Parental divorce in childhood tied to 61% higher stroke risk in older adults

Childhood family upheaval casts decades-long shadow: Older adults with divorced parents face stroke risks rivaling diabetes—even without abuse, finds groundbreaking U.S. study.

Study: Parental divorce’s long shadow: Elevated stroke risk among older Americans. Image Credit: VGstockstudio / Shutterstock

In a recent study published in the journal PLoS ONE, a group of researchers investigated the association between parental divorce in childhood and stroke risk in older adulthood among individuals without a history of childhood abuse while examining sex-based differences in this relationship.

Background

Stroke affects nearly 795,000 individuals annually in the United States (U.S.), with significant mortality and long-term disability rates. Survivors often face cognitive decline, reduced mobility, and loss of independence, adding to the $56.5 billion annual economic burden. Known risk factors include sociodemographic variables, unhealthy behaviors, and adverse childhood experiences (ACEs), such as abuse and neglect, which are linked to stroke and cardiovascular disease. Parental divorce, a form of ACE, has been associated with adverse adult health outcomes, yet its specific link to stroke remains underexplored. The study focused on a cohort of older adults born before 1969, a time when divorce was less common and more socially stigmatized than in later decades. Further research is needed to clarify this relationship and inform prevention strategies.

About the Study

The present study utilized a secondary analysis of the publicly available Behavioral Risk Factor Surveillance System (BRFSS) dataset, funded by the Centers for Disease Control and Prevention (CDC). The BRFSS is an annual cross-sectional survey that collects health-related data from non-institutionalized U.S. adults aged 18 and older. The dataset used was accessed on July 16, 2024, and included de-identified data, which did not require ethics approval. The 2022 BRFSS survey featured an optional module on ACEs administered in eight states, targeting respondents aged 18 and above. The initial sample size was 445,132 individuals, reduced to 13,205 after applying inclusion criteria, excluding individuals with missing data, a history of childhood physical or sexual abuse, or other childhood adversities such as parental substance abuse, incarceration, or domestic violence.

The outcome variable was self-reported physician-diagnosed stroke, measured as a binary response to whether respondents had ever been diagnosed with a stroke. The main exposure variable was parental divorce, assessed based on whether respondents experienced parental separation or divorce before age 18, excluding individuals whose parents were never married. Covariates included socioeconomic status (SES), additional ACEs, social support, health behaviors, and chronic health conditions.

Descriptive, bivariate, and multivariable analyses were conducted. Binary logistic regression models were used to examine the association between parental divorce and stroke, adjusting for covariates. Interaction effects between parental divorce and sex were tested but found to be non-significant. Adjusted odds ratios (AOR) and 95% confidence intervals (CI) were reported, with statistical significance set at p < .05. Stata MP 17 was used for data analysis, employing survey weights to account for the BRFSS’s complex sampling design and ensure population-representative results.

Study Results

Among the 13,205 adults aged 65 and older who had never experienced childhood physical or sexual abuse, 7.3% self-reported a physician diagnosis of stroke, with 13.9% reporting parental divorce during childhood. Significant bivariate associations were observed between stroke prevalence and multiple sample characteristics. Respondents who experienced parental divorce had a significantly higher prevalence of stroke (11.2%) compared to those who did not (7.5%). Stroke prevalence was also associated with being older, male, living in a rural area, having a lower income, and experiencing adverse circumstances such as neglect, household substance use, domestic violence, or a lack of safety and protection during childhood. However, other ACEs—including emotional abuse, household mental illness, and parental incarceration—lost significance in adjusted models.

Marital status influenced stroke prevalence, with separated/divorced individuals showing the highest rates (10.7%) compared to those widowed (8.8%), single/never married (7.9%), or married (6.7%). Smoking status also demonstrated significant variation, with current smokers showing a higher prevalence of stroke (11.2%) compared to former smokers (8.5%) and never-smokers (7.1%). Physical inactivity and chronic health conditions like depression and diabetes were strongly associated with stroke. Respondents who were physically inactive had a significantly higher stroke prevalence (10.9%) compared to those who were active (6.7%). Similarly, stroke prevalence was higher among individuals diagnosed with depression (11.2%) or diabetes (11.9%) than among those without these conditions.

Multivariable logistic regression analysis showed that parental divorce remained significantly associated with stroke after adjusting for demographic, socioeconomic, and other covariates. In the fully adjusted model, respondents who experienced parental divorce had 1.61 times higher odds of having a stroke compared to those who did not (AOR = 1.61, p = .005). The association was partially attenuated with the inclusion of socioeconomic and other covariates, but it remained statistically significant.

Other significant findings from the analysis included an increased likelihood of stroke among older respondents. Compared to those aged 65–69, individuals aged 75–79 had 1.74 times higher odds of stroke, while those aged 80 and above had 2.11 times higher odds. Males were 1.47 times more likely to have a stroke than females. Low household income, depression, and diabetes were also significant predictors of stroke. Specifically, depression was associated with 1.76 times higher odds of stroke, while diabetes increased the odds by 1.37 times.

Conclusions

To summarize, this study explored the association between parental divorce and stroke among older adults aged 65+ with no history of childhood physical or sexual abuse. Parental divorce was linked to 1.61 times higher odds of stroke, comparable to established risk factors like depression (1.76) and diabetes (1.37). The association was consistent across sexes but remained unexplained by other factors such as poverty, stress, or sleep disorders. Hypothesized mechanisms include stress-related dysregulation of the hypothalamic-pituitary-adrenal axis. The authors also noted potential survivor bias, as the sample excluded individuals who may have died prematurely from stroke or other causes before age 65.