Table 2

MI-associated factors for men found significant by at least three of the four statistical analyses employed by this study, (A) who ever had an MI by 72 years of age and (B) who had a MI between 65 and 72 years of age.

Factor description (age at survey year)MI case/non-case
(n)
Random forest ‘important’ variablesLogistic regression OR (95% CI)χ2 (adjusted p value)Recursive partitioning (tree nodes)
(A) Males, ‘MI by 72 years of age’
High cholesterol (by 65 years)269/893Yes3.29 (2.59 to 4.18)***<0.0001CPresent
High cholesterol (by 72 years)221/987Yes3.32 (2.50 to 4.42)***<0.0001CPresent
Diabetes (by 65 years)124/254Yes3.24 (2.53 to 4.15)***<0.0001CPresent
Diabetes (by 72 years)122/458Yes2.26 (1.78 to 2.88)***<0.0001C
Stroke (by 65 years)48/56Yes5.01 (3.36 to 7.48)***<0.0001C
High blood pressure (by 72 years)269/1428Yes2.16 (1.67 to 2.79)***<0.0001C
High blood pressure (by 65 years)283/10232.39 (1.92 to 2.96)***<0.0001CPresent
Exposed to dangerous conditions at work (by 65 years)174/7841.41 (1.11 to 1.79)**0.0080CPresent
Parent or sibling had a MI before age 55 (by 65 years)82/2671.89 (1.43 to 2.49)***<0.0001CPresent
(B) Males, ‘MI between 65 and 72 years of age’
Stroke (by 65 years)13/55Yes4.08 (2.17 to 7.65)***0.0002F
Diabetes (by 65 years)36/254Yes2.71 (1.81 to 4.04)***<0.0001C
  • The ‘age at survey year’ of 65 and 72 represent survey years 2004 and 2011, respectively. Analyses associated with the ‘MI by 72 years of age’ data set included independent variables from all WLS survey years, 1957–2011, while analyses associated with the ‘MI between 65 and 72 years of age’ data set included only independent variables from WLS survey year 2004 or earlier.

  • ‘***’=p<0.001, ‘**’=p<0.01.

  • C, χ2 test; F, Fisher's Exact test.