R01 Size Growth and the Modular Cap

May 28 2014 Published by under Uncategorized

In a previous post, I presented data on the growth of average sizes of Research Project Grants (RPGs) from FY1983 to FY2013. The growth in the average RPG size over this period exceeded inflation, even measured by the Biomedical Research and Development Price Index, BRDPI. However, I noted that the average may or may not be a representative measure of growth since the distribution of RPG sizes has changed with the addition of large grants on one end and small grants (R03s, R21s) on the other end. I proposed the hypothesis that the growth in average RPG size could be due to growth in the number of and size of larger awards with less due to the growth in the bulk of the RPG distribution which is made up of primarily R01s.

To address this hypothesis, I have now examined full R01 distributions for years FY2003 and FY2013. From these years, the average size of an RPG grew from $380K in annual total costs to $445K, a 17% increase in nominal value. Correcting for inflation measured by BRDPI, this corresponds to a 19% drop in buying power.

How does this RPG average compare with the R01 average? This distribution of R01 sizes for FY2003 is shown below:

2003 Histogram

In FY2003, the average annual total cost for an R01 was $338K and the median was $306K. For comparison, in FY2013, the average total cost was $430K, a 27% increase (or a 9% drop in buying power, correcting for BRDPI). In FY2013, the median total cost had grown to $360K, a more modest 18% increase (or a drop of 18% in buying power).

2003-2013 Histograms

Note that the left side of the distribution shifted by approximately $50K from FY2003 to FY2013 whereas the right side of the distribution was essentially fixed. This observation will be discussed further below.

Examination of the distribution reveals the presence of more than 250 R01s with annual total costs exceeding $1M in FY2003. Removing these large R01s has only a small effect on the growth in R01 size with the average increasing from $326K in FY2003 to $401K in FY2013, a 23% nominal increase and the median increasing from $305K in FY2003 to $357K, a 17% nominal increase.

These data reveal that the growth in the average R01 size actually exceeded the growth in the average RPG size over this period of time whereas the growth in the median R01 size closely matched the growth in average RPG size. Thus, my hypothesis was not correct. The growth in the average RPG size corresponds to the growth in R01 sizes in many regards.

The distribution of the sizes of R01s in FY2003 is approximated as a Gaussian remarkably well.

2003 Gaussian Fit

A Gaussian fits the size distribution on the left side quite well. The fit is less good on the right side with an "excess" number of R01s with annual total costs larger than approximately $400K. Based on a comparison of the distribution of annual total costs with annual direct costs, presented in my previous post, the boundary near $400K in annual total costs corresponds to the boundary of $250K in annual direct costs, that is, the cap for modular grants.

The size distribution for FY2013 is fit by a Gaussian less well.

2013 Gaussian

As noted above, the left side of the distribution shifted upwards by approximately $50K from FY2003 whereas the right side was essentially fixed at the annual total cost level corresponding to the modular cap. Consistent with the notion that R01 sizes are being limited by this cap, the actual distribution is slightly narrower than the best fit Gaussian.

The "excess" of grants above the modular cap has increased by 2.5 fold from FY2003 to FY2013. This excess corresponds to 14% of the area of the Gaussian distribution in FY2003 and 34% of that area in FY2013.

This analysis reveals at least two important points. First, the modular cap on direct costs is clearly influencing the distribution of R01 grant sizes awarded and this effect is much more pronounced in FY2013 than it was in FY2003. Second, the overall trends in the R01 grant pool matches those in the overall RPG pool so that issues related to the balance between grant sizes and stipend levels are not primarily related to the choice of average RPG size as a metric. However, the influence of the modular cap as well as the impact of increased non-stipend costs (tuition, fringe benefits) raised by commenters on previous posts must still be considered.

22 responses so far

  • DrugMonkey says:

    17% nominal increase = 19% drop in purchasing power.

    You are underlining a very clear recommendation to the NIH- raise the modular cap to match inflation. This will have a tremendous stabilizing influence on grant churn and PI expectations

    • datahound says:

      NIH has tried to cope with the flat (or worse) budgets since the end of the "doubling" by trying to avoid too much of a decline in success rates. This has been accomplished by keeping average grant sizes from growing by various means including not raising the modular cap. However, BRDPI-measured inflation is approximately 36% over this 11-year period. This loss in purchasing power is real and the cost of maintaining the success rate by these means is the churn and frustrations that you note. On the other hand, larger grants means fewer funded investigators.

      The fraction of the NIH budget going to R01-equivalent grants has also fallen to 34%. The R01 success rate could also be managed by cutting back on other spending categories.

      • drugmonkey says:

        NIH has tried to cope with the flat (or worse) budgets since the end of the "doubling" by trying to avoid too much of a decline in success rates. This has been accomplished by keeping average grant sizes from growing

        But this is a strategy that is almost comically blind to reality. Declining purchasing power just makes the PIs submit even more grant applications, thereby contributing to the denominator and lowering success rates. Cutting budgets does the same thing. It can artificially prop things up for a year or so but then the inevitable tail-chasing catches up.

        On the other hand, larger grants means fewer funded investigators.

        This is not at all clear and some of the data analyses that you've been conducting support my assertions on this. There is a pool of approximately continually funded PIs. They seek to maintain, across time, a more or less consistent lab size. Maybe with a slight upward trend across the career...but this should be predictable. If the effective spending power of a particular collection of grants declines, then those individuals need more grants. They will seek them by submitting more applications.

        This really is not rocket science.

        The flip side of this is the necessary recognition that the NIH *expects* a certain degree of productivity. This is not adjusted in the minds of all participants (from peer review on up to Francis Collins) in step with declining purchasing power.

        • datahound says:

          I agree with you to some extent, but take a more nuanced view. First, these issues are clearly in the retrospectoscope than they are on a year-to-year basis. Through much of the last decade, there was hope with some basis that the NIH appropriation would get back onto a track that at least kept up with inflation (or BRDPI inflation). Of course, with the exception of the ARRA years, this did not occur. There seems to be less basis for hope in the present climate, at least for the next couple of years.

          Second, your analysis is most correct for established laboratories with multiple sources of funding. For a new lab or an established lab with only 1 major grant, the total lack of funding can have very negative consequences and introduce major inefficiencies (e.g. loss of and then rehiring of key personnel). Thus, stretching appropriations to cover more laboratories can prevent these inefficiencies assuming, of course, that the lab would recover in a reasonable period of time. I agree that approach of limiting the growth in the size of grants (well below BRDPI) does lead to the behavior that you aptly describe as tail chasing for labs that require a fixed amount of buying power to survive.

          Third, the diversity of "business models" in scientific research leads to diversity in approach from investigators. For a researcher at a free-standing research institute or at one of many private medical schools, the need to obtain grants to stay in the game is substantially greater than that for an investigator at a school of arts and sciences with a largely hard money salary and student support available through teaching fellowships.

          Finally, the statement that "larger grants means fewer funded investigators" is certainly true on a one-year basis. Over time, you are correct that this need not be true. I gather that there are discussions underway at NIH to develop a grant mechanism to allow a lab "program" to be funded rather than individual "projects." Such grants would be available only if the investigator agreed not to apply for other grants. I have mixed feelings about this approach, but it is certainly worthy of discussion.

          • drugmonkey says:

            Not sure I get your rationale wrt keeping as many labs afloat as possible. The end game there is lots of people "funded" to do nothing but sit in empty labs. Large labs that lose funding drop staff and productivity too. Not sure how the loss of funding hitting a 2 person lab is more of an impact than a 5 person lab? Either way, headway is lost.

            Just pointing out that your argument presupposes one distribution is better without any justification. Do you agree with me that a major problem is too many mouths at the trough? Or do you think we can solve the problem by reducing per PI funding so that more people have less money?

  • lurker says:

    Another very Nice analysis. I was wondering if you can overlay how many distinct PIs fall within each data point of the X-axis? In other words, is that big bump on the right in 2013 above the Gaussian concentrated in a few set of investigators? Or is it supporting a very broad number of separate PIs, such as in P and U mechs? Is the bulk of the money supporting as diverse and wide as possible a number of PIs, or concentrated in the few and powerful?

    In the case of sustainability, it would seem a better goal to distribute funds widely, plant as many seeds, let all seeds have a chance to play out, than to funnel resources to the few. Seems like 2003 was better at this than recent past?

  • datahound says:

    Thanks, lurker.

    If I take $450K in annual total costs as the cutoff for the bump, there are approximately 6780 R01s above this level. This represents 5840 PIs. Of these PIs, approximately 4000 have only 1 R01, 1400 have 2 R01s, 350 have 3 R01s, and 90 have 4 or more R01s. Noted that none of the analysis includes the effects of multi-PI grants, but this is likely to be a small correction.

    Is this the right distribution? Clearly having a large number of investigators, all of whom are underfunded, is unlikely to lead to optimal productivity. On the other hand, I have written extensively in the past on my views that very well funded investigators should be subjected to considerable scrutiny to try to ensure that their productivity is commensurate with their level of funding.

    • drugmonkey says:

      try to ensure that their productivity is commensurate with their level of funding.

      Meaning all sources, correct? HHMI, training grants, NRSA trainees galore, etc, etc.

      • datahound says:

        I agree with the all sources approach (which contrasts with the current NIH-wide policy which includes only direct NIH support.

  • Joe says:

    I think this decision of NIH to try to fund more people with less money is problematic. The post points out that inflation has eaten away at purchasing power, and a previous post describes increases in stipend costs (and benefits costs and tuition costs). Furthermore, with states cutting funds to universities, things that were once provided by the school are being charged to PIs or are just going away, including equipment maintenance fees, central services like dishwashing and autoclaving, and administrative support. We're getting squeezed from all sides. Keeping the modular cap at this low level seems to be only for PR purposes, so that they can say they are supporting more scientists.

    The same kind of thing happened with the sequester and subsequent budget. Awarded grants were paid at a reduced percentage during the sequester, and when half that money was restored for the new budget, that restored money did not result in restored award amounts. I assume it must have gone for new awards. It is clear to me now that I should have had a 15% inflated budget for my proposal, because now I am left with not enough money to do the experiments proposed and am putting off expensive experiments until I see if another proposal will get funded.

    • datahound says:

      Joe: I think you are being a little harsh saying that keeping the cap fixed is just for PR purposes but, as you note and as I think the data in this post indicate, the cap is limiting grant size growth to keep up with research costs and I would think that NIH should serious examine the benefits and costs to the present policy.

      Investigators do have the ability to submit non-modular grants up to annual direct costs of up to $500K without permission from NIH. With justification, this seems appropriate to help ensure that the science proposed aligns with the requested budget. Of course, NIH staff (with input from study sections) may not award this budget, but if one does not ask, one is guaranteed not to get the requested amount.

      • Joe says:

        If it's not PR, then is it that they are trying to maintain a large number of underfunded projects or that they think that we were all inflating our budgets and didn't really need the requested funds? Those are the only interpretations I've come up with. Maybe they are keeping a lot of labs going waiting for the day when the NIH budget will improve.

        I have a non-modular grant and always submit non-modular because that is the budget I need for the work. A number of times I have gotten back a review saying "reduce to modular, nothing exceptional about expenses for this project". I think it would be the exceptional project that could be done on a modular budget.

        • datahound says:

          Joe: I think your last hypothesis is correct. They are trying to avoid losing labs that have been productive or have great potential to be productive for a time when the fiscal climate improves. As I noted in a reply to DM, for much of the last decade, there are some reasons to be hopeful, but the increased budgets never materialized (except for the ARRA years).

          It is interesting the see the responses you have gotten to submitting non-modular budgets. I imagined that would happen but I had not heard specific anecdotes. Again, I think NIH does need to take a serious look at this policy.

          • drugmonkey says:

            Remember there was a *reason* for modular budgeting. To keep study sections from ticky tacking around with itemized budgets. My experience is that this is a good thing. I propose itemized budgets that are straight up my reality (without, mind you, the real inflation built in b/c rules). And what so you know, reviewers want to complain and cut (based on god knows what justification).

            Abandoning modular or saying PIs can just go above if they want to is not in step with reality. The only step forward is to adjust the modular limit with BRDPI.

      • drugmonkey says:

        It is a bit too easy to say PIs are free to go above cap. You know perfectly well only some can get away with that. And some ICs cut them extra compared with modular grants so it can be a useless effort to propose say $380K (to account for inflation since the modular cap was established) since you'll draw a heavier cut and your odds are lower b/c study sections.

  • Note that the left side of the distribution shifted by approximately $50K from FY2003 to FY2013 whereas the right side of the distribution was essentially fixed. This observation will be discussed further below.

    In 2003, people submitted modular budgets below the cap. Now no one ever submits a modular budget that isn't maxed at the cap.

    • datahound says:

      I think a few people still do submit below the cap, but it is small percentage and these proposals are likely mostly from investigators with substantial hard money support. This percentage has certainly dropped dramatically since 2003.

    • drugmonkey says:

      One would have to be insane...oh wait, I did that.

  • datahound says:

    DM: Thanks for your various comments. I certainly do agree that the major problem is "too many mouths at the trough." However, I think it takes some careful thinking to figure out what determines the "right" number. As is usually the case, both limiting cases are clearly wrong. Having a larger number of investigators all of whom are substantially underfunded is certainly underfunded. At the same time, having an essentially closed system with very few opportunities for new investigators and ideas because a relatively small number of investigators consume the bulk of the funds is also problematic.

    I also agree with you about the modular cap issue. If NIH needs or wants to limit grant sizes, there are ways to do that, but the lack of adjustment of the modular cap is almost certainly introducing inequities and unstated inequities in the system.

    From my perspective, there are at least three sources of the problems. First, the doubling and associated policies (e.g. with regard to facilities cost recovery), encouraged many institutions to build new research space with the model that the space would be filled with well-funded investigators bringing in both direct and indirect costs. This led to the expansion of faculty (and increases in the sizes of training programs but that it is separate issue). Second, the doubling was followed by very limited increases in the NIH budget in nominal terms and substantial decreases in buying power (measured by BRDPI). This has limited the ability of the new pools of investigators to achieve adequate funding and has decreased the ability of institutions to cover the costs that they incurred in anticipation of and during the doubling. Third, the NIH has moved away (as a percentage of the budget) from R01s toward both larger mechanisms (some of which have been quite productive but other not) and toward smaller mechanisms (such as R21s which, in my experience do not produce much sustainable research effort). At this point, 40% or less of the NIH budget is going to R01-like funding mechanisms. This is contributing to all of the problems above.

  • […] - $318K) = $146K. This distribution is very similar to that for all R01s for FY2013 presented in an earlier post.  Approximately 1/3 of the grants fall outside of a Gaussian distribution centered around the mode […]

  • […] may be exacerbated by the modular cap which has remained unchanged since it was created and evidence indicates that PIs may be requesting $250K at the cap rather than requesting a larger budget and running the […]

  • […] years ago, I posted an analysis of the distributions of the sizes of R01 awards from NIH and the relationship of this distribution […]

Leave a Reply