long covid has been around for years, why aren't there more cases of proven recovery

The short version: the evidence on long COVID recovery is thin, and the number of proven, documented recoveries is low because most studies do not follow people long enough, use weak definitions, and lack unvaccinated comparison groups.

The best data here comes from a 2025 Australian study of 1,234 adults with long COVID at 3 months after Omicron infection. At 6 months, 58.7% still had symptoms and only 41.3% had recovered [19]. Among those who did not recover, the number of symptoms stayed flat: a median of 7 symptoms at both 3 and 6 months [19]. That is not a recovery trajectory. It is a plateau.

A 2025 systematic review on post-COVID swallowing problems found that about 15-30% of patients still had trouble swallowing at 3-12 months, and 10-25% had it beyond 6 months [4]. A 2026 review of retinal, nerve, and muscle biomarkers found objective damage that persisted at one year: thinning of retinal nerve layers, small fiber neuropathy confirmed by skin biopsy in 56.5% of tested patients, and abnormal nerve signals on EMG in 56% at follow-up [6]. These are not symptoms that resolve on their own.

The evidence does not include a single large study that tracks recovery beyond one year with a true unvaccinated comparison group. The studies here are mostly observational, with no control for natural recovery or placebo effects. The most informative design—vaccinated versus unvaccinated over several years—has not been run.

My call: proven, documented long COVID recovery is uncommon in the studies that actually measure it, and the small fraction who do recover are not tracked long enough to know if recovery is real or just fluctuation. Confidence: moderate. The evidence is consistent across multiple studies, but the lack of long-term controlled data and the wide variation in definitions limits precision.