Comparing the “Life's Essential 8” Scores of Older Adults Living With Cardiovascular Diseases: NHANES, 2013 to 2018

Abstract

Background

The American Heart Association's Life's Essential 8 (LE8) framework quantifies cardiovascular health. Prior studies have focused on individuals without clinical cardiovascular disease (CVD). We sought to examine recent trends in LE8 scores among older adults with prevalent CVD.

Methods

We included noninstitutionalized older US adults (ages 65+) from NHANES (National Health and Nutrition Examination Survey 2013–18). Overall LE8 scores (range 0–100; higher is better cardiovascular health) were calculated for all participants. CVD diagnoses were self‐reported, including coronary heart disease, stroke, heart failure, hypertension, angina, and myocardial infarction. Percent change in LE8 scores from 2013 to 2018 was calculated for each diagnosis group, and a linear regression tested for significance of changes within groups.

Results

The 3050 participants represented 37 908 305 US adults (54.7% women; mean age 72.6). LE8 scores tended to stay stable or decline between 2013 to 2014 and 2017 to 2018 for those with and without CVD. Significant decreases in mean LE8 scores occurred in those with hypertension, with a 4.1% decline (from 59.6 to 57.1 [P<0.01]); stroke, with an 11.5% decline (from 60.6 to 53.6 [P=0.01]); and heart failure, with a 15.2% decline (from 60.9 to 51.6 [P<0.001]).

Conclusions

LE8 scores remained stable or declined for older US adults before the COVID‐19 pandemic. Populations with hypertension, stroke, and heart failure had significant LE8 score declines. Because LE8 metrics include behavioral and physiologic metrics associated with CVD risk, these data indicate concerning trends and primary and secondary prevention opportunities in older US adults, who are at highest risk for incident and recurrent CVD events.

Nonstandard Abbreviations and Acronyms

CVH

cardiovascular health

LE8

Life's Essential 8

NHANES

National Health and Nutrition Examination Survey

Research Perspective.

What Is New?

• These data demonstrate 2 independent findings: (1) older adults with heart failure, stroke, and hypertension had significant decreases in cardiovascular health leading up to the COVID‐19 pandemic (2013–2018) and (2) greater numbers of prevalent cardiovascular disease diagnoses are associated with significant declines in Life's Essential 8 scores.

What Question Should be Addressed Next?

• Future research should examine (1) which behavioral and physiologic drivers of cardiovascular health have the biggest impact on older adult cardiovascular health and (2) peri‐ and postpandemic cardiovascular health trends in older US adults with cardiovascular disease.

Cardiovascular disease (CVD) prevalence is forecasted to rise as the US population ages over the next few decades. ¹ , ² In adults ages 65 years and above, CVD is projected to account for 40% of all‐cause mortality. ¹ Simultaneously, the economic burden of CVD in the United States, estimated to be $400 billion in 2020, is projected to triple to $1344 billion by 2050. ³ , ⁴ Despite the increasing financial and resource burden of CVD in older adults, little is known about opportunities for improving prevention strategies in older adults. Given the growing burden of CVD morbidity, mortality, and costs driven by the aging population, it is critical to understand the cardiovascular health (CVH) of older adults in the United States. ⁵

An individual's overall CVH is difficult to quantify, which led the American Heart Association to create Life's Simple 7 in 2010. ⁶ This paradigm, composed of 7 health behaviors and factors that are causally related to CVD, provided a standardized, quantifiable framework for researching and promoting CVH. In 2022, the American Heart Association updated its approach to measuring CVH with Life's Essential 8 (LE8), adding sleep and a more granular 0 to 100 scale to the existing framework. ⁶ LE8 includes 4 health behaviors (diet, physical activity, nicotine, sleep) and 4 health factors (body mass index, non‐high‐density lipoprotein blood cholesterol, blood glucose, and blood pressure [BP]); higher scores indicate better CVH.

Previous research employing LE8 has overwhelmingly investigated individuals without CVD to understand primary and primordial CVD prevention and CVH in younger adults and children. ⁷ Additionally, some studies have analyzed associations between higher LE8 scores and risk of all‐cause and CVD‐specific mortality. ⁸ A significant research gap, however, persists in understanding LE8 scores among older adults (ages 65+) with CVD. The growth of this high‐risk population demonstrates the increasing importance of understanding what opportunities exist to promote CVH and limit CVD progression in older adults. ⁹ , ¹⁰ Furthermore, despite more than a 150% increase in Medicare Advantage enrollment since 2010, little is understood about whether Medicare Advantage versus traditional Medicare may be differentially associated with CVD in older adults. ¹¹ It remains unclear how the COVID‐19 pandemic affected population‐level CVH in older adults. ¹² , ¹³ , ¹⁴ As we continue to uncover the impact of the pandemic, it will be important to understand the baseline CVH in this population. Accordingly, we used data from the NHANES (National Health and Nutrition Examination Survey), 2013 to 2018, to examine the CVH of older adults living with CVD in the United States. The LE8 paradigm provided a means for characterizing CVH variations based on specific CVD diagnosis, number of CVD diagnoses, and type of insurance provider, as well as trends in CVH over time.

METHODS

Study Participants

All data and guidance on analytical approaches are publicly and freely available from the US Centers for Disease Control and Prevention's National Center for Health Statistics and can be accessed at https://www.cdc.gov/nchs/nhanes/index.html. This serial cross‐sectional study examined 2013 to 2018 data from NHANES, which collects data in 2‐year cycles. ¹⁵ The sampling methodology of participants is designed to create a nationally representative sample of the civilian, noninstitutionalized US population within each cycle. To maximize national representativeness, NHANES data are collected using a multistage, probabilistic design to select participants from strata defined by geography and proportions of minority populations. Primarily, the strata are single or contiguous counties selected with probability proportional to size. Interviews were conducted at participants' homes, and they were invited to visit a mobile examination center to undergo various anthropometric and physiologic examinations and provided a blood sample. NHANES protocols dictated the collection of all data. ¹⁵ The total combined sample of NHANES 2013 to 2018 was 28 061 participants. This study analyzed 3050 older adults, following exclusions for age <65 years (n=24 102) or incomplete interview or exam (n=909). All participants gave written informed consent. The present study was exempt from institutional review board approval given all data are publicly available and deidentified.

Demographic Characteristics

Demographic characteristics (age, sex, race, ethnicity, poverty index, household income, and educational attainment) were gathered during the home interview. Participants were stratified by self‐reported CVD diagnosis: coronary heart disease, stroke, heart failure, hypertension, angina, myocardial infarction, and no CVD reported. Self‐reported race or ethnicity was categorized as non‐Hispanic Asian, non‐Hispanic Black, non‐Hispanic White, Mexican, Other Hispanic, or Other race, including multiracial, non‐Hispanic American Indian or Alaska Native, Native Hawaiian, or Other Pacific Islander, according to NHANES protocol. Poverty index ratio, total family income divided by the US poverty threshold, was stratified into 4 categories based on US federal assistance program eligibility criteria: ≤1.30, 1.31 to 1.85, 1.86 to 3.50, >3.50. ¹⁶ Family income was categorized as <$45 000 or ≥$45 000. Educational attainment was categorized as <9th grade, 9th to 11th grade, high school graduate or test for General Educational Development, some college or associate degree, college graduate or above. Participants were stratified into 2 insurance categories, private and public, with Medicare covering the vast majority of publicly insured individuals due to the ≥65‐years‐old age cutoff of this analysis. A third category for uninsured individuals was not included due to an insufficient sample size. To incorporate the complex multistage sampling design of NHANES in the statistical analysis, SAS procedures SURVEYFREQ, SURVEYMEANS, and SURVEYREG were used. Sample weights for laboratory and physical examination data were used to estimate the number of noninstitutionalized US adults in each group.

Cardiovascular Health Score

Detailed explanations for applying LE8 scoring to NHANES data are provided in the Supplemental Material and in the American Heart Association Presidential Advisory. ⁶ Table S1 outlines definitions and scoring for the 8 CVH component metrics, including the 4 health behaviors (diet, physical activity, smoking, and sleep) and 4 health factors (body mass index, non‐high‐density lipoprotein cholesterol, blood glucose, and BP). For each participant, each CVH metric was scored on a scale of 0 to 100 points according to the American Heart Association algorithm. To calculate an overall CVH score from 0 to 100, LE8 scores for each of the 8 metrics were summed and divided by 8.

Statistical Analysis

We used survey procedures in SAS analytics software, version 9.4 (SAS Institute, Cary, NC), to account for the complex NHANES design; and we applied survey weights to generate US population‐level estimates. To create a larger sample, data from 3 2‐year cycles of the continuous NHANES were combined for 2013 through 2018. Per NHANES analytical guidelines for combining data across cycles, sample weights were constructed with rescaling of the weights such that the sum of weights matched the survey population at the midpoint of each survey period. Sample weights for laboratory and physical examination data were used to estimate the number of individuals in the US population overall and in each subgroup as appropriate. Sample weights and design were incorporated in calculating prevalence estimates and standard errors. We used weighted linear regression to compare the estimates across strata and the trend test was performed by including the number of CVD as the continuous variable. The linear regression assumption of linearity, normality of residuals, and homoscedasticity were adequately met. For all analyses, we tested hypotheses using a statistical significance level of 0.05 based on a 2‐tailed P value.

RESULTS

Sample Characteristics

In the NHANES samples, there were 3050 participants over age 65, representing 37 908 305 noninstitutionalized adults, with at least one of the following self‐reported CVDs (coronary heart disease [CHD], stroke, heart failure [HF], hypertension, angina, and myocardial infarction [MI]) or no self‐reported CVD in the United States. Table displays the self‐reported demographic characteristics (ie, sex, age, race, ethnicity, poverty index, family income, and education level) of the weighted sample by subtype of CVD. The weighted sample was 54.7% female, had a mean age of 72.6 years, and included individuals self‐identifying as Non‐Hispanic Asian (3.4%), Non‐Hispanic Black (7.2%), Non‐Hispanic White (79.9%), Mexican (3.9%), other Hispanic (3.3%), and other race/multiracial (2.4%). The average poverty index was 2.52, and 47.6% of the study population had a family income <$45 000. Excluded participants included a greater share of non‐Hispanic, Black, and Asian individuals and individuals with lower educational attainment and lower income (Table S1).

Table 1.

Cardiovascular Health Characteristics of Older US Adults (Ages 65+ Years; Not Institutionalized) by CVD

Characteristics	Overall	Coronary heart disease	Stroke	Heart failure	Hypertension	Angina	Myocardial infarction	No CVD reported	1 CVD condition	2 CVD conditions	≥3 CVD conditions
Prevalence % (n [in millions], weighted) of total sample	…	12.1 (4.6)	7.8 (2.9)	6.9 (2.6)	64.9 (24.6)	5.1 (1.9)	9.3 (3.5)	28.2 (10.7)	51.3 (19.4)	11.5 (4.3)	9.0 (0.3)
Age, y, mean (95% CI)	72.6 (72.2–73.0)	73.8 (73.0–74.7)	74.0 (73.2–74.9)	74.6 (73.9–75.3)	73.2 (72.8–73.5)	74.0 (72.7–75.2)	73.8 (73.1–74.5)	71.3 (70.7–71.9)	72.8 (72.4–73.2)	73.7 (73.1–74.3)	74.3 (73.4–75.2)
Self‐reported sex, prevalence % (n [in millions], weighted)
Men	45.3 (17.2)	64.3 (3.0)	43.4 (1.3)	51.1 (1.3)	42.1 (10.4)	60.3 (1.2)	66.2 (2.3)	47.5 (5.1)	40.4 (7.8)	50.3 (2.2)	59.6 (2.0)
Women	54.7 (20.7)	35.7 (1.6)	56.6 (1.7)	48.9 (1.3)	57.9 (14.3)	39.7 (0.8)	33.8 (1.2)	52.5 (5.6)	59.6 (11.6)	49.7 (2.2)	40.4 (1.4)
Self‐reported race and ethnicity, prevalence % (n [in millions], weighted)
Non‐Hispanic White	79.9 (30.2)	84.5 (3.9)	77.9 (2.3)	79.0 (2.1)	77.3 (19.0)	86.3 (1.7)	80.4 (2.8)	83.8 (9.0)	77.8 (15.1)	78.8 (3.4)	80.8 (2.7)
Non‐Hispanic Black	7.2 (2.7)	3.8 (0.2)	8.3 (0.2)	10.2 (0.3)	8.7 (2.2)	4.0 (0.1)	6.1 (0.2)	4.6 (0.5)	8.4 (1.6)	8.2 (0.4)	6.6 (0.2)
Non‐Hispanic Asian	3.4 (1.3)	2.9 (0.1)	2.6 (0.1)	1.7 (0.0)	3.6 (0.9)	1.4 (0.0)	1.6 (0.1)	3.3 (0.4)	3.8 (0.7)	3.7 (0.2)	1.3 (0.04)
Mexican	3.9 (1.5)	2.8 (0.1)	2.5 (0.1)	3.8 (0.1)	4.2 (1.0)	3.4 (0.1)	3.3 (0.1)	3.7 (0.4)	4.4 (0.8)	2.6 (0.1)	3.5 (0.1)
Other Hispanic	3.3 (1.2)	2.7 (0.1)	2.3 (0.1)	3.7 (0.1)	3.5 (0.9)	3.5 (0.1)	3.8 (0.1)	3.0 (0.3)	3.5 (0.7)	2.6 (0.1)	3.6 (0.1)
Other race, including multiracial	2.4 (0.9)	3.4 (0.2)	6.4 (0.2)	1.7 (0.0)	2.6 (0.6)	1.4 (0.0)	4.8 (0.2)	1.6 (0.2)	2.1 (0.4)	4.1 (0.2)	4.2 (0.1)
Poverty index, prevalence % (n [in millions], weighted)
≤1.30	15.1 (5.2)	14.6 (0.6)	21.1 (0.6)	20.5 (0.5)	17.2 (3.9)	16.7 (0.3)	19.0 (0.6)	10.9 (1.1)	15.7 (2.8)	20.2 (0.8)	17.8 (0.6)
1.31–1.85	13.0 (4.5)	16.1 (0.7)	24.2 (0.7)	26.7 (0.7)	14.5 (3.2)	19.6 (0.3)	21.9 (0.7)	8.1 (0.8)	12.9 (2.3)	16.5 (0.7)	23.3 (0.7)
1.86–3.50	29.8 (10.3)	33.1 (1.4)	29.6 (0.8)	28.0 (0.7)	30.2 (6.8)	28.6 (0.5)	24.6 (0.8)	28.4 (2.7)	30.9 (5.5)	28.6 (1.2)	29.0 (0.9)
>3.50	42.2 (14.6)	36.3 (1.6)	25.1 (0.7)	24.8 (0.6)	38.1 (8.6)	35.1 (0.6)	34.4 (1.2)	52.5 (5.1)	40.4 (7.2)	34.7 (1.4)	29.9 (0.9)
Family income, prevalence % (n [in millions], weighted)
≥$45 000	52.4 (18.3)	49.6 (2.2)	34.3 (0.9)	33.4 (0.8)	48.5 (11.0)	44.5 (0.8)	42.4 (1.4)	61.3 (6.0)	51.9 (9.3)	44.6 (1.8)	38.4 (1.2)
<$45 000	47.6 (16.7)	50.4 (2.2)	65.7 (1.8)	66.6 (1.7)	51.5 (11.7)	55.5 (1.0)	57.6 (2.0)	38.7 (3.8)	48.1 (8.6)	55.3 (2.3)	61.6 (2.0)
Education levels, prevalence % (n [in millions], weighted)
<9th grade	5.8 (2.2)	5.9 (0.3)	5.4 (0.2)	8.0 (0.2)	6.3 (1.6)	5.1 (0.1)	9.1 (0.3)	4.4 (0.5)	6.3 (1.2)	5.5 (0.2)	7.8 (0.3)
9th–11th grade	8.8 (3.3)	12.3 (0.6)	14.6 (0.4)	13.1 (0.3)	10.0 (2.5)	9.2 (0.2)	13.0 (0.5)	6.0 (0.6)	8.8 (1.7)	11.4 (0.5)	14.5 (0.5)
High school graduate or General Educational Development	23.4 (8.9)	21.5 (1.0)	31.8 (0.9)	27.0 (0.7)	25.7 (6.3)	30.3 (0.6)	26.6 (0.9)	19.3 (2.1)	24.1 (4.7)	26.9 (1.1)	28.2 (1.0)
Some college or associate degree	30.2 (11.5)	26.0 (1.2)	27.0 (0.8)	29.9 (0.8)	30.4 (7.5)	28.8 (0.6)	27.6 (1.0)	28.9 (3.1)	32.6 (6.3)	25.6 (1.1)	26.7 (0.9)
College graduate or above	31.7 (12.0)	34.2 (1.6)	21.1 (0.6)	22.0 (0.6)	27.6 (6.8)	26.5 (0.5)	23.8 (0.8)	41.4 (4.4)	28.2 (5.5)	30.6 (1.3)	22.8 (0.8)

Prevalence of cardiovascular disease diagnoses in older US adults (ages 65+, not institutionalized), National Health and Nutrition Examination Survey 2013 to 2018. CVD indicates cardiovascular disease.

Hypertension (64.9%) was the most common CVD diagnosis, followed by CHD, MI, stroke, HF, and angina. Over a quarter of the weighted sample had no reported CVD. Men made up the majority of participants with CHD, angina, HF, and MI, whereas women made up the majority of participants with stroke, hypertension, and no reported CVD. Prevalence of various CVDs also varied by race and ethnicity. The majority of participants with any CVD had family incomes <$45 000, whereas only 38.7% of those without CVD had a family income <$45 000. About 41% of participants without CVD had an undergraduate or graduate degree, whereas the prevalence of undergraduate and graduate degrees ranged from 21% to 34% for those with CVD.

Status of CVH in Older Adults With Cardiovascular Disease

Figure 1 displays LE8 scores (possible range 0–100) overall and by LE8 metric for participants from 2013 to 2018, stratified by CVD diagnoses. Overall LE8 scores were highest for participants with CHD (60.4) and lowest for participants with HF (56.1). Physical activity scores tended to be the lowest among all LE8 metrics across CVD diagnoses; participants with stroke had the lowest score (27.8). BP scores also tended to be low; participants with hypertension had the lowest score, and participants with CHD had the highest score. Of note, the LE8 scoring algorithm applies a “treatment penalty” to the BP score for individuals taking antihypertensive medications, thereby contributing to the low BP scores of participants medicated for their hypertension. ⁶ Body mass index and diet scores were low to moderate, as defined by the LE8 guidelines. ⁶ Non‐high‐density lipoprotein cholesterol and blood glucose scores were moderate and had narrow ranges, 65.2 to 69.1 and 56.7 to 61.5, respectively. Lastly, sleep scores were the highest for all participants, followed closely by nicotine exposure scores. Excluded participants had similar physiologic LE8 measures to included individuals, but their LE8 health behavior scores were slightly lower (Table S1). When examining LE8 scores by sex, race, and CVD, men and women had no significant differences, and non‐Hispanic Asian participants had the highest overall LE8 scores (Table S2).

Figure 1. LE8 scores by cardiovascular disease.

Mean is represented by the thick white bar and 95% CI by the thin white bars. Life's Essential 8 scores of older US adults (ages 65+ years; not institutionalized) by cardiovascular disease: National Health and Nutrition Examination Survey 2013 to 2018. CVD indicates cardiovascular disease; CVH, cardiovascular health; HDL, high‐density lipoprotein; and LE8, Life's Essential 8.

Association Between LE8 CVH Score and Number of CVD Diagnoses

Figure 2 depicts the association between LE8 CVH scores and the number of CVD diagnoses per individual participant. The mean LE8 scores decreased from 68.4 to 53.5 as the number of CVD diagnoses increased (P value <0.0001). Significant differences in mean LE8 score existed between participants with 0 and 1 CVD and between those with 3 and 4 CVDs. Sensitivity analyses were performed to remove hypertension as a CVD diagnosis and separately to remove BP as a component of LE8 score to limit any confounding associated with having BP contribute to the LE8 score of participants with hypertension (Figures S1 and S2).

Figure 2. Association between LE8 CVH score and number of CVD diagnoses.

Mean overall LE8 score is represented by the purple dot, 95% CI is represented by the black bars. Individuals with 6+ CVD diagnoses were removed due to the small sample size (n [in millions] weighted=0.08) and wide 95% CI range (LE8=33–67). P value assesses the significance of the relationship between mean overall LE8 score and number of CVD diagnoses. CVD indicates cardiovascular disease; CVH, cardiovascular health; and LE8, Life's Essential 8.

Trends in CVH in Adults With Cardiovascular Disease

Figure 3 displays the trend in mean overall LE8 scores from 2013 to 2018. Generally, LE8 scores declined or stayed stable across these 3 biennial NHANES samples. Participants with HF, stroke, and hypertension had significant declines of −15.2% (P<0.001), −11.5% (P=0.01), and −4.1% (P<0.01), respectively. LE8 scores also changed somewhat for angina (−6.6%), no CVD (−4.0%), MI (−1.3%), and CHD (+1.5) but these differences were not statistically significant. A sensitivity analysis removing BP from the LE8 score to assess for confounding found no significant differences in trends for individuals with CVD from 2013 to 2018 (Figure S3). Splitting participants into 2 age subgroups, 65–74 (Figure S4) and 75+ (Figure S5), demonstrated that the declines in LE8 scores from 2013 to 2018 were driven primarily by the older subgroup.

Figure 3. Change in Life's Essential 8 scores over time for older US adults (aged 65+ years, not institutionalized) by cardiovascular disease: National Health and Nutrition Examination Survey 2013 to 2018.

P value reflects the test of whether the slope of change in overall LE8 score between 2013 to 2014 and 2017 to 2018 differs significantly from 0. P values <0.05 are bolded for significance. CHD indicates coronary heart disease; CVD, cardiovascular disease; HF, heart failure; LE8, Life's Essential 8; MI, myocardial infarction; and NHANES, National Health and Nutrition Examination Survey.

Status of CVH in Adults With Cardiovascular Disease by Insurance

Figure 4 displays the means and 95% CIs of LE8 CVH scores (possible range, 0–100) for adults from 2013 to 2018 stratified by CVD diagnosis and type of insurance. For all CVD except for HF, participants with CVD with private insurance had higher overall LE8 scores than those with public insurance. By CVD, the largest difference in LE8 scores between participants with CVD with private insurance (including Medicare Advantage) versus public insurance (ie, Medicare) was angina (7.0) whereas the smallest difference was stroke (0.1).

Figure 4. Overall LE8 scores by CVD by insurance coverage.

Mean is represented by the thick white bar and 95% CI by the thin white bars. Overall Life's Essential 8 scores of older US adults (ages 65+ years; not institutionalized) by CVD by insurance coverage: National Health and Nutrition Examination Survey 2013 to 2018. CVD indicates cardiovascular disease; CVH, cardiovascular health; and LE8, Life's Essential 8.

DISCUSSION

This study provides an assessment of the CVH of older adults with CVD using the American Heart Association's LE8 metrics, offering insight into the current state of CVH in this vulnerable population. ⁶ Our findings highlight significant heterogeneity in the CVH of older adults with and without CVD. Participants with 1 CVD had a mean LE8 score of 59.5, which was 8.9 points lower than those without any CVD. Prior research has shown that per 10‐point increase in LE8 score, CVD risk decreases by 17% and total mortality decreases by 10% in individuals with type 2 diabetes. ¹⁷ LE8 scores decreased as the number of CVD diagnoses increased. These findings complement geriatric CVD research that has shown the high prevalence of multimorbidity in older adults with CVD. ¹⁸ , ¹⁹ The substantial difference in LE8 score between the group with 0 and 1 CVD may help explain the “accelerated aging” seen in older adults with CVD. ¹⁸ , ²⁰ Therefore, although this association is only correlative, it highlights an opportunity to focus on primary prevention opportunities in older adults.

The 8.9‐point gap between mean overall LE8 score for those with 0 and 1 CVD diagnosis was primarily driven by 2 factors: BP and physical activity scores (Figure 1). BP scores for participants with CVD were 26.6 to 45.1 points lower than those without CVD, and physical activity scores were 11.2 to 24.1 points lower. Put in context, a 20‐point increase in BP LE8 score has been shown to be associated with a 0.90 hazard of developing CVD in individuals with type 2 diabetes. ¹⁷ Notably, when BP was removed as a component of LE8 score, the mean LE8 scores for each number of diagnoses increased a few points, suggesting that the participants appeared healthier when BP was not considered. BP control has long been a focus of CVH management in geriatric populations, and these findings suggest that even greater emphasis may be warranted. ²¹ , ²² , ²³ , ²⁴ It is hoped that the results of the SPRINT (Systolic Blood Pressure Intervention Trial) and ESPRIT (Effects of Intensive Systolic Blood Pressure Lowering Treatment in Reducing Risk of Vascular Events) trials demonstrating that intensive BP control (systolic BP <120 mm Hg) reduces major adverse cardiac events compared with standard BP control (systolic BP <140 mm Hg) will lead to clinical improvements in BP control in older patients with CVD. ²⁵ , ²⁶ , ²⁷ Finally, the difference in mean blood glucose scores (9.1–13.9 points) also contributed to the gap between overall LE8 scores those with CVD and those with no CVD, which is consistent with the well‐demonstrated impact of blood glucose on metabolic health and CVD development. ²⁸ , ²⁹ , ³⁰

The study also revealed a decline in LE8 scores over time for older adults with HF, stroke, and hypertension. Participants with stroke and HF had the lowest physical activity scores among all diagnoses, which is consistent with the functional limitations commonly associated with these conditions. Although formal thresholds for clinically significant change in LE8 score have not been established and existing literature focuses on people without CVD, the magnitude of decline observed, particularly in the groups with stroke and HF, suggests additional CVD management strategies may be needed to improve CVH in this population. ³¹ There is particular urgency considering the growing prevalence of CVD among older adults and the associated burden of morbidity and mortality. ⁴ Although this declining trend may be due in part to worsening health in our older adult population, it may also be driven by improvements in cardiovascular care prolonging the lives of groups with lower CVH. ¹⁸ In other words, as medical care improves, sicker individuals live longer than before. Thus, over time, the average CVH of adults over age 65 in the United States may decline due to greater longevity for individuals with low CVH. However, given the short time course of this study, it is unlikely that cardiovascular care improvements (and thus longer life expectancy for individuals with lower CVH) can fully account for the substantial declines seen in HF, stroke, and hypertension LE8 scores.

Finally, the study observed that older adults with CVD and Medicare Advantage had higher LE8 scores compared with those with public Medicare, except for those with HF (Figure 4). This disparity in CVH by insurance status highlights the need to investigate how type of insurance affects the cardiovascular care that older adults receive. Many other socioeconomic factors, such as income, education, and access to health care, have been shown to significantly affect CVH status. ³² , ³³ , ³⁴ However, insurance status and type of insurance are often excluded from these types of analyses. A better understanding of how variations in health insurance (or ability to afford it) may play a role in CVH is important for achieving equitable CVD outcomes and improving the overall health of older adults.

Limitations

First, the cross‐sectional nature of these data precludes them from demonstrating causal relationships. To show trends in individual CVH over time, longitudinal cohort studies would be required. Furthermore, conclusions from this cross‐sectional study must remain at a population level and cannot be directly applied to individual patients in a clinical setting. Any inconsistencies in the serial cross‐sectional samples could affect the reliability of the findings. Second, the self‐reported nature of CVD diagnoses in NHANES may introduce some degree of misclassification. Third, our study is limited to 6 CVD diagnoses (ie, CHD, stroke, HF, hypertension, angina, MI) and excludes other, less common CVD diagnoses. Fourth, the excluded group contained a slightly higher proportion non‐Hispanic Black and Asian participants, which may limit the generalizability of our findings to these populations. Fifth, diet scoring, measured using the Dietary Approaches to Stop Hypertension score, omits explicit assessment of refined carbohydrates such as added sugar, which have been linked to atherogenic dyslipidemia. ³⁵ Sixth, a large portion of participants had multiple CVD diagnoses, so the LE8 scores calculated for each CVD include individuals who have more than just that sole diagnosis. Although this may limit the utility of applying these results to individual patients, it accurately reflects the US population, making it suitable for epidemiological conclusions. ³⁶ , ³⁷

Future Directions

Our findings are intended to lay the groundwork for future investigation of the drivers of CVH discrepancies in older adults. Longitudinal studies will be required to understand the time course and factors that differentiate the LE8 scores of older adults with prevalent CVD diagnoses. Furthermore, observational analysis of the peri‐ and postpandemic population will be required to determine how CVH trends may have changed in older US adults with CVD.

Conclusions

This study of a nationally representative sample of older US adults demonstrates those with CVD had significantly lower LE8 scores than those without CVD. Furthermore, an increasing number of CVD diagnoses was shown to be significantly correlated with a decreasing LE8 score. Finally, we observed that CVH was suboptimal and declining in older adults before the COVID‐19 pandemic. As our society continues to age, these findings highlight a need for further research aimed at preventing the onset and progression of CVD in our older adult population.

Sources of Funding

None.

Disclosures

None.

Supporting information

Tables S1–S2

Figures S1–S5