Comment on Proposed Pilot to Support NIGMS Investigators’ Overall Research Programs

29 comments

We’re planning an experiment in how we fund research, and we want your input. As outlined in the Request for Information (RFI) included below, we propose to create a pilot program called Maximizing Investigators’ Research Award (MIRA) that would support all of the projects in an investigator’s lab that are relevant to the NIGMS mission.

We expect that the MIRA program will offer a number of benefits. For instance, investigators would not have to break their work into smaller, strictly prescribed increments. In addition, the program could improve funding stability and enhance grantees’ flexibility to follow new research directions as opportunities and ideas arise.

It’s important to note that MIRAs are not intended to be a method for supporting only a perceived elite group of investigators or promoting only high-risk, high-potential-reward research.

Our intent is to pilot a program that might transform how we support fundamental biomedical research, creating a more productive, efficient and sustainable enterprise. I encourage you to read the proposal and share your comments using the RFI input form (no longer available) by the August 15 deadline. We welcome responses from both individuals and organizations.


Request for Information (RFI): Soliciting Comments on a Potential New Program for Research Funding by the National Institute of General Medical Sciences

Notice Number: NOT-GM-14-122

Key Dates

Release Date: July 17, 2014

Response Date: August 15, 2014

Related Announcements

None

Issued by

National Institute of General Medical Sciences (NIGMS)

Purpose

This is a time-sensitive Request for Information (RFI) directed at obtaining input to assist the National Institute of General Medical Sciences (NIGMS) in its planning for a potential new program tentatively named Maximizing Investigators’ Research Award (MIRA). This award would be a grant in support of all of the research supported by NIGMS in an investigator’s laboratory.

Background

Supporting basic research by funding individual projects has a number of consequences for the efficiency and effectiveness of the basic biomedical research enterprise in the U.S. (Alberts, 1985; Ioannidis, 2011;  Vale, 2012; Bourne, 2013; Alberts et al., 2014). To address these issues and increase the efficiency and efficacy of its funding mechanisms, NIGMS is considering a pilot program to fund investigators’ overall research programs, which represents a compilation of the investigator’s research projects. It is hoped that this new funding mechanism will achieve the following:

  • Increase the stability of funding for NIGMS-supported investigators, which could enhance their ability to take on ambitious scientific projects and approach problems creatively.
  • Increase flexibility for investigators to follow important new research directions as opportunities arise, rather than being bound to specific aims proposed in advance of the studies.
  • Improve the distribution of funding among the nation’s highly talented and promising investigators to increase overall scientific productivity and the chances for important breakthroughs.
  • In the long term, reduce the time spent by researchers writing and reviewing grant applications, allowing them to spend more time conducting research.

Overview of the proposed NIGMS MIRA program

  • An NIGMS MIRA would provide support for a lab’s research program, which represents a compilation of the investigator’s NIGMS research projects (research areas supported by NIGMS are outlined at our website). Researchers would have the freedom to explore new avenues of inquiry that arise during the course of their work as long as those avenues are relevant to the mission of the Institute and do not require additional review for regulatory compliance (e.g., new human subjects research).
  • An NIGMS MIRA would be renewable.
  • Funding would range from $150,000-$750,000 (direct costs/year), depending on recommendations of the study section and the National Advisory General Medical Sciences Council as well as staff evaluation of the needs and expected productivity and impact of the program. Support for the investigator from sources other than NIGMS would be taken into consideration when deciding on funding levels for an NIGMS MIRA.
  • Up to $150,000 in administrative supplement support for the purchase of new equipment could be requested by an NIGMS MIRA grantee per grant cycle. Decisions on these requests would be made by NIGMS staff and the National Advisory General Medical Sciences Council based on an assessment of need and the potential impact of the new equipment on the research. The number of supplements given would depend on the available funds.
  • The median direct costs for NIGMS MIRAs would be higher than the current median R01 direct costs at NIGMS.
  • The length of an NIGMS MIRA would be 5 years, which is longer than the current average for an NIGMS R01 of close to 4 years.
  • A researcher funded by an NIGMS MIRA would not be given any other sources of NIGMS funding with the following exceptions:
  • Grants supporting research resources
  • Grants supporting training, workforce development or diversity building
  • Funding for clinical trials
  • SBIR/STTR grants
  • Conference grant
  • The Program Director/Principal Investigator (PD/PI) would be expected to commit at least 50 percent research effort to the NIGMS MIRA.
  • Revision applications to allow new collaborative work might be included as part of this program.
  • Review of the application would emphasize a holistic evaluation of the investigator’s track record and the overall potential importance of the proposed research program, without focusing on specific project details. Specific aims would not be required. The process would include peer review using existing criteria and processes but would be tailored to address the particular features of the MIRA (see the section below on the Possible Peer Review Process).
  • To avoid the abrupt termination of research groups from an adverse round of peer review, NIGMS MIRAs could be ramped down from one funding level to a lower one that is more consistent with the recent and perceived future productivity of the group and the importance of the work, as assessed by the study section. Conversely, a renewing program could have its budget increased if the perceived productivity, impact and needs merited it. As per standard grants policy, NIGMS program staff would make decisions on funding levels, guided by the recommendations from study sections and the National General Medical Sciences Council.

Possible Peer Review Process

In addition to the standard review criteria, among the considerations reviewers would be asked to address in reviewing MIRA applications are whether:

  • The proposed research effort is substantive, broad and ambitious.
  • A PD/PI’s record shows evidence of productivity, creativity, adaptability, service and excellence in mentoring.
  • For Early stage Investigators (ESIs), there is evidence of productivity, independent research and contributions to the design and direction of past research efforts.
  • The proposed research includes evidence of creativity and the incorporation of novel approaches as appropriate.
  • There are sound bases and generally well-thought-through and reasoned approaches for the proposed research effort.
  • There is evidence that the PD/PI has considered alternative approaches, outcomes, models and directions that might inform the scientific questions being posed.
  • The work will be conducted carefully and cost-effectively, with good stewardship of the data generated.
  • For ESIs, there is evidence of institutional support and mentoring.

Possible Implementation Plan

Because this is a pilot program, implementation must be carefully phased in and outcomes and unintended consequences assessed along the way. One possible implementation plan, consisting initially of two paths, is outlined below.

  • In lieu of a competitive renewal (Type 2), PDs/PIs who currently have two or more NIGMS R01s could apply for an NIGMS MIRA. Application for a MIRA would be evidence of a willingness to relinquish all other NIGMS research grants upon award. Award of the MIRA would be contingent on relinquishing other current NIGMS research grants in favor of the MIRA. Applicants proposing to consolidate their NIGMS awards would have to submit a MIRA application that would undergo peer review. The budget would be higher than that for any of the individual awards the PD/PI has, but usually less than the total of all of his or her NIGMS support. The length of the NIGMS MIRA would be 5 years.
  • The program could be open to applications from ESIs. This would bring a cadre of ESIs into the system who could be directly compared to other NIH-supported ESIs funded through traditional mechanisms. MIRA would be considered a substantial, independent NIH research award that disqualifies an individual from classification as an ESI.

The MIRA Funding Opportunity Announcement, when ultimately published, will include metrics that will be used to evaluate the success of the program. Once the program is established and indications of success have been measured, additional groups of investigators would be invited to apply for NIGMS MIRAs. If the program becomes successful, it would ultimately be open to applications from all investigators working on topics relevant to the mission of NIGMS and could become the primary research funding mechanism used by the Institute.

Information Requested

NIGMS is planning to issue a Funding Opportunity Announcement (FOA) to test this new program on a pilot scale. To aid in planning, the Institute is seeking feedback from the scientific community. NIGMS invites comments on the topics below; however, comments are not limited to these topics.

  1. The merits of this funding program for established and early stage investigators.
  2. The likelihood that established and early stage investigators would apply for NIGMS MIRAs.
  3. Concerns about the NIGMS MIRA proposal.
  4. Suggestions for changes to improve the NIGMS MIRA proposal or associated processes.

Submitting a Response

All responses must be submitted to https://www.research.net/s/NewGrantProgram_gov (no longer available) by August 15, 2014. Responses are limited to 500 words per topic.

This RFI is for planning purposes only and should not be construed as a solicitation for applications or an obligation on the part of the government. The government will not pay for the preparation of any information submitted or for the government’s use of that information.

The NIH will use the information submitted in response to this RFI at its discretion and will not provide comments to any responder’s submission. However, responses to the RFI may be reflected in future funding opportunity announcements. The information provided will be analyzed and may appear in reports. Respondents are advised that the Government is under no obligation to acknowledge receipt of the information or provide feedback to respondents with respect to any information submitted. No proprietary, classified, confidential, or sensitive information should be included in your response. The Government reserves the right to use any non-proprietary technical information in any resultant solicitation(s).

Inquiries

Please direct all inquiries to:

Peter C. Preusch, Ph.D.
National Institute of General Medical Sciences (NIGMS)
Telephone: 301-594-0827

Helen R. Sunshine, Ph.D.
National Institutes of General Medical Sciences (NIGMS)
Telephone: 301-594-2881

29 Replies to “Comment on Proposed Pilot to Support NIGMS Investigators’ Overall Research Programs”

  1. A good idea, but why not extend it to those who have had long-term NIGMS R01 funding, but do not have it right now?

  2. I fully support the NIGMS MIRA pilot program. I also hope it is successful and that it will become, if not the norm, at least a significant fraction of NIGMS funding. The current system is artificially constrictive, as is recognized by many of the most productive researchers. The current system also forces PIs to apply for more than one grant to cover disparate research that is appropriate for a creative researcher to wish to pursue.
    Regarding the specific questions asked of responders: MIRA would be ideal for established investigators. It would be more difficult for early stage investigators to break into, but more young investigator grants could compensate.

  3. Sounds like a fabulous idea. Any program that places the emphasis on individual productivity of the disaster that traditional peer-review has become is most welcome.

    I imagine that you will have all sorts of issues such as whether to include grants held from other Institutes, the impact on collaborative projects across Institutions that an investigator already holds, and whether a P01 sub-grant of the same size as an R01 counts toward the eligibility. Two of these would affect me, but in principle it is definitely something I would apply to in the future.

    Good luck with it

    Greg

  4. This is a surprising response to the ongoing crisis. NIGMS has so few dollars to spread among so many wonderful scientists pursuing so many interesting topics. Colleagues whose empty labs and promising ideas have lost all funding are the major issue. Relieving the burden of those writing grants to maintain a lab of a dozen pales by comparison.

  5. Congratulations on your vision. These are the best news I have heard from NIH. I’ve been funded from nigms for about 7 years and this institute has been absolutely fantastic to me and helped me tremendously in my career, however I find individual grants are too narrowly defined. Supporting a broad research program will help tremendously to balance funding and plan ahead better.
    Thank you for such innovative idea.
    Antonio

  6. The idea is outstanding. This type of grant should have been instituted many years ago.
    There is a problem with the first path. Investigators may have had their second grant directed to another institute. Will that actually prevent them to apply to this initiative? Investigators coming off an HHMI award may have one or no NIGMS grants; will they be not eligible? It would be more reasonable to have the first path open to everybody and have a path for special consideration for ESIs.

  7. Great idea. Would give PIs much more flexibility than chasing down the data to support that specific aim or holding back publishing data just to look good on a grant.

    I have a concern about the % effort since some institutions require the salary recovery to match the effort. Might want to consider a capping the direct cost that can be applied to the PI salary (e.g 20% of the award). If that hits 50% recovery without exceeding the 20% then that’s fine – but what you are looking to do is avoid cases where the majority of the award is spent for recovery of the PIs salary…

  8. This sounds like a good idea to me. My only potential concern is with the proposed possible implementation plan: starting out by considering only PIs who have 2 or more existing R01s runs counter to the statement that “It’s important to note that MIRAs are not intended to be a method for supporting only a perceived elite group of investigators.”

  9. Many thanks for the input so far. Please keep it coming—and please remember to share it using the RFI form (no longer available).

    Regarding Steve Kron’s concerns, the goals of the NIGMS MIRA program include increasing funding stability for investigators and improving the Institute’s ability to optimally distribute funding among the many highly talented researchers studying problems relevant to the NIGMS mission. We hope that these improvements in stability and efficiency, combined with our ongoing efforts to rebalance our portfolio to renew our commitment to investigator-initiated research, will help ameliorate the problems Steve describes.

    The question of who should be eligible for MIRA during the first part of the pilot phase is an important one. Ultimately, if the program is successful it will be open to anyone working in a scientific area supported by NIGMS. During the initial roll-in phase, however, there is a danger that we could exacerbate the problems Steve brings up if not enough of the most highly funded NIGMS investigators enter the program. We will continue to think carefully about how to make the MIRA pilot available to as many investigators as possible, as soon as possible, without causing unintended, negative consequences, and we welcome your suggestions for approaches we could take.

  10. Although I think MIRA is a wonderful idea, I also urge you to rethink the proposed implementation plan. The labs in most need of stable support are those supported by a single R01, as funding gaps are most devastating for those that lack other significant resources. It is also my impression that NIGMS-funded female scientists are more likely to run their labs on a single R01. (These data should be readily available). If this is indeed the case, limiting the implementation to the most highly funded investigators could also have the unintended effect of increasing gender disparities.

  11. I think it is important to work through the numbers about this pilot program. By my quick search, there are approximately 500 PIs with more than one active NIGMS R01 grant. How many of these are potentially eligible to consider applying for a MIRA grant and how many would you expect/hope to participate in the program? Also, as Anonymous notes above, by my estimation, about 80% of these PIs are male so potential gender disparities need to be considered carefully.

    1. Jeremy,
      This is a very interesting statistic. What is the total number of PIs with NIGMS R01 grants? Are the PIs with multiple R01s the top 10%? 1%?

  12. I am SO excited to see this type of funding approach being considered by NIGMS. I wrote a blog earlier this year describing basically the same approach.

    http://watersheding.wordpress.com/new-nih-funding-model-fund-people-not-projects/

    I really believe the system would be improved if we strove to fund people not projects. I think a key consideration when reviewing the applicants is to normalize productivity to funds received. Obviously, a researcher receiving more funds, including grants AND graduate and postdoctoral fellowship support, will have greater overall productivity. We must make sure we are not comparing apples to oranges or a system like this would just become a positive feedback loop where “the rich get richer”.

  13. This is an OUTSTANDING idea. The concept of funding the researcher rather than a specific project is what we should be doing. Furthermore, the ability to ramp up and down funding depending on success makes perfect sense. I have never understood the “all or nothing” approach to funding that currently exists, where funding just stops at a certain score, as opposed to modulating the amounts of the awards based on the scores or success. There will be kinks to work out, and the numbers should be carefully analyzed to make sure that this will not result in the rich getting richer or a loss in total number of funded investigators. But overall I really like this!

  14. Overall this is an excellent, forward-looking proposal to simplify the support of productive research groups. This mechanism would have been ideal for me, since I have used two NIGMS grants to support my research for more than 40 years. Over this period of time each grant has been reviewed about 10 times. If the MIRA mechanism with five years of funding had been available from the beginning, only about 8 applications would have been reviewed and processed, saving everyone time and effort.

    I see two problems that will have to be managed:

    1. MIRA would have precluded my participation in a NIGMS program project grant with a variety of collaborators over the past ten years. You may want to think about how a MIRA participant might contribute to that sort of larger scale effort.

    2. The requirement to devote 50% effort to the MIRA program will be problematic for those at institutions that provide faculty with 9 months of salary. Managing the extra 25% effort on the MIRA will require one of two equally negative strategies: (i) The MIRA investigator could ask for 50% salary on the grant and negotiate with their employer to recover the 25% of their hard money salary to support their research. (ii) The MIRA investigator could request only 25% of their salary on the grant but commit 50% effort. In this case the employer would have to pay the overhead on the 25% of “contributed effort.” Employers are very reluctant to do this. Obviously employers should be rewarded not penalized for paying faculty salary to work on federally funded research, but this would require a change in the federal accounting practices.

    1. It’s 51% of the research effort a PI must commit… not 51% of their total effort. Not unlike the Pioneer requirements. Look at total research effort and over half must go to the MIRA. I believe this is not a problem for 9 month salaried folks.

  15. Five years is insufficient to affect the PI perception (and let us face it, reality) of instability. If you really want to affect that aspect the key is to extend the interval of high-confidence funding.

    I favor an increase in the use of R37/MERIT, myself. Could also be done with clearly enhanced paylines for competing continuations but that runs the risk of study section pushback (see ESI paylines) that ruins the outcome.

  16. The program sounds potentially very valuable both for established investigators who prefer not to artificially segment their research programs and for new investigators who have a set of great ideas and a strong track record of making new and innovative science happen. The discussion of the program online is also tremendously valuable and well thought out. Details of how to administer the program, how to ensure that institutions accept a single-award model for investigator productivity (rather than pressuring investigators to obtain additional grants ‘at any cost’ in time and effort), and of how to avoid a obvious-superstar-already-annointed-by-many-other-accolades focus of the program will all be challenging but worthwhile.

    I have just one additional suggestion, not related to this program but related to the process. I wonder if ALL new RFAs can be subject at a draft stage to an initial comment period as this program has. Such a process would be tremendously useful in making sure that any science-specific RFAs that are truly needed are indeed optimized for best impact on the scientific future, and (undoubtedly) to obtain feedback from a larger community before a commitment is made to weed out any RFAs that are too narrow or are mis-focused in terms of a true return on investment.

    Andy Fire (Stanford University School of Medicine, Departments of Pathology and Genetics)

  17. The main problem (see Kron comment) is that as proposed, it is not likely to do much to help the smaller labs that are being shut out of the research funding loop. And although it will reduce the burden of grant writing for multi-grant PIs, it will effectively lock up funds of up to 3 standard R01s ($750K/yr), making it even harder for smaller labs to compete. The program should be altered to target two types of laboratories. Example, have L-MIRA (Large MIRA) and S-MIRA (Small-MIRA) applications, broken down into two dollar ranges, funded by two separate pools of money. The amount available in each pool needs to be scrutinized carefully.
    S-MIRA $150-300K per year.
    L-MIRA $300-750K per year.
    And, importantly begin the PILOT program with S-MIRA grant applications. This will both test the program and address the burning-research-house problem by saving small labs before they perish. Hold off on L-MIRA applications for a couple of years while the research enterprise is stabilized.

  18. FANTASTIC! I’ve been pushing a PI-based (as opposed to project-based) basis for evaulation for more than a decade, and never thought it had any chance of happening. Past performance is by far the best predictor of future accomplishments, not poor guesses by grant reviewers who are often biased and/or marginally knowledgeable. A PI-based mechanism has numerous advantages; much less time spent writing and reviewing grants; easy dealing with overlap issues and gaming the system; allows innovation by putting the risk on the investigator who has the greatest stake in the outcome and the best judgment; avoiding the quantized nature of individual project grants; must more stable funding for individual labs.

    A few ideas:
    1. PI record should be biased towards more recent accomplishments but also include (on a sliding scale over time) past ones. This helps balance the new vs. senior investigator issue. First-time investigators have significant track records over a 5-10 year period which helps evaluation, long-time very successful investigators get deserved benefits, and long-time investigators whose output has diminished get some credit for their earlier work but also lose points on recent productivity.
    2. I like the idea that a PI is given a score analogous to a chess rating, the score being based on comment 1 above. Over time, the score can go up or down, depending on what happened during the previous granting period. And, I think the simplest mechanism is that their is a simple scale that links the score to dollars. This way, if the PI improves of declines during a grant period, their future funding goes up or down, but not all-or-none (unless there isn’t any productivity)
    3. There is an overlap issue that needs to be dealt with, namely funding from other sources. For example, if a person has both GM support and CA support, the PI evaluation should either be done only on the GM part, or it can be done on all support, but the dollars given by GM reflect the percent effort for the GM part. Similar issues will occur with private foundation money or project-based grants from other NIH institutes or programs.

    I can’t say how much I think a PI-based evaluation should be implemented. It would be the best thing the NIH has done in 50 years for funding research.
    Kevin Struhl

  19. Thanks for pursuing this and requesting feedback! I think the MIRA program could help solve several major problems – incontinuity of funding, the burden of writing and reviewing increasing numbers of grants, and the constraints of aims-based awards on research. I saw Dr. Lorsch give a great talk on the MIRA program. He noted that the research return-on-investment declines after a PI receives more than ~ 300k a year, and starts to dip negative after about ~$750k / yr. It would be wonderful if MIRA could address this, leading to more funded labs with a bit less money each. The current situation has 5% of the PIs receiving 25% of NIH funding, and the top 20% get 50% of funds (Dr. Lorsch’s numbers).

    To be successful, MIRAs should be structured such that scientific “elite” don’t get most of these grants. I second Steve Hanes’s idea of two separate MIRA programs (Large and Small) with separate competition pools. I would also suggest that the size of MIRA awards should be scaled in consideration of a PI’s current funding from all public and private sources (e.g. 1M / year HHMI awardees probably shouldn’t also get $750k / yr MIRA awards). Again, this is due to the lower ROI.

    For ESI’s, their reputation is tied up closely to those of their mentors (for better or worse). I suggest ESI MIRAs should focus on the ideas presented and the likelihood of strongly impacting one or more fields.

  20. My initial reaction to this was positive (I have 2 R01’s in very different areas). However, if someone went down this path, they would be putting all their eggs in one basket. Loss of 1 grant is a disaster, but loss of 2 would be a catastrophe. It is much like any issue of diversification – it would be better to have the totals/grant increase and the funding %tile more reasonable despite what appears to be less grant writing/reviewing. Of course the winners would think it’s a great idea!

  21. I think this is the best idea to come out of the NIH in all of the time I have been in Academia. My worry would be that it would be rolled out on a limited basis only with too little money devoted to it and it would fail to impress then. The success rates and the impact would be too low. My hope is that you go all the way and really give us approach a full try. I am certainly applying for it. Can’t wait!

  22. The NCI has recently announced the Outstanding Investigator Award (R35) to support investigators with outstanding records of productivity in cancer research with up to $600,000 in direct costs per year for 7 years to provide funding stability. The Canadian Institutes of Health Research program has already implemented a similar design, with established investigators being awarded 7-year grants and new/early career investigators being awarded 5-year grants. Why is MIRA for only five years? And, perhaps more importantly, why are regular GM grants still for only four years? Regular grants should be funded for 5 years and MIRA grants for 7. This will help everyone to save time and money.
    The average period for completing and publishing scientific studies in the biomedical field has increased tremendously. (In parallel, the average period for finishing a Ph.D. or a postdoctoral fellowship has also increased from approximately two years in the 1980s to more than five years currently.) This increase is mostly due to the fact that it is difficult/impossible to publish smaller, sequential scientific observations that will only form a complete study over time. Currently, each paper needs to be a complete study (or “story,” as editors like to call them). Because of this situation, the GM practice of awarding only four years is antiquated and creates a burden for scientists (who need to apply for renewal every three years), reviewers, and NIH administrators. Dr. Lawrence Tabak, in a recent piece published in “Peer Review Notes May 2014,” says that the NIH should “Reduce perverse incentives by… …providing longer-term support for investigators.” Similarly, in a Nature commentary, coauthored by Dr. Tabak and Dr. Francis Collins (January 2014), it is stated that “The NIH is also considering providing greater stability for investigators at certain, discrete career stages, utilizing grant mechanisms that allow more flexibility and a longer period than the current average of approximately four years of support per project.” Therefore, the new MIRA program planned by GM could function as a part of this strategy. However, it would be even more beneficial to a larger scientific community to increase the number of years funded by each regular GM R01 from four to five. Funding grants for only four years is anachronistic, and it overwhelms a system that is already above and beyond its capacity.
    I personally would favor the use of systems similar to the R37/MERIT (which, when was implemented, used to reward grantees who obtained high scores twice in a row), or any other system that provide stability to awardees who have done well during the previous cycle. For example, competing renewals could be reviewed separately from new applications and judged in large part on the impact of the work produced during the previous funding cycle.

Leave a Reply to Dmitri Petrov Cancel reply

Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.