Table 5.

Maximum likelihood estimates of binary logistic and probit regression models for a CHD event for subjects in the Framingham offspring study in the 20-year observation period predicted by the demographic, behavioral and anthropometric variables and diabetes in Exam 1¹

	Dependent variable: CHD event in the 20-year period n = 3718
Independent variable	Binary logistic model				Probit model
	Coefficient	SE	95% CI for exp (β)		Coefficient	SE	Marginal effect	SE
Constant	-3.531	1.902	-	-	-2.072*	0.996	-	-
Sex	-0.979*	0.209	0.249	0.566	-0.507*	0.107	-0.051*	0.011
Age	0.080*	0.008	1.067	1.100	0.041*	0.004	0.004*	0.0004
Physical activity score	-0.021	0.017	0.947	1.013	-0.010	0.009	-0.001	0.001
Alcohol index	-0.004	0.011	0.974	1.019	0.002	0.006	-0.0002	0.0006
Cigarettes smoked	0.015*	0.004	1.008	1.023	0.009*	0.002	0.0009*	0.0002
Systolic Blood Pressure	-0.008	0.007	0.978	1.005	-0.003	0.004	-0.0003	0.0004
Diastolic Blood Pressure	0.033*	0.010	1.013	1.055	0.017*	0.005	0.002*	0.0006
LDL	0.007*	0.002	1.004	1.011	0.004*	0.001	0.0004*	0.0001
HDL	-0.020*	0.005	0.969	0.991	-0.009*	0.002	-0.0009*	0.0003
Weight	0.000	0.006	0.988	1.012	0.0005	0.003	0.0001	0.003
Height	-2.077	1.075	0.015	1.031	-1.075	0.564	-0.109	0.057
Diabetes in Exam 1	1.195*	0.333	1.721	6.343	0.695*	0.196	0.070*	0.020
Chi-square statistic2, df = 1	4.65*	-			2.46	-

¹ Values are slope coefficients and standard errors; see equation (5) in the text. ² Chi-square test for combining height and weight as the BMI based on a logarithmic specification of the model. * P < 0.05.