The Inequality of Science

FEATUREThe Inequality of Science   © JOELLE BOLTIn 2004, close to one in five extramural NIH dollars went to only 10 of the 3,000 institutions that received grants. Five US states get almost half of all funding. What about everyone else? BY ALISON MCCOOKThe ceiling in the medical school at the University of South Dakota (USD) in Vermillion is visibly water-stained and falling down in spots. Walking through the fac

Aug 1, 2006
Alison McCook
FEATURE
The Inequality of Science
 
The Inequality of Science
© JOELLE BOLT

In 2004, close to one in five extramural NIH dollars went to only 10 of the 3,000 institutions that received grants. Five US states get almost half of all funding. What about everyone else?
BY ALISON MCCOOK

The ceiling in the medical school at the University of South Dakota (USD) in Vermillion is visibly water-stained and falling down in spots. Walking through the facility, researcher Robin Miskimins admits that she can't plug in two hoods and three incubators in the tissue culture room without blowing a circuit. Everyone shares key equipment, such as a confocal microscope, a cell culture facility, and an aging scanning electron microscope.

Four hundred miles due north through largely farm country, Diane Darland waited for months after moving from Boston to the University of North Dakota (UND) in Grand Forks to get equipment - a small set of cages for transgenic mice, and a $10,000 hood (which the Dean of Arts and Science eventually bought for her). All these resources, now tucked into a sparse, roughly 18 square meter room in the basement of the biology building, are "standard at most universities," she says with a shrug. Her husband, also a biologist, accepted a position at the university around the same time, and is now "scrubbing tanks" to create the school's only zebrafish facility. The school has no operational transgenic animal core, nor a BSL-3 lab, so Matthew Nilles, a plague researcher, can work with only an attenuated strain, limiting the applicability of his results.

In 2004, USD and UND received $10 million and $8 million from the National Institutes of Health, respectively, amounting to barely one percent of that given to the top-funded school that year, Johns Hopkins University. UND and USD don't get less money just because they're smaller than Johns Hopkins: Per capita, the research staff and full-time faculty at USD received about $24,000, and UND got almost $13,000. Johns Hopkins' per capita figure: $137,000.1 "It was pretty obvious when I got [to South Dakota] that I was going to have to limit my ambitions," says bacterial genetics professor Keith Weaver.

North and South Dakota aren't the only states feeling the inequity. A group of 23 traditionally disadvantaged states, along with Puerto Rico, collectively receive less than seven percent of the total NIH budget each year. In 2005, the principal investigator on the biggest NIH grant, Eric Lander at Massachusetts Institute of Technology, received more than $50 million, nearly seven times the total amount given to all institutions in the state of Wyoming the year before.

THE STATE OF THINGS

Not surprisingly, universities at the top would apparently like to keep things that way. Within a few months after accepting the role as director of the National Science Foundation, Neal Lane was approached by a group of approximately five provosts from the nation's top universities. They informed him that their schools should receive the lion's share of federal funding for research. "The message definitely was: 'We're where it's at,'" Lane recalls.

NIH Awards to Institutions by Rank,
Fiscal Year 2004
Rank
Organization Location
Amount
1
Johns Hopkins University Baltimore, Maryland
$599,151,309
2
University of Washington Seattle, Washington
$473,432,138
3
University of Pennsylvania Philadelphia, Pennsylvania
$464,076,925
4
University of California, San Francisco San Francisco, California
$438,778,831
5
Science Applications International Corp* San Diego, California
$403,213,237
6
Washington University St. Louis, Missouri
$388,307,875
7
University of Michigan Ann Arbor, Michigan
$368,176,446
8
University of California, Los Angeles Los Angeles, California
$361,593,433
9
University of Pittsburgh Pittsburgh, Pennsylvania
$360,635,035
10
Duke University Durham, North Carolina
$343,825,304
*SAIC is the US's largest employee-owned research
and engineering firm.
Note: Some independent hospitals are affiliated with schools, but received funding that is not included in the institution total. This is the case with Harvard, which is affiliated with the five top-funded independent domestic hospitals in 2004, which brought in a total of $852 miillion extramural awards.
NIH Awards to State of
Recipient Institution, Fiscal Year 2004
Rank
State
Amount
1
California
$3,619,589,540
2
Massachusetts
$2,265,512,043
3
New York
$1,964,889,285
4
Maryland
$1,415,908,552
5
Pennsylvania
$1,394,474,505
6
Texas
$1,147,992,873
7
North Carolina
$985,446,510
8
Washington
$815,255,551
9
Ohio
$691,538,371
10
Illinois
$689,658,870

Inequality in funding has always been a problem, says Lane, who won't disclose the identity of the schools who approached him. But there are signs that the gap between the "haves" and "have-nots" may be widening even further. Between 1994 and 2004, in the rankings of universities and colleges according to total R&D expenditures in biological sciences, the differences between the number one school and the 100th school more than doubled.

Lane, who directed the NSF from 1993 to 1998, agrees that the problem has "gotten even worse" because the equipment needed to do elegant experiments has become more expensive, while NIH funding has, in recent years, remained relatively flat. And without easy access to cutting edge equipment, researchers "don't really have a chance of competing at the NIH or NSF," Lane says; grant funding reviewers are "very sensitive" to the equipment applicants have at their fingertips.

There's "no question that states like ours are disadvantaged," says UND's Jonathan Geiger, chair of the department of pharmacology, physiology and therapeutics. Researchers at every institution can typically find a way to share everything they need, and collaborate with off-site scientists who have something they don't, but all of that takes extra time and effort. "Since time equals money in the current competitive environment, having to go elsewhere to complete experiments can be challenging and costly in the long run," says UND's Darland.

And you can share equipment, but you can't share people, so scientists at both Dakota schools say finding good postdocs, graduate students, and technicians is a constant challenge. "A talented undergrad doesn't want to come to Vermillion for grad school," says Weaver. Despite the fact that both Dakota schools come with short commutes, manicured lawns, and stately brick buildings, faculty at both say they've offered jobs to promising candidates who turned them down because their families didn't want to live in North or South Dakota. Postdocs leave after a few months. Guest speakers decline to make the trip, or simply ignore invitations.

Researchers at both schools also say they encounter bias from their colleagues, some of whom expect little to come out of either state. When Michael Chaussee, a researcher in bacterial pathogenesis, told his colleagues he was leaving NIH for USD, their reaction was largely "surprised and negative," he recalls. He finds he often has to promote and defend the school at meetings, and people assume he's looking for another position elsewhere. (He's not.)

At both UND and USD, scientists will readily provide a list of equipment they'd like to have: UND, for instance, has no fluorescence capacity in its laser-capture microscopy, no multiple-photon excitation fluorescence microscopy system, and no small-animal behavioral testing facility. At USD, researchers would love to add a positron emission tomography scanner, and functional magnetic resonance imaging.

THE VIEW FROM THE TOP

At the University of Pennsylvania, the stately buildings and manicured lawns are crisscrossed by the busy traffic of Philadelphia. In a sleek office tucked away from the noisy streets, the school's vice dean for research and research training, Glen Gaulton, is bursting with enthusiasm, gesturing wildly with his hands and arms to explain how Penn climbed its way to the "tippy-top" of the NIH funding charts. The school took the number three spot on the NIH funding charts in 2004, bringing in nearly half a billion dollars. In 2003, it ranked number one for Health and Human Services R&D expenditures (which includes NIH dollars).

"My view is that mass spectrometers are like handbags," says Penn's Ian Blair. "You can't have too many."

But when Gaulton moved to Philadelphia from Harvard in 1986, Penn wasn't even in the top 10, he recalls. The school was strong in certain areas, such as retroviruses, but was considered a "sleepy Ivy League medical school," lacking the energy characteristic of truly elite institutions. Then in the late 1980s, the school of medicine got a new dean, William Kelley, who decided to transform Penn from being "very good" to "great," says Gaulton, excelling in all areas, not just a few. To do that, he formed a team to support this vision, worked to increase alumni and philanthropic donations, borrowed millions, and boosted income from the Penn health systems. The school used this extra cash to build or renovate approximately 65,000 square meters of research space (bringing it to today's total of approximately 100,000 square meters), at a cost of likely up to $500 million.

Penn also spends $10 million a year recruiting top faculty, and works to keep them, he says. Two to three times every month, a competitor tries to recruit one of the 1,700 faculty members at the school of medicine, and Penn will sometimes offer the scientist more money or resources. All this, on top of the approximately $10 million needed each year for equipment and running core facilities. The school has spent "an enormous amount of money," Gaulton says.

It shows. Ian Blair, a professor and scientific director of the school's proteomics core facility, has a lab filled with multiple versions of state-of-the-art tools constantly humming away, while newly displaced instruments that many labs would kill to own are repurposed for lesser projects or simply tucked away in a dark corner. His newest addition: a $1 million Thermo Electron LTQ-FT, a Fourier transform ion cyclotron resonance mass spectrometer, with a maximum resolution of 500,000 which he purchased using a high-end instrumentation grant from NIH. Just to install the instrument required $100,000 of renovations, which included raising the ceiling and installing liquid nitrogen to remove the massive amounts of nitrogen gas that would be released if the magnet quenches. His lab and the proteomics core alone contain 12 mass spectrometers. (In contrast, the University of North Dakota is about to add its third overall.)

COURTESY OF THE UNIVERSITY OF PENNSYLVANIA
University of Pennsylvania's John Morgan building, home of science offices and labs.

Blair's eyes twinkle when he speaks about the equipment he's accumulated. "My view is that mass spectrometers are like handbags," he says. "You can't have too many." Still, he says even he is dwarfed by the private sector. "It makes you feel ill when you go into a pharmaceutical company," he complains, where there is roughly one mass spec for every two people working on related projects, such as drug metabolism.

Blair says he's become a leader in the use of mass spectrometers by pushing the technology forward, refusing to be just a user. Currently, he's trying to investigate lipid peroxidation damage in DNA from particular cell types, where the amount of DNA damage is relatively small: on the order of three modified bases surrounded by 10 million normal ones. "No one has ever been able to do that [using mass spectrometry]; my challenge is trying to think of ways to do that."

But having the right equipment is only one ingredient, he insists; the other is hiring the right people who know how to use the equipment to its fullest capability, and who can devise interesting biological questions that drive the field forward. Of course, having the right technology attracts top people, and Blair's staff is constantly winning awards, publishing in top-tier journals, and submitting winning grants. Blair says his success rate for NIH grants, accounting for resubmissions, is roughly 50%.

Now, Penn is firmly in the "tippy-top" of the grant recipients, all of which excel in all areas of medicine, not just a few, says Gaulton. Things are easier now, he admits. "Once you're there, it's not that difficult to stay there," since the school has built up the right infrastructure and administrative support for being one of the elite. "But to get there requires an enormous amount of effort." (See How to get to the top)

MAKING THINGS WORK

The top schools are not the only places that do good work, and there are many signs that the Dakotas are churning out strong science. Both schools received multimillion dollar grants in 2005: At USD, Anthony Gerdes was awarded $1.8 million from the NIH to study mechanisms of cardiovascular remodeling, and at UND, Geiger received close to $2 million to study the pathophysiology of neurodegenerative diseases.

There's no way the University of North Dakota can fairly compete with major players. "They've got tanks for weaponry, and we've got BB guns."
-Van Doze

The schools have also used extra assistance provided by the government to help disadvantaged regions kick-start their research programs. The Experimental Program to Stimulate Competitive Research (EPSCoR), a program that NSF offers, helps underprivileged states build up their research infrastructures to better compete with California, Massachusetts, and other typically well-funded states.

The NIH's Institutional Development Award (IDeA) program has also given almost $1 billion to disadvantaged regions. Awards for IDeA states increased three-fold between fiscal years 1999 and 2005, a result of the infrastructures the program helped build. Grant funds from IDeA enabled USD to purchase a mass spectrometer, which allowed the school to be "more on the cutting edge," says Barbara Goodman, a professor and the principal investigator on the grant.

Indeed, research funding at USD has doubled in the last five years. Next to USD's run-down medical school is an enormous construction site, where workers are buzzing away, building a $37 million facility equipped with more electrical capacity, larger rooms with more useable space, and new furniture for offices. There will be a BSL-3 facility, and the basement will contain animal rooms constantly monitored for pressure, temperature, and humidity. As a sign of enthusiasm, people working at the medical school have written dates on the walls of the old building predicting when they can move in. (see "Intelligent Redesign" on p. 40.)

As for UND, the school has recently added labs and an atrium onto the older school of medicine, and has taken advantage of the extra help that NIH provides disadvantaged states to build a shiny new neuroscience building. The school is now home to the state's first cyclotron, constructed using a $3.9 million grant. UND has also doubled its research expenditure in the last five years, and has $300 million worth of pending proposals, according to Peter Alfonso, the school's first vice president for research.

UND has also raised half the money it needs to construct the first building in a research and technology park, which would include a BSL-3 lab facility. Alfonso says UND wants to create a "life sciences corridor" that would include Minneapolis-St. Paul and Rochester in Minnesota, and Winnipeg and Saskatoon in Canada.

© ALISON MCCOOK


University of South Dakota's Robin Miskimins (top) at the site of the new medical school, still under construction. University of North Dakota's Van Doze (bottom), standing in front of a patch-clamp set-up used to record electrical activity from visually-identified neurons in live brain tissue.

One of the best resources the Dakota schools have to offer is their highly trained staff. All but a handful lack the singsong accent typical of the American Midwest; the vast majority studied outside the region, often in top-level institutions, and chose to come to North and South Dakota to have more comfortable personal and professional lives. The schools offered them tenure-track faculty positions (not requiring researchers to rely on grant money for salary), and often two positions for a working couple. Researchers enjoy a relaxed research environment without administrators' constant monitoring, and the advantage of living in a safe neighborhood for raising their families. People are valued at UND and USD, the researchers say (it's so difficult to replace them).

Visitors are impressed by what they see, says UND's Geiger, and scientists have been sent to UND for training in equipment not often found elsewhere, such as an instrument that uses microwaves to sacrifice animals, which Geiger says is more humane than traditional methods. And many scientists say they haven't had any trouble getting the grants they need to keep their research going.

Their secret? Finding a niche that makes competition from major players less likely. When other researchers looked at general cell activity to investigate a protein involved in cell-cycle control in glial cells, USD's Miskimins focused her research on one gene that turns on when the cell is turned on. Researchers looking at the big picture didn't see any effect of the protein, but she found that when the protein is active, the gene she focused on was also active. She eventually published four papers out of the project. "In order to stay funded, you just have to be different," says Miskimins. "Small shops like USD and UND can't go head to head with bigger schools. It would just anger them, she laughs, like forcing them to "swat an ant."

It makes sense to also specialize as an institution, the scientists say. At UND, there is a clear research focus on neuroscience, while a disproportionate number of USD researchers work on Gram-positive pathogens. Focusing on one area makes the schools more attractive to scientists working in that area; it's what enticed Indra Biswas, who works with Gram-positive bacteria, to come to USD from Atlanta's Emory University. However, focusing on one area can cause problems when research leads to unexpected directions. When Weaver's Gram-positive project yielded clear implications for Escherichia coli, which is Gram-negative, he couldn't just "walk down the hall" to discuss the findings.

Biswas also chose USD partly because he attended a seminar by Weaver at Emory, and says it was one of the best he's seen. Indeed, Weaver says he believes attending meetings, giving interesting talks, writing book chapters, and providing insightful comments during study sections has shown many scientists that good science can come out of less well-known institutions. "You're from Yale, and everyone always automatically listens to what you say," he says. "Here, you have to work."

USD and UND scientists say there's some benefit to being at these less well-known schools. Kevin Young, a UND professor, says he wears a North Dakota t-shirt when attending meetings, since people always remember the guy from North Dakota. "It's like going to see a panda."

But all too often, the limitations of working at a disadvantaged school can be frustrating. The schools may have beefed up their facilities and narrowed the gap separating them from the top, but it's easy to look like you're advancing in leaps and bounds when you had very little to start with, reminds UND's Van Doze. The divide is still very wide, he says. A few years ago, the school had "little chance of being competitive." Now, there's "some chance."

Doze says many of his former classmates are at Johns Hopkins University, and when he visits them, he's amazed by what they have. He has yet to get a grant on the first try since leaving Stanford University, where "it was not all that uncommon to receive a grant on your first attempt." He has not received an award letter on a grant recommended for funding eight months ago, making him doubt whether it will ever be funded. After five years, Doze says he has accumulated $500,000 worth of equipment, the bare "minimum" to do his work on neuromodulation in mammals. When he was at Stanford, he was one of approximately 100 electrophysiologists; at UND, there are two. He hasn't been able to recruit a PhD-level electrophysiologist, so his lab uses mostly undergraduate and graduate researchers. There's no way the school can fairly compete with major players, he says. "They've got tanks for weaponry, and we've got BB guns."

SHOOTING FOR PHYSICS?

The cost of doing sophisticated science shows no signs of slowing, suggesting research may become even more consolidated into the hands of a few key gunners. There is a "trend" in this direction, says Samuel Miller of the University of Washington, Seattle, one of last year's top grantees (see Top 20 NIH grants of 2005). For instance, in the near future, Miller suggests that answering a genomics question will likely involve more than just genotyping assays; it could also require phenotyping assays, along with human studies and complex data analyses. This monopolization of science funding is already happening, according to an editorial in April's Journal of Clinical Investigation, which asked the NIH to stop funding "large clinical studies that divert hundreds of millions of dollars away from hypothesis-driven scientific research."

Both Miller and Lawrence Corey, another top-20 grant recipient based at the University of Washington, are arguably benefiting from any consolidation of science into the hands of a select few. But both are concerned about the trend, and what it means for scientists working on smaller projects. "Some problems cannot be done by a group getting $200,000 a year," Corey says. "But there are some problems that can only be solved that way." Biology is "not like physics yet" - meaning, enormous projects in a few locations, forcing young scientists to simply join an existing lab, not start their own, says Miller. "As our tools become more sophisticated in biological science ... it might eventually reach that point," predicts Miller. "We're not there yet. But we're closer than we were 20 years ago."

1. Research staff and faculty numbers provided by the Carnegie Foundation for the Advancement of Teaching, as of May 2006.