Fig. 1

Trial profile. ^aThe number of patients who declined to participate did not differ between the intervention and usual care groups after adjusting for cluster randomisation (P = 0.28). ^bAt the measurement 3 months after the first visit, 459 participants had complete data available for calculation of the 10-year CVD risk score. At the measurement 24 months after the first visit, 403 participants had complete data available for calculation of the 10-year CVD risk score. ^cAt the measurement 3 months after the first visit, 546 participants had complete data available for calculation of the 10-year CVD risk score. At the measurement 24 months after the first visit, 499 participants had complete data available for calculation of the 10-year CVD risk score. ^dMissing values of (systolic blood pressure and/or total cholesterol/HDL-cholesterol ratio and/or smoking status) of 88 participants in the intervention group and 76 in the usual care were imputed. One participant in the intervention group died of an unknown cause without having attempted to have her medication deprescribed and was not included in the analysis. CVD cardiovascular disease, EMR electronic medical records, FH familial hypercholesterolemia, GP general practitioner, HDL high-density lipoprotein