Depression linked to higher heart attack risk in NHANES 2005–2020 analysis

Depression has long been viewed as an emotional burden, but a sweeping new analysis of U.S. health data suggests it may also weigh heavily on the heart. In a study drawing on more than 37,000 American adults, researchers report that people with more severe depressive symptoms are substantially more likely to have suffered a myocardial infarction, the medical term for a heart attack, and that the relationship rises steadily and linearly with each additional point of depressive symptom burden. The findings, published in Annals of General Psychiatry, add weight to the growing argument that mental health screening belongs squarely inside cardiovascular prevention.

The research team, led by investigators at Hangzhou Hospital of Traditional Chinese Medicine in China, mined the National Health and Nutrition Examination Survey, or NHANES, a continuous program run by the U.S. National Center for Health Statistics that assesses the health of the civilian population through stratified, multistage probability sampling. From an initial pool of 76,496 individuals spanning survey cycles from 2005 through 2020, the investigators retained 37,139 adults aged 20 or older after excluding those lacking data on depressive symptoms or heart attack history. Within that final sample, 1,574 participants reported having had a myocardial infarction.

Depression was quantified with the nine-item Patient Health Questionnaire, or PHQ-9, a widely validated instrument administered in person at mobile examination centers. Each of its nine items captures how often a respondent has experienced a given depressive symptom over the preceding two weeks, scored from zero for “not at all” to three for “nearly every day,” yielding a total between 0 and 27. Participants were grouped into four severity bands: no depression (0 to 4), mild (5 to 9), moderate (10 to 14) and severe (15 to 27). A total score of 10 or higher, which previous studies have shown to achieve roughly 88 percent sensitivity and specificity for diagnosing depression, served as the study’s clinical threshold.

The heart attack outcome came from the survey’s Medical Conditions Questionnaire, in which participants were asked whether a doctor had ever told them they had a heart attack. Crucially, because NHANES employs a complex survey design, the researchers applied the appropriate sample weights, strata and primary sampling units so that their estimates represent the entire non-institutionalized U.S. adult population rather than merely the tens of thousands who walked into the examination centers.

The gradient that emerged was striking. The prevalence of myocardial infarction rose progressively across depression severity categories, from 2.89 percent among participants with no depressive symptoms to 4.39 percent with mild, 5.80 percent with moderate and 7.08 percent with severe depression, a pattern that was statistically robust at p < 0.001. In the unadjusted logistic regression model, every one-point increase in PHQ-9 score corresponded to a 6.2 percent increase in the odds of having had a heart attack, with an odds ratio of 1.062 and a 95 percent confidence interval of 1.047 to 1.077. Participants with severe depression faced approximately 2.56-fold higher odds of myocardial infarction compared with those without depressive symptoms.

Critically, the association survived increasingly aggressive statistical adjustment. The researchers built three logistic regression models: the first unadjusted, the second controlling for sex, age and race/ethnicity, and the third adding education, marital status, poverty income ratio, body mass index, systolic and diastolic blood pressure, triglycerides, total cholesterol, LDL-C, HDL-C, uric acid, smoking, alcohol use, vigorous or moderate physical activity, congestive heart failure, coronary heart disease and hypertension. Even under this fully saturated model, the link between depressive burden and heart attack history remained significant. To probe whether the relationship might bend at higher symptom levels, the team employed restricted cubic spline regression, a technique that fits flexible curves across the exposure range. No significant nonlinearity appeared in any model, supporting an approximately linear dose-response pattern between depression scores and heart attack odds.

Not every subgroup told the same story, however. The association was directionally consistent across most population strata, including those defined by sex, race, education, income, body mass index, smoking, drinking and hypertension, with no significant statistical interactions. But the researchers detected significant interactions for three variables: congestive heart failure, coronary heart disease and vigorous or moderate physical activity, all with interaction p-values below 0.05. In participants with pre-existing heart failure or coronary disease, and in those who engaged in higher levels of moderate-to-vigorous activity, the apparent association between depression and myocardial infarction was attenuated. The authors suggest that in people already carrying high baseline cardiovascular risk, the incremental signal attributable to depression may be diluted, while regular physical activity may buffer some of the inflammatory and autonomic consequences of depression.

Those buffering hypotheses point to the biological machinery that plausibly connects mood and myocardium. Depression is known to activate the hypothalamic–pituitary–adrenal axis and the sympathetic nervous system, elevating blood pressure and heart rate and thereby increasing cardiac workload. It also drives systemic low-grade inflammation, reflected in elevated C-reactive protein and interleukin-6, promotes endothelial dysfunction and platelet hyperreactivity, all of which accelerate atherosclerosis and plaque destabilization. Meta-analytic evidence cited by the authors shows that coronary patients with comorbid depression exhibit both higher inflammatory markers and lower heart rate variability, a signature of vagal withdrawal that heightens vulnerability to events. Emerging work also implicates the tryptophan–kynurenine pathway as a metabolic bridge linking neuroinflammation to coronary pathology, while platelet activation markers such as soluble P-selectin are elevated in coronary artery disease populations, lending biological plausibility to a depression-driven prothrombotic mechanism. Even the gut–heart axis may participate: microbial metabolites such as trimethylamine N-oxide and short-chain fatty acids can modulate immune and neuroendocrine pathways in ways that influence both atherosclerosis and depressive phenotypes.

Behavioral pathways likely compound the biology. In the study’s baseline data, rising depression severity tracked with higher body mass index, higher triglycerides, lower HDL cholesterol, more smoking, less physical activity and lower income, a cluster of shared vulnerabilities that simultaneously raise the probability of depressive symptoms and cardiovascular events. The authors are careful, though, about what their design cannot show. Because the analysis is cross-sectional, it cannot establish causation or temporal order. The PHQ-9 captures symptoms over just the prior two weeks, whereas the myocardial infarction outcome is a lifetime self-report, so reverse causation, particularly the well-documented phenomenon of post-heart-attack depression, cannot be excluded. Self-reported outcomes also introduce recall, misclassification and survival bias, and despite exhaustive adjustment, residual confounding from factors such as medication adherence, psychosocial stress and cardiac rehabilitation remains possible.

The broader context sharpens the urgency. Between the 2005–2006 and 2017–2020 survey cycles, the prevalence of myocardial infarction, depression, and their co-occurrence nearly doubled, a parallel secular trend that, while not itself evidence of individual-level association, mirrors a wider global rise in depression-related cardiovascular mortality. Clinical literature consistently finds depression in 20 to 40 percent of patients with coronary heart disease or heart failure, and recent observational work has linked depression in cardiac patients to a three- to five-fold higher risk of major adverse cardiovascular events within a year. Yet whether treating depression improves hard cardiovascular endpoints remains unresolved; some reviews note that antidepressants and exercise improve mood without proven survival benefits, while other data suggest that the persistence or worsening of depressive symptoms over time is more closely tied to adverse outcomes than any single snapshot of symptom severity.

The study’s authors argue their findings support integrating routine depression screening, using tools as simple as the PHQ-9, into both general and high-risk cardiovascular care, alongside holistic mind–body management that addresses physical activity, smoking and metabolic risk in patients with depression. They also stress the need for prospective cohort studies and interventional trials to clarify temporal sequence, test whether modifying depression can genuinely reduce heart attack risk, and validate whether the linear relationship observed here holds across diverse populations and health care systems. For now, the message from this nationally representative sample is clear: the burden on the mind and the burden on the heart rise together, and ignoring one may mean underestimating the other.