Material and methods. In outpatient cardiology clinic, 245 randomly selected patients of both sexes aged 35-60 years with prior MI were examined with complex questionnaire for education (Ed), income (Inc), occupation (Oc), family status (FS), and subdepression (SbD, defined as 50-59 points at Zung’s scale). Patients with more serious mood disorders (60+ points at Zung’s scale) were excluded from further analysis. Univariate, bivariate and multivariate stepwise backward logistic regression analysis was done with SbD as dependent categorical variable, and sex, age, Ed, Inc, Oc, FS as independent categorical factors without and with interaction modes. Odds ratios (OR) were calculated from regression coefficients with 95% confidence intervals (CI).
Results. Overall SbD was found in 29 (22.8%) men and 42 (35.6%) women. In univariate analysis, ORs of SbD were 2.2 (1.3-3.9) for female sex, 2.8 (1.5-5.0) for older age (above median), 2.8 (1.4-5.5) for lower Ed (Ed other than higher), 3.9 (2.1-7.4) for lower Inc (below median), 4.5 (2.5-8.3) for lack of Oc, and 4.4 (2.4-8.2) for living without family. In multivariate model (overall accuracy 0.75), independent associations with SbD were confirmed for female sex (OR 1.8; 1.1-3.0), older age (OR 2.3; 1.2-4.6), lower Inc (OR 2.4; 1.2-4.9), lack of Oc (OR 2.9; 1.5-5.8), and lack of family (OR 3.8; 1.9-7.7). Positive multiplicative interactions were found in bivariate logistic regression models for SdD between female sex and all above mentioned factors: older age, lower Inc, lower Ed, lack of Oc, and lack of family, – stressing the fact of excessive female vulnerability to numerous hazardous factors for mood disorders. In multivariate regression models for SbD that were separately built for men and women the above mentioned variables behaved somewhat differently: in men the most important factors associated with SbD were lower Inc (OR 3.2; 1.2-8.4) and lack of family (OR 2.6; 1.2-5.8), in women – lack of Oc (OR 13.2; 2.6-66.4), lack of family (OR 8.9; 2.9-27.3), and older age (OR 3.0; 1.1-8.8). It has to be pointed out that in females the regression model explained the SbD variability to much greater extent than in males (pseudo-R square by Neigelkerke 0.515 vs 0.118).
Conclusion. Special attention should be reasonably paid to gender aspects of SbD and its multiple associated and predisposing factors to improve medical care and potentially prevent the post-MI patients from overt depression.