CBT and rehabilitation show modest benefits for long COVID recovery, study finds

CBT, rehabilitation, and exercise show moderate benefits for long COVID recovery, while most other interventions lack strong evidence

In a recent living systematic review published in British Medical Journal, researchers examined 24 randomized trials, including 3695 patients to investigate the effectiveness of drug and non-drug interventions for managing long coronavirus disease 2019 (COVID-19) symptoms.

They found that cognitive behavioral therapy (CBT) and combined physical and mental health rehabilitation showed moderate evidence of improving fatigue, concentration, depression, quality of life, and overall health in long COVID patients.

About the study

The analysis included 24 randomized trials of interventions for managing long COVID, defined by the World Health Organization as symptoms persisting for at least three months after COVID-19 diagnosis. Additionally, 239 trials were identified as ongoing or completed but unpublished.

The trials predominantly took place in the Americas and Europe, with most funded by government sources or unfunded, and were typically published in 2022 or 2023. Trials were excluded if they involved <25 participants per arm, pseudorandomized designs, or acute COVID-19 treatments. Data from published and preprint trials were extracted on demographics, interventions, and key outcomes like fatigue, cognitive function, and quality of life.

A total of 3695 adult patients with long COVID were included, and drug or non-drug interventions were compared to placebos, usual care, or alternative treatments. Patients had a history of laboratory-confirmed COVID-19, with about a third hospitalized for severe disease. Vaccination status was reported in three trials, and most participants were fully vaccinated. The interventions included drugs, physical activity or rehabilitation, behavioral therapies, dietary supplements, medical devices, and combined treatments.

The review used the grading of recommendations assessment, development, and evaluation (GRADE) approach to assess evidence certainty and tailored the Cochrane Risk of Bias 2.0 tool for quality evaluation. Subgroup analyses based on factors like vaccination status and COVID-19 variants were planned but not conducted due to limited data. Patient advisors helped prioritize outcomes and interpret findings.

Results and discussion

Drug interventions (4 trials, 708 patients) involved vortioxetine, leronlimab, glucosaminyl muramyl dipeptide, and actovegin, with high-certainty evidence showing no benefit of vortioxetine for cognitive function and moderate-certainty evidence suggesting minimal impact on depressive symptoms and quality of life. Physical activity and rehabilitation (8 trials, 985 patients) showed moderate-certainty evidence that intermittent aerobic exercise improved physical function, while other interventions provided low-certainty evidence.

Behavioral interventions (3 trials, 314 patients) demonstrated that CBT likely reduced fatigue and improved concentration. Dietary supplements (4 trials, 794 patients) such as synbiotic SIM01, coenzyme Q10, vitamin C, and L-arginine were supported by low-certainty evidence, with no compelling quality-of-life improvements.

Medical devices (4 trials, 309 patients) and combined physical and mental health rehabilitation (1 trial, 585 patients) were evaluated. Moderate-certainty evidence suggests that combined rehabilitation potentially improved recovery rates and quality of life. Still, it showed limited effects on fatigue and cognitive function.

The risk of bias was a concern in many trials due to lack of blinding, selective reporting, and trial design issues. About half of the results were deemed high-risk for bias, limiting the certainty of conclusions.

The study is strengthened by involving people with lived experience, a rigorous trial search, and a focus on patient-important outcomes. However, the study is limited by the potential for missed trials, subjective interpretation of GRADE assessments, reliance on self-reported data, and insufficient evidence to explore variations in intervention effects.

Conclusion

In conclusion, the findings suggest that CBT and physical and mental health rehabilitation likely improve long covid symptoms but require active patient engagement, which may be challenging. Evidence was at high risk of bias, and future trials should compare these interventions with other active treatments.

The synbiotic formulation SIM01 showed promise but needs independent replication. Future studies should include diverse long covid phenotypes, as only one trial supported each effective intervention.