The COVID-19 pandemic has affected hundreds of millions of people worldwide, with major consequences for health and economies. While most patients recover, evidence suggests that as many as 15% might experience long-term health effects from COVID-19, including fatigue, myalgia, and impaired cognitive function, called post-COVID condition or long COVID.
Long COVID is a complex, multisystemic condition that can impact multiple organ systems and significantly impair a person’s functional and cognitive capacity. Estimates suggest that at least 65 million people globally have experienced long COVID symptoms, with cases increasing daily. Despite the scale of this public health crisis, diagnostic and treatment options remain insufficient, and clinical trials to rigorously test potential interventions are urgently needed.
This systematic review of 24 randomized trials comprising 3,695 patients with long COVID identified moderate certainty evidence that a program of cognitive behavioral therapy (CBT) and a program of physical and mental health rehabilitation can be effective for improving symptoms. Specifically:
-
Moderate certainty evidence suggests that an online CBT program probably reduces fatigue and improves concentration compared to usual care.
-
Moderate certainty evidence suggests that a combined program of physical and mental health rehabilitation delivered online over 8 weeks probably increases the proportion of patients who experience recovery or important improvements, and probably improves quality of life, compared to a single session of advice and support.
However, the effectiveness of these interventions was modest, just reaching the minimal important difference for most outcomes. We did not find compelling evidence to support the effectiveness of other interventions, including drug treatments, dietary supplements, and various medical devices and technologies.
These findings come with important caveats. Long COVID may be a heterogeneous condition, and it is unclear whether these interventions are broadly effective across all patients. Additionally, the success of interventions such as CBT and physical/mental health rehabilitation may depend on the fidelity with which they are replicated in future trials and settings, along with the experience of the therapists delivering them.
Concerningly, one in four of the trials we reviewed raised doubts about the integrity of the study results or execution, which may not be immediately apparent to evidence users. This highlights the critical need for rigorous, transparent, and well-designed clinical trials to inform the management of long COVID.
Understanding Long COVID
Long COVID, also known as post-acute sequelae of COVID-19 (PASC), is a multisystemic condition comprising a wide range of often severe symptoms that persist for at least 3 months after the initial SARS-CoV-2 infection. It can occur after both mild and severe acute COVID-19, and affects people of all ages.
The prevalence of long COVID is difficult to establish because most symptoms are non-specific, and many studies lack sufficiently rigorous designs to confidently attribute symptoms to the COVID-19 infection. However, conservative estimates suggest that at least 65 million individuals worldwide have experienced long COVID, with cases increasing daily as the pandemic continues.
Patients with long COVID may experience a wide variety of symptoms across multiple organ systems, including:
- Fatigue and post-exertional malaise
- Shortness of breath and respiratory issues
- Cognitive impairment (“brain fog”)
- Neurological symptoms like headaches, dizziness, and sensory disturbances
- Cardiovascular issues such as chest pain and heart palpitations
- Gastrointestinal problems including nausea, abdominal pain, and changes in appetite
- Musculoskeletal pain
- Mental health challenges like depression and anxiety
These symptoms can be debilitating, leading to significant functional impairment and reduced quality of life. In many cases, patients experience a relapsing and remitting pattern, with symptoms waxing and waning over time.
The pathophysiology of long COVID remains uncertain, but several potential mechanisms have been proposed, including:
- Persistent viral reservoirs or reactivation
- Immune dysregulation and autoimmunity
- Microbiome disruption
- Microvascular dysfunction and coagulation abnormalities
- Neuroinflammation and autonomic nervous system dysregulation
Additionally, long COVID appears to share many similarities with other post-viral syndromes, such as myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and postural orthostatic tachycardia syndrome (POTS). This overlap suggests that long COVID may involve mechanisms seen in these other conditions, including mitochondrial dysfunction, metabolic disturbances, and reactivation of latent viral infections.
Evaluating Interventions for Long COVID
Given the significant public health impact of long COVID and the lack of effective treatments, our systematic review aimed to compare the effectiveness of various interventions for the management of this condition.
We searched the medical literature for randomized controlled trials that enrolled adults with long COVID and compared any drug or non-drug intervention to placebo, sham, or usual care. Our search identified 24 eligible trials with a total of 3,695 participants.
These trials investigated a diverse range of interventions, including:
- Drug treatments (e.g., vortioxetine, leronlimab, glucosaminyl muramyl dipeptide)
- Physical activity and rehabilitation programs
- Behavioral interventions like cognitive behavioral therapy
- Dietary supplements and probiotics
- Medical devices and technologies (e.g., transcranial direct current stimulation, hyperbaric oxygen)
- Combination programs involving physical and mental health rehabilitation
For each intervention, we assessed the certainty of the evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, which considers factors like risk of bias, inconsistency, indirectness, imprecision, and publication bias.
Key Findings
Cognitive Behavioral Therapy
Moderate certainty evidence from a single trial suggests that an online CBT program probably reduces fatigue and improves concentration in patients with long COVID, compared to usual care.
The 17-week CBT program, called “Fit after COVID,” was developed based on existing protocols for severe fatigue in chronic health conditions. It addressed disruptive sleep-wake patterns, unhelpful beliefs about fatigue, low activity levels, social support, fears/worries about COVID-19, and pain coping mechanisms.
Compared to usual care, the CBT program probably reduced fatigue by a mean difference of 8.4 points on the Checklist of Individual Strength fatigue subscale (range 8-56, higher scores indicate greater impairment). It also probably improved concentration by a mean difference of 5.2 points on the Checklist of Individual Strength concentration problems subscale (range 4-28, higher scores indicate greater impairment).
These findings are encouraging, as CBT has been shown to be effective for similar post-viral syndromes like ME/CFS. However, the modest magnitude of the effects means that CBT alone may not be sufficient to fully address the range of long COVID symptoms.
Physical and Mental Health Rehabilitation
Moderate certainty evidence from a single trial suggests that a combined program of physical and mental health rehabilitation delivered online over 8 weeks probably increases the proportion of patients who experience recovery or important improvements, and probably improves quality of life, compared to a single session of advice and support.
The rehabilitation program consisted of weekly live, supervised group sessions with exercise physiologists, physiotherapists, and health psychologists. It focused on motivation, fear avoidance, managing emotions, fatigue, and stress/anxiety, in addition to graduated physical activity.
Compared to the control group, the rehabilitation program probably led to 161 more patients per 1,000 experiencing recovery or important improvements. It also probably reduced symptoms of depression by a mean difference of 1.5 points on the Hospital Anxiety and Depression Scale depression subscale (range 0-21, higher scores indicate greater impairment), and probably improved quality of life by a mean difference of 0.04 points on the PROMIS 29+2 Profile (range -0.022 to 1, higher scores indicate less impairment).
These results highlight the potential benefit of a holistic, multimodal approach to long COVID rehabilitation, targeting both physical and mental health. However, as with CBT, the magnitude of the effects was modest, suggesting that this intervention alone may not fully address the complex needs of patients with long COVID.
Other Interventions
For other interventions investigated in the trials, we did not find compelling evidence of effectiveness:
-
Drug treatments: Moderate certainty evidence suggests vortioxetine, a FDA-approved antidepressant, has little to no effect on cognitive function, depressive symptoms, or quality of life in long COVID. The evidence for other drug interventions was limited by concerns about study integrity.
-
Physical activity and rehabilitation: While one trial found that intermittent aerobic exercise probably improves physical function compared to continuous exercise, the evidence for other physical activity programs was low or very low certainty.
-
Dietary supplements and probiotics: Low to moderate certainty evidence suggested potential benefits of a synbiotic (probiotic-prebiotic) formula on fatigue, concentration, and dyspnea, but these findings require replication by independent researchers.
-
Medical devices and technologies: The evidence for interventions like transcranial direct current stimulation, hyperbaric oxygen, and others was limited to low or very low certainty, or raised concerns about study integrity.
Importantly, one in four of the trials we reviewed had issues that raised doubts about the integrity of the study results or execution, such as retrospective trial registration, implausibly large benefits, or unusually small variability in outcomes. This highlights the critical need for rigorous, transparent clinical trials in the field of long COVID.
Implications and Future Directions
Our systematic review provides important insights into the current state of evidence for managing long COVID. While the findings suggest that CBT and physical/mental health rehabilitation programs may offer modest benefits for some patients, the overall landscape of effective interventions remains limited.
Several factors should be considered when interpreting these results:
-
Heterogeneity of Long COVID: Long COVID may encompass distinct phenotypes or subtypes, and the effectiveness of interventions may vary depending on the predominant symptoms experienced by patients. Further research is needed to better characterize long COVID subgroups and tailor treatment approaches accordingly.
-
Importance of Intervention Fidelity: The success of complex interventions like CBT and rehabilitation programs may depend heavily on how well they are implemented and the expertise of the providers delivering them. Ensuring high-quality, consistent delivery will be crucial for replicating the benefits observed in trials.
-
Need for Rigorous, Transparent Clinical Trials: The high proportion of trials with integrity concerns highlights the urgent need for well-designed, registered, and reported clinical studies to establish a robust evidence base for long COVID management. Funding bodies and researchers must prioritize transparent, high-quality trials that can withstand scrutiny.
-
Alignment with Existing Post-Viral Illness Research: Given the similarities between long COVID and conditions like ME/CFS, researchers should build on the extensive body of knowledge from these related fields. Incorporating insights from the ME/CFS community, including patient-centered approaches, may accelerate progress in long COVID research and care.
-
Addressing Health Inequities: Long COVID disproportionately affects marginalized populations, including racial/ethnic minorities and those of lower socioeconomic status. Future research must be inclusive and representative, and interventions must be designed with accessibility and equity in mind.
As the long COVID crisis continues to unfold, a multipronged approach will be essential. In addition to further research, healthcare providers, policymakers, and the general public must work together to raise awareness, improve access to care, and provide comprehensive support for those affected by this debilitating condition.
Only by learning from past mistakes, engaging patients as true partners, and leveraging the best available evidence and expertise can we hope to effectively manage the long-term consequences of the COVID-19 pandemic.
