Skip to main content

Table 1 Linear regressions that predict the three chronic diseases from individual nutrients and socio-demographic control variables (income, gender, age, education level)

From: Large-scale and high-resolution analysis of food purchases and health outcomes

Feature Coefficient Std. error p-value
Hypertension
α (intercept) 0.3241 0.050 <0.001
Carbs 0.7652 0.075 <0.001
Fats 0.3225 0.066 <0.001
Sugar 0.2402 0.072 0.001
Proteins −0.2868 0.050 <0.001
Fibre −0.0627 0.051 0.216
Income 0.0477 0.042 0.259
%Females −0.2124 0.038 <0.001
Average age 0.1664 0.037 <0.001
Education 0.2451 0.042 <0.001
Durbin–Watson stat. = 2.048 Adj \(R^{2} = 0.388\)
Cholesterol
α (intercept) 0.2645 0.047 <0.001
Carbs 0.5877 0.070 <0.001
Fats 0.3382 0.062 <0.001
Sugar 0.2441 0.067 <0.001
Proteins −0.2745 0.046 <0.001
Fibre −0.0268 0.047 0.569
Income −0.0184 0.039 0.640
%Females −0.2322 0.036 <0.001
Average age 0.1272 0.035 <0.001
Education 0.1751 0.039 <0.001
Durbin–Watson stat. = 2.001 Adj \(R^{2} = 0.345\)
Diabetes
α (intercept) 0.5073 0.041 <0.001
Carbs 0.57659 0.061 <0.001
Fats 0.5002 0.054 <0.001
Sugar 0.4992 0.059 <0.001
Proteins −0.5137 0.041 <0.001
Fibre −0.1312 0.041 0.002
Income −0.1222 0.034 <0.001
%Females −0.3536 0.031 <0.001
Average age −0.0290 0.030 0.342
Education −0.0947 0.035 <0.006
Durbin–Watson stat. = 2.000 Adj \(R^{2} = 0.598\)
\