Skip to main content
. 2023 Nov 24;23:773. doi: 10.1186/s12877-023-04483-z

Table 3.

The association between BMI trajectories in late-middle age and risk of dementia in older age

Trajectory Model1 Model2 Model3 Model 4
No.participants No.event RR (95% CI) P value No.participants No.event ARR (95% CI) P value No.participants No.event ARR (95% CI) P value No.participants No.event ARR (95% CI) P value
Ascend BMI with low variation 3932 264 Reference 3920 262 Reference 3902 258 Reference 3902 258 Reference
Decline BMI with low variation 2690 226 1.25 (1.05–1.49) 0.013 2681 224 1.13 (0.95–1.36) 0.17 2674 223 1.16 (0.97–1.39) 0.113 2674 223 1.18 (0.98–1.42) 0.076
Ascend BMI with high variation 2761 225 1.21 (1.02–1.45) 0.033 2752 224 1.28 (1.07–1.54) 0.007 2741 224 1.27 (1.06–1.53) 0.01 2741 224 1.27 (1.06–1.52) 0.011
Decline BMI with high variation 1464 185 1.88 (1.56–2.27)  < 0.001 1460 185 1.79 (1.48–2.16)  < 0.001 1451 184 1.75 (1.45–2.12)  < 0.001 1451 184 1.76 (1.45–2.13)  < 0.001

BMI Body mass index

Model 1 was a univariate model for the BMI trajectory

Model 2 was additionally adjusted for sex, education, ethnicity, year of aged 65, and marital status

Model 3 was additionally adjusted for smoking status, alcohol consumption, and longstanding illness status. Model 4 was additionally adjusted for the mean BMI group