“Everybody complains about the weather, but nobody does anything about it.”[1]
Changing scientists’ aversion to sharing data may be even more difficult than changing the weather, but the means to accomplish the change would be relatively easy to implement. In a recent issue of Nature, one scientist confesses that outright fraud is all too common, and cries out for sharing quantitative, as well as audio and video data.
Many of the leading scientific journals have been cacophonously banging their drums about corporate conflicts of interest for some time, to the detriment of focusing their readers on the merits of scientific disputes. The focus on industry has been so one-sided that when a leading journal acknowledges the
biases or conflicts of scientists who self-style themselves “environmentalists,” the event becomes newsworthy.
Last week, in an editorial in Nature, an environmental scientist rendered the drum banging a little bit more harmonious by identifying a prevalent motive for glibly ignoring the prevalence of fraud in academic science, as well as refusing to share underlying data and information from research.
In his Nature article, Timothy Clark notes that scientists sometimes act badly or tolerate fraud and bad science out of “green” motives.[2] Clark cites a survey that reports that one out of seven scientists has witnessed fraud, which suggests that outright fraud is widely prevalent. Of course, everyone has seen really bad science done and published, much more commonly than outright fraud. Finding fraud and misrepresentations so common, Clark urges the adoption of strict requirements for data sharing and production as the only therapy of the widespread fabrication and massaging of data.
Unlike many of the narratives advanced for data sharing, which focus on perceived or real abuses by corporate-sponsored science, Clark acknowledges that the pressure to advance environmental causes and policy, leads to the need for “simple stories” about damage to wildlife. Scientists in this field argue that the stories told are more important than the validity or correctness of the scientific investigations. Mon Dieu!
Clark invokes the nomenclature of “White Hat Bias” for the phenomenon he has observed in the environmental sciences. Of course, this bias is not limited to environmental science; it has crept into all policy laden areas of science and medicine, and probably shows its greatest stranglehold in occupational and environmental epidemiology. If you have wandered into courtroom involving an issue of occupational or environmental disease, you may have seen smug arrogance topped with a heavy dose of white hat bias.
Clark argues that scientists get away with questionable methods and practices because the publication of scientific findings is imbued with trust, and the burden of proof is on those who challenge the reported findings. Because the authors of studies have rarely shared their protocols, statistical analysis plans, and their underlying data, challengers to their findings often cannot carry a burden of proving that something untoward has occurred. Clark suggests that only by shifting the burden of proof onto the authors themselves, and by requiring routine sharing and documentation of data and findings, will the scientific community heal itself from self-inflicted wounds of fraud and misconduct.
Clark’s focus on “White Hat Bias”[3] is unusual for its candor, in the world of environmental science. He identifies the phrase as having been coined by two industry-funded scientists, in their 2010 papers on obesity research.[4] The bias, of course, has been with us forever. Although I have not found the exact phrase used before 2010, the phenomenon was described aptly in the pages of the Journal of the National Cancer Institute, 25 years ago, in 1992, when a science reporter described a paradigmatic occurrence of White Hat Bias in the administration of President Jimmy Carter. In 1978, Joseph Califano, then Secretary of the Department of Health, Education, and Welfare, began a massive effort to exaggerate the health effects of asbestos. In an April 1978 press conference, Califano predicted that 17% of all future cancers would be caused by asbestos. Califano, a non-scientist, supposedly relied upon an unpublished report by scientists from the National Cancer Institute, the National Institute of Environmental Health Sciences, and the National Institute of Occupational Safety and Health. These scientists in turn relied uncritically upon thse work of Irving Selikoff and his colleagues at Mt. Sinai School of Medicine.
Not all scientists were snookered by the Selikoff and his allies in government. Sir Richard Doll and Sir Richard Peto wrote a monograph designed primarily to debunk the Selikoff mythology.[5] For their efforts and insights, Doll and Peto were rewarded with the unremitting enmity of the occupational medicine establishment in the United States. And the snookering continues in asbestos personal injury cases and many other types of litigation, in courtrooms all around the nation.
Although it is fortunate that these dire asbestos claims and predictions never came to pass, it is sad that public officials, scientists, and judges were so credulous and uncritical. In 1992, one of the authors of the unpublished report, Marvin Schneiderman, Ph.D., confessed that he and his colleagues had used Selikoff’s estimates without questioning them.[6] Another scientist, Philip Enterline was honest enough to explain the motivation behind the baseless exaggerations of asbestos hazards. Enterline acknowledged that the Zeitgeist of the 1970s, fueled by sensationalism over allegations of industry misconduct, fostered muddled thinking and outright fabrication:
“It was sort of the ‘in’ thing to exaggerate . . . [because] that would be good for the environmental movement. … At the time it looked like you were wearing a white hat if you made these wild estimates.”[7]
Scientists who are more interested in advancing “simple stories,” than in knowing the validity or correctness of the scientific investigations involved are commonplace in litigation and regulatory proceedings. In the world of science, Clark’s reforms would help rob the exaggerations and the fraud of some of their cache. In the courtroom, judges would do well to have expert witnesses take off their hats, even their white hats, and disclose all their data.
[1] Charles Dudley Warner, 6 The Book Buyer: A Monthly Review of American & Foreign Literature 56, 57 (1889); entry for “Charles Dudley Warner,” in Wikipedia.
[2] Timothy D. Clark, “Science, lies and video-taped experiments,” 542 Nature 139 (2017).
[3] White Hat Bias,” in Wikipedia, available at <https://en.wikipedia.org/wiki/White_hat_bias>.
[4] Mark B. Cope and David B. Allison, “White hat bias: examples of its presence in obesity research and a call for renewed commitment to faithfulness in research reporting,” 34 Internat’l J. Obesity 84 (2010); Mark B. Cope and David B. Allison, “White Hat Bias: A Threat to the Integrity of Scientific Reporting,” 99 Acta Paediatrica 1615 (2010). See also Trevor Butterworth, “The Wrongs Of Righteous Research,” Forbes (Dec 3, 2010); John A. Dawson, “Lessons from a Silly Yet Serious Study,” 2 Obesity & Eating Disorders (2016).
[5] Richard Doll & Julian Peto, “The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today,” 66 J. Nat’l Cancer Instit. 1191 (1981); Richard Doll & Julian Peto, Asbestos: Effects on health of exposure to asbestos (1985).
[6] Tom Reynolds, “Asbestos-Linked Cancer Rates Up Less Than Predicted,” 84 J. Nat’l Cancer Instit. 560, 561 (1992).
[7] Id. at 562 (quoting Philip Enterline).
