Table 4.
Multivariable-adjusted odds ratios (95% CIs) for BPH across tertile categories of DPI
| Tertiles of DPI | P-trend | |||
|---|---|---|---|---|
| T1 ≤ 19 |
T2 19 < DPI < 28 |
T3 ≥ 28 |
||
| Crud Model | 1.00 | 0.21(0.1–0.4) | 0.3(0.15–0.61) | 0.001 |
| Model I | 1.00 | 0.21(0.1–0.44) | 0.29(0.14–0.58) | 0.001 |
| Model II | 1.00 | 0.18(0.08–0.39) | 0.3(0.15–0.63) | 0.001 |
| Model III | 1.00 | 0.15(0.06–0.36) | 0.23(0.1–0.54) | 0.001 |
Binary logistic regression was used to obtain OR and 95% CI. The overall trend of OR across increasing tertiles was examined by considering each category’s median score as a continuous variable. Model I: Adjusted for Energy intake, waist circumference, BMI, weight. Model II: Model I plus education level and family history of BPH, smoking status, physical activity, marital status. Model III: Model I and II plus red meat, fish, cholesterol, egg, coffee, total fat, refined grains