Looking for Long Covid: A Clash of Definition and Study Design (SARA TALPOS, 07.25.2024, UnDark)
Few experts dispute that long Covid can be debilitating, or that it warrants careful study. But in interviews with Undark, a number of experts said that it is misleading to frame long Covid as an increasing threat. The best data, they say, suggest that most people recover from the disorder and that long Covid rates will decline as people develop immunity. (A July study by the VA St. Louis team also found that rates of long Covid declined over the course of the pandemic.)
The work produced by Al-Aly and his colleagues, which relies on electronic health records of U.S. veterans, is also a key point of contention. In interviews, several experts questioned the VA St. Louis’ methods. At the request of Undark, Harvard epidemiologist Marc Lipsitch reviewed the group’s first long Covid study and raised a range of concerns. Many of them related to the handling of negative controls, a statistical technique that, when deployed properly, can help researchers detect problems in the analysis of their dataset. Some of the negative controls “are simply misused in the paper,” Lipsitch wrote in an email to Undark.Additionally, some experts suggested that the VA St. Louis studies are not truly measuring long Covid. “They’re not studying post-viral illness, in my opinion, in these VA studies,” said Anders Hviid, a professor at the University of Copenhagen and head of the Department of Epidemiology Research at the Statens Serum Institut. Post-viral syndromes, said Hviid, are relatively rare and are usually characterized by fatigue and cognitive difficulties. Al-Aly’s research, meanwhile, looks at what Hviid described as a gamut of outcomes: dementia, thromboembolisms, psychiatric diseases, kidney diseases — “everything under the sun,” he said.
At best, the studies are detecting health problems known to occur when people with poor baseline health experience a severe infection of any kind, said Hviid. At worst, the findings simply reflect bias in the study design, and are picking up on symptoms that are not caused by Covid-19 at all. “It’s a disappointment that not more U.S. scientists have spoken up about this,” said Hviid.
Heck, they still pretend fibromyalgia is a thing.