What future position are current MD/PhD students training for?

[okemba]

It always struck me when I was considering applying to MD/PhD programs that... the current batch of MD/PhD trainees seem to be training for positions that didn't yet exist. This isn't critical or anything, I'm coming from a place of trying to achieve understanding.

I will post bullet points instead of prose:

1. Clinical research positions (IMO) are not a desired outcome of MD/PhD programs, and are not a great return on training investment for either programs or trainees.
2. If current trainees are not interested (generalizing here) in clinical research positions, and tenure-track positions are not available, then their other option to remain engaged in "real" research (most but not all of which has a large wet-lab/experimental component) would be some kind of part-time staff scientist role, which would require 20-30+h/week of pipetting.
3. Most people above the age of 40 do not want to be pipetting near full-time as part of their career.

It's these three points, 1>2>3 which make me not understand how career progression will work for the current batch of MD/PhD trainees. It's one thing if trainees can look forward to a higher-level role surveying the literature and thinking of new projects, advising students and analyzing their experimental results, but this part-time staff scientist role is quite a different beast...

Are people just going to drop out of research en masse, or what? I don't understand. Having many MD/PhD-bearing 40+ year olds doing labwork nearly full-time seems ridiculously unlikely to me. How are these people going to remain engaged with research?

---

Members don't see this ad.

 

[Fencer]

The NIH still has an annual budget of $ 31 billion dollars for research. Who would be working on that? The MD/PhD workforce hasn't grown much. Indeed, nationally, there has been a decrease number of openings for tenure-track positions and a delay in retirements from tenured faculty. However, there is a lot of heterogeneity across the country, my institution has doubled the number of MD/PhD faculty over the past decade, and at least half of the new recruits are doing bench research.

 

[URHere]

For most MD/PhDs, the decision tree doesn't ever take you down the staff scientist path. If you want a significant research component to your career, you look for a research-intensive fellowship where you have a year or two of protected research time and you use that time to apply for a K-award. If you get one, you have the funding to continue conducting research in the quest for an R01. Obviously, not everyone gets to that point. It's possible to fail at either the K or R award stage...but don't forget that many MD/PhDs choose not to take this path in the first place, either because they enjoy clinical work too much to cut their hours or because the attending-level paycheck is too tempting to pass up for a more meager research salary.

Back up plans for many MD/PhDs do include clinical research, but for most the back up plan is to simply be a full-time clinician. After all, one big benefit of the combined program is that you have a career safety net that is not available to your PhD-only colleagues. Ultimately, I have met more than a handful of bitter MD/PhD interviewers on the residency trail telling me all about why basic science research isn't feasible anymore...but "not guaranteed" is significant different than "not feasible", and if you want to be a basic science PI, there is certainly still a path that gives you a chance of making it happen.

 

[dl2dp2]

1. Clinical research positions (IMO) are not a desired outcome of MD/PhD programs, and are not a great return on training investment for either programs or trainees.

I'm not sure this is true. Also, there is not a stark division between what is basic, translational or clinical, given that a lot of basic research is done on human subjects, and a lot of clinical research is done on human tissues. People do research that leverages their background, and apply for funding mechanisms to sponsor the research they want to do. I'm pretty sure MD/PhD training would not have been a waste either individually or institutionally if the graduate continues to contribute to research.

2.If current trainees are not interested (generalizing here) in clinical research positions, and tenure-track positions are not available, then their other option to remain engaged in "real" research (most but not all of which has a large wet-lab/experimental component) would be some kind of part-time staff scientist role, which would require 20-30+h/week of pipetting.

The variety of research in biomedicine vastly exceeds "pipetting". Similarly, the administrative possibilities of a career goes between 100% hard money ("tenure track") to 100% soft money to 100% clinical ("clinician educator"). You can apply for grants on a project by project basis and live off soft money and clinical revenue -- in fact this is the dominant model for PIs for most large centers. However, this is NOT "20-30 hours of week of pipetting."

3.Most people above the age of 40 do not want to be pipetting near full-time as part of their career.
It's these three points, 1>2>3 which make me not understand how career progression will work for the current batch of MD/PhD trainees. It's one thing if trainees can look forward to a higher-level role surveying the literature and thinking of new projects, advising students and analyzing their experimental results, but this part-time staff scientist role is quite a different beast...

There aren't that many MD/PhD trainees. Generally speaking, if you *really* want a job as a "basic" PI and you got into an MD/PhD program, you have a pretty good shot of being able to do that. On an institutional basis however, "tenure track" jobs don't really exist. This does not mean though that you'll function as a technician when you are a PI on an R01.

The dominant reason for the leaky pipeline isn't that MD/PhDs really want to be academic PIs and can't be one, but because other possibilities become incredibly attractive in comparison to a job as a PI. MD/PhDs with a full residency training can often command 2-3x salary in clinical or administrative positions, and 3x+ salary in the private sector, with very similar responsibilities on a day to day basis (i.e. research, clinical, administrative, supervisory). Furthermore, because of paucity of "tenure track" jobs, if you really want one, you'll take a huge pay cut to relocate to an undesirable location. Nevertheless, the jobs are there, you can go for it if you want.

This is in contrast to a PhD-only graduates, where alternative careers are not significantly better than a career in PI-hood. It becomes a matter of trade-offs. Indeed, a large proportion of PhD-onlys will never get academic jobs no matter how low they drop their institutional, location or salary requirements.

You need more experience talking to mentors. Your comments reflect a relative ignorance of the possible career tracks and long term outcomes of MD/PhDs.

 

[Neuronix]

n=1, but both URHere and sluox claimed that there is a research position available for all MD/PhD grads, even if undesirable. I take issue with that assertion, as I had no majority research faculty or fellowship option after residency.

 

[dl2dp2]

n=1, but both URHere and sluox claimed that there is a research position available for all MD/PhD grads, even if undesirable. I take issue with that assertion, as I had no majority research faculty or fellowship option after residency.

Hmm. Do you think if you told the dept chair that you want a "research fellowship" that pays 80% research and spend 20% clinical, and get paid let's say 50k+300k*0.2 = 90k, such a position would be unavailable?

Still. If you "really wanted it", you could just VOLUNTEER your 80% time as a researcher, which would boil down to getting a part time clinical job. What you are saying is that in some specialties, part time is not available. I can see that. This may be an issue ===> what this means is that what I'm saying is only true for certain cognitive specialties. I guess if you "really want to be a basic PI", you should not pick some specialties because it blocks your career progression.

Usually, you need some period of time to prepare and write a K award application, at which point you get paid 90k+300k*0.25 = 165k. NCI currently has K funding rate at around 25-35%, which means you have a reasonably good shot to get it, especially if you resubmit. This is what I mean by "if you really want it". Then subsequent to the K, if you go for an R multiple times, eventually you'll get it. Meanwhile, you'll subsist on whatever partial clinical involvement you have to float your salary. Maybe sometimes you'll have a funding gap, which means for a year or two you'll do entirely clinical work, but perhaps at a reduced salary. Eventually your grant comes through, and some proportion of your FTE gets "bought out" to do research. If you are a star and get lots of grants you get to spend most of your time doing research. This seems to be basically how it all works these days no matter what specialty you are in.

But of course, why the HELL would you? I know I wouldn't.

 

[Neuronix]

Hmm. Do you think if you told the dept chair that you want a "research fellowship" that pays 80% research and spend 20% clinical, and get paid let's say 50k+300k*0.2 = 90k, such a position would be unavailable?

Yes. I discussed this possibility with several departments. I applied for a similar NIH fellowship and was rejected.

I do have the caveat of my research only being available at maybe 10 instutions, so I didn't have a huge number of places to which I could apply. Still, that is the curse of having a PhD and a specific skillset and knowledgebase. The MD->post-doc types have much more flexibility in that they aren't as differentiated. I suppose I could "de-differentiate", and change my research area. I still feel that my research area is topical, potentially high impact, and I have no desire to start over at age 35.

I should add as well that I couldn't be happier with my clinical academic position. I couldn't have asked for a better clinical assistant professor appointment. But I think it's incorrect to assert that I had an option to obtain a majority research position.

 

[dl2dp2]

I should add as well that I couldn't be happier with my clinical academic position. I couldn't have asked for a better clinical assistant professor appointment. But I think it's incorrect to assert that I had an option to obtain a majority research position.

Zackly. And I think that this is one of the optimal outcomes for the MSTPs. Someone has to staff these clinical positions at academic medical centers.

I guess it all boils down to money and security. With MSTP, you can get paid a solid six figured salary doing "mostly" research to the extent that projects get funded by NIH, and also at the same time get the security that you'll always have a job that pays this much and potentially a lot more.

With a PhD-only, that's literally not true. Your options are perpetual postdoc at ~ 50k for a decade vs. a private sector job that pays marginally more with no security. Literally about 5% of PhD-onlys get a tenure track position. And not all tenure track positions are created equal! I'd say a lower tier tenure track position as a basic PI is a much less desirable job in almost every single respect to the majority of individuals compared to a clinical academic medicine job. For an optimal outcome in that track (tenure track at a reputable R1 with sufficient environment to sustain long term funding success, startup package, etc.), you need pedigree and luck every step of the way. The odds are more like 1 out of a 500-1000. And even THEN the job isn't great. I know people who are on that track (i.e. managing a large high profile, well funded basic science lab at a top institution), and personally I don't think it's a necessarily better job than being an academic physician-scientist (i.e. the soft money 80/20 job), but of course that's an argument that can be made both ways and depends on the person. To the average lay person the 80/20 job is better: sounds better (prestige), higher salary, "real doctor" doing research to cure diseases, better lifestyle/hours. To the basic scientist the job is worse: have to do "irrelevant" clinical work, no "tenure", disease oriented work not as "fundamental" and "interesting", lower impact journals, not as flashy, etc.

The student OP asks this question: should I do MD/PhD with a low probability of becoming a basic PI? My reply is: what other options do you have? Do a PhD-only with an even LOWER probability (sometimes by an order of magnitude) of becoming a basic PI? ==> If you want to be a basic PI, have the grades to get into med school, and like disease oriented work, then by all means do the MD/PhD. Going to med school was the BEST thing I've done professionally, and I think you'd agree with that as well.

 

[URHere]

n=1, but both URHere and sluox claimed that there is a research position available for all MD/PhD grads, even if undesirable. I take issue with that assertion, as I had no majority research faculty or fellowship option after residency.

I'll admit that I was mainly speaking about options available in the IM or Peds pathways. This is especially true for people like myself who are interested in less competitive fellowship areas (ID, nephrology, etc). Things are likely much different in more competitive or sub-specialized fields where fewer positions are available.

Thankfully, clinical practice is a viable back-up plan in nearly all cases.

 

[Neuronix]

I know medical oncologists who feel similarly. I can't comment on ID or nephro.

 

[tr]

Yes. I discussed this possibility with several departments. I applied for a similar NIH fellowship and was rejected.

I do have the caveat of my research only being available at maybe 10 instutions, so I didn't have a huge number of places to which I could apply. Still, that is the curse of having a PhD and a specific skillset and knowledgebase. The MD->post-doc types have much more flexibility in that they aren't as differentiated. I suppose I could "de-differentiate", and change my research area. I still feel that my research area is topical, potentially high impact, and I have no desire to start over at age 35.

I should add as well that I couldn't be happier with my clinical academic position. I couldn't have asked for a better clinical assistant professor appointment. But I think it's incorrect to assert that I had an option to obtain a majority research position.

I don't think it's incorrect to say this. By your own admission you had the option to 'de-differentiate' and do a postdoc with an established PI. This is the only option for the vast majority of PhDs who want to stay in research and keep trying to find a tenure-track position. But it's such a vastly inferior option to the clinical assistant professor route that you don't even really consider it an option. For which I don't blame you. I don't consider it one either. But I'm in complete agreement with everything sluox has said here.

For some reason there's this idea on this board that clinical assistant professor jobs are subpar, 'not what we trained for,' etc. Huh? Clinical assistant professor is fantastic. You have job security, interesting work, fantastic salary compared to PhDs (and almost everyone else except private practice MDs - let's not forget that the mean FAMILY income in the US hovers around $50K/year), and an affiliation with a university that gives you the option to keep doing your own research and applying for funding at whatever % you can make it work, filling the rest out with clinial. Versus the PhD only who is competing for that 1/1000 chance of a 100% basic research position, with other options being only endless postdocs or opting out of academia to industry jobs (which have their own merits but largely don't allow you to continue to pursue your own research agenda in any form).

I do think that in the case of someone who is dead set on running a basic science lab and would not enjoy clinical work, there is not really an advantage to the MD, because you lose a lot of time to clinical training that could be put to building a publication record, and thus you are at a disadvantage when applying for funding and tenure-track jobs. But the vast majority of people who start out with this intention won't make it anyway, and those who don't have an MD don't have an alternative option to stay in academia at all. It's tenure-track, else postdoc or out.

 

[eteshoe]

I actually want to be a medical oncologist with a clinical academic appointment. My dad was a basic science PI and so I've seen that life from both sides. There's pros and cons to every path but I figure I enjoy what I'm doing so might as well keep going.

 

[GadRads]

To the average lay person the 80/20 job is better: sounds better (prestige), higher salary, "real doctor" doing research to cure diseases, better lifestyle/hours. To the basic scientist the job is worse: have to do "irrelevant" clinical work, no "tenure", disease oriented work not as "fundamental" and "interesting", lower impact journals, not as flashy, etc.
.

So true. I know an MD/PhD faculty who was very pro-basic science who thought this way. He found clinical research/medicine very boring.

 

[chalamilasidney]

Can you receive multiple NIH grants (Ks / Rs) at the same time? do you know someone in residency or fellowship or postdoc who did this?

 

[Fencer]

I know people who obtained a K08 late in their fellowship (in a hybrid fellow/instructor position), who then moved and used their K as leverage for an Assistant Professor position with a package. K awards are 5 years, and you can receive a R award in the last 2 years of your NRSA K fellowship.
FYI: I did move with a K award between two institutions.

 

[SurfingDoctor]

Can you receive multiple NIH grants (Ks / Rs) at the same time? do you know someone in residency or fellowship or postdoc who did this?

No, you can only be on one type of K award (K08, K23, KL2, K12) at a time and spend a maximum of 5 years on a K award. Rs can be multiple however and as mentioned, you can apply for Rs in the latter half of a K award and maintain the salary offered by the K. Ks are possible during fellowship training, though personally I have seen it more for junior faculty (and some K12s, and institutional KL2s are only available to faculty). The more common path that I have seen for fellows are T32s and F32s, the former being more common.

The instructor thing as mentioned above is possible (ie, not a fellow but not an assistant professor with the tenure clock ticking) though those types of positions are highly institutional dependent. For instance, when I started on faculty, that was the recommended path for physician-scientists to be promoted onto the tenure track. However, in the past 3 or 4 years, that position has been phased out across the university.

 

[CheGuevaraMD]

I don't think it's incorrect to say this. By your own admission you had the option to 'de-differentiate' and do a postdoc with an established PI. This is the only option for the vast majority of PhDs who want to stay in research and keep trying to find a tenure-track position. But it's such a vastly inferior option to the clinical assistant professor route that you don't even really consider it an option. For which I don't blame you. I don't consider it one either. But I'm in complete agreement with everything sluox has said here.

For some reason there's this idea on this board that clinical assistant professor jobs are subpar, 'not what we trained for,' etc. Huh? Clinical assistant professor is fantastic. You have job security, interesting work, fantastic salary compared to PhDs (and almost everyone else except private practice MDs - let's not forget that the mean FAMILY income in the US hovers around $50K/year), and an affiliation with a university that gives you the option to keep doing your own research and applying for funding at whatever % you can make it work, filling the rest out with clinial. Versus the PhD only who is competing for that 1/1000 chance of a 100% basic research position, with other options being only endless postdocs or opting out of academia to industry jobs (which have their own merits but largely don't allow you to continue to pursue your own research agenda in any form).

I do think that in the case of someone who is dead set on running a basic science lab and would not enjoy clinical work, there is not really an advantage to the MD, because you lose a lot of time to clinical training that could be put to building a publication record, and thus you are at a disadvantage when applying for funding and tenure-track jobs. But the vast majority of people who start out with this intention won't make it anyway, and those who don't have an MD don't have an alternative option to stay in academia at all. It's tenure-track, else postdoc or out.

While it's true that academia is brutally competitive these days, I don't think that a clinical professorship is a great substitute for a regular tenure track position. Here's an analogy: just because you sell paintings at the local coffeeshop or on the boardwalk doesn't mean you're actually on your way to become Ai Weiwei, Jeff Koons, or Banksy. George Bush is not a great artist of our time.

Science, like fine art, is about the communication of ideas, and these ideas can only be communicated if you have not only talent but also substantial resources and a platform for national or international recognition. If you don't have substantial federal funding (or equivalent) and don't regularly publish in high impact journals, your science is probably more of a hobby than a vocation. Is that unfair? Sure, but we all can't be famous modern artists either.

 

[tr]

While it's true that academia is brutally competitive these days, I don't think that a clinical professorship is a great substitute for a regular tenure track position. Here's an analogy: just because you sell paintings at the local coffeeshop or on the boardwalk doesn't mean you're actually on your way to become Ai Weiwei, Jeff Koons, or Banksy. George Bush is not a great artist of our time.

Science, like fine art, is about the communication of ideas, and these ideas can only be communicated if you have not only talent but also substantial resources and a platform for national or international recognition. If you don't have substantial federal funding (or equivalent) and don't regularly publish in high impact journals, your science is probably more of a hobby than a vocation. Is that unfair? Sure, but we all can't be famous modern artists either.

I'm not sure what it would mean for one position to 'substitute' for another, but I would say a clinical professorship with research activity is an appropriate outcome of MSTP training, and one with some significant practical advantages to the individual.

If you are someone who only wants to run a basic science lab, and has no interest in clinical/translational work, then yeah, a clinical professorship most likely won't get you what you want. In that case, as I said, MSTP training is no advantage to you, and may be a disadvantage because of the years lost to clinical training that result in a smaller publication record.

 

[OddNath]

For some reason there's this idea on this board that clinical assistant professor jobs are subpar, 'not what we trained for,' etc. Huh? Clinical assistant professor is fantastic. You have job security, interesting work, fantastic salary compared to PhDs (and almost everyone else except private practice MDs - let's not forget that the mean FAMILY income in the US hovers around $50K/year), and an affiliation with a university that gives you the option to keep doing your own research and applying for funding at whatever % you can make it work, filling the rest out with clinial. Versus the PhD only who is competing for that 1/1000 chance of a 100% basic research position, with other options being only endless postdocs or opting out of academia to industry jobs (which have their own merits but largely don't allow you to continue to pursue your own research agenda in any form).

As someone who has just had a K rejected, and looking at likely starting a 100% clinical faculty position in a few months-- this is some comforting perspective. It's easy to constantly feel like a failure in this training pathway. Thank you.

 

[tr]

As someone who has just had a K rejected, and looking at likely starting a 100% clinical faculty position in a few months-- this is some comforting perspective. It's easy to constantly feel like a failure in this training pathway. Thank you.

Yeah I couldn't get a K funded either. Honestly it's fine. I'm actually still doing the project I proposed for the K, just at a slightly smaller scale and with foundation/internal funding and some generous help from senior faculty. Obviously I would have liked to have had my K funded but I enjoy my job, I still have a good mix of different things that I'm doing, and I'm happy with my life. This alternative works for me. I hear you about feeling like a failure but hey, life goes on. Most of the people who are a few years ahead of me and coming off Ks at my institution are ending up funnelling into clinician-educator jobs at that point anyway.

 

[StilgarMD]

Yeah I couldn't get a K funded either. Honestly it's fine. I'm actually still doing the project I proposed for the K, just at a slightly smaller scale and with foundation/internal funding and some generous help from senior faculty. Obviously I would have liked to have had my K funded but I enjoy my job, I still have a good mix of different things that I'm doing, and I'm happy with my life. This alternative works for me. I hear you about feeling like a failure but hey, life goes on. Most of the people who are a few years ahead of me and coming off Ks at my institution are ending up funnelling into clinician-educator jobs at that point anyway.

So I think during the PhD, we don't get a good look at the "non-traditional professorship". What kind of operation can you run in circumstances like these? I know there is going to be variability, but if you could give some ideas about whats possible that would be great. I'm guessing expensive model organisms are out. I am acutely aware of the whole mess that may be worse or better by the time I'm in the shoes of someone starting a lab, and I'd be interested to hear what is possible.

 

[tr]

So I think during the PhD, we don't get a good look at the "non-traditional professorship". What kind of operation can you run in circumstances like these? I know there is going to be variability, but if you could give some ideas about whats possible that would be great. I'm guessing expensive model organisms are out. I am acutely aware of the whole mess that may be worse or better by the time I'm in the shoes of someone starting a lab, and I'd be interested to hear what is possible.

There is just so much variation, but I can talk about my own experience.

I did a basic science PhD and by the time I was done with it I knew I never wanted to run a basic science lab. After residency I did a fellowship/postdoc with a clinically focused project. Towards the end of that I connected with a faculty member who runs a basic science lab, and who was able to help me pursue some more translational questions I had, that grew out of the clinical research I had been doing. I piggyback on his lab and his expertise. When I was still in fellowship and had protected time I was pipetting for my project myself. Now I pay his tech a few percent salary to do the work for my project.

As I mentioned, since finishing fellowship I was never able to get NIH money . I have a clinician-educator appointment and have been pushing my agenda forward with foundation and internal seed money.

In terms of where this is going, I honestly don't know. It may be that I can keep this up long enough to scrape my way into some more reliable funding. It may be that I am never able to do that and will end up with an essentially clinical career plus a few papers on the side. I am OK with both of those outcomes, although if course I enjoy my research and don't want to let it die if I can help it.

I do have a mentor who has a setup that I would like to replicate if possible. She runs a research group. You could call it a lab but there's no physical lab. At any given time she usually has a couple of RAs, sometimes a grad student or fellow, sometimes not. Undergrads, med students, and other students who work for credit hours rotate through frequently. There are several junior clinical faculty in the group as well. Things expand and contract flexibly depending on funding.

She usually has NIH funding at any given time but also has stretches where she doesn't, but might have foundation or pharma funding instead.

She collaborates extensively with people who do have basic science labs, just like I am doing, although her interests and collaborators are different. She also makes use of the excellent core facilities at our institution to get at mechanistic questions.

If I stay here, keep being co-I on her grants, and keep pushing my own stuff forward on a shoestring, at some point it's possible I will figure out how to crack the code of federal money, in which case it's possible I will be able to replicate her setup. I'm not sure though. One issue is that my research interests are not appealing to pharma companies, which seem to be important stopgaps in this model for the times without federal funds.

I hope that answers your question? I don't myself know how this is going to turn out. I'm feeling my way forward one step at a time, which is kind of the way I operate.

 

[StilgarMD]

There is just so much variation, but I can talk about my own experience.

I did a basic science PhD and by the time I was done with it I knew I never wanted to run a basic science lab. After residency I did a fellowship/postdoc with a clinically focused project. Towards the end of that I connected with a faculty member who runs a basic science lab, and who was able to help me pursue some more translational questions I had, that grew out of the clinical research I had been doing. I piggyback on his lab and his expertise. When I was still in fellowship and had protected time I was pipetting for my project myself. Now I pay his tech a few percent salary to do the work for my project.

As I mentioned, since finishing fellowship I was never able to get NIH money . I have a clinician-educator appointment and have been pushing my agenda forward with foundation and internal seed money.

In terms of where this is going, I honestly don't know. It may be that I can keep this up long enough to scrape my way into some more reliable funding. It may be that I am never able to do that and will end up with an essentially clinical career plus a few papers on the side. I am OK with both of those outcomes, although if course I enjoy my research and don't want to let it die if I can help it.

I do have a mentor who has a setup that I would like to replicate if possible. She runs a research group. You could call it a lab but there's no physical lab. At any given time she usually has a couple of RAs, sometimes a grad student or fellow, sometimes not. Undergrads, med students, and other students who work for credit hours rotate through frequently. There are several junior clinical faculty in the group as well. Things expand and contract flexibly depending on funding.

She usually has NIH funding at any given time but also has stretches where she doesn't, but might have foundation or pharma funding instead.

She collaborates extensively with people who do have basic science labs, just like I am doing, although her interests and collaborators are different. She also makes use of the excellent core facilities at our institution to get at mechanistic questions.

If I stay here, keep being co-I on her grants, and keep pushing my own stuff forward on a shoestring, at some point it's possible I will figure out how to crack the code of federal money, in which case it's possible I will be able to replicate her setup. I'm not sure though. One issue is that my research interests are not appealing to pharma companies, which seem to be important stopgaps in this model for the times without federal funds.

I hope that answers your question? I don't myself know how this is going to turn out. I'm feeling my way forward one step at a time, which is kind of the way I operate.

This does mostly answer my questions. It is interesting to think on alternative sources of funding and what kind of work survives the gaps. I sort of have this idea in the back of my head that developing research methods and interests which don't require anything besides data I can access and a computer (Modeling clinic data or publically available data sets).

 

[CheGuevaraMD]

I'm not sure what it would mean for one position to 'substitute' for another, but I would say a clinical professorship with research activity is an appropriate outcome of MSTP training, and one with some significant practical advantages to the individual.

If you are someone who only wants to run a basic science lab, and has no interest in clinical/translational work, then yeah, a clinical professorship most likely won't get you what you want. In that case, as I said, MSTP training is no advantage to you, and may be a disadvantage because of the years lost to clinical training that result in a smaller publication record.

I don't think that a clinical professorship is a bad outcome, but one is certainly overtrained if they earn a PhD and complete a postdoc and never become a PI. Given that many clinical professors are successful without a PhD, MSTP training just doesn't make sense. Though I think overtraining in research is a problem in general. The K is considered a training grant! And some people fail to secure an R after their K -- what a waste of training!

 

[tr]

I don't think that a clinical professorship is a bad outcome, but one is certainly overtrained if they earn a PhD and complete a postdoc and never become a PI.

Hm. That describes over 99% of postdocs. Are you suggesting that all the world's industry, staff scientist, etc positions should be filled by people without postdoc training?

On the other side, many people with MDs but not PhDs run research laboratories. Does that mean dual degrees are unnecessary?

Given that many clinical professors are successful without a PhD, MSTP training just doesn't make sense.

I think that depends on what they are actually doing, not on their job title.

For myself, I can say for sure there is no way I would be able to ask the kinds of questions I am asking without neuroscience training. I just wouldn't have the perspective to come up with them. The teaching I do also requires graduate level neuroscience knowledge. I hated my PhD but there's no way I would say I am not using the training.

Beyond that, at what % research involvement would you say that a clinician-educator appointment is a waste of a PhD? At my institution, % research for CE faculty ranges from 0-100%. Heck, the statistician I work with is a CE, though I don't think she has any actual clinical capabilities. It just means the university can demand that her entire salary go on soft money, and they never have to give her tenure. What a deal for the U, right? No wonder they're shedding tenure lines like there's no tomorrow.

Though I think overtraining in research is a problem in general. The K is considered a training grant! And some people fail to secure an R after their K -- what a waste of training!

Have to agree with you there.

 

[tr]

This does mostly answer my questions. It is interesting to think on alternative sources of funding and what kind of work survives the gaps. I sort of have this idea in the back of my head that developing research methods and interests which don't require anything besides data I can access and a computer (Modeling clinic data or publically available data sets).

From a career survival perspective, this is absolutely a great set of skills to cultivate.

From the perspective of getting actual useful information, I have to say I'm pretty disillusioned with what's come out of data mining from these huge datasets that weren't specifically designed to answer a given question.

The huge numbers look really shiny and tend to attract attention from journal editors, but the ability to account for confounders is abysmal, and beyond that, the huge numbers tend to mean that even tiny and clinically irrelevant effects make it under the significance bar, which is misleading.

My feeling is that people should hold off from data mining unless they already have a solid idea about potential causality for what they're testing.

Otherwise you have these huge studies that find associations between A and B that are actually completely related to unmeasured confounders, yet nonetheless get interpreted to mean that megadoses of vitamin D prevent cancer or antidepressants cause autism or birth control pills cause suicide. Then they get out in the popular press and cause hysteria and interfere with proper care.

 

[StilgarMD]

From a career survival perspective, this is absolutely a great set of skills to cultivate.

From the perspective of getting actual useful information, I have to say I'm pretty disillusioned with what's come out of data mining from these huge datasets that weren't specifically designed to answer a given question.

The huge numbers look really shiny and tend to attract attention from journal editors, but the ability to account for confounders is abysmal, and beyond that, the huge numbers tend to mean that even tiny and clinically irrelevant effects make it under the significance bar, which is misleading.

My feeling is that people should hold off from data mining unless they already have a solid idea about potential causality for what they're testing.

Otherwise you have these huge studies that find associations between A and B that are actually completely related to unmeasured confounders, yet nonetheless get interpreted to mean that megadoses of vitamin D prevent cancer or antidepressants cause autism or birth control pills cause suicide. Then they get out in the popular press and cause hysteria and interfere with proper care.

100% agree with all of those problems with this approach. In an ideal world, you'd have a strong relationship with the clinicians at your institute capable of doing confirmatory studies on any possible associations, or at least working with them to make the data as complete as needed to avoid potential confounds you suspect will be present.

 

[tortuga87]

Wanted to pitch in here, since I have been developing statistical research methods for many years now. There are both pros and cons to working in this area...

Pros:
- you are often the only limiting factor (do not have to wait for others, animals or things)
- you can work from anywhere including in the clinic during downtime
- finance wise, you basically need to cover your salary and a good computer
- your methods get taken up by scientists around the world in unexpected areas

Cons:
- very steep learning curve especially for ppl who are not trained in a mathematical discipline
- few ppl understand what you do or can read your papers. Journals thus do not have high impact factors because the community is small
- the medical community assumes your grants/publications work like theirs, when it usually doesnt. This makes it hard to write a grant because, once the grant is finished, you have finished the work. Also, the review time for statistics journals is a few years. Conferences are valued highly in computer science but not in medicine which makes it harder to get promoted under a medical department.
- you are stuck somewhere in between medicine and stats/CS/math, and can have trouble feeling like you have a home base if your research does not fit nicely within your institution's biostats department

 

[StilgarMD]

Wanted to pitch in here, since I have been developing statistical research methods for many years now. There are both pros and cons to working in this area...

Pros:
- you are often the only limiting factor (do not have to wait for others, animals or things)
- you can work from anywhere including in the clinic during downtime
- finance wise, you basically need to cover your salary and a good computer
- your methods get taken up by scientists around the world in unexpected areas

Cons:
- very steep learning curve especially for ppl who are not trained in a mathematical discipline
- few ppl understand what you do or can read your papers. Journals thus do not have high impact factors because the community is small
- the medical community assumes your grants/publications work like theirs, when it usually doesnt. This makes it hard to write a grant because, once the grant is finished, you have finished the work. Also, the review time for statistics journals is a few years. Conferences are valued highly in computer science but not in medicine which makes it harder to get promoted under a medical department.
- you are stuck somewhere in between medicine and stats/CS/math, and can have trouble feeling like you have a home base if your research does not fit nicely within your institution's biostats department

yea the learning curve is something I'm feeling quite a bit, but I'm hoping it will pay off. Thanks a lot for the input.

 

[pbrocks15]

There is just so much variation, but I can talk about my own experience.

I did a basic science PhD and by the time I was done with it I knew I never wanted to run a basic science lab. After residency I did a fellowship/postdoc with a clinically focused project. Towards the end of that I connected with a faculty member who runs a basic science lab, and who was able to help me pursue some more translational questions I had, that grew out of the clinical research I had been doing. I piggyback on his lab and his expertise. When I was still in fellowship and had protected time I was pipetting for my project myself. Now I pay his tech a few percent salary to do the work for my project.

As I mentioned, since finishing fellowship I was never able to get NIH money . I have a clinician-educator appointment and have been pushing my agenda forward with foundation and internal seed money.

In terms of where this is going, I honestly don't know. It may be that I can keep this up long enough to scrape my way into some more reliable funding. It may be that I am never able to do that and will end up with an essentially clinical career plus a few papers on the side. I am OK with both of those outcomes, although if course I enjoy my research and don't want to let it die if I can help it.

I do have a mentor who has a setup that I would like to replicate if possible. She runs a research group. You could call it a lab but there's no physical lab. At any given time she usually has a couple of RAs, sometimes a grad student or fellow, sometimes not. Undergrads, med students, and other students who work for credit hours rotate through frequently. There are several junior clinical faculty in the group as well. Things expand and contract flexibly depending on funding.

She usually has NIH funding at any given time but also has stretches where she doesn't, but might have foundation or pharma funding instead.

She collaborates extensively with people who do have basic science labs, just like I am doing, although her interests and collaborators are different. She also makes use of the excellent core facilities at our institution to get at mechanistic questions.

If I stay here, keep being co-I on her grants, and keep pushing my own stuff forward on a shoestring, at some point it's possible I will figure out how to crack the code of federal money, in which case it's possible I will be able to replicate her setup. I'm not sure though. One issue is that my research interests are not appealing to pharma companies, which seem to be important stopgaps in this model for the times without federal funds.

I hope that answers your question? I don't myself know how this is going to turn out. I'm feeling my way forward one step at a time, which is kind of the way I operate.

Is it that difficult to get a cut of the pie (the NIH grants)? I know this is a broad question and that the answer is typically yes, but I thought though that MD/PhDs are more likely to get NIH grants and have less funding issues.

 

[tr]

Is it that difficult to get a cut of the pie (the NIH grants)? I know this is a broad question and that the answer is typically yes, but I thought though that MD/PhDs are more likely to get NIH grants and have less funding issues.

I think the letters after your name are not very relevant to determining your ability to get NIH money. The financial advantage of the MD is that you can fall back on clinical work if you can't get grants. Not that funders will give you a grant because you have an MD. Statistically I think you are right that MSTPs are slightly more successful than single-degree holders. But it's not a direct effect right, it goes through a bunch of mediators like research field, question, and mentor, and probably also moderators like intrinsic ability or prestige factors.

Given my track record I'm probably not the best person to say what the NIH is actually looking for, but my observations for what they're worth are that the NIH seems to like projects that are conservative, obviously highly feasible, and proposed by people who are known/reliable quantities (i.e. have had past NIH funding and demonstrated productivity) or, in the case of a training grant, mentees of people who are known/reliable quantities, with their proposal closely enough related to the mentor's work that reliability is not in question. And of course you should be working on something the NIH is actually interested in. And there's also a large stochastic factor in there for individual proposals.

Foundation/seed sources seem, IME, more willing to make a reach and fund stuff that's less conservative, a step or two farther out there. But those funding amounts are much smaller.

 

[dl2dp2]

Is it that difficult to get a cut of the pie (the NIH grants)? I know this is a broad question and that the answer is typically yes, but I thought though that MD/PhDs are more likely to get NIH grants and have less funding issues.

This is variable depending on the field. The rough number goes something like this, out of everyone who applies for a K about 30-40% get one. Out of all the Ks who apply for an R, about 50% get an R. 30% of Ks don’t apply for Rs. Depending on where you are institutionally, once you have an R the median survival for ongoing grand support is something like 10ish years. The intuition on that number here is comparable to median survival of a cancer, say. Unlike in cancer tho, it’s often people can remediate funding gap by getting money later. Also, due to favorable indirect rates, typically once people have two R01s you start to get some small amount of hard money support. As above and below said, these numbers are generally unaffected by the letters behind your name. However, my intuition is that it's MUCH more "difficult" to get to the K stage if you "only" have a PhD or an MD, but for different reasons. For PhD, the sheer number of trainees in the denominator overwhelms the overall rate of success. For MDs, the alternative career path is so attractive that the numerator is much smaller. As I said in a different thread, the overall number is something like 5-10% of PhDs ever get to the pre-K stage, whereas about 20-30% of MD/PhDs do.

This is all public information you can find on google.

 

[Neuronix]

I thought though that MD/PhDs are more likely to get NIH grants and have less funding issues.

This is an often repeated assertion that is not supported by the data. If you spin the data in certain ways you can get a *slight* advantage to having an MD/PhD when applying for NIH grants.

 

[Sir William Osler]

My opinion on this topic is that MD/PhD can help mostly in the sense that you have more years of research experience and hopefully more publications than a straight MD. In some cases especially for early career grants, you may get more respect with the extra PhD from other PhDs especially in some fields that are more basic such as molecular biology of bioengineering. For translational topic areas, it may be less important since MDs are common in many translational areas.