Depressive Symptoms and Prevalent Cardiovascular Disease: A Cross-Sectional Analysis in Port-au-Prince, Haiti

Technical Summary: Depressive Symptoms and Prevalent Cardiovascular Disease in Port-au-Prince, Haiti

Problem Statement
Cardiovascular disease (CVD) is the leading cause of mortality globally and in low- and middle-income countries (LMICs) like Haiti, where it accounts for an estimated 36% of adult deaths. While traditional risk factors (hypertension, diabetes, obesity) are well-documented, the role of mental health, specifically depression, remains underexplored in LMIC contexts. Existing literature, largely derived from high-income countries, suggests a link between depression and CVD through physiological, behavioral, and social pathways. However, data regarding this association in Haiti is scarce. Furthermore, sex differences in the depression-CVD relationship remain inconclusive, with some studies suggesting women may be more susceptible due to hormonal and psychosocial factors. This study addresses the gap by investigating the association between depressive symptoms (DS) and prevalent CVD in an urban Haitian cohort, specifically examining whether sex modifies this relationship.

Methodology
The study utilized a cross-sectional analysis of enrollment data from the Haiti Cardiovascular Disease (HCVD) Cohort, a longitudinal community-based study conducted in Port-au-Prince between March 2019 and August 2021.

• Study Population: The analysis included 2,995 adults (≥18 years) after excluding 10 participants who did not complete the depression assessment. The mean age was 41.9 years, with 58.0% female participants.
• Exposure (Depressive Symptoms): DS were measured using the Patient Health Questionnaire-9 (PHQ-9), translated into Haitian Creole. Scores were categorized as "none–mild" (<10) and "moderate–severe" (≥10). The tool demonstrated good internal consistency (Cronbach's α = 0.78).
• Outcome (Prevalent CVD): CVD was defined as a composite of heart failure, stroke, transient ischemic attack, myocardial infarction, and angina. Cases were adjudicated by a three-physician committee using epidemiologic criteria aligned with the American Heart Association, WHO, and European Society of Cardiology, classifying events as "definite" or "probable."
• Statistical Analysis: The authors employed generalized estimating equation (GEE) Poisson models with a log link to estimate prevalence ratios (PRs), accounting for household clustering. Models were adjusted for age, sex, education, income, food insecurity, smoking, alcohol use, physical activity, BMI, and stress. Traditional cardiovascular risk factors (hypertension, diabetes) were excluded from the primary adjustment set to avoid overadjustment, as they may lie on the causal pathway. Effect modification by sex was assessed on both multiplicative and additive scales (using Relative Excess Risk due to Interaction, RERI). Sensitivity analyses included multiple imputation for missing data and models excluding physical activity and BMI.

Key Results

• Prevalence: 16.2% of participants had moderate-to-severe DS. Prevalence was significantly higher in females (22.0%) than males (8.5%). The overall prevalence of CVD was higher among those with moderate-to-severe DS (21.5% in males, 23.3% in females) compared to those with none-to-mild DS (10.6% in males, 15.3% in females).
• Association: Moderate-to-severe DS were independently associated with a higher prevalence of CVD compared to none-to-mild DS. The fully adjusted prevalence ratio (aPR) was 1.36 (95% CI: 1.08–1.71).
• Sex Differences: In sex-stratified models, the association remained significant for females (aPR = 1.38; 95% CI: 1.06–1.78) but was not statistically significant for males (aPR = 1.25; 95% CI: 0.75–1.99).
• Effect Modification: While the association appeared stronger in women, statistical tests for interaction were not significant on the multiplicative scale (p > 0.9). On the additive scale, the RERI was 0.07 (95% CI: −0.74 to 0.88), indicating limited evidence for effect modification by sex.
• Sensitivity: Results were robust across sensitivity analyses, including multiple imputation and models excluding physical activity and BMI (which yielded slightly stronger associations, aPR = 1.48).

Key Contributions and Claims
The paper contributes to the literature by providing evidence from a resource-limited, urban Haitian setting that moderate-to-severe depressive symptoms are independently associated with prevalent CVD, even after adjusting for socioeconomic determinants and lifestyle factors.

• Independence from Traditional Risk Factors: The study highlights that the DS-CVD association persists even when traditional risk factors like hypertension and obesity are not included in the adjustment set, suggesting depression may act as a distinct risk factor or that its effects are mediated through pathways not fully captured by standard clinical metrics.
• Contextual Relevance: The findings underscore the impact of social determinants of health (food insecurity, poverty, stress) in LMICs, where these factors may exacerbate both mental and cardiovascular health burdens.
• Sex-Specific Trends: While the study did not find statistically significant effect modification, it observed a consistent pattern of stronger associations among women, aligning with global trends of higher depression prevalence and vulnerability in females.

Significance and Limitations
The authors claim that these findings support the integration of depression screening and management into CVD prevention efforts in resource-limited settings. They argue that addressing mental health is crucial for mitigating the growing burden of CVD, particularly among women.

The paper explicitly acknowledges limitations:

• Causality: The cross-sectional design precludes establishing causality or the temporal sequence between DS and CVD.
• Generalizability: Findings are specific to an urban cohort in Port-au-Prince and may not extend to rural Haiti.
• Measurement: DS were self-reported via a screening tool administered by health workers, which may introduce bias.
• Unmeasured Confounding: Factors such as healthcare access, health literacy, family history, and neighborhood violence were not fully accounted for.
• Selection Bias: Individuals with severe DS may have been less likely to enroll.

The authors conclude that prospective studies are warranted to better understand the underlying mechanisms and causal pathways linking depressive symptoms and cardiovascular disease.