Table 5.
Associations between Body Mass Index (BMI) and cardiovascular mortality and all-cause mortality, Poisson regression coefficients and incidence rate ratio (IRR) from Poisson regressions (additive residuals).
| Cardiovascular mortality | Coefficient (95% CI) | IRR | Coefficient (95% CI) | IRR |
|---|---|---|---|---|
| Men | Women | |||
| Early adulthood BMI | 0.09 (0.07, 0.11) | 1.10 | 0.06 (0.01, 0.11) | 1.06 |
| Midlife BMI | 0.07 (0.05, 0.08) | 1.07 | 0.05 (0.01, 0.08) | 1.05 |
| Early adulthood BMI as instrumental variable for BMI | 0.10 (0.08, 0.13) | 1.09 | 0.07 (0.02, 0.12) | 1.04 |
| All-cause mortality | Men | Women | ||
| Early adulthood BMI | 0.05 (0.04, 0.06) | 1.05 | 0.03 (0.01, 0.04) | 1.03 |
| Midlife BMI | 0.01 (0.01, 0.02) | 1.01 | 0.01 (−0.00, 0.02) | 1.01 |
| Early adulthood BMI as instrumental variable for BMI | 0.06 (0.04, 0.07) | 1.05 | 0.03 (0.01, 0.05) | 1.03 |
Instrumental variable analyses, using early adulthood BMI as an instrument for midlife BMI, ivpoisson regressions (multiplicative residuals) in STATA. All associations adjusted for education.