Associations of healthy aging index with dementia: a prospective cohort study of 305,566 individuals

doi:10.21203/rs.3.rs-4999254/v1

Download PDF

Research Article

Associations of healthy aging index with dementia: a prospective cohort study of 305,566 individuals

https://doi.org/10.21203/rs.3.rs-4999254/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Purpose: To prospectively assess whether Healthy Aging Index (HAI) is associated with the risk of dementia, and to investigate the potential relationship between HAI and genetic susceptibility in the development of Alzheimer's disease (AD).

Methods: The study included 305,566 participants from the UK Biobank, with a median follow-up of 14.01 years. HAI components were systolic blood pressure, reaction time, forced vital capacity, serum cystatin c, and serum glucose. Cox proportional hazard regression models assessed the associations between HAI and dementia risk.

Results: Adjusted analyses showed that each one-point increase in HAI was associated with a 15% higher risk of all-cause dementia (HR 1.15, 95% CI 1.13-1.17), a 10% higher risk of AD (HR 1.10, 95% CI 1.07-1.13), and a 29% higher risk of VD (HR 1.29, 95% CI 1.24-1.34). Restricted cubic spline (RCS) analyses confirmed these results. Further, the combined effects of HAI and AD genetic risk score (AD-GRS) were assessed. Participants with HAI scores of 7-10 had a higher risk of AD at intermediate (HR 1.68, 95% CI 1.16-2.44) and high AD-GRS levels (HR 2.43, 95% CI 1.66-3.56).

Conclusion: Our study indicates that a higher HAI score is associated with an increased risk of dementia. Early intervention to lower HAI scores in early adulthood, such as adopting a healthy lifestyle, can attenuate these associations.

Healthy aging index

dementia

genetic susceptibility

UK Biobank

cohort study

Amid a global shift towards an aging population, dementia has emerged as a critical public health challenge, with the number of affected individuals projected to soar from 574 million in 2019 to 1.58 billion by 2050(1). This sharp increase in prevalence places immense economic and caregiving burdens on both families and society, with associated costs expected to more than double, from $1.3 trillion annually in 2019 to $2.8 trillion by 2030(2). Furthermore, a staggering 75% of caregivers struggle to balance the demands of dementia care with other responsibilities, underscoring the pervasive impact of this condition on quality of life(3). Currently, no reliable pharmaceutical interventions exist to delay the onset or progression of dementia(4), highlighting the critical and urgent need for effective primary prevention strategies.

Advancements in evidence-based medicine have identified numerous risk factors for dementia, facilitating the development of prevention strategies(5). While aging is the most significant risk factor for chronic diseases, including dementia(6), the aging process itself varies considerably among individuals. This variation is influenced by living conditions, lifestyle, and genetic composition, all of which contribute to differences in disease susceptibility(7). To better understand the impact of aging on health, the Healthy Aging Index (HAI) was introduced by Newman et al.(8) and later refined by Sanders et al.(9) and Wu et al.(10). This comprehensive index, constructed from easily measurable biomarkers, provides an integrated assessment of an individual's health status through five non-invasive clinical checks: systolic blood pressure (SBP), reaction time (RT), forced vital capacity (FVC), serum glucose (SG), and serum cystatin C (SCysC). These markers, which span the circulatory, nervous, respiratory, urinary, and endocrine systems, have proven valuable in predicting adverse outcomes such as death, disability, limited activity, or cardiovascular events in elderly populations in both China and the United States(11–14).

Despite the growing body of research on HAI, its relationship with dementia remains unexplored. Given that elevated SBP or abnormal fasting SG in early adulthood has been linked to cognitive decline in later years(15), and that declining lung function increases the risk of all-cause dementia (ACD) and its subtypes(16). It is plausible that higher HAI correlate with an increased risk of dementia. Exploring this potential association is crucial, as it may provide a new avenue for early intervention and prevention.

Moreover, the heterogeneity of dementia, potentially arising from genetic variations or the interplay between genetic and environmental factors(17), underscores the importance of investigating gene-environment interactions in Alzheimer's disease (AD). Understanding these interactions is essential for developing personalized early interventions that could reduce the global incidence of AD(18). During aging, epigenetic changes, particularly DNA methylation patterns, have emerged as key indicators for assessing biological age(19). However, the relationship between the HAI and genetic susceptibility to AD risk remains unexplored.

This study aims to bridge existing gaps by exploring the association between the HAI and ACD, including its subtypes—AD and vascular dementia (VD)—using UK Biobank (UKB). Additionally, we seek to quantify the contribution of HAI to dementia and explore the correlation between genetic susceptibility and the risk linked with HAI and AD.

Study population

We analyzed data from the UKB, a long-term research project initiated by the UK government, expected to span 30 years(20). The UKB is a prospective epidemiological study that recruited over 500,000 participants between 2006 and 2010(21). It collects longitudinal follow-up data on genotypes, extensive phenotypes, physical measurements, health records, and multimodal imaging through touchscreen questionnaires, interviews, body measurements, and biometric samples(22). The UKB project resources are openly accessible to researchers adhering to ethical and scientific standards. All participants provided informed consent via electronic signature at baseline registration. The study received approval from the National Health and Social Care Information Management Committee and the NHS Northwest Multicenter Research Ethics Committee(23). For more comprehensive details on the biobank data, refer to https://www.ukbiobank.ac.uk/. This study obtained approval from the UKB under application ID 76636.

After excluding participants diagnosed with dementia at baseline and those with missing HAI components or covariate information, the primary analysis included 305,566 participants (Fig. 1).

Components of the Healthy Aging Index

The exposure factor in our study is the HAI, a composite measure derived from five parameters: SBP, FVC, SG, RT, and SCysC(24, 25). Each parameter was scored from 0 to 2, with higher scores reflecting poorer health. SG levels were base on the American Diabetes Association's clinical cutoff values for random blood glucose: 0: ≤140 mg/dL, 1:140–200 mg/dL, and 2:≥200 mg/dL(26). Detailed Field IDs and cutoff values are provided in Supplementary Table 1. According to previous studies, the cumulative scores were used to categorize participants into four health groups: 0–2 (healthiest), 3–4 (relatively healthy), 5–6 (less healthy), and 7–10 (least healthy)(11, 12, 27).

Outcomes

The primary endpoint is ACD, with secondary endpoints including AD and VD. The outcome events were defined using UKB’s algorithmically-defined outcomes (Field ID: 42018–42025) and hospital diagnosis records. Dementia cases were identified using International Classification of Diseases (ICD)-9 and ICD-10 codes, as detailed in Supplementary Table 2. The most recent follow-up was conducted on March 12, 2023.

Confounders

This study considers potential confounding variables, including demographic factors like age, sex (female/male), ethnicity (white/other), lifestyle factors including Townsend Deprivation Index (TDI), education level (college or university degree or other professional qualifications/other)(28), smoking status (never/previous/current), alcohol drinker status(never/previous/current), daily sleep duration, level of physical activity (high/moderate/low), and health diet score (0–5 points). Biomedical factors such as body mass index (BMI) are also accounted for. The health diet score evaluates overall dietary patterns, incorporating factors such as daily consumption of at least three servings of fruits, a minimum of four tablespoons of vegetables per day, consumption of fish at least twice a week, and limitations on processed and unprocessed red meat intake. Scores for each dietary factor are aggregated to derive a final dietary score ranging from 0 to 5(29, 30).

Assessment of genetic susceptibility to Alzheimer's disease

The Polygenic Risk Score (PRS) serves as a common tool for genetic risk evaluation in AD. For this investigation, we identified 29 SNP sites highly correlated with AD from previous Genome-wide Association Studies (GWAS)(31–33). Corresponding weights (β coefficients) were assigned to each SNP locus (Supplementary Table 3), resulting in the derivation of the Alzheimer's Disease Genetic Risk Score (AD-GRS) for each participant(34). Participants were stratified into three classes of AD genetic risk: high (quartile 4), moderate (quartiles 2–3), or low (quartile 1).

Statistical analyses

A Cox proportional hazards model was employed to estimate hazard ratios (HR) and 95% confidence intervals (CIs) of HAI in relation to ACD and its subtypes (AD and VD)(35). Follow-up person-years were calculated from baseline to the date of death or the end of follow-up, whichever came first. Separate models were developed to examine potential confounding effects on the association between HAI levels and dementia risk.The specific covariates were detailed in Supplementary Table 4. Kaplan-Meier (K-M) curves were also applied to assess cumulative risk across different HAI groups (0–2, 3–4, 5–6, 7–10) for ACD, AD and VD.

To elucidate the nonlinear relationship between HAI scores and the three dementia outcomes, we utilized a restricted cubic spline (RCS) model with four knots positioned at the 5th, 35th, 65th, and 95th percentiles. The placement of knots was guided by Harrell's recommendations for optimal model fitting(36). The RCS model was adjusted for the three sets (Model 3) of covariates previously mentioned, and nonlinear p-values were calculated for each model.

To enhance the robustness of our findings and minimize confounding errors, we conducted three sensitivity analyses. First, we refined the original model by accounting for the impact of cardiovascular risk factors and AD genetic risk factors on dementia risk. The specific covariates included were detailed in Supplementary Table 4.To minimize reverse causation, we excluded individuals with follow-up periods of less than 5 years post-baseline dementia diagnosis. We also retained and imputed missing data for a population of 385,556 individuals. Considering the stability of variables such as age, sex, race, education level, smoking status, alcohol intake, physical activity, and BMI during follow-up, we performed a rigorous subgroup analysis for All-cause dementia risk. Likelihood ratio tests were separately conducted for All-cause dementia, AD, and VD risks to generate overall interaction results.

We employed a Cox proportional hazards model to analyze the combined association of HAI and AD-GRS with AD risk, adjusting for age, gender, race, TDI, education level, sleep duration, smoking status, alcohol consumption, dietary health score, physical activity level, and BMI. Interaction between HAI scores and AD-GRS on AD risk was assessed using likelihood ratio tests. To further investigate the direct association between gene-environment combinations and AD risk, participants were categorized into 12 groups based on HAI scores (0–2, 3–4, 5–6, 7–10) and AD-GRS (low, medium, high). The healthiest group with low or medium AD-GRS served as reference groups for evaluating the AD risk associated with other combinations.

All statistical analyses were performed using R (version 4.3.1), with a two-tailed p-value < 0.05 indicating statistical significance.

Baseline characteristics

Our study comprised 301,518 participants with a median follow-up of 14.01 years. Dementia patients were typically older, male, Caucasian, and had higher TDI and BMI (Table 1). Compared to the control group, dementia patients exhibited higher SBP, RT, SG, and SCysC levels, while FVC was lower (all P-values < 0.01).

Table 1

Baseline characteristics of study participants by incident dementia status.
Characteristic	Overall	Non-dementia	Dementia	P value
N	301518	296739	4779
Demographic factors
Age(year)	57 (50,63)	57 (49,63)	65 (62,68)	< 0.001
Sex (%)				< 0.001
Female	158188 (52.5)	156040 (52.6)	2148 (44.9)
Male	143330 (47.5)	140699 (47.4)	2631 (55.1)
Ethnicity (%)				< 0.001
White	288091 (95.5)	283468 (95.5)	4623 (96.7)
Other	13427 (4.5)	13271 (4.5)	156 (3.3)
Lifestyle factors
Education (%)				< 0.001
High	142166 (47.2)	140438 (47.3)	1728 (36.2)
Low	159352 (52.8)	156301 (52.7)	3051 (63.8)
TDI	-2.3 (-3.7,0.2)	-2.3 (-3.7,0.2)	−2.1 (−3.6,0.7)	< 0.001
Smoke status (%)				< 0.001
Current	29474 (9.8)	29025 (9.8)	449 (9.4)
Previous	105243 (34.9)	103115 (34.7)	2128 (44.5)
Never	166801 (55.3)	164599 (55.5)	2202 (46.1)
Alcohol drinking status (%)				< 0.001
Current	280854 (93.1)	276647 (93.2)	4207 (88)
Previous	9682 (3.2)	9390 (3.2)	292 (6.1)
Never	10982 (3.6)	10702 (3.6)	280 (5.9)
Sleep hours(h)	7 (7,8)	7 (7,8)	7 (7,8)	< 0.001
TPA group (%)				0.102
Low	55406 (18.4)	54529 (18.4)	877 (18.4)
Moderate	123423 (40.9)	121400 (40.9)	2023 (42.3)
High	122689 (40.7)	120810 (40.7)	1879 (39.3)
Healthy diet score (%)				< 0.001
0(Unhealthy)	6073 (2)	5987 (2)	86 (1.8)
1	29423 (9.8)	29037 (9.8)	386 (8.1)
2	63185 (21)	62277 (21)	908 (19)
3	82725 (27.4)	81392 (27.4)	1333 (27.9)
4	77393 (25.7)	76141 (25.7)	1252 (26.2)
5(Healthy)	42719 (14.2)	41905 (14.1)	814 (17)
BMI(kg/m² )	26.6(24,29.6)	26.6(24,29.6)	26.9(24.2,29.9)	< 0.001
Single item of HAI
SBP (mmHg)	136(124.5,149)	135.5(124,148.5)	142.5(130,155.5)	< 0.001
RT(ms)	532 (477,605)	532 (477,605)	581 (516,665)	< 0.001
FVC(liters)	3.7 (3.1,4.5)	3.7 (3.1,4.5)	3.4 (2.8,4.2)	< 0.001
SG (mg/dl)	4.9 (4.6,5.3)	4.9 (4.6,5.3)	5.1 (4.7,5.5)	< 0.001
Cystatin C(mg/dl)	0.9 (0.8,1)	0.9 (0.8,1)	0.9 (0.9,1.1)	< 0.001
Note: TDI, Townsend deprivation index; BMI, body mass index; SBP, systolic blood pressure;RT, reaction time;FVC, forced vital capacity; and SG, serum glucose.

The hazard ratios for all-cause dementia, AD, and VD

The median follow-up duration for participants was 14.01 years (IQR 13.27–14.72; total follow-up 4,121,814 person-years). During this period, 4,779 participants developed ACD, including 2,172 with AD and 1,040 with VD. As presented in Table 2, after adjusting for multiple variables (Model 3), higher HAI scores were significantly associated with increased dementia risk. Specifically, each additional HAI point raised the risk of ACD by 15% (HR = 1.15, 95%CI:1.13–1.17), AD by 10% (HR = 1.10, 95%CI:1.07–1.13), and VD by 29.2% (HR = 1.29, 95%CI: 1.24–1.34). The strongest associations were observed in the least healthy group. RCS models, adjusted for confounders, indicated a non-linear relationship between HAI and dementia risk, with significant p-values for non-linearity in ACD (p < 0.0001), AD (p = 0.0049), and VD (p = 0.0016). The RCS curves displayed a reverse L-shape, showing a marked increase in HR values for the least healthy group (Fig. 2). K-M curves demonstrated a significantly higher cumulative risk for the least healthy group compared to other groups (Supplementary Fig. 1).

Table 2

Hazard ratios for outcomes associated with HAI Categories.
	Per score point (n = 301,518)	Total HAI score				P
	Per score point (n = 301,518)	0–2 (n = 82,182)	3–4 (n = 97,209)	5–6 (n = 82,022)	7–10 (n = 40,105)	P
All-cause dementia
Person-years at risk	4,121,814	1,150,413	1,340,343	1,109,240	521,818	-
Crude cumulative incidence, cases (%)	4,779(1.58)	313(0.38)	1,055(1.09)	1,787(2.18)	1,624(4.05)	-
Incidence per 100,000 person-years	115.94	27.21	78.71	161.10	311.22	-
Model 1	1.153(1.135,1.172)	Reference	1.221(1.074,1.387)	1.479(1.305,1.676)	2.129(1.873,2.421)	< 0.001
Model 2	1.135(1.117,1.154)	Reference	1.196(1.052,1.360)	1.422(1.254,1.612)	1.961(1.723,2.232)	< 0.001
Model 3	1.146(1.126,1.166)	Reference	1.213(1.066,1.379)	1.460(1.286,1.659)	2.038(1.783,2.330)	< 0.001
AD
Person-years at risk	4,129,137	1,150,891	1,341,957	1,111,960	524,329	-
Crude cumulative incidence, cases (%)	2172(0.72)	141(0.17)	521(0.54)	815(0.99)	695(1.73)	-
Incidence per 100,000 person-years	52.60	12.25	38.82	73.29	132.55	-
Model 1	1.093(1.067,1.119)	Reference	1.217(1.008,1.470)	1.293(1.075,1.556)	1.694(1.401,2.050)	< 0.001
Model 2	1.076(1.050,1.102)	Reference	1.186(0.983,1.433)	1.230(1.021,1.481)	1.548(1.277,1.875)	< 0.001
Model 3	1.098(1.070,1.127)	Reference	1.227(1.015,1.483)	1.314(1.089,1.586)	1.720(1.410,2.097)	< 0.001
VD
Person-years at risk	4,131,695	1,151,132	1,342,806	1,112,932	524,825	-
Crude cumulative incidence, cases (%)	1,040(0.34)	39(0.05)	180(0.19)	360(0.44)	461(1.15)	-
Incidence per 100,000 person-years	25.17	3.39	13.40	32.35	87.84	-
Model 1	1.328(1.283,1.375)	Reference	1.554(1.096,2.205)	2.137(1.523,2.999)	4.222(3.007,5.928)	< 0.001
Model 2	1.304(1.259,1.351)	Reference	1.527(1.077,2.166)	2.071(1.475,2.907)	3.892(2.768,5.473)	< 0.001
Model 3	1.292(1.244,1.341)	Reference	1.491(1.051,2.116)	1.970(1.400,2.772)	3.569(2.519,5.055)	< 0.001
Note: HR, hazard ratio; CI, confidence interval; Model 1 adjusted for age; Model 2 adjusted for age, sex, Townsend deprivation index, ethnicity, education; Model 3 adjusted for age, sex, Townsend deprivation index, ethnicity, education, alcohol drinking status, smoking status, sleep hours, Body Mass Index, physical activity group, healthy diet score.

Sensitivity and subgroup analysis

We conducted three sensitivity analyses to ensure robustness. First, in Model 6, which included additional cardiovascular risk factors and AD genetic covariates, HAI remained significantly associated with dementia risk, increasing by 8.9% for ACD, 5.2% for AD, and 15.2% for VD per HAI point (Supplementary Table 5). Second, we excluded individuals with follow-up within 5 year after a dementia diagnosis at baseline. This exclusion did not lead to a weakening of the statistical difference for ACD and VD (Supplementary Table 6). Lastly, imputing missing covariate values yielded results consistent with the main analysis (Supplementary Table 7).

In subgroup analyses, we stratified participants by age, sex, ethnicity, education, smoking status, alcohol consumption, BMI, and physical activity. For ACD, the effect estimates were more pronounced in the younger group (< 65 years) compared to the older group (≥ 65 years; P interaction < 0.001) (Supplementary Table 9). Notably, individuals with the least healthy group, who were current or former smokers, current or former drinkers, and had lower physical activity levels, showed an increased risk of all-cause dementia. No significant differences were observed in relation to sex, education level, or BMI (Supplementary Fig. 2).

The Joint association between HAI and AD‑GRS for AD risk

A Cox proportional hazards model was employed to investigate the association between HAI and AD-GRS with respect to AD risk. In the fully adjusted model, a significant interaction between HAI and AD-GRS was observed (P = 0.044). After adjusting for multiple variables, an increasing trend in HR was noted with elevated AD-GRS and HAI. Stratifying risk based on AD-GRS and grouping HAI resulted in 12 groups. Comparing the least healthy group to the healthiest group within the context of low AD-GRS risk, the least healthy group showed a significantly higher risk of AD in both moderate AD-GRS (HR = 1.679, 95% CI:1.16–2.44, P = 0.006) and high AD-GRS (HR = 2.43, 95% CI:1.66–3.56, P < 0.001) (Fig. 3A). Similar findings were observed when using the healthiest group with moderate AD-GRS as the reference (Fig. 3B).

In this study, harnessing the extensive dataset and prolonged tracking period of the UKB, we explored the relationship between HAI and dementia, encompassing its subtypes. Initially, we noted that heightened HAI levels corresponded to escalated risks of ACD, AD, and VD. Specifically, each incremental HAI point amplified risks by 15% for ACD, 10% for AD, and 29% for VD. Subsequent nonlinear scrutiny unveiled a progressively ascending reverse L-shaped pattern across all outcomes, with the apex risk evident in the the least healthy group. Moreover, individuals with higher AD-GRS within the the least healthy group exhibited notably heightened susceptibility to AD in contrast to those in the healthiest group. Of equal importance is the discernment from subgroup analysis, accentuating the more pronounced impact of HAI in the younger demographic compared to the elderly. In essence, our findings underscore the pivotal role of HAI as a pivotal gauge for gauging dementia risk.

Prior studies have associated variations in individual HAI components with dementia risk. For instance, a Rotterdam study involving over 4,765 adults above the age of 65 investigated the relationship between FEV1/FVC, FEV1, FVC, and ACD, revealing a connection between decreased FVC and dementia risk(37). Similarly, a combined analysis of the CARDIA, MESA, CHS, and Health ABC cohorts found that elevated early adulthood SBP or abnormal fasting glucose correlates with later-life cognitive decline(15). However, research specifically addressing the relationship between HAI and dementia risk is lacking. Additionally, individual HAI components such as FVC, SBP, and SG, when amalgamated to form HAI, surpass the confines of representing signs of functional decline in a single organ, offering a more nuanced and potentially clinically relevant metric for unhealthy aging(11). Our primary discovery underscores that an elevated HAI score augments the risk of dementia, marking the first exploration of the correlation between HAI scores and ACD and its subtypes to our knowledge.

Numerous potential pathways connect HAI with dementia. Firstly, chronic inflammation and neuroinflammation play pivotal roles in dementia development(38). Brain chronic inflammation releases various pro-inflammatory and toxic agents, including cytokines, chemokines, reactive oxygen species, and nitric oxide(39), amplifying immune responses, leading to neurotoxicity, and escalating neurodegenerative risks, notably AD. Aging-related chronic inflammation is a low-grade chronic inflammatory state that develops with age, with elevated levels of inflammatory factors like C-reactive protein (CRP) in the serum(40). Our study found that as HAI increased, serum CRP levels also rose. Elevated CRP is associated with acute and chronic inflammation from various diseases, including infectious and non-infectious inflammatory conditions, and is a recognized inflammatory marker(41). This suggests that individuals with high HAI may be more susceptible to chronic inflammation, potentially leading to neuronal damage, neurodegeneration, and brain inflammation, thereby increasing the risk of dementia. Secondly, elevated systolic blood pressure and blood glucose levels in HAI components serve as cardiovascular risk markers, provoking systemic microvascular damage, which affects brain structures, culminating in cognitive decline and the development of dementia(42, 43). Thirdly, diminished FVC indicates reduced respiratory capacity, leading to decreased oxygen intake, a critical factor given the brain's high oxygen consumption(44). Hypoxia-induced cognitive impairment and dementia entail Aβ protein accumulation, heightened phosphorylated Tau expression, and synaptic dysfunction(45, 46). FVC reduction partly contributes to elevated HAI. By assessing multiple physiological systems, the HAI offers a comprehensive appraisal of individual health status, displaying promising discriminatory potential for dementia cohorts.

Our study suggests expanding the use of HAI indicators beyond the elderly, as younger individuals face a heightened dementia risk. Subgroup analysis shows that younger individuals (≤ 44 years) have a nearly 30% greater risk of cardiovascular events associated with higher HAI levels, compared to a 20% increase in those aged ≥ 60 years(24). Similarly, younger participants (< 60 years) exhibit a 19% higher mortality risk, versus a 15% increase in older participants(25). These findings underscore the potential influence of both the intensity and duration of HAI exposure on the lifelong risk of dementia events. Younger individuals burdened with higher HAI levels are predisposed to prolonged exposure to aging-associated factors. Studies indicate that early-onset conditions like hypertension, diabetes, and obesity lead to a cognitive decline rate more than twice that of their healthier counterparts over the subsequent decade(15). Furthermore, enhancing cardiovascular health in early adulthood is linked to diminished dementia risk in later life(47). Thus, early intervention or amelioration strategies are advisable for individuals with elevated HAI scores to mitigate future dementia risks.

Strengths and limitations

The primary strengths of this study include: Firstly, it draws upon prospective cohort data from the UKB, offering extensive and enduring follow-up. Secondly, meticulous adjustment for diverse dementia-related confounders and comprehensive sensitivity analyses enhance result accuracy and robustness. Thirdly, it pioneers the investigation into the relationship between HAI scores, genetic predisposition, and dementia risk. Nonetheless, we recognize certain limitations. Primarily, the predominant representation of white individuals in the UKB warrants caution when generalizing findings to other racial demographics. Additionally, despite meticulous adjustment for numerous dementia-related confounders, the potential for unidentified confounding effects remains. Moreover, HAI components are assessed solely at baseline, and although averaging multiple measurements minimizes errors, variations over time may lead to misclassification. Future research should track HAI trajectories and explore their supplementary prognostic utility.Thirdly, relying on hospital inpatient and death records to identify dementia could lead to misclassification bias, as some dementia cases may go unrecognized if they are not captured through self-reports or ICD-9/10 codes.

In summary, our study has revealed for the first time a non-linearly increasing association between HAI and the risk of dementia. HAI can be a useful tool for screening high-risk populations for dementia. Moreover, individuals with a high AD-GRS demonstrated an increased risk of developing AD, particularly with a high HAI score (7-10 point). It is crucial to adopt a healthy lifestyle early on to prevent the continuous increase of HAI.

ACD=all-cause dementia

AD = Alzheimer’ s disease

AD-GRS = Alzheimer’ s disease genetic risk score

BMI = body mass index

CIs = confidence intervals

CRP = C-reactive protein

FVC = forced vital capacity

GWAS = genome-wide association study

HAI = healthy aging index

HR = hazard ratio

ICD = International Classification of Diseases

IQR = interquartile range

K-M = Kaplan-Meier

PRS = Polygenic Risk Score

RCS = restricted cubic spline

RT = reaction time

SBP = systolic blood pressure

SCysC = serum cystatin C

SG = serum glucose

SNPs = single nucleotide polymorphisms

TDI = Townsend deprivation index

UKB=UK Biobank

VD = vascular dementia

Ethics approval and consent to participate: UK Biobank data has approval from the North West Multi-centre Research Ethics Committee (MREC) (REC reference: 21/NW/0157). This research has been conducted with the UK Biobank Resource under projec no. 76636.

Consent for publication:Written informed consent for publication was obtained from all participants.

Availability of data and materials：Not applicable.

Competing interests：The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding：Not applicable.

Authors' contributions：

Shanyuan Tan:Conceptualization,Methodology,Validation,Writing - Original Draft Xiaxuan Huang: Methodology,Validation,Writing - Review & Editing

Yitong Ling:Formal analysis,Data Curation,Visualization

Shiqi Yuan: Software

Zihong Bai:Investigation

Hongtao Cheng:Resources

Anding Xu:Supervision,Project administration

Jun Lyu:Supervision,Funding acquisition

Acknowledgements：We are grateful to the original data creators, depositors, copyright holders, and the funders of the data collections for their efforts and contributions, which have enabled access to data from the UK Biobank team. (project no. 76636).

Collaborators. GDF. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. The Lancet Public health. 2022;7(2):e105-e25.
Long S, Benoist, C., Weidner, W. World Alzheimer Report 2023: Reducing dementia risk: never too early, never too late. In: London EAsDI, editor. 2023.
International. AsD. World Alzheimer Report 2019: Attitudes to dementia. In: International LAsD, editor. 2019.
Sexton CE, Anstey KJ, Baldacci F, Barnum CJ, Barron AM, Blennow K, et al. Alzheimer's disease research progress in Australia: The Alzheimer's Association International Conference Satellite Symposium in Sydney. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2022;18(1):178-90.
Morovic S, Budincevic H, Govori V, Demarin V. Possibilities of Dementia Prevention - It is Never Too Early to Start. Journal of medicine and life. 2019;12(4):332-7.
Wahl D, Solon-Biet SM, Cogger VC, Fontana L, Simpson SJ, Le Couteur DG, et al. Aging, lifestyle and dementia. Neurobiology of disease. 2019;130:104481.
Bae CY, Kim IH, Kim BS, Kim JH, Kim JH. Predicting the incidence of age-related diseases based on biological age: The 11-year national health examination data follow-up. Archives of gerontology and geriatrics. 2022;103:104788.
Newman AB, Boudreau RM, Naydeck BL, Fried LF, Harris TB. A physiologic index of comorbidity: relationship to mortality and disability. The journals of gerontology Series A, Biological sciences and medical sciences. 2008;63(6):603-9.
Sanders JL, Boudreau RM, Penninx BW, Simonsick EM, Kritchevsky SB, Satterfield S, et al. Association of a Modified Physiologic Index with mortality and incident disability: the Health, Aging, and Body Composition study. The journals of gerontology Series A, Biological sciences and medical sciences. 2012;67(12):1439-46.
Wu C, Smit E, Sanders JL, Newman AB, Odden MC. A Modified Healthy Aging Index and Its Association with Mortality: The National Health and Nutrition Examination Survey, 1999-2002. The journals of gerontology Series A, Biological sciences and medical sciences. 2017;72(10):1437-44.
McCabe EL, Larson MG, Lunetta KL, Newman AB, Cheng S, Murabito JM. Association of an Index of Healthy Aging With Incident Cardiovascular Disease and Mortality in a Community-Based Sample of Older Adults. The journals of gerontology Series A, Biological sciences and medical sciences. 2016;71(12):1695-701.
O'Connell MDL, Marron MM, Boudreau RM, Canney M, Sanders JL, Kenny RA, et al. Mortality in Relation to Changes in a Healthy Aging Index: The Health, Aging, and Body Composition Study. The journals of gerontology Series A, Biological sciences and medical sciences. 2019;74(5):726-32.
Li K, Xu W, Hu B, Shan Q, Liu L, Cao J. The modified Healthy Aging Index is associated with mobility limitations and falls in a community-based sample of oldest old. Aging clinical and experimental research. 2021;33(3):555-62.
Tu R, He H, Wang S, Ding J, Zeng Q, Guo L, et al. The association between healthy aging index and trajectories of disability: a population-based cohort study. Aging clinical and experimental research. 2022;34(8):1893-900.
Yaffe K, Vittinghoff E, Hoang T, Matthews K, Golden SH, Zeki Al Hazzouri A. Cardiovascular Risk Factors Across the Life Course and Cognitive Decline: A Pooled Cohort Study. Neurology. 2021;96(17):e2212-e9.
Ma YH, Shen LX, Li YZ, Leng Y, Yang L, Chen SD, et al. Lung function and risk of incident dementia: A prospective cohort study of 431,834 individuals. Brain, behavior, and immunity. 2023;109:321-30.
Xu W, Tan L, Wang HF, Jiang T, Tan MS, Tan L, et al. Meta-analysis of modifiable risk factors for Alzheimer's disease. Journal of neurology, neurosurgery, and psychiatry. 2015;86(12):1299-306.
Dunn AR, O'Connell KMS, Kaczorowski CC. Gene-by-environment interactions in Alzheimer's disease and Parkinson's disease. Neuroscience and biobehavioral reviews. 2019;103:73-80.
Duan R, Fu Q, Sun Y, Li Q. Epigenetic clock: A promising biomarker and practical tool in aging. Ageing research reviews. 2022;81:101743.
Xu Y, Zheng X, Li Y, Ye X, Cheng H, Wang H, et al. Exploring patient medication adherence and data mining methods in clinical big data: A contemporary review. Journal of evidence-based medicine. 2023;16(3):342-75.
Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS medicine. 2015;12(3):e1001779.
Cox N. UK Biobank shares the promise of big data. Nature. 2018;562(7726):194-5.
Collins R. What makes UK Biobank special? Lancet (London, England). 2012;379(9822):1173-4.
Huang N, Zhuang Z, Song Z, Wang W, Li Y, Zhao Y, et al. Associations of Modified Healthy Aging Index With Major Adverse Cardiac Events, Major Coronary Events, and Ischemic Heart Disease. Journal of the American Heart Association. 2023;12(5):e026736.
Zhuang Z, Zhao Y, Huang N, Li Y, Wang W, Song Z, et al. Associations of healthy aging index and all-cause and cause-specific mortality: a prospective cohort study of UK Biobank participants. GeroScience. 2024;46(1):1241-57.
ElSayed NA, Aleppo G, Aroda VR, Bannuru RR, Brown FM, Bruemmer D, et al. Classification and Diagnosis of Diabetes: Standards of Care in Diabetes-2023. Diabetes care. 2023;46(Suppl 1):S19-s40.
Rosso AL, Sanders JL, Arnold AM, et al. Multisystem physiologic impairments and changes in gait speed of older adults. J Gerontol A Biol Sci Med Sci. 2015;70(3):319-324.
Huang SY, Li YZ, Zhang YR, Huang YY, Wu BS, Zhang W, et al. Sleep, physical activity, sedentary behavior, and risk of incident dementia: a prospective cohort study of 431,924 UK Biobank participants. Molecular psychiatry. 2022;27(10):4343-54.
Pazoki R, Dehghan A, Evangelou E, Warren H, Gao H, Caulfield M, et al. Genetic Predisposition to High Blood Pressure and Lifestyle Factors: Associations With Midlife Blood Pressure Levels and Cardiovascular Events. Circulation. 2018;137(7):653-61.
Wang M, Zhou T, Li X, Ma H, Liang Z, Fonseca VA, et al. Baseline Vitamin D Status, Sleep Patterns, and the Risk of Incident Type 2 Diabetes in Data From the UK Biobank Study. Diabetes care. 2020;43(11):2776-84.
Leng Y, Ackley SF, Glymour MM, Yaffe K, Brenowitz WD. Genetic Risk of Alzheimer's Disease and Sleep Duration in Non-Demented Elders. Annals of neurology. 2021;89(1):177-81.
Jansen IE, Savage JE, Watanabe K, Bryois J, Williams DM, Steinberg S, et al. Genome-wide meta-analysis identifies new loci and functional pathways influencing Alzheimer's disease risk. Nature genetics. 2019;51(3):404-13.
Marioni RE, Harris SE, Zhang Q, McRae AF, Hagenaars SP, Hill WD, et al. GWAS on family history of Alzheimer's disease. Translational psychiatry. 2018;8(1):99.
Fan M, Sun D, Zhou T, Heianza Y, Lv J, Li L, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. European heart journal. 2020;41(11):1182-9.
Che WQ, Li YJ, Tsang CK, Wang YJ, Chen Z, Wang XY, et al. How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology. Military Medical Research. 2023;10(1):50.
Li DJ, Tsai CS, Hsiao RC, Chen YL, Yen CF. Associations between Allergic and Autoimmune Diseases with Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder within Families: A Population-Based Cohort Study. International journal of environmental research and public health. 2022;19(8).
Xiao T, Wijnant SRA, Licher S, Terzikhan N, Lahousse L, Ikram MK, et al. Lung Function Impairment and the Risk of Incident Dementia: The Rotterdam Study. Journal of Alzheimer's disease : JAD. 2021;82(2):621-30.
Ghosh A, Comerota MM, Wan D, Chen F, Propson NE, Hwang SH, et al. An epoxide hydrolase inhibitor reduces neuroinflammation in a mouse model of Alzheimer's disease. Science translational medicine. 2020;12(573).
Zhang W, Xiao D, Mao Q, Xia H. Role of neuroinflammation in neurodegeneration development. Signal transduction and targeted therapy. 2023;8(1):267.
López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: An expanding universe. Cell. 2023;186(2):243-78.
Qu J, L X, Liu Y, Wang X. Evaluation of procalcitonin, C-reactive protein, interleukin-6 & serum amyloid A as diagnostic biomarkers of bacterial infection in febrile patients. The Indian journal of medical research. 2015;141(3):315-21.
Veldsman M, Tai XY, Nichols T, Smith S, Peixoto J, Manohar S, et al. Cerebrovascular risk factors impact frontoparietal network integrity and executive function in healthy ageing. Nature communications. 2020;11(1):4340.
Walker KA, Sharrett AR, Wu A, Schneider ALC, Albert M, Lutsey PL, et al. Association of Midlife to Late-Life Blood Pressure Patterns With Incident Dementia. Jama. 2019;322(6):535-45.
Burtscher J, Mallet RT, Burtscher M, Millet GP. Hypoxia and brain aging: Neurodegeneration or neuroprotection? Ageing research reviews. 2021;68:101343.
Lall R, Mohammed R, Ojha U. What are the links between hypoxia and Alzheimer's disease? Neuropsychiatric disease and treatment. 2019;15:1343-54.
Hassan H, Chen R. Hypoxia in Alzheimer's disease: effects of hypoxia inducible factors. Neural regeneration research. 2021;16(2):310-1.
Vu TT, Zhao L, Liu L, Schiman C, Lloyd-Jones DM, Daviglus ML, et al. Favorable Cardiovascular Health at Young and Middle Ages and Dementia in Older Age-The CHA Study. Journal of the American Heart Association. 2019;8(1):e009730.

No competing interests reported.

SUPPLEMENTALMATERIAL.docx

Download PDF

Editorial decision: Revision requested
02 Sep, 2024
Editor assigned by journal
02 Sep, 2024
Submission checks completed at journal
30 Aug, 2024
First submitted to journal
29 Aug, 2024

You are reading this latest preprint version

Associations of healthy aging index with dementia: a prospective cohort study of 305,566 individuals

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study population

Components of the Healthy Aging Index

Outcomes

Confounders

Assessment of genetic susceptibility to Alzheimer's disease

Statistical analyses

Results

Baseline characteristics

The hazard ratios for all-cause dementia, AD, and VD

Sensitivity and subgroup analysis

The Joint association between HAI and AD‑GRS for AD risk

Discussion

Strengths and limitations

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1