Steering Science from a High Altitude

It's a given that peer reviewers influence research-funding decisions. But behind the scenes, another group of scientists holds considerable sway over what research gets funded and what does not: scientific program officers. Sometimes called program managers or scientific officers, "they're catalysts who play a key role in the scientific leadership in their fields," says Norka Ruiz Bravo, deputy director for extramural research at the U.S. National Institutes of Health in Bethesda, Maryland.

Funding agencies and organizations have a cadre of program officers whose most pressing responsibility is to oversee peer review for a portfolio of grants. For example, program officers at the National Science Foundation (NSF) handle about 100 proposals a year. The work involves planning program initiatives, issuing calls for proposals, advising researchers, reading proposals, picking reviewers, running review panels, organizing reviews by mail, synthesizing reviewers' evaluations, and making funding recommendations to a division director who usually has final say over funding decisions.

In addition to portfolio management, program officers keep an eye out for new opportunities and facilitate connections among researchers. "The best program officers survey their disciplines and ask, 'What does the field need from me?' " says Irene Eckstrand, a program director at the National Institute of General Medical Sciences in Bethesda. "It may be that the field just needs money, or it may need a symposium to move it along, or it may need deeper collaborations within the field. To make that happen, you bring the scientists together and give them opportunities to talk."

Another way program officers exert influence is by guiding discussions during panel reviews, says Katharine Covert, a program director in NSF's Division of Chemistry. She says she asks a lot of questions during review meetings to stimulate productive conversation. But, she says, "I try very hard to not put my own opinion of a proposal out there. It's very easy to bias a group."

Program officers must also make tough judgment calls about proposals that receive mixed reviews, or when--as is often the case--there are far more high-scoring proposals than there is money to fund them. Program officers can also override reviewers, recommending that lower ranked proposals get funded, for example, to support new investigators or encourage high-risk but potentially transformative research. "It's really tricky business, because the community could rise up against you," observes a former NSF director, Neal Lane, who is now a physicist at Rice University in Houston, Texas.

Learning the Job

Because program officers have so much latitude in decision-making, Lane says, "the job really can't be done by someone who doesn't understand the science or the engineering that he or she is going to be overseeing." Program officer positions in the federal government usually require a Ph.D. and at least 6 to 10 years of experience in independent research. Formal training for new program officers covers a wide range of rules and procedures designed to ensure fairness and consistency, including instruction on federal ethics laws, meeting facilitation, and specialized software used for handling grants.

For many program officers, being a steward of taxpayer money is something new. "It takes a while to get used to people coming to you as an authority and as the person who holds the purse strings," says Paul Hertz, chief scientist in the Science Mission Directorate at NASA. Making a smooth transition, he says, demands not only solid scientific credentials but also strong administrative skills. "You need to work well with the people in your field, because you are now the face of the agency," Hertz says. "You need to be a good communicator. You need to be decisive and willing to make hard decisions. If you're the kind of person who wants to make everybody happy, you're probably not going to enjoy your job, since only about one out of three or four proposals gets accepted."

Working in a large government bureaucracy also means giving up some freedom. "You have to be very careful about what you say because you speak for the government," says Eckstrand. "There's nothing that you say, even cocktail chatter, that isn't on the record. And there are times where as an agent of the government, you have to implement policy that you don't agree with. I know people who have left government because they couldn't administer policy in a way that was consistent with their principles."

Whether in federal government or at a private foundation, being a good program officer also takes organization and versatility. Brian Quinn, a program officer in the Robert Wood Johnson Foundation's Research and Evaluation group, says that in a given month he puts his hands on 30 or 40 different grants. "When the phone rings, you've got to be able to shift gears quickly," he says.

For program officers in federal agencies, planning around an uncertain federal budget is also a perennial challenge. "I have to get solicitations out for next year, but we don't know what the budget for next year is going to be," says Covert.

A False Economy

Some science-policy experts argue that program officers have been hampered by ever-increasing workloads and recent administrative budget cuts. In a report released this June, a blue-ribbon panel appointed by the American Academy of Arts and Sciences wrote, "It is false economy to deny program officers who manage millions of taxpayer dollars the resources necessary to engage fully with the professional communities they fund and for whom they are responsible."

"Increasingly, program officers feel there are constraints on what they can do," says Lane, who served on the panel. "The program officers' commitment hasn't changed, but doing the job well has been made more difficult. ... My worry is that unless we reverse the trend, it will be harder to find program officers in the future."

He adds that dwindling coffers have restricted program officers' ability to attend scientific conferences and visit universities and even to devote time to reading literature that allows them to keep up-to-date in their fields. All of this has undermined some program officers' confidence, he says, making them more reluctant to take flyers on high-risk, high-reward proposals.

As evidence of the effects of how program officers are stretched too thin, Howard Hughes Medical Institute President Thomas Cech, who chaired the American Academy panel, notes that discretionary funds set aside for program officers to allocate to promising projects at critical junctures went mostly unspent at NSF last year. "Those are the moments when the right program officer should be able to find that money within days and say, 'You know, I'm going to stick my neck out,' " he says. "That hasn't been happening."

An Aerial View of Science

For all its challenges, science administration also presents unique rewards. "If you come to scientific administration as a fallback because you couldn't get funding for your lab or because nothing else worked out, you're not going to like it," Eckstrand says. "But if you really like developing science from an aerial view, and if you have a certain tolerance for bureaucracy, then science administration can be very fun and rewarding."

"It's an opportunity to work with dedicated, smart, perceptive people and to be a part of science in a very different way," agrees Covert. Most gratifying, she says, is when "things happen that wouldn't have otherwise happened if I hadn't been involved. I might casually mention an opportunity to an investigator, who then goes out and explores it and ends up getting funded." At the other extreme, she says, "There are difficult situations--say, where an investigator falls ill or has other pressing needs--and the person on the other end of the phone line matters. A lot of our job is very bureaucratic and prescribed, because we have to maintain fairness, but there are other times when we just have to be human."

Siri Carpenter is a freelance science writer in Madison, Wisconsin.

10.1126/science.caredit.a0800138