NIH addresses challenges in the submission and review of grants
Gary Boas, News Editor, firstname.lastname@example.org
It was a hazy early April morning, my coffee was still brewing, and I was talking on the phone with Eugene G. Arthurs, CEO of SPIE, about the boosts provided by the Challenge Grants and other National Institutes of Health (NIH) mechanisms supported by the recently passed stimulus bill.
Arthurs described the grants as a sort of lifeline for a generation of investigators dispirited by the declining fortunes of medical research in the US. As an example, he noted the recent “appalling” trends in the rates of successful grant applications. “If one were to sit down and calculate how much effort went into the forty or fifty thousand applications, every year, that do not get funded …,” he said, trailing off before finishing the thought.
The National Institutes of Health has released a report with recommendations to improve the grants submission and review process, which has become unwieldy for applicants and reviewers alike. Shown here are the NIH campus (top) and paper grant applications in the NIH Center for Scientific Review mailroom (bottom). The latter photo was taken in June 2006. Most applications now are submitted electronically. (Printed with permission of the National Institutes of Health/Department of Health and Human Services.)
A couple of hours later, he e-mailed me with a few additional comments. Although lauding the efforts of those at the NIH and the National Science Foundation to prioritize the available funds for best use in the scientific community, he included some back-of-the-envelope calculations underscoring the need to revisit the application and review processes: If investigators spend an average of two months preparing a grant, then they are devoting 100,000 months of effort, annually, to those 50,000 unfunded applications. This translates into 8300 years of researcher time, every 12 months.
“That’s a poor use of a high-quality resource,” he said.
I had heard a great deal, anecdotally, about the inordinate amount of time devoted to preparing and reviewing grants. And having spent the better part of a decade hanging around academic research centers, I had seen investigators struggle mightily with the monumental task of assembling applications: writing and rewriting aims, rubbing away all the rough edges, coordinating the flow of preliminary data from the many arms of the multidisciplinary research team and, somehow, making it all come together in the final few hours. And then, after jetting off to D.C. or perhaps to somewhere in California to attend a review session, doing it all over again for the next grant.
Yet I had never heard it expressed in such plain, sobering terms.
I thought there might be a story in this. I knew that the NIH was implementing a new, 12-page application – intended to reduce the amount of time spent preparing the grants, among other things. And after digging a bit (i.e., mentioning it to someone at a party), I learned that the institutes were rolling out a far more comprehensive set of changes: changes designed to streamline the application and review processes, to make them more focused, more manageable.
I decided to look a little deeper.
NIH reviews the review (and more)
In 2007, Dr. Elias Zerhouni, then director of the NIH, ordered a study of the ways in which the application and review processes – mostly review – could be improved. “There were concerns that a system that worked well when thirty to fifty percent of applications were being funded might not be optimal for a time when only five to twenty percent of applications were being funded,” said Alan L. Willard, chief of the Scientific Review Branch, NIH-National Institute of Neurological Disorders and Stroke. Other factors in this decision included the enormous increase in the absolute number of applications that had to be reviewed, the increased “hassle factor” involved in air travel, and the increased demands on the time of people who serve as reviewers.
“Until the early 1990s,” Willard explained, “a typical study section met for two to three days, discussed all applications in detail and prepared extensive critiques that often had a mentoring as well as an evaluative component. Currently, very few reviewers are able to spend that many days at a meeting or to prepare such detailed comments – they simply have too many other demands on their time.”
In March 2008, after a yearlong study, the NIH released a report detailing its findings. The report identified significant challenges inherent to the peer review system and made a number of recommendations as to how to address them.
Many of the changes have been implemented already. The criteria used in reviewing applications now emphasize the overall impact of a project rather than the technical details, as was the case previously. Also, critiques are shorter and more focused, noting the strengths and weaknesses of an application in bullet form. Finally, the NIH is making greater use of the Internet as well as teleconference and videoconference technology to minimize the amount of time spent traveling to review sessions in Washington and elsewhere.
The NIH is planning additional changes in the coming year, including introducing the significantly shorter applications already seen with the Challenge Grants. The idea, Willard said, is to place less emphasis on preliminary data and more emphasis on identifying high-impact projects from well-qualified investigators using a “logical but not excessively detailed technical approach.”
Some investigators might lament this latter change. The preliminary results section of a grant allows researchers to demonstrate not only that they have done the necessary legwork for the proposed study, but also that they can complete the tasks outlined in the application. In lieu of this section, reviewers likely will look to publications and to what they know about the applicants’ work otherwise (read: reputation) to decide whether what has been proposed can be achieved. Thus, the new shorter applications might privilege high-profile investigators with years of publications in important journals – essentially, the big shots in any given field.
To an extent, however, this has always been the case. Reviewers can be influenced, subconsciously or not, by the experience and perceived stature of the investigator. This is not entirely surprising – on some level, past achievement would seem to imply future success – but neither is it entirely fair. The NIH has acknowledged this potential bias and, in the recent report, has taken steps to address it (more on this below).
Possible biases aside, the shorter applications can help researchers focus on the aims and possible impact of their projects. With 25 – or even more – pages, as in current applications, it can be easy to get lost in the details – in the preliminary results or methods sections, for example. When limited to 12 pages with a format encouraging emphasis on the impact of the project, applicants must work out exactly what their goals are and present them in as clear and concise a manner as possible.
And lest we forget, the shorter applications likely will help reduce the amount of time spent preparing grants. In the past, much of the effort devoted to grant writing involved finalizing preliminary results and working out the precise methods to be used. Although these still will be important, applicants may now find the overall process less daunting and, ideally, less time-consuming.
Encouraging new and early-stage investigators
As I explored the issue of the amount of time spent writing and reviewing grants, I heard additional concerns about how the peer-review system has evolved over the years. One of the most significant is the fact that researchers often are well into their careers before they are awarded their first R01 grants: The average age is now 42. Naturally, this has become a source of tremendous frustration.
Recognizing this, the NIH report delved into the myriad reasons for the increased amount of time a researcher spends in the “training phase” of his or her career. A major factor, Willard said, is the volume of information that investigators must master, as research is increasingly multidisciplinary in nature, drawing equally on physics, engineering and biology, for example. Also, the rate of faculty retirements has been slower than anticipated, which translates into less turnover of principal investigators. Finally, as the competition for grants has grown ever more intense in recent years, many young investigators “feel discouraged from even trying to apply for grants,” Willard explained. Instead, they remain co-investigators on grants submitted by senior scientists.
Others I spoke with noted the perceived bias favoring the latter investigators. One researcher told me a story of the same multiproject grant submitted twice, with the same personnel responsible for the same projects – only in the second submission, a well-known name was attached as one of the principal investigators. This second time around (the application was not funded the first time), a reviewer expressed confidence that the grant would be successful with the senior scientist at the helm. The researcher with whom I spoke was frustrated by this, she said, because she would have spearheaded the project in either case.
The NIH has taken several steps to address the imbalance. First, the institutes will achieve comparable success rates for new and “early stage” investigators by selectively paying R01 applications. “At some NIH institute, this could mean that the pay line for new/early-stage investigator R01 applications will be more generous than for experienced investigators,” Willard said. Other efforts will include ensuring that the same numbers of R01 grant applications from new/early-stage and experienced investigators are discussed at study section meetings, and clustering the discussion of new/early-stage R01 applications.
Finding a balance
The peer-review system is not perfect, but it provides the best opportunity to have grants critiqued by reviewers who are capable of understanding the science, recognizing the potential impact of the project and assessing whether the applicants will be able to achieve what they have proposed.
Still, as with any endeavor, some kind of balance must be struck between “exhaustive” and “efficient.” With the recent recommendations, the NIH has taken significant steps toward achieving the right balance.
With respect to the amount of time spent preparing grants – the question that set me on this path to begin with – the shorter applications likely will contribute to fewer headaches and will allow researchers to devote more time to actual research. Other changes, such as the new limits as to the number of resubmissions permitted – one instead of two – could lead to fewer overall submissions and, thus, to less time spent writing.
But the larger problem might still be the intense competition for grants. As long as investigators are in need of funding, they will submit as many applications as they can. Only when the outlook begins to improve – and there are signs that it will – will they feel they can take a break.
Even if just a short one.