NIH, NSF Move Ahead Slowly On Electronic Submissions

New joint effort doesn't erase differences in how the two agencies view the long-term impact of getting proposals on the wire WASHINGTON -- The National Institutes of Health and the National Science Foundation have teamed up to study how the government might develop a system for scientists to submit their grant applications via modem. University grants administrators predict that such submissions are inevitable, and speculate about the possibility of a single application for all federal grants. But the track records of the two agencies suggest that an electronic future is a long way off. "My goal is to provide a tool that supplies the NIH with applicant data in a much more useful form than now exists," says John Mathis, assistant to the chief of the referral and review branch of NIH's Division of Research Grants, who has been given the job of computerizing the applications process. "We're looking ahead to the 21st century, but I want to stress that we are moving very slowly and carefully." His NSF counterpart, information specialist Gerald Stuck, is equally cautious in his forecast. "Our goal is to get proposals into the hands of decision-makers as quickly as possible," says Stuck. "But they still have to read the proposals, for the most part on paper, and make decisions based on the science. Our job is merely to provide the necessary support for that process." The two agencies so far have taken different approaches to the problem. For Mathis, a former executive secretary to the pathobio-chemistry study section, the long-term goal is to construct an enormous electronic database drawn from the 40,000 proposals sent annually over high-speed communications lines. Such a database, checked closely for accuracy, would nearly eliminate the need for paper throughout the review process, he says. It would also offer the opportunity for an unprecedented level of analysis of the state of biomedical research in the United States. Stuck's vision is a bit more modest. He'd like to get NSF applicants to submit their proposals electronically so that the foundation can ease the physical demands on its mail room and nudge institutions toward a single standard for transmitting their material. The possibility of manipulating the information contained in the 30,000 applications that pour in each year is only a dim shadow on his horizon. So far, the institutions that are participating in the experiment appear to be doing it more for the experience than for any immediate gain. "I think that it's the direction that grantsmanship is going, no question about it," says Celia Walker of the Office of Research Initiation at Colorado State University. "Right now there's a tremendous amount of paper and time expended on the production of a proposal, resources that would be much better spent on thinking up new ideas. "But at this point I'd have a hard time justifying why a P.I. should use it," she adds. "Until a proposal can be produced within a system that links up everybody electronically -- and that includes the department head, the dean, the vice president, the animal care committee, and the sponsored research office -- it's not going to save anybody any time." This month's opening meeting of an interagency task force -- made up of representatives from NIH, NSF, and two research universities -- reflects a blending of what began as separate efforts. Although both agencies have traveled part of the way down the road to electronic submissions, so far it has been a bumpy ride. Cedric Minkin doesn't mind being a guinea pig in NSF's experiment to submit grant applications electronically. In fact, the cell biologist, a professor in the school of dentistry at the University of Southern California, considers his participation to be part of his civic duty as a scientist. "I think it's the right thing to do," he says about the grant proposal that he submitted last December. (His proposal deals with the factors that regulate the development of osteoclasts, cells involved in the removal of bone material.) "I think it is clearly something that needs to happen, and anything that I can do to help the community is worth the time it takes." The 48-year-old Minkin, who has been supported by a variety of federal grants since 1967, also admits to a sense of curiosity. "I did it because I'm adventurous, and I wanted to see how it worked." What he found was that the process took a few days longer, and required much closer cooperation with the university's grants officer, than does the typical paper application. He also discovered that an application could get lost -- at least for a few days -- if NSF officials failed to log it in properly. But none of that seems to faze him. "I remember when I was on study section for NIH in the early 1980s, that I was the only member writing reviews on a word processor," he says. "But don't get me wrong -- I'm not a computer freak who has to have the latest technology. It's just that I appreciate the advantages of a PC, and I enjoy producing a document electronically." His enthusiasm may also be linked to his belief that a scientist must keep up with the times, or run the risk of being left behind. "I remember the comments I received on my last proposal to NIH," he recalls. "They told me that it was good work, but that it didn't make use of the latest advances in genetic engineering. I had a choice -- either I could become filled with righteous indignation and rail against the tide, or I could bite the bullet and learn something new. So I decided that change wasn't such a bad thing." --J.M. Two years ago NIH asked scientists at some 150 universities to test a piece of software written to handle incoming applications (The Scientist, May 30, 1988, page 21). "They said it was a terrible program," recalls Mathis. "But they thought it was a great idea, and they gave us lots of suggestions on how to improve it." Similarly, NSF last year pulled back from its multimillion-dollar attempt to create a nationwide standard for electronic submissions, admitting that the idea was overly ambitious (The Scientist, March 6, 1989, page 8). However, as a result of that project, called EXPRES, NSF is currently far ahead of NIH in the number of applications that it is handling electronically. Since January 1989 the foundation has received some 115 proposals from 50 institutions, most of them from individual investigators. In addition, last summer the foundation encouraged electronic submissions for the second round of its competition for university-based science and technology centers. But only 14 institutions -- some 10 percent of the total pool -- took NSF up on its offer. (Because of reduced funding from Congress in the current fiscal year, NSF has delayed making any awards until this fall, after it knows how much it will be getting in fiscal year 1991.) One institution that accepted the challenge was Bowling Green (Ohio) State University's Center for Photochemical Sciences. The procedure went smoothly, says administrative director Pat Green, and officials were able to avoid the tedious process of inserting tabs and making 25 copies of a complex, 40-page proposal. But she suspects that electronic submissions aren't the answer for everybody. "You need to be willing to go through all the steps," she says. "The biggest problem was figuring out the instructions - - it's like reading a computer manual. It's definitely not for people who are intimidated by computers, and who don't already have experience with the technology. "On the other hand, I think that it's the only way to go in the long run. We have PCs and modems in our office, and there was never any hesitation on our part that it was the right thing to do." The University of Southern California also bought into the experiment (see Close-Up, below). "I went to a workshop that Gerry held in February 1989," explains Barbara Davey of the university's grants office, "and a few months later we announced to the faculty that we were ready to give it a try. We didn't get much of a response at first," she says, "but we hope to do more as time goes on. For Davey, electronic submissions give her a chance to serve her constituents, university scientists. It's also an opportunity to stay on the cutting edge. "It's a good way for grants officers to be helpful to the P.I.," she says. "It doesn't really save us any time because the individual departments handle the mailing, but we are bolstering our image by gaining this kind of experience. And we like to be leaders in a field." Davey also offers an explanation for why NSF and NIH have taken different approaches to solving the same problem. The reasons, she says, are rooted in their present methods of handling grants, as well as in their organizational cultures. "NIH has a bigger problem with tracking grant applications and with ensuring accuracy," she says. "They are also more bureaucratic -- there are so many more levels for them to go through to get approval for something like this. On the other hand, NSF already does a reasonably good job tracking the proposals it receives, and its application is less cumbersome and requires less information." That difference, she suggests, even extends to the personnel assigned to the task. "John is doing this on temporary assignment, as a scientist, while Gerry is full-time, and a computer person to begin with. Now it's nice for NIH to have somebody like John, who's so interested that he's willing to take on the challenge. But it probably means that things are going to move a lot slower." Mathis pleads guilty to such charges. "The key question for NIH is, How can institutions communicate with us in a way that we can understand?" he says. "That includes being able to capture [the information], store it, and retrieve it for use by the program staff and the reviewers. My goal is to make us function better, to cut down on processing time, cost, and omissions in applications, and to include 100 percent of the applicant's information in a database." But achieving that goal will take time. It will be sometime in 1992, he predicts, before NIH is ready to advance beyond the test stage and receive some "real" applications. Until then, he counsels patience. "This morning I got a call from a scientist who wanted the software to meet an October 1 filing deadline for an RO1 [an investigator-initiated grant]," Mathis says. "I had to tell her that it wouldn't do her much good, because we're not yet ready to handle what she wants to send. My biggest nightmare is that we'll get all this information and not be able to use it."

NIH, NSF Move Ahead Slowly On Electronic Submissions

Interested in reading more?

Become a Member of