Association of Life’s Crucial 9 with cognitive function and stroke risk: insights from the NHANES 2011–2014 study

Abstract

Background

Cognitive impairment and stroke constitute major health challenges for the aging global population, adversely impacting quality of life and increasing healthcare burdens. The American Heart Association’s “Life’s Essential 8” (LE8) framework has served as a key tool for evaluating cardiovascular health (CVH); however, it omits mental health, a critical factor influencing both cognitive function and stroke risk. The introduction of “Life’s Crucial 9” (LC9), which includes depressive symptoms, provides a more comprehensive approach. This study investigates the relationship between LC9, cognitive function, and stroke risk.

Methods

Utilizing the National Health and Nutrition Examination Survey (NHANES) dataset from 2011 to 2014, cross-sectional data from 2,327 participants were analyzed. Stratified analyses were performed according to demographic and health-related factors. A Restricted Cubic Spline (RCS) model was employed to examine potential threshold effects. Additionally, weighted linear regression models were used to evaluate cognitive performance, and logistic regression models were applied to assess stroke risk.

Results

Higher LC9 scores were positively associated with better cognitive function and lower odds of stroke. Within the cognitive function analysis, higher LC9 scores were significantly associated with superior performance on the Digit Symbol Substitution Test (DSST) (β = 0.18, 95% CI: 0.11– 0.26, P < 0.001). In the stroke analysis, individuals with higher LC9 scores exhibited decreased odds of experiencing a stroke (OR = 0.97, 95% CI: 0.95–0.99, P = 0.005). RCS analysis identified a non-linear relationship between LC9 scores and the odds of stroke, with the greatest decreases in stroke odds observed at lower LC9 scores, plateauing around a score of 70.

Conclusions

Higher LC9 scores are associated with better cognitive function and lower odds of stroke. These findings suggest that incorporating mental health metrics, such as depression, into cardiovascular health assessments enhances the predictive power for cognitive outcomes and stroke prevention.

Introduction

Cognitive impairment and stroke represent significant threats to the health and well-being of the global elderly population, substantially reducing patients’ quality of life and life expectancy while imposing considerable burdens on society and healthcare systems [1, 2]. Cognitive impairment encompasses various stages, starting from mild cognitive impairment (MCI), which serves as a bridge between normal cognition and dementia, to severe dementia, often represented by Alzheimer’s disease [3]. In 2015, approximately 47 million individuals worldwide were living with dementia, and this figure is projected to increase to 131 million by 2050, with annual global economic losses from dementia exceeding significant thresholds [4]. Likewise, by 2030, the incidence of stroke is anticipated to increase most notably among Hispanic men, with direct healthcare expenses having surged by over 300% since 2012 [5]. The focus on cognitive dysfunction and stroke, as well as their negative outcomes, has emerged as a central concern in global public health and gerontological research. This underscores the critical importance of identifying and modifying health behaviors and biomarkers to enhance prevention and management strategies.

In recent years, the American Heart Association (AHA) introduced ‘Life’s Essential 8’ (LE8) as a comprehensive framework for evaluating cardiovascular health (CVH) [6]. Despite the association of higher LE8 scores with reduced cardiovascular disease risk and improved cognitive function [7, 8, 9], several limitations remain. Notably, LE8 does not incorporate mental health factors, such as depression, which significantly influence cognitive dysfunction and stroke risk [10], [11]. To address this gap, the Life’s Crucial 9 (LC9) framework was recently proposed, incorporating mental health into the LE8 construct to provide a more comprehensive assessment of overall health [12]. Given that depression has emerged as a major global public health challenge—with the World Health Organization (WHO) projecting it to become the foremost contributor to the global burden of chronic diseases by 2030 [13]—integrating mental health considerations into CVH assessments has become increasingly imperative.

Currently, no studies have investigated the link between LC9 with cognitive function and stroke. This study aims to addresses the gap by analyzing data from the National Health and Nutrition Examination Survey (NHANES) to investigate the relationship between LC9 with cognitive function and stroke.

Methods

The data is publicly accessible on the NHANES website(https://wwwn.cdc.gov/nchs/nhanes/Default.aspx).

Study population

This analysis utilized data from NHANES, a continuous, nationwide, and periodic health survey program designed to evaluate the health and nutritional status of individuals in the United States. The design and methodology of NHANES have been detailed in prior publications.

This study employed data from the 2011–2012 and 2013–2014 NHANES cycles, applying a stringent screening process to identify the final cohort for analysis. Cognitive function data were exclusively gathered during three NHANES cycles: 2011–2012, 2013–2014, and 2019–2020. The selection of the 2011–2014 period enabled the inclusion of two consecutive cycles, thereby facilitating a larger and more robust sample size. This methodological choice augmented the statistical power of the study and ensured consistency in data collection protocols across the selected cycles. Initially, 19,931 participants were screened, and 3,014 individuals with complete Digit Symbol Substitution Test (DSST) data were included. Subsequently, 448 participants without LC9 data were excluded, leaving 2,566 participants. Further screening excluded 239 participants due to missing covariate data, resulting in 2,327 participants being included in the cognitive function analysis. Among these participants, they were divided into groups based on stroke status, including 158 participants with stroke and 2,169 participants without stroke (Fig. 1). This screening process ensured the integrity of the analysis sample and the reliability of the study results.

Fig. 1

Flowchart of the sample selection. Flowchart illustrating selection of the study population and analysis in NHANES from 2011 to 2014. NHANES, National Health and Nutrition Examination Survey

Fig. 2

Subgroup analyses of the association between LC9 scores and cognitive function. The figure presents β coefficients and 95% CIs for the association between LC9 scores and cognitive function (assessed by DSST) across various subgroups, including sex, race, education level, PIR, hypertension status, and DM status. PIR, poverty-to-income ratio; DM, diabetes mellitus

Fig. 3

Restricted cubic spline analyses showing the association of LC9 scores with cognitive function and stroke. (A) The association between LC9 scores and cognitive function. (B) The association between LC9 scores and the odds of stroke. LC9, Life’s Crucial 9

Fig. 4

Subgroup analyses of the association between LC9 scores and stroke. The figure presents odds ratios and 95% CIs for the association between LC9 scores and stroke across different subgroups, including sex, race, education level, PIR, hypertension status, and DM status. LC9, Life’s Crucial 9; PIR, poverty-to-income ratio; DM, diabetes mellitus

Measurement of LC9

The LC9 score, as described in Supplementary Table S1, is calculated by combining the depression score with the LE8 score. Eight parameters make up the LE8 score: four physiological—body mass index (BMI), blood glucose, non-high‐density lipoprotein cholesterol (non-HDL-C), and blood pressure—and four behavioral—physical activity, nicotine exposure, sleep duration, and dietary habits. Each of these metrics reflects an important aspect of cardiovascular health and is evaluated based on specific guidelines or standardized measures [14, 15, 16]. The depression score was derived from the Patient Health Questionnaire‐9 (PHQ‐9), a validated instrument for depression screening. Elevated PHQ-9 scores correspond to increased severity of current depressive symptoms. The depression score were categorized as 100, 75, 50, 25, and 0, corresponding to PHQ-9 score ranges of 0–4, 5–9, 10–14, 15–19, and 20–27, respectively 17.

An individual’s LC9 score was determined by averaging the sum of the eight LE8 component scores and the depression score [18]. Since there are currently no established classification criteria specifically for LC9, the cutoff points from the LE8 framework were applied. According to the AHA guidelines [6], participants were classified into LC9 groups as low (< 50), moderate (50–79), and high (≥ 80) based on their scores.

Outcomes

Outcomes of the study included cognitive function and stroke. The DSST, a validated and effective measure, was employed to assess cognitive function [19]. The DSST, a component of the Wechsler Adult Intelligence Scale (WAIS-III), assessed processing speed, sustained attention, and working memory. Participants in the study need to accurately code a set of symbols within 120 s to finish the DSST. The DSST scoring system spans from 0 to 105, where higher scores reflect enhanced cognitive function.

Stroke was defined as a physician-diagnosed condition reported by the individual during an in-person interview. Individuals who responded ‘yes’ to the question, “Have you ever been told by a physician or a health professional that you had stroke?” were classified as having had a stroke. While the use of self-reported measures may introduce recall bias, which could potentially affect data interpretation, this approach has been commonly used in other studies utilizing the NHANES database [20, 21].

Covariates

Drawing from earlier studies and biological insights, we gathered numerous covariates known to influence cognitive function and stroke. The covariates included age, sex (female, male), race (Mexican American, non-Hispanic black, non-Hispanic white, other race), education (< high school, high school, > high school), poverty-to-income ratio (PIR) (< 1.3, 1.3–3.5, > 3.5), and alcohol use (never, former, heavy, mild, moderate), which were collected through standardized questionnaires. The diagnosis of hypertension and diabetes mellitus (DM) was based on index measurements, the use of medications, and self-reporting.

Statistical analysis

The baseline characteristics of the study participants were described using the mean ± standard deviation (SD) for continuous variables and frequencies with corresponding percentages for categorical variables. Comparative analyses were conducted using either analysis of variance or the Kruskal-Wallis test for continuous variables, while categorical variables were assessed using the chi-square test. Weighted linear regression was employed to investigate the association between the LC9 score and cognitive function, whereas logistic regression was utilized to assess its relationship with stroke. The Crude models were unadjusted for any potential confounders. Model 1 adjusted for age, sex, race, education, and PIR, while Model 2 further adjusted for alcohol consumption. RCS was also applied to further explore potential threshold effects in the relationship between LC9 scores, cognitive function, and stroke.

To investigate various subpopulations at baseline, analyses were stratified according to sex, age, race, education, PIR, hypertension, and DM. The interaction between stratification factors and LC9 scores was analyzed using multiplicative interaction tests. Similarly, the interaction between health behavior scores and health factor scores was evaluated using the same methodological approach.

Statistical analyses for this study were carried out using R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria; http://www.r-project.org) and statistical significance was ascertained by a two-sided P value of < 0.05.

Results

Baseline characteristics of the study participants

The baseline characteristics of the study participants, categorized stratified by LC9 scores are presented in Table 1. Overall, the mean age was 68.88 ± 0.2 years, 54.12% were women, and 81.77% were non-Hispanic White. The average LC9 score was 67.26 ± 0.48. Participants classified within the high LC9 scores group had superior financial status, lower BMI, higher levels of educational attainment, and a greater proportion of individuals who were light alcohol consumers and non-smokers. Additionally, this group showed a significantly lower prevalence of hypertension and stroke. Furthermore, individuals in the high LC9 scores group achieved higher scores on the DSST, reflecting enhanced cognitive function (P < 0.001).

Table 1 Baseline characteristics of participants by LC9 quartile

Association between LC9 and cognitive function

As illustrated in Table 2, weighted linear regression models were employed to assess the relationship between LC9 and cognitive function. In Model 1, a higher LC9 score was associated with improved DSST performance (β = 0.16, 95% CI: 0.008–0.23, P < 0.001). After adjusting for covariates such as age, sex, ethnicity, education, PIR, and alcohol use in Model 2, the positive association between LC9 and DSST performance persisted (β = 0.18, 95% CI: 0.11–0.26, P < 0.001).

Table 2 Weighted linear regression showing the relationship between LC9 and DSST scores

Figure 2 illustrated the results of a subgroup analysis examining the association between LC9 and cognitive function. The findings indicated that LC9 was significantly associated with cognitive function across different genders, major ethnic groups (e.g., White, Mexican), PIR, DM, and hypertension statuses (all P < 0.05), with no significant interactions detected (P > 0.1). The RCS analysis demonstrated a significant linear association between LC9 and cognitive function, with an overall P-value of less than 0.001. Furthermore, the test for nonlinearity yielded a P-value of 0.545, indicating a lack of significance. This suggested that the relationship between LC9 and cognitive function was predominantly linear, as illustrated in Fig. 3A.

Association between LC9 and stroke

Table 3 presents the association between LC9 scores and the likelihood of experiencing a stroke, as evaluated through logistic regression analysis across three models: the Crude model, Model 1, and Model 2. In the Crude model, higher LC9 scores were significantly correlated with reduced odds of stroke (OR = 0.97, 95% CI: 0.95–0.98, P < 0.001). This inverse relationship remained statistically significant in Model 1, which accounted for confounding variables such as age, sex, race, educational attainment, and the PIR (OR = 0.97, 95% CI: 0.95–0.99, P = 0.004), and persisted in Model 2 with additional adjustment for alcohol use (OR = 0.97, 95% CI: 0.95–0.99, P = 0.005).

Table 3 Weighted logistic regression showing the relationship between LC9 score and the odds of stroke

When LC9 scores were stratified into low, moderate, and high categories, the Crude model indicated that individuals with moderate LC9 scores exhibited significantly lower odds of stroke (OR = 0.37, 95% CI: 0.22–0.64, P < 0.001), with those possessing high LC9 scores demonstrating even further reduced odds (OR = 0.18, 95% CI: 0.08–0.43, P < 0.001) relative to the reference group. These associations remained significant in Model 1 (OR = 0.39, 95% CI: 0.22–0.70, P = 0.003 for moderate LC9, and OR = 0.23, 95% CI: 0.10–0.56, P = 0.002 for high LC9). Further adjustments in Model 2 did not substantially modify the results for both moderate and high LC9 categories. Subgroup analyses presented in Fig. 4 revealed a statistically significant association between higher LC9 scores and lower odds of stroke, with the most pronounced associations observed among male participants, White individuals, those with at least a high school education, participants with moderate income, and individuals diagnosed with hypertension (P < 0.05). No significant interactions were identified among the subgroups (P for interaction > 0.1). These findings provide compelling evidence supporting the relationship between increased LC9 scores and decreased odds of stroke.

The RCS analysis revealed a significant overall association between LC9 scores and the odds of stroke (P for overall < 0.001), with the evidence of nonlinearity (P for nonlinear = 0.008). As the LC9 score increased, the odds of stroke decreased sharply at lower LC9 scores, plateaued around an LC9 score of 70, and then showed slight fluctuations while remaining below an OR of 1, indicating consistently lower odds of stroke with higher LC9 scores (Fig. 3B).

Discussion

This study investigated the associations between LC9 scores with cognitive function and stroke using data from NHANES. The findings suggest that higher LC9 scores are associated with better cognitive performance and lower odds of stroke. Specifically, as LC9 scores increased, participants demonstrated improved performance on the DSST, indicating enhanced cognitive ability, and were less likely to report a history of stroke.

Our findings are consistent with previous research on the LE8 framework, which indicates that several key cardiovascular health factors - such as diet, exercise, blood glucose, and blood pressure - are closely associated with cognitive health and stroke risk [22, 23]. The LE8 model has been extensively applied in studies investigating the relationship between CVH and chronic diseases. Nonetheless, a notable limitation of the LE8 frame is its exclusion of mental health factors, such as depression. To enhance the comprehensiveness of this framework, the AHA introduced LC9 framework, which adds depressive symptoms as an independent metric on top of the original LE8 components [12]. Depression is widely acknowledged as an independent risk factor for both cognitive impairment and stroke, and it interacts with cardiovascular risk factors such as blood pressure and blood glucose levels[10, 11]. Numerous cohort studies have demonstrated a negative association between depressive symptoms and cognitive function [24, 25]. For example, Sawyer et al. investigated the relationships among depressive symptoms, hippocampal volume, and cognitive decline, proposing that depression may trigger a glucocorticoid cascade that damages the hippocampus, thereby impairing memory formation and storage and increasing the risk of cognitive decline [26]. However, some studies have reported conflicting findings, indicating no significant association between depression and cognitive decline in older adults [27]. These discrepancies may be attributed to both variations in study populations and the inherent limitations of focusing exclusively on the association between depression and cognitive function.

Furthermore, the association between depression and stroke has garnered increasing scholarly attention. A meta-analysis, which synthesized findings from 28 prospective cohort studies, identified depression as an independent risk factor for stroke [28]. In addition, studies have indicated that depression was not only associated with an increased risk of stroke onset but also significantly influenced the recovery process, the occurrence of complications, and reduced survival rates post-stroke [29]. Various mechanisms have been proposed to explain these associations, including alterations in brain and neuronal function affecting neuroendocrine pathways, autonomic nervous system dysfunction, immune responses, platelet activation and thrombosis, lifestyle behaviors, and cardiac metabolic risk factors [30, 31]. A pooled analysis encompassing 563,255 participants from 22 prospective demonstrated an association between baseline depressive symptoms and the incidence of CVD, including stroke [32]. Nonetheless, the magnitude of these associations was found to be modest. This underscores the necessity of expanding research beyond merely examining the relationship between depression, stroke, and cognitive function.

Current evidence indicates that depression does not occur in isolation but interacts with other cardiovascular risk factors, such as hypertension, high cholesterol, and diabetes, all of which have significant effects on cognitive impairment and stroke risk [33]. Depression may amplify the impact of these risk factors by affecting the neuroendocrine system, particularly the hypothalamic-pituitary-adrenal axis, thereby triggering inflammatory responses and altering vascular reactions [34]. These biological pathways may enhance susceptibility to cognitive decline and stroke. In consideration of these complexities, the current study emphasizes the importance of adopting a more comprehensive framework for assessing cognitive function and stroke risk. Consequently, we employed the recently proposed LC9 framework by the AHA and further investigated the associations between LC9 scores, cognitive function, and stroke risk. Furthermore, this study utilized multiple analytical methods to ensure the robustness and generalizability of our findings. Initially, a weighted linear regression analysis revealed a significant positive association between LC9 scores and cognitive function. This association remained statistically significant even after adjusting for various potential confounders. To further explore the relationship between LC9 scores and cognitive function, we conducted the RCS analysis, which confirmed a linear association. This result further substantiates the validity of the LC9 framework as a reliable tool for evaluating the connection between CVH and cognitive function. Secondly, regarding stroke risk, our analyses revealed that elevated LC9 scores were significantly correlated with a decreased probability of stroke occurrence. This association persisted across various statistical models. Notably, our stratified analyses demonstrated that the relationship between LC9 scores and stroke risk was robust across multiple subgroups, including gender, race, educational attainment, income levels, and hypertension status. These findings imply that LC9 is not only effective within the general population but also exhibits broad applicability across diverse demographic and health-related subgroups.

While this study offers valuable insights, several limitations must be acknowledged. Firstly, the reliance on self-reported stroke data may introduce recall bias, potentially compromising the accuracy and interpretation of the results. Secondly, although adjustments were made for a wide array of potential confounders, the possibility of residual or unmeasured confounding cannot be entirely dismissed. Thirdly, the investigation of the relationship between LC9, cognitive function, and stroke was conducted using a cross-sectional design, which precludes the establishment of temporality and limits the ability to infer causality between LC9 scores and stroke outcomes. Future research should seek to validate these findings through the use of more precise, clinically confirmed stroke diagnoses and the adoption of longitudinal study designs to better establish temporal relationships and causal inferences. Furthermore, studies employing innovative methodologies and involving larger, more diverse populations are necessary to enhance the robustness and generalizability of the findings.

Conclusions

This research illustrates that higher LC9 scores were associated with better cognitive function and lower odds of stroke. These relationships were consistent across diverse demographic and health subgroups, underscoring the wide-ranging applicability of LC9 framework. The results underscore the significance of integrating mental health variables, such as depression, into cardiovascular health evaluations. Employing the LC9 framework facilitates a more holistic comprehension of the interplay between cardiovascular health, cognitive function, and stroke, thereby providing critical insights for early prevention and intervention strategies.