See the Glossary for more terms.
Strategy for NIH Funding
This page supplements Choose the Grant in Part 2. Pick and Design a Project.
Exploratory/developmental research grants (R21) provide investigators with funding to introduce novel scientific ideas, model systems, tools, agents, targets, and technologies that have the potential to substantially advance biomedical research.
On this page we lay out some considerations you'll want to pay attention to if you're thinking about submitting an R21.
If you’ve been thinking about submitting an R21, you’ll want to know the pros and cons.
R21s can serve different purposes for different types of research and career stages. But they’re not for everyone, and knowing the caveats will help you decide whether an R21 is a good option for you.
Starting with the reasons to apply for an R21, people use them for at least three purposes. The first one follows NIH's definition (see the sidebar) to encourage exploratory and even high-risk research.
The second purpose is for more mainstream applications with projects that are smaller than would be appropriate for an R01.
And the third purpose is for investigators seeking funds to generate preliminary data for a future R01 (we discuss that topic in detail in the next section).
While any one of those approaches may be appealing, consider the drawbacks.
If you're applying for an R21 to gather preliminary data, you will not have an easy time getting funded, as you'll read below. Even if you do succeed, you could end up coming up short for time with a two-year grant.
People often find that a two-year grant is not long enough to complete a project that yields enough data for publication or preliminary data for an R01 application. To avoid a funding gap, you'll need to plan how to continue supporting your research if your R21 funds run out before a future R01 can begin.
Applying successfully for your follow-on R01 can take longer than you may think.
Estimate from 5 to 20 months from application to award depending on several factors, including the weightiest: whether you will need to resubmit. If you end up on the long end of that continuum, your R21, which you can’t renew, may end well before we could fund your R01.
If you’re a new or early-stage investigator, you’ll need to pay particular attention to timing issues.
While you work on your R21, time will march on, moving you closer to the end of the 10-year period when you qualify as an early-stage investigator.
Further, your R21 won't benefit from our higher R01 payline for new (including early-stage) investigators. Your R21 application will also not be eligible for our R56-Bridge award or selective pay programs.
Despite these hitches, R21s can be a useful part of a grant portfolio. You can find samples of successful applications at Sample Applications and Summary Statements. The first four on the list are examples of smaller projects, and the last two are examples of high-risk applications.
Preliminary data are not required for an R21 application, but they correlate with funding success.
In 2012, we randomly selected and reviewed 200 scored R21 applications (roughly 25 percent of scored 2011 R21 applications assigned to NIAID), counting resubmissions once, to assess the presence or absence of preliminary data. To count as having preliminary data, the application had to include actual data, not just a reference to earlier studies.
Starting with the big picture, our cohort had the following characteristics:
These data show that people overwhelmingly (93.5 percent) describe preliminary data in their R21 applications.
We don't know why. Perhaps they believe that, regardless of requirements, preliminary studies strengthen their case, or they are meeting reviewers' expectations.
Whatever the rationale, the behavior speaks for itself. And that behavior is supported by the differential success rates shown in the bullets above.
NIH did not intend the R21 to be a means for new investigators to obtain their first NIH grant, and there is no evidence that R21s provide a path to an independent research career. So should a new investigator apply for an R21?
Be aware that, even though NIH does not require preliminary data for R21s, most applications include it, and reviewers tend to expect it.
In 2012, we looked at the fate of new investigators submitting an R21 application and the relationship between preliminary data and success. (Note that officially NIH new investigator status—never funded with an R01 equivalent grant—applies only to R01s; we are using it here as an analysis factor only.)
First observation: the R21 is popular with new PIs.
In our 200-application cohort, 69 (34.5 percent) of scored applications were from PIs (or sets of investigators) that qualified as an NIH new investigator.
In data from another study that included unscored applications from new PIs, that number was 50 percent. For the sake of comparison, NIH 2010 data show that new investigators submitted 31.8 percent of competing R01 applications.
But closer scrutiny reveals a negative for new PIs: in our R21 cohort, success rates were considerably lower.
Was the absence of preliminary data a significant factor?
Though the numbers are small, a greater proportion of applications from new PIs lacked preliminary data, compared to the cohort as a whole. Out of the 13 applications that had no preliminary data, 7 (53.8 percent) were from new investigators, none of which were funded.
Our comparison of success rates is particularly salient because it counters a belief among some new investigators that seeking an R21 as a first grant is an easier route than applying for an R01.
In fact, R01s are a better choice for a first independent grant. You benefit from a special payline for new investigators, and the R01 is a more robust award.
For more advice, go to What Award Should You Apply For? in our New Investigator Guide to NIH Funding. For a comparison of several key characteristics of R21s compared to R01s, see R21 or R01? below.
In our cohort of 200 applications, 47 (23.5 percent) were from multiple PIs, and they did roughly as well as single PI applications.
Next we looked at the effect of new versus experienced investigators on a multiple PI application's success. Though the numbers are very small, the presence of one or more experienced investigators had a positive impact.
For this group, the effect of preliminary data is hard to determine because only four multiple PI applications did not have preliminary data, none of which were funded.
This table compares some features of the parent R21 and R01 announcements. Note that institutes may publish their own funding opportunities using the R21 or R01 with different characteristics.
To introduce novel scientific ideas, model systems, tools, agents, targets, and technologies that have the potential to substantially advance biomedical research.
To support a discrete, specified, circumscribed project...in an area representing the investigator's specific interest and competencies, based on NIH's mission.
Up to two years.
Up to five years.
Up to $275,000 in direct costs over two years. At most $200,000 for any year.
Unlimited. (But note that most new investigators submit a modular grant of $250,000 or less—see graph at How to Plan an Appropriate Budget in the New Investigator Guide to NIH Funding.)
Yes, as a renewal application.
Not required, but read Know the Importance of Preliminary Data on this page.
See the list of R21 Participating Institutes and Centers in the NIH Exploratory/Developmental Research Grant Program, Parent R21.
Most NIH institutes and centers. Find details in the R01 Parent Program Announcement.
New investigator benefits
Higher payline than non-new PIs, lower reviewer expectations, and other benefits. Read more at Are You "New"? in the New Investigator Guide to NIH Funding.
NIAID success rate for FY 2015*
* Data come from RePORT's Research Project Success Rates for NIAID.
For more information, see the following:
Return to Choose the Grant in Part 2. Pick and Design a Project.
Table of Contents for the Strategy
We welcome your comments, questions, or suggestions. Email firstname.lastname@example.org.
Last Updated May 06, 2016
Last Reviewed June 21, 2012