Fig. 2

Factors associated with persistent symptoms following COVID-19 lasting i) ≥12 weeks and ii) ≥52 weeks versus those who reported being asymptomatic or symptoms resolved within 4 weeks. Logistic regression models with one or more COVID-19 symptoms lasting ≥12 weeks (y/n) or ≥52 weeks (y/n) as the binary outcome variables. Modelling of persistent symptoms as a function of biological and demographic variables. In the forest plot, data were presented as adjusted odds ratios (central dot) and 95% confidence intervals (bars). Adjusted odds ratios compare participants with persistent symptoms lasting i) ≥12 weeks (black) or ii) ≥52 weeks (blue) with those who reported being asymptomatic or symptoms resolved within 4 weeks (n = 126,016 participants for ≥12 weeks analysis / n = 121,142 participants for ≥52 weeks analysis). Mutually adjusted for age, sex, ethnicity, IMD, comorbidities, smoking status, severity of initial infection, dominant variant at the time of infection, and vaccination status. Data used: Supplementary Table 2.