Table 2—

Linear regression modeling of the relationship between insulin and covariates

ModelParametersF value 2 degrees of freedomP value for ethnicity × BMIParameter estimate for ethnicity (black) × BMIParameter estimate for ethnicity (Mexican-American) × BMI
Model 1: Insulin = ethnicity + BMI + ethnicity × BMI
1Ethnicity × BMI interaction term
 Women14.23<0.0001−0.0100.005
 Men4.540.01−0.0020.008
All of the below modeled for women only
2Model 1 + diabetes (binary)5.090.006−0.0210.021
3Model 1 + education (binary, <high school vs. >high school)11.94<0.0001−0.0110.007
4Model 1 + % calories from carbohydrates (highest quartile vs. rest)8.540.0002−0.0090.003
5Model 1 + age (in 20-year age groups)11.72<0.0001−0.018−0.001
6Model 1 + income (family income <$20,000 vs. >$20,00012.63<0.0001−0.0100.000
7Model 1 + physical activity (no physical activity vs. any)16.32<0.0001−0.0190.006
8Model 1 + family history of diabetes11.71<0.0001−0.0070.021
  • Modeling in SAS. Models 2–8 contain covariate and all first- and second-order interaction terms. Interaction term contains 2 degrees of freedom. BMI was modeled continuously.