Because of the small number of cases and controls with diabetes, the wide confidence intervals, and the borderline significance, however, a type I error cannot be ruled out [ 1 ].
1
—
—
Because of the small number of cases and controls with diabetes, the wide confidence intervals, and the borderline significance, however, a type I error cannot be ruled out [ 1 ].