Table 4.
Sensitivity analysis of linear regression between dietary diversity score (continuous) and phenotypic age acceleration, weighted.
| Adjustment methods | Model 1 | Model 3 | Model 3 | |||
|---|---|---|---|---|---|---|
| β (95% CI) | p-value | β (95% CI) | p-value | β (95% CI) | p-value | |
| No multiple interpolation | −0.39 (−0.42, −0.36) | <0.001 | −0.39 (−0.42, −0.36) | <0.001 | −0.33 (−0.36, −0.30) | <0.001 |
| Alternative age acceleration a | −0.26 (−0.31, −0.21) | <0.001 | −0.13 (−0.18, −0.08) | <0.001 | −0.07 (−0.12, −0.01) | 0.008 |
| Datasets with non-illnesses b | −0.42 (−0.46, −0.39) | <0.001 | −0.33 (−0.37, −0.30) | <0.001 | −0.32 (−0.35, −0.29) | <0.001 |
| Changed age acceleration c | −0.46 (−0.50, −0.42) | <0.001 | −0.46 (−0.50, −0.42) | <0.001 | −0.38 (−0.42, −0.34) | <0.001 |
β (95% CI): regression coefficient (95% confidence interval).
a
Using the Klemera–Doubal method biological age to calculate age acceleration.
b
Cardiovascular disease or hypertension or diabetes mellitus or cancer.
c
Increase C-reactive protein to calculate aging (n = 11,886).