The Kern Controversy

These questions lie at the heart of a controversy that has pitted David Kern, MD, and his supporters against three institutions: Brown University Medical School in Providence, Rhode Island, where Kern is a faculty member; Memorial Hospital of Rhode Island (an affiliate of Brown University) in Pawtucket. Rhode Island, where Kern's occupational medicine unit is based; and Microfibres, Inc., a Pawtucket textile company that hired Kern as a consultant to probe the possible causes of a mysterious lung disease that was affecting workers. This controversy is destined to become a landmark in the emerging literature about secrecy in science. But, as with all recent cases in which suppression of scientific information is alleged, the facts of the situation and their interpretation are being disputed.

In 1990, a man working at a Microfibres, Inc. plant in Kingston, Ontario, known there as Hartford Fibres, presented to his doctor with breathlessness and cough. The manufacturing process at all Microfibres plants involves gluing finely cut nylon filament, known as flock, onto cotton and polyester fabric to produce a velour-type material. This material is used, among other things, as a seat covering for furniture and automobiles. By 1991, several workers at the Ontario plant had developed the same condition, a rare form of interstitial lung disease now known as flock worker's lung (MMWR Morb Mortal Wkly Rep. 1997; 46:897-901). The disease caused hypoxemic respiratory failure and was quite severe. One man required mechanical ventilation for several weeks and came close to dying. Two were left dependent on supplemental oxygen and forced to curtail their daily activities. None of those affected were able to return to work. An investigation by the Ontario Ministry of Labor resulted in interventions to prevent the disease.

In 1994, a worker at the Microfibres plant in Rhode Island developed shortness of breath that worsened over several months. He was referred to Kern, an occupational physician at nearby Memorial Hospital, who followed up on the referral by visiting the plant. In an interview, Kern explained, "At the door, we were greeted by the personnel director, who asked us to sign a trade secret confidentiality agreement." Kern signed the agreement and spent about an hour at the plant but found nothing to explain the worker's symptoms.

However, 1 year later, when a second worker at the plant was referred with similar symptoms, Kern began to suspect an occupational illness. He contacted the company and suggested that because two cases of a rare disease had occurred at the same plant, it was likely that something at the plant was causing the illness. Microfibres requested an investigation by the National Institute on Occupational Safety and Health (NIOSH) and asked Kern to investigate the potential causes of the illness.

Confidentiality Issues

In late 1996, while these investigations were taking place, Kern prepared an abstract for submission to the 1997 meeting of the American Thoracic Society. "Perhaps this matter had already been addressed somewhere else in the world, and there was an answer and a solution," he said, when asked why he wanted to present the information. "We might be helped then, by fellow scientists and physicians. Secondly, we wanted to warn other physicians and scientists to be on the lookout for disease of this type in this type of a work force."

But the company asked that the abstract not be submitted. John Woodhouse, a spokesperson for Microfibres, explained in an interview that the draft of the abstract contained a list of the chemicals used in the company's manufacturing process, which the company considered proprietary information. Publication of the abstract would violate the confidentiality agreement that Kern had signed on his first visit to the plant in 1994. In addition, Woodhouse noted that the company believed the abstract reached premature conclusions.

For his part, Kern was shocked that the agreement he had signed in 1994 was being applied to an investigation he had not begun until 1995. Confidentiality agreements of the type signed by Kern are routine, he asserted. "Such agreements cover trade secrets and confidential production details and have never been construed to limit the reporting of occupational disease" (Med Decis Making. 1998; 18:239).

Kern's understanding of the scope of jurisdiction conferred by signing a confidentiality agreement is shared by other occupational physicians, according to Katherine Kirkland, MPH, Executive Director of the Association of Occupational and Environmental Clinics in Washington, D.C. Kirkland said that physicians at occupational clinics nationwide told her that they have signed such agreements when making site visits to companies without first getting legal approval from their universities. Legal experts told Kern that the signed agreement was irrelevant to his attempt to submit an abstract to a scientific meeting.

Was the company actually concerned about a leak of trade secrets? Or was it attempting to suppress data that, although vital from a public health perspective, could damage the company's image? Microfibres argues that its request for the involvement of NIOSH ensured that any health information that emerged from the investigation would eventually be released to the public. But Woodhouse acknowledged that managers were worried that their competition might learn about the workers' lung disease. Microfibres acquired its first plant in Europe in 1994, which brought it into direct competition with European companies that manufacture the same velour-type material by using natural fibers. Microfibres managers believed that if the competing manufacturers learned of the problems in Rhode Island, they might try to strengthen their marketing position by arguing that synthetic velour material made from flock would cause lung disease in anyone who used it.

Attempts at Negotiation

Kern wanted the investigation to proceed, so he offered to forgo submission of the abstract if the company would agree to certain terms of the investigation. He specifically requested a policy of open communication between himself and NIOSH, but the company rejected the proposal. He also made changes to the abstract that would make it difficult for readers to identify the company. The company's name was not mentioned, the plant was described as "a synthetic textile manufacturing plant" (the word flock did not appear), and the plant in Canada was identified only as "a similar plant"-its location was not disclosed. But when the revised abstract did not alter the company's stance, Kern contacted the administrations at Memorial Hospital and at the medical school and informed them of the situation.

The hospital's counsel identified two legal problems. First, Kern's signature on the 1994 confidentiality agreement might have placed the hospital at risk for breach of contract action if he presented his abstract. Second, no formal consulting agreement between Kern and the company existed. According to Kern, Microfibres had rebuffed efforts by hospital staff to procure such a contract, despite having paid the hospital over $100 000 for his consultations. Lawyers for the hospital concluded that any information generated by Kern belonged to the company and advised Kern to withdraw the abstract.

In an interview, H. Denman Scott, MD, physician-in-chief at Memorial Hospital, explained, "The company was worried ... We know they went to a lot of lawyers. If the information could be used by their competitors to discredit the product and if they'd actually seen their sales decline as a result, they were all set to come after the hospital. Our hospital lawyers thought it was a risk, and we were relying on them."

At Brown University, Peter Shank, PhD, Associate Dean for Research, reviewed the confidentiality agreement that Kern had signed at the time of the plant visit in 1994 and declared in a letter to Kern, "I see no way in which you can publish results of your studies at the company without their written approval." Shank noted in an interview that the administration at Brown did not have a unified stance because another dean believed that Kern did have the right to publish the abstract.

Repercussions of Presentation

Kern did not withdraw the abstract. Instead, he terminated his consulting relationship with the company. (The company hired another physician to continue the investigation.) In the spring of 1997, Kern presented his findings at the American Thoracic Society meeting in San Francisco. One week later, he received letters from the hospital and from the university informing him that his 5-year employment contract would not be renewed after its expiration in 1999.

Hospital administrative staff deny that the dismissal was in retaliation for Kern's publication of the abstract. They claim that the occupational medicine program, in which he was the sole physician, had been running at a loss and that plans to terminate the program had been made before the controversy arose. Scott said that Kern was one of two members of the Department of General Internal Medicine whose contracts would not be renewed. But Scott admitted that the controversy over the abstract affected Kern's employment opportunities. "In the wake of all this, there were many interpersonal factors and issues to take into account," Scott said. "Could he [Kern] operate successfully in an environment that he thought was corrupt?"

Donald Marsh, MD, dean of the medical school, said that because Kern was employed by Memorial Hospital and not by the university, the university could not continue his appointment if the hospital discontinued his contract. Marsh claimed that efforts to find Kern a post at another affiliated hospital were unsuccessful.

Kern says his program had not been running at a loss, and that, in any case, his own salary as a clinician-teacher was paid by the Department of Medicine. He believes that the hospital did not want to bite the hand that feeds it. He claims that the family that owns Microfibres has contributed substantially to the hospital and that members of that family serve as members of the hospital corporation. Spokespersons for the company and for the hospital termed these charges "insulting."

The Responsibility To Report

More than 70 occupational physicians have written letters protesting Brown University's treatment of Kern. They share Kern's belief that he had a moral duty as an occupational physician to communicate with the larger scientific community by presenting preliminary findings at an international meeting.

C. Andrew Brodkin, MD, MPH, of the University of Washington, Seattle, who has written on codes of ethics for occupational physicians (J Occup Environ Med. 1996; 38:869-81), offered his perspective in an interview. "One of the important principles is the obligation to report," he asserted. "Consider how we learned about asbestosis: A few professionals started observing that workers had interstitial lung disease, and eventually it led to Selikoff's study. Without the case reports, it would have taken much longer to put two and two together. The obligation is certainly to report to industry, but there is a further obligation to report to the scientific community and to regulatory agencies."

But Kern's opponents claim that his reporting responsibilities were never blocked. He acknowledges that the company did not prevent him from reporting the cluster of cases of interstitial lung disease to the Rhode Island Department of Health. He was also able to submit a summary of the investigation, which later appeared in Morbidity and Mortality Weekly Report, and the NIOSH report on Microfibres was published in April 1998. Investigators from NIOSH concluded that occupational exposure to flock-associated dust is a health hazard and recommended reduction of worker exposure to dust at the Rhode Island plant (Washko R, Burkhardt J, Piacitelli C. Health Hazards Report 96-0093, April 1998, NIOSH Publications, Cincinnati, Ohio).

Yet Microfibres, Memorial Hospital, and Brown University all encouraged Kern to withdraw the abstract he submitted to the American Thoracic Society. Marsh said that "The university cannot advise its faculty to breach contracts, because then it assumes a liability which would be inappropriate for it to assume. He [Kern] should have had the right to publish [the abstract], but he compromised it. The reality, from our standpoint, is that Kern signed a contract that compromised that right away."

As academic scientific research becomes increasingly supported by private industry, the debate about the ethical duties of physician-scientists will not be resolved soon. The debate essentially revolves around who holds the higher moral ground: scientists who share their findings in an effort to protect patients from harm or scientists who honor signed contracts by guarding the confidentiality of their sponsors. Even when secrecy is a contractual obligation, medical scientists may believe that the more pressing obligation is to prevent the potential harms to patients that could result from secrecy. And it would be up to a court to decide whether a company's claim of proprietary information was outweighed by public interest in disclosure.

Other Cases

Whichever side is perceived to be correct, Kern's case and others like it suggest that disputes over confidentiality agreements between medical researchers and industry may be relatively common. In 1995, a similar controversy arose at the University of California, San Francisco, and subsequently received substantial media coverage.

A drug company had contracted with one of the university's professors to compare one of its products, Synthroid, with three less expensive levothyroxine compounds. The company paid $250 000 for the research, which it believed would demonstrate the superiority of Synthroid. Instead, Betty Dong, PharmD, and her colleagues found the other agents to be equally effective. The research was accepted for publication in JAMA, but the company, New Jersey-based Knoll Pharmaceutical, disputed the conclusions and objected to publication of the research. University of California lawyers believed that Dong's contract gave Knoll the right to prevent publication and said they could not defend her if the company sued. Dong then withdrew the article from the journal. After the story became public, the company relented, and the article eventually appeared (JAMA. 1997; 277:1205-13). Individuals who have taken Synthroid have since brought a class action suit against the company for trying to withhold the information; that suit is ongoing.

In 1996, another dispute over the right to present industry-sponsored research occurred at the University of Toronto when Nancy Olivieri, MD, and her colleagues reported on a clinical trial of an experimental medication called deferiprone. Deferiprone has been proposed as an oral treatment for the iron overload that develops in patients with thalassemia who require repeated transfusions. The medication is being developed by the Canadian pharmaceutical company Apotex, which sponsored Olivieri's study.

The first published results of the trial looked promising (N Engl J Med. 1995; 332:918-22). But in the year after the study was published, Olivieri grew concerned over evidence suggesting that the drug could become ineffective after prolonged use (Blood. 1997; 89:739-54). She wanted to present the new findings at the American Society of Hematologists meeting in December 1996. But Apotex disagreed with her interpretation of the data and refused to give permission for the presentation. Instead, the company terminated the clinical trial at the site where Olivieri was the principal investigator and removed her from another trial of the drug in which she had served on the steering committee (Medical Post, Toronto. 21 Jan 1997; 33:1).

A confidentiality agreement that Olivieri had signed gave the company the right to block discussion of the findings at an open scientific meeting or in print. She violated that agreement by presenting her findings at the American Society of Hematologists meeting and later by writing about her concerns in Blood. The company then threatened to sue her for making untrue and damaging statements about their product (National Public Radio, Morning Edition, 17 May 1997. Report by Joe Palca). The Hospital for Sick Children (where Olivieri's study was based), which is affiliated with the University of Toronto, declined to provide legal support to defend her against the company's threats. But she continued the research, and her findings will appear this month in Blood.

Innovative Solution Required

Some experts have suggested that such controversies could be avoided if scientists and their institutions were more attentive to the contracts they sign, but that may not be possible. Kern's hospital requested a formal consulting contract; Microfibres did not offer one and instead held Kern to a confidentiality agreement signed more than a year before he began his investigations. At the University of California, San Francisco, a spokesperson said that faculty are prohibited from signing contracts that restrict their right to publish new findings, yet Dong signed such a contract. The University of Toronto also has a publications policy that prohibits faculty from signing contracts that forbid publication of results. However, Peter Munch, MD, Toronto's Assistant Vice President for Technology Transfer, notes that the university policy does not bind faculty whose contracts are signed through an affiliated hospital. Olivieri's contract was signed through the Hospital for Sick Children.

Recent data show that relationships between medical academics and industry are sometimes intentionally designed to compromise university standards. The data, from a survey of faculty at 50 universities, suggest one mechanism for this: Companies bypass university administrative scrutiny by giving academic scientists gifts instead of grants to support their research (JAMA. 1998; 279:995-9). Brown's Peter Shank describes another mechanism: Companies write misleading contracts for scientists to sign. According to Shank, "These contracts frequently contain contradictory statements, such as the scientist ‘has the right to publish’ but the companies ‘own all the data generated in the study.’" Today's academic standards are operating in an environment where the lines between public and private science have blurred. In this setting, the protection of public interest on one hand and of proprietary interests on the other require more innovative solutions than are currently offered by universities or industry.