Table 2.
Adjusted Associations Between Physical Activity Levels at ARIC Visit 1 and 25(OH)D Levels at ARIC Visit 2
| Model 1a | Model 2b | Model 3c | |
|---|---|---|---|
| Relative risk ratio (95% CI)d for deficient 25(OH)D <20 ng/mLe by PA levels | |||
| AHA-defined PA categories | |||
| Poor (reference) | Reference (1) | Reference (1) | Reference (1) |
| Intermediate | 0.91 (0.84, 1.00) | 0.94 (0.86, 1.02) | 0.95 (0.87, 1.04) |
| Recommended | 0.69 (0.63, 0.75) | 0.71 (0.65, 0.77) | 0.72 (0.66, 0.79) |
| METS*min/wk (per 100) | 0.97 (0.97, 0.98) | 0.97 (0.97, 0.98) | 0.97 (0.97, 0.98) |
| Differences [β coefficient (95% CI)]f in continuous 25(OH)D levels (ng/mL)e by PA levels | |||
| Poor (reference) | Reference (0) | Reference (0) | Reference (0) |
| Intermediate | 0.78 (0.39, 1.18) | 0.66 (0.26, 1.05) | 0.49 (0.10, 0.89) |
| Recommended | 2.40 (2.03, 2.77) | 2.18 (1.81, 2.54) | 1.97 (1.61, 2.34) |
| METS*min/wk (per 100) | 0.17 (0.15, 0.19) | 0.15 (0.13, 0.17) | 0.14 (0.12, 0.16) |
Data in bold type are statistically significant.
Model 1: Age, sex, race/center, education, smoking status, and alcohol intake.
Model 2: Model 1 plus additional potential mediating variables (of the association between PA or vitamin D and ASCVD risk): SBP, antihypertensive medication use, diabetes, total and HDL cholesterol levels, use of lipid-lowering medications, and eGFR.
Model 3: Model 2 plus BMI.
Relative risk ratios for vitamin D deficiency [<20 ng/mL (<50 nmol/L)] vs adequate 25(OH)D by PA levels were derived from Poisson regression models.
To convert 25(OH)D levels to nmol/L from ng/mL, multiply by 2.496.
β coefficients were derived from linear regression models and represent the differences (in ng/mL) in 25(OH)D levels by PA levels.