Brand new T32 outcome data

[dl2dp2]

NIH R01 Grant Outcomes of T32 Postdoctoral Participants: Associations with Demographics, Cohort, Degree, and Career Development (K) Awards – NIH Extramural Nexus

nexus.od.nih.gov

Pretty self-explanatory there. Basically verifying every thing I've ever said on this forum based purely on "clinical observation". Big cohort effect for MDPhD vs. PhD in the last 20 years (this is in apparent contradiction but in reality complementary to @Neuronix earlier data that MDPhDs don't have a higher success rate once an application is submitted--the effect mainly relates to lower drop out rates). Keep in mind T32 postdocs are perhaps at most 1/2 of total postdoc spots, as most postdocs are ad-hoc spots offered by R01 budgets rather than dedicated T32 training programs. So suffice it is to say, if you take any arbitrary postdoc in America right now, her chance of eventually getting an R01 is below 10%.

---

Members don't see this ad.

 

[Fencer]

I saw that yesterday. The Fig. 2 graphing the outcome of MD/PhD vs PhD vs MD for R01 success rate is quite self-explanatory, but lower than the survey of MD/PhD graduates data set (43%) or the PhD outcomes study from NIH (18%). It does not mean that the other >75% are not doing quality research in bench, translational, clinical, or health care outcomes. It also does not include those investigators supported by other means like VA Merit awards.

 

[tortuga87]

My gut reaction with this data is that T32s are very risky for 5-15 years of your working life. With the MD/PhD program, you get research training, basic clinical training, med school paid off as well as a shot at residency. I see it as equivalent to making $75K per year (stipend + med school payments) to improve yourself. With the T32, you get to do more research but at a significantly reduced income without job security (~43% of getting any funding per above, tenure discussions in previous threads). Some people choose to pay a premium for job security, but this pathway does not even offer that.

It seems like most MD/PhDs would choose not to do T32s but for some reason still so many apply. Are there any benefits to a T32 besides getting ready to apply for grants?

 

[SurfingDoctor]

My gut reaction with this data is that T32s are very risky for 5-15 years of your working life. With the MD/PhD program, you get research training, basic clinical training, med school paid off as well as a shot at residency. I see it as equivalent to making $75K per year (stipend + med school payments) to improve yourself. With the T32, you get to do more research but at a significantly reduced income without job security (~43% of getting any funding per above, tenure discussions in previous threads). Some people choose to pay a premium for job security, but this pathway does not even offer that.

It seems like most MD/PhDs would choose not to do T32s but for some reason still so many apply. Are there any benefits to a T32 besides getting ready to apply for grants?

I can't speak for PhD and post-docs, but for MD and MD/PhDs, the T32 gives you extra protected time in your training that you wouldn't be afforded otherwise. For instance, just went straight MD and didn't do a PhD. In fellowship training, which is clinically heavy, I was on a T32. Now, that extended my fellowship training by 1 year (obviously at a financial cost) but I view that extra time as crucial to any later successes I had. Most clinical training programs offer "research" electives or require "scholarly work", but frankly they are halfa-s and mostly useless. That's fine for a majority of physicians because they don't care anyway and research to them is about sending out surveys or learning how to be better teachers or none of the above. But for those you actually want to pursue research, it's woefully inadequate and thus any extra, protected time one can get is a must.

However, saying all of that, while I appreciated the time I had on a T32 and think I learned a lot, none of that specifically alters the facts that getting R01 funding is mostly a matter of persistence, timing and luck.

 

[tr]

It seems like most MD/PhDs would choose not to do T32s but for some reason still so many apply. Are there any benefits to a T32 besides getting ready to apply for grants?

Because nobody is going to hand you a research assistant professor job straight out of residency unless you are some kind of wunderkind. Applying for grants is the point, they are pretty much a prerequisite for a research position.


Separately: Gee, look at the dropping success rates on the Ks over time. Same number of awards, more and more applicants.

Application	Yes	1898 (16.3)	2423 (20.6)	2602 (21.4)
PI on K Award	Yes	1313 (11.3)	1512 (12.8)	1411 (11.6

 

[Neuronix]

I am reserving comment because I have questions about the data. I e-mailed the author of the blog post.

 

[SurfingDoctor]

Because nobody is going to hand you a research assistant professor job straight out of residency unless you are some kind of wunderkind. Applying for grants is the point, they are pretty much a prerequisite for a research position.


Separately: Gee, look at the dropping success rates on the Ks over time. Same number of awards, more and more applicants.

Application	Yes	1898 (16.3)	2423 (20.6)	2602 (21.4)
PI on K Award	Yes	1313 (11.3)	1512 (12.8)	1411 (11.6

This is also a good point and another "reason" that one would want to do a T32. This isn't really directed at you tr, but I'm using your comment as a jumping off point.

For better or worse, after graduating training and looking for a job, and more specifically a research position (and I speaking more to tenure track assistant professor as that is what I know), there is a little bit of having to justify why an applicant is worth the departmental and institutional cost to fund such a position. I suppose it wouldn't be always the case, but entering the TT assistant professor position is usually associated with direct, start up costs to the department/institutional. Now, maybe someone straight out of training is a wunderkind and viewed as a relative slam dunk, but that is a rarity. Instead, the institution is making an investment on someone that their hope it will pay off for. In my experience, that can kinda be a crapshoot, but being on a T32 during training is another tick that they can use as a measurement to hedge they potential investment, as it is usually associated with some sort of personal sacrifice. I mean, I'm in a field with a very low number of physician-scientists, but speaking from experience, if one doesn't go on a T32 during training to more of less "prove" themselves, they probably won't be offered a TT assistant professor position. Again, that is not to say it's the right way to do it (and honestly, I think it probably misses out on potentially good or better applicants), but in my own department, if you aren't on a T32 during training, the chair will not write a LOS for a K award.

 

[SurfingDoctor]

Also:
Table 3: Cox Proportional Hazards Model for First Competing R01 Award among T32 Post-doctoral Matriculants for Those with at Least 7 Years of Follow-up

Predictor	HR (univariable)	HR (multivariable)
Post-Doctoral Cohort
1995-1999	-	-
2000-2004	1.02 (0.95-1.11, p=0.558)	0.97 (0.90-1.05, p=0.458)
2005-2009	0.89 (0.81-0.99, p=0.025)	0.84 (0.77-0.93, p=0.001)

Not surprising but its nice to have data that backs it up. It just ain't what it used to be.

 

[CheGuevaraMD]

Table 3 shows that K awards are the real bar. T/F32 is not that special anymore. They’re almost as big as PhD programs at some institutions.

You have to win a K before you’ve ‘made it’ (have a good chance of winning a R01) in biomedical academia these days, which is rough, because most people are 35 (straight PhD or MD) to 40 (MD/PhD) when they apply.

 

[Neuronix]

The person who wrote the article would not answer my questions at this time.

Figure 2 and 7 taken together do not make sense unless MD/PhDs get more R01 grants without first receiving K grants.

The discussion states that MD/PhDs apply more often for grants. Yet, rates of MD, PhD, and MD/PhD grant applications are not presented.

I can only assume that MD/PhDs are more likely to apply for R01 grants and more likely to receive R01 grants without first receiving a K grant. I think sluox is probably right based on the other existing data that T32 funded MD/PhDs are more likely to write R01s than other T32 funded degrees, though their rate of success in receiving those grants is no different.