Table 4.
The association of 3+ vs. 0–2 Ideal Cardiovascular Health Metrics with Incident Diabetes Stratified by Age, Sex, and Estimated Glomerular Filtration Rate.
| Hazard Ratios (95% CI) | Ideal vs non-ideal | p-value* |
|---|---|---|
| All participants | 0.63 (0.47, 0.84) | |
| Sex | 0.0058 | |
| Men (N = 990) | 0.90 (0.61, 1.33) | |
| Women (N = 1801) | 0.39 (0.25, 0.61) | |
| Age (median=53 years) | 0.0338 | |
| ≤median (N = 1389) | 0.80 (0.56, 1.15) | |
| >median (N = 1402) | 0.40 (0.24, 0.68) | |
| Estimated Glomerular Filtration Rate (median=97 ml/min per 1.73 m2) |
||
| ≤median (N = 1395) | 0.43 (0.27, 0.70) | 0.0625 |
| >median (N = 1396) | 0.77 (0.53, 1.12) |
p-value for the interactions.
Significant interactions existed for sex, age and estimated glomerular filtration rate in the association of ideal vs. non-ideal cardiovascular health with incident diabetes (all p<0.10). There was greater magnitude of the association in women vs. men, age greater than 53 years compared to less than or equal to 53 years and estimated glomerular filtration rate less than or equal to 97 ml/min per 1.73 m2 compared to greater than 97 ml/min per 1.73 m2. For instance, the 3+ ideal cardiovascular health metrics in women was associated with a 61% lower risk of incident diabetes compared those in the 0–2 group.