When Politics Corrupts ScienceBy Yonoson Rosenblum | SEPTEMBER 2, 2020
How politics polluted our search for a COVID cure
Lawrence Krauss writes in the Wall Street Journal ("The Ideological Corruption of Science," July 12) how as a young physics professor at Yale, he and his colleagues in the hard sciences looked with bemusement at the dominant deconstructionism of the comparative literature department, which denied the existence of objective truth itself. That could never happen in the sciences, they assured themselves, except under a totalitarian regime such as Stalin's.
That idealized view of science as a separate realm devoted to the pursuit of truth and devoid of all political bias, Krauss notes, is no longer sustainable. In June, the American Physical Society, representing 55,000 physicists, declared a one-day "strike for black lives" to eradicate "systemic racism" in science. No evidence was adduced for the latter, other than the underrepresentation of blacks in the sciences.
One of the day's activities was to organize a protest campaign that resulted in the removal of physicist Stephen Hsu as vice president for research at Michigan State University. His crimes: his own studies in computational genomics to study how human genetics might be related to cognitive ability, and research by MSU psychology professors that did not support the narrative of racial bias in police shootings.
A distinguished Canadian chemist was censured by his university provost for calling for merit-based hiring, and the editors of a journal that accepted an article by him were suspended. Meanwhile, Francis Collins, the director of NIH, declared that he will no longer attend scientific conferences where white males, like him, predominate, regardless of their professional merit.
The pure objectivity of science is further clouded by fact that scientists are also human beings, prey to normal human temptations, such as the billions of dollars at stake in the race to produce medical cures or the quest for academic advancement. In 2005, Stanford professor Dr. John Iaonnidis published a paper titled "Why Most Published Research Findings are False," analyzing how bias creeps into study designs; it quickly became the most downloaded article in the history of the Public Library of Science. And in 2014, his group argued in the Journal of the American Medical Association that 35 percent of the results of controlled clinical trials could not be replicated upon reanalysis of their raw data.
THE POLITICALLY CHARGED field of climate science has been beset by data manipulation by leading research centers. And now there is evidence that politics has crept into the search for cures to COVID-19, argues Dr. Norman Doidge in a lengthy article in Tablet, "Hydroxychloroquine: A Morality Tale."
On March 21, President Trump tweeted that a combination of hydroxychloroquine (HCQ) and azithromycin might be a real "game changer" in treating COVID-19. Since that moment, the mainstream media, led by the Washington Post and CNN, has trumpeted every piece of evidence that HCQ is ineffective against COVID-19 and/or potentially dangerous, and downplayed or ignored all evidence to the contrary, in order to establish that Trump is a dangerous idiot. The MSM has, uniquely in the history of pandemics, engaged in what Doidge terms "unwishful thinking" — fervently hoping that a drug with the potential to save tens of thousands of lives, at a low cost, and without dangerous side effects, would turn out to be a bust.
Now, I would not recommend getting one's medical information from the president's Twitter feed. And his tweet was, in any event, premature. At most, there was, at the time of his tweet, tantalizing evidence of a "proof of concept." A study from China published in Nature, a respected science journal, showed that HCQ inhibits COVID-19 in cells in test tubes. As often happens in medicine, the idea of testing HCQ arose when front-line physicians in Wuhan noticed serendipitously that none of those admitted to hospital for COVID-19 were being treated with HCQ for diseases of the connective tissues.
In May, however, Dr. Didier Raoult, the most cited microbiologist in Europe, and a researcher with long experience repurposing existing generic drugs for new diseases, published a study of 1,061 COVID-19 patients given a combination of HCQ and azithromycin, which showed that over 90 percent showed a significant decrease in viral load over the course of treatment. Around that time, a survey of 6,000 front-line physicians in 30 countries showed that a large plurality — 37 percent — chose HCQ, out of 15 possible medicines, as the best response to a diagnosis of COVID-19.
But a non-peer-reviewed Veterans Administration study, a month after President Trump's tweet, showed a much higher percentage of patients treated with HCQ died than those who were not treated with it. The CNN headline trumpeted: "No Benefits; Higher Death Risks."
That VA study, however, proved highly flawed. It had ignored a crucial confounding factor: Those patients receiving HCQ in the study were much sicker than those who did not. At that time, HCQ was only approved for use as a desperation measure for seriously ill patients. Dr. Anthony Fauci was still recommending doing nothing for patients quarantined at home.
Yet no proponent of the HCQ-azithromycin combination had ever suggested it is anything more than an early intervention remedy to reduce the viral load and thus the severity of the disease. No one ever proposed it was a wonder drug that could repair failing organ systems at a late stage of the illness.
If the VA study was flawed, sister studies published in Lancet and the New England Journal of Medicine, two of the world's leading medical journals, and whose lead author was an eminent Harvard professor, constituted, according to Lancet editor Richard Horton, "monumental fraud." The studies, which purported to be based on data obtained from 96,000 patients on six continents, showed a 30 percent higher mortality rate for patients treated with HCQ and a greater danger of adverse cardiac events. But when 100 scientists around the world wrote to Lancet seeking the underlying data, the studies' authors immediately withdrew the two articles.
By labeling the articles a "monumental fraud," Lancet's editor sought to divert attention from the no less monumental failure of the peer-review process. Anyone with even a rudimentary knowledge of medical record-keeping, argues Doidge, would have been extremely skeptical that there existed comparable data sets from around the world. (Incidentally, the week before publishing the study, Lancet called editorially for Trump's defeat in November.)
DEEPLY FLAWED, EVEN FRAUDULENT, negative studies of HCQ do not establish either its efficacy or safety. But if I tested positive for COVID-19, I would not hesitate to take the HCQ-azithromycin combination.
HCQ has been in use for 65 years and has been given to at least one billion people to treat malaria and lupus. Physicians know what dosages are safe. Yale Medical School epidemiologist Dr. Harvey Risch terms the risk from proper doses, administered over a ten-day period, negligible — 9/100,000. In addition, it is cheap — sixty cents per tablet — and can be taken at home with water.
No doubt other early intervention drugs will be developed: No single drug is appropriate for every patient, and the possibility of better drugs is always there.
On July 1, the Henry Ford Medical Center in Detroit published in the International Journal of Infectious Diseases a study of patients in which the severity of illness was fully taken into account, which showed that HCQ reduced the mortality hazard (mortality over a fixed period of time) by 66 percent. Another study from Italy at the end of July found the same 66 percent reduced mortality rate.
And yet Margaret Sullivan, the Washington Post's media critic, opened her July 31 column (nearly a month after the Henry Ford study was published), ridiculing "fringe doctors spouting dangerous falsehoods about HCQ as a COVID-19 wonder cure." She was engaged in "unwishful thinking" that Donald Trump would prove right about something.
Doidge's enumeration of the multiple errors of experts and institutions we trusted to help us solve our most pressing scientific and medical problems includes "academics who increasingly see all human activities as 'political' power games, and so in good conscience can justify inserting their own politics into academic pursuits and reporting."
What, for instance, besides "implicit bias," at a minimum, can explain a presumably intelligent Harvard medical school professor making the following self-contradictory statements in one breath: HCQ is possibly dangerous, and we must save it for patients suffering from lupus or rheumatoid arthritis. Be wary of the experts.