Psychosocial risk factors and nonfatal myocardial infarction
N J O'Connor,J E Manson,G T O'Connor,J E Buring
DOI: https://doi.org/10.1161/01.cir.92.6.1458
IF: 37.8
1995-09-15
Circulation
Abstract:Background: Numerous psychosocial factors have been hypothesized to play a role in coronary heart disease. However, existing studies have yielded inconsistent results. Methods and results: The relations between type A personality as well as suppressed versus expressed anger and risk of nonfatal myocardial infarction (MI) were studied in 340 patients and 340 age-, sex-, and community-matched control subjects. Subjects were interviewed at home to assess behavioral and medical cardiovascular risk factors, and fasting blood samples were obtained. Type A personality was associated with nonfatal MI in crude matched-pair analysis (OR, 1.57; 95% CI, 1.12 to 2.20; P = .008). Adjusting for known cardiovascular risk factors (including treated hypertension, body mass index, treated diabetes, family history of premature MI, physical activity, smoking, alcohol, total calories per day, and saturated fat) did not substantially change the magnitude of the point estimate, although the finding was no longer statistically significant (OR, 1.43; 95% CI, 0.97 to 2.09; P = .069). Further adjustment for lipids, including total cholesterol, total HDL, its subfractions (HDL2, HDL3), LDL, VLDL, and triglycerides, markedly attenuated the association (OR, 1.12; 95% CI, 0.66 to 1.90; P = .687), an effect due almost entirely to HDL cholesterol. Suppressed anger was positively but not statistically significantly associated with increased risk of MI in crude matched-pair analysis (OR, 1.33; 95% CI, 0.98 to 1.81; P = .065), in analysis adjusted for behavioral and medical cardiovascular risk factors (OR, 1.26; 95% CI, 0.89 to 1.78; P = .193), or after adjustment for lipids (OR, 1.11; 95% CI, 0.67 to 1.82; P = .695). Conclusions: These findings suggest a possible association of type A but not suppressed anger with risk of nonfatal MI that may be mediated by alterations in HDL cholesterol level. If decreases in HDL are not in the same causal pathway, then the apparent association between type A personality and risk of MI is due to confounding, principally by HDL.