TABLE 4.
Median intake | Cases | Person-years | All participants | Confirmed diabetes | |
g/d | n | n | |||
Fresh milk | |||||
None | 0 | 1248 | 98,917.5 | 1.00 (reference) | 1.00 (reference) |
<100 g/d | 34.3 | 336 | 59,380.3 | 0.61 (0.54, 0.69)2 | 0.64 (0.55, 0.75) |
100–200 g/d | 150.0 | 652 | 130,002.7 | 0.56 (0.50, 0.62) | 0.64 (0.57, 0.72) |
>200 g/d | 250.0 | 34 | 9443.9 | 0.46 (0.32, 0.64) | 0.60 (0.41, 0.88) |
P for trend | <0.001 | <0.001 | |||
Powdered milk | |||||
No | 0 | 1784 | 20,8921.9 | 1.00 (reference) | 1.00 (reference) |
Yes | 4.10 | 486 | 88,822.4 | 0.74 (0.67, 0.82) | 0.85 (0.75, 0.96) |
P for trend | <0.001 | <0.01 |
Adjusted for age, energy intake (in kcal/d), BMI (in kg/m2), waist-hip ratio, smoking status, alcohol consumption, physical activity, income level, education level, occupation, and hypertension. A Cox proportional hazards model was used to assess the effect of milk intake on the incidence of type 2 diabetes. Tests for trend were performed by entering the categorical variables as continuous parameters in the models.
Relative risk; 95% CI in parentheses (all such values).