Chapter 4 114 SUPPLEMENTARY MATERIALS Supplementary Materials contain Supplementary Methods, Supplementary Tables S1 and S2 and Supplementary Figures S1-5. Supplementary Methods 1. Analysis of determinants of time to recovery from symptoms We assessed the effect of age, sex, BMI, and number of comorbidities at illness onset on time to complete recovery from symptoms using univariable and multivariable Cox proportional hazards models in which these four variables were included. Interactions between BMI and the other variables were assessed. Clinical severity was not explored in this analysis due to collider bias with the other variables. To explore the impact of both survival bias and recall bias resulting from retrospective enrolment, a sensitivity analysis was performed by restricting the Cox PH models to prospectively-enrolled participants. The proportional hazards assumption of each HR was assessed using Schoenfeld residuals and time-varying covariates were used if this assumption was not met.
RkJQdWJsaXNoZXIy MTk4NDMw