For nearly two decades, the NFL ran a series of scientific experiments. The league formed its own research arm and published 16 papers about football and head injuries. The central conclusion -- that NFL players don't get brain damage -- led to public criticism, Congressional hearings and, in 2009, the abandonment of the project.
But the NFL hasn't abandoned the science of concussions. Over the past three and a half years, the league has transformed itself into one of the largest funders of brain research in the United States, allowing it to maintain a powerful role in science that could affect millions of people and, not incidentally, the bottom line of America's richest and most popular sport.
How the NFL wields that power is revealed in funding data compiled by Outside the Lines, interviews with scientific experts inside and outside the federal government and documents that have not been previously disclosed. With little fanfare, the NFL and its corporate partners have committed more than $100 million for scores of research studies -- a figure that exceeds the National Institutes of Health's budget for traumatic brain injury research. At a time when funding is scarce and the list of former NFL players diagnosed with brain disease continues to grow -- this week adding Raiders great Ken Stabler -- many researchers say the money is desperately needed.
"I'm extremely grateful to the NFL for work they've done; they have changed the public perception of brain injury," said Dr. Uzma Samadani, a neurosurgeon and researcher who has not received NFL money for her work on traumatic brain injury at Hennepin County Medical Center in Minneapolis.
But beneath the surface of the NFL's largesse is a secretive funding apparatus with its own set of rules, one that often rewards league doctors, punishes critics and, some researchers believe, steers research away from potentially uncomfortable truths about the relationship between football and brain disease.
In at least six instances over the past two years, NFL-affiliated grants totaling several million dollars have gone to scientists or institutions directly connected to the league, the data show. The NFL and its partners awarded nearly $4 million for projects tied to the co-chairman of its powerful Head, Neck and Spine Committee, Dr. Richard Ellenbogen, including a $2.5 million concussion clinic affiliated with Ellenbogen and another top NFL adviser.
Outside the Lines has also reviewed documents that show Ellenbogen and three other NFL health and safety advisers were applicants for a $16 million research project on football and brain disease that was to be funded by the NFL through a $30 million "unrestricted gift" to the NIH. The research proposal was led by Kevin Guskiewicz, a prominent concussion researcher who chairs the NFL's Subcommittee on Safety Equipment and Playing Rules. The proposal would have directed millions of dollars to research institutions that employ the league's advisers.
After an NIH review selected a competing proposal, Ellenbogen and two other senior NFL health and safety officers challenged the decision in a conference call, a senior agency official recently told Outside the Lines. Ellenbogen's participation on the call as an NFL adviser who was also an applicant for the NFL-funded grant "undermines the integrity of the entire peer-review process," said one researcher who works closely with the NIH and requested anonymity because of the sensitivity of the issue.
In a series of emails to Outside the Lines, Ellenbogen, a neurosurgeon, denied he complained about the selection process, saying he had been on a call with the NIH discussing "a wide range of research issues" but "dropped off" to go to the operating room. He initially denied he was part of Guskiewicz's grant application but later confirmed he was a "minor consultant."
At the time, the NFL argued that the researcher who was selected for the study, Robert Stern of Boston University, is biased and that the review was marred by a conflict of interest. According to a source familiar with the complaint, the NFL initially raised its concerns in an email from Dr. Elliot Pellman, a rheumatologist and NFL adviser who once led the NFL's discredited concussion research program.
After the NIH concluded that the NFL's complaints were unfounded, the NFL reversed a commitment to fund the project. The NIH used taxpayer money to keep the study alive.
NFL politics have clouded at least two other scientific initiatives over the past two years. The league funded a recent study that raised questions about the accuracy of helmet sensors -- devices that track the frequency, location and magnitude of hits to the head. The expert who edited the paper told Outside the Lines that the researchers hired by the NFL created a standard for accuracy that was unattainable. Last year, the league and the NFL Players Association cited the results of the research as justification to indefinitely suspend the implementation of sensors in the NFL -- even though the instruments have been used to study and prevent concussions for more than a decade.
Prior to the sensor study, a dispute between the NFL and the union significantly diminished a previously announced $100 million Harvard study on player health, with the league ultimately declining to participate, which was previously reported by Outside the Lines.
Outside the Lines requested an interview with Jeff Miller, the NFL's senior vice president for health and safety policy, but an NFL spokesperson said Miller was not available. In an email response to questions about the league's extensive research funding, the spokesperson wrote: "As it relates to head health, the league is funding scientific and medical research by leading independent experts around the world to advance understanding of brain injuries. This research is moving forward, and as it does, we will use it to improve how our athletes train and play the game."
Ellenbogen said it's wrong to suggest the NFL is trying to control research.
"What you are seeing is just how well-intentioned scientists debate the different approaches, how competitive the grants are, and fortunately how much interest there is in this research," he wrote. "This is not about the NFL stopping [brain research]. Quite the opposite. ... The NFL directly or indirectly funded much of it to the tune of millions."
Ellenbogen and Guskiewicz emphasized that they are not paid by the league as members of the NFL Head, Neck and Spine Committee, which they say operates independent of the league office. Outside the Lines has previously reported that the committee is funded by the NFL, reports to the commissioner and filters communication through the NFL's media office, which in the past has monitored interviews and correspondence with committee members. None of the committee members is paid by the league, but they are reimbursed for expenses through the league office.
"The interesting thing -- unless you expect the NFL to be altruistic -- is that from a business perspective, they're doing exactly what they should do," said Eric Nauman, a professor of mechanical and biomedical engineering at Purdue University, whose research has shown that football-related head trauma can lead to dramatic changes in brain chemistry in high school players.
Nauman and his Purdue colleague Tom Talavage previously conducted research through an NFL-General Electric research initiative, but Talavage said they have since decided not to take any additional NFL money because "we've seen too many examples" of influence and scientific bias.
"A while back, we talked about all the folks who have ended up on the Head, Neck and Spine Committee, and we think there's a pretty clear demarcation between the research they did before they were funded by the league or tied to the NFL, and the research they did after they were tied to the league," Talavage said. "And basically, there's a sense that we really, really want to make sure we aren't in that position."
The NFL is funding a broad spectrum of brain research using neuroimaging (at least $4 million), blood samples (at least $2.7 million) and a host of other tools. It is funding the development of new technologies -- a soft-shelled football helmet ($1.5 million), more forgiving synthetic turf ($500,000), and goggles that measure changes in eye movement associated with head trauma ($500,000).
In many ways, researchers say the league is filling gaps left by other funding sources. The NIH, the nation's largest biomedical research institution, awarded grants for just 18.1 percent of all research applications in 2014, according to agency data.
"I have to say that the NIH, until recently, wasn't advertising ... that they're interested in this topic," said Dr. Jennifer Coughlin, a researcher at Johns Hopkins Medical Institutions, who led a team that received $300,000 through a partnership between the NFL and General Electric.
"As a scientist, when you have an interest in a particular scientific topic, you go to foundations or groups who are interested in giving money to answer that scientific question," said Coughlin, whose group is using cutting-edge scanning to track how inflammation might cause brain disease in current and former NFL players.
Dr. Samadani, the Minneapolis neurosurgeon, agreed that the NFL's role has been critical.
"Up until the NFL came along and made this into a disease, people didn't think of raising money for it," she said. "Now all of a sudden, everybody wants to get into the brain -- to find out the consequences of brain injury and figure out how to prevent and treat it. And I'm glad about that. I care very deeply about people with brain injury. There's nothing better than seeing people recover: It's like they're born again."
The NFL's financial interest in the outcome of the science "does trouble me," Dr. Samadani said. "But I don't know the answer."
The potential pitfalls of industry-funded research have been known for decades. From the early 1950s, the tobacco industry spent hundreds of millions of dollars to buy off scientists, set up its own research committee and fund thousands of studies -- all with the goal of denying the connection between smoking and lung disease.
In 1994, the NFL formed its own research committee, led by Pellman, which repeatedly denied the link between football and brain disease and attacked independent scientists who asserted otherwise. After commissioner Roger Goodell disbanded the committee in 2009, the league brought in respected researchers like Ellenbogen, the chairman of the neurological surgery department at the University of Washington, and Dr. Mitch Berger, who holds the same position at the University of California San Francisco.
After taking the job, Ellenbogen was critical of the previous committee's work and at one point told Outside the Lines: "You can't have the NFL doing studies. ... You gotta get people who don't owe us anything."
According to Goodell, the NFL was "going to let the medical people decide" whether football causes brain damage.
But some researchers wonder whether the league has already drawn its own conclusions.
Last summer, 100 Ivy League and Big Ten scientists met in Chicago for a concussion summit. The keynote speaker was Guskiewicz, the concussion expert who led the proposal for the NFL-funded NIH grant. A former Pittsburgh Steelers assistant athletic trainer, Guskiewicz was named dean of the University of North Carolina's College of Arts and Sciences last year, an achievement that followed two decades of research that helped turn sport-related concussions into a major public health issue.
Once a harsh NFL critic who compared the league's "industry-funded research" under Pellman to an airport security breach, Guskiewicz is now part of the NFL's Head, Neck and Spine Committee, which helps shape league policy. Some researchers believe that since affiliating with the league, Guskiewicz has softened his previous positions, including the conclusions of his own pioneering research. Guskiewicz said he has remained consistent: that previous concussions can lead to long-term consequences, including depression and mild cognitive impairment, but that much is still unknown about the connection between football and brain disease.
During his speech in Chicago, Guskiewicz downplayed the importance of "sub-concussive" hits, which some researchers cite as the likely source of brain disease associated with football. The theory is especially thorny for the NFL, because it would mean the damage is produced by the unavoidable helmet-to-helmet contact that occurs on every play and in practice drills.
Guskiewicz, in an interview this week, said the title of his speech was, "The State of Sports Concussion: Legitimate Concerns vs. Paranoia," and that its intent was to be provocative for the audience of clinicians and researchers. He said he pointed out that previous research, including his own, indicates that repeated diagnosed concussions may lead to long-term problems, but that "the pendulum has swung so far on this notion of repetitive sub-concussive impact just because it, quote-unquote, 'should make sense.' And what I keep emphasizing is, 'Well, let's figure it out.' Where's the evidence to support this notion or theory?"
Dr. Hans Breiter, a Northwestern University psychiatrist and behavioral scientist, was in the audience. Breiter has worked closely over the past two years with Purdue's Talavage and Nauman, who have researched the effects of sub-concussive blows on players.
"The whole group of us at our table were going, 'What? You gotta be kidding me,'" Breiter said. "It was really bizarre. We all started to look at each other and say, 'This is what happened with Big Tobacco.' It felt like we were going back to the stage where the people who were funded by Big Tobacco were saying smoking is not harmful."
The NFL made its $30 million pledge to the NIH in September 2012 -- four months after Junior Seau, the Hall of Fame linebacker, committed suicide by shooting himself in the chest.
The grant was the largest individual donation in NFL history.
Inside the NIH, senior officials debated whether to accept the NFL's money, Outside the Lines has learned. One official urged caution: Kathy Hudson, a biologist who spearheads major scientific initiatives for the NIH as deputy director for science, outreach and policy.
"She said, essentially, 'Don't do this. This is a recipe for disaster,'" said a person familiar with the discussions who requested anonymity because of the sensitivity of the issue. "She was concerned about two things: Would the NFL provide the entire amount? And would they try to control how the money was spent? In a sense, she was right, on both counts." Hudson could not be reached for comment.
Other officials, including Story Landis, the former director of the National Institute for Neurological Disorders and Stroke, and her deputy, Dr. Walter Koroshetz, a champion of traumatic brain injury research, believed the donation was an opportunity to make significant progress on concussion research that would benefit not only athletes but also the general public. The NFL couldn't intervene, they believed, because of written agreements that placed a nonprofit foundation between the league and the NIH.
That organization, the Foundation for the NIH, raises private money for NIH research. The NFL donation was a coup: The league became one of the FNIH's 13 largest donors, along with pharmaceutical and biotech companies and the Gates Foundation. The FNIH had hoped to recruit additional donors for the NFL-funded research initiative, dubbed the Sports and Health Research Program, from the NHL, the NCAA and other leagues, but efforts were unsuccessful.
An NFL media release described the gift as "unrestricted." But the written agreements state otherwise.
A September 2012 agreement between the NFL and the FNIH outlined multiple research areas that were preapproved by the league, according to a source with access to the document. Pellman, the NFL's medical director, signed the agreement and is listed as the league's primary contact. Pellman was lead author on nine of the 16 NFL papers that described concussions as minor injuries. He and his co-authors once wrote: "Professional football players do not sustain frequent repetitive blows to the brain on a regular basis."
According to a memorandum of understanding between the FNIH and NIH, released under the Freedom of Information Act, the NFL can offer input on "research concepts" through a "stakeholder board." FNIH financial statements described the $30 million donation as a "conditional contribution" that allows the NFL to cancel its funding.
Tom Murray, president emeritus at the Hastings Center, a nonprofit bioethics research institute, called the claim that the NFL donation came with no strings attached "borderline ludicrous."
"This looks more like a thick stainless steel cable than a string," Murray said.
In December 2013, the NIH used the NFL money to award two $6 million grants for a "comprehensive investigation" into chronic traumatic encephalopathy and six pilot studies totaling $2 million. The recipients of the CTE grants were Dr. Wayne Gordon, a professor of rehabilitation medicine at New York's Mount Sinai Hospital, and Dr. Ann McKee, a neuropathologist with Boston University and the Department of Veterans Affairs, who, like Stern, is part of BU's CTE Center.
The parallel studies produced a consensus definition of CTE as a distinct disease caused by repetitive head trauma -- even though the NFL had rejected a connection between football and CTE for years.
McKee, who has clashed with the NFL over her assertion that a large percentage of pro football players are likely to get CTE, said she was grateful for the league's funding: "They've been an absolute lifeline for my work. They have funded me in huge ways."
At the same time, she said, she still worries about the NFL's conflicted role.
"I think you always feel like you're on tenuous ground with them, always concerned that their commitment may not be long term, that it may be conditional," said McKee. "You do feel like you take two steps ahead, and you have to take two steps back. So it's a mixed message."
The NIH moved to expand on McKee's and Gordon's research in 2014 -- backed by the NFL donation. The league agreed to use the remaining $16 million for a seven-year study to detect CTE -- which currently can be diagnosed only after death -- in living patients. That discovery has been described as the holy grail of concussion research, with broad implications that almost certainly would transform sports, including football.
The NFL "agreed in principle to the study," a source close to the discussions said.
The White House announced the NFL's role in the study during a 2014 "Concussion Summit." The league trumpeted the project in its 2015 NFL Health and Safety Report: "This study will aim to use longitudinal data to better understand the neurological mechanisms of CTE and enable the development of an evidence-based clinical diagnosis," the report said.
Top NFL advisers were applying to run the NFL-funded project.
In addition to Guskiewicz, the proposal's lead investigator, three other NFL advisers were listed on the application as having potential roles in the study: Ellenbogen; Michael McCrea, a Head, Neck and Spine Committee member and Green Bay Packers consultant who directs brain injury research at the Medical College of Wisconsin; and Dr. Bruce Miller, a brain trauma expert at the University of California San Francisco who consults with the league. Also listed was Dr. Geoff Manley, a UCSF neurosurgeon who has worked as an unaffiliated sideline consultant at Oakland Raiders games.
Miller and Manley are colleagues of Berger, the UCSF neurosurgeon who also leads the NFL Subcommittee on Former Players and Long-term Effects of Brain and Spine Injury. According to the document submitted to the NIH, Guskiewicz, McCrea and Miller were listed as co-principal investigators, meaning that millions of research dollars and funding for other indirect costs would flow to their institutions. Ellenbogen was listed as a "co-investigator."
Ellenbogen, who wrote in an email that he will receive no financial benefits from NFL-funded projects, defended the idea of members of the Head, Neck and Spine Committee applying for research money coming from the league: "We spend thousands of hours pushing the science forward for free because we want to answer essential scientific questions pertinent to our clinical practices. We are often in the best position to do it."
Guskiewicz told Outside the Lines that when he was invited to join the Head, Neck and Spine Committee, he would not have accepted the position "if my research team would be ineligible to apply for funding. Our livelihood depends on being eligible to apply for grants to keep our research programs going."
He said the Head, Neck and Spine Committee is independent of the NFL funding decisions. "I'm a member of the Head, Neck and Spine Committee, which is not the NFL," Guskiewicz said. "I can assure you that the Head, Neck and Spine Committee never had to approve anything or draw anything up. That's the distinction: I'm part of a seven-member committee, which is different from the NFL."
The Stern proposal included McKee and one NFL-connected researcher: Dr. Robert Cantu, a clinical professor of neurosurgery at Boston University and a concussion expert who is a senior adviser to the Head, Neck and Spine Committee.
The competing proposals included researchers with starkly different views on the relationship between football and brain disease.
Stern and his Boston University colleagues believe that repetitive head trauma associated with football has led to an alarming number of CTE cases, suggesting, as McKee once said, that "a shockingly high percentage" of NFL players are likely to get the disease.
Guskiewicz and other top NFL advisers believe the research is inconclusive and that the Boston University researchers have oversold their findings, feeding a growing hysteria over the risks of playing football. In 2013, Berger accused the Boston researchers of scientific bias, charging that "their whole existence, their funding, relies on this [idea] that it's a fact if you play football you're going to have some form of cognitive impairment."
In 2014, Stern filed a 61-page affidavit opposing the settlement of a lawsuit in which thousands of former players accused the NFL of covering up the link between football and brain damage. He wrote that the settlement would deny benefits to the most severely disabled former players. Stern said he was not paid for his findings.
Last May, the NIH informed Stern that his group had won the grant after an exhaustive process that included a panel of peer reviewers and a separate evaluation by leading experts on an NIH advisory council. Multiple sources close to the process said the competition wasn't close.
In June, after Pellman's email challenging the decision, Ellenbogen, Berger and Jeff Miller, the NFL's senior vice president for health and safety policy, confronted the NIH in a call with Koroshetz, who that month succeeded Landis as director of the National Institute for Neurological Disorders and Stroke, Koroshetz told Outside the Lines.
Under the agreement that governs the NFL donation, the FNIH can work to resolve any donor issues that come up. Ellenbogen and Berger would have been appealing as donors -- representatives of the NFL -- but Ellenbogen was also an applicant for the same grant, and Berger's department stood to benefit if the proposal had been approved.
"The idea that you would have anybody make that call who stands to gain, that would be just insane. You would never do that," said Dr. Steven DeKosky, the deputy director of the McKnight Brain Institute at the University of Florida, who once came under fire from the NFL after he and a former colleague, Dr. Bennet Omalu, reported the first case of brain disease in an NFL player. "That's just inappropriate, if they were competing for the same grant and on the call to protest something about the guy who got the grant. They should be nowhere near that."
Ellenbogen denied that he challenged the Stern grant.
"I do not know Dr. Stern and therefore do not have an opinion of him but assume he is a very capable scientist if he was awarded the grant," Ellenbogen wrote in an email. Berger did not respond to an interview request.
An FNIH spokeswoman declined to comment. The NIH did not respond to questions about potential conflicts of interests or potential violations of protocol. Discussion of grant reviews is generally confined to the principal investigator, according to a source familiar with the process, and applicants are usually prohibited from discussing competing grant proposals.
Guskiewicz said he was unaware that the NFL challenged the NIH's decision to award the grant to Stern. Asked if he thought the call was appropriate, he responded: "I'm not going to answer that question. I just don't want to comment."
In September, an NIH advisory council determined that the allegations of bias were unfounded. With no indication that the NFL intended to fund the study, the NIH announced that it would pay for it on its own.
The NFL has said the NIH makes its own funding decisions and has termed prior Outside the Lines stories about the controversy as "inaccurate." The league said NFL funds have "always" been available for the Boston University study.
In response to Outside the Lines' reporting on the issue, four Democrats on the House Energy and Commerce Committee have requested information about the NFL's funding relationship from the NIH and the FNIH. Both agencies said they are working to comply with the request.
NFL politics also have clouded essential research into helmet sensors -- tiny devices that measure the frequency, force and location of hits to the head.
Dozens of college and high school football programs use helmet sensors to help identify concussions and track the number of times players are hit in the head. The U.S Department of Defense uses sensors to study blast injuries. Guskiewicz, who won a 2011 MacArthur "Genius Grant," uses the devices at North Carolina. His sensor studies were the basis for the NFL's decision to move kickoffs up to the 35-yard line, a rule change the league made in an effort to reduce concussions.
Guskiewicz told Outside the Lines three years ago: "If we're sitting here a year from now and we're not any closer to on-field, real-time biomechanics being measured, I'll be real frustrated, and perhaps ready to throw the towel in."
Instead, the NFL funded its own study, which was used by the league and the players' association to justify the decision to suspend implementation of devices that, at minimum, would tell the league how many times its players are struck in the head over the course of a season.
The league commissioned MEA Forensics, a British Columbia firm that specializes in legal research and expert witnesses, to study two sensors. One was produced by X2 Biosystems, a Seattle firm, and was endorsed by Dr. Stan Herring, the Seattle Seahawks team doctor who recently resigned as a member of the Head, Neck and Spine Committee.
Herring is listed on X2's website as chairman of the company's medical advisory board: "The data we collect provides immediate utility to trainers, coaches, and physicians right now, and will be invaluable in supporting the future research that will revolutionize our understanding of this important injury."
The other sensor studied by MEA, called the HITS System, had been used for years in Guskiewicz's concussion research program at North Carolina.
The study showed that the sensors detected 96 percent of head hits but concluded that the sensors' ability to determine the point of impact was "significantly" flawed -- even when the sensor was just off by 2.3 degrees, a difference measured in millimeters.
The analysis was quickly seized upon by both the NFL and the players union, which blasted the devices as useless.
During a peer review for Annals of Biomedical Engineering, the scientific journal that published the study, MEA was asked whether "NFL selected consultants" had created standards for accuracy so restrictive that "it raises questions on the overall objective of this study (i.e., no sensor could ever pass this)," according to emails obtained under the Virginia Freedom of Information Act.
After the initial submission, the co-editor on the paper, Stefan Duma, the head of Virginia Tech's department of biomedical engineering and mechanics, asked the authors to "clarify that the NFL paid MEA to perform this analysis." A disclosure statement submitted with the paper "obfuscates this direct relationship," Duma wrote.
Among the researchers who worked as "outside consultants or as resources" on the study, according to another email, were Pellman; Dave Viano, a biomechanics specialist who once co-led the NFL's since-disbanded MTBI committee; and Kristy Arbogast, a specialist in pediatric injury biomechanics who is also a consultant to the NFL Players Association.
In an interview with Outside the Lines, Duma said he believed that the study set standards for accuracy that are unattainable.
"I'm not accusing them of fabricating anything, but they are looking at it from a very particular viewpoint," Duma, who uses sensors in his research at Virginia Tech, told Outside the Lines. "They have created a system that's impossible to pass."
The paper's lead author, Gunter Siegmund, MEA's research director and an independent consultant to the NFL Head, Neck and Spine Committee, did not respond to a request for comment. An NFL spokesperson did not answer a question about the sensor study.
After receiving comments from reviewers, the authors added the sentence: "Despite not meeting our validity criteria, both sensors can provide useful information about the number, direction and magnitude of head impacts."
The authors also included the disclosure: "This study was funded by the National Football League."
The league hasn't been the only impediment to sensor use; the NFL Players Association also has cited questions about their accuracy. "The NFL and NFLPA did a feasibility study, and the results of that study, based on the opinion of independent researchers, was that the data was too unreliable," said Sean Sansiveri, the union's vice president of business and legal affairs. The union also has expressed concerns that rather than gathering data for research purposes, teams would use the information for personnel decisions.
"This whole idea that we don't need to use sensors at all, because they don't give us 100 percent accuracy is really the wrong way to go about science," said Purdue's Talavage. "It's not what science is about. Science is about, 'Give me all the data, and I'll figure it out.'"
Asked how valuable it would be if the NFL were to gather sensor data, he said: "Ridiculously valuable. Because if nothing else ... regardless of the system, you'll have a good idea of how much exposure in terms of raw numbers. ... How often do [players] get hit? How often do they get hit in practice? And how often do they get hit in games?"
Guskiewicz said he continues to stand by the sensors as "very good at measuring the number of impacts that a player receives in practice and games, and for that reason I continue to use them." He said the devices are more "limited" in detecting the point of impact and magnitude of the hit but still provide useful information.
Guskiewicz was an author on the NFL-funded sensor study.
Asked how he felt about the NFL and the NFLPA citing it to keep sensors out of the league, he replied: "Maybe they should go back and read the findings of the paper."