Table 3.
Hazard ratios for mortality risk associated with affect and life satisfaction
| Individual-level analyses, clustered variance estimators | Within-pair analyses, stratified by twin pair | |||||
|---|---|---|---|---|---|---|
| All twins | DZ twins | MZ twins | ||||
| HR (95% CI) | P-value | HR (95% CI) | P-value | HR (95% CI) | P-value | |
| Affect | N = 3,966 | N = 800 | N = 548 | |||
| Affect | 0.82 (0.79, 0.86) | 0.000 | 0.71 (0.61, 0.83) | 0.000 | 0.76 (0.61, 0.94) | 0.010 |
| Age at intake | 1.12 (1.11, 1.13) | 0.000 | - | - | - | - |
| Sex (Female) | 0.60 (0.55, 0.66) | 0.000 | - | - | - | - |
| Affect adjusted | N = 3,714 | N = 745 | N = 510 | |||
| Affect | 0.91 (0.86, 0.95) | 0.000 | 0.80 (0.67, 0.97) | 0.020 | 0.79 (0.63, 0.99) | 0.040 |
| Age at intake | 1.11 (1.10, 1.12) | 0.000 | - | - | - | - |
| Sex (Female) | 0.57 (0.52, 0.63) | 0.000 | - | - | - | - |
| Number of medications | 1.08 (1.05, 1.10) | 0.000 | 1.02 (0.94, 1.12) | 0.611 | 1.01 (0.91, 1.13) | 0.807 |
| Number of illnesses | 1.04 (1.02, 1.07) | 0.000 | 1.09 (1.00, 1.20) | 0.057 | 1.04 (0.94, 1.15) | 0.478 |
| Cognitive composite | 0.80 (0.76, 0.83) | 0.000 | 0.64 (0.52, 0.79) | 0.000 | 0.78 (0.60, 1.02) | 0.074 |
| Life satisfaction | N = 3,966 | N = 800 | N = 548 | |||
| Life satisfaction | 0.80 (0.77, 0.83) | 0.000 | 0.68 (0.58, 0.80) | 0.000 | 0.78 (0.65, 0.94) | 0.009 |
| Age at intake | 1.12 (1.11, 1.13) | 0.000 | - | - | - | - |
| Sex (Female) | 0.62 (0.57, 0.67) | 0.000 | - | - | - | - |
| Life satisfaction adjusted | N = 3,714 | N = 745 | N = 510 | |||
| Life satisfaction | 0.87 (0.83, 0.91) | 0.000 | 0.77 (0.64, 0.92) | 0.005 | 0.80 (0.65, 0.98) | 0.034 |
| Age at intake | 1.11 (1.10, 1.12) | 0.000 | - | - | - | - |
| Sex (Female) | 0.58 (0.53, 0.63) | 0.000 | - | - | - | - |
| Number of medications | 1.07 (1.05, 1.10) | 0.000 | 1.02 (0.93, 1.11) | 0.638 | 1.01 (0.90, 1.13) | 0.884 |
| Number of illnesses | 1.04 (1.02, 1.07) | 0.000 | 1.09 (1.00, 1.20) | 0.056 | 1.04 (0.93, 1.15) | 0.503 |
| Cognitive composite | 0.80 (0.76, 0.83) | 0.000 | 0.64 (0.51, 0.79) | 0.000 | 0.79 (0.60, 1.04) | 0.093 |
Sample sizes differ for univariate and multivariate models due to missing covariate data. The effects of age and sex are controlled by design in within-pair analyses.