Long Covid Recovery Depends Sharply on Symptom Pattern, US Study Finds

A study published on Nov 2 in the Annals of Internal Medicine offers a more divided picture of long Covid recovery than the headline numbers usually suggest. Researchers followed 1,019 adults who still had probable long Covid a year after their initial infection, then tracked them for three more years.

Over that period, about half of the participants had sustained improvement in their symptoms, and roughly a quarter saw their symptoms disappear entirely. Taken alone, those figures sound moderately encouraging. The study's more important finding is that the averages hide sharp differences between patients.

The pattern of illness mattered most. Among people whose long Covid was dominated by changes in smell or taste, about 60% had improved by the three-year mark. Among those whose symptoms spread across multiple body systems, fewer than 19% had. Fewer than one in five of the multi-system group experienced sustained improvement within three years of infection.

The condition can bring fatigue, cognitive impairment and post-exertional malaise, in which symptoms worsen after even modest physical or mental activity, along with problems in numerous organ systems. There is no approved treatment. The study is part of the Recover initiative, funded by the US National Institutes of Health, and tracked symptoms including fatigue, brain fog, dizziness, palpitations, shortness of breath and changes in smell or taste. A participant counted as improved when their symptom score stayed below the threshold for likely long Covid at two successive study visits.

That scoring system drew criticism from Ziyad Al-Aly, director of the Clinical Epidemiology Centre at the VA St. Louis Health Care System, who was not involved in the research. He noted that the index gives substantial weight to loss of smell or taste and to chronic cough. Those symptoms are characteristic of Covid-19 infection, he said, but they are not central to the chronic multi-system illness he associates with long Covid. In his view the multi-system group "is a different disease altogether." He added: "This is the real long Covid that will endure, and these patients do not appear to show a robust trajectory of recovery."

The study also points to a wider difficulty: deciding what counts as recovery. Nisreen Alwan, a professor of public health at the University of Southampton, said the symptom index was developed using data from the Recover initiative itself and has not been independently validated in other populations. Long Covid symptoms can also fluctuate, with stretches of improvement followed by relapse.

Alwan and a colleague made that argument in the BMJ in June, writing that inconsistent definitions make it hard to tell lasting recovery from temporary remission. The new findings land inside that unresolved debate. Until recovery is defined consistently and the index is tested in other groups, estimates of how many people get better will depend heavily on who is counted and how.