Family history, socioeconomic factors, comorbidities, health behaviors, and the risk of sudden cardiac arrest

Genetic, environment, and behaviour factors have a role in causing sudden cardiac arrest (SCA). We aimed to determine the strength of the association between various risk factors and SCA incidence. We conducted a multicentre case-control study at 17 hospitals in Korea from September 2017 to December 2020. The cases included out-of-hospital cardiac arrest aged 19–79 years with presumed cardiac aetiology. Community-based controls were recruited at a 1:1 ratio after matching for age, sex, and urban residence level. Multivariable conditional logistic regression analysis was conducted. Among the 1016 cases and 1731 controls, 948 cases and 948 controls were analysed. A parental history of SCA, low educational level, own heart disease, current smoking, and non-regular exercise were associated with SCA incidence (Adjusted odds ratio [95% confidence interval]: 2.51 [1.48–4.28] for parental history of SCA, 1.37 [1.38–2.25] for low edication level, 3.77 [2.38–5.90] for non-coronary artery heart disease, 4.47 [2.84–7.03] for coronary artery disease, 1.39 [1.08–1.79] for current smoking, and 4.06 [3.29–5.02] for non-regular exercise). Various risk factors related to genetics, environment, and behaviour were independently associated with the incidence of SCA. Establishing individualised SCA prevention strategies in addition to general prevention strategies is warranted.


Variable and measurements
The main items were a parental history of SCA (no or yes), socioeconomic factors including medical aid (no or yes), a low educational level (less than or equal to high school) (no or yes), underlying heart disease (no, CAD [myocardial infarction and angina pectoris], non-CAD [heart failure, arrhythmia, structural disease including valvular heart disease, congenital heart disease, and other heart diseases]), health behaviors including current smoking (≥1 cigarette per day within the past month) (no or yes), and non-regular exercise, defined as engaging in moderate-to-vigorous physical activity less than once per week over the past year.The CAPTURES-II project uses the same questionnaire for both the cases and controls 25 .In the case group, after a patient arrives at the ED, physicians at the ED conduct face-to-face interviews with the patients' families to collect patient information and recruit study participants for the community-based control group.Information about patients' demographics, socioeconomic factors, health behaviors, and comorbidities was collected.Comorbidities were entered as 'yes' only when a doctor or clinic diagnosed them, and treatment was also investigated.In addition to own heart diseases, comorbidity information for hypertension, diabetes mellitus, dyslipidaemia, and stroke was also collected.

Statistical analysis
The demographic findings of the SCA and community-based control groups are described.Continuous variables were compared using the Wilcoxon rank-sum test, and categorical variables were compared using the Chi-square test.For the case-control dataset, conditional logistic regression analysis was conducted to estimate the association of a parental history of SCA, medical aid, low educational level, own heart disease, current smoking, and non-regular exercise with OHCA incidence and to calculate the adjusted odds ratios (AORs) and 95% confidence intervals (CIs) after adjusting for potential confounders, including hypertension, diabetes mellitus, dyslipidaemia, and stroke.We also conducted a stratified analysis based on age (≥65 years and <65 years) and sex.All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).All p-values were two-tailed, and statistical significance was set at P < 0.05.

Main results
The results of the conditional multivariable logistic regression models, including AORs (95% confidence intervals [CIs] for SCA are shown in  2).
In the sex-specific multivariable analysis, a parental history of SCA, low educational level, and heart disease were associated with SCA, regardless of sex.However, current smoking was the only risk factor for SCA in women (AOR [95% CI] 8.63 [3.50-21.29])but not in men (AOR [95% CI] 1.05 [0.80-1.38])(Table 3).

Discussion
We found that parental SCA, a low educational level, the patient's own heart disease, current smoking, and nonregular exercise were independently associated with SCA after adjusting for comorbidities.Additionally, the strength of the association of risk factors for SCA differed according to the patient's sex and age.Current smoking in women, medical aid and a low educational level in young individuals and non-regular exercise in older adults showed stronger associations with SCA.Our findings suggest that various risk factors related to genetic, environmental, and behavioral factors independently contribute to the incidence of SCA, even if they adjust for each other's effects.In addition, a general prevention strategy for SCA can be effective; however, individual priorities among risk factors could also vary according to the characteristics of the population.The finding that a parental history of SCA is a risk factor for SCA is consistent with the findings of other investigations 16,26 .Previous reports have shown that SCA is more prevalent in individuals with two or more SCAs among first-degree relatives, suggesting that the increased risk of SCA associated with parental SCA may be related to a genetic background 27 .Spooner et al. 28 suggest that genetic variation in pathologic and physiologic mechanisms may contribute to SCA incidence through three main pathways: (1) atherosclerosis and thrombosis, (2) electrogenesis, and (3) neural regulation and control.However, the underlying genetic mechanisms predisposing individuals to SCA are multifactorial, and detecting essential genetic mutations and polymorphisms can be difficult 27 .Previous studies and our study indicate that a parental history of SCA is independently associated with the risk of OHCA after adjusting for other risk factors.However, there may also be factors related to familial influence, such as hypertension, diabetes, smoking, or physical activity, which increase vulnerability to SCA.In the stratified analysis, we found that parental history of SCD was the only significant risk factor in the male sex group.Moreover, the adjusted odds ratio of a parental history of SCA for SCA was slightly higher in the younger age group than in the older group.A previous study of autopsy results for SCD patients 29 showed that myocardial hypertrophy, a high genetic predisposition and cause of SCA, was more common in men and younger patients, which may explain our finding.Nonetheless, a definitive explanation cannot be provided.
Socioeconomic factors are also well known risk factors for cardiovascular disease, including SCA.Previous studies also reported a greater disparity in SCA incidence in the younger than in the older age group 17,18 .We also found that medical aid and a low educational level were significantly associated with SCA only in the young but not in the older age group.The trend of diminishing socioeconomic differences according to age, selective survival in the group with a lower socioeconomic status (i.e.only the healthiest individuals from the lower socioeconomic status group survived until old age), and earlier onset of other diseases or death in the low educational level group could be some possible explanations for our findings 30,31 .
In our study, patient's own heart disease, including non-CAD and CAD, showed relatively higher odds for OHCA in the female and younger groups.In general, it was an important risk factor for SCA regardless of age and sex, which is consistent with the results of previous studies 9,32 .
Similar to previous studies 33,34 , current smoking status was significantly associated with OHCA in the present study.However, the smoking prevalence was lower in women than in men, and the adjusted odds ratio of current smoking status was only significant in women.A previous study reported that smoking had a higher risk of acute coronary events and cardiovascular death in women than in men 35 .A previous study reported that smoking had a higher risk of acute coronary events and cardiovascular death in women than in men, with findings suggesting that this increased risk in females may be linked to genetic factors related to thrombin signaling 36 .Moreover, the risk of smoking on SCA was only significant in younger patients aged < 65 years, consistent with the findings from a British case-control study, which showed that the association between smoking and myocardial infarction weakened with every 10-year age group 37 .One explanation for this phenomenon could be the 'depletion of susceptibles' effect 38 .These results further emphasise the importance of smoking cessation for the female and younger age groups because smoking cessation significantly reduces OHCA risk 33 .Further studies regarding the association of smoking and its cessation with OHCA incidence according to age and sex are needed.
Exercise has a multifactorial effect on OHCA, which occurs more frequently during or shortly after vigorous exercise 39 .However, regular exercise reduces the risk of CAD, including sudden cardiac arrest 39,40 .Our study also found that regular exercise showed a stronger protective effect against SCA incidence in the age and sex groups.Regular exercise also proved to be a particularly effective effort to reduce OHCA in the group over 60 years of age.
Our study demonstrated the effects of a parental history of SCA, socioeconomic factors, heart disease, and health behavior on SCA incidence, which were previously known to be related to SCA.It also showed that each risk factor had a different effect size according to the sex and age group.Based on these results, risk management according to patient characteristics is required to reduce the burden of SCA in addition to common preventive strategies such as chronic disease management.
Our study had certain limitations.First, our study was a case-control study and not an interventional study.There may have been a significant potential bias that could not be controlled.For example, recall bias could have led to an underestimation or overestimation of the association between risk factors and SCA due to reliance on patient or family member recollections.Selection bias in our control group may have compromised comparability with cases, potentially affecting the strength of our findings.Additionally, misclassification of matching variables could have skewed our risk estimates, altering the apparent magnitude of certain risk factors.In addition, a control group was selected from the same risk population; however, there was a possibility of misclassification of the matching variables.Second, we could not investigate specific information about risk factors such as the cause of parental SCA, diagnosed period of heart disease, smoking period, and intensity of exercise.Third, the historical data on patient risk factors may have been underestimated or overestimated due to reliance on self-reports from patients and guardians during hospital visits.We did not match these self-reports with more objective methods or secondary data sources, such as health insurance records, which could have provided a more balanced and validated dataset.Fourth, in our study, we implemented 1:1 matching using variables such as age, sex, and level of urban residence.However, we were unable to completely overcome issues like selection bias.Fifth, during the data collection phase of our study, we encountered challenges in precisely achieving the planned 1:2 matching ratio based on age, sex, and level of urban residence.Fewer participants than expected were recruited for the control group, which meant that we could not fully meet the optimal comparison framework set out in our research design.Sixth, our study presumed cardiac origin for SCA cases unless a clear non-cardiac origin was evident.This assumption carries the potential for misclassification errors regarding the underlying cause of SCA.Finally, the investigators of our project were not blinded to the study hypotheses, which could have led to biased data collection.

Conclusion
In our study, a parental history of SCD, low educational level, heart disease, health behavior including smoking, and regular exercise were associated with OHCA incidence.Additionally, each risk factor showed a different effect size according to sex and age.It is necessary to develop a strategy that considers the strength of the association with OHCA risk factors and their different effects according to patient characteristics to establish a prevention program for SCA.

Table 1 .
Characteristics of the sudden cardiac arrest case group and age and residence level-matched control group.

Table 2 .
Risk of parental sudden cardiac death, socioeconomic factors, own heart disease, and health behaviours for sudden cardiac arrest.CI confidence interval.*Adjusted for diabetes, hypertension, dyslipidaemia, and stroke.

Table 3 .
Risk of parental sudden cardiac death, socioeconomic factors, own heart disease, and health behaviours for sudden cardiac arrest by sex.CI confidence interval.*Adjusted for diabetes, hypertension, dyslipidaemia, and stroke.