Table 4.
Multivariate logistic regression models examining potential associations between soy milk consumption status and sex, race/ethnicity, education level, and age.
| Independent Variables | OR | CI | p | OR | CI | p |
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | |||||
| Sex | ||||||
| Female | 1.57 | [0.97, 2.56] | 0.067 | 1.55 | [0.96, 2.51] | 0.071 |
| Ethnicity | ||||||
| Mexican American | 4.26 | [2.07, 8.76] | <0.001 | 4.16 | [2.52, 10.18] | <0.001 |
| Other Hispanic | 3.21 | [1.61, 6.41] | 0.002 | 3.22 | [1.85, 6.91] | 0.002 |
| Non-Hispanic Black | 2.55 | [1.56, 4.17] | 0.001 | 2.62 | [1.69, 4.44] | <0.001 |
| Non-Hispanic Asian | 5.70 | [3.82, 8.53] | <0.001 | 5.60 | [4.23, 9.30] | <0.001 |
| Other Race a | 1.85 | [0.73, 4.72] | 0.185 | 1.95 | [0.78, 4.99] | 0.150 |
| Education level | ||||||
| Less than 9th grade | 1.51 | [0.86, 2.65] | 0.143 | 1.60 | [0.93, 2.77] | 0.087 |
| 9–11th grade | 1.36 | [0.78, 2.35] | 0.265 | 1.47 | [0.85, 2.57] | 0.159 |
| Some college or AA degree | 1.83 | [0.93, 3.62] | 0.079 | 1.75 | [0.88, 3.47] | 0.105 |
| College graduate or above | 2.14 | [1.21, 3.80] | 0.011 | 1.84 | [1.01, 3.33] | 0.045 |
| Moderate activity | ||||||
| Yes | 1.65 | [1.14, 2.40] | 0.011 | |||
| Smoking | ||||||
| Yes | 0.82 | [0.51, 1.32] | 0.402 | |||
Legend: a = includes multi-racial. Significant regression equations were found for both models: F(10,16) = 21.98 (model 1) and F(12,14) = 21.16 (model 2), respectively, with a p-value < 0.001 for both. Reference categories were as follows: Male sex; Non-Hispanic White; High school graduate/GED; Moderate recreational activities in a typical week: “no”; Smoking: “no”. OR = odds ratio. CI = confidence interval. The model is based on a total n of 8511 participants.