Time to recovery from long COVID: a longitudinal analysis of symptom duration and risk factors using accelerated failure time models

초록

Objectives: Long COVID comprises heterogeneous symptoms that may persist for months to years after acute SARS-CoV-2 infection. While prevalence is well described, less is known about symptom duration and determinants of recovery. Methods: We conducted a multicenter longitudinal cohort study across 12 institutions in Korea (December 2022-March 2025), enrolling adults with confirmed infection and uninfected controls. Participants were followed for up to 22 months with repeated assessments of symptoms, vaccination, and comorbidities. Eight common symptoms were analyzed. Time to resolution was estimated using Kaplan-Meier methods and log-normal accelerated failure time models, adjusting for demographic and clinical factors. Results are reported as time ratios (TRs). Results: Among 757 infected participants and 558 controls, respiratory and physical symptoms resolved similarly between groups, whereas systemic and neurological symptoms persisted longer in infected individuals. At one year, 20-25% reported fatigue and 15-20% sleep disturbance versus similar to 10% and similar to 5% in controls. Infection was associated with similar to 50% longer duration of fatigue and sleep disturbance. Individuals aged 40-59 recovered more slowly; other factors showed minimal associations. Conclusion: Persistent long COVID burden is driven by systemic neuro-related symptoms, highlighting the need for targeted, symptom-specific management strategies. (c) 2026 The Author(s). Published by Elsevier Ltd on behalf of International Society for Infectious ( http://creativecommons.org/licenses/by-nc-nd/4.0/ )

키워드