Do Corporate Dollars Strangle Scientific Research?

In 1954, when Jonas Salk announced that he had developed a vaccine for polio, he appeared in a television interview with Edward R. Murrow. The newscaster asked Salk, then a bacteriologist at the University of Pittsburgh, why he had failed to patent an invention clearly worth millions. Salk blinked and replied, "How can you patent the sun?" This question sounds quaint today, when everything from genes to vaccines is patented by universities and their researchers. In 1995 the University of California system alone pocketed $57 million from patents established since 1980, when the Bayh-Dole Act permitted universities and scientists to lay financial claim to faculty inventions created with federal research funds. The biologist Peter Medawar once quipped that the "dire equation" of scientific research was "Useless=Good." But the Bayh-Dole Act has changed the formula, engendering an explosion of collaboration between academia and the business world. While corporate funding accounts for only around 10 percent of today's research dollars, the vast majority of working scientists carry some kind of industry grant; and as government wells dry up, the percentage keeps growing. Add in entanglements like honoraria, consultancies, and those highly profitable patent licenses -- one UC geneticist received $521,276 last year in royalties from his share of a university patent on genetically engineered strawberries -- and the line between academia and industry becomes increasingly difficult to draw. All this raises a question: Might scientists aligned with companies be lured off the path of dispassionate inquiry? Sheldon Krimsky, a professor of environmental policy at Tufts, recently warned in The Boston Globe that the scientific community was draining itself of "a reservoir of disinterested experts who can be called upon to advise government policy makers on the safety and efficacy of treatments, on the hazards of chemicals, [and] on the safety of technology." Nowhere is this concern more visible than in the precincts of scientific publishing-a world almost obsessively keyed to screening out bias. Increasingly, editors have begun to ask uncomfortable questions: Does the profit motive lead scientists to downplay unfavorable results or overemphasize success? Do industry funders secretly massage manuscripts? And, after years of resistance, many editors are now requiring contributors to disclose any financial interest in research results. Krimsky gives the reason as follows: "If I were to tell you that a couple of academics had just published a positive study of a 'safe' cigarette," he asks, "wouldn't your first question be: Who funded that research? Could these scientists make millions from the product's success?" In response, many scientists argue that financial incentives pose no direct threat to the integrity of their research. "There's a real trumping up of this issue," complains Boston University's Kenneth Rothman, editor of the journal Epidemiology. "We're educating readers that credentials, not science, is what they should focus on. Trust the report if it's funded by X, but not if it's funded by Y. That's just unhealthy."So has industry polluted the academic pond? A recent study by Krimsky highlights, if nothing else, the extent of the corporate presence. In the report, presented this past February at the annual meeting of the American Association for the Advancement of Science, Krimsky's research team scrutinizes 789 articles written by authors from Massachusetts universities published in leading scientific journals in 1992. The results: In one out of three cases, at least one author had a vested interest in his research. These interests took a variety of forms-owning a patent directly related to the published work; being a major stockholder or executive in a company with commercial interests tied to the research; or serving on the board of directors of such a company. Consultancies and honoraria weren't considered because they're all but impossible to track. "If you factor those in, the percentage is likely higher," says Krimsky. All of these relationships are perfectly legal-and largely unregulated. But that's not to say they haven't stirred up controversy. Government concern about conflicts of interest first emerged in 1988, when a scandal broke out over a National Institutes of Health-funded study on Genentech's heart-attack medication, TPA. The genetically engineered wonder drug, an extremely expensive anticlotting agent, was heralded in a report in the Journal of the American Medical Association (JAMA) as substantially more effective than cheaper medications like streptokinase. But as physicians were busy combing through the study's data, a reporter from Newsday revealed that at least thirteen researchers were long-term Genentech stockholders-some to the tune of $100,000. (More than thirteen were likely involved, but many of the scientists on the research team refused to reveal what they termed "confidential information.") Genentech, a nascent biotech firm, marketed only a slim roster of products; therefore, its success depended heavily on TPA's acceptance by the medical community. Following several conflicting studies, the medical community today remains divided over whether TPA is significantly more effective than streptokinase. While the exposed scientists angrily denied that their financial interests could bias their objectivity -- after all, what scientifically literate investor wasn't bullish about biotech in the Eighties? -- Congress held hearings on the controversy. Ultimately, though, most of the proposed remedies, such as banning federally funded scientists from owning stock in firms affected by their research, were shot down as too invasive. After protracted wrangling, the NIH finally adopted a loose set of rules in 1995: Researchers receiving federal moneys are now required to disclose -- to their universities, not the public -- any involvements exceeding $10,000 in companies that might benefit from their research. Journals have no legal obligation to disclose such information about their contributors. Krimsky's study of journal articles attempted to reopen the very questions the NIH had hoped to close. In order to complete his work, Krimsky combed through annual reports filed by companies and consulted worldwide patent records. Sound sneaky? In fact, he had few options -- for only one of the 268 relevant articles contained a disclosure statement. (To be fair, a few of the journals in Krimsky's study have adopted some kind of disclosure policy since 1992.) In his March Boston Globe piece, Krimsky urged journals without disclosure policies to jump on the ethical bandwagon: "The public has a right to know whether scientists who are employed at universities...have any financial interests in their published work." Especially, critics say, when such conflicts are legion. The editor of a leading medical journal says, "My guess is that some pharmaceutical companies try to buy up every investigator in sight." Jerome Kassirer, editor of The New England Journal of Medicine, reports that he had to go to his "sixth choice" for a recent editorial before he could find someone fully free of financial ties. ("The biggest problem comes when you're trying to evaluate equipment or drugs," he says.) This past April in JAMA, West Coast editor Drummond Rennie detailed his difficulty in finding unbiased experts on thyroid disease, since virtually all of the leading experts have financial ties to Knoll Pharmaceuticals, the leading thyroid-drug manufacturer. Over the past several years, the press has picked up on the conflict-of-interest problem and had a field day exposing the financial ties of supposedly impartial scientists. In a high-profile case in 1994, Harvard rheumatologist Peter Schur, the editor of Arthritis and Rheumatism, rejected several submissions claiming a link between silicone breast implants and autoimmune diseases like rheumatoid arthritis. At the same time, Schur published a rival study (of which he was a co-author) denying any such connection. Litigants against breast-implant manufacturers revealed that during this same period, Schur had acted as a paid consultant to lawyers defending implant manufacturer Dow Corning-to the tune of $300 an hour. (At the time, Boston attorney Frederic Ellis complained, "The evidence suggests that the manufacturers have attempted to control science by hiring experts who are key players in the decision making about what gets published.") Schur is not alone. Last spring, the environmental group Ozone Action released documents revealing that Arizona State climatologist Robert Balling, one of the most vocal naysayers of global warming, had received over $300,000 in research funding and honoraria from oil companies-not to mention the Kuwaiti government. And last summer Michael Macknin, a physician at the Cleveland Clinic, published a study indicating that zinc gluconate lozenges reduced the length of a cold by almost half. What the journal that published the study, The Annals of Internal Medicine, did not report was that Macknin purchased nine thousand shares of stock in a private offering by Quigley, the company manufacturing the lozenges-and that Macknin made a whopping $145,000 when Quigley stock jumped after his data were publicized.While such revelations surely embarrassed researchers dragged through the media mud, it's not so clear that their science was skewed. Nobody, for example, has argued that Macknin's zinc-lozenge study was improperly conducted; his decision to cash in on promising results (after the trials were finished) smacks more of insider trading than scientific fraud. Balling's opinions on ozone, though by no means the dominant view, have been published in some peer-reviewed journals. And the publications committee of Arthritis and Rheumatism ultimately judged Schur's editorial actions "not prudent" but, after looking at the rejected manuscripts' data, deemed his decisions scientifically sound; he finished out his term as editor. However, Schur did agree to resign from a larger Harvard study on breast implants. Harvard's Charles Hennekens, the chief of that study, emphasized that the decision was "based more on the perception of how the results of the studies might be viewed, rather than any reality." But if all the media bluster is by and large a perception problem, why cave in to facile cynicism? That's the argument advanced by a recent editorial in the venerable British journal Nature. In the printed equivalent of an exaggerated yawn, Nature's editorial staff published a response to Krimsky's study, saying the results were "no surprise" and noting that "virtually every good paper with a biotechnological relevance" most likely "has at least one author with a financial interest -- but what of it?" Krimsky's study, noted the editorial, "makes no claim that the undeclared interests lead to any fraud, deception, or bias in presentation, and until there is evidence that there are serious risks of such malpractice, this journal will persist in its stubborn belief that research as we publish it is indeed research, not business." In other words: Science is science, dammit! Krimsky is frustrated by Nature's argument. "Science doesn't take place in a vacuum," he insists. "Disclosure simply allows readers to ask if a researcher's financial interest may have led him to emphasize certain data or neglect other paths of inquiry." Failing to disclose, Krimksy reasons, allows industry to trade on the presumption of objectivity. (Perhaps that's why no companies require affiliated researchers to reveal their ties when publishing their findings.) Many science editors have finally started coming around to Krimsky's point of view. JAMA now demands disclosure from prospective authors, as do Science, The Lancet, and NEJM. George Lundberg, editor-in-chief of JAMA, says, "I understand the frustration of honest scientists, but from an editorial standpoint, the price of not being duped is eternal skepticism." Lundberg finds Nature's blithe posture "extraordinarily naive, possibly representing editorial inertia." Jerome Kassirer of NEJM will just say, "I only hope they reconsider before they're embarrassed."While many journals now consider financial disclosure statements worthwhile, NEJM has taken the battle against bias one step further. In 1990 the journal's editors banned researchers from penning editorials on subjects that could in any way benefit companies they have alliances with -- or hurt rivals. Kassirer says he's "proud" of NEJM's tough stand and "hopes the others will soon follow our lead." Depending on how you see these issues, NEJM is either at the vanguard of responsible science or, as one rival editor snipes, "whatever's beyond holier-than-thou." In any event, a recent crisis at NEJM reveals the headaches that can result from all this hand-wringing. This past August, the journal published a lengthy report indicating that Redux, a potent new weight-loss drug that's already racked up more than $200 million in sales, might carry a potentially lethal side effect-pulmonary hypertension. The study's troubling conclusions were given a more heartening spin, however, in an editorial commissioned by Kassirer appearing in the same issue. The editorial, co-written by Harvard's JoAnn Manson and Penn's Gerald Faich, turned a critical eye on the research report, arguing that the drug's minimal mortality risk was not substantially greater than that of everyday medications like penicillin and the Pill. The duo also pointed to methodological flaws that they claimed "may have inflated the estimates of risk." Considering the unquestionable health risks associated with obesity, Manson and Faich concluded that "the possible risk of pulmonary hypertension associated with dexfenfluramine is small and appears to be outweighed by benefits." The day the journal was released, the stock price of Interneuron Pharmaceuticals, the Massachusetts-based maker of Redux, jumped 13 percent. Within days of the issue's publication, word spread among scientists that Manson and Faich had consulting ties to companies producing and distributing Redux, in direct violation of NEJM policy. While Manson's relationship with Interneuron had ended one year before she ever put pen to paper, Faich admitted to an ongoing consultancy with American Home Products, the U.S. distributor for Redux. (Faich told the Boston Globe his failure to disclose was "an error.") The NEJM editors, claiming surprise, hastily published a statement saying that they had been misled-and that publishing the editorial was a mistake. "When considered as the opinion of unbiased experts," the editorial explained, "this is just the sort of practical summary we want from editorialists. But when considered as the conclusion of people who were paid consultants..., it raises troubling questions." Kassirer asserts that the revelation of the duo's consulting history dramatically affected his perception of their argument. "The paper seemed to take a unique perspective-to take the data in the research article and compare it with the dangers of obesity in general," he says. "But when I found out that Faich, at least, was an ongoing consultant, well, it began to look different, more like risk analysis or something." Many find Kassirer's logic baffling. In a blistering response published in Epidemiology this past May, Kenneth Rothman wrote: "The premise that we cannot take an editorial comment at face value, that we can only interpret it if we understand the state of mind of the author...actually compromises rather than enhances objectivity, by...encouraging readers to apply ad hominem evaluations." Others worry that NEJM will lose intellectual vigor if top experts like Manson and Faich are barred from scientific debates. Lundberg, for one, strongly believes that "disclosure is sufficient." He adds: "Rather than barring the door, we think our readers want to hear the opinions of the most knowledgeable workers in any field." As an example, Lundberg envisions an eye surgeon who develops a striking new technique for treating cataracts. Though the surgeon may strike it rich if his novel method is adopted, "you'd still want him to disseminate his ideas. That's in the best tradition of Hippocrates." As for Manson, she's infuriated at any aspersions cast her way. "I didn't stand to make a cent from this editorial," she says. She says that her involvement with Interneuron consisted of "testifying before the FDA in support of pharmacological approaches to obesity. I was paid for my time, but I was mainly doing my civic duty. I believe these drugs can help people." And Manson insists she wasn't hiding anything. "I'm a proponent of disclosure, for God's sake!" she says. "I've always revealed links when they're relevant." To wit, in 1995 Manson published a research report in NEJM underscoring Redux's effectiveness in promoting weight loss; the published paper mentions her Interneuron consultancy. So why, then, did Kassirer invite Manson to write the Redux editorial? "By their logic, they shouldn't have asked for my opinions in the first place," Manson says. "But they did, because I know a lot about this subject."Aside from the matter of whether financial ties bias researchers when they write up their results, scientists and editors have lately begun pointing to an even more troubling phenomenon: industry's manhandling of manuscripts during the period before publication. When scientists sign on for industry funding, their contracts typically require them to keep findings confidential until any possible patents are fully secured. Firms need to protect their investments, after all, so it's only natural that a company would want to peruse a researcher's data before, say, the readers of Science do. But when company executives examine an early manuscript -- as is customary today -- it appears they often use the opportunity to suggest, or even demand, alternative ways to frame data. Unflattering data get minimized; promising results hyped. Frank Davidoff, editor of the Annals of Internal Medicine, says that he "sometimes gets submissions that have the marketing department written all over them, certain adjectives that just betray some kind of spin." David Sharp, deputy editor of The Lancet, agrees. "All the time, I look at manuscripts where I'm absolutely sure that the medical division of some company has looked over the draft," he says. "You wonder sometimes: Who actually owns the data?" In fact, universities are careful not to turn over total control to industry funders. The top research institutions do allow companies to review results before they are published (and delay publication of an article for up to one month after a patent has been secured). But they simultaneously insist that all contracts guarantee scientists' ultimate right to publish their data as they see fit. However, as a much-discussed case at UC-San Francisco recently established, the system isn't entirely leakproof -- not when millions are at stake. Ten years ago, Betty Dong, a UCSF pharmacologist, was asked by London's Boots Pharmaceuticals to compare their profitable medication Synthroid with three cheaper, knockoff medications. At the time, Synthroid was the only drug of its type considered truly effective in combating hypothyroidism; other drug companies, eager to carve up Synthroid's monopoly, claimed that their variants worked equally well. Boots officials hoped that Dong would confirm Synthroid's unique properties. The company's selection of Dong wasn't accidental: She had just published an article emphasizing the different effects of various hypothyroid medications. Dong agreed to perform a rigorous clinical trial-and ended up confirming Boots's worst fears. Synthroid, she found, wasn't any better than its rivals. Since Synthroid dominates a $600 million U.S. market, Dong's study had huge financial implications-both for Boots (big losses) and for patients (big savings). Company executives immediately disputed her results, and Dong spent a full four years wrangling with the company over her interpretation. Despite Boots's objections, Dong ultimately wrote up her findings, which were accepted for publication in JAMA in January 1995. But the article failed to appear. It was quashed at the last minute by Boots. In signing her contract, Dong had accepted a clause stating, "Data obtained by the investigator while carrying out this study is also considered confidential and is not to be published or otherwise released without written consent." When confronted with a possible lawsuit, UCSF officials caved in, forcing an appalled Dong to yank her manuscript. Only after The Wall Street Journal exposed the suppression did Knoll Pharmaceuticals (which purchased Boots in 1995) relent and allow publication of the report this past April. The resulting front-page scandal tarnished the reputation of Knoll -- not to mention that of UCSF administrators, who were criticized for failing to support Dong's right to publish. For the record, the view of Knoll president Carter Eckert is: "I stopped a flawed study that would have put millions of patients at risk." But why did Dong sign away her autonomy? Despite some news reports to the contrary, she says that at first she refused to sign the contract-alerting a UCSF patent officer about the troublesome clause. This university official, Dong says, "gave me pretty bad advice." In her words, the administrator "told me not to worry about it," suggesting that no pharmaceutical company would have the gall to violate so flagrantly the norms of academic research. "Well, he was wrong in this case," Dong laughs ruefully.With Dong's study, public health was affected in a largely financial sense: Nobody claims that Synthroid doesn't work, just that cheaper variants work equally well. But another recent uproar underscores the potential health dangers of granting industry control over publication of scientific data. This past January, Canadian blood specialist Nancy Olivieri risked a lawsuit when she broke her confidentiality agreement with Toronto-based Apotex Pharmaceuticals. Apotex produces defer-iprone, an experimental drug designed to treat thalassemia, a rare disorder in which iron slowly (and fatally) builds up in the bloodstream. Olivieri, who works at the Hospital for Sick Children in Toronto, drew on Apotex funds to study deferiprone-with what seemed like exciting results. Previously, thalassemics had to undergo extensive blood transfusions in order to filter out excessive iron. The drug seemed to perform the same task, releasing patients from the shackles of the transfusion process. In 1995 Olivieri published her preliminary findings in NEJM (with Apotex's blessing). But as time progressed, she found that her patients who continued treatment with deferiprone weren't doing so well. Iron levels began creeping upward again. She informed Apotex officials who discounted her findings and, following their contract, prohibited her from publishing her new results. Apotex eventually appointed an "independent" review committee of four scientists, all of whom promptly adopted the company line. Granted, Olivieri might well be wrong about deferiprone's long-term efficacy. But why wouldn't Apotex allow her to submit her results to a peer-reviewed journal? If her findings were so clearly off base, they wouldn't likely pass muster with outside reviewers. Frustrated by the impasse, Olivieri broke the confidentiality agreement by allowing her lawyer to release the following statement to the press: "Dr. Olivieri wants to put her views on the record and allow scientific debate to take place. She is the leading authority in the world on this and she felt it important to make the scientific community aware."Universities are supposed to protect their employees from being put in Olivieri's unhappy situation. But when the financial stakes are really high, even the strongest contractual protections may not be enough to shield academics from aggressive pharmaceutical executives. Consider the example of a three-year study of calcium channel blockers, a high-blood-pressure medication that in 1995 was the most frequently prescribed drug in the U.S., yielding over $3.7 billion in sales. The study was designed to compare DynaCirc, a popular calcium channel blocker, with a traditional diuretic to determine which was more effective in reducing hypertension. Substantial funding came from Swiss pharmaceutical giant Sandoz, which produces DynaCirc. The study, headquartered at Wake Forest University, included contractual language ensuring academic freedom for the nine researchers involved. But, says Curt Furberg, a professor at Wake Forest, Sandoz "began to intervene anyway" when the study revealed something unexpected: Not only was DynaCirc no more effective than the diuretic, but patients taking DynaCirc experienced nearly twice as many strokes and "vascular events" like angina and heart failure. Furberg was alarmed-the millions of patients who took calcium channel blockers were possibly at risk. While the study was specifically designed to compare efficacy, not safety, many of the researchers were nonetheless troubled by the implications. Sandoz research staff, Furberg says, became "agitated by these findings" and "introduced themselves into the drafting of the report," where "they put in language designed to downplay the problems." In addition, Sandoz executives bestowed gifts on researchers that made Furberg and several other group members uncomfortable --such as posh lodging in Lake Tahoe for a group meeting and snazzy ski jackets emblazoned with the company logo. The last straw, Furberg says, was when Sandoz circulated a final draft of the paper that "had eliminated key phrases that already had been agreed upon by the group," such as words highlighting safety questions in the paper's abstract. Last summer, four of the study's major authors quit in protest over what they considered excessive pressure from Sandoz. To be sure, some members of the research team, including lead author Vardaman Buckalew of Wake Forest, apparently agreed with Sandoz's milder interpretation of the data. (Writing in JAMA, Buckalew underscored that he felt the results "should be interpreted with caution and, at most, indicate a need for further study.") But even Buckalew notes that "Sandoz officials involved themselves in discussions as to what language to use in the manuscript." So what happened to the article? The report appeared in JAMA last September, with an addendum revealing that several researchers had left the study and removed their names from the manuscript --something virtually unheard of in scientific papers. So were the editors of one of the most prestigious medical journals duped into publishing spin-controlled science? Not quite. Furberg's story has a surprise ending: Peer review led to the removal of many of Sandoz's changes. "Peer review worked," he says jubilantly. "The reviewers must have seen that the Sandoz-approved draft downplayed the bad and up-played the neutral." Furberg chuckles. "I know I'm happier with the published study than Sandoz is." Even lead author Buckalew agrees that "the original drafts had serious problems. I'm more pleased with the peer-reviewed version." This past January, Furberg and the others who left the study published a follow-up letter in JAMA. "We feel it important to state publicly why we dropped out of the investigative group while the final paper was under preparation," they wrote. "We believed that the sponsor of the study was attempting to wield undue influence on the nature of the final paper. This effort was so oppressive that we felt it inhibited academic freedom." When Lingua Franca attempted to contact Sandoz officials, they did not return calls. "The company has wisely decided not to challenge our JAMA statement," Furberg says.Of course, most scientists never butt heads with the companies who fund them. If anything, such relationships tend to be quite cozy. Harvard public health professor David Blumenthal says, "People focus on these high-profile conflicts, but what may be more common is that a researcher willingly toes the company line-delaying publication, underplaying certain data, and so on." After all, companies seek out researchers whose previous publications indicate a sympathy with their goals. As one biologist puts it, "The fuel industry isn't going to sign contracts with environmentalist researchers, and vice versa." Is it a coincidence, say, that the only recent high-profile study arguing against the link between salt and high blood pressure-appearing last fall in JAMA-was funded by the research offshoot of Campbell's soup? Jeremiah Stamler, a hypertension expert at Northwestern, doesn't think so. In fact, he claims that Campbell's and other food companies have labored hard to "create the impression that there's a scholarly dissensus on salt, when there isn't one-at least among federally funded researchers." Stamler's 1988 INTERSALT study, the most comprehensive look into the matter, shows a clear link between salt and hypertension. His findings, he says, have been nitpicked by scientists with financial ties to the food industry. One of these salvos, funded by the Salt Institute, was so feeble that the British Medical Journal published it last May with a mocking editorial introduction: "This analysis is of service only to illustrate the lengths to which a commercial group will go to protect its market when presented with clear evidence detrimental to its interest." Some scientists take issue with the anti-salt argument. Suzanne Oparil, a cardiologist at the University of Alabama, says: "In my opinion, if you required dramatically lower salt levels across the board, you'd make food inedible by our standards," she says. "It's a very costly and unrealistic idea and leads to reduced lifestyle satisfaction. There are other, better ways to prevent heart disease." Oparil, who currently consults with Campbell's in its efforts "to develop a heart-healthy food line," claims that Stamler isn't playing fair when he points to his opponents' dependence on industry funding sources. "He's emeritus," she points out. "He doesn't have to support a research group anymore." Oparil recently edited a supplement to the American Journal of Clinical Nutrition, titled "Dietary Sodium & Health," representing what she terms a "wide range of views" on salt. The volume is the result of a conference funded by the International Life Sciences Institute, a lobbying group for the food industry with a board of trustees that includes both academics and executives from Kraft, Procter & Gamble, and Kellogg's. Stamler calls the supplement "misleading," noting that the conference was "heavily stacked toward the pro-salt position." (In the end, he grudgingly agreed to submit a paper to the conference "to lend some kind of balance.") Stamler further observes that the journal has "no conflict-of-interest policy-so readers have no way to tell which papers were industry funded." Oparil naturally bristles at what she calls Stamler's attempt to "discredit alternate views" on sodium. Does she worry, though, that her associations with Campbell's and the food industry will lead colleagues to doubt her objectivity? "I'm sure that's what some people say about me behind my back," she sighs.Whether or not industry funding leads to significant bias, experts say research supported by industry is less innovative and more product oriented than federally funded science. In her new book Prescriptions for Profits (Scribner), Linda Marsa reports that in 1991 the Food and Drug Administration approved 327 new drugs and products, "yet only five of them were considered a significant advance. All five of the therapies...were devised with federal funds." Jonathan King, a protein-folding expert at MIT, sees a similar pattern in his field of molecular biology. "I devote most of my energies to studying the complex ways that neurological diseases like Alzheimer's develop," he says. "But some of my colleagues seem primarily interested in hooking up with some biotech firm to develop therapies that may help alleviate the condition. That's important work, without a doubt, but academic research is supposedly all about rooting around for the fundaments of scientific problems." Other scientists think the notion of scientific research's "good old days" is a canard. Harvard's JoAnn Manson is concerned that "misguided ethics police" will clamp down on important alliances that work to benefit patients. "Collaboration between science and industry is crucial, particularly in the medical realm," she says. "This is how the public gets safe and effective therapies. If scientists refuse to talk to companies, progress will slow down." In the end, says Manson, "The public can't have it both ways -- demanding both speedy medical advances and some fairy-tale vision of academic purity." Suzanne Oparil adds that it's perfectly natural for industry-funded research to be less about stargazing and more about "technology and applied research." She notes that "Harold Varmus, director of the NIH, often says that he wants to keep government funds focused on pure bench research, since clinical trials have become so expensive. But clinical trials are crucial for public health." A drug trial for a heart medication, Oparil says, can ultimately cost upwards of $200 million. "So industry has taken up the slack for research that's of a different character, but equally important." No doubt. And yet there's a reason many scientists look back wistfully to the time when the ivory tower and industry remained separate worlds. As Curt Furberg puts it, "When you read a journal article that's funded by the government, you don't immediately ask yourself: Can I trust this?" Even so, Furberg admits, he continues to do research with a mixture of government and industry funds. How does he feel about the compromise? "Fortunately," Furberg chuckles, "some companies are nicer than others."Daniel Zalewski is a senior editor of Lingua Franca.