Letter from the U.K. | The New Yorker

The Lancet Editor’s Wild Ride Through the Coronavirus Pandemic | The New Yorker
▻https://www.newyorker.com/news/letter-from-the-uk/the-lancet-editors-wild-ride-through-the-coronavirus-pandemic

https://media.newyorker.com/photos/5ef77e8525aad12acaa83061/16:9/w_1280,c_limit/Knight-TheLancet.jpg

I spoke to Horton on Zoom at his home, in Muswell Hill, in North London, where he has been since March 23rd, when Johnson announced Britain’s lockdown. Because of his health, Horton has scarcely left the property. He sat at a garden table, wearing a dark T-shirt, in the shade of a deep-red umbrella. The leaves of a large bush framed an empty summer sky. I asked Horton to describe editing The Lancet during the pandemic. “We’ve been deluged with research papers and communications from all over the world,” he said. Submissions to the journal are currently running at four or five times the usual rate; Horton and the editorial team reject about ninety-five per cent of them. “My constant anxiety is, Have we let something go that could be really important?” he said. “I don’t think I’ve ever been in a situation where so much knowledge has been produced in such a short space of time.” He and the journal have struggled to cope. “I don’t think we’ve had the capacity easily to deal with it, and that has stretched all of us,” Horton said. “Inevitably, in moments like that, you get very, very anxious about mistakes.”

On May 22nd, The Lancet published a striking paper about hydroxychloroquine, the antimalarial drug touted, and taken, by President Trump, as a potential treatment for COVID-19. Unlike other studies, which had merely questioned the drug’s effectiveness, The Lancet article claimed that the use of hydroxychloroquine carried a greater risk of heart arrhythmia and death. The paper’s stark conclusions and huge sample size—it purported to use data from 96,032 patients on six continents—halted hydroxychloroquine trials around the world. But, within days, reporters and public-health experts noticed anomalies in the study’s data set, which was provided by Surgisphere, a small tech company outside Chicago. Surgisphere supplied almost real-time “cloud-based health-care data” from 4,402 COVID-19 patients in Africa, which other researchers found improbable. It overstated the number of deaths from the disease in Australia. Thirteen days after the paper was published, The Lancet retracted it. An hour later, The New England Journal of Medicine, the world’s other preëminent medical journal, also retracted a COVID-19 study that relied on Surgisphere data.

Horton described the episode as “a monumental fraud.” (On June 3rd, Sapan Desai, the chief executive of Surgisphere, told the Guardian that there was “a fundamental misunderstanding about what our system is and how it works.”) Horton said that something like this happens every few years. “In some ways, this is normal science,” he said. “Science is not immune to having bad people. There are bad people in society, and there are bad people in science. Science is very vulnerable to deceit. . . . When somebody submits a paper to The Lancet, the first thing I think is not, Do I need to consider research misconduct?” He acknowledged the political appeal of the hydroxychloroquine study, in light of Trump’s remarks. “It certainly excited our editors and peer reviewers about the possibility of answering that question,” Horton said. “And we all made a collective error, and that collective mistake was to believe what we were being told.”

Over the years, Horton’s politics have come to be expressed in studies that The Lancet has chosen to publish. He told me that he chose to make “reparations” for the Wakefield paper with a focus on child and adolescent health. Last year, Horton received the Roux Prize, an award that comes with a hundred thousand dollars, for his contributions to population health.

But there have been plenty of scrapes. In 2006, three weeks before the U.S. midterm elections, The Lancet published a paper claiming that there had been six hundred and fifty thousand excess deaths as a result of the invasion of Iraq, a much higher figure than most estimates. Horton has been a severe critic of Israel. In 2014, he printed “An Open Letter for the People of Gaza,” signed by twenty-nine Palestinian doctors and scientists, which was widely seen as simplistic and one-sided. Horton backs the environmental movement Extinction Rebellion, which stages acts of civil disobedience around the world. In February, quoting extensively from President Trump’s State of the Union address, Horton launched Lancet Migration, a project to improve the health of migrants and oppose the rise of populism, “which is fuelling racism, xenophobia, and hate.” There can be an all-encompassing quality to Horton’s activism. He questions the business of scientific publishing itself, including the all-important “impact factor,” which preserves the dominance of journals such as his own. “We aid and abet the worst behaviours,” Horton wrote in a Lancet editorial, in 2015. “Our love of ‘significance’ pollutes the literature with many a statistical fairy-tale. We reject important confirmations.”

During the pandemic, Horton has sought to merge almost entirely the scientific mission of The Lancet with a political purpose, while allowing each side to proceed by a different method. “One part of that story, we’re trying to deal with it in as objective a way as possible, and make judgments only about the science,” he told me. “But, at the same time, we’re trying to constantly assess and arrive at some preliminary conclusion or verdict about the political response. And that is obviously not objective. That is clearly political and requires a subjective and often deeply emotional response.”

#Covid #The_Lancet #Richard_Horton

Life Inside S.C.L., Cambridge Analytica’s Parent Company | The New Yorker
▻https://www.newyorker.com/news/letter-from-the-uk/life-inside-scl-cambridge-analyticas-parent-company

https://media.newyorker.com/photos/5ab5727676329f082506cdeb/16:9/w_1200,h_630,c_limit/Knight-Life-At-SCL-Parent-Company-of-Cambridge-Analytica.jpg

There were times during our conversation when the employee seemed as bemused as anybody that a company that was started in the early nineteen-nineties with some intuitive but eccentric ideas about group psychology—one of Oakes’s first ventures was selling aromas to stores, to persuade customers to buy more—was now at the center of a transatlantic conversation about voter rights, data privacy, and the integrity of the world’s most important social network. But the employee was also clear that access to big data, particularly in the form of Facebook, combined with S.C.L.’s long interest in psychological profiling and audience segmentation, had been able to equip political campaigns with digital weapons that most voters were unaware of. “You can get philosophical about this and say that Facebook being an advertising platform masquerading as a social platform is the start of the rot and the tool was always there,” the employee said. S.C.L.’s executives were the wrong people who came along at the wrong time. “There were always going to be dodgy fuckers willing to work for rich people, and the S.C.L. was just an example of the dodgy fucker.” (S.C.L. did not immediately respond to a request for comment, but the company claims that it destroyed the Facebook data, in October, 2015, and that it played no role in the Presidential election.)

The employee welcomed the current attention on S.C.L.’s methodology and behavior, whether it was illegal, or whether it should have been. The leaders of the company were not interested in these questions. “Alexander is not constrained by the sort of worries we are seeing expressed right now,” the employee said. “It really is about getting money together.” The employee continued, “What is wonderful about now is this bit of it is being opened, and I think it is bloody important, because something as catastrophic as Brexit and Trump—the technical possibility of that—is achieved through this dark shit. And this dark shit can be done by fucking cowboys. And, for lots of people who worked for the organization, it wasn’t supposed to be this way.”

Some of S.C.L.’s methods had merit. Oakes’s insight in forming the B.D.I. was to aim messages at social groups—rather than at individuals—and to place a low expectation on persuading people to change their minds. In a “classic S.C.L. project,” the employee explained, the company would use subcontractors, survey companies, and academics in the run-up to an election to create what it called a “super sample.” “We would speak to sixty thousand people, and we wouldn’t say, ‘Who are you going to vote for?’ ” the employee said. “We would say, ‘How do you feel about life?’ ” S.C.L.’s data concentrated on local concerns, such as housing, water shortages, or tribal conflict. “With all of that, we would delineate a strategy for them to win by focussing on targeted groups that we had identified within the population,” the employee said. “It is not so much, let’s make these people do this thing; it is, can we take this thing in such a way that the people who should get it do get it?”

#Facebook #Cambridge_analytica