Table 3.
Association between smoking status and incident dementia in competing-risk models
| Smoking status | No. of dementia cases | Competing-risk model of death | Competing-risk model of other types of disability | Competing-risk model of other types of disability and death | |||
|---|---|---|---|---|---|---|---|
| No. of competing events | Multivariable modela | No. of competing events | Multivariable model | No. of competing events | Multivariable model | ||
| Never smokers | 673 | 501 | 1.00 (ref.)b | 836 | 1.00 (ref.) | 1167 | 1.00 (ref.) |
| Current smokers | 154 | 248 | 1.35 (1.08, 1.68) | 149 | 1.41 (1.14, 1.75) | 338 | 1.34 (1.08, 1.66) |
| Ex-smokers (years since smoking cessation) | |||||||
| ≤ 2 | 34 | 63 | 1.34 (0.92, 1.94) | 33 | 1.41 (0.97, 2.05) | 82 | 1.36 (0.94, 1.96) |
| 3–5 | 29 | 72 | 0.94 (0.63, 1.40) | 36 | 1.04 (0.70, 1.55) | 93 | 0.94 (0.63, 1.40) |
| 6–10 | 43 | 87 | 0.95 (0.67, 1.34) | 70 | 0.97 (0.69, 1.37) | 129 | 0.94 (0.66, 1.33) |
| 11–15 | 40 | 63 | 1.18 (0.83, 1.66) | 36 | 1.24 (0.88, 1.74) | 82 | 1.20 (0.85, 1.70) |
| > 15 | 137 | 214 | 0.89 (0.70, 1.12) | 170 | 0.91 (0.72, 1.14) | 313 | 0.88 (0.70, 1.12) |
aMultivariable model was fully adjusted for the same covariates as Model 2 in the Table 2
bHazard ratios (HRs) and 95% confidence intervals (95% CIs) were calculated by Fine and Gray’s subdistribution hazards regression model