The 18th century philosopher and economist Adam Smith observed, "People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices." That view is one justification for giving federal agencies so much money, control and discretion over commerce and other aspects of our lives. And when it comes to research on the risks of products or activities, because the government is here to protect us, so the thinking goes, the research it funds or performs is more trustworthy than that emanating from the private sector.
Consider the following scenario. A federally-funded researcher at a state university hospital finds a link between a certain pesticide and a debilitating disease whose cause(s) has been a mystery. After learning of the findings, the pesticide's manufacturer, a huge multinational company, spends millions of dollars to replicate the data. Yet the experiments performed by the company's scientists fail to find the potentially deadly link reported by the university scientist.
Whom should you trust? The mantra among activists, the media and some politicians is that studies by industry can't be trusted, while government-funded science is more disinterested and, therefore, more reliable.
Therefore, you might be inclined to go with the latter. But you'd be wrong.
All research data should be scrutinized, especially when the stakes are high. Industry researchers have potential conflicts of interest, to be sure â€“ they're not eager to find problems with their employer's products â€“ but government-funded or NGO "independent" researchers often have conflicts of interest of their own. For a number of reasons, they might even be tempted to fabricate or manipulate research results. Finding something new and potentially harmful to the public is a proven way to get more government grants and possibly to climb the academic ladder. Moreover, regardless of the source of their funding, some scientists simply have an ideological axe to grind, and fudging the data might further their political agenda.
Any of these reasons might explain the extraordinary news that broke in The Scientist in June: "A former assistant professor at the University of Medicine and Dentistry, New Jersey (UMDNJ) committed research misconduct by fabricating data, according to an investigation by the university and the Department of Health and Human Services' Office of Research Integrity (ORI). The ORI, which announced its findings on Thursday (June 28), determined that Mona Thiruchelvam falsified cell count data published in two papers in 2009 in Environmental Health Perspectives and Journal of Biological Chemistry, both of which she has agreed to retract." The retracted papers "investigate the neurological response to the combined pesticides paraquat and maneb, and suggest the pesticide atrazine also has a role in disrupting dopamine pathways," which could be related to the pathophysiology of Parkinson's disease.
The article provides additional details: "Thiruchelvam fabricated stereological cell count data in two studies on how pesticides influence neuronal mechanisms involved in Parkinson's disease (PD). The studies reported the results of 13 new experiments that supposedly counted nigrostriatal neurons in the brains of mice and rats, but an investigation spearheaded by the UMDNJ determined those counts were never taken. The nigrostriatal pathway is a major dopamine circuit in the brain, and loss of neurons in this area is one of the main features of Parkinson's disease."
An honest error? Unlikely, because according to The Scientist, Thiruchelvam not only fabricated data but provided a witness who testified falsely to the origin of some of it. Finally, "[a]fter the investigation was passed back to UMDNJ and the findings were confirmed, Thiruchelvam was notified and provided with an opportunity to respond. She never did, and in February 2010, she left the UMDNJ. The ORI sent her a Voluntary Exclusion Agreement, which excludes her from federal funding and serving on advisory committees for seven years, which she signed without comment."
The studies funded by the pesticide's maker that failed to replicate Thiruchelvam's findings were thus far more credible than the government-funded university research.
Another example of similar, dubious university research is that of Tyrone Hayes, a University of California, Berkeley, biology professor who claims to have found that "developing male frogs exhibited female characteristics after exposure to atrazineâ€¦at exposure levels deemed safe by the US Environmental Protection Agency" and that the chemical induces mammary and prostate cancer in laboratory rodents and is a potential cause of reproductive cancers in humans. However, other investigators and regulatory agencies in the United States and elsewhere have been unable to reproduce his claims and have found his experimental designs to be flawed, and the Environmental Protection Agency has complained that Hayes has failed to make all of his raw data available for their review as he is ethically required to do.
These extreme scenarios should by no means be seen as a condemnation of government-funded or academic research, but they should cause us to question the conventional wisdom that tends to favor government-funded research over that performed by industry.
All scientific research, regardless of how it is funded, should be regarded with skepticism. As Caltech molecular biologist James F. Bonner once said about the research results of a competitor that did not agree with his own, "There are a thousand ways to do an experiment wrong." Every study should be evaluated to determine whether the methodology and the implementation are sound, and whether the results fully support the conclusions.
Henry Miller, a physician and molecular biologist, is the Robert Wesson Fellow in Scientific Philosophy and Public Policy at Stanford University's Hoover Institution. He was the founding director of the FDA's Office of Biotechnology. Jeff Stier is a senior fellow at the National Center for Public Policy Research.