“We will never accept a world with thalidomide in it,” wrote Randolph Warren on July 17, 1998, the day after the US Food and Drug Administration licensed a chemical that had, between 1956 and 1962, caused birth defects in as many as twelve thousand children. Warren heads the Thalidomide Victims’ Association of Canada. He was born with lower-body phocomelia after his mother was given thalidomide for nausea during pregnancy: the bones of his legs failed to develop, leaving his feet to articulate directly with his pelvis. His arms are shorter than usual, and each of his hands has only four fingers. Randolph Warren describes himself as a thalidomider.
The drug was rehabilitated in 1998 to treat a rare but serious condition called erythema nodosum leprosum (ENL), a complication of Hansen’s disease (also known as leprosy). No more than one hundred new cases are diagnosed each year in the US. But in making their intensely controversial decision, FDA regulators were in effect also saying that thalidomide could now be prescribed for unapproved use in many other diseases as a drug of last resort. Endorsement by a US government agency appeared like a betrayal to many thalidomiders. For while other countries had suffered epidemics of birth defects some forty years before, the US had uniquely protected itself from harm through the stubborn integrity of a legendary FDA official, Dr. Frances O. Kelsey, who in the Sixties prevented the FDA from approving the drug.
But it was the FDA itself which, in 1995, invited the manufacturers of thalidomide to look again at a drug that had, in their words, “the capacity to cure.”1 By 1996, an advisory committee set up by the FDA was reconsidering thalidomide’s capacity to cause birth defects. It concluded that if thalidomide was to be licensed, the FDA somehow had to “ensure that no pregnancy occurs while a patient is taking the drug.” In September 1997, the FDA, together with the National Institutes of Health and Centers for Disease Control and Prevention, convened a public workshop to discuss the evidence for and against thalidomide. Randy Warren was an invited speaker, along with Frances Kelsey.
The renewed use of thalidomide was based on the drug’s encouraging effects on patients with AIDS, various cancers, and diseases in which the body’s immune system attacks its own tissues. At the conference Kelsey set out the history of drug regulation in the US, and the way in which the thalidomide scandal helped to bring about more stringent rules for evaluating drugs. Warren spoke about the difficult balance between thalidomiders’ twin fears of unregulated access to the drug and the wider use that would inevitably follow its regulated availability. In his view, the first fear outweighed the second. He supported temporary approval of thalidomide but with several stipulations. The word “thalidomide” should always be used next to any other name for the drug. Patients must be fully informed about side effects. Doctors who prescribed thalidomide must be certified to do so. Research should seek to find safe alternatives. And “someone must be accountable for dealing with any new victims.”
In July 1998, Janet Woodcock, a director of the FDA’s Center for Drug Evaluation and Research, completed her supervisory review of New Drug Application 20-785 submitted by the Celgene Corporation. She described evidence, some of which was over twenty-five years old, from clinical trials examining thalidomide’s efficacy in treating ENL. Her findings were unambiguous: thalidomide consistently and rapidly improved symptoms for most patients with ENL. She also set out proposed “extraordinary restrictions” on the distribution of thalidomide.
These unprecedented measures included a System for Thalidomide Education and Prescribing Safety (STEPS); fully informed patient consent; creation of a mandatory registry of those taking thalidomide; monthly surveys to detect toxicity; a ban on sharing medications and on blood or sperm donation; agreement by women taking the drug to use two methods of birth control and by men to use barrier contraception; regular pregnancy testing; and limiting drug prescriptions to no more than twenty-eight-day supplies. The aim of these conditions was to “deter any prescribing that is not carefully considered.”
Still, Woodcock concluded that “the Celgene application meets the standard for demonstrating effectiveness of the drug for the intended use.” The first US new drug application for thalidomide was withdrawn on March 8, 1962. This same teratogenic poison was now given official government blessing for clinical use. “The goal,” according to Celgene, was “zero defects.”
Medical journals in the late 1950s did much to create widespread clinical enthusiasm for thalidomide. On November 22, 1958, The Lancet ran an advertisement claiming that Distaval—the UK trade name for thalidomide—had “no known toxicity” and was “free from untoward side-effects.” This strong and unqualified statement came at a time when doctors were increasingly worried about the dangers of barbiturates as sedative agents. Thalidomide was, according to the advertiser’s copy, a sedative “particularly suitable for children.”
The hyperbole adopted by Distillers, the company that made Distaval, would eventually have a grotesque irony. In 1960 and 1961, The Lancet published advertisements depicting an inquisitive blond child playing with open medicine bottles. The headline asserted that “this child’s life may depend on the safety of ‘Distaval,'” and the company went on to claim that “there is no case on record in which even gross over-dosage with ‘Distaval’ had harmful results.” The drug was, Distillers said, “outstandingly safe.”
In their thoughtful account of the rise, fall, and subsequent rise again of thalidomide’s fortunes, Trent Stephens and Rock Brynner place the origins of this confidence firmly within prevailing 1950s naiveté about science. The introduction of thalidomide became the single most important event in ending the postwar mirage of an impending technological utopia. Its consequences were deep and long-lasting—yet they are in danger of being forgotten.
Thalidomide was inadvertently stumbled across in a series of experiments aimed at developing new antibiotics. The man who led this work was Wilhelm Kunz, chief of chemical research at the German pharmaceutical house Chemie Grünenthal. Company executives puzzled over what to do with this substance which did not kill bacteria and had the tantalizing quality of seeming unusually safe. They pushed hard to find a disease for thalidomide to treat since no matter how high the dose given, laboratory animals would not die from it. Thalidomide was tested without success as a treatment for, among other complaints, constipation, seizures, and influenza. Eventually, in 1957, after observing the drug’s calming effects, Chemie Grünenthal decided to market thalidomide as a sedative. Distillers signed up as UK dis-tributors the following year, avoiding any further human trials to check either safety or efficacy. Thalidomide would eventually be sold, usually without prescription, in forty-six countries.
Early experience with the drug proved promising. Doctors reported a surprising lack of toxicity despite accidental overdoses. Such anecdotal observations were confirmed by Distillers in the laboratory. Experiments on mice and rats showed that thalidomide caused far fewer adverse effects than a barbiturate.2 But by the end of 1960, questions were being asked about the drug’s safety. In the British Medical Journal, Dr. A. Leslie Florence described four patients who developed nerve and muscle disorders while taking thalidomide. The Distillers company played down this report by labeling the reaction “rare.” But further cases of neuropathy were described the following year. The typically wry British conclusion was merely that “the perfect hypnotic has not yet been discovered.”3
A gruesome twist came in December 1961, when Distillers first alerted doctors to thalidomide’s “harmful effects on the fetus in early pregnancy.” They withdrew the drug from the UK market immediately, a few days after it had been withdrawn in Germany.4 Two weeks later, on December 16, Dr. William McBride, an obstetrician working in New South Wales, Australia, reported in a letter to The Lancet severe congenital abnormalities in a fifth of women taking thalidomide during pregnancy.5 In his brief one-hundred-word letter, McBride described the beginning of a tragedy that continues today in the lives of approximately five thousand thalidomide survivors.6
Looking through the weekly scientific correspondence in medical journals during 1962 detailing thalidomide’s devastating toxicity, we can relive the panic of the time. Distillers made pleas for calm. Researchers undertook desperate efforts to find other causes for the reported birth defects. All those trying to secure more certain information about the evolving disaster were in a state of despair. The overriding impression left by these publications is one of the incomparable fragility of human embryonic development, a process that we usually take for granted. Otherwise spare scientific prose is punctuated with photographs of infants displaying thalidomide-induced deformities. Sometimes these newborn babies are screaming into the camera lens, their contracted limbs bent into seemingly impossible angles. Or else they are looking away, eyes fixed on a doctor perhaps, their gaze trusting. Occasionally, faces of children are blanked out, rightly to preserve privacy, but in a way that erases the human reality—and cost—of thalidomide’s legacy.
Stephens and Brynner justly credit an often forgotten German pediatrician, Dr. Widukind Lenz, as being the first to tie thalidomide publicly to human malformations at a conference in November 1961. Lenz mixed a passion for astute observation and inquiry with a desire to hold accountable those ultimately responsible for harming so many children. When Distillers urged that “the problem [of thalidomide] can be solved without emotion and alarm,” Lenz shot back, “I can hardly imagine that any person will be able to face the facts without emotion and alarm.”
When a criminal trial began in 1968 against executives of Chemie Grünenthal, Lenz was a key prosecution witness. But the German courts deemed his testimony unduly biased in favor of the thalidomide children, and his evidence was held to be inadmissible. Lenz’s response was quietly implacable: “I decided to take it as a compliment to my moral engagement, rather than an offense to my scientific honesty.”
After initial clinical reports by Lenz, McBride, and others, belated experiments investigating thalidomide’s toxicity showed that the drug could harm mouse, rabbit, chicken, and monkey (but not rat) embryos; any of these experiments, if performed earlier, might have alerted doctors to the risks of the drug in pregnant women. Indeed, it was the same G.F. Somers of Distillers, the person who had earlier praised thalidomide’s limitless safety, who later described multiple deformities in rabbits “remarkably similar to those seen in humans.”7
The debate quickly moved on from whether or not thalidomide was a teratogen—an agent that causes malformation of the fetus—to what should be done for the children and families affected. Terrible parental distress, stigma, and the child’s personal and medical needs all became pressing concerns. In a rush to make amends for this iatrogenic epidemic of disability, surgeons took knives to limbs with often damaging effect. Prostheses were forced on children when they had little use for them. And providing children with ridiculously complex (and wholly unworkable) power-assisted arms simply reinforced their feelings of physical inadequacy. Worst of all, many hospital staff members, themselves unable to confront such extreme deformities, provided little comfort. One mother wrote:
The staff at the hospital had not met such a situation as ours and, I think, were at a loss to know what to do. They did not tell me; and, although I didn’t know quite what was wrong, I was sure that I had done something horrible. Had I been told at once that, although my baby was unusual, he was still just a baby I would have accepted it more easily….8
Thalidomide soon began its slow return to respectability. Only three months after McBride’s letter was published, Gerard Rogerson, a doctor from Shropshire, England, wondered
whether this drug, which seems to have such remarkable inhibitory powers on growing tissues in certain cases, is being investi-gated for possible anti-carcinogenic properties.9
By 1965, preliminary studies with thalidomide had revealed its potent effects on ENL. The thalidomide story then divided into two separate histories, one concerning the drug’s obscure mechanism of action, the other exploring its extraordinary therapeutic range. Dark Remedy unites both histories in an unusual way.
Stephens is a professor of anatomy and embryology at Idaho State University. He has devoted his career to the question of how an organism comes to be the particular shape it is, and his research has concentrated on the development of limbs. Thalidomide proved an important experimental tool for studying this delicate system. Stephens is also a Mormon bishop. In his book Evolution and Mormonism,10 he argues, together with two Mormon biologist colleagues, that it is satisfying for them as scientists to believe that God created the processes by which life is formed rather than individual organisms themselves. Last year he published a detailed theory, which owed nothing explicitly to Mormonism, concerning the process by which thalidomide exerts its effects on the developing embryo.11 Brynner, by contrast, is a historian who contracted a disease—pyoderma gangrenosum—that could only be treated with thalidomide. Thanks to the FDA’s license, he was able to take the drug and made a good recovery, although he has suffered nerve damage in his fingers and toes. Both Brynner and Stephens have become captivated by thalidomide’s possibilities. And with good reason.
Following Rogerson’s hunch, Robert D’Amato and his colleagues at Harvard Medical School discovered that thalidomide could stop the growth of new blood vessels critical to tumor formation.12 Early work indicates that thalidomide produces encouraging results when used against some types of brain cancer; it has had variable effects on Kaposi’s sarcoma, leukemia, and kidney, prostate, colorectal, and liver cancers.
Thalidomide also influences the body’s immune system by switching off a substance—tumor necrosis factor alpha—that provokes inflammation. In patients infected with HIV, this switching-off effect can reverse weight loss and wasting, relieve diarrhea, and improve the symptoms of tuberculosis. Results such as these caused those with AIDS to argue, against thalidomiders, that “the real danger is keeping a possible life-saving treatment from people who need it.”
But the most striking effects of tha-lidomide are to be found in patients with multiple myeloma, a cancer affecting bone marrow. When existing drugs fail, thalidomide can produce sustained benefits in some patients with this incurable disease.13 The latest results from the Mayo Clinic even indicate that thalidomide may no longer be a drug of last resort. Instead, it seems to be an effective first-line treatment. Since, in the US, eleven thousand people die from myeloma each year and fourteen thousand people are newly diagnosed, the impact of thalidomide, if these preliminary results are confirmed, could be substantial.
This optimism has been largely fostered by Celgene, a New Jersey–based pharmaceutical company that describes thalidomide as its “flagship product.” The firm’s belief is that thalidomide is a precursor to “new families” of agents that modulate immune responses and selectively inhibit chemicals—cytokines, such as TNF—that do much of the immune system’s work. The company’s Web site (www. celgene.com) lists over seventy ongoing clinical studies involving thalidomide. In 2000, Celgene made $62 million from sales of thalidomide, almost three quarters of its total revenue and a 158 percent increase on 1999 sales figures. Sol Barer, the company’s president, said in December of last year that “we will continue to expand our clinical trial programs” with the drug. He plans to submit “a supplemental new drug application to the FDA in 2001” to take thalidomide’s license well beyond ENL. Jerome Zeldis, the chief medical officer of Celgene, promised that he would “proceed aggressively” with future clinical trials. In January of this year, Celgene became listed on the Nasdaq biotechnology index. The prospects for thalidomide have never seemed brighter.
But what of thalidomide’s toxicity? Few other substances have such a clearly documented record of adverse effects. The most recent reports about thalidomide’s effectiveness all repeat the well-known warnings about possible complications of nerve damage, sedation, dizziness, confusion, depression, tremor, headache, nausea, low blood pressure, hypersensitivity, rash, severe skin reactions, paradoxically raised HIV load, constipation, muscle aches, and blood clots. For people taking thalidomide, the burden of toxicity can be high.
As the uses of thalidomide broaden, we can expect that sooner or later another child will be born with an entirely preventable congenital malformation. Stephens and Brynner ask, “Who is responsible if another baby is born with birth defects?” I would replace “if” with “when.” Perhaps our consciences are spared if the responsibility is spread among many—government regulators, Celgene executives, advocacy groups, doctors, and scientists. But is the success and stock price of an innovative pharmaceutical company sufficient justification for the FDA to extend its license? For those patients with conditions refractory to existing treatments, perhaps it is. If a patient with resistant myeloma was a member of my family, and thalidomide was the only hope for prolonging life, I think I would be the first to consider it and call on the FDA to relax its rules. But arguments such as these, although emotionally powerful, are not a good basis for designing national policies on drug regulation. If I became a parent of a thalidomide-damaged child, I might equally recall the searing reports from the early 1960s and the present-day testi-monies of thalidomiders. And then ask, why?
If thalidomide is to offer hope that is without undue risk, it can only be through its own chemical offspring—analogues that are more powerful and safer than their parent. Celgene is searching for such analogues. At this year’s annual meeting of the American Association for Cancer Research, Celgene scientists reported that thalidomide analogues did indeed have “anti-cancer properties.” In view of this promising early work, it seems to me that thalidomide’s use should continue to be tightly circumscribed.
What are the lessons of thalidomide? Frances Kelsey has emphasized the “profound effect” the tragic use of the drug has had on drug regulation worldwide.14 Almost at a stroke, the sciences of pharmacology and toxicology assumed central positions in medical practice. Animal testing became a necessary condition for appraisals of safety. Medical authorities quickly agreed on the need for carefully conducted and reported clinical trials, including a patient’s informed consent. Careful monitoring to warn of adverse events was instituted, and information about drug safety was shared internationally. It became clear that the FDA needed to be strengthened through new laws—as it was in the 1962 Kefauver-Harris amendments to the Food, Drug, and Cosmetic Act of 1938—to withstand the severe commercial pressures placed upon it.
Kelsey had an important part in this process. She was honored by the Kennedy administration with a distinguished federal civil service award. On October 7 of last year, she was inducted into the National Women’s Hall of Fame. A Canadian by birth, she has a school on Vancouver Island named after her. And most recently her name was appended to an asteroid—minor planet 6260.
The FDA today is proud of its impact on public health. Its statement of its mission places safety before efficacy for foods, drugs, devices, and cosmetics. But this rhetoric conceals a conflict at the heart of the use of pharmaceuticals in medicine. Big pharma is big business, and the politics of drugs brings constant pressure to bear on the FDA to be more favorable to the industry in its deliberations. For example, as Stephens and Brynner point out, it was only in 1996 that the US Senate proposed to make it legal for companies to promote the use of drugs for purposes different from those shown on the label. Newt Gingrich called the FDA “the leading job-killer in America.” Even Kelsey once admitted there was legitimate debate to be had about whether the FDA “impeded the flow of new and important drugs.”
But if the FDA is so conservative and safety conscious, how could the Lotronex debacle have taken place last year? Lotronex was the trade name of a drug made by Glaxo Wellcome for the treatment of women with irritable bowel syndrome, a non-life-threatening condition that affects as many as one in five adults. The drug was recommended for approval in 1999 by an advisory committee of the FDA based on data from two clinical trials.15 The committee’s decision was hailed as an important advance for patients. Glaxo Wellcome pronounced itself “extremely pleased” with the result. Lotronex was finally approved for use in February 2000.
Glossy six-page advertisements for the drug soon appeared in specialist medical journals and were lavish in their claims. “Announcing a breakthrough approach,” said one, above a picture of a magic lantern and the words “Grant your IBS patients three wishes.” The advertisement explained, “Relief from pain and discomfort, relief from urgency, re-lief from frequency…relief, once achieved, continues throughout treatment.” Doctors were reassured that this new drug had a “favorable safety profile.”
Enthusiasm for Lotronex waned as cases of severe constipation, some requiring surgery, were reported. Several patients also developed ischemic colitis, a complication in which the gut is damaged by a shutdown in blood supply. The FDA responded in August 2000, not by withdrawing Lotronex on the grounds that its earlier decision may have been hastily made on incomplete evidence, but rather by issuing a “medication guide” to set out the drug’s risks more clearly. By November, however, five people taking Lotronex had died. Glaxo Wellcome finally capitulated and took its lethal drug off the market.
When the FDA approved Lotronex, the agency knew about its adverse effects. Why did officers at the FDA not act sooner? The answer is that, even today, the threshold for drug approval in the US is too low. When evidence does point to risk, the FDA often gives the company the benefit of its doubt, with death being the only event that forces regulators to think again. The Lotronex episode might well have been predicted. In 1997, Congress weakened US drug regulation by passing the FDA Modernization Act. One of the act’s most important provisions was to reauthorize “user fees,” which are paid by companies in order to shorten the average time for the drug review process to be completed. A system of regulation that depends financially on the very entities it is supposed to regulate is ripe for error. User fees that pay for almost seven hundred FDA employees to produce relatively quick reviews of new drugs invite regulators to take a lax attitude toward safety concerns. After all, it is industry money that pays their salaries. The FDA Modernization Act has had a damaging impact on the independence of the FDA’s safety review practices and standards. But the issue of pharmaceutical power goes far beyond drug licensing.
Two tendencies are currently injuring relations between medicine and the drug industry. First, companies frequently manipulate information in a way that goes beyond the bounds of acceptable marketing. Examples are legion.16 But the pattern is often the same. A new drug is designed and must be tested. A contract is drawn up between the sponsor company and a research team. In return for money to pay for the trial, the investigators will relinquish a degree of control over their findings to the firm. Let us say that the study shows the drug is less effective than the company hoped for. The investigators will still want to publish their results. But often the company will now stall. There are several possible outcomes.
The company’s contractual rights—mixed with the threat of litiga-tion—might persuade the investigators to keep quiet. The data could be published, but with a heavy spin in favor of the sponsor’s point of view. Or the conflict may escalate until one side or the other blinks. The strategy of suppressing information, promoting disinformation, and sowing confusion is identical to that practiced by Chemie Grünenthal when the first hints of thalidomide toxicity surfaced.
Second, the relationship between many doctors and pharmaceutical companies has now become close to corrupt. The UK’s equivalent of the FDA is called the Committee on Safety of Medicines. Its members review drug applications and decide on new licenses. Whether personally or professionally, they often depend on industry money. Each of the committee’s thirty-six members must declare personal and nonpersonal—i.e., research—interests in drug and device manufacturers. As of December 2000, twenty-seven members had declared industry interests. These included shareholdings, fees, consultancies, non- executive directorships, grants, and financial support to attend meetings. No member with a conflict of interest is allowed to take part in discussions that relate to a product he or she has an interest in. But the culture of the regulatory process is one in which it is acceptable for participants to profit personally from industry. The FDA’s rules on conflict of interest are more nuanced than those in the UK. For example, exclusions from reviewing committees are subject, in certain specified circumstances, to waivers. But the cultural norm, which is overtly pro-industry, is largely the same.
The research process itself is also immersed in a financial quagmire of conflicts of interest. A recent study at the University of California–San Francisco found that a third of faculty investigators received payments from companies for delivering lectures and accepting consultancies.17 Ownership of shares in pharmaceutical companies and personal financial ties are common. Prestigious medical conferences organized by some of the world’s most respected specialist societies—e.g., the European Society of Cardiology—are now packed with industry-sponsored symposia promoting a product, a company, or both.
There is, moreover, convincing evidence that in some cases the opinions of medical experts can be bought by the highest bidder. Doctors who take money from drug companies are more likely to sing the company line—hiding anxieties about safety—than those who keep their hands firmly in their pockets.18 Such is the atrocious venality of modern academic medicine.
The notion of pharmaceutical research as a curiosity-driven enterprise in the service of medicine has become a comforting but erroneous myth. As the cost—and risk—of drug development in a post-genome era soars, and competition between companies intensifies in ever-narrower markets, public health inevitably emerges as the first casualty. How many remember thalidomide now?
May 17, 2001
The documents from which I have extracted quotations can be found on the FDA’s Web site, www.fda.gov. ↩
G.F. Somers, “Pharmacological Properties of Thalidomide, a New Sedative Hypnotic Drug,” British Journal of Pharmacology, Vol. 15 (1960), pp. 111–116. ↩
Editorial, “Thalidomide Neuropathy,” British Medical Journal, September 30, 1961, pp. 876–877. ↩
Thalidomide was not withdrawn so quickly elsewhere. In Canada, for example, thalidomide remained available until March 1962, and in Japan until May 1962. ↩
W.G. McBride, “Thalidomide and Congenital Abnormalities,” The Lancet, December 16, 1961, p. 1358. As the current editor of The Lancet, I should declare a potential conflict here. McBride has alleged, and Stephens and Brynner repeat his claim, that The Lancet rejected an earlier paper, submitted in June 1961, describing the link between thalidomide and birth defects. To my knowledge, there is no record of this submission, and McBride has never produced any evidence to back his claim. If what he says is true, however, Stephens and Brynner are right to caution that “had such a paper been published in July 1961, much suffering would have been prevented.” I would like to take this opportunity to invite McBride to submit a copy of the letter of rejection to The New York Review for independent verification. If such a letter exists, it is a vitally important historical document. ↩
In 1993, McBride was found guilty of scientific fraud in a case unrelated to his work on thalidomide. He was barred from medical practice, only returning to the Australian medical registry in 1998. Stephens and Brynner call McBride’s later career “a study in hubris.” He was continuing to publish papers on thalidomide up to 1999. ↩
G.F. Somers, “Thalidomide and Congenital Abnormalities,” The Lancet, April 28, 1962, pp. 912–913. ↩
Quoted in Elizabeth Whatley, “Thalidomide-Damaged Babies,” The Lancet, July 7, 1962, p. 46. ↩
Gerard Rogerson, “Thalidomide and Congenital Abnormalities,” The Lancet, March 31, 1962, p. 691. A later editorial, entitled “Another Chance for Thalidomide?,” argued that the drug should not be “relegated to the ‘lumber room of medical history.'” (See The Lancet, November 20, 1965, p. 1059). ↩
Signature, 2001. ↩
Trent D. Stephens and Bradley J. Fillmore, “Thalidomide Embryopathy—Proposed Mechanism of Action,” Teratology, Vol. 61 (2000), pp. 189–195. ↩
Robert J. D’Amato et al., “Thalidomide Is an Inhibitor of Angiogenesis,” Proceedings of the National Academy of Sciences USA, April 1994, pp. 4082– 4085. ↩
Seema Singhal et al., “Antitumor Activity of Thalidomide in Refractory Multiple Myeloma,” New England Journal of Medicine, November 18, 1999, pp. 1565–1571. ↩
Frances O. Kelsey, “Thalidomide Update: Regulatory Aspects,” Teratology, Vol. 38 (1998), pp. 221–226. ↩
One of these trials, published in The Lancet, was later criticized by Ralph Nader’s Public Citizen Health Research Group as exaggerated, misleading, and in danger of causing overprescription. ↩
See Drummond Rennie, “Thyroid Storm,” JAMA, April 16, 1997, pp. 1238–1243; and Catherine DeAngelis, “Conflict of Interest and the Public Trust,” JAMA, November 1, 2000, pp. 2237–2238.May 17, 2001 ↩
Elizabeth A. Boyd and Lisa A. Bero, “Assessing Faculty Financial Relationships with Industry,” JAMA, November 1, 2000, pp. 2209–2214. ↩
Henry Thomas Stelfox et al., “Conflict of Interest in the Debate over Calcium-Channel Antagonists,” New England Journal of Medicine, January 8, 1998, pp. 101–106. ↩