Fig. 2. Linear mixed effects regression models for patient reported outcomes in relation to SIMD.
Linear mixed effects regression models were fitted to assess the association between each PROM measure and deprivation (SIMD quintile) between enrolment and follow-up. Models included a random subject effect, with fixed effects for age (linear), sex (binary), SIMD quintile (linear), study visit (binary), and a SIMD-by-visit interaction. In each pair of figures, the left panel shows the predicted mean PROM score (with 95% confidence bands), in relation to SIMD quintile, at Visit 1 (enrolment) and Visit 2 (28–60 days post discharge). The right panel shows the predicted mean score in relation to study visit, for each SIMD quintile; confidence bands are shown for most and least deprived quintiles only. The data relating to this figure are available in Supplementary Data.