Health outcomes up to 3 years and post-exertional malaise in patients after hospitalization for COVID-19: a multicentre prospective cohort study (CO-FLOW)

Abstract

Background: Many patients experience long-lasting health problems after COVID-19. The study aimed to assess 3-year trajectories of a comprehensive set of patient-reported outcome measures (PROMs) in patients hospitalized for COVID-19, particularly focusing on the 2- to 3-year trajectory. Additionally, we evaluated prevalence of post-exertional malaise (PEM) at 3 years, its risk factors, co-occurring health problems, and the 3-year trajectories of patients with and without PEM.

Methods: The CO-FLOW multicentre prospective cohort study followed up adults hospitalized for COVID-19 in 7 hospitals, located in the Netherlands. Study assessments were performed at 3, 6, 12, 24, and 36 months post-discharge, conducted between July 1, 2020, and May 22, 2024. PROMs on recovery, symptoms, fatigue, mental health, cognition, participation, sleep quality, work status, health-related quality of life (HRQoL), and PEM were collected. Generalized estimating equations were used to assess health trajectories and multivariable logistic regression to identify risk factors for PEM.

Findings: In total, 299/344 (87%) patients completed the 3-year follow-up and were included in the analysis. Complete recovery rates increased (p < 0.001), from 12% at 3 months to 24% at 3 years. Symptoms of impaired fitness, fatigue, and muscle weakness (all p < 0.0019) and PROMs for fatigue score, participation, return to work, and HRQoL (all p < 0.005) improved significantly over time, while PROMs for cognitive failures worsened (p < 0.001). Between the 2- and 3-year visits, memory problems (OR 1.4 [1.1-1.7], p < 0.001), and scores of fatigue (MD +1.0 [0.4-1.6], p = 0.002), cognitive failures (MD +2.2 [0.9-3.4], p < 0.001), and SF-36 mental component summary (-2.2 [-3.1 to -1.3], p < 0.001) significantly worsened. At 3 years, 66% of patients experienced fatigue, 63% impaired fitness, 59% memory problems, and 53% concentration problems. PROMs showed that 62% reported poor sleep quality, 55% fatigue, and 28% cognitive failures. PEM was reported by 105/292 (36%) patients at 3 years; risk factors were female sex (OR 3.4 [95% CI 1.9-6.0], p < 0.001), pre-existing pulmonary disease (3.0 [1.7-5.6], p < 0.001), physical inactivity pre-COVID-19 (2.3 [1.2-4.1], p = 0.008), and ICU treatment for COVID-19 (1.8 [1.02-3.0], p = 0.04). Concurrent fatigue, cognitive failures, and dyspnea were more common in patients with (42%) than without (6%) PEM. Patients with PEM showed poor health outcomes throughout the entire follow-up period, including worsening fatigue and HRQoL during the third year.

Interpretation: Many health problems persisted up to 3 years post-discharge, with self-reported fatigue and cognitive problems worsening in the third year. PEM was common, and linked to a more severe phenotype of long COVID. These findings highlight the urgent need to optimize treatment options and investigate underlying pathological mechanisms of COVID-19.

Funding: The Netherlands Organisation for Health Research and Development (ZonMw); Rijndam Rehabilitation; Laurens.

Keywords: COVID-19; Long COVID; Long-term health outcomes; Post-exertional malaise.

Fig. 1

Flowchart of participants included in analysis.

Fig. 2

Trajectories of the ten most prevalent symptoms in patients with COVID-19 up to 3 years after hospital discharge. P values are obtained from Generalized Estimating Equations analysis, and are presented for changes over the overall follow-up period at the top and specifically from 2 to 3 years follow-up above the columns. A p value less than 0.0019 was considered statistically significant and is indicated in bold.

Fig. 3

Trajectories of PROMs in patients up to 3 years after hospitalization for COVID-19. Data are presented as estimated means with standard errors obtained from Generalized Estimating Equations analysis. The trajectories from 3 to 36 months post-discharge are presented for health outcomes of fatigue (a), cognitive failures (b), sleep quality (c), anxiety [denoted by A] and depression [D] (d), and participation (e). In panel e, USER-P includes the subscales restriction [R], satisfaction [S], and frequency [F]. In the panels a–e, p values are presented for the overall time effect from 3 to 36 months (top left corner) and specifically for the trajectory between the 2- and 3-year visits (above trajectory line). A p value less than 0.00417 was considered statistically significant and is indicated in bold. For HRQoL, data are presented for each SF-36 domain in a spider plot (f), ∗ indicates a p value < 0.00417 for the trajectory between the 2- and 3-year visits. FAS, Fatigue Assessment Scale; CFQ, Cognitive Failures Questionnaire; PSQI, Pittsburgh Sleep Quality Index; HADS, Hospital Anxiety and Depression Scale; USER-P, Utrecht Scale for Evaluation of Rehabilitation-Participation; SF-36, 36-item Short Form Health Survey with the domains: GH, General Health; PF, Physical Functioning; RP, Physical Role Impairment; RE, Emotional Role Impairment; VT, Vitality; MH, Mental Health; SF, Social Functioning; BP, Bodily Pain.

Fig. 4

Co-occurring health problems and health outcome trajectories in patients with COVID-19 with and without PEM at 3 years. Co-occurring health problems in patients with PEM (a) at 3 years after hospital discharge. Group comparisons were performed for the 3-year trajectories of fatigue (b), cognitive failures (c), and HRQoL physical component (d) and mental component summary (e) scores, adjusted for sex, pre-COVID-19 employment, pre-COVID-19 education level, pre-COVID-19 physical activity level, pre-existing comorbidities obesity, pulmonary disease, and cardiovascular disease, and intensive care unit admission during hospitalization. P values are presented for within group differences between the 2- and 3-year visit, obtained from Generalized Estimating Equations analysis. PEM, Post-Exertional Malaise; FAS, Fatigue Assessment Scale; CFQ, Cognitive Failures Questionnaire; HRQoL, Health-Related Quality of Life; SF-36, 36-item Short Form Health Survey; PCS, Physical Component Summary; MCS, Mental Component Summary.