Total serum cholesterol, atherogenic indices and their longitudinal association with depressive symptoms among US adults

Abstract

Serum cholesterol, both total and lipoprotein fractions, has been associated with mid- and late-life depression. Using longitudinal data on a large and ethnically diverse sample of urban adults, the associations of serum lipid profile measured by high or low total cholesterol (TC; >200 mg dl(-1); <160 mg dl(-1)) and by atherogenic indices, namely high total cholesterol and low-density lipoprotein cholesterol relative to high-density lipoprotein cholesterol, with change in total and domain-specific depressive symptoms over time were examined. Findings were compared by sex. (Hypothesis 1) In addition, baseline depressive symptoms as predictors for longitudinal change in lipid profile trajectory were tested. (Hypothesis 2) Mixed-effects regression analyses stratified by sex was used. Sample sizes of participants (n) and repeated observations (n') were: Hypothesis 1 (Men: n=826 ; n'=1319; Women: n=1099 ; n'=1817); Hypothesis 2 (Men: n=738; n'=1230; Women: n=964; n'=1678). As hypothesized, a higher level of atherogenic indices was linked to faster increase in depressive symptom scores, particularly depressed affect and interpersonal problems, though this relationship was found only among women. Among men a U-shaped relationship between baseline TC and longitudinal increase in somatic complaints and a direct link between low TC and longitudinal putative improvement in positive affect was found. On excluding statin users among women, low TC was associated with slower increase in depressed affect over time, whereas high TC was associated with faster increase in interpersonal problems. In summary, atherogenic indices were directly linked to faster increase in depressive symptoms among women only. More studies are needed to explain these sex-specific associations.

Figure 1

(a) Predictive margins of CES-D total score, mixed-effects regression model with TC:HDL-C atherogenic index, controlling for selected covariates, among women. (b) Predictive margins of CES-D total score, mixed-effects regression model with LDL-C:HDL-C atherogenic index, controlling for selected covariates, among women. CES-D, Center for Epidemiologic Studies-Depression; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol.

Figure 2

(a) Predictive margins of CES-D component 2 score (depressed affect), mixed-effects regression model with TC:HDL-C atherogenic index, controlling for selected covariates, among women. (b) Predictive margins of CES-D component 4 (interpersonal problems), mixed-effects regression model with TC:HDL-C atherogenic index, controlling for selected covariates, among women. (c) Predictive margins of CES-D component 2 score (depressed affect), mixed-effects regression model with LDL-C:HDL-C atherogenic index, controlling for selected covariates, among women. CES-D, Center for Epidemiologic Studies-Depression; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol.

Figure 3

(a) Predictive margins of CES-D domain 1 score (somatic complaints), mixed-effects regression model with TC, categorical, controlling for selected covariates, among men. (b) Predictive margins of CES-D domain 3 score (positive affect), mixed-effects regression model with TC, categorical, controlling for selected covariates, among men. CES-D, Center for Epidemiologic Studies-Depression; TC, total cholesterol.