Impact of childhood adversity on cardiovascular disease among middle-aged and older Canadians: Differential protective role of social support across sex and nativity statuses


Jalal Uddin, Dalhousie University

It is now well established that childhood toxic stress is associated with morbidity and mortality decades later. Research has also demonstrated differential exposure to and impact of adverse childhood experiences (ACE) for different population groups. Although research to date has clearly outlined the causal effects of ACE on adult health outcomes, fewer studies have explicitly examined how resilience resources in adult life may guard against childhood stressors that impact cardiovascular disease risk. This study examines the associations of ACEs with cardiovascular disease among middle-aged and older Canadians. Further, we examine how social support availability buffers against childhood stressors in shaping the risk of cardiovascular disease (CVD) and test whether the buffering process differs by sex and nativity status. Guided by the sociological framework of the stress process, we hypothesize that protective resources, such as social support, buffer ACE’s harmful impacts on CVD risk, and social stratification factors, such as sex and nativity, may modify the buffering process. This is a secondary analysis of the Canadian Longitudinal Study on Aging (CLSA). The CLSA is a nationally representative prospective longitudinal study that collects psychosocial, clinical, and biomarker data from a cohort of approximately 50,000 individuals aged 45 or above. The current analysis is based on an analytic sample of 32,688 participants aged 45-89 from the baseline (2011-2015). The outcome is a binary measure of any CVD event self-reported in the baseline tracking and comprehensive cohorts of the CLSA. The exposure is the number of ACEs experienced in childhood. The ACE module consists of retrospectively reported 14 items on household dysfunctions, economic hardship, and physical and emotional abuse. Perceived social support is measured using the 19-item Medical Outcomes Study (MOS) questionnaire. The analysis uses logistic regression to estimate ACEs’ associations with CVD in sequentially adjusted models controlling for sociodemographic (e.g., age, sex, marital status, nativity, education, income, and home ownership), CVD risk factors (e.g., smoking, alcohol drinking, physical exercise, and BMI), and perceived social support. The analysis examined whether ACEs’ effect on CVD risk differs by social support, sex, and nativity status by employing statistical interactions. Overall, 13% of participants reported any CVD. The prevalence of CVD was higher among males (14.5%) versus females (10.7%) and slightly higher among those born in Canada (12.6%) versus foreign-born participants (11.9%). We found a consistent association of ACE categories with any CVD outcome. Across models with sequential adjustment of covariates, participants with a higher number of ACEs were more likely to have any CVD. The magnitude of the association of 3 or more ACEs with any CVD was higher in males (Odds ratio: 1.43, CI: 1.08-1.89) than in females (Odds ratio: 1.24, CI: 0.94-1.63) and among participants born in Canada (Odds ratio: 1.33, CI: 1.08-1.66) than those were foreign-born (Odds ratio: 1.29, CI: 0.80-2.08). Further, we observed significant effect modification by perceived social support, broadly suggesting buffering of the harmful associations of ACEs with CVD outcome at higher social support scores. The buffering processes of social support further differ by sex and nativity categories, suggesting a stronger protective role of social support for females and foreign-born participants. This study finds differential associations of ACE with CVD outcomes among population subgroups in a nationally representative sample of middle-aged and older Canadians. Social support strongly buffers the impact of ACE on CVD risk and may differ for males and females and by nativity status. Our findings have policy implications for investing in social interventions that help develop nurturing, resilient, and protective communities in Canada.

Non-presenting authors: Emran Hasan, School of Health Administration, Dalhousie University; Mario Ulises Pérez-Zepeda, Dalhousie University; Melissa Andrew, Dalhousie University; Susan Kirkland, Dalhousie University

This paper will be presented at the following session: