Table 2.
Association between chronic diseases considered individually and risk of subsequent dementia. Values are hazard ratios (95% confidence intervals)
| Chronic disease | Prevalent at 55 years* † | Prevalent at 60 years* † | Prevalent at 65 years* † | Prevalent at 70 years* † | Overall† ‡ |
|---|---|---|---|---|---|
| Coronary heart disease | 1.35 (0.94 to 1.95) | 1.46 (1.11 to 1.92) | 1.31 (1.03 to 1.66) | 1.16 (0.93 to 1.45) | 1.25 (1.04 to 1.50) |
| Stroke | NA | 3.55 (1.46 to 8.59) | 2.68 (1.38 to 5.20) | 1.62 (0.89 to 2.95) | 3.37 (2.54 to 4.49) |
| Heart failure | NA | NA | 3.44 (1.41 to 8.35) | 1.39 (0.69 to 2.81) | 2.49 (1.88 to 3.32) |
| Diabetes | 2.31 (1.56 to 3.43) | 2.21 (1.63 to 2.99) | 1.98 (1.54 to 2.55) | 1.60 (1.25 to 2.05) | 1.63 (1.35 to 1.97) |
| Hypertension | 1.37 (1.15 to 1.62) | 1.29 (1.10 to 1.51) | 1.29 (1.10 to 1.52) | 1.31 (1.10 to 1.56) | 1.55 (1.27 to 1.88) |
| Cancer | 0.97 (0.50 to 1.88) | 0.99 (0.60 to 1.64) | 0.69 (0.44 to 1.09) | 0.88 (0.63 to 1.24) | 1.06 (0.85 to 1.31) |
| Chronic kidney disease | NA | NA | NA | 5.18 (2.29 to 11.70) | 2.98 (2.22 to 3.99) |
| Chronic obstructive pulmonary disease | NA | NA | 3.64 (1.49 to 8.90) | 1.12 (0.46 to 2.72) | 2.20 (1.63 to 2.97) |
| Liver disease | NA | NA | NA | NA | 1.89 (1.11 to 3.21) |
| Depression | 1.42 (0.94 to 2.15) | 1.71 (1.25 to 2.35) | 1.79 (1.36 to 2.35) | 1.93 (1.50 to 2.50) | 2.95 (2.44 to 3.57) |
| Mental disorders | NA | 13.51 (6.53 to 27.95) | 6.66 (3.61 to 12.27) | 2.05 (0.97 to 4.35) | 5.73 (4.48 to 7.34) |
| Parkinson’s disease | NA | NA | 21.24 (8.66 to 52.09) | 19.31 (8.45 to 44.11) | 8.16 (5.68 to 11.73) |
| Arthritis/Rheumatoid arthritis | NA | 1.66 (0.68 to 4.05) | 1.51 (0.91 to 2.51) | 1.14 (0.78 to 1.67) | 1.49 (1.22 to 1.81) |
NA=not applicable (insufficient number of cases (<5) to allow analysis).
Four separate analyses were run for prevalent disease at age 55, 60, 65, and 70 years, with median follow-up of 19.6, 14.8, 10.1, and 6.7 years, respectively. All chronic diseases were studied in separate models.
Analyses were stratified on birth cohort (5 year groups) and adjusted for age (as time scale), sex, ethnicity, education, marital status, and health behaviours (smoking, alcohol consumption, physical activity, and diet). Covariate measurement was concurrent with measure of chronic diseases.
All chronic diseases were studied in separate models, and covariates were entered as time varying measures. Median follow-up was 31.7 years.