People with long COVID show natural recovery over time

2 giờ trước
Stephen Padilla
Stephen PadillaSenior Editor; MIMS
Stephen Padilla
Stephen Padilla Senior Editor; MIMS
People with long COVID show natural recovery over time

A recent study involving people with likely long COVID 12 months after initial infection with SARS-CoV-2 has observed meaningful symptom reductions in half of these individuals by 36 months, with one in four experiencing complete remission.

“These findings underscore both the potential for natural recovery and the urgent need for placebo-controlled therapeutic intervention trials,” said the authors, led by Dr Harrison Reeder, Massachusetts General Hospital and Harvard Medical School, Harvard University, Boston, Massachusetts, US.

Reeder and his team conducted this prospective, multisite cohort study in 83 US-based sites across 33 states, Washington, DC, and Puerto Rico. They included 1,019 persons with suspected long COVID 12 months after first SARS-CoV-2 infection and reported unadjusted estimates of cumulative incidence and adjusted marginal cumulative incidence across all follow-up time points.

Of the participants, 36.2 percent met the primary outcome of time until two consecutive study visits with Long COVID Research Index (LRCI) <11 by 24 months following infection, which increased to 50.1 percent by 36 months. [Ann Intern Med 2026;doi:10.7326/ANNALS-25-05275]

Furthermore, 56.4 percent met the secondary outcome of time until LCRI measure decreased by ≥50 percent and improved quality of life by 36 months, while 26.9 percent had LCRI reduction to 0.

The relative timing of subsequent symptom improvement was comparable among individuals who still reported long COVID at 18 and 24 months after their initial infection.

“These data serve as important preliminary findings to inform the design of interventional trials aimed at long COVID improvement or resolution, showing that a sizeable proportion of persons with prevalent likely long COVID exhibit some symptom reduction over time,” Reeder and colleagues said.

Collider bias

This study centred on explaining directly interpretable cumulative probabilities of improvement across key demographic and illness-related characteristics in persons living with long COVID, according to Reeder and his team.

Rather than generalizing to the broad population of people with a history of SARS-CoV-2 infection, “the analysis conditions by design on prevalent long COVID status and therefore corresponds to the population of persons with a history of infection who are symptomatic with likely long COVID at 12 months after initial infection,” Reeder and colleagues said.

“As a result, it is not intended to disentangle causal relationships between individual characteristics and timing of symptom improvement because a cohort conditioned on prevalent long COVID may exhibit noncausal associations between health characteristics measured before infection and timing of improvement—that is, collider bias,” they added.

“For example, if a preinfection health characteristic reduces risk for long COVID, then those who nevertheless develop long COVID may be those with disproportionately poorer health,” Reeder and colleagues said.

This situation would suggest an association between such health characteristics and longer time to symptom improvement. [BMJ 2023;381:1135]

“[G]iven the remaining large number of adults still living with long COVID, our study suggests a need for therapeutic interventional trials to improve recovery outcomes, designed with concurrent control groups to distinguish the impact of treatment from natural symptom resolution,” Reeder and his team said.

Long COVID is an infection-related chronic condition occurring after SARS-CoV-2 infection. It persists for at least 3 months as a continuous relapsing and remitting or progressive disease state that affects one or more organ systems. [https://nap.nationalacademies.org/catalog/27768/a-long-covid-definition-a-chronic-systemic-disease-state-with]