The characteristics of those included 29,050 participants by depressive symptoms status were depicted in Table S2. The prevalence of depression symptoms (PHQ-9 scale ≥10) was 9.07%. There were significant differences in demographic characteristics between the depressive symptoms group and the non-depressive symptoms group. People with depressive symptoms were more likely to females, younger, smokers, drinkers, obese, subjects with hypertension, diabetes, less education, lower family income, lower work activity and recreational activity, lower total energy intake, higher caffeine intake. In addition, individuals with ischemic heart disease or its subtypes had a higher percentage in the depressive symptoms group.
Table 1 showed the results of logistic regression analyses. Depressive symptoms were positively associated with the risk of ischemic heart disease, coronary heart disease, angina, and heart attack, based on crude odds ratios (ORs) and 95% confidence intervals (CIs) for depressive symptom. When the PHQ scores were analyzed in the logistic regression as a continuous variable, the results were still significant. After adjustment for age and sex (model1), the results were similar to the crude model. Further adjustment in model2, there was significant correlation between depressive symptoms and ischemic and its subtypes (coronary heart diseases, heart attack and angina) with the multivariate-adjusted ORs (95% CIs) were 2.44(1.91,3.10), 2.32(1.67,3.23), 2.18(1.71,2.78), and 2.72(1.96,3.79), respectively. We used a forest map to show the results (Figure S1).
Table 1
Weighted odds ratios (95% confidence intervals) of ischemic heart disease coronary heart disease heart attack and anginas across depressive symptoms, NHANES 2007–2018 (N = 29050)
| Crude(N=29050) | Model1(N=29050) | Model2(N=22598) |
Ischemic heart disease | | | |
OR (95% CI) | 2.10(1.75,2.53) | 3.01(2.46,3.68) | 2.44(1.91,3.10) |
P-value | <0.001 | <0.001 | <0.001 |
Coronary heart disease | | | |
OR (95% CI) | 1.91(1.49,2.46) | 2.80(2.12,3.69) | 2.32(1.67,3.23) |
P-value | <0.001 | <0.001 | <0.001 |
Heart attack | | | |
OR (95% CI) | 2.13(1.72,2.66) | 2.93(2.35,3.66) | 2.18(1.71,2.78) |
P-value | <0.001 | <0.001 | <0.001 |
Angina | | | |
OR (95% CI) | 2.63(2.01,3.45) | 3.31(2.49,4.40) | 2.72(1.96,3.79) |
P-value | <0.001 | <0.001 | <0.001 |
aCalculated using binary logistic regression. Model 1 adjusted for age, gender. Model 2 adjusted for age, gender, race/ethnicity, educational level, household income, caffeine intake, total energy intake, smoking, alcohol consumption, work activity, recreational activity, diabetes, hypertension and BMI. |
Table S3 estimated associations between depressive symptoms and ischemic heart disease and its subtypes stratified by age, sex, obesity, and central obesity. In sex stratification, the positive associations between ischemic disease and depressive symptoms were significant in both male and female groups. In age-stratified analysis, we found no meaningful association between depressive symptoms and coronary heart disease in the 20-39 age group. In other age groups, depressive symptoms were positively associated with ischemic heart disease and its subtypes. In stratified analyses by obesity and central obesity, depressive symptoms were significantly associated with risk of ischemic heart disease and its subtypes in all levels.
Table 2 showed the combined effect of depression and obesity (BMI≥30) and central obesity (WC≥102/88cm for males/females) on ischemic heart disease. The combined effect of depression and central obesity was significantly greater than the sum of the individual effect. Compared with the reference group, the OR (95% CI) of only central obesity was 1.16 (0.93,1.46), and the OR (95% CI) of only depression was 1.88(1.24,2.88), and 3.15 (2.32,4.48) for it with both central obesity and depression. The RERI and AP with 95% CIs were 1.10(0.01,2.19) and 0.35(0.06,0.64) for depressive symptoms and central obesity. However, the additive interaction between depression and obesity was not significant. The RERI and AP were 0.90(-0.31,2.12) and 0.26(-0.03,0.55), respectively. Table S4 described the synergic effect of depression and obesity on secondary outcomes, the three types of ischemic heart disease. Depressive symptoms and obesity were not significant for all three subtypes of ischemic heart disease. Depressive symptoms and central obesity could have a meaningful synergistic effect on heart attack. The RERI and AP were 0.84(-0.28, 1.96) and 0.31(0.00, 0.69), respectively. The synergistic effect was not significant when the outcome variables were coronary heart disease and angina pectoris.
Table 2
Synergic effect of depression and obesity on ischemic heart disease incidence, NHANES 2007–2018 (N = 29050)
| Incidence (%) | OR (95%CI) | RERI (95%CI) | AP (95%CI) |
BMI category | | | | |
BMI<30.0kg/m2 &Depression- | 55.92 | 1 | 0.90(-0.31,2.12) | 0.26(-0.03,0.55) |
BMI ≥30.0 kg/m2 & Depression- | 34.91 | 1.33(1.13,1.57) |
BMI <30.0 kg/m2 & Depression+ | 4.48 | 2.21(1.59,3.09) |
BMI ≥30.0 kg/m2 & Depression+ | 4.69 | 3.49 (2.57,4.73) |
Waist category | | | | |
Central obesity- &Depression- | 38.37 | 1 | 1.10(0.01,2.19) * | 0.35(0.06,0.64) ** |
Central obesity+ &Depression- | 55.46 | 1.16(0.93,1.46) |
Central obesity- & Depression+ | 2.78 | 1.88(1.24,2.85) |
Central obesity+ & Depression+ | 6.39 | 3.15(2.32,4.48) |
The model adjusted for age, gender, race/ethnicity, educational level, household income, caffeine intake, total energy intake, smoking, alcohol consumption, work activity, recreational activity diabetes and hypertension. |
*P < 0.05; **P < 0.01 |
The dose-response relationships between PHQ-9 scores and ischemic heart disease, coronary heart disease, heart attack, and angina pectoris were shown in Figure 2. In the restricted cubic spline model, we found linear relationships between PHQ score with the risk of ischemic heart disease (P for linearity<0.0001), coronary heart disease (P for linearity=0.015), heart attack (P for linearity<0.0001), and angina (P for linearity=0.001).