Heart disease and depression may affect brain aging in an unexpected way

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Nov 3 2026

How do years of depressive symptoms affect the relationship between heart disease and cognitive aging? A 14-year study of more than 5,400 adults reveals a complex relationship between cardiovascular health, mental health, and changes in memory and thinking.

Study: The association between cardiovascular disease and cognitive decline varies by long-term depressive symptom trajectories in older adults. Image Credit: Lightspring / Shutterstock

In a recent study published in the journal Scientific Reports, a group of researchers investigated whether long-term depressive symptom trajectories modify the observed association between cardiovascular disease (CVD) and cognitive decline among middle-aged and older adults.

Background

Could persistent depressive symptoms influence how CVD affects brain health as people age? Projections indicating that dementia cases will exceed 150 million worldwide by 2050 highlight the importance of identifying factors associated with cognitive decline. CVD and depressive symptoms are independently associated with poorer cognitive outcomes in older adults. Depressive symptoms have been linked to poorer memory, slower information processing, and reduced verbal fluency. CVD may affect cognition by reducing cerebral blood flow and altering brain structure. Depressive symptoms change over time, but whether depressive symptom trajectories modify the association between CVD and cognitive decline remains unclear.

About the study

The researchers conducted a longitudinal cohort study using data from the English Longitudinal Study of Aging (ELSA), a nationally representative study of health, socioeconomic circumstances, cognitive function, and social relationships among adults aged 50 and older in England. Data analyzed were collected from 2004–2005 to 2018–2019, covering waves 2-9. Of the initial 9,432 participants, the final analysis included 5,423 after excluding individuals with missing CVD information, insufficient cognitive assessments, fewer than three depression assessments, or incomplete baseline characteristics. CVD included self-reported physician diagnoses of angina, heart attack, heart failure, stroke, and other heart problems.

Depressive symptoms were assessed repeatedly using the eight-item Center for Epidemiologic Studies Depression Scale (CES-D). The eight items were scored as binary responses, yielding a total CES-D score ranging from 0 to 8; higher total scores indicated more severe symptoms. A score of 3 or higher indicated clinically relevant symptoms, but the trajectories reflected symptom patterns rather than depression diagnoses. Participants were categorized into low-stable, moderate-stable, and high-stable depressive symptom trajectories. Depressive symptoms and cognitive changes were assessed over overlapping follow-up periods, so the analysis could not establish whether persistent symptoms preceded cognitive decline.

The researchers assessed cognitive function through memory evaluation, orientation, and tests of semantic fluency. Memory was assessed by determining immediate and delayed recall of 10 unrelated words, and orientation was evaluated by asking participants to identify the present date. Semantic fluency was tested using a task in which participants named as many animals as they could in one minute. Scores for each domain were standardized and combined to calculate global cognition.

The analysis accounted for demographic characteristics, body mass index (BMI), marital status, socioeconomic position, alcohol consumption, smoking, physical activity, activities of daily living (ADL), hypertension, diabetes, and lipid-lowering therapy. Growth mixture modeling (GMM) identified depressive symptom trajectories, while linear mixed-effects models evaluated longitudinal associations between CVD, depressive symptoms, and cognitive changes. Cross-sectional regression, sensitivity analyses, and multiple imputation were also performed. Statistical analyses were conducted using Stata 18.0 and R version 4.4.

Study results

Among the 5,423 participants, 3,027 (56%) were women, and the mean age was 65.2 years. At baseline, 1,080 participants (19.9%) had CVD. Participants with CVD had poorer baseline performance in memory, orientation, and semantic fluency than those without the condition. The mean memory scores were 9.9 and 10.7, respectively, and the mean semantic fluency scores were 19.9 and 21.0.

The researchers identified three distinct depressive symptom trajectories. The low-stable group represented 76.8% of participants, followed by the moderate-stable group at 18.1% and the high-stable group at 5.1%. CVD prevalence increased across these groups, accounting for 17.5%, 26.2%, and 33.7%, respectively.

CVD was linked to faster cognitive deterioration across the overall study population. After adjusting for potential confounders, participants with baseline CVD showed an additional annual decline of 0.045 standard deviations (SD) in global cognition compared with participants without CVD. Differences in annual decline associated with CVD were also observed in semantic fluency (0.034 SD/year), memory (0.034 SD/year), and orientation (0.018 SD/year).

Long-term depressive symptom burden was also associated with poorer cognitive performance and faster decline over the follow-up period. Compared with the low-stable group, participants with moderate-stable depressive symptoms had lower baseline global cognition, memory, and semantic fluency scores. Their baseline orientation scores did not differ significantly. Participants in the high-stable group demonstrated poorer baseline performance across all four cognitive measures.

Both moderate-stable and high-stable groups experienced faster longitudinal declines in global cognition, memory, and semantic fluency than the low-stable group. The moderate-stable group also demonstrated significantly faster deterioration in orientation, whereas the corresponding association in the high-stable group was not statistically significant.

The researchers also found that the association between CVD and cognitive decline varied according to depressive symptom trajectories. Among participants in the low-stable group, CVD was associated with a particularly pronounced decline in global cognition, with an additional annual decline of 0.051 SD compared with participants without CVD. This association was weaker in the moderate-stable and high-stable groups. The interaction coefficients indicated significant attenuation in the moderate-stable group (β = 0.031, 95% confidence interval [CI] = 0.002–0.060) and high-stable group (β = 0.059, 95% CI = 0.011–0.107).

Similar patterns were observed for orientation, whereas the corresponding interactions were weaker or statistically nonsignificant for memory and semantic fluency. In the moderate-stable group, participants with CVD experienced significantly faster memory decline, with an additional annual decline of 0.028 SD compared with participants without CVD. The corresponding associations with global cognition, semantic fluency, and orientation were not statistically significant. In the high-stable group, CVD-related differences were not statistically significant across the examined cognitive domains. These findings did not establish an absence of CVD-related cognitive risk, especially because the high-stable group was relatively small.

Participants with persistent depressive symptoms generally had poorer baseline cognition and faster overall cognitive deterioration. The weaker CVD association in these groups did not indicate better cognitive outcomes. Sensitivity analyses, including multiple imputation and models incorporating random intercepts and slopes, broadly supported the primary findings. Some differences were also observed by sex and socioeconomic position.

The observational design cannot establish causation. The researchers lacked information on CVD severity, depression treatment, and brain imaging, while participant attrition and mortality may have influenced estimates in higher-risk groups.

Conclusions

Long-term depressive symptom trajectories were associated with baseline cognitive performance and subsequent cognitive decline among middle-aged and older adults. Baseline CVD was also associated with faster deterioration in global cognition, memory, semantic fluency, and orientation. The additional decline associated with CVD was more pronounced in individuals who had low depressive symptoms than in individuals who had moderate to high levels of depressive symptoms. A smaller CVD-related difference in the higher-symptom groups did not mean their overall cognitive outcomes were better.

These findings suggest that cognitive risk assessments in older adults should consider long-term depressive symptoms alongside cardiovascular health.