Associations of threat- and deprivation-related adverse childhood experiences with risk of cardiovascular disease in later life

Abstract

Exposure to adverse childhood experiences (ACEs) has consistent associations with increased long-term risk of cardiovascular disease (CVD). However, links of distinct ACE subtypes and particular ACEs with later-life CVD are insufficiently understood. This longitudinal cohort study initially recruited 20,452 participants from the China Health and Retirement Longitudinal Study and Life History Survey. Follow-up data were obtained from five waves conducted between 2011 and 2020. ACEs were measured with ten items adapted from the Life Stressor Checklist-Revised, a measure that includes five threat-related ACEs and five deprivation-related ACEs. The outcome measure was CVD (including heart disease and stroke) during the follow-up period. Multivariate Cox proportional hazards regression models assessed links of cumulative ACEs, ACE subtypes, and individual ACEs with CVD incidence. Of 13,920 included participants, 8434 (60.6 %) reported exposure to at least one ACE. During the study period of 9 years, 2689 participants (19.3 %) received a clinical diagnosis of CVD, including 2098 (15.1 %) with heart disease and 683 (4.9 %) with stroke. Both threat-related ACEs (hazard ratio [HR], 1.08; 95 % CI, 1.03–1.12) and deprivation-related ACEs (HR, 1.08; 95 % CI, 1.01–1.15) were independently associated with increased risk of CVD, with the strongest association observed for stroke. Findings underscore the importance of considering threat- and deprivation-related ACEs in assessments and targeted intervention studies as potential means of yielding long-term cardioprotective benefits.

1. Introduction

Exposure to adverse childhood experiences (ACEs) is linked to poorer mental and physical health and higher risk of cardiovascular disease (CVD). CVD refers to conditions affecting the heart and blood vessels, typically those involving atherosclerotic plaque buildup that narrows the arteries (e.g., heart disease, stroke). However, most studies used cumulative ACE scores, assuming that all ACE types exert similar mechanisms. Nonetheless, previous research [1] has proposed an alternative conceptualization of ACEs that differentiates deprivation-related ACEs (i.e., the absence of expected cognitive and social inputs) from threat-related ACEs (characterized by the presence of experiences that endanger one’s physical integrity); this framework is supported by neurobiological evidence suggesting that experiences of threat and deprivation differentially affect neural structure and function in humans. Social deprivation is associated with higher CVD incidence [2], and the American Heart Association incorporates a social deprivation index into its CVD risk equations. However, whether deprivation- versus threat-related ACEs differentially predict CVD remains unclear.

Using data from the China Health and Retirement Longitudinal Study (CHARLS), we examined the relative impact of five deprivation- versus five threat-related ACEs on late-life CVD after adjusting for confounders. Furthermore, we conducted stratified analyses to clarify associations with heart disease and stroke.

2. Methods

Participants were recruited from the CHARLS [3]. The initial cohort was established between June 2011 and March 2012, with follow-up surveys conducted every two years. The Life History Survey (LHS) collected data on childhood experiences retrospectively during the 2013–2014 wave of the CHARLS. Only participants without missing data on adverse childhood experiences (ACE) were retained. After excluding participants without follow-up, who had cardiovascular diseases at baseline, or were younger than 45 years old, 13,920 participants were included (Supplement eFig. 1). Ethical approval for the CHARLS was provided by the Peking University Research Ethics Committee (IRB00001052–11015). Participants provided written informed consent.

Fig. 1.

Associations between specific adverse childhood experiences (ACEs) and cardiovascular disease subtypes

Note: The graphs show adjusted hazard ratios (HRs) for any cardiovascular disease (CVD) by deprivation-related ACEs (A) and threat-related ACEs (B), adjusted for age, sex, residence, marital status, education, smoking status, drinking status, hypertension, hyperlipidemia, diabetes, and kidney disease. Panel (C) presents the specific incidence of ACEs for CVD, while panels (D) and (E) show adjusted HRs for CVD and heart disease, respectively, and panel (F) presents adjusted HRs for stroke, all adjusted for the same confounders.

ACEs were measured with ten items, including five threat-related and five deprivation-related adversities (Supplement eTable 1). Cumulative ACEs were operationalized as a continuous variable, and as a categorical variable divided into four groups: 0, 1, 2, or 3 or more ACEs. The primary outcome was incident cardiovascular disease (CVD) based on whether participants had been diagnosed with heart disease or stroke by a clinician during the follow-up period from 2013 to 2020. Following previous published research [4], the CVD diagnosis date was recorded as being between the date of the last interview and that of the interview reporting a CVD incident. Both demographic characteristics (age, sex, residence, marital status, educational level) and health-related lifestyle conditions (smoking status, drinking status, and presence of hypertension, hyperlipidemia, diabetes, and kidney diseases) were assessed at baseline.

Table 1.

Cox pH models of association between numbers of adverse childhood experience (ACE) and subsequent cardiovascular diseases in late life.

Outcome	Model 1 (Cardiovascular Diseases)a			Model 2 (Heart Diseases)a			Model 3 (Stroke)a
	Participants, No			Participants, No			Participants, No
	Events, No.	Incidence per 1000 PYs	HR (95 % CI)	Events, No.	Incidence per 1000 PYs	HR (95 % CI)	Events, No.	Incidence per 1000 PYs	HR (95 % CI)
No. of ACEs
0	1018	29.9	1 [Reference]	814	23.9	1 [Reference]	244	7.2	1 [Reference]
1	820	28.7	0.96 (0.87–1.05)	640	22.4	0.95 (0.85–1.05)	200	7.0	0.93 (0.77–1.12)
2	523	35.0	1.20 (1.08–1.33)	404	27.0	1.17 (1.04–1.32)	137	9.2	1.23 (1.00–1.52)
≥3	328	35.8	1.25 (1.11–1.42)	239	26.1	1.16 (1.00–1.34)	101	11.0	1.53 (1.21–1.94)
ACE, No.	-	-	1.07 (1.04–1.11)	-	-	1.06 (1.02–1.10)	-	-	1.12 (1.05–1.20)
Threat-Related ACEs, No.	-	-	1.08 (1.03–1.12)	-	-	1.06 (1.01–1.11)	-	-	1.12 (1.03–1.21)
Deprivation-Related ACEs, No.	-	-	1.08 (1.01–1.15)	-	-	1.0 (0.98–1.14)	-	-	1.13 (1.00–1.27)

Abbreviation: HR, hazard ratio; PY, person-year.

^aAll of three models were adjusted for age, sex, residence, marital status, educational level, smoking, drinking, hypertension, hyperlipoidemia, diabetes, and kidney diseases.

Incidence rates of CVD events, along with heart disease and stroke rates per 1000 person-years (PYs) were calculated based on categorical ACE groups. Missing baseline covariate values were imputed using the multiple imputations of chained equations method. Three sets of Cox proportional hazards (Cox pH) models were constructed to estimate adjusted hazard ratios (HRs) with 95 % confidence intervals (CIs) for CVD, heart disease, and stroke. All models adjusted for demographic characteristics and health-related lifestyle conditions listed above. A 2-tailed P < 0.05 significance level was considered to be statistically significant.

3. Results

Cumulative ACE scores ranged from zero to ten. As summarized in eTable 2, of 13,920 participants, 8434 (60.6 %) reported exposure to at least one ACE. Among these, 1472 (10.6 %) reported three or more ACEs and had a CVD incidence of 22.3 %. Among 1018 participants with no ACEs, the CVD incidence was 18.6 %.

As shown in Table 1, participants with two ACEs had an incidence rate of 35.0 per 1000 person-years and a 20 % increased hazard (HR, 1.20; 95 % CI, 1.08–1.33), while those with three or more ACEs had an incidence rate of 35.8 per 1000 person-years and a 25 % increased hazard (HR, 1.25; 95 % CI, 1.11–1.42). The risk of incident CVD among participants with one ACE did not differ significantly from those with no ACEs. Each additional ACE was associated with a 7 % increased risk of CVD (HR, 1.07; 95 % CI, 1.04–1.11), with similar trends for heart disease and stroke. Overall, total ACEs, deprivation-related and threat-related ACEs were positively associated with an 8 % increased risk of CVD (Figs. 1A and 1B), with stroke showing the steepest gradient and highest risk from both ACE subtypes.

As presented in Figs. 1C and 1D, domestic violence had a strong association with CVD, with an incidence rate of 40.6 per 1000 person-years and a 32 % increased hazard (HR, 1.32; 95 % CI, 1.17–1.50). Household mental illness was associated with a 24 % increased hazard (HR, 1.24; 95 % CI, 1.11–1.38) and an incidence rate of 37.6 per 1000 person-years. Physical abuse and bullying showed modest associations, with CIs crossing unity, while household substance abuse, emotional neglect, and unsafe neighborhood did not show statistically significant associations with CVD. Stratified analyses (Figs. 1E and 1F) indicated that domestic violence was associated with a 35 % increased hazard of heart disease (HR, 1.35; 95 % CI, 1.17–1.54) while household mental illness showed the strongest association with stroke, corresponding to a 50 % increased hazard (HR, 1.50; 95 % CI, 1.22–1.84).

4. Discussion

This national cohort study found that both threat- and deprivation-related ACEs were independently and significantly associated with increased risk of new-onset CVD, with stronger associations for stroke than for heart disease. These findings support both ACE dimensions as potentially important risk factors for later-life CVD.

Our findings further support the notion that cumulative ACEs may contribute to new-onset CVD risk among older adults. Higher ACE exposure has been linked to elevated inflammatory activity and stress hormones [5], indicated by hypothalamic–pituitary–adrenal (HPA) axis dysregulation and impaired blood pressure regulation, thereby increasing risk for CVD. Specifically, deprivation-related ACEs are associated with faster declines in later-life cognitive function [6]. These findings underscore the importance of distinguishing between different subtypes of ACEs that may initiate distinct biological pathways with varied long-term health implications.

This study further advanced the understanding of relationships between ACEs and CVD incidence by highlighting associations between specific ACEs and CVD risk. Notably, exposure to domestic violence (a threat-related ACE) and living in a household with mental illness (a deprivation-related ACE) were particularly strongly predictors of developing CVD. Although the remaining ACEs did not show significant associations with CVD incidence, this may be related to smaller numbers of events. Inconsistent associations across individual ACE exposures and CVD outcomes highlight the need for additional empirical evidence to better understand how distinct ACE profiles relate to clinical CVD events as well as established cardiovascular risk factors (e.g., diabetes, hypertension, obesity, dyslipidemia) and composite downstream outcomes (e.g., major adverse cardiovascular events).

This study had several limitations. First, retrospective ACEs assessment may have been affected by recall bias. Second, substantial attrition occurred over time, potentially resulting in selection bias when analyzing complete-case data. To mitigate this issue, we employed multiple imputation using a range of relevant covariates associated with both trauma exposure and CVD outcomes under the missing-at-random assumption.

In conclusion, this national cohort study documented positive relationships of threat- and deprivation-related ACEs with new-onset CVD, particularly stroke. These findings support incorporating childhood adversity into CVD risk assessments and highlight prevention and treatment of cumulative threat- and deprivation-related ACEs as a potential strategy to reduce long-term CVD burdens.

Financial disclosure

The study was supported by Beijing High Level Public Health Technology Talent Construction Project (Discipline Backbone-01–028), the Beijing Municipal Science & Technology Commission (No. Z181100001518005), the Capital's Funds for Health Improvement and Research (CFH 2024–2-1174) and the University of Macau (MYRG-GRG2023–00141-FHS; CPG2025–00021-FHS).

Author agreement

All authors have read and approved the final version of the manuscript and consent to its publication.

Authors statement

Study design: Zi-Mu Chen, Yuan Feng, Yu-Tao Xiang. Data collection, analysis and interpretation: Zi-Mu Chen, Yan-Bo Zhang, Sha Sha, Qinge Zhang, Zhaohui Su, Teris Cheung, Gabor S. Ungvari. Drafting of the manuscript: Zi-Mu Chen, Yu-Tao Xiang. Critical revision of the manuscript: Todd Jackson. Approval of the final version for publication: all co-authors.

Data availability

The data in this study were sourced from the China Health and Retirement Longitudinal Study and are available at: https://charls.charlsdata.com. Access to the dataset can be obtained from Peking University upon reasonable request to the research team. Similarly, the code used for analysis will be made available by the corresponding author upon reasonable request.

CRediT authorship contribution statement

Zi-Mu Chen: Writing – original draft, Formal analysis, Data curation, Conceptualization. Yan-Bo Zhang: Formal analysis, Data curation, Conceptualization. Sha Sha: Methodology, Data curation, Conceptualization. Qinge Zhang: Methodology, Data curation, Conceptualization. Zhaohui Su: Validation, Supervision, Methodology, Formal analysis. Teris Cheung: Supervision, Methodology, Investigation. Gabor S. Ungvari: Supervision, Methodology, Investigation. Todd Jackson: Writing – review & editing, Supervision. Yu-Tao Xiang: Writing – review & editing, Writing – original draft, Supervision, Methodology. Yuan Feng: Visualization, Validation, Supervision.

Declaration of competing interest

The authors have no conflicts of interest to declare.