Lockdown ended. In this study, poor sleep didn't.

Cohort study. 2,390 people. two surveys about 5 months apart: October 2021 (baseline, during lockdowns per the authors) and March 2022 (follow-up, after all UK lockdown measures ended on 27 January 2022), timed near the autumn and spring equinoxes.

People were watched for years. Nobody was told what to do. A clue, not proof.

Who: UK adults from the COVID Symptom Study Biobank (recruited via the KCL/ZOE COVID Symptom Study app, with extra recruitment of people with long symptom duration); analysed cohort 80.7% women, middle-aged (mean age 57 in text, 58.2 in Table 1), 96.8% white, a third from the least deprived areas; about half had tested positive for SARS-CoV-2 by baseline

• Sleep score, lockdown75+ counts as poor
• Sleep score, after7.15 months later
• Had COVID vs notno gapno clear effect

More to know, from our read of the paper:

• Observational, all self-reported questionnaires, not linked to health records; authors: results are associations only
• Only ~5 months between surveys; authors say this may be too short to detect recovery
• Self-selected volunteer cohort deliberately enriched for long COVID; 2,390 of 8,349 invited; IPW cannot fix unmeasured confounding (income, job changes, furlough, lifestyle); not representative of men, younger people or ethnic minorities
• The how-to (get lingering poor sleep checked) is the authors' recommendation in Abstract/Discussion, not something the study tested

UK researchers surveyed the same 2,390 adults twice: in October 2021, during lockdowns, and in March 2022, after they ended. Most were middle-aged women, and all answers were self-reported. The average sleep score was 7 at the start and 7.1 five months later, above the cutoff of 5 the authors use for poor sleep. Having COVID made no clear difference. There was no data from before the pandemic, so the study can't show that lockdown caused the poor sleep.

We read the abstract and the full text of this paper and checked every number on this page against it (16 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.