VCU survey finds over half of COVID smell-loss respondents still symptomatic after 5 years

Virginia Commonwealth University researchers reported in Otolaryngology–Head and Neck Surgery that more than half of respondents in a long-running national survey of adults who self-reported smell loss after COVID-19 still had smell dysfunction five years after their initial symptoms. The preliminary report describes follow-up surveys begun in 2020 among more than 2,000 adults who tested positive and reported smell loss. In late 2025, 419 respondents answered questions about current smell status, with 44.2% reporting complete recovery and 48.4% reporting partial recovery.

At the two-year follow-up, 267 respondents reported complete recovery at 38.2%, partial recovery at 54.3%, and no recovery or worsening at 7.5%. Five years later, complete recovery rose to 44.2%, partial recovery was 48.4%, and no recovery was 7.4%. Researchers said recovery appeared to plateau after two years: people who regained smell by then generally kept it, while those who had not often saw little change. Complete recovery was 61.7% among participants younger than 40 and 34% among older participants.

The study is an observational, self-reported survey, not a clinical trial, so it can describe persistence but cannot establish causes, treatment effects, or population-wide prevalence. It follows a selected group who sought or reported smell loss, which may overstate symptoms among all infected people. The provided report summary did not list funding. Still, the pattern challenges a simple recovery narrative: early improvement does not necessarily mean full restoration. The age difference is consistent with slower olfactory cell regeneration in older adults, but the report does not measure regeneration directly.

For people with persistent smell dysfunction, the stakes extend beyond taste. Researchers said inability to detect body odor, gas leaks, smoke, or spoiled food can create safety concerns, while distorted smells can reduce appetite, cause weight loss and contribute to depressive symptoms. The survey can show persistence, but not cause or treatment response; larger objective testing and treatment trials are needed.