Association of mid-age Life's Essential 8 score with digital cognitive performance and incident Alzheimer's disease: The Framingham Heart Study

Abstract

Background

Cardiovascular health (CVH) is a modifiable risk factor for Alzheimer's disease (AD). However, studies examining the association between mid-age CVH, as indicated by Life's Essential 8 (LE8) health metrics, and digital cognitive performance or AD risk are limited.

Objective

To examine the associations between mid-age CVH, assessed by LE8 scores during ages 45 to 65, and digital Clock Drawing Test (dCDT) performance as well as the incidence of AD.

Methods

We included 1198 participants (51.6% women) from the Framingham Heart Study (FHS) Offspring cohort. Linear regression and Cox proportional hazards models were applied to examine the associations between mid-age CVH and dCDT performance, as well as the incidence of AD.

Results

Over a median follow-up of 17.5 years, 45 participants developed AD. Each standard deviation (SD) higher mid-age LE8 total score was associated with a 0.16 SD higher level of the dCDT total score (p < 0.001) and a 0.35-fold lower risk of incident AD (HR = 0.65, 95% CI: 0.49–0.87, p = 0.003). The dCDT measures showed stronger associations with mid-age LE8 and AD risk compared to the conventional CDT (cCDT). For example, the drawing score on copy tasks was more strongly associated with LE8 (beta = 0.10, p = 0.007 versus beta = 0.08, p = 0.27) and had higher discrimination for incident AD (C-statistic = 0.89 versus 0.83) compared to the cCDT.

Conclusions

Our results highlight the potential of digital cognitive assessments for evaluating AD risk and emphasize the importance of mid-age CVH in shaping cognitive outcomes and the development of AD.

Keywords

Alzheimer's disease cardiovascular health conventional Clock Drawing Test digital Clock Drawing Test digital cognitive assessment Life's Essential 8

Introduction

Alzheimer's disease (AD) presents a significant challenge to public health due to its gradual onset and progressive deterioration of cognitive functions.¹ With effective treatments for AD being scarce, the focus has shifted towards early detection as a key strategy for managing this disease.² Studies have highlighted the importance of mitigating major risk factors, particularly cardiovascular health (CVH), during middle age before cognitive decline begins, to lower AD risk.^3–6 The American Heart Association (AHA) had put forward Life's Simple 7 (LS7) as a CVH metric targeting modifiable risk factors including blood pressure, blood sugar, cholesterol levels, body mass index (BMI), physical activity, diet, and smoking.⁷ Many studies have reported a significant association between higher LS7 scores and a decreased AD risk.^8–10 Building on LS7, the AHA recently introduced Life's Essential 8 (LE8), refining the model to incorporate sleep quality and a 0 to 100 scale for a more precise CVH quantification.¹¹ However, the effect of mid-age LE8 scores on cognitive health and their influence on AD risk remains an underexplored area.

Understanding the association between mid-age LE8 scores and early cognitive decline requires accurately tracking these initial, subtle cognitive changes. Advances in technology have led to the development of precise digital methods for assessing cognitive function, significantly surpassing traditional neuropsychological (NP) tests in terms of objectivity and sensitivity to early signs of cognitive decline. The digital Clock Drawing Test (dCDT) exemplifies such innovation by utilizing a digitizing ballpoint pen and grid-patterned paper to perform the conventional Clock Drawing Test.¹² This approach utilizes advanced software to track pen movements, thereby refining the conventional CDT (cCDT) and NP evaluations with standardized administration, objective scoring, and reduced biases. Additionally, dCDT covers a broad spectrum of cognitive domains and utilizes a comprehensive scoring system for an in-depth analysis of cognitive abilities.^12,13 Therefore, determining the association between mid-age CVH, as indicated by LE8 scores, and the capacity of dCDT for early cognitive assessment is essential for timely intervention strategies.

The main objective of this study was to investigate associations between mid-age LE8 scores, dCDT performance, and incident AD in the Framingham Heart Study (FHS), aiming to provide essential insights for the prevention of cognitive decline. Additionally, we compared the capabilities of cCDT and dCDT in evaluating cognitive status by conducting association analyses with mid-age CVH and incident AD in the FHS cohort.

Methods

Study population

The FHS, established in 1948, is a community-based cohort study.¹⁴ The Offspring (Gen2) cohort of the FHS began in 1971, initially enrolling 5124 participants, including children of the Original cohort and the spouses of the children. These participants undergo routine health examinations approximately every four to six years, with ten exams (Exams 1–10) conducted so far, to collect a broad array of clinical data at each exam.^15,16 The dCDT measurements were collected during a neuropsychological examination conducted around Exam 9. The mid-age LE8 scores were calculated using variables collected from Exam 5 through Exam 9. Participation in the dCDT test was based on participant consent, with a total of 1413 Gen 2 participants having the dCDT measurements. We further excluded participants under 60 years of age (n = 67) and those diagnosed with non-AD dementia (n = 25) at the time of dCDT measurement. Additionally, 123 participants were excluded due to incomplete data on independent and outcome variables, as well as covariates, resulting in a cohort of 1198 participants eligible for subsequent analysis (Supplemental Figure 1). Of the 1198 participants, all had their mid-age LE8 assessments completed before the dCDT assessment, except for four participants whose mid-age LE8 assessments included variables measured at Exam 9, when the dCDT was also measured.

The study's procedures and protocols were approved by the Institutional Review Board at Boston University Medical Campus. All participants provided written informed consent.

Construction of mid-age LE8 score

The LE8 total score was derived based on the AHA guidelines,¹¹ with modifications made to fit the FHS cohorts.¹⁷ Briefly, individuals who quit smoking less than two years ago were assigned a score of 25 for smoking. The Dietary Approaches to Stop Hypertension (DASH)¹⁸ diet score was calculated based on components including the intake of vegetables, fruits, nuts, legumes, whole grains, and low-fat dairy, as well as the intake of red and processed meat, sugar-sweetened beverages, and sodium.¹⁹ The physical activity score was computed by considering the duration and intensity of activities, including sleep, sedentary behavior, and different levels of physical activity (i.e., light, moderate, and vigorous).²⁰ The other scores such as sleep quality, blood sugar levels, BMI, blood lipid, and blood pressure were derived based on AHA standards.¹¹ The LE8 total score was calculated by averaging the scores across all eight LE8 components. Additionally, two sub-scores, the LE8 health behavior score and the LE8 health factor score, were also obtained.¹¹

We defined “middle age” as the age range of 45 to 65 years across Exams 5–9,^21–23 due to the measurement of most LE8 components starting from Exam 5. Considering participants could have attended multiple examinations during middle age, their LE8 scores were averaged across the attended exams to calculate the mid-age LE8 scores. Furthermore, we classified the mid-age LE8 total score into three CVH categories: Poor-CVH group (mid-age LE8 total score below 50), Intermediate-CVH group (mid-age LE8 total score between 50 and 70), and Ideal-CVH group (mid-age LE8 total score above 70).¹¹ The mid-age LE8 scores were standardized to z-scores with a mean of 0 and a standard deviation (SD) of 1.

Construction of dCDT scores

Since 2011, FHS has conducted the dCDT, a digital version of the conventional Clock Drawing Test (cCDT). The details of the dCDT can be found in previous studies.^{12,13,24–26} This study utilized the dCDT total score, which serves as a composite measure of overall cognitive function. Additionally, we analyzed eight domain-specific scores: COM/COP Drawing Efficiency scores, which reflect the efficiency with which participants drew clocks in both the command and copy tasks; COM/COP Simple Motor scores, which assess the graphomotor skills involved in drawing the clocks; COM/COP Information Processing scores, which evaluate the non-motor cognitive functions used during the process of the command and copy clock drawing tasks; and COM/COP Spatial Reasoning scores, which gauge the spatial abilities used in arranging the elements of the clock during the drawing tasks.²⁵ All scores were standardized to a mean of 0 and an SD of 1.

Construction of cCDT scores

The comparison of cCDT and dCDT measures in their associations with mid-age CVH and AD is lacking. Due to differences in data collection, we included total drawing time (in seconds) and clock face drawing (spatial reasoning) scores from both the command (COM) and copy (COP) tasks, and assessed their associations with the mid-age LE8 total score and incident AD. To enhance comparability, we selected 16 features for the command task, and 15 features for the copy task from cCDT (Supplemental Table 10). We aligned the features to ensure that higher COM and COP scores consistently reflected better performance. Scores across all features were then summed and standardized to a mean of 0 and an SD of 1 for both tasks.

Diagnostic criteria for AD

The cognitive status of FHS participants has been consistently monitored through a comprehensive surveillance system, which includes cognitive screening at core health exams (e.g., using the Mini-Mental State Examination (MMSE)) and periodic assessments with the NP test battery.^6,27,28 For participants flagged with potential cognitive impairment, based on poor MMSE performance (≤24), significant declines in MMSE scores between successive examinations, or self- or family-reported concerns, additional follow-up NP exams are administered more frequently, typically every one to two years, with some participants also receiving neurological evaluations by an expert team.²⁹ This team, which includes at least one neurologist and one neuropsychologist, reviews all available data and reaches a consensus on whether participants meet the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria for dementia, assigning specific subtype diagnoses as needed.³⁰ The diagnostic criteria for AD adhere to the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA criteria).³¹

Covariates

Covariates included age at dCDT, delta age, sex, education level, and apolipoprotein E (APOE) genotype. The delta age was calculated as the difference in years between participants’ age at the time of the dCDT and their age at the last LE8 measurement during their middle-age period. Participants with APOE ε2/ε4 genotype were excluded. The remaining participants were stratified into two groups based on the presence (ε3/ε4 and ε4/ε4) or absence (ε2/ε3, ε2/ε2, and ε3/ε3) of the APOE ε4 allele.

Statistical analysis

Univariate analysis was conducted between each covariate and the z-scores of the mid-age LE8 total score and dCDT total score. Linear regression models were used to examine the associations of the dCDT total score with the LE8 total score and CVH categories. Furthermore, the pairwise associations between LE8 sub-scores and dCDT sub-scores were analyzed using linear regression models (Figure 1). All models were adjusted for age at dCDT, sex, education level, delta age, and APOE ε4 status. Multiple comparisons were adjusted using false discovery rate (FDR) correction.³²

To determine whether sex and the presence of the APOE ε4 allele modify the association between mid-age CVH and dCDT performance, we conducted analyses that included interaction terms (sex*CVH categories) and (APOE ε4 status*CVH categories) in the regression models. Sex-specific analyses were performed to further assess the variance in the effect size of the association between mid-age CVH and dCDT total score among different sex groups. Additionally, stratified analyses were conducted for APOE ε4 allele carriers and non-carriers to evaluate variations in effect size based on APOE ε4 status.

The Cox proportional hazards model was used to examine the association between the mid-age LE8 total score and incident AD. Covariates included age at the time of the last LE8 measurement during the middle-age period, sex, education level, and APOE ε4 status (Figure 1). For participants who developed incident AD, the follow-up period was calculated from the end of middle age until the earliest occurrence of AD onset. For those without incident AD diagnosis, the follow-up period was calculated from the end of middle age to death or until July 12, 2023. We assumed that participants who had no diagnosis of AD by July 12, 2023, had not developed the condition as of that date. We conducted a sensitivity analysis, considering death as a competing event to test the validity of the association between mid-age LE8 and incident AD. We also applied the Cox model using the categorical mid-age LE8 variable as the main predictor, while adjusting for the covariates. Additionally, we determined whether sex and the presence of the APOE ε4 allele modified the association between mid-age CVH and incident AD by conducting analyses that included the same interaction terms in the Cox models.

The association between dCDT performance and incident AD was examined using the Cox model, with the dCDT total score as the primary predictor and AD incidence as the dependent outcome (Figure 1). The covariates included age at dCDT, sex, education level, and APOE ε4 status. Follow-up time spanned from the dCDT assessment to the earliest occurrence of AD onset, death, or July 12, 2023. As a sensitivity analysis, we conducted an additional analysis accounting for death as a competing event between dCDT and incident AD. In the secondary analysis, we conducted the Cox model with the eight dCDT subdomain scores as the main predictors and AD onset as an outcome, adjusting for the same set of covariates. Additionally, we compared the performance of cCDT and dCDT for their associations with mid-age LE8 and incident AD. To predict AD using Cox models, we evaluated and compared the C-statistics for models incorporating cCDT and dCDT (Supplemental Methods). The proportional hazard assumptions for all Cox models were examined using the Schoenfeld residuals test. All analyses were performed using RStudio version 4.2.1.

Results

Participant characteristics

This study included 1198 participants with a mean age of 73.1 ± 6.9 years at dCDT measurement. The cohort comprised 51.6% of women, and 47.5% of the participants had attained a college education or higher. Detailed clinical characteristics are presented in Table 1. There was a significantly higher percentage of women in the ideal-CVH group compared to the other two CVH groups (68.0% versus 48.9% versus 48.8%, p < 0.001). Significant differences in educational level were observed across CVH categories, with a higher percentage of participants receiving advanced education in the ideal-CVH group (p < 0.001). Additionally, the mean dCDT total score differed significantly between CVH groups, with the group having better CVH exhibiting a higher mean dCDT total score (p < 0.001). During the median follow-up of 17.5 years from the last core exam in middle age, 45 participants developed AD. A significantly lower percentage of AD cases was observed in the ideal-CVH group compared to the other two groups (p < 0.015). The average time interval was approximately 10.7 ± 6.5 years from middle age to the dCDT measurement.

Table 1.

Characteristics of the 1198 participants in the Framingham heart study offspring cohort.

Variable	Poor CVH^a	Intermediate CVH^a	Ideal CVH^a	Total	p ^b
Variable	(n = 84)	(n = 617)	(n = 497)	(n = 1198)	p ^b
Age at last FHS exam in mid-age (y), mean (SD)	62.1 (2.3)	62.5 (2.1)	62.3 (2.1)	62.4 (2.1)	0.165
Age at dCDT (y), mean (SD)	73.5 (7.0)	73.4 (6.8)	72.6 (7.0)	73.1 (6.9)	0.128
Delta Age (y), mean (SD)	11.4 (6.5)	11.0 (6.5)	10.3 (6.5)	10.7 (6.5)	0.150
Women (%)	41 (48.8)	302 (48.9)	338 (68.0)	618 (51.6)	<0.001*
Education, n (%)
No high School	1 (1.2)	19 (3.1)	12 (2.4)	32 (2.7)	<0.001*
High School	19 (22.6)	143 (23.2)	85 (17.1)	247 (20.6)
Some College	38 (45.2)	182 (29.5)	130 (26.2)	350 (29.2)
College and higher	26 (31.0)	273 (44.2)	270 (54.3)	569 (47.5)
APOE ε4+, n (%)	19 (22.9)	151 (25.0)	99 (20.3)	269 (22.5)	0.194
Number of FHS exams^c in mid-age, median [IQR]	4 [3, 4]	4 [3, 4]	4 [3, 4]	4 [3, 4]	0.869
Mid-age LE8 scores, mean (SD)
Total score	44.69 (4.70)	61.25 (5.40)	77.93 (5.35)	67.0 (11.4)	<0.001*
Behavior score	45.31 (10.53)	60.67 (12.52)	74.17 (9.31)	65.2 (14.0)	<0.001*
Factor score	43.92 (11.17)	61.71 (12.35)	81.47 (10.20)	68.7 (16.3)	<0.001*
dCDT total score, mean (SD)	65.65 (26.67)	71.05 (23.25)	75.83 (20.88)	72.7 (22.7)	<0.001*
AD cases, n (%)	6 (7.1)	29 (4.7)	10 (2.0)	45 (3.8)	0.015*
Time-to-event from last FHS exam in mid-age^d (y), median [IQR]	17.6 [16.3, 23.9]	17.6 [12.2, 24.0]	17.4 [11.9, 23.8]	17.5 [12.1, 23.9]	0.251
Time-to-event from dCDT^e (y), median [IQR]	7.2 [5.9, 10.2]	8.9 [5.7, 9.9]	8.9 [5.7, 9.8]	8.9 [5.7, 9.9]	0.753

dCDT: digital Clock Drawing Test; y: year; SD: standard deviation; IQR: interquartile range.

CVH categories were derived based on LE8 total score: Ideal (>70), Intermediate (50–70), and Poor (<50).

p value, comparison of variables among the three CVH categories

The number of FHS core exams in middle age, where the LE8 information collected is utilized to calculate mid-age LE8 scores.

The time in years from the last mid-age FHS exam (LE8 measurement) to the onset of AD.

The time in years from the time of dCDT examination to the onset of AD. *indicates statistical significance.

Association of demographic variables with mid-age LE8 scores and dCDT scores

In the univariate analysis, each additional year of age corresponded to a 0.07 SD lower dCDT total score (95% CI: −0.08, −0.06; p < 0.001), suggesting reduced cognitive performance in older participants (Supplemental Table 1). Women demonstrated superior cognitive performance than men, compared to men, scoring on average 0.19 SD higher (95% CI: 0.03, 0.36) on the dCDT total score (p = 0.02). Additionally, a clear educational gradient was observed in dCDT performance. No significant association was observed between age or education level and the mid-age LE8 total score in the study participants (Supplemental Table 2).

Association between mid-age LE8 scores and dCDT scores

Every one-SD higher level of the mid-age LE8 total score was associated with a 0.16-SD higher dCDT total score (95% CI: 0.08, 0.24; p < 0.001) (Figure 2). Using participants in the intermediate CVH group as the reference, those within the ideal-CVH group had 0.17 SD higher dCDT total score (95% CI: 0.01, 0.34; p = 0.04). In contrast, those categorized into poor-CVH group had 0.36 SD lower dCDT total score (95% CI: −0.67, −0.05; p = 0.02) (Table 2).

Table 2.

Association of dCDT total score with LE8 total score and CVH categories.

Main predictor	Beta	95% CI	SE	p
LE8 total z-score	0.16	(0.08, 0.24)	0.04	<0.001*
CVH categories^a
Intermediate	Ref	–	–	–
Poor	−0.36	(−0.67, −0.05)	0.16	0.023*
Ideal	0.17	(0.01, 0.34)	0.08	0.040*

The association was examined by linear regression model adjusting for age at dCDT examination, delta age, sex, education, and APOE ε4 status.

CVH categories were derived based on LE8 total score: Ideal (>70), Intermediate (50–70), and Poor (<50). *indicates statistical significance.

A higher mid-age LE8 total score was associated with enhanced dCDT performance (Supplemental Figure 2). The mid-age LE8 total score was significantly associated with most dCDT sub-scores (FDR-adjusted p ≤ 0.03), except for the scores of drawing efficiency and information processing under the command task. Compared to the mid-age LE8 health behavior score, the mid-age LE8 health factor score showed significant associations with more dCDT sub-scores (4 versus 6) (Supplemental Figure 2).

Effect modification by sex

Sex significantly influenced the association between mid-age CVH categories and dCDT performance (Supplemental Figure 3), with a p-value of 0.018 for the interaction term (Supplemental Table 4). Every one-SD higher level of the mid-age LE8 total score was significantly associated with a 0.20-SD higher level of the dCDT total score among women (95% CI: 0.10, 0.30; p < 0.001) (Supplemental Table 3). For men, this association did not reach statistical significance (p = 0.18). Women in the poor-CVH group had, on average, a 0.79-SD lower dCDT total score than those in intermediate-CVH group (95% CI: −1.22, −0.35; p < 0.001) (Supplemental Table 3). This association was not observed in men.

Effect modification by APOE ε4 status

From the interaction analysis, we did not find that APOE ε4 status significantly modified the association between mid-age continuous (interaction p = 0.053) and categorical (interaction p = 0.25) LE8 variables with the dCDT total score (Supplemental Table 4). Given that APOE ε4 status is a major genetic risk factor for AD, we investigated the association of mid-age LE8 total score and dCDT total score in stratified samples (i.e., APOE ε4 carriers and non-carries). We observed variations in the magnitude of the association between dCDT scores and LE8 variables, in APOE ε4 carriers compared to non-carriers (Supplemental Figure 4). For example, among APOE ε4 non-carriers, a one-SD higher level of the mid-age LE8 total score was associated with a 0.12-SD higher level of the dCDT total score (95% CI: 0.03, 0.21; p = 0.009). In contrast, among APOE ε4 carriers, the association between mid-age LE8 total score and dCDT showed a larger effect size (beta = 0.29; 95% CI: 0.11, 0.47; p = 0.002) (Supplemental Table 5) compared to that among APOE ε4 non-carriers.

Association of mid-age LE8 scores with incident AD

Every one-SD higher mid-age LE8 total score was associated with a 35% lower risk of incident AD (HR: 0.65; 95% CI: 0.49, 0.87; p = 0.003) (Table 3). After accounting for death as a competing risk, the association remained similar (HR: 0.64; 95% CI: 0.49, 0.82; p < 0.001) (Supplemental Table 6). Compared with participants in intermediate-CVH group, ideal-CVH group appeared to have a 50% lower risk of AD (95% CI: 0.23, 0.98; p = 0.06). Conversely, participants in poor-CVH group in mid-age showed a 44% higher risk of incident AD (Supplemental Figure 5). For every one-SD higher in the mid-age LE8 behavior score, there was a 38% reduction in the risk of incident AD (95% CI: 0.47, 0.82; p < 0.001) (Supplemental Table 7). Neither sex nor APOE ε4 status significantly modified the relationship (all p > 0.05) between mid-age CVH, measured by both continuous LE8 total score or the categorical mid-age CVH variable, and AD risk. (Supplemental Table 8). The proportional hazards assumption was satisfied in all Cox models (all p > 0.14).

Table 3.

Association of incident AD with LE8 total score, CVH categories, and dCDT total score.

Main predictor	HR	95% CI	p
LE8 total z-score^a	0.65	(0.49, 0.87)	0.003*
CVH categories^b
Intermediate	Ref	–	–
Poor	1.44	(0.59, 3.53)	0.428
Ideal	0.50	(0.23, 0.98)	0.063
dCDT total score^c	0.52	(0.43, 0.62)	<0.001*

The association of AD with mid-age LE8 total z-score was examined by adjusting for age at last included FHS exam, sex, education, and APOE ε4 status.

CVH categories were derived based on LE8 total score: Ideal (>70), Intermediate (50–70), and Poor (<50). The association of AD and CVH categories (intermediate was used as the reference group) was examined by adjusting for age at last included LE8 measurement, sex, education, and APOE ε4 status.

The association of AD and dCDT total z-score was examined by adjusting for age at dCDT, sex, education, and APOE ε4 status. *indicates statistical significance.

Association of dCDT scores with incident AD

Each SD higher level of the dCDT total score was associated with a 48% lower hazard of incident AD (HR: 0.52; 95% CI: 0.43, 0.62; p < 0.001) (Table 3) with a median follow-up of 8.9 years (Table 1). In sensitivity analysis, we considered the competing risk of death. We found similar results (HR: 0.54; 95% CI: 0.46, 0.64; p < 0.001) (Supplemental Table 6). Significant associations between incident AD and digital cognitive performance in four domains under both command and copy tasks were observed (FDR-adjusted p < 0.001). Higher domain-specific cognitive scores were associated with lower hazards of developing AD, with HRs ranging from 0.52 to 0.68 (Supplemental Table 9).

Comparison between cCDT and dCDT

Three dCDT measures demonstrated more significant associations with mid-age LE8 total score compared to cCDT measures (Figure 3(a)). Specifically, the digital spatial reasoning domain scores (the drawing score) under both tasks showed stronger associations with mid-age LE8 total score compared to their cCDT counterparts (COM: beta = 0.10, p = 0.007 for dCDT versus beta = 0.03, p = 0.27 for cCDT; COP: beta = 0.08, p = 0.02 for dCDT versus beta = −0.04, p = 0.049) (Figure 3(a)).

Similarly, stronger discriminative power was observed with three of the dCDT measures in predicting AD compared to their cCDT counterparts (Figure 3(b)). For example, for total clock drawing time under the copy task, a higher C-statistic (0.87 versus 0.83) was found using dCDT for predicting incident AD compared to cCDT. For clock face drawing performance scores of both tasks, the models incorporating dCDT scores consistently had higher C-statistics compared to those using cCDT scores (COM: 0.89 versus 0.83; COP: 0.88 versus 0.81) (Figure 3(b)).

Discussion

Our study investigated the association between mid-age LE8 scores, dCDT performance, and incident AD. The results indicated that individuals with higher LE8 total scores in their middle ages tend to exhibit better cognitive performance and a reduced risk of incident AD later in life. In addition, higher mid-age LE8 health factor and lifestyle behavior scores were positively correlated with specific cognitive functions in older age. Furthermore, dCDT measures demonstrated stronger associations with the mid-age LE8 total score and better predictive power for AD risk compared to cCDT.

AD, a neurodegenerative condition, may progress for years before noticeable symptoms appear.³³ Previous studies have highlighted the importance of addressing mid-age risk factors to delay and prevent AD.^34,35 Our study emphasizes mid-age CVH as a critical period for implementing interventions before significant disease progression. Various methods for measuring CVH in middle age have been established, with evidence linking mid-age CVH to AD.^36,37 One study found that the mid-age Framingham Stroke Risk Profile was significantly associated with memory decline.³⁸ Other research using LS7 to quantify CVH, has indicated that higher LS7 scores protect cognition.^8,9,39–41 LE8 incorporates eight modifiable risk factors on a 100-point scale, offering a comprehensive CVH assessment.¹¹ Our findings indicate that superior mid-age LE8 scores are associated with better cognitive performance and reduced AD risk. We also observed differences in the associations between mid-age CVH and dCDT performance between men and women, suggesting that sex may play an important role in CVH and AD risk. Further studies are needed to validate our findings in larger and more diverse populations.

Digital cognitive assessment tests are becoming widely used and hold the potential to replace traditional paper-based tests as a rapid and more sensitive tool for detecting early cognitive decline. Establishing the relationship between CVH and these new modalities of cognitive performance is essential. Studies have explored the potential of dCDT to detect subtle changes in cognitive functions, demonstrating its capability for early detection of cognitive decline.^42–45 As a digital version of the traditional Clock Drawing Test, dCDT offers a more detailed capture of participant interactions through digital tools, such as a digital pen. However, the dCDT primarily assesses specific cognitive domains, such as visuospatial function, information processing abilities, simple motor function, and memory. In contrast, a conventional NP battery includes a broader range of tasks designed to measure various aspects of cognition, including language, orientation, attention, visuoperceptual skills, and more detailed aspects of memory. While dCDT provides valuable insights into specific cognitive functions, it may not capture the full spectrum of cognitive abilities as comprehensively as a conventional NP battery.

This study presents the first comparison of how cCDT and dCDT are associated with mid-age CVH and AD. We found that dCDT measures demonstrated stronger associations with the mid-age LE8 total score and outperformed cCDT in predicting AD, indicating that dCDT better reflects the impact of mid-age CVH on cognitive function and AD risk. Our findings are consistent with a prior study that compared the performance of dCDT with conventional NP tests, which found that composite scores based on dCDT were superior to their corresponding NP tests for predicting mild cognitive impairments.⁴⁶ Future studies are warranted to compare dCDT and additional digital measures (e.g., digital voice analysis) with conventional cognitive measures, to evaluate whether digital measures more effectively capture the effects of mid-age CVH on cognitive changes and better predict AD.

Genetic factors, particularly the APOE ε4 allele, are significantly associated with AD onset.^47,48 Our study found that APOE ε4 carriers with suboptimal CVH tend to have a higher risk of AD compared to those with ideal CVH. While a better mid-age LE8 score was positively associated with dCDT scores in both carriers and non-carriers, the association was stronger in carriers, though no significant interaction effects were found (p = 0.053 for LE8*APOE ε4). This differs from previous research, which found that higher LS7 scores protect against dementia in non-carriers, with no significant interaction between APOE ε4 status and LS7 scores.¹⁰ These discrepancies may reflect differences in our study's focus on mid-age CVH (mean age 62.4) over a longer follow-up period (17.5 years), versus previous studies examining older populations (mean age 75.3) with shorter follow-up periods (5.8 years). Additionally, the impact of modifiable risk factors, such as BMI, may vary by age.⁴⁹ Given the absence of significant interactions in both our study and prior research, larger studies are needed to draw definitive conclusions. Based on our findings, we recommend promoting healthy lifestyle habits and maintaining optimal CVH from an early age, especially for APOE ε4 carriers, to reduce the risk of AD.

Our study has several limitations. First, the FHS Offspring cohort primarily includes non-Hispanic Whites from an affluent town, which may limit the generalizability of our findings to more diverse populations.³⁶ Additionally, the associations between LE8, dCDT, and AD are based on a small number of incident cases. Larger studies with more AD cases and more diverse populations are needed. Since the dCDT measures specific cognitive functions, future research should explore the relationship between LE8 scores and a broader range of neuropsychological tests to better understand how CVH influences cognitive outcomes.^42,50 The main strengths of this study include the use of the longitudinal FHS Offspring cohort, which allows for the continuous assessment of AD diagnosis and related measurements, as well as the investigation of digital cognitive measures in relation to mid-age CVH and AD risk.

In summary, our results highlight the potential of digital cognitive assessments for evaluating AD risk and emphasize the importance of mid-age CVH in shaping cognitive outcomes and the development of AD.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251317734 - Supplemental material for Association of mid-age Life's Essential 8 score with digital cognitive performance and incident Alzheimer's disease: The Framingham Heart Study

Supplemental material, sj-docx-1-alz-10.1177_13872877251317734 for Association of mid-age Life's Essential 8 score with digital cognitive performance and incident Alzheimer's disease: The Framingham Heart Study by Jian Yang, Huitong Ding, Yi Li, Ting Fang Alvin Ang, Sherral Devine, Yulin Liu, Wendy Qiu, Rhoda Au, Jiantao Ma and Chunyu Liu in Journal of Alzheimer's Disease

Footnotes

Acknowledgments

We express our gratitude to the participants of the FHS for their commitment, acknowledging that this research would not have been possible without their involvement. Additionally, we extend our thanks to the FHS researchers for their sustained dedication to conducting subject examinations over the years.

ORCID iDs

Jian Yang

Huitong Ding

Rhoda Au

Chunyu Liu

Author contributions

Jian Yang (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing – original draft; Writing – review & editing); Huitong Ding (Investigation; Writing – original draft; Writing – review & editing); Yi Li (Data curation; Writing – review & editing); Ting Fang Alvin Ang (Resources; Writing – review & editing); Sherral Devine (Resources; Writing – review & editing); Yulin Liu (Resources; Writing – review & editing); Wendy Qiu (Resources; Writing – review & editing); Rhoda Au (Resources; Writing – review & editing); Jiantao Ma (Resources; Writing – review & editing); Chunyu Liu (Conceptualization; Funding acquisition; Methodology; Project administration; Supervision; Visualization; Writing – original draft; Writing – review & editing).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Heart, Lung, and Blood Institute contract (N01-HC-25195; HHSN268201500001I) and grants from the National Institute on Aging (AG008122, AG062109, AG068753).

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Au is a scientific advisor to Signant Health and Novo Nordisk. The remaining authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The data used in this study can be requested by submitting an application to the Framingham Heart Study ().

Supplemental material

Supplemental material for this article is available online.

References

Tiwari

Atluri

Kaushik

, et al. Alzheimer's disease: pathogenesis, diagnostics, and therapeutics. Int J Nanomedicine 2019; 14: 5541–5554.

Pase

Satizabal

Seshadri

. Role of improved vascular health in the declining incidence of dementia. Stroke 2017; 48: 2013–2020.

Stefanowski

Kucharski

Szeliga

, et al. Cognitive decline and dementia in women after menopause: prevention strategies. Maturitas 2023; 168: 53–61.

Povova

Ambroz

Bar

, et al. Epidemiological of and risk factors for Alzheimer's disease: a review. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2012; 156: 108–114.

Crous-Bou

Minguillon

Gramunt

, et al. Alzheimer's disease prevention: from risk factors to early intervention. Alzheimers Res Ther 2017; 9: 71.

Norton

Matthews

Barnes

, et al. Potential for primary prevention of Alzheimer's disease: an analysis of population-based data. Lancet Neurol 2014; 13: 788–794.

Han

You

, et al. National trends in American Heart Association revised life's simple 7 metrics associated with risk of mortality among US adults. JAMA Netw Open 2019; 2: e1913131.

Pase

Beiser

Enserro

, et al. Association of ideal cardiovascular health with vascular brain injury and incident dementia. Stroke 2016; 47: 1201–1206.

Zhao

, et al. Association between life's simple 7 and cerebrospinal fluid biomarkers of Alzheimer's disease pathology in cognitively intact adults: the CABLE study. Alzheimers Res Ther 2022; 14: 74.

10.

Guo

Brickman

Manly

, et al. Association of life's simple 7 with incident dementia and its modification by the apolipoprotein E genotype. Alzheimers Dement 2021; 17: 1905–1913.

11.

Lloyd-Jones

Allen

Anderson

CAM

, et al. Life's essential 8: updating and enhancing the American Heart Association's construct of cardiovascular health: a presidential advisory from the American Heart Association. Circulation 2022; 146: e18–e43.

12.

Davis

Libon

, et al. THink: inferring cognitive status from subtle behaviors. Proc AAAI Conf Artif Intell 2014; 2014: 2898–2905.

13.

Rentz

Papp

Mayblyum

, et al. Association of digital clock drawing with PET amyloid and tau pathology in normal older adults. Neurology 2021; 96: e1844–e1854.

14.

Kannel

Feinleib

McNamara

, et al. An investigation of coronary heart disease in families. The Framingham offspring study. Am J Epidemiol 1979; 110: 281–290.

15.

Feinleib

Kannel

Garrison

, et al. The Framingham offspring study. Design and preliminary data. Prev Med 1975; 4: 518–525.

16.

Benjamin

Brown

Burke

, et al. American heart Association cardiovascular genome-phenome study: foundational basis and program. Circulation 2015; 131: 100–112.

17.

Carbonneau

Liu

, et al. Abstract 13855: Epigenetic aging markers mediate the association of Life’s Essential Eight with cardiovascular disease and all-cause mortality. J Am Heart Assoc 2023; 148(Suppl_1). DOI: https://doi.org/10.1161/circ.148.suppl_1.13855.

18.

Appel

Moore

Obarzanek

, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH collaborative research group. N Engl J Med 1997; 336: 1117–1124.

19.

Mellen

Gao

Vitolins

, et al. Deteriorating dietary habits among adults with hypertension: DASH dietary accordance, NHANES 1988–1994 and 1999–2004. Arch Intern Med 2008; 168: 308–314.

20.

Kannel

Sorlie

. Some health benefits of physical activity. The Framingham study. Arch Intern Med 1979; 139: 857–861.

21.

Burgio

Matthews

Engel

. Prevalence, incidence and correlates of urinary incontinence in healthy, middle-aged women. J Urol 1991; 146: 1255–1259.

22.

Prineas

Soliman

, et al. United States National prevalence of electrocardiographic abnormalities in black and white middle-age (45- to 64-year) and older (>/=65-year) adults (from the reasons for geographic and racial differences in stroke study). Am J Cardiol 2012; 109: 1223–1228.

23.

Jiang

Zhu

Qin

. Relationships between chronic diseases and depression among middle-aged and elderly people in China: a prospective study from CHARLS. Curr Med Sci 2020; 40: 858–870.

24.

Gomes-Osman

Solana-Sanchez

Rogers

, et al. Aging in the digital age: using technology to increase the reach of the clinician expert and close the gap between health span and life span. Front Digit Health 2021; 3: 755008.

25.

Matusz

Price

Lamar

, et al. Dissociating statistically determined normal cognitive abilities and mild cognitive impairment subtypes with DCTclock. J Int Neuropsychol Soc 2023; 29: 148–158.

26.

Thompson

Cummings

Emrani

, et al. Digital clock drawing as an Alzheimer’s disease susceptibility biomarker: associations with genetic risk score and APOE in older adults. Prev Alzheimers Dis 2024; 11: 79–87.

27.

Chene

Beiser

, et al. Gender and incidence of dementia in the Framingham heart study from mid-adult life. Alzheimers Dement 2015; 11: 310–320.

28.

Ahl

Beiser

Seshadri

, et al. Defining MCI in the Framingham heart study offspring: education versus WRAT-based norms. Alzheimer Dis Assoc Disord 2013; 27: 330–336.

29.

Yuan

Maserejian

Liu

, et al. Severity distribution of Alzheimer's disease dementia and mild cognitive impairment in the Framingham heart study. J Alzheimers Dis 2021; 79: 807–817.

30.

Guze

. Diagnostic and statistical manual of mental disorders, (DSM-IV). Am J Psychiatry 1995; 152: 1228–1228.

31.

McKhann

Drachman

Folstein

, et al. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA work group* under the auspices of department of health and human services task force on Alzheimer's disease. Neurology 1984; 34: 939–939.

32.

Benjamini

Hochberg

. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc 1995; 57: 289–300.

33.

Wolk

Detre

. Arterial spin labeling MRI: an emerging biomarker for Alzheimer's disease and other neurodegenerative conditions. Curr Opin Neurol 2012; 25: 421–428.

34.

Zhao

Liu

, et al. Favorable cardiovascular health at young and middle ages and dementia in older age-the CHA study. J Am Heart Assoc 2019; 8: e009730.

35.

Lee

Back

Kim

, et al. Systematic review of health behavioral risks and cognitive health in older adults. Int Psychogeriatr 2010; 22: 174–187.

36.

Armstrong

Bangen

, et al. Associations between midlife (but not late-life) elevated coronary heart disease risk and lower cognitive performance: results from the Framingham offspring study. Am J Epidemiol 2019; 188: 2175–2187.

37.

Bleckwenn

Kleineidam

Wagner

, et al. Impact of coronary heart disease on cognitive decline in Alzheimer's disease: a prospective longitudinal cohort study in primary care. Br J Gen Pract 2017; 67: e111–e117.

38.

Gupta

Preis

Beiser

, et al. Mid-life cardiovascular risk impacts memory function: the Framingham offspring study. Alzheimer Dis Assoc Disord 2015; 29: 117–123.

39.

Peloso

Beiser

Satizabal

, et al. Cardiovascular health, genetic risk, and risk of dementia in the Framingham heart study. Neurology 2020; 95: e1341–e1350.

40.

Tin

Bressler

Simino

, et al. Genetic risk, midlife life's simple 7, and incident dementia in the atherosclerosis risk in communities study. Neurology 2022; 99: e154–e163.

41.

Malik

Georgakis

Neitzel

, et al. Midlife vascular risk factors and risk of incident dementia: longitudinal cohort and Mendelian randomization analyses in the UK Biobank. Alzheimers Dement 2021; 17: 1422–1431.

42.

Souillard-Mandar

Davis

Rudin

, et al. Learning classification models of cognitive conditions from subtle behaviors in the digital clock drawing test. Mach Learn 2016; 102: 393–441.

43.

Ding

Sunderaraman

Devine

, et al. The utility of the digital clock drawing test as an accurate early screening tool for those with preclinical Alzheimer’s disease. Alzheimers Dement 2023; 19(Suppl 15): e075294.

44.

Muller

Herde

Preische

, et al. Diagnostic value of digital clock drawing test in comparison with CERAD neuropsychological battery total score for discrimination of patients in the early course of Alzheimer's disease from healthy individuals. Sci Rep 2019; 9: 3543.

45.

Souillard-Mandar

Penney

Schaible

, et al. DCTclock: clinically-interpretable and automated artificial intelligence analysis of drawing behavior for capturing cognition. Front Digit Health 2021; 3: 750661.

46.

Yuan

Libon

Karjadi

, et al. Association between the digital clock drawing test and neuropsychological test performance: large community-based prospective cohort (Framingham heart study). J Med Internet Res 2021; 23: e27407.

47.

Crean

Ward

Mercaldi

, et al. Apolipoprotein E epsilon4 prevalence in Alzheimer's disease patients varies across global populations: a systematic literature review and meta-analysis. Dement Geriatr Cogn Disord 2011; 31: 20–30.

48.

Jefferson

Beiser

Seshadri

, et al. APOE And mild cognitive impairment: the Framingham heart study. Age Ageing 2015; 44: 307–311.

49.

Pedditzi

Peters

Beckett

. The risk of overweight/obesity in mid-life and late life for the development of dementia: a systematic review and meta-analysis of longitudinal studies. Age Ageing 2016; 45: 14–21.

50.

Santisteban

Iadecola

. Hypertension, dietary salt and cognitive impairment. J Cereb Blood Flow Metab 2018; 38: 2112–2128.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.19 MB