This study aimed to investigate the role of cardiovascular health (CVH) and vascular events as potential contributors to socioeconomic inequalities in dementia using causal mediation analyses. We used data from the Three-City Cohort, a French population-based study with 12 years of follow-up, with active search of dementia cases and validated diagnosis. Individual socioeconomic status was assessed using education, occupation and income. A CVH score as defined by the American Heart Association and incident vascular events were considered separately as mediators. We performed multi-level Cox proportional and Aalen additive hazard regression models to estimate the total effects of socioeconomic status on dementia risk. To estimate natural direct and indirect effects through CVH and vascular events, we applied two distinct weighting methods to quantify the role of CVH and vascular events: Inverse Odds Ratio Weighting (IORW) and Marginal Structural Models (MSM) respectively. Among 5581 participants, the risk of dementia was higher among participants with primary education (HR 1.60, 95%CI 1.44–1.78), blue-collar workers (HR 1.62, 95%CI 1.43–1.84) and with lower income (HR 1.23, 95%CI 1.09–1.29). Using additive models, 571 (95% CI 288–782) and 634 (95% CI 246–1020) additional cases of dementia per 100 000 person and year were estimated for primary education and blue-collar occupation, respectively. Using IORW, the CVH score mediate the relationship between education or income, and dementia (proportion mediated 17% and 26%, respectively). Yet, considering vascular events as mediator, MSM generated indirect effects that were smaller and more imprecise. Socioeconomic inequalities in dementia risk were observed but marginally explained by CVH or vascular events mediators.

本研究旨在使用因果中介分析调查心血管健康 (CVH) 和血管事件作为痴呆症社会经济不平等的潜在因素的作用。我们使用了三城市队列的数据，这是一项基于人群的法国研究，进行了 12 年的随访，积极搜索痴呆病例并进行了验证诊断。使用教育、职业和收入评估个人社会经济地位。美国心脏协会定义的 CVH 评分和突发血管事件被分别视为中介因素。我们进行了多层次 Cox 比例和 Aalen 加性风险回归模型，以估计社会经济地位对痴呆风险的总体影响。通过 CVH 和血管事件估计自然直接和间接影响，我们应用了两种不同的加权方法来量化 CVH 和血管事件的作用：分别是逆优势比加权 (IORW) 和边际结构模型 (MSM)。在 5581 名参与者中，受过小学教育 (HR 1.60, 95%CI 1.44–1.78)、蓝领工人 (HR 1.62, 95%CI 1.43–1.84) 和收入较低 (HR 1.23, 95%CI 1.09–1.29)。使用加法模型，分别估计每 10 万人和每 10 万人每年新增 571 例（95% CI 288-782）和 634 例（95% CI 246-1020）痴呆病例用于小学教育和蓝领职业。使用 IORW，CVH 评分调节教育或收入与痴呆症之间的关系（比例分别为 17% 和 26%）。然而，考虑到血管事件作为中介，MSM 产生的间接影响更小、更不精确。观察到痴呆风险的社会经济不平等，但 CVH 或血管事件介质可以稍微解释这一点。