The effects of sleep disturbance on dyspnoea and impaired lung function following COVID-19 hospitalisation: a prospective multi-centre cohort study

Background Sleep disturbance is common following hospitalisation both for COVID-19 and other causes. The clinical associations are poorly understood, despite it altering pathophysiology in other scenarios. We, therefore, investigated whether sleep disturbance is associated with dyspnoea along with relevant mediation pathways. Methods Sleep parameters were assessed in a prospective cohort of patients (n=2,468) hospitalised for COVID-19 in the United Kingdom in 39 centres using both subjective and device-based measures. Results were compared to a matched UK biobank cohort and associations were evaluated using multivariable linear regression. Findings 64% (456/714) of participants reported poor sleep quality; 56% felt their sleep quality had deteriorated for at least 1-year following hospitalisation. Compared to the matched cohort, both sleep regularity (44.5 vs 59.2, p<0.001) and sleep efficiency (85.4% vs 88.5%, p<0.001) were lower whilst sleep period duration was longer (8.25h vs 7.32h, p<0.001). Overall sleep quality (effect estimate 4.2 (3.0–5.5)), deterioration in sleep quality following hospitalisation (effect estimate 3.2 (2.0–4.5)), and sleep regularity (effect estimate 5.9 (3.7–8.1)) were associated with both dyspnoea and impaired lung function (FEV 1 and FVC). Depending on the sleep metric, anxiety mediated 13–42% of the effect of sleep disturbance on dyspnoea and muscle weakness mediated 29-43% of this effect. Interpretation Sleep disturbance is associated with dyspnoea, anxiety and muscle weakness following COVID-19 hospitalisation. It could have similar effects for other causes of hospitalisation where sleep disturbance is prevalent. Funding UK Research and Innovation, National Institute for Health Research, and Engineering and Physical Sciences Research Council.