Table 2.
Multivariable-adjusted hazard ratios and population attributable risk percent for all-cause and cause-specific mortality by healthy lifestyle factors
| No. of deaths in healthy group | Mortality rate per 1000 person-years in healthy group | Hazard ratio (95% CI) | Population attributable risk percent (95% CI) | |
|---|---|---|---|---|
| All-cause mortality | ||||
| Non-smoking | 21,140 | 6.12 | 0.77 (0.75, 0.79) | 10.6 (9.5, 11.7) |
| Non-excessive alcohol intake | 33,911 | 7.51 | 0.88 (0.85, 0.91) | 1.3 (0.9, 1.7) |
| Being physically active | 16,454 | 6.76 | 0.78 (0.76, 0.80) | 13.9 (12.9, 15.0) |
| Healthy dietary habits | 2233 | 5.43 | 0.89 (0.85, 0.93) | 11.8 (8.0, 15.5) |
| Healthy body weight and fat | 24,755 | 7.12 | 0.76 (0.75, 0.78) | 8.2 (7.5, 8.9) |
| Ischaemic heart disease | ||||
| Non-smoking | 3013 | 0.87 | 0.75 (0.69, 0.80) | 10.4 (7.6, 13.1) |
| Non-excessive alcohol intake | 4767 | 1.06 | 1.01 (0.90, 1.13) | NAa |
| Being physically active | 1892 | 0.78 | 0.78 (0.73, 0.83) | 15.3 (12.4, 18.3) |
| Healthy dietary habits | 304 | 0.74 | 0.85 (0.75, 0.95) | 15.8 (6.0, 25.2) |
| Healthy body weight and fat | 3027 | 0.87 | 0.73 (0.68, 0.77) | 11.1 (9.2, 12.9) |
| Ischaemic stroke | ||||
| Non-smoking | 917 | 0.27 | 0.72 (0.63, 0.82) | 11.6 (6.6, 16.5) |
| Non-excessive alcohol intake | 1453 | 0.32 | 0.89 (0.73, 1.08) | 0.9 (−0.7, 2.5) |
| Being physically active | 538 | 0.22 | 0.63 (0.56, 0.71) | 25.7 (20.9, 30.4) |
| Healthy dietary habits | 91 | 0.22 | 0.87 (0.70, 1.08) | 14.0 (−4.1, 31.2) |
| Healthy body weight and fat | 988 | 0.28 | 0.81 (0.72, 0.90) | 6.7 (3.1, 10.3) |
| Haemorrhagic stroke | ||||
| Non-smoking | 2600 | 0.75 | 0.93 (0.86, 1.01) | 3.5 (0.2, 6.8) |
| Non-excessive alcohol intake | 4071 | 0.90 | 0.76 (0.68, 0.84) | 2.4 (1.3, 3.5) |
| Being physically active | 2088 | 0.86 | 0.76 (0.71, 0.81) | 13.3 (10.3, 16.2) |
| Healthy dietary habits | 149 | 0.36 | 0.78 (0.66, 0.93) | 23.1 (10.3, 35.2) |
| Healthy body weight and fat | 2975 | 0.86 | 0.82 (0.77, 0.88) | 5.3 (3.2, 7.5) |
| Cancer | ||||
| Non-smoking | 6712 | 1.94 | 0.70 (0.67, 0.74) | 13.7 (11.8, 15.6) |
| Non-excessive alcohol intake | 10,887 | 2.41 | 0.75 (0.71, 0.80) | 3.4 (2.6, 4.2) |
| Being physically active | 5984 | 2.46 | 0.90 (0.87, 0.94) | 6.3 (4.5, 8.1) |
| Healthy dietary habits | 931 | 2.26 | 0.95 (0.89, 1.02) | 4.3 (−1.9, 10.5) |
| Healthy body weight and fat | 8715 | 2.51 | 0.91 (0.87, 0.95) | 2.5 (1.2, 3.8) |
| Respiratory diseases | ||||
| Non-smoking | 982 | 0.30 | 0.70 (0.62, 0.79) | 13.6 (8.8, 18.4) |
| Non-excessive alcohol intake | 1657 | 0.39 | 0.93 (0.79, 1.10) | 0.6 (−1.0, 2.3) |
| Being physically active | 771 | 0.34 | 0.68 (0.61, 0.75) | 19.8 (15.3, 24.1) |
| Healthy dietary habits | 110 | 0.28 | 1.00 (0.81, 1.22) | 2.5 (−15.8, 20.6) |
| Healthy body weight and fat | 1074 | 0.33 | 0.56 (0.51, 0.61) | 19.1 (16.8, 21.3) |
| Other causes | ||||
| Non-smoking | 6157 | 1.78 | 0.87 (0.82, 0.91) | 4.8 (2.7, 6.8) |
| Non-excessive alcohol intake | 9392 | 2.08 | 0.95 (0.88, 1.01) | 0.5 (−0.2, 1.2) |
| Being physically active | 4467 | 1.83 | 0.76 (0.73, 0.80) | 14.8 (12.8, 16.7) |
| Healthy dietary habits | 589 | 1.43 | 0.86 (0.79, 0.93) | 14.7 (7.7, 21.6) |
| Healthy body weight and fat | 6918 | 1.99 | 0.76 (0.73, 0.79) | 7.9 (6.6, 9.2) |
Low-risk lifestyle factors were defined as follows: non-smoking or having stopped for reasons other than illness; non-daily drinking or drinking < 30 g (men)/15 g (women) of pure alcohol per day; engaging in an age- (< 50 years, 50–59 years, and ≥ 60 years) sex-specific median or higher level of physical activity; eating a diet rich in vegetables, fruits, legumes and fish, and low in red meat; having a BMI between 18.5 and 27.9 kg/m2 and a WC < 90 cm (men)/85 cm (women). Multivariable model was adjusted for sex (men or women); education (no formal school, primary school, middle school, high school, college, or university or higher); marital status (married, widowed, divorced or separated, or never married); family histories of heart attack, stroke and cancer (presence, absence, or unknown); and hip circumference (continuous). All five lifestyle factors were included simultaneously in the same model
aNo meaningful PAR% estimate was obtained because the estimated relative risk for this factor in the model had a negative coefficient