Skip main navigation

Association of Joint Genetic and Social Environmental Risks With Incident Myocardial Infarction: Results From the Health and Retirement Study

Originally publishedhttps://doi.org/10.1161/JAHA.122.028200Journal of the American Heart Association. 2023;12:e028200

Abstract

Background

Myocardial infarction (MI) is a significant clinical and public health problem worldwide. However, little research has assessed the interplay between genetic susceptibility and social environment in the development of MI.

Methods and Results

Data were from the HRS (Health and Retirement Study). The polygenic risk score and polysocial score for MI were classified as low, intermediate, and high. Using Cox regression models, we assessed the race‐specific association of polygenic score and polysocial score with MI and examined the association between polysocial score and MI in each polygenic risk score category. We also examined the joint effect of genetic (low, intermediate, and high) and social environmental risks (low/intermediate, high) on MI. A total of 612 Black and 4795 White adults aged ≥65 years initially free of MI were included. We found a risk gradient of MI across the polygenic risk score and polysocial score among White participants; no significant risk gradient across the polygenic risk score was found among Black participants. A disadvantaged social environment was associated with a higher risk of incident MI among older White adults with intermediate and high genetic risk but not those with low genetic risk. We revealed the joint effect of genetics and social environment in the development of MI among White participants.

Conclusions

Living in a favorable social environment is particularly important for people with intermediate and high genetic risk for MI. It is critical to developing tailored interventions to improve social environment for disease prevention, especially among adults with a relatively high genetic risk.

Nonstandard Abbreviation and Acronym

HRS

Health and Retirement Study

Clinical Perspective

What Is New?

  • We found a joint effect of intermediate/high genetic risk and high social environmental risk in the development of incident myocardial infarction.

  • For older adults at intermediate and high genetic risk, a favorable social environment is associated with a lower risk of incident myocardial infarction.

What Are the Clinical Implications?

  • Our findings suggest the importance of developing comprehensive social intervention strategies to attenuate the intermediate/high genetic risk of myocardial infarction.

Myocardial infarction (MI) has become a significant clinical and public health challenge worldwide.1 The prevalence of MI increases substantially with advancing age, placing a disproportionally high burden on older adults.2 It has been well documented that MI is associated with a higher risk of disability,3 premature mortality,4 and greater use of health care resources and expenditure.5

Both genetic and social environmental factors play critical roles in the development of MI.6, 7, 8 Genome‐wide association studies (GWAS) have identified >200 genomic loci associated with the risk of MI.6, 9 A polygenic risk score has been created to comprehensively assess an individual's genetic risk of MI.10 Previous studies have also identified a few social environmental factors contributing to the development of MI.8, 11, 12, 13 An increasing number of studies have examined how genetic risk and lifestyle (eg, smoking, physical activity, diet) jointly influence MI.14, 15 However, little is known about the interplay between genetic predisposition and social environment in the development of MI. Social factors often coexist; exploring their synergistic effects is crucial. Motivated by the polygenic risk score,16 we developed a polysocial score to capture the aggregate health effects of a wide range of social factors among US older adults.17 This novel tool provides an opportunity to examine how the social environment influences the risk of MI across different levels of genetic susceptibility and whether the social environment could attenuate the genetic risk.

The present study aimed to investigate whether a disadvantaged social environment was associated with an increase in the risk of incident MI among older adults with low, intermediate, or high genetic risk. We also examined possible interactions between genetics and the social environment in the development of incident MI. Data were from the HRS (Health and Retirement Study), where we previously developed and validated a polysocial score to quantify the aggregated effect of social variables and a polygenetic risk score for MI is available. The findings would advance our understanding of the health benefits of social environmental factors among individuals with different levels of genetic susceptibility for MI and elucidate whether older adults living in a favorable social environment could counteract the risk effect of genetic factors on MI.

METHODS

Study Population and Analytic Sample

The HRS is a biennial longitudinal cohort study of a nationally representative sample of noninstitutionalized residents aged ≥50 years in the United States. The HRS collected health and economic condition information among US adults.18 Ethical approval was obtained from the institutional review board at the University of Michigan. Written informed consent was collected from all HRS participants. The data that support the findings of this study are available from the corresponding author on reasonable request. Details about study design, recruitment protocol, and data collection have been recorded elsewhere.19 The HRS study began in 1992. Additional cohorts were recruited in some of the follow‐up waves. Since 2006, the HRS has used a mixed‐model design for follow‐up in which a random half of the sample is assigned an enhanced face‐to‐face interview, including a supplementary psychosocial questionnaire and a saliva collection for DNA extraction. The other half of the participants completed the same questionnaire and saliva collection in 2008. The psychosocial questionnaire recorded participants' subjective evaluations of life conditions, well‐being, lifestyles, and social relationships.

We included 612 Black and 4795 White participants who (1) completed the psychosocial questionnaire at baseline (2006 or 2008), (2) had a polygenic risk score for MI, (3) were free of MI at baseline, and (4) had data to ascertain the incidence of MI in the follow‐up waves (2010, 2012, or 2014). Figure 1 shows how the analytic sample was derived.

Figure 1. Flowchart of participants.

Myocardial Infarction

MI was identified according to responses to the following questions in each wave: “Since [last interview date] has a doctor told you that you had a heart attack, coronary heart disease, angina, congestive heart failure, or other heart problems?” [If so] “Did you have a heart attack or myocardial infarction?” In the 2010, 2012, and 2014 waves, participants who responded “Yes” to the 2 questions were considered to have MI. The outcome was time to incident MI, calculated as the number of years from the interview date to the date reporting MI. Participants were censored at the date of the last contact or by the end of the follow‐up period, whichever came first. We did not use follow‐up data after the 2014 wave because participants' social environments might have changed during a relatively long period.

Polysocial Score for MI

Guided by our previous work,17 we used a polysocial score developed based on 14 social determinants from the following 5 domains: (1) economic stability, (2) neighborhood and physical environment, (3) education, (4) community and social contexts, and (5) health care system (Table S1). This polysocial score was developed among 7383 adults aged ≥65 years and was shown to be predictive of all‐cause mortality.17

Economic Stability

Total household income, total wealth, and out‐of‐pocket medical expenditure were categorized into quartiles of the sample distribution. Employment status was classified into working (full‐time or part‐time) or retired (retired or partly retired).

Neighborhood and Physical Environment

Housing types were classified as mobile house, 1‐family house, 2‐family house/duplex, or 34‐family house. Regions of residence were categorized as Northeast, South, Midwest, or West.

Education

Education level was categorized as less than high school, high school graduate, or postsecondary. The language used was classified into English or Spanish.

Community and Social Contexts

Marital status was categorized as widowed, married/partnered, or single (never married, separated, and divorced). Neighborhood social cohesion was measured by 7 conditions of the neighborhood areas within a 20‐minute walk.20 We assessed social engagement and stress (life‐threatening illness, natural disasters, addiction, unemployment, and crime) per the guidelines of the previous study.17

Health Care System

Long‐term care insurance included nursing home care and in‐home care. Life insurance included both whole life insurance policies and term life insurance policies.

The polysocial score ranged from 12 to 56, with a mean of 32.8 (SD=7.2); the distribution was nearly normal (Figure S1). We classified the polysocial score into the following 3 categories based on the sample distribution: low = 0 to 29, intermediate = 30 to 39, and high = ≥40.

Polygenic Risk Score for MI

The HRS polygenic risk score for MI was established using results for coronary artery disease from the 2015 GWAS.6 Specifically, the polygenic risk score contained all accessible MI‐related single‐nucleotide polymorphisms (SNPs) that overlapped between the GWAS meta‐analysis and the HRS genetic data (1 257 292 overlapped SNPs for European ancestry and 1 261 702 overlapped SNPs for African ancestry).21 For each overlapped SNP, the number of MI‐related alleles was weighted by the GWAS meta‐analysis effect size. The polygenic risk scores were constructed by standardizing the weighted sum scores of the overlapping SNPs for Black and White participants. We classified the polygenic risk score into low (lowest quintile), intermediate (quintiles 2–4), and high (highest quintile) risk, a routinely used approach for modeling the polygenic risk score in epidemiological studies.22, 23, 24, 25

Covariates

Covariates included demographics, healthy lifestyles, and the first 5 principal components of ancestry to account for potential population stratification and possible ancestry differences in genetic background. Demographics included age in years and male or female sex. Healthy lifestyles included body mass index (BMI), calculated as body weight in kilograms divided by standing height in meters squared and classified into underweight or normal (BMI ≤24.9 kg/m2), overweight (BMI 25.0–30.0 kg/m2), or obese (BMI ≥30.0 kg/m2); smoking status (never, former, or current); and alcohol consumption (0 drink per week, 1–2 drinks per week, or ≥3 drinks per week).

Statistical Analysis

We described participants' demographic and lifestyle characteristics by each polysocial score category (low, intermediate, and high) using analysis of variance for continuous variables and the χ2 test for categorical variables. We used Cox proportional hazard regression models to examine the race‐specific association of polygenic risk score and polysocial score with MI. As a result of the absence of a graded association between polygenic risk score and MI among Black participants, we restricted the subsequent analyses to White participants only. In each polygenic risk score category, we merged participants with an intermediate or high polysocial score into 1 group because of the limited event number of MIs in the high polysocial score category, and we examined the association between polysocial score and MI. We then examined the joint effect of genetic (low, intermediate, and high) and social environmental risks (low/intermediate, high) on MI by introducing the model term for the combined polygenic risk score and polysocial score categories (6 categories with low genetic risk and low/intermediate social environmental risk as reference). We used the summary(survfit()) function in R software after fitting a Cox model to estimate an 8‐year survival probability and its CI for each of the 6 categories. We then computed an 8‐year risk of incident MI and its CI for each of the 6 categories. We measured the additive and multiplicative interactions between the genetic and social environmental risks using the polygenic risk score categories and the grouped polysocial score categories. Studies have shown sex differences in the incidence of MI.26, 27 As a secondary analysis, we examined the effect modification of the association between polygenic score and incident MI by sex among Black and White participants separately. We conducted stratified analyses by sex if significant effect modification was observed.

We conducted several sensitivity analyses. First, we evaluated the robustness of the association of the combined polygenic risk score and polysocial score categories with MI, dividing the combined intermediate/high polysocial score group into intermediate and high polysocial score categories. Second, we conducted a sensitivity analysis in which sampling weights were incorporated. Third, we used a Cox model with penalized splines to examine the nonlinear or irregular shape of the association between continuous polygenic score and MI among Black and White participants. Finally, we used the Firth correction to a Cox model to handle monotone likelihood, an issue that could arise with sparse or heavily censored survival data. These analyses were adjusted for age, sex (except for the stratified analyses by sex), BMI, smoking status, alcohol consumption, and the first 5 principal components of ancestry.

All tests were 2 sided with a significance level of 0.05. All statistical analyses were conducted in R 4.1.2 and Stata/SE 15.0/17.0.

RESULTS

Sample Characteristics

Of 5407 participants, the average age was 74.2 years (SD = 6.8 years), 54.4% were female sex, and 88.7% were White participants; 1786 (33.0%), 2629 (48.6%), and 992 (18.3%) were classified as low, intermediate, and high polysocial score, respectively (Table 1). Compared with participants in the low and intermediate polysocial score categories, those with a high polysocial score were more likely to be younger male participants and less likely to be obese and smoke (Table 1).

Table 1. Baseline Characteristics of Participants by Polysocial Score Categories

CharacteristicsPolysocial scoreTotal, N=5407P value*
0–29, n=178630–39, n=2629≥40, n=992
Race, n (%)<0.001
White participants1475 (82.6)2387 (90.8)933 (94.1)4795 (88.7)
Black participants311 (17.4)242 (9.2)59 (5.9)612 (11.3)
Age, y, mean±SD76.0±7.474.0±6.471.8±5.574.2±6.8<0.001
Female sex, n (%)1046 (58.6)1388 (52.8)509 (51.3)2943 (54.4)<0.001
BMI, n (%)0.001
Underweight/normal531 (30.0)827 (31.8)329 (33.5)1687 (31.5)
Overweight702 (39.5)1049 (40.3)406 (41.3)2157 (40.2)
Obese542 (30.5)726 (27.9)248 (25.2)1516 (28.3)
Smoking status, n (%)<0.001
Never658 (37.0)1108 (42.5)477 (48.3)2243 (41.7)
Former902 (50.7)1269 (48.6)458 (46.4)2629 (48.9)
Current218 (12.3)233 (8.9)52 (5.3)503 (9.4)
Alcohol use, n (%)<0.001
0 drinks per wk1332 (74.7)1683 (64.1)530 (53.4)3545 (65.6)
1–2 drinks per wk207 (11.6)400 (15.2)193 (19.5)800 (14.8)
≥3 drinks per wk245 (13.7)544 (20.7)269 (27.1)1058 (19.6)

BMI indicates body mass index.

*Obtained from the analysis of variance or χ2 tests.

BMI is calculated as body weight in kilograms divided by standing height in meters squared and is classified into underweight or normal (≤24.9), overweight (25.0–30.0), or obese (≥30.0).

Association of Polygenic Risk Score and Polysocial Score With MI

White participants with intermediate and high polygenic risk scores had a significantly higher incidence rate of MI than those with a low polygenic risk score, with adjusted hazard ratios (HRs) of 1.61 (95% CI, 1.06–2.42) and 2.02 (95% CI, 1.27–3.20), respectively. The association of polygenic score with incident MI was stronger among female participants than male participants (Table S2; P=0.04 for intermediate polygenic score×female sex; P=0.01 for high polygenic score×female sex). We did not find an obvious risk gradient across categories of the polygenic risk score among Black participants. Black participants with intermediate and high polygenic risk scores had adjusted HRs of 1.18 (95% CI, 0.35–3.97) and 0.99 (95% CI, 0.22–4.50), respectively. There was no evidence of effect modification in the relationship between polygenic score and MI by sex among Black participants.

White participants in the intermediate and low polysocial score categories had a significantly higher incidence rate of MI than those with a high polysocial score, with adjusted HRs of 1.69 (95% CI, 1.10–2.59) and 2.64 (95% CI, 1.68–4.13), respectively (Figure 2). There was no evidence of effect modification in the relationship between polysocial score and MI by sex (Table S3).

Figure 2. Cumulative incidence rate of myocardial infarction by polygenic risk score and polysocial score.

Standardization was performed to the average for each covariate.25 Cox regression models were adjusted for age, sex, smoking, alcohol use, body mass index, and the first 5 principal components of ancestry. The parentheses include 95% CIs for the hazard ratios.

Table 2 presents the association between polysocial score and MI by each genetic risk category among White participants. In the low genetic risk category, people with a low polysocial score had a lower incidence rate of MI than those in the combined high and intermediate polysocial score category (4.4 versus 5.1 per 1000 person‐years). In the intermediate genetic risk category, people in the lower polysocial score category had a 5.1 higher incidence rate than those with a high/intermediate polysocial score (11.7 versus 6.6 per 1000 person‐years), and the difference in incidence rate increased to 6.8 in the high genetic risk category (14.7 versus 7.9 per 1000 person‐years). The polysocial score was a strong risk factor for MI among participants with intermediate and high genetic risk. Among people with intermediate genetic risk, a low polysocial score was associated with a 92% (95% CI, 34%–175%) higher hazard of MI than those with a high/intermediate polysocial score. Among people with high genetic risk, having a low polysocial score was associated with an adjusted HR of 1.78 (95% CI, 1.04–3.07).

Table 2. Association of Polysocial Score With Myocardial Infarction Across Polygenic Risk Score Categories Among White Participants

Polygenic risk score
Low genetic riskIntermediate genetic riskHigh genetic risk
Polysocial scoreHigh/intermediateLowHigh/intermediateLowHigh/intermediateLow
Events/sample size23/7156/24681/197856/90131/62726/328
Incidence rate*5.14.46.611.77.914.7
HR (95% CI)Reference0.90 (0.36–2.30)Reference1.92 (1.34–2.75)Reference1.78 (1.04–3.07)
P valueReference0.83Reference<0.001Reference0.04

HR indicates hazard ratio.

*Unadjusted incident rates were reported per 1000 person‐year.

Standardization was performed to the average for each covariate.25 Cox regression models were adjusted for age, sex, smoking, alcohol use, body mass index, and the first 5 principal components of ancestry.

Association of the Joint Effect of Genetic and Social Environmental Risks on MI

We present the absolute risk of incident MI for 8 years among White participants for each combination of genetic and social environmental risks in Figure 3. For participants with low/intermediate social environmental risk, participants with intermediate and high genetic risk had 3.5% and 4.3% higher risks of incident MI than those with low genetic risk, respectively. The risk of incident MI for participants with intermediate and high genetic risk in the high social environmental risk category was 7.1% and 8.0% higher than participants with low genetic risk in the low/intermediate social environmental risk category, respectively. Table 3 demonstrates the joint effect of genetic and social environmental risks on MI with the reference category being low genetic risk and low/intermediate social environmental risk. Among participants with intermediate genetic risk, those with high social environmental risk had a significantly higher (HR, 2.38 [95% CI, 1.43–3.95]) risk of incident MI than participants in the reference category, whereas those with low/intermediate social environmental risk did not have a significantly higher (HR, 1.28 [95% CI, 0.79–2.06]) risk of incident MI than the reference group. Similarly, among participants at high genetic risk, the risk of incident MI for those with high social environmental risk was significantly higher (HR, 2.87 [95% CI, 1.60–5.21]) compared with participants in the reference category, whereas the risk of incident MI for those with low/intermediate social environmental risk was not significantly higher (HR, 1.57 [95% CI, 0.90–2.73]) compared with the reference group. In the high polygenic score category, the joint effect of genetic and social environment risks on MI among female participants was stronger than male participants (Table S4). We found no significant additive or multiplicative interaction between genetic and social environment risks.

Figure 3. The 8‐year risk of incident myocardial infarction by polygenic risk score and polysocial score among White participants.

Standardization was performed using the average for each covariate.25 Cox regression models were adjusted for age, sex, smoking, alcohol use, body mass index, and the first 5 principal components of ancestry. The absolute risk represents the risk of myocardial infarction when all covariates in the multivariable model are set to their mean values. The bars indicate 95% CIs.

Table 3. Combined Effects of Genetic and Social Environmental Risks on Myocardial Infarction

Polygenic risk scorePolysocial scoreEvents/sample sizeIncidence rateHR (95% CI)
UnadjustedAdjusted*
Low genetic riskHigh/intermediate23/7155.1ReferenceReference
Low genetic riskLow6/2464.40.90 (0.37–2.21)0.90 (0.36–2.24)
Intermediate genetic riskHigh/intermediate81/19786.61.29 (0.81–2.06)1.28 (0.79–2.06)
Intermediate genetic riskLow56/90111.72.39 (1.47–3.88)2.38 (1.43–3.95)
High genetic riskHigh/intermediate31/6277.91.55 (0.91–2.67)1.57 (0.90–2.73)
High genetic riskLow26/32814.72.99 (1.71–5.24)2.87 (1.60–5.12)

HR indicates hazard ratio.

*Standardization was performed to the average for each covariate.25 Cox regression models were adjusted for age, sex, smoking, alcohol use, body mass index, and the first 5 principal components of ancestry.

Results were robust to several sensitivity analyses. First, dividing the combined intermediate/high polysocial score group did not significantly change the joint effect of genetic and social environmental risks on MI (Tables S5 and S6, Figures S2 and S3). Therefore, we found that a higher genetic susceptibility exacerbated the risk of incident MI, whereas a favorable social environment (an intermediate/high polysocial score) mitigated the risk of incident MI (Figure 4). Second, the results did not change appreciably after incorporating sampling weights in the analyses (Tables S7 through S12). In addition, when modeling the polygenic score as a continuous variable in the Cox model with penalized splines, we found a nonlinear relation between continuous polygenic score and incident MI among White participants (Figure S4) and an inverted U‐shape relation among Black participants (Figure S5). Results were similar with and without multivariable adjustment. Finally, using the Firth correction method did not appreciably change the association of polygenic score with MI (Tables S13 and S14) or additive/multiplicative interaction between genetic and social environment risks (Table S15).

Figure 4. Association of genetic susceptibility and polysocial score with 8‐year risk of incident myocardial infarction.

DISCUSSION

The aim of the present study was to understand the joint effect of genetic and social environmental risks in the development of incident MI among older Black and White adults. We found that higher genetic risk, quantified by a polygenic risk score, and unfavorable social environment, assessed by a polysocial score, were independently associated with a higher risk of incident MI among White participants. We did not find an obvious risk gradient of incident MI across categories of genetic or social environmental risk among Black participants; however, these unexpected findings should be interpreted with caution. Among White older adults in the intermediate and high genetic risk categories, an unfavorable social environment (low polysocial score) was significantly associated with a higher risk of incident MI. Among people with a low genetic risk, the difference in incident rate of MI between high/intermediate and low polysocial score substantially attenuated and was no longer significant. We also revealed the joint effect of genetic and social factors on MI; the incident risk of MI among people with a high genetic risk and low polysocial score was >3 times as much as those with a low genetic risk and high/intermediate polysocial score. Taken together, these findings suggest that living in a favorable social environment is particularly important for people with intermediate and high genetic risk for MI. The polysocial score approach provides a novel insight for understanding the joint effect of genetic and social environmental risks in the development of MI.

The cumulative effect of SNPs on MI has been well demonstrated using genetic risk scores. In the middle‐aged population, previous studies found the utility of polygenic risk score in identifying genetic risk of MI.25, 28 Recent studies revealed a similar role of polygenic risk score among the older population. Neumann et al29 and Menniti et al30 identified an independent association between polygenic risk score incident MI or coronary heart disease mortality for 5 years among older adults. We found that a polygenic risk score was independently associated with incident MI among White participants, which is in line with previous findings. We did not find an obvious risk gradient of incident MI across categories of genetic risk among Black participants. In the HRS, the genetic loci used to calculate the polygenic risk score were derived from GWAS on European ancestry and therefore might not be generalizable to other ancestry groups.31 A similar phenomenon was observed in a recent study examining the predictive value of polygenic risk score for type 2 diabetes in the HRS.32

Although the value of polygenic risk score has been increasingly recognized, the personal and clinical utility of an individual's genetic susceptibility to chronic conditions is limited. Recently, more attention has been paid to the synergistic effects of genetic susceptibility and modifiable factors with a focus on healthy lifestyles. Previous studies investigated the association of genetic risk and lifestyle with chronic diseases, including heart diseases and diabetes,14, 15, 25, 33 and found the correlation between unfavorable lifestyle and higher risk of chronic diseases across genetic risk categories. Among the studies, Said et al14 and Pazoki et al15 examined interactions between genetic risk and healthy lifestyles in relation to heart diseases; neither of the studies found a significant interaction between genetic and lifestyle factors. However, little is known about whether and how social environment interacts with genetic susceptibility in influencing the risk of MI. Our study provides evidence that social environment and genetic susceptibility jointly influence the risk of MI. The risk of incident MI among White participants with a high genetic risk and low polysocial score was >3 times as much as those with a low genetic risk and high/intermediate polysocial score. In addition, White participants with different genetic risks benefited differently from living in a favorable social environment. Our findings suggest that individuals with an intermediate or high genetic risk would benefit more from having a favorable social environment.

Prior studies also showed several social environmental factors associated with MI. Kim et al34 and Giurgescu et al13 showed that lower neighborhood social cohesion was associated with a higher risk of MI. A prospective cohort study and a meta‐analysis consistently showed that lower household income and education level enhanced the incidence rate of MI.8, 12 A cross‐sectional study in Spain identified the role of stress in the development of MI.35 Our findings were consistent with the previous investigations. We assessed the association of social environmental risk with MI using the polysocial score, which included neighborhood social cohesion, household income, education, and stress. Using a developed polysocial score,17 we not only found the association of a disadvantaged social environment with a high risk of incident MI in the intermediate and high genetic risk categories but also found the joint effect of intermediate/high genetic risk and high social environmental risk in the development of incident MI.

Structural racism, which refers to “the totality of ways in which societies foster racial discrimination, through mutually reinforcing inequitable systems,”36 is a fundamental driver of the persistent cardiovascular health disparities.37 In the present study, living in a more favorable social environment, indicated by a higher polysocial score, was associated with a lower risk of MI for both older Black and White adults. However, disproportionately more Black participants were in the low polysocial score category; thus, even if interventions targeting increasing polysocial score turn out to be effective in reducing disease risk, Black individuals might not be able to bootstrap themselves out of unfavorable social environments because of structural racism. Although the polysocial score, which quantifies the aggregate effects of multiple social environmental factors on health, could capture the multidimensional and interactive features of structural racism, it reflects the outcome and processing of structural racism instead of directly measuring the construct per se. Future studies should develop methodologically sound and theoretically driven approaches to accurately and precisely measure structural racism to achieve health equity.

Our study has several strengths. First, although previous studies have explored the interactions between genetic susceptibility and modifiable lifestyle factors, we undertook a new direction, focusing on whether and how social environment, which is modifiable as well, interacts with genetic variation. Future studies could extend this line of work to other chronic conditions. Second, although several social factors have been identified to be important in the development of MI in cross‐sectional and longitudinal studies, we adopted a novel tool—a polysocial score approach—to comprehensively quantify the aggregated effects of a wide range of social factors belonging to different domains. Lastly, the relatively large sample size enabled us to examine the joint effect of genetic variation and social factors among White participants.

We acknowledge several limitations. First, we used self‐reported data to ascertain MI status, which might be subject to misclassification; however, self‐reported MI has been shown to have a satisfactory agreement with medical record diagnoses.38 Second, changes in social factors over time were not considered in the present analyses. Future research is needed to examine how changes in the polysocial score influence the risk of MI. These findings would provide an answer to the question of whether improvement in the social environment could bring health benefits. Third, only 10% of Black participants were in the most favorable polysocial score category, which might limit the statistical power in analyzing the association of polysocial score with MI among Black participants. Similarly, because of the small sample size of Black participants, we might be underpowered to identify a risk gradient of MI across categories of polygenic risk score among Black participants. Future research might further explore whether there would be risk gradients of MI classified by polygenic risk score and polysocial score using a larger sample size of Black participants. In addition, because participants with a high genetic risk for MI might have died at an earlier age (<65 years), they were, therefore, not included in the present study. As a result, participants included in our work likely represent the healthier and more robust segment of the population, potentially leading to a more conservative estimate of the association between polygenic score and the risk of MI. Our results were echoed by a previous study that found an association between higher genetic risk and higher risk of early‐ and late‐onset coronary heart disease.39 Future research might consider including younger participants for a better representation of those with a high genetic risk of MI. Finally, compared with White participants, our study might only consist of Black participants with a relatively low susceptibility of MI. Because the risk of incident MI in middle‐aged Black people was significantly higher than White people,40 Black people with a high risk of MI may have been excluded because of prevalent MI or MI‐related premature death at early ages. At the same time, because the HRS polygenic score for MI was created using results from a meta‐analysis of 48 studies of mainly European, South Asian, and East Asian populations,6, 32 the polygenic score available in the HRS might inadequately represent the genetic determinants of MI among Black participants.41 Previous studies demonstrated that the HRS polygenic score for type 2 diabetes and the HRS non‐APOE (apolipoprotein E) polygenic score for Alzheimer's disease lacked predictive power among Black participants.32, 42 Similarly, our study showed that the HRS polygenic score did not play a significant role in predicting MI among Black participants. Nevertheless, our work serves as another piece of evidence suggesting that a greater inclusion of participants of African ancestry in genetic studies could significantly advance health and medical sciences.43 The polygenic score in the HRS would be routinely updated as meta‐analyses of GWAS are published. These updates, particularly from studies including a sufficiently large number of participants of African ancestry, would help encapsulate the underlying genetic risk of MI for Black older adults. This is a critical step to address racial disparities in health.

In summary, using an available polygenic risk score and a developed polysocial score, we found that a disadvantaged social environment was associated with a higher risk of incident MI among older White adults at intermediate and high genetic risk, but not those with a low genetic risk. In addition, we found a joint effect of intermediate/high genetic risk and disadvantaged social environment in the development of incident MI. Living in a favorable social environment is particularly important for people with intermediate and high genetic risk for MI. The polysocial score approach provides a novel insight for understanding the joint effect of genetic and social environmental risks in the development of MI. It is critical to develop tailored interventions to comprehensively improve the social environment for disease prevention, especially among adults with a relatively high genetic risk.

Sources of Funding

Duke Kunshan University Faculty Scholarship and Travel Award provided research funding.

Disclosures

None.

Footnotes

*Correspondence to: Chenkai Wu, PhD, MPH, MS, Global Health Research Center, Duke Kunshan University, Academic Building 3038, No. 8 Duke Avenue, Kunshan, Jiangsu, China 215316. Email:

*J. Tang and C. Sheng contributed equally.

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.122.028200

For Sources of Funding and Disclosures, see page 10.

References

  • 1 Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, et al. ACC/AHA guidelines for the management of patients with ST‐elevation myocardial infarction—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1999 guidelines for the Management of Patients with Acute Myocardial Infarction). Circulation. 2004; 110:588–636. doi: 10.1161/o1.CIR.0000134791.68010.FALinkGoogle Scholar
  • 2 NHANES ‐ National Health and Nutrition Examination Survey Homepage. 2022. Accessed July 5, 2022. https://www.cdc.gov/nchs/nhanes/index.htmGoogle Scholar
  • 3 Bokenberger K, Rahman S, Wang M, Vaez M, Dorner T, Helgesson M, Ivert T, Mittendorfer‐Rutz E. Work disability patterns before and after incident acute myocardial infarction and subsequent risk of common mental disorders: a Swedish cohort study. Sci Rep. 2019; 9:16086. doi: 10.1038/s41598-019-52487-wCrossrefMedlineGoogle Scholar
  • 4 Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: global burden of disease study. Lancet. 1997; 349:1498–1504. doi: 10.1016/S0140-6736(96)07492-2CrossrefMedlineGoogle Scholar
  • 5 Shiyovich A, Gilutz H, Arbelle JE, Greenberg D, Plakht Y. Temporal trends in healthcare resource utilization and costs following acute myocardial infarction. Isr J Health Policy Res. 2020; 9:6. doi: 10.1186/s13584-020-0364-yCrossrefMedlineGoogle Scholar
  • 6 Nikpay M, Goel A, Won HH, Hall LM, Willenborg C, Kanoni S, Saleheen D, Kyriakou T, Nelson CP, Hopewell JC, et al. A comprehensive 1000 genomes–based genome‐wide association meta‐analysis of coronary artery disease. Nat Genet. 2015; 47:1121–1130. doi: 10.1038/ng.3396CrossrefMedlineGoogle Scholar
  • 7 Mayer B, Erdmann J, Schunkert H. Genetics and heritability of coronary artery disease and myocardial infarction. Clin Res Cardiol. 2007; 96:1–7. doi: 10.1007/s00392-006-0447-yCrossrefMedlineGoogle Scholar
  • 8 Fretz A, Schneider AL, McEvoy JW, Hoogeveen R, Ballantyne CM, Coresh J, Selvin E. The association of socioeconomic status with subclinical myocardial damage, incident cardiovascular events, and mortality in the ARIC study. Am J Epidemiol. 2016; 183:452–461. doi: 10.1093/aje/kwv253CrossrefMedlineGoogle Scholar
  • 9 Hartiala JA, Han Y, Jia Q, Hilser JR, Huang P, Gukasyan J, Schwartzman WS, Cai Z, Biswas S, Tregouet DA, et al. Genome‐wide analysis identifies novel susceptibility loci for myocardial infarction. Eur Heart J. 2021; 42:919–933. doi: 10.1093/eurheartj/ehaa1040CrossrefMedlineGoogle Scholar
  • 10 Iribarren C, Lu M, Jorgenson E, Martínez M, Lluis‐Ganella C, Subirana I, Salas E, Elosua R. Weighted multi‐marker genetic risk scores for incident coronary heart disease among individuals of African, Latino and East‐Asian Ancestry. Sci Rep. 2018; 8:6853. doi: 10.1038/s41598-018-25128-xCrossrefMedlineGoogle Scholar
  • 11 Testa A, Jackson DB, Semenza DC, Vaughn MG. Food deserts and cardiovascular health among young adults. Public Health Nutr. 2021; 24:117–124. doi: 10.1017/S1368980020001536CrossrefMedlineGoogle Scholar
  • 12 Khaing W, Vallibhakara SA, Attia J, McEvoy M, Thakkinstian A. Effects of education and income on cardiovascular outcomes: a systematic review and meta‐analysis. Eur J Prev Cardiol. 2017; 24:1032–1042. doi: 10.1177/2047487317705916CrossrefMedlineGoogle Scholar
  • 13 Giurgescu C, Nowak AL, Gillespie S, Nolan TS, Anderson CM, Ford JL, Hood DB, Williams KP. Neighborhood environment and DNA methylation: implications for cardiovascular disease risk. J Urban Health. 2019; 96:23–34. doi: 10.1007/s11524-018-00341-1CrossrefMedlineGoogle Scholar
  • 14 Said MA, Verweij N, van der Harst P. Associations of combined genetic and lifestyle risks with incident cardiovascular disease and diabetes in the UK Biobank study. JAMA Cardiol. 2018; 3:693–702. doi: 10.1001/jamacardio.2018.1717CrossrefMedlineGoogle Scholar
  • 15 Pazoki R, Dehghan A, Evangelou E, Warren H, Gao H, Caulfield M, Elliott P, Tzoulaki I. Genetic predisposition to high blood pressure and lifestyle factors. Circulation. 2018; 137:653–661. doi: 10.1161/CIRCULATIONAHA.117.030898LinkGoogle Scholar
  • 16 Agerbo E, Sullivan PF, Vilhjalmsson BJ, Pedersen CB, Mors O, Borglum AD, Hougaard DM, Hollegaard MV, Meier S, Mattheisen M, et al. Polygenic risk score, parental socioeconomic status, family history of psychiatric disorders, and the risk for schizophrenia: a Danish population‐based study and meta‐analysis. JAMA Psychiatry. 2015; 72:635–641. doi: 10.1001/jamapsychiatry.2015.0346CrossrefMedlineGoogle Scholar
  • 17 Ping Y, Oddén MC, Stawski RS, Abdel Magid HS, Wu C. Creation and validation of a polysocial score for mortality among community‐dwelling older adults in the USA: the health and retirement study. Age Ageing. 2021; 50:2214–2221. doi: 10.1093/ageing/afab174CrossrefMedlineGoogle Scholar
  • 18 Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, Weir DR. Cohort profile: the Health and Retirement Study (HRS). Int J Epidemiol. 2014; 43:576–585. doi: 10.1093/ije/dyu067CrossrefMedlineGoogle Scholar
  • 19 Heeringa SG, Connor J H. Technical Description of the Health and Retirement Study Sample Design. Ann Arbor, MI: University of Michigan; 1995.CrossrefGoogle Scholar
  • 20 Latham K, Williams MM. Does neighborhood disorder predict recovery from mobility limitation? Findings from the Health and Retirement Study. J Aging Health. 2015; 27:1415–1442. doi: 10.1177/0898264315584328CrossrefMedlineGoogle Scholar
  • 21 Ware EB, Schmitz LL, Faul J, Gard A, Mitchell C, Smith JA, Zhao W, Weir D, Kardia SL. Heterogeneity in polygenic scores for common human traits. BioRxiv. 2017 Feb 5:106062. doi: 10.1101/106062CrossrefGoogle Scholar
  • 22 Lourida I, Hannon E, Littlejohns TJ, Langa KM, Hypponen E, Kuzma E, Llewellyn DJ. Association of lifestyle and genetic risk with incidence of dementia. JAMA. 2019; 322:430–437. doi: 10.1001/jama.2019.9879CrossrefMedlineGoogle Scholar
  • 23 Mega JL, Stitziel NO, Smith JG, Chasman DI, Caulfield M, Devlin JJ, Nordio F, Hyde C, Cannon CP, Sacks F, et al. Genetic risk, coronary heart disease events, and the clinical benefit of statin therapy: an analysis of primary and secondary prevention trials. Lancet. 2015; 385:2264–2271. doi: 10.1016/S0140-6736(14)61730-XCrossrefMedlineGoogle Scholar
  • 24 Tada H, Melander O, Louie JZ, Catanese JJ, Rowland CM, Devlin JJ, Kathiresan S, Shiffman D. Risk prediction by genetic risk scores for coronary heart disease is independent of self‐reported family history. Eur Heart J. 2016; 37:561–567. doi: 10.1093/eurheartj/ehv462CrossrefMedlineGoogle Scholar
  • 25 Khera AV, Emdin CA, Drake I, Natarajan P, Bick AG, Cook NR, Chasman DI, Baber U, Mehran R, Rader DJ, et al. Genetic risk, adherence to a healthy lifestyle, and coronary disease. N Engl J Med. 2016; 375:2349–2358. doi: 10.1056/NEJMoa1605086CrossrefMedlineGoogle Scholar
  • 26 Millett ERC, Peters SAE, Woodward M. Sex differences in risk factors for myocardial infarction: cohort study of UK Biobank participants. BMJ. 2018; 363:363. doi: 10.1136/bmj.k4247CrossrefGoogle Scholar
  • 27 Peters SAE, Colantonio LD, Chen L, Bittner V, Farkouh ME, Rosenson RS, Jackson EA, Dluzniewski P, Poudel B, Muntner P, et al. Sex differences in incident and recurrent coronary events and all‐cause mortality. J Am Coll Cardiol. 2020; 76:1751–1760. doi: 10.1016/j.jacc.2020.08.027CrossrefMedlineGoogle Scholar
  • 28 Inouye M, Abraham G, Nelson CP, Wood AM, Sweeting MJ, Dudbridge F, Lai FY, Kaptoge S, Brozynska M, Wang T, et al. Genomic risk prediction of coronary artery disease in 480,000 adults: implications for primary prevention. J Am Coll Cardiol. 2018; 72:1883–1893. doi: 10.1016/j.jacc.2018.07.079CrossrefMedlineGoogle Scholar
  • 29 Neumann JT, Riaz M, Bakshi A, Polekhina G, Thao LTP, Nelson MR, Woods RL, Abraham G, Inouye M, Reid CM, et al. Prognostic value of a polygenic risk score for coronary heart disease in individuals aged 70 years and older. Circulation. 2022; 15:e003429. doi: 10.1161/CIRCGEN.121.003429LinkGoogle Scholar
  • 30 Menniti G, Paquet C, Han HY, Dube L, Nielsen DE. Multiscale risk factors of cardiovascular disease: CLSA analysis of genetic and psychosocial factors. Front Cardiovasc Med. 2021; 8:599671. doi: 10.3389/fcvm.2021.599671CrossrefMedlineGoogle Scholar
  • 31 Martin AR, Gignoux CR, Walters RK, Wojcik GL, Neale BM, Gravel S, Daly MJ, Bustamante CD, Kenny EE. Human demographic history impacts genetic risk prediction across diverse populations. Am J Hum Genet. 2017; 100:635–649. doi: 10.1016/j.ajhg.2017.03.004CrossrefMedlineGoogle Scholar
  • 32 Wu YY, Thompson MD, Youkhana F, Pirkle CM. Interaction between physical activity and polygenic score on type 2 diabetes mellitus in older Black and White participants from the Health and Retirement Study. J Gerontol A Biol Sci Med Sci. 2021; 76:1214–1221. doi: 10.1093/gerona/glab025CrossrefMedlineGoogle Scholar
  • 33 Merino J, Guasch‐Ferré M, Li J, Chung W, Hu Y, Ma B, Li Y, Kang JH, Kraft P, Liang L, et al. Polygenic scores, diet quality, and type 2 diabetes risk: an observational study among 35,759 adults from 3 US cohorts. PLoS Med. 2022; 19:e1003972. doi: 10.1371/journal.pmed.1003972CrossrefMedlineGoogle Scholar
  • 34 Kim ES, Hawes AM, Smith J. Perceived neighbourhood social cohesion and myocardial infarction. J Epidemiol Community Health. 2014; 68:1020–1026. doi: 10.1136/jech-2014-204009CrossrefMedlineGoogle Scholar
  • 35 Pais JL, Coronel BI, Gil DG, Pascual MJE, Peredo CGM, García PA, Magaña JG, Caballero RM, Sanz AF, Muñiz J, et al. Psycho‐emotional disorders as incoming risk factors for myocardial infarction with non‐obstructive coronary arteries. Cardiol J. 2018; 25:24–31. doi: 10.5603/CJ.a2017.0139CrossrefMedlineGoogle Scholar
  • 36 Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. Lancet. 2017; 389:1453–1463. doi: 10.1016/S0140-673630569-XCrossrefMedlineGoogle Scholar
  • 37 Churchwell K, Elkind MS, Benjamin RM, Carson AP, Chang EK, Lawrence W, Mills A, Odom TM, Rodriguez CJ, Rodriguez F, et al. Call to action: structural racism as a fundamental driver of health disparities: a presidential advisory from the American Heart Association. Circulation. 2020; 142:e454–e468. doi: 10.1161/CIR.0000000000000936LinkGoogle Scholar
  • 38 Okura Y, Urban LH, Mahoney DW, Jacobsen SJ, Rodeheffer RJ. Agreement between self‐report questionnaires and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure. J Clin Epidemiol. 2004; 57:1096–1103. doi: 10.1016/j.jclinepi.2004.04.005CrossrefMedlineGoogle Scholar
  • 39 Mars N, Koskela JT, Ripatti P, Kiiskinen TT, Havulinna AS, Lindbohm JV, Ahola‐Olli A, Kurki M, Karjalainen J, Palta P, et al. Polygenic and clinical risk scores and their impact on age at onset and prediction of cardiometabolic diseases and common cancers. Nat Med. 2020; 26:549–557. doi: 10.1038/s41591-020-0800-0CrossrefMedlineGoogle Scholar
  • 40 Jurgens PT, Carr JJ, Terry JG, Rana JS, Jr. Jacobs DR, Duprez DA. Association of abdominal aorta calcium and coronary artery calcium with incident cardiovascular and coronary heart disease events in Black and White middle‐aged people: the Coronary Artery Risk Development in Young Adults Study. J Am Heart Assoc. 2021; 10:e023037. doi: 10.1161/JAHA.121.023037LinkGoogle Scholar
  • 41 Freedman BI, Divers J, Palmer ND. Population ancestry and genetic risk for diabetes and kidney, cardiovascular, and bone disease: modifiable environmental factors may produce the cures. Am J Kidney Dis. 2013; 62:1165–1175. doi: 10.1053/j.ajkd.2013.05.024CrossrefMedlineGoogle Scholar
  • 42 Bakulski KM, Vadari HS, Faul JD, Heeringa SG, Kardia SL, Langa KM, Smith JA, Manly JJ, Mitchell CM, Benke KS, et al. A non‐APOE Polygenic score for Alzheimer's disease and APOE‐ε4 have independent associations with dementia in the Health and Retirement Study. medRxiv. 2020 Feb 11:2020‐02. doi: 10.1101/2020.02.10.20021667CrossrefGoogle Scholar
  • 43 Oni‐Orisan A, Mavura Y, Banda Y, Thornton TA, Sebro R. Embracing genetic diversity to improve Black health. N Engl J Med. 2021; 384:1163–1167. doi: 10.1056/NEJMms2031080CrossrefMedlineGoogle Scholar
  • 44 Bugliari D, Carroll J, Hayden O, Hayes J, Hurd M, Karabatakis A, Main R, Marks J, McCullough C, Meije E, et al. RAND HRS Longitudinal File 2018 (V1) Documentation. RAND Center for the Study of Aging; 2021.Google Scholar

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.