As you know from a previous post, in April we issued Investing in the Future: NIGMS Strategic Plan for Biomedical and Behavioral Research Training. Since then, several NIGMS working groups have been very busy devising practical ways to implement the plan’s 15 action items. Last month, we brought together about 25 stakeholders—training grant directors, other university researchers, deans, department chairs and others—to give us a reality check on some of our proposals. We were gratified to see that we are on course to implement most of the measures the plan calls for by early 2012.
In a complementary effort, the Advisory Committee to the NIH Director established a working group to examine and develop a model for the future of the biomedical research workforce in the United States. To inform the working group in its deliberations, NIH has just issued a request for information (RFI). I encourage you to respond to the RFI and send in your input by October 7 to http://grants.nih.gov/grants/guide/rfi_files/bmw/add.cfm.
Over a decade ago, the Commission on Life Sciences laid out very well many of the problems with the current training model; there is no need to go over them here. As the NIH budget doubled and no action was taken, the problems have only worsened. In the current funding environment, inaction is not an option.
NIH supports the overwhelming majority of PhD trainees through individual research grants to PIs. It would be in the interest of the NIH and the public for the NIH to develop mechanism to ensure that this money is being wisely spent. When a higher proportion of graduate student funding was coming from training grants, NIH had comparatively more oversight on the training process, including the qualifications of the trainee, the aggregate accomplishments of the trainees and their faculty mentors, and the training environment provided by the host programs. Foreign students were excluded from training grants (the rationale for this is unclear, especially as foreign trainees are supported by NIH money through individual research grants). However, as PhD programs expanded, training grants became more scarce, the domestic:international student ratios changed, and the size of individual awards increased, NIH has lost oversight of PhD training in the US.
NIH needs to exert some control on how universities are spending public resources.
There are many institutions that continue to do a fine job training future scientists, even as the career opportunities for these trainees are shifting (aside: the well-trained can and will adjust, and contribute in ever-innovative ways). The 2010 NRC rankings are data-based, rather than purely reputational, and I think it is fair to say that programs that bothered to respond to the survey, and were ranked in the top half or perhaps top two thirds are, by and large, doing a decent job with training.
Many institutions are not doing a decent job, however. Perhaps they cannot recruit trainees with a high probability of success, or do not apply rigorous standards for retention and completion. The result is that the Bursars offices of these universities are the primary beneficiaries of NIH funding, in tuition support for trainees paid from individual research grants. The students are not properly trained for the (ever scarcer) jobs for which their degrees should credential them, and their faculty mentors have squandered time and resources unsuccessfully attempting to draw blood from stones.
Does every university have to have a PhD program? Does every NIH-funded scientist need to have graduate students in their labs? Is our failure to envision other models for scientific productivity shunting public resources to unaccountable institutions?
Anyone who has ever applied for, received and renewed a training grant knows that this is a demanding process. Perhaps NIH should apply the very same rigor (with a more realistic outlook on international students) to deciding which programs should be permitted to allow ANY graduate students to be supported by NIH funds. To reiterate, there are a many programs, public and private, large and small, north, south, east and west, coastal and heartland, with strong track records for providing rigorous training for highly qualified trainees who then to on to contribute their expertise in academia, industry, intellectual properties, government and journalism. These programs would easily survive the kind of scrutiny that would be needed to make decisions on allocation of public resources. Others, quite frankly, would not.
There’s no getting around the fact that hard decisions will have to be made. A fair, rigorous, data-driven approach to determining, and supporting, training environment fitness is preferable to across the board reductions.
I would welcome opportunities to discuss this further.
While looking for updates regarding this NIH ROI, I ran across your well written comment. I think your questions:
“Does every university have to have a PhD program?
Does every NIH-funded scientist need to have graduate students in their labs?”, are excellent and I agree they should be considered.
One additional question I would add is:
“How many NIH funded graduate students and postdocs should one PI train?”
There is definitely a supply and demand issue around PhD training. Rather than focusing solely on reducing supply I think we might also investigate ways to increase demand. There are numerous PIs who have been active for over 30+ years now, continue to train many times in excess of replacement, and show no signs of retiring. Limiting the number of trainees a PI can fund with NIH dollars might serve to both reduce the number of PhDs and encourage senior scientists to retire.
Certainly technicians could be hired in place of graduate students but technicians are often more expensive than graduate students, straining a senior scientists resources once they reach their trainee cap.
There are numerous implications to imposing such a cap and I’d be happy to discuss the idea further.
Thanks for the reply, Young Scientist.
Without having put too much thought into this, I will say that I am generally not in favor of caps. If a PI can demonstrate that they are doing a good job with trainees (and, in my model, the institution is “approved” for allowing R01 funds to support trainees), then why place a limit? I do understand the supply and demand issues-there are not enough good trainees to go around, and this can disadvantage promising young PIs. But I think that systematic overhauls, both the planned ones and the ones that are necessitated by flat pay lines, might organically limit AND redistribute the number of trainees per lab, without NIH having to put explicit caps in place.
My original comment was a bit more rambling than I would want it to be (but thanks for the kind words!), so I don’t know how well I articulated my central thesis that it is institutions that need to be held more accountable to the NIH for the quality of the training environment. I was a little disappointed with the emphasis that the NIGMS Strategic Plan placed on individual PI efforts. The best efforts of the best educators will be hamstrung if the institution has not properly invested in ensuring a good training environment (non-rigorous admissions standards, no process for evaluation and dismissal of underperforming students, book-keeping tricks to ensure that everyone gets a B or better, no incentivisation of faculty to do a good job training, no good standards for completion). They will waste their time, and the public’s money, and sacrifice their own productivity, imperiling future funding. They will produce trainees who are unqualified for future contributions to the US scientific enterprise. On the other hand, good students can survive a sub-optimal trainee-mentor relationship if they are in a good training environment, where the faculty and admin are accountable to the NIH to demonstrate that they are effectively training the next generation. If programs and institutions can not demonstrate this, they should not be able to use NIH funds to support trainees. Successful programs can be given wide latitude to decide such things as which faculty members can and can’t take PhD students (in other words, make a distinction between Faculty and Graduate Faculty), or how many students a PI can take. If institutions are held accountable, then it is in the interest of the individual programs to decide these issues wisely.
So what I am advocating is something fairly draconian on the institutional level-NIH evaluates grad programs and decides which ones can continue to use NIH funds to support trainees, but then individual institutions and programs, and to a lesser extent PIs, are free to make the best decisions that will, under periodic review, sustain their status as approved training institutions. This is not to say that NIH funds (R awards, etc) will no longer go to the losing institutions-productive researchers should be funded wherever they are. But, as I said, not every university needs to have a PhD program, and not every productive PI needs to be Graduate Faculty.
Happy to keep this going. I’d love to get more attention to this topic. Thanks for engaging.
The Advisory Committee to the NIH Director Working Group on Diversity in the Biomedical Research Workforce has issued a new Request for Information. Comments are due by February 24.