We have a dispiriting shortage of high-quality health research for many reasons, including the fact that it’s expensive, difficult and time-intensive. But one reason is more insidious: Sometimes groups seek to intimidate and threaten scientists, scaring them off promising work.
By the time I wrote about the health effects of lead almost two years ago, few were questioning the science on this issue. But that has not always been the case. In the 1980s, various interests tried to suppress the work of Dr. Herbert Needleman and his colleagues on the effects of lead exposure. Not happy with Needleman’s findings, the lead industry got both the federal Office for Scientific Integrity and the University of Pittsburgh to conduct intrusive investigations into his work and character. He was eventually vindicated — and his discoveries would go on to improve the lives of children all over the country — but it was a terrible experience for him.
I often complain about a lack of solid evidence on guns’ relationship to public health. There’s a reason for that deficiency. In the 1990s, when health services researchers produced work on the dangers posed by firearms, those who disagreed with the results tried to have the National Center for Injury Prevention and Control shut down. They failed, but getting such work funded became nearly impossible after that.
I have also discussed the too-slowly changing approach to back pain. There’s a reason for that, too. When research was published, also in the early 1990s, arguing that the proper treatment of back pain was nonsurgical, some with a financial interest in surgical intervention tried to have the Agency for Health Care Policy and Research defunded. They failed, too, but left researchers skittish about focusing on this topic.
The area I complain about most, though, concerns nutrition, including supplements. That domain allows us to focus on another type of intimidation: lawsuits.
In 2013, scientists at the Food and Drug Administration published a study in The Journal of Pharmaceutical and Biomedical Analysis showing that nine brands of dietary supplements sold in the United States contained a synthetic analogue of amphetamine. The authors noted that the efficacy and safety of this stimulant, beta-methylphenylethylamine, or BMPEA, had never been studied in humans.
A year later, Canadian health authorities recalled supplements containing the stimulant, noting the potential for “serious cardiovascular complications.” The FDA, inexplicably, remained silent. The agency did not warn the public, recall products or warn manufacturers.
Pieter Cohen, an associate professor of Medicine at Harvard Medical School, replicated aspects of the 2013 study and came to the same conclusion as the FDA experts: The stimulant was available in multiple brands of supplements, and a comprehensive review of the biomedical and chemistry literature found not a single scientific study of the stimulant’s efficacy or safety in humans. These results were published in Drug Testing and Analysis in 2015 and widely disseminated by national and international media outlets. Two weeks after that, the FDA alerted consumers that the stimulant was potentially dangerous and warned manufacturers to remove it from their products.
One of the companies that received an FDA warning letter, in turn, sued Cohen for $200 million in damages for libel, alleging that statements in the peer-reviewed article, and subsequent interviews with the media, were false. The company asserted, without supporting scientific evidence, that while the article said the stimulant was not “natural,” it had extracted it from a Mexican shrub. Company officials also said they had evidence of the stimulant’s efficacy and safety in humans. The lawsuit, initially filed in Georgia, was dismissed because of lack of jurisdiction there, then refiled in federal court in Massachusetts.
During the lawsuit’s discovery phase, the supplement company demanded and received access to emails related to the study, including those with co-authors, journal editors, the FDA, outside experts and the news media. The company also demanded and received all revisions of the manuscript, as well as peer reviewers’ comments and the authors’ responses. Despite the absence of evidence of wrongdoing, the judge allowed the case to go to trial.
Cohen got entangled in what legal scholars call a strategic lawsuit against public participation, or SLAPP. Anti-SLAPP laws are intended to prevent people from using courts, and even the threat of a lawsuit, to intimidate people who are exercising their First Amendment rights. But in Cohen’s case, the court refused to give full weight to Massachusetts’ anti-SLAPP statute on the ground that dismissing the case would undermine the supplement company’s constitutional right to a jury trial.
Although the jury eventually found for the defense, the experience was extremely unsettling. “Preparation for the trial included a six-hour deposition, a mock trial and a review of more than 4,000 pages of studies, emails, correspondences, drafts and depositions,” Cohen told me. “The trial itself lasted seven days, and put my family through the wringer.” He was fortunate to have the full support of his university in defending his work.
Cohen and I, along with Nicholas Bagley, a law professor at the University of Michigan, recently wrote in JAMA Internal Medicine about the damage such suits inflict on scientific inquiry. We pointed out that the peer-review process provides a way to question a study’s conclusions before publication, and that less formal peer review continues afterward in the form of letters to the editor and editorials.
If errors or mistakes are believed to be fraud, mechanisms for review exist in university systems. Only if evidence of fraud surfaces does it make sense for courts to be brought into play.
“Courts aren’t equipped to referee scientific disputes,” Bagley said. “And they have an obligation to prevent unscrupulous plaintiffs from abusing the machinery of justice to stifle science.”
Lawsuits like these are too common in health research. Bagley did a fairly comprehensive search of the reported opinions over the past 40 years. He found two cases in the 1980s and two in the 1990s. Since 2000, however, there have been 10. These numbers greatly understate the number of filed cases, however, since the vast majority are settled.