Xuxi Zhang

Supplementary Table S3 Multivariate linear regression models (12 ‐ month change in physical activity and follow ‐ up scores of frailty) (Continued) Items 12 ‐ month follow ‐ up overall frailty 12 ‐ month follow ‐ up physical frailty 12 ‐ month follow ‐ up psychological frailty 12 ‐ month follow ‐ up social frailty Crude Model B(95%CI) Adjusted Model B(95%CI) Crude Model B(95%CI) Adjusted Model B(95%CI) Crude Model B(95%CI) Adjusted Model B(95%CI) Crude Model B(95%CI) Adjusted Model B(95%CI) Living situation Living with others Reference Reference Reference Reference Living alone 0.15( ‐ 0.10,0.39) ‐ 0.08( ‐ 0.25,0.09) ‐ 0.11 a ( ‐ 0.21, ‐ 0.02) 0.62 c (0.53,0.70) Smoking No Reference Reference Reference Reference Yes  ‐ 0.02( ‐ 0.41,0.38) ‐ 0.17( ‐ 0.45,0.12) 0.09( ‐ 0.07,0.26) 0.06( ‐ 0.06,0.18) Alcohol risk No Reference Reference Reference Reference Yes  ‐ 0.21( ‐ 0.45,0.04) ‐ 0.07( ‐ 0.24,0.10) ‐ 0.11 a ( ‐ 0.21, ‐ 0.01)  ‐ 0.05( ‐ 0.12,0.02) Multi ‐ morbidity No Reference Reference Reference Reference Yes 0.50 a (0.12,0.87) 0.45 b (0.18,0.71) 0.06( ‐ 0.10,0.21) 0.05( ‐ 0.06,0.16) Intervention condition Control group Reference Reference Reference Reference Intervention group ‐ 0.46 c ( ‐ 0.68, ‐ 0.25) ‐ 0.39 c ( ‐ 0.55, ‐ 0.24)  ‐ 0.04( ‐ 0.13,0.05)  ‐ 0.01( ‐ 0.08,0.05) Adjusted R 2 , % 56.6 57.8 51.6 53.0 39.2 39.7 41.8 48.5 a p <.05, b p <.01, c p <.001, significant effect estimates in bold 3 71 Longitudinal association between physical activity and frailty