Associations of cardiovascular health with arterial health in adults free of cardiovascular disease

Abstract

Background

In adults without cardiovascular disease (CVD), there is limited understanding of the association between overall cardiovascular health (CVH) and arterial health.

Methods

In 2330 Framingham Heart Study Offspring participants free of CVD (60±9 years; 57% women) with Life’s Essential 8 (LE8) and applanation tonometry data (Exam 7), we calculated CVH scores per American Heart Association’s LE8 guidelines. Multivariable-adjusted regression analyses examined the relations of LE8 with aortic stiffness and pressure pulsatility [negative inverse carotid-femoral pulse wave velocity (niCFPWV), central pulse pressure (CPP), respectively], and examined effect modification by age and sex. We also evaluated niCFPWV and CPP as mediators of the relation between LE8 and death outcomes (CVD, all-cause mortality).

Results

Higher LE8 scores (better CVH) were associated with lower niCFPWV [standardized (std) β= -0.20±0.02, p<0.0001] and CPP (std β= -0.11±0.02, p<0.0001). While age- and sex- interactions were not significant, stratified analysis revealed stronger association of LE8 with arterial health in women (niCFPWV: std β= -0.11±0.02, p<0.0001 vs. std β= -0.06±0.03, p=0.04; CPP: std β= -0.13±0.03, p<0.001 vs. std β= -0.06±0.03, p=0.07 in women vs. men, respectively). niCFPWV and CPP mediated 19% and 10% of the association between LE8 and CVD mortality, respectively, and 17% and 15% of the association between LE8 and all-cause mortality, respectively.

Conclusion

Better CVH measured by LE8 was associated with lower arterial stiffness and pressure pulsatility, both of which mediate a significant proportion of the associations between CVH and CVD/death outcomes. These findings underscore the importance of optimal cardiovascular health behaviors and factors in maintaining arterial health.

1. Introduction

Cardiovascular disease (CVD) remains the leading cause of death both in the United States[1] and globally[2], despite declining trends in age-standardized death rates from CVD and stroke. The persistence of suboptimal cardiovascular health (CVH) metrics across all age groups in the United States, with few maintaining favorable CVH, likely contributes to this issue[3]. The association of single risk factors with CVH has been well demonstrated[[4], [5], [6], [7], [8], [9]]. However, because risk factors do not function in isolation, efforts have turned to the evaluation of the joint influence of multiple risk factors. Evidence from several studies suggests an inverse relation between LE8 scores and poor outcomes, indicating that higher LE8 scores are associated with lower risks of CVD and mortality, while consistently low scores confer the highest risks[[10], [11], [12], [13], [14], [15]]. As an extension of composite risk factors including lifestyle for CVD risk, the American Heart Association's Life's Essential 8 (LE8) was created to amalgamate key health behaviors and CVD risk factors including diet (updated), physical activity, nicotine exposure (updated), sleep health (new), body mass index (BMI), blood lipids (updated), blood glucose (updated), and blood pressure[16].

High arterial stiffness and pressure pulsatility, defining characteristics of poorer vascular health[[17], [18], [19]], are leading contributors to CVD-related morbidity and mortality[[20], [21], [22], [23]]. While individual CVH factors, such as blood pressure, are linked to vascular health[4,24], the relationship between overall CVH, as captured by the LE8 score, and vascular health has been less extensively studied. Addressing this gap, the objective of our study is to (a) explore the relation between LE8 score and markers of arterial health and aging, such as carotid-femoral pulse wave velocity (niCFPWV) and central pulse pressure (CPP); (b) quantify the mediation effect of niCFPWV and CPP on the relation between LE8 and CVD/all-cause mortality outcomes, in individuals free of CVD from the Framingham Heart Study (FHS) Offspring cohort.

2. Methods

2.1. Study sample

The FHS was established in 1948 by the National Heart, Lung, and Blood Institute (NHLBI) to identify risk factors for heart disease[25]. The Offspring Cohort includes the children of the original Framingham cohort, as well as their spouses, who were enrolled in 1971 and followed subsequently with repeated examinations every 4–8 years[26]. For this investigation, we included participants of the FHS Offspring cohort who attended exam 7 (1998–2001) and had assessments of lifestyle measures and applanation tonometry. Of the 3002 participants, we excluded participants with prevalent CVD (n = 303, defined as coronary artery disease, heart failure, stroke, or peripheral artery disease) and missing LE8 or tonometry variables (n = 369), resulting in our final sample size of 2330 participants (Fig. 1). No additional exclusions were made for chronic conditions such as obstructive sleep apnea, chronic kidney disease, or inflammatory disorders (e.g., systemic lupus erythematosus, rheumatoid arthritis, vasculitis). The study protocol was approved by the Institutional Review Board of Beth Israel Deaconess Medical Center. The study included participants of the FHS Offspring cohort, with research conducted in accordance with the principles outlined in the Declaration of Helsinki. All participants provided written informed consent.

Fig. 1.

Sample selection. Flowchart of final sample size after exclusion of individuals with prevalent CVD and missing variables.

2.2. Measurement of lifestyle and CVH factors

The study sample comprised individuals aged 33 to 86 years who completed questionnaires containing questions about food frequency, physical activity, sleep duration, and nicotine exposure. Objective measures of BMI, seated blood pressure after a 5-minute rest, plasma total cholesterol, and fasting blood glucose levels were also evaluated. Each factor is rated on a scale from 0 to 100, and these scores are averaged to reflect overall CVH[16]. CVH was categorized based on the total score of 0–100, with a higher score indicating better CVH. LE8 was analyzed both as a continuous variable (total score) and by tertiles (< 63, 63–74, and ≥75 defining low, intermediate, and high CVH, respectively), due to unequal distribution when categorized according to the AHA LE8 groups (Supplemental Table 1).

2.3. Measures of arterial stiffness and health

Applanation tonometry was used to measure non-invasive arterial hemodynamic properties, with the electrocardiogram R-wave serving as a reference point. The systolic and diastolic cuff pressures were averaged and used to calibrate the peak and trough of the brachial artery pressure waveform. Brachial artery mean and diastolic pressure were used to calibrate pressure waveforms in other parts of the body, including the carotid (neck), radial (wrist), and femoral (thigh) arteries. Calibrated pressure in the neck was used to estimate the pressure in the heart. The systolic ejection period and reflected wave transit time were measured from the blood pressure waveforms in the neck, and the effective reflecting distance was calculated from these measurements. CPP was calculated using the carotid pressure waveform[17]. niCFPWV was assessed from carotid and femoral tonometry and body surface measurements, accounting for parallel transmission by taking the difference between distances from the suprasternal notch to femoral and carotid sites, with previously demonstrated excellent reproducibility (intraclass correlation coefficients of 0.93 to 0.95 for repeated measures and inter-observer correlations of 0.97 or higher)[17,18,27].

2.4. Statistical analysis

The Shapiro-Wilk test was employed to evaluate the normality of the continuous variables. To normalize the distribution of niCFPWV, this variable was transformed to negative inverse PWV inverse-transformed [1/(raw CFPWV)], then multiplied by −1000 to correct for the direction of effect, as in previous studies[28]. Continuous variables are presented as mean ± standard deviation (SD) and categorical variables as percentages. We used linear regression models to analyze continuous and categorical CVH (LE8 score), respectively. For the analysis of continuous variables, both independent (LE8 score) and dependent variables (niCFPWV, CPP) were standardized to a mean of 0 and SD of 1 to facilitate comparisons across analyses. Results for continuous outcomes are presented as the effect size ± standard error on niCFPWV or CPP, per standard deviation unit increment in LE8. Categorical analysis was conducted using low CVH (LE8 score <64) as the reference group. We first adjusted for age and sex and separately adjusted for additional confounding factors including heart rate and mean arterial pressure (MAP). Although MAP and CPP are mathematically intertwined, they reflect different aspects of hemodynamics: MAP represents the steady-state (resistance-related) pressure influenced by overall blood flow and resistance, while CPP reflects the pulsatile load, driven by arterial compliance and stroke volume. Adjusting for MAP allows isolation of the association between LE8 and pulsatile arterial load, without confounding from the resistance load[[29], [30], [31], [32]]. This approach aligns with previous research investigating pulse pressure and central hemodynamics in arterial stiffness[[33], [34], [35]]. We conducted analyses to evaluate for effect modification by age and sex in the relations between LE8 and arterial health measures. To address multicollinearity issues identified in the interaction terms, we centered LE8 by subtracting its mean value to reduce correlation between the main effects and the interaction terms. Variance inflation factors were calculated to assess multicollinearity. Recognizing the demonstrated biological differences[36] in arterial stiffness with sex, we additionally analyzed by sex-stratified results of the relations of lifestyle to arterial health. To understand the contributions of individual risk factors within the composite LE8 variable, we also examined the relation of each individual LE8 component to niCFPWV and CPP. Finally, we conducted analyses to evaluate the extent to which abnormal arterial hemodynamics mediate the relation of CVH factors with CVD mortality and all-cause mortality. These analyses were based on a counterfactual framework using causalmed procedure in SAS 9.4, with a binary outcome (all-cause mortality; CVD-cause mortality), continuous mediator (niCFPWV or CPP), and logit and identity links, respectively. We estimated the natural direct effect (NDE), natural indirect effect (NIE), total effect (TE), and the proportion mediated, while adjusting for age, sex, heart rate, and mean arterial pressure. Confidence intervals and p-values were computed using the delta method. This approach is consistent with causal mediation methodology described in the epidemiological literature[[37], [38], [39], [40]]. Only participants with complete data were eligible for the analysis. Participants with missing values for any variable involved in the exposure, mediator, outcome, or covariate set were excluded from the relevant models. No imputation was performed. The threshold for statistical significance was set at two-sided p < 0.05. Statistical analyses were completed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

3. Results

Baseline demographics and participant characteristics are outlined in Table 1. At the FHS Exam 7 study visit, the study cohort included 2330 participants, categorized into low, intermediate, and high CVH tertiles (n = 781, n = 795, and n = 754, respectively). Participants were middle-aged and older adults (60±9 years), with no significant age difference among the CVH groups. Women represented 57 % of the overall sample, with higher proportions in the high (69 %) and intermediate (53 %) CVH groups compared to the low CVH group (49 %). Individuals in the high CVH group demonstrated more favorable health metrics, including lower mean BMI, blood glucose, blood pressure, total cholesterol, LDL levels, and better diet scores compared to those in the low and intermediate CVH groups. The LE8 components with the highest percentage of participants achieving the maximum score of 100 were blood glucose (79 %), sleep (74 %), and smoking (57 %). Conversely, only a very small percentage of individuals (5 %) demonstrated a diet score of 100 (Fig. 2). Mean estimated glomerular filtration rate was 89.2 ± 14.9 mL/min/1.73 m² in the overall sample and remained similar across CVH categories. C-reactive protein levels, reported as median (interquartile range), were highest among participants with low CVH [3.3 (1.5, 6.8) mg/L] and progressively lower in those with intermediate [2.0 (1.0, 4.2) mg/L] and high CVH [1.4 (0.7, 3.1) mg/L] status (Table 1).

Table 1.

Characteristics Of Study Sample by Cardiovascular Health Group.

Variable	Full Sample (n = 2330)	Low CVH (n = 781)	Intermediate CVH (n = 795)	High CVH (n = 754)
Age, years	60 ± 9	60±9	60 ± 10	60 ± 10
Women	57	49	53	69
Current smoking	14	27	10	3
Former smoking	35	41	37	28
Never smoking	51	32	53	69
Sleep hours	7.2 ± 1.0	7.2 ± 1.8	7.31 ± 0.91	7.32 ± 0.86
BMI, kg/m2	28 ± 5	31 ± 6	28 ± 4	25 ± 3
Blood glucose, mg/dl	101 ± 21	109 ± 28	100 ± 16	94 ± 11
SBP, mmHg	126 ± 19	133 ± 18	128 ± 18	118 ± 16
DBP, mmHg	74 ± 10	78 ± 10	75 ± 10	71 ± 8
Total cholesterol, mg/dl	202 ± 36	213 ± 37	204 ± 34	190 ± 31
LDL, mg/dl	122 ± 32	132 ± 34	124 ± 31	109 ± 28
Heart Rate, bpm	65 ± 11	68 ± 11	65 ± 10	63 ± 10
Mean Arterial Pressure, mmHg	92 ± 12	96 ± 13	92 ± 12	87 ± 11
eGFR, mL/min/1.73 m²	89.2 ± 14.9	89.4 ± 15.3	88.7 ± 15	89.4 ± 14.3
CRP, mg/L	2 (1, 4.8)	3.3 (1.5, 6.8)	2 (1, 4.2)	1.4 (0.7, 3.1)
Diabetes Medication	5	10	2	1
Hypertension Medication	28	37	28	19
CFPWV, m/s	9.6 ± 3.3	10.5 ± 3.7	9.8 ± 3.5	8.7 ± 2.6
CPP, mmHg	51 ± 16	55 ± 17	51 ± 16	49 ± 15

Continuous variables are presented as mean ± standard deviation (SD), except for C-reactive protein (CRP), which is reported as median (interquartile range) due to its skewed distribution, categorical variables are expressed as percentage ( %), eGFR= estimated glomerular filtration rate, CPP = central pulse pressure (mmHg), CFPWV = carotid femoral pulse wave velocity (m/s), SBP= systolic blood pressure, DBP = diastolic blood pressure, Low CVH group = <63 score, Intermediate CVH group = 63–74 score, High CVH group ≥75 score.

Fig. 2.

Scores achieved in each LE8 component. The graph depicts the distribution of scores among participants for each LE8 component, which is assessed using the published AHA scoring scale.

3.1. Relations of LE8 with niCFPWV

Age- and sex-adjusted regression analyses assessing the cross-sectional relations between continuous LE8 scores and niCFPWV demonstrated that higher LE8 score was associated with lower niCFPWV [standardized (std) β = −0.24±0.02, p < 0.0001)]. We observed a similar trend in multivariable models adjusting for additional confounders, where a higher LE8 score, indicating better CVH, was associated with lower niCFPWV, consistent with more favorable vascular health (std β = −0.10±0.02, p < 0.0001) (Table 2).

Table 2.

Association of LE8 score with vascular health measures.

	niCFPWV				CPP
		β ± SE	p-value	β ± SE		p-value
Continuous LE8
Age-, sex-adjusted*		−0.24±0.02	<0.0001	−0.18±0.02		<0.0001
Multivariable-adjusted*		−0.10±0.02	<0.0001	−0.11±0.02		<0.0001
Categorical LE8 (Low CVH = referent)
Age-, sex-adjusted
High CVH		−0.52±0.05	<0.0001	−0.41±0.05		<0.0001
Intermediate CVH		−0.25±0.05	<0.0001	−0.22±0.05		<0.0001
Multivariable-adjusted
High CVH		−0.21±0.04	<0.001	−0.23±0.05		<0.0001
Intermediate CVH		−0.09±0.04	0.017	−0.14±0.05		0.002

Variables in multivariable model: age, sex, heart rate and mean arterial pressure, niCFPWV = carotid femoral pulse wave velocity ; niCFPWV = (1/raw CFPWV)*(−1000) to normalize the skewed distribution and correct the direction of effect and then standardized, CPP = central pulse pressure; Continuous LE8 β = standardized β presented as results per standard deviation unit (SDU) change in outcome variable per SDU change in LE8 score; Categorical LE8 β = represents the difference in the outcome (in SDU) for individuals in the intermediate or high CVH category compared to the low CVH reference group; score thresholds: high CVH ≥ 75, intermediate CVH = 63–74, low CVH ≤ 63; p-value for trend across LE8 groups, niCFPWV: p = 0.0002; CPP: p = 0.0009.

We next related CVH groups to vascular stiffness, with the lowest LE8 tertile (LE8 score<63.75) serving as the referent group. In analyses adjusting for age and sex, both intermediate and high tertiles were associated with lower effect size for niCFPWV (std β= −0.25±0.05, p < 0.0001, and std β= −0.52±0.05, p < 0.0001, respectively). In multivariable models, a similar trend was observed, individuals with intermediate and high CVH had lower niCFPWV (std β = - 0.09±0.04, p = 0.017; std β = −0.21±0.04, p < 0.001, respectively) (Table 2).

3.2. Relations of LE8 with CPP

We evaluated the association between LE8 and central pressure pulsatility, measured by CPP. Higher LE8 score was associated with lower CPP in both age- and sex-, and multivariable-adjusted models (std β = −0.18±0.02 and −0.11±0.02, respectively, both p < 0.0001). In categorical analysis with age- and sex-adjustment, participants with intermediate CVH had lower CPP compared to the lowest tertile of LE8 (std β = −0.22±0.05, p < 0.0001). Individuals with a high CVH score also had lower CPP compared to the low CVH group (std β = −0.41±0.05, p < 0.0001). The association between higher CVH scores with lower pressure pulsatility persisted in multivariable-adjusted analyses for both intermediate and high tertiles (compared to low CVH tertile, intermediate CVH: std. β= −0.14±0.05, p = 0.002 and high CVH std β= −0.23±0.05, p < 0.0001, respectively) (Table 2). Evaluating standardized results allows comparison of the magnitude of effects across CFPWV and CPP. Results per unit change in LE8 (non-standardized results) similarly demonstrate lower CFPWV and CPP with higher LE8 and are presented in Supplemental Table 2.

3.3. Age and sex in LE8—arterial health relations

We did not observe effect modification by age (niCFPWV: p = 0.67; CPP: p = 0.35) or by sex (niCFPWV: p = 0.51; CPP: p = 0.31). However, based upon a priori hypothesis that lifestyle and risk factors may have differential associations with vascular hemodynamics between men and women, we conducted sex-stratified analyses of LE8-arterial health measures. Though exploratory, we observed greater effect sizes and significance in the relations between LE8 and arterial health in women compared with men (niCFPWV: std β= −0.11±0.02, p < 0.0001 in women vs. −0.06±0.03, p = 0.04 in men, and CPP, std β=−0.13±0.03, p < 0.001 in women vs. −0.06±0.03, p = 0.07 in men).

3.4. Relations between individual risk factors and arterial health

We analyzed the association of individual LE8 factors with aortic stiffness (Supplemental Table 3). Blood pressure exhibited the largest effect size with both niCFPWV (std β = −0.13±0.02, p < 0.0001) and CPP (std β = −0.21±0.02, p < 0.0001). Weaker significant associations were present for body mass index (niCFPWV: std β = −0.04±0.02, p = 0.04 and CPP: std β = −0.07±0.02, p < 0.001) and blood glucose (std β for niCFPWV = −0.05±0.02, p = 0.008). Diet, physical activity, smoking, sleep health, and lipids were not significantly associated with either niCFPWV or CPP.

3.5. Arterial health as mediator of relations between LE8 and mortality

There were 346 CVD deaths and 507 all-cause deaths. Associations of LE8 with aortic stiffness and pressure pulsatility, and with the outcomes of CVD mortality and all-cause mortality are presented in Fig. 3. After adjusting for confounding covariates (age, sex, heart rate, and mean arterial pressure), niCFPWV and CPP respectively mediated 21 % and 10 % of the effect of LE8 on CVD mortality and 19 % and 17 % of the effect of LE8 on all-cause mortality.

Fig. 3.

Arterial health as mediators of the relations between LE8 with incident outcomes. ab represents the natural indirect effect of LE8 via arterial health on the outcomes of either CVD mortality or all-cause mortality. c represents the natural direct effect odds ratio between LE8 with the outcomes of CVD mortality or all-cause mortality.

4. Discussion

In our community-based study of adults free of CVD, we examined the relation of several lifestyle and CVH factors encompassed by AHA’s LE8 metric, with aortic stiffness and pressure pulsatility, which are measures of arterial health and aging. In multivariable-adjusted analyses, higher (more favorable) LE8 scores were associated with lower niCFPWV and CPP. While exploratory stratified analyses were conducted by age and sex, only sex showed nominal differences in associations between LE8 and arterial health. However, formal interaction testing was not statistically significant for either factor. Additionally, niCFPWV and CPP significantly mediated the effects of LE8 on CVD mortality. niCFPWV is a strong predictor of all-cause mortality[21,41] and both niCFPWV and CPP mediated a sizeable proportion of the relationship between LE8 and all-cause mortality. Analysis of individual LE8 factors demonstrated the strongest association of niCFPWV and CPP with blood pressure. Overall, our findings are supportive of the hypothesis that healthier levels of multiple modifiable factors are associated with more favorable CVH.

Our findings both corroborate and extend the existing literature on the relations between CVH indices and arterial stiffness. Previous investigations have largely evaluated the association of individual risk factors, such as blood pressure or smoking, with vascular stiffness[4,[42], [43], [44], [45], [46], [47], [48]]. Other studies have demonstrated that comprehensive CVH scores that combine risk factors, including AHA’s Life’s Simple 7, are associated with arterial stiffness[[37], [38], [39], [40], [41]]. We observed that these relations persist in the updated and detailed LE8 framework for CVH.

We did not observe effect modification by age in the relations between LE8 and arterial health measures. Arterial stiffness and higher pressure pulsatility increase with age due to structural and functional changes in the arterial wall, such as collagen accumulation, elastin fragmentation, and abnormal endothelial function[[49], [50], [51]]. The relationship of CVH with arterial health by age has varied in samples studied. While some studies have reported detecting relations between CVH and arterial health more in younger compared with older individuals[52,53], others have demonstrated that the association between better CVH and lower arterial stiffness indices was more marked in older adults[54]. The discrepancies observed between studies highlight potential variations in the impact of CVH across different populations and may be attributed to factors such as genetic predispositions, lifestyle behaviors, or methodological differences in the measurement of CVH or arterial health. Larger, diverse population samples utilizing longitudinal designs may be better suited to dissect these relations further.

Our analysis did not indicate sex as an effect modifier in the relationship between LE8 and arterial health. However, in the sex-stratified analysis, potential differences emerged, with women showing a stronger association between LE8 and both niCFPWV and CPP, whereas the relationship between LE8 and CPP was not significant in men. While data are limited on sex differences between overall CVH and arterial health, sex-specific analyses using individual risk factors have been conducted, suggesting differential associations between certain CVD risk factors in women (e.g. obesity, dyslipidemia, hyperglycemia) as compared with men (blood pressure, smoking), though these findings may be influenced by the prevalence of these conditions by sex[[55], [56], [57]]. While our results suggest that women may derive greater vascular health benefits from favorable CVH, these findings require further confirmation. Given the higher proportion of women across low to high CVH categories in our study, we cannot exclude influence in the results due to sample size differences between the sexes among the CVH groups. Nevertheless, these results should be interpreted as exploratory, given the non-significant interaction terms and the potential for limited statistical power to detect modest effect modification. Future studies in larger and more diverse populations are needed to determine age- and sex-based effects, if any, of the relations between cardiovascular health and arterial stiffness and pulsatility.

Limited prior work has quantified the role of arterial stiffness as a mediator in the effects of modifiable risk factors on CVD outcomes. Arterial stiffness only mediated 9 % of the association between LE8 and stroke in the large Kailuan I and II cohorts [58], while this mediation was nearly 24 % in another study of 15,297 individuals including all LE8 measures except for sleep[59]. These differences may relate to variations in the assessment of CVH and the baseline health profiles of various samples. The role of arterial stiffness in mediating the risk of similar lifestyle and risk factors with CVD measures other than stroke has not been well studied. In this study, we demonstrate that meaningful proportions of both CVD mortality and all-cause mortality may be attributable to the effect of unfavorable arterial stiffness and pressure pulsatility resulting from suboptimal CVH. The observed significant indirect effects through these arterial stiffness measures highlight their roles as crucial intermediaries in the pathway from cardiovascular health to all-cause mortality and CVD mortality. Together with prior studies, the collective evidence suggests that modifiable risk factors and their impact on arterial health play significant roles in not only cardiovascular but also all-cause mortality.

Our findings have significant public health and clinical implications. A contemporary NHANES study reported a mean LE8 score of 65 (95 % CI, 63.9–65.6) and 66 (95 % CI, 64.4–66.6) for adults and children, respectively, consistent with moderate CVH according to the AHA’s LE8 categorization and thus, suboptimal cardiovascular health across the population, with 17.9 % (≈30 million) having CVH scores of <50 (3). Scores were particularly low for diet, physical activity, and body mass index metrics. Given our results on the connection of modifiable and lifestyle factors with arterial health and outcomes in this context, there is a critical need to target interventions improving CVH and address the specific areas where health behaviors and factors are most suboptimal.

4.1. Limitations

Some limitations deserve consideration in our study. While the AHA LE8 score is comprehensive through encompassing 8 key modifiable lifestyle and CVH factors, it weighs CVH scores across categories similarly, though these factors may have differing degrees of detail in their measurements. This is an intrinsic limitation of the score, and further advancements may enable more granularity in the quantification of CVH. In addition, several LE8 components including diet, sleep, and physical activity are based on self-report, which may introduce recall or social desirability bias. This could partially explain the high proportion of participants with perfect LE8 scores and may attenuate true associations. Moreover, while the transition from LS7 to LE8 offers a more comprehensive evaluation of cardiovascular health through the addition of sleep, information on sleep is only partially captured by self-reported nightly hours, and comparative studies that reconcile the findings based on these two sets of criteria are limited. Additionally, while we controlled for several variables, there may still exist unmeasured confounders that could affect the associations observed. Ultimately, the complexity of CVD and all-cause mortality involves a multitude of physiological and genetic factors not measured or able to be addressed in this study. Finally, the Framingham Heart Study cohort participants are predominantly White and of European descent, which limits the generalizability of our findings to broader and more diverse populations. This homogeneity restricts our ability to assess how the associations between LE8 scores and measures of arterial stiffness and pressure pulsatility might differ by race or ethnicity. Prior research has demonstrated that social determinants of health, healthcare access, environmental exposures, and genetic background can all influence cardiovascular risk profiles and arterial aging trajectories[[60], [61], [62], [63], [64]]. As a result, the strength, direction, or even presence of the associations we observed may differ in populations with different ancestral backgrounds or lived experiences. Generalizing these findings without validation in racially and ethnically diverse cohorts could potentially obscure subgroup differences or lead to inequitable health recommendations. Thus, replication of our analyses in multiethnic cohorts is essential to better inform population-level cardiovascular health strategies that are inclusive and equitable.

5. Conclusions

Higher LE8 scores, reflecting better CVH, are associated with lower arterial stiffness and pulsatility, both of which are significant mediators of CVD and all-cause mortality. Given the high proportion of people having suboptimal CVH in the general U.S. population, our study findings have public health implications and support measures to improve CVH and arterial health in reducing poor outcomes.

Funding

This study was supported by NHLBI contracts N01-HC-25,195, HHSN268201500001l, and 75N92019D00031 (R.S.V) and by DK080739 (R.S.V), HL107385, HL126136, HL93328, HL142983, HL143227, HL131532 and AG079390 (R.S.V., G.F.M.), and HL60040, HL70100, HL092577, 2U54HL120163, AG066010 (E.J.B.), and R01HL155717 (C.W.T.). E.J.B was supported in part by R01HL092577 and R01HL60040.

CRediT authorship contribution statement

Mawra Jha: Writing – review & editing, Writing – original draft, Methodology, Conceptualization. Zhiyong Dong: Formal analysis. Maria M. Ruda: Writing – review & editing. Brenton Prescott: Formal analysis. Vanessa Xanthakis: Writing – review & editing. Matthew Nayor: Writing – review & editing. Priya Gajjar: Writing – review & editing. Martin G. Larson: Writing – review & editing. Emelia J. Benjamin: Writing – review & editing, Supervision. Ramachandran S. Vasan: Writing – review & editing, Supervision. Gary F. Mitchell: Writing – review & editing, Supervision. Connie W. Tsao: Writing – review & editing, Supervision.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Gary F. Mitchel, MD is the owner of Cardiovascular Engineering, Inc., a company that designs and manufactures devices that measure vascular stiffness. The company uses these devices in clinical trials that evaluate the effects of diseases and interventions on vascular stiffness. G.F.M. also serves as a consultant to and receives grants and honoraria from Novartis, Merck, Bayer, Servier, Philips, and deCODE genetics. G.F.M. is a co-inventor on 2 pending patent applications that disclose methods for predicting various measures of biological age from a pressure waveform by using a convolutional neural network. The remaining authors have no disclosures to report.