Table 1.
NGT | T2D | P for trend | |||
---|---|---|---|---|---|
n | n | OR | 95% CI | ||
Men (IGFBP1, μg/l) | |||||
< 11 | 13 | 36 | 1.00 | ||
11–18 | 29 | 29 | 0.30 | 0.12–0.76 | |
19–28 | 28 | 25 | 0.33 | 0.14–0.77 | |
> 28 | 37 | 17 | 0.14 | 0.05–0.38 | < 0.001 |
Women (IGFBP1, μg/l) | |||||
< 22 | 3 | 26 | 1.00 | ||
22–32 | 14 | 15 | 0.17 | 0.04–0.85 | |
33–48 | 18 | 12 | 0.07 | 0.01–0.39 | |
> 48 | 25 | 7 | 0.02 | 0.01–0.13 | < 0.001 |
Odds ratios (ORs) and 95% confidence intervals (CIs) for the association of IGFBP1 serum levels measured at baseline and the development of T2D at 10‐year follow‐up. Men and women were grouped into quartiles according to their baseline level of IGFBP1.
Conditional logistic regression analysis was performed. Cases (T2D; type‐2 diabetes) were pair matched to their respective controls (NGT; normal glucose tolerance) by exact age and by family history of diabetes (negative or positive).