Primary outcomes at 2-y follow-up: health behaviours and adherence to preventive care recommendations.

Modified intention-to-treat analysis based on all participants surviving in the community, with multiple imputation for missing values (intervention group, n = 827; control group, n = 1,320). Odds ratios and p-values based on z-statistics from generalised estimating equation models. For analysis with complete case dataset alone (i.e., dataset without imputed data), see S7 Table. Control group is reference group.aDenominator includes individuals aged n = 482; control group, n = 784 (individuals aged ≥75 y were excluded since the recommendation for cholesterol measurement was given to individuals aged bDenominator includes individuals aged n = 680; control group, n = 1,089 (individuals aged ≥80 y were excluded since the recommendation for faecal occult blood testing was given to individuals aged Primary outcomes at 2-y follow-up: health behaviours and adherence to preventive care recommendations.