant, etc? My committee still doesn't seem to want to let me finish?!? But that’s a story for a later time.



Still, there’s that nagging anxiety about returning to medical school. I remember every time we, as newbie 1st and 2nd years, used to ask a MD/PhD graduate if it was hard going to grad school, the response was almost invariably, “No. The hard part is coming back to medical school.” I mean, do you remember with minute detail what you learned 3,4,5 or more years ago? I spent 4 mornings during my entire PhD shadowing a physician—and that was a Radiologist in the reading room! Of course, I’m not the only one. During the meeting we heard advice from the Medicine clerkship and MD/PhD directors. The advice was surprisingly useful until the director joked—“If you don’t remember what Wegner’s or Goodpasture’s is, come see me after the meeting.” We all giggled nervously and whispered to one another. I think this was the intended effect. But, finally one of the students exclaimed “Wegman’s is a supermarket!” We burst into laughter.


As we sat there I thought about who was in attendance. We sat in clusters, those who had just become 6th years like me, 7th years, 8th years, and even a bunch of people I didn’t recognize. You see, if I return to the clinics this October, I will graduate in 7 years total. But looking around I realized that I am one of a lucky few. There are exactly 5 out of 16 students in my year who have a strong change of graduating in 7 years. Well… That’s a little bit unfair. One classmate took a year leave of absence. I’m strongly considering doing the same. One classmate dropped the PhD entirely in my year and another seems to be heading in the same direction. Still, there were far more students in that room who will graduate in 9 or more years than there was who will graduate in 7 years total. Or less?! I had never personally seen a 6 year student. 



Wait, so how long does the MD/PhD program take?
I write about this because it surprises me. When I applied, I remember asking every program what their average time to graduation was. Conventional wisdom was that a program should take 7-8 years. We heard some strange rumors about some programs that kept students for 9 or more years. We heard in person and on the forums that these students were rare and had serious personal problems, and these few bad programs that kept more than one were passed in whispers. The directors at those programs of course claimed any problems were “fixed” and that long graduation times were due to personal problems on behalf of the student. One even told me that one of his 9 year graduates stayed in graduate school because he didn’t want to leave (LIAR!). But when I asked what the average time to graduation was, most directors dodged the question or feigned ignorance. One big exception was when I asked the assistant MD/PhD director at URochester. Once he told me the average was 8.5 years there, I, in my typical insanity, asked him why it was so long at his school. The answer was, “We get a real PhD at URochester.” Please, evaluate this statement for yourself. My response got me a rejection letter that beat me home from the interview. 

Yet now I’ve come to realize that there are many 9 and even 10 year graduates. I’ve spoken to several personally and asked them if it was for personal reasons. The unanimous response has been—“No, I’ve been working my ass off!” The average time to graduation is trending upwards. If you average over the lifetime of the program, I was told when I applied that the average time to graduation here was 7.53 years. The reality is, that over the past 5 years the average is just shy of 8 years. In this batch graduating this year, I am certain from seeing the crop with my own eyes that the average graduation time will be well over 8 years. Shhh. Don’t tell anyone I told you all this. But it’s not just us. I wish I could tell you who some of the more knowledgeable forum posters are. One who would know this sort of thing informs us that nationally the average is up to about 8 years. In this thread http://forums.studentdoctor.net/showthread.php?t=534722 for example we learn that another school does similar things—posting an overall average of 7.3 years over the lifetime of the program, but they are also now up to 7.7 years over the past 5 years. So what was the national average time to graduation 20 years ago?


The average was as low as 6.8 just that long ago (in 1985)! So we are trending sharply upwards and there is no end in sight. 

Why does this matter?
This hurts us MD/PhDs in numerous ways:

1) The average time until a MD/PhD gets their first R01 grant is 43 years Physician-Scientist Training—Reply Ley, Rosenberg JAMA 295 (2006):623-4. For those of you who don’t know about grants, the R01 grant is essentially the grant you need to establish yourself as an independent investigator. At my school for example, word of mouth is that once you have 2 of these grants, you can be eligible for tenure. You have 7-10 years to accomplish this feat, at a rate of funding of about 10% depending on the institute. Good luck, ye who wants to be a basic scientist. At your mid 40s do you still want to be fighting for tenure—a stable job? Imagine, obtaining tenure at the age of 50, just in time to retire at 65. At face this seems silly—training for a lifetime for a job you spend less than two decades performing. How far will this timeline stretch as R01s get harder to get, residencies get longer, and MD/PhD programs get longer? We can already tack on a year to the 43 number because the MD/PhD programs themselves have gotten at least a year longer. 

2) Women feel this more than ever. In my opinion, long training times account for why there are more male MD/PhD students http://www.aamc.org/members/great/mdphd/presentations/garrisonhandout.pdf

Nationally, MD and PhD programs have equalized by gender and yet MD/PhD programs still have not. Everyone notes that training is a difficult time to have children due to time requirements of residency, graduate school, medical school, fellowship, etc etc etc. I think it’s a sad reflection on society that women still shoulder more of this burden than men, but regardless of that, such a long a grueling training time is a difficult time for families. Women also seem more likely in my limited experience to drop out of MD/PhD programs (does anyone have data on this?).

3) Does this long training time contribute to an increasing number of MD/PhDs not persuing basic science? The Hopkins website has some great pie charts on career choice by decade of their graduates-- http://www.hopkinsmedicine.org/mdphd/images2/Career_Trajectory.png. Note the upward trend of the private practice graduates (1995-1999 and beyond still has a sizeable percentage “In Training).

I want to move through my program as quickly as possible, what can I do?

So, what can you do individually to try to graduate in 7 years? A quick glance at the 5 MD/PhDs in my cohort who can graduate in 7 years reveals a simple truth. Here’s a list of the graduate programs the 7 year students are in by graduating student/# of students in that department in my year:

Economics – 1/1 Student

Pharmacology – 1/1 Student

Genomics (GCB) – 1/1 Student

Immunology – 1/? Students (I think we have 3?)

Biochemistry and Molecular Biophysics (BMB) – 1/2 (That’s the imaging guy in Biophysics by the way. The biochemist was the once cheerful fellow I was referring to earlier).

The most popular graduate programs at my school are Cell and Molecular Biology, Neuroscience, and Immunology. Hmmmmmm… So why are these atypical PhDs getting done sooner? Simple. BMB requires one course and lets you take the rest as electives. That’s a full year of classes at most, and you can cut that if you take a bunch of classes during med school. The same goes for Pharmacology. Economics PhDs seem to be mostly simulations and analysis of existing data, and there’s no publishing requirements. Genomics is a mixed bag, as can be imaging, but in general we always hear it’s easier to publish when you do those. I mean, I know one guy who got 3 publications and a PhD in 14 months from a Genomics lab with no prior experience. Let’s contrast this with one other graduate program for example, which at my school requires a full year of teaching undergrads (one course a semester), and more or less requires 2 years full time of classes. The graduate chair tells MD/PhD students openly you should expect it to be a 8-9 year program. The two dropouts in my year were both in that department.

So young student—choose your graduate department wisely. Nobody at the school is ever going to tell you teaching is a waste of time. How can they? They are faculty and their job is not particularly stable as it is. A big part of being faculty is politics! Well, I’ll stick my neck out for you and tell you—it’s a big waste of time. If you ask faculty at schools that don’t require teaching how they feel about teaching they will also tell you this. Besides, if you’re not required to teach, you’ll always have opportunities to teach for money. I’ve never seen a single MD/PhD take that money. But, from a departmental standpoint money from the NIH is getting tighter and someone needs to teach those courses… Hence the true motivation; let’s force the students to do it!

Now we should all recognize that two big parts of getting your PhD quickly, or even at all, is hard work and luck. But we’re talking about averages, so we’ll assume that students today aren’t lazier or have more problems than their earlier counterparts, and let’s talk some more about how to stack the deck in your favor. First, pick a PI who has had MD/PhD students before and moved them through in a reasonable amount of time. Ask them straight out—how do you feel about MD/PhD students? Many will be upfront—whether that means they don’t think MD/PhDs are real PhDs or they recognize the pressure we’re under. Then, grab a project that will likely yield data, and keep a few side projects going just in case. With money as tight as it is, it simply isn’t practical to flounder around for a few years trying to find a project that works like some hardcore old-time PhDs seem to feel is necessary. In many cases you need to produce data for your lab and yourself so it can stay afloat. Finally, be sure to stack your thesis committee with those who aren’t going to badger you and as many MD/PhDs as possible. Some projects, heck, most in molecular biology, are just freaking hard to do. They take a long time and are risky. You aren’t going to cure cancer in your PhD.

The cries are coming: "don’t listen to Neuronix!"

Now the more hardcore among us will flame me into oblivion. What you’re describing isn’t a real PhD! This sort of lax attitude won’t get you the training you need to succeed! You will be undesirable to your future employers! To all this criticism, I have a simple reply: I’m not buying it. During your PhD, what you really want are skills so you can move on, thinking like a scientist (For a similar point of view, see: How to succeed in science: a concise guide for young biomedical scientists. Part I: taking the plunge, Yewdell, Nature Rev Mol Cell Bio, 9 (2008):413-6. 

Residency programs will want to see you published once—preferably in their area of specialization. Most are more concerned with Step I scores and clinical grades anyway. For Radiology, the 4th most popular MD/PhD specialty choice: Key criteria for selection of radiology residents: results of a national survey., Otero et al, Acad Radiol, 13 (2006):1155-64. Research intensive fellowships are practically begging for applicants, given the salary increase given by going into mostly clinical practices. Then when it’s time to be hired as an assistant professor, do you really think they’re going to scrutinize your PhD record? Did med school scrutinize your high school record? Will residency care about your undergrad record? Never forget that whole system is a “what have you done for me lately” system. Similarly, you likely won’t be working on the same thing in fellowship that you were for your PhD. You probably won’t even be in the same field. So what does all that extra toil do for you in the long run? 

This all leads me to advise you to get a PhD, get done, and get on with life. That is what every recent MD/PhD graduate has told me.

We have a problem, here's how I would fix it.
So how do we fix the system? This is where I’m going to get even more controversial. I think we should redefine what the PhD is. In my opinion, the rule of three publications to PhD is outdated and sets the student up to too much abuse and politics. It is simply getting harder to publish, especially in any reputable journal. Many projects don’t lend themselves to publishing quickly. Further, publication acceptions and rejections are often up to fickle and overly critical reviewers and political decisions. Everyone knows this is true and everyone says it is a normal part of research life. And still I have seen MD/PhD students held up because their third paper was not yet submitted for publication (http://drslounge.studentdoctor.net/showthread.php?t=280743)! And still this rule of three publications still remains in many cases. Why?! It benefits the lab PI and the department, but not you.

I propose we go to a four year fixed system. One year of classes and three years of full time research. You can complete it faster if you do manage to get your three papers quickly, but if you make satisfactory progress you’re done in four years by default. Does this hurt the integrity of the PhD? In my opinion no based on two factors:



First, many schools in Europe typically give PhDs in a shorter and often fixed amount of time. Yet, the percentage of publications worldwide are increasing from Europe while the percentage from America is decreasing: Biomedical publication—global profile and trend, Public Health, 117 (2003):274-80. If their training is so poor, how could this be so?

Second, there are many residency programs which now offer you a PhD. They typically claim to offer the PhD in 3-4 years. As a few examples:




Why does our PhD have to be longer? I’m not going to claim it would be easy to make this change. I just think it’s the right thing to do. Fortunately, I’m a student, and I can make these wild suggestions without actually having to implement them. Unfortunately, many graduate programs already view us as MD/MS students or that we don’t get a “real” PhD (whatever the **** that means), so making this happen is unlikely.

So, just how long is an MD/PhD program, anyhow? Count on it being 8 years, possibly 9. Eight years is the new average. If you get done in 7, the old norm, consider yourself lucky.