What is it?
Long COVID describes symptoms that continue or emerge after SARS-CoV-2 infection and cannot be explained by another diagnosis. Experiences range from smell changes to post-exertional malaise and cognitive dysfunction.
Why does it matter?
Millions of people are affected, yet routine tests may not capture the biological processes behind fluctuating symptoms.
What did researchers find?
Studies have reported immune changes, viral material in tissues, vascular abnormalities, autonomic dysfunction, and altered metabolism. The findings vary between studies and participants.
Where is the science right now?
This remains an active research field with observational studies and treatment trials. Many mechanistic findings are early and require replication.
Is this available to patients?
Clinical evaluation and symptom-focused care are available. No single FDA-approved drug currently treats Long COVID as a whole.
What are the limitations?
Definitions, comparison groups, timing, variants, vaccination, and pre-existing health all complicate studies. Long COVID may include several biological subtypes.
What happens next?
Researchers are working toward validated biomarkers, better trial designs, and treatments matched to biological subgroups.
Research Horizon
Updated September 25, 2026
Two large, well-run trials of treatments many people hoped would help came back negative in 2026. That is useful information, not a dead end: it narrows where researchers look next.
| Approach | What was tested | Result | Evidence stage |
|---|---|---|---|
| Antiviral for existing long COVID | RECOVER-VITAL: nirmatrelvir-ritonavir for 15 or 25 days vs placebo in 959 adults | No benefit in cognitive, autonomic or exercise-intolerance groups at day 90; no new safety signals | Tested, not supported |
| Brain training and stimulation for cognitive symptoms | RECOVER-NEURO: online cognitive training, structured rehabilitation and brain stimulation in 328 adults | No approach beat its comparison; every group improved somewhat over time | Tested, not supported |
| Antiviral during acute COVID to prevent long COVID | PANORAMIC Norway: 5 days of nirmatrelvir-ritonavir vs placebo | Long COVID at 3 months in 26% vs 43%, but only 144 of a planned 2,000 people enrolled | Early, promising, not settled |
| POTS after COVID | RECOVER-AUTONOMIC: IVIG, ivabradine and coordinated non-drug care | Underway; trial design published in 2026 | Active research |
"Not supported" means the treatment did not beat placebo in that trial, at that dose and duration, in those participants. It does not rule out benefit for every subgroup, and it is not a reason to stop anything a clinician prescribed.
Sources
- CDC: Long COVID overview
- NIH RECOVER Initiative
- Nature Reviews Microbiology: Long COVID mechanisms
- RECOVER-VITAL: nirmatrelvir-ritonavir for long COVID (Lancet Infectious Diseases, 2026)
- RECOVER-NEURO: interventions for cognitive symptoms (JAMA Neurology, 2026)
- PANORAMIC Norway: nirmatrelvir during acute COVID (Lancet Infectious Diseases, 2026)
- RECOVER-AUTONOMIC trial design (American Heart Journal, 2026)
What This Means
Multiple credible biological leads exist. None should be mistaken for a universal explanation or a clinically validated personal test.
What We Still Don't Know
It remains unclear which mechanisms drive which symptoms, how they interact, and which treatments meaningfully change long-term outcomes.