BehindLines

Long COVID, about half of followed patients improve lastingly within three years

· BehindLines newsroom

Among participants still living with long COVID a year after their initial infection, about one in two sees lasting symptom improvement within three years of infection, according to a cohort study published in the Annals of Internal Medicine. The American College of Physicians (ACP) press release went out on 2 November 2026; EurekAlert gives 3 November as the article's publication date. The study is observational: it describes how symptoms evolve, not the effect of any treatment.

Who was followed

According to the release, a team from Massachusetts General Hospital and collaborators followed 1,019 adults with probable long COVID from the RECOVER-Adult cohort, starting twelve months after their initial COVID-19 diagnosis. It is a prospective study: participants were assessed over time rather than reconstructed after the fact. Symptoms were measured with a research index commonly used for long COVID, which the release does not name. The corresponding author is Harrison T. Reeder. The article is titled "Timing of Long COVID Improvement in Prospectively Followed RECOVER-Adult Participants."

The reported figures

According to the release, sustained symptom improvement applied to 36.2% of participants at two years and 50.1% at three years. At three years, 56.4% had a substantial reduction in symptoms along with good quality of life, and about 26.9% were in complete remission. These percentages describe different criteria and should not be added together; by themselves, they do not establish whether the same people meet several criteria. The release does not spell out the thresholds that define "sustained improvement," "substantial reduction" or "complete remission"; those definitions are in the paper, which we could not access.

What the authors conclude

According to the ACP, the results show that recovery is possible for many patients. The release adds that the findings suggest a need for placebo-controlled studies evaluating treatments. This is a cautious conclusion: since the study tests no intervention, it cannot say what helps or does not help recovery.

How to read these results

Several points call for nuance. First, the 50.1% figure describes a group, not the fate of any one person: it cannot predict how a given patient will fare. Second, the other half had not reached that threshold of sustained improvement at three years, a reminder that the condition remains disabling for many people. Finally, the release mentions neither the study's limitations, nor its funding, nor the share of participants lost to follow-up, all points to check in the publication. An earlier analysis from the same RECOVER initiative (3,659 adults, with symptoms assessed from three to fifteen months after infection) reported, according to its release, that 81% of patients with long COVID at three months still had persistent or intermittent symptoms a year later; populations, durations and definitions differ, and we could not compare the two studies.

Key points

Sources