AHA Scientific Sessions 2026 Abstract · Nov 2, 2026

Depressive Symptom Burden Increases with Cardiometabolic Multimorbidity Among United States Adults

By Desai K, Dalal N, Divya A

Background: Depressive symptoms and cardiometabolic disease frequently co-occur. Prior work has characterized cardiometabolic multimorbidity among adults with depression; less is known about whether depressive-symptom burden scales with accumulated cardiometabolic conditions among U.S. adults. Methods: We performed a cross-sectional analysis of 41,471 adults from NHANES 2005-2023 with complete PHQ-9 data and ascertainable cardiometabolic status. Moderate-to-severe depressive symptoms were defined as PHQ-9 ≥10. A cardiometabolic multimorbidity count ranging from 0 to 4 was created by summing cardiovascular disease, hypertension, diabetes, and hyperlipidemia. Survey-weighted logistic and linear regression models adjusted for age, sex, race/ethnicity, and survey cycle. Results: Weighted prevalence of moderate-to-severe depressive symptoms increased monotonically across conditions 0, 1, 2, 3, and 4: 6.9%, 7.9%, 9.0%, 13.3%, and 18.0% , respectively. Each additional condition was associated with higher adjusted odds of depressive symptoms (OR 1.55; 95% CI, 1.47-1.62), reaching OR 6.04 (95% CI, 4.75-7.68) for four versus zero conditions. The mean PHQ-9 score increased by 0.71 points per additional condition, and the cognitive/affective subscale increased by 0.30 points per condition, suggesting that the association was not explained solely by somatic-symptom overlap. In subgroup analyses, the highest prevalence was observed among women with CVD: 20.2% (95% CI, 17.9-22.6). Conclusion: Among U.S. adults, depressive-symptom burden increased stepwise with cardiometabolic multimorbidity. The gradient persisted using continuous PHQ-9 scores and cognitive/affective symptoms, supporting the robustness of the association. These findings highlight the potential value of integrated depression screening as cardiometabolic conditions accumulate. Cross-sectional design precludes causal inference.

doi.org/10.1161/circ.154.suppl_1.8863Populations

AgingResearch.News