Abstract
Background
Cardiovascular-kidney-metabolic syndrome (CKM) is associated with an increased risk of mortality among older adults, but the relationship between cognitive function and mortality across different stages of CKM remains unclear. This study investigated the associations between cognitive performance and all-cause mortality among older adults with different stages of CKM syndrome.
Methods
The current cohort analyzed data from 1155 United States older adults aged ≥ 60 years. Cognitive function was measured using typical cognitive tests. Participants were classified into non-advanced CKM and advanced CKM groups. Weighted regression models were used to explore the association between cognitive function and mortality.
Results
Following full adjustment for potential confounders, higher cognitive function scores were associated with a decreased risk of all-cause mortality risk in both advanced and non-advanced CKM groups. The highest quartile group exhibited a decreased risk of mortality in cognitive tests ( hazard ratio < 1, P < 0.05). Significant interactions were observed between subgroups of cognitive scores, smoking status, and body mass index.
Conclusion
Improving cognitive function may lower the risk of mortality among older adults with non-advanced CKM and advanced CKM.
Graphical Abstract
Cognitive function and mortality among adults with CKM syndrome. CKM: Cardiovascular-kidney-metabolic syndrome; CERAD-WL: Consortium to Establish a Registry for Alzheimer's Disease - Word List; CERAD-DR: Consortium to Establish a Registry for Alzheimer's Disease - Delayed Recall; DSST: Digit Symbol Substitution Test
Supplementary Information
The online version contains supplementary material available at 10.1186/s12877-026-07055-z.
Keywords: Cardiovascular-Kidney-Metabolic syndrome (CKM), Cognitive function, All-cause mortality
Introduction
Cognitive frailty can lead to disability and impose a heavy burden on healthcare systems [1]. Cardiovascular-Kidney-Metabolic syndrome (CKM) develops in older adults and may be affected by metabolic conditions, social determinants of health (SDOH), and inflammation [2–4]. Studies have shown that the risk of cognitive impairment and mortality is higher among individuals with cardiovascular diseases, kidney diseases, metabolic disorders [5–8]. This finding suggests a link between cognitive dysfunction in older adults with CKM and mortality risk, although the relationship among different CKM stages remains unclear.
Cognitive frailty is a hallmark of aging and is positively correlated with the incidence and progression of frailty and the risk of all-cause mortality among older adults [9]. During CKM progression, exposure to numerous risk factors can damage the vascular endothelium of critical organs, potentially exacerbating cognitive frailty and leading to poor clinical outcomes [10]. Comprehensive assessment and risk stratification of CKM in older individuals play a crucial role in health management. More studies are urgently needed to focus on older adults with CKM to guide early interventions tailored to different CKM stages. Cognitive function assessment is the cornerstone of health promotion among older adults.
The Consortium to Establish a Registry for Alzheimer’s Disease - Word List (CERAD-WL) primarily assesses immediate memory functions. Animal fluency test assesses language fluency and executive function. The Digit Symbol Substitution Test (DSST) measures processing speed and executive function. Lastly, CERAD - Delayed Recall (CERAD-DR) evaluates delayed memory function, specifically after an interval of 8 to 10 min. The four tests provide a focused evaluation, which is quick, simple, and easy to implement, allowing for a multidimensional assessment of cognitive performance. These methods are practical and cost-effective, making them suitable for routine follow-ups without necessitating expensive equipment.
With the increasing burden of aging, it is particularly important to reveal the correlation between cognitive performance and mortality in different CKM stages. Several studies on the association between mortality and cognitive performance have primarily focused on community-dwelling older adults. These studies were not nationally representative or relied on a single cognitive assessment method, such as the mini-mental state examination [11, 12]. These limitations restrict the application of the four cognitive tests as potential indicators of cognitive impairment in clinical practice.
To address this knowledge gap, the current study sought to uncover the potential correlation between cognitive performance and all-cause mortality among US overall CKM population, individuals with advanced CKM and individuals with non-advanced CKM.
Methods
The design of the current study
The present study enrolled individuals aged ≥ 60 years old from the National Health and Nutrition Examination Survey (NHANES) in 2011–2014. The detailed introduction of this database can be accessed via the official website. all participants signed informed consent for participation. Of the 19,931 participants enrolled in 2011–2014, 1155 aged ≥ 60 years were included in this study. Individuals were classified into five stages based on the scientific advisory of the 2023 American Heart Association on the CKM framework [13]. Referring to the methodology and assessment of Aggarwal’s study on the CKM [14], exclusions were made for participants: [1] lack of cognitive tests final scores (n = 16997); [2] missing information of CKM components (n = 1576); [3] lacking valid fasting measurement weights (n = 164); [4] missing demographic data and other biochemical data (n = 38); [5] missing follow-up data (n = 1). After excluding participants who met these criteria, 1155 remained (Fig. 1).
Fig. 1.
Flowchart of inclusion and exclusion
Assessment of CKM components
Data on demographic factors, physiological, and biochemical indicators, and chronic disease status including age, sex, ethnicity, body mass index (BMI), fasting glucose, hemoglobin A1c (HbA1c), triglyceride, high-density lipoprotein, serum creatinine and self-reported disease questionnaires were collected. The Predicting Risk of Cardiovascular Disease Events (PREVENT) Equations is a powerful tool to assess cardiovascular risk. The 2021 CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) creatinine equation was utilized to calculate the estimated glomerular filtration rate (eGFR). Participants with the value of 10-year cardiovascular risk ≥ 20% or very high-risk CKD were regarded as having subclinical CVD.
According the definition, different process and severity of CKM was divided into 5 stages. The detailed stage criteria are provided in Table S1.
The concise staging framework was defined as follows:
Stage 0: No CKM syndrome risk factors
Stage 1: Prediabetes or BMI / waist circumference exceeding the threshold
Stage 2: Presence of hypertension, diabetes, elevated fasting serum triglycerides, CKD, or metabolic syndrome
Stage 3: High 10-year CVD risk predicted by the PREVENT or very high-risk CKD.
Stage 4: Presence of CVD, characterized by a documented history of angina, coronary heart disease, myocardial infarction, heart failure, or stroke.
This cohort was divided into two groups: the non-advanced CKM group and the advanced CKM group. The advanced CKM group included participants with CKM stages 3 and 4, whereas the non-advanced CKM group included those with CKM stages 0–2.
Assessment of cognitive function
During the 2011–2014, the NHANES survey focused on cognitive function tests to assess the cognitive health of older adults, using four cognitive function tests.
CERAD tests consist of three consecutive trials requiring immediate recall of as many words as possible, followed by a delayed recall trial conducted 8–10 min later to assess the number of words that can be remembered. The scoring range is 0–30 and 0–10 for the two tests, respectively. In both tests, higher scores indicate better memory performance. The animal fluency test assesses the ability to categorize animals and demonstrates language fluency, thereby differentiating individuals with varying levels of cognitive function. Participants are asked to name animals within one minute, with the scoring of each correct interaction receiving one point. Higher scores indicate better verbal fluency and semantic retrieval. The DSST presents nine specific digits paired with corresponding symbols on the paper. Participants are asked to match 133 digits to their corresponding symbols in 133 boxes within 2 min. Correct matches are scored, with the total score ranging from 0 to 133. Higher scores indicate better processing speed, sustained attention, and working memory. For the CERAD, animal fluency test and DSST, higher scores suggest better cognitive function [15, 16].
Definition of mortality outcomes
Mortality data were obtained from the National Death Index (NDI), a comprehensive resource that enables public health and medical researchers to link study data with US death records. The NDI was employed to ascertain participants’ mortality by linking baseline data to death certificate records up to December 31, 2019. Although the NDI provides data on various types of mortality, this study focused specifically on all-cause mortality.
Other study covariates
Household interviews were conducted to collect demographic information, including the level of education. Smoking status was categorized as never (individuals who had never smoked), former (≥ 100 cigarettes in the past but no longer smoked), and current (currently smoked and had a cumulative history of smoking more than 100 cigarettes). Laboratory data included the measurements of uric acid, total cholesterol (TC), low-density lipoprotein cholesterol (LDL), white blood cell count, lymphocyte count, monocyte count, neutrophil count, platelet count, and the percentages of lymphocytes, monocytes, and neutrophils. The following formula was used to calculate the Systemic Immune-Inflammation Index (SII) :
 |
Statistical analyses
R version 4.4.0 was used to conduct the overall analysis with appropriate weights. Continuous variables with a normal distribution are reported as weighted means with standard deviations. Four test scores were divided into quartiles, ranging from the lowest (Q1) to the highest (Q4). Categorical variables were compared using the Chi-square test and are presented as counts with weighted percentages. The student’s t-test was employed to compare continuous data with a normal distribution, whereas the Mann-Whitney U test was used to compare non-normally distributed continuous data.
Cox regression analysis was conducted to estimate the risk of mortality and 95% confidence intervals (CI). Model 1 was an unadjusted model. Model 2 was adjusted for the age, sex and ethnicity. Model 3 was adjusted for additional factors, including educational level, TC, uric acid, smoking status, and SII. Trend tests were conducted across quartiles of cognitive function scores. Multiplicative interaction terms between CKM group (non-advanced vs. advanced) and cognitive function quartiles were added to the fully adjusted Cox models. P values for interaction were obtained from likelihood ratio tests comparing models with and without the interaction terms. Kaplan-Meier curves showed cumulative mortality across quartiles of cognitive function test scores.
To explore the effect of different severities of CKM, participants were divided into two groups: non-advanced CKM and advanced CKM. Restricted cubic spline regression with full covariate adjustment was conducted to analyze the inverse relationship. Subgroup analyses were conducted based on sex (male and female), body mass index (BMI: <30 kg/m² and ≥ 30 kg/m²), smoking status, and the presence of common chronic diseases (diabetes mellitus and hypertension) and interaction tests were performed. The false discovery rate correction was employed to account for multiple tests in subgroup analyses. To examine the independent effect of CKM severity on mortality within comparable levels of cognitive function, we further stratified participants by tertiles of each cognitive test. Within each tertile, we fitted survey-weighted Cox proportional hazards models with CKM group (advanced vs. non-advanced) as the exposure and estimated HR for all-cause mortality, adjusting for the same covariates as in Model 3. To characterize the dose–response relationship between CKM severity and cognitive performance, survey-weighted linear regression models were fitted to examine the cross-sectional association between CKM stage and baseline cognitive test scores. To assess potential selection bias, we additionally compared baseline characteristics between age-eligible participants who were included in the analytic sample and those who were excluded due to missing cognitive tests or CKM parameters, using survey-weighted linear regression for continuous variables and survey-weighted chi-square tests for categorical variables.
Results
Analysis of baseline characteristic
The present cohort included 1155 individuals with CKM syndrome undergoing four cognitive tests. Baseline demographic data, components of CKM syndrome, and other potential risk factors are presented in Table 1. The participants had a mean age of 69.03 ± 6.55 years and 44.87% of older adults were male. The mean follow-up length was 79.45 ± 19.15 months. The participants were classified into two group (non-advanced CKM group and advanced CKM group for stage 0–2 and stage 3–4, respectively). Participants in the non-advanced CKM group had a higher level of education and lower levels of serum uric acid, serum creatinine, HbA1c, BMI, triglyceride, and SII. In addition, compared to individuals in advanced CKM group, participants in the non-advanced CKM group were less likely to be current smokers. The average scores of all tests were higher in the non-advanced CKM group than in the advanced CKM group (P < 0.001). Among all age-eligible participants with non-missing survey weights, those included in the analytic sample differed from those excluded because of missing cognitive tests or CKM parameters (Supplementary Table S5). Compared with excluded participants, included individuals were slightly younger (mean age 69.0 vs. 71.1 years, P < 0.001) but generally exhibited a higher cardiometabolic burden. In particular, included participants had higher levels of TG, TC and LDL, Scr, uric acid, and SII, and were more likely to have diagnosed hypertension, ASCVD, and congestive heart failure.
Table 1.
Weighted baseline characteristics of CKM syndrome of adult Americans from NHANES 2011–2014
| Variables | Total cohort | Non-advanced CKM | Advanced CKM | P value |
|---|---|---|---|---|
| Age, years | 69.03(0.28) | 66.90(0.26) | 72.74(0.48) | < 0.0001 |
| BMI, kg/m2 | 29.17(0.34) | 28.72(0.34) | 29.94(0.52) | 0.03 |
| Waist circumference, cm | 102.81(0.80) | 100.67(0.74) | 106.53(1.28) | < 0.0001 |
| Glucose, mg/dl | 112.15(1.61) | 107.73(1.65) | 119.82(2.25) | < 0.0001 |
| HbA1C | 5.95(0.05) | 5.80(0.05) | 6.20(0.08) | < 0.0001 |
| Triglyceride, mg/dl | 121.65(3.30) | 117.76(3.41) | 128.40(4.75) | 0.03 |
| HDL, mg/dl | 57.00(1.00) | 59.98(1.09) | 51.84(1.24) | < 0.0001 |
| TC, mmol/L | 4.96(0.03) | 5.19(0.04) | 4.56(0.06) | < 0.0001 |
| LDL, mmol/L | 2.86(0.03) | 3.03(0.04) | 2.56(0.05) | < 0.0001 |
| Serum creatinine, umol/L | 85.94(1.69) | 77.16(0.90) | 101.16(4.81) | < 0.0001 |
| Uric acid, umol/L | 335.71(3.52) | 324.25(4.16) | 355.58(5.69) | < 0.0001 |
| WBC number, 1000 cells/uL | 6.57(0.11) | 6.25(0.11) | 7.12(0.12) | < 0.0001 |
| Lymphocyte number, 1000 cells/uL | 1.82(0.04) | 1.81(0.03) | 1.83(0.06) | 0.73 |
| Monocyte number, 1000 cells/uL | 0.56(0.02) | 0.53(0.02) | 0.60(0.01) | 0.003 |
| Neutrophil number, 1000 cells/uL | 3.93(0.08) | 3.66(0.08) | 4.40(0.10) | < 0.0001 |
| Platelet number, 1000 cells/uL | 223.24(1.75) | 228.33(2.24) | 214.42(3.98) | 0.01 |
| SII | 546.54(17.29) | 505.44(12.07) | 617.80(32.19) | 0.001 |
| CERAD-WL | 20.06(0.28) | 20.87(0.28) | 18.64(0.28) | < 0.0001 |
| CERAD-DR | 6.42(0.13) | 6.80(0.15) | 5.77(0.12) | < 0.0001 |
| Animal Fluency | 18.25(0.22) | 18.95(0.32) | 17.04(0.33) | < 0.001 |
| DSST | 52.58(0.81) | 57.15(0.95) | 44.66(0.98) | < 0.0001 |
| Sex | < 0.0001 | |||
| Female | 587(55.13) | 411(63.51) | 176(40.61) | |
| Male | 568(44.87) | 282(36.49) | 286(59.39) | |
| Ethnicity | 0.002 | |||
| Mexican American | 108( 3.52) | 75(3.81) | 33(3.00) | |
| Non-hispanic White | 587(79.81) | 316(79.72) | 271(79.99) | |
| Non-hispanic Black | 231( 8.14) | 140(7.71) | 91(8.88) | |
| Non-hispanic Asian | 94( 3.01) | 73(3.81) | 21(1.62) | |
| Other Hispanic | 118( 3.59) | 81(3.76) | 37(3.30) | |
| Other Race | 17( 1.93) | 8(1.19) | 9(3.21) | |
| Education | < 0.0001 | |||
| Less than 9th grade | 124( 5.57) | 66(4.49) | 58(7.44) | |
| 9-11th grade | 165(10.62) | 95( 9.64) | 70(12.31) | |
| High school graduate | 275(22.38) | 145(17.45) | 130(30.92) | |
| Some college or AA degree | 330(32.18) | 210(35.03) | 120(27.25) | |
| College graduate or above | 261(29.25) | 177(33.38) | 84(22.08) | |
| Smoke | < 0.0001 | |||
| Never | 572(49.25) | 388(55.70) | 184(38.06) | |
| Former | 447(40.64) | 242(36.88) | 205(47.18) | |
| Now | 136(10.11) | 63( 7.43) | 73(14.76) | |
| Hypertension | < 0.0001 | |||
| No | 260(25.44) | 198(30.95) | 62(15.87) | |
| Yes | 895(74.56) | 495(69.05) | 400(84.13) | |
| Diabetes mellitus | < 0.0001 | |||
| No | 741(69.61) | 520(79.91) | 221(51.76) | |
| Yes | 414(30.39) | 173(20.09) | 241(48.24) | |
| ASCVD | < 0.0001 | |||
| No | 920(78.25) | 693(100.00) | 227( 40.54) | |
| Yes | 235(21.75) | 0( 0.00) | 235(59.46) | |
| Congestive heart failure | < 0.0001 | |||
| No | 1076(93.01) | 693(100.00) | 383( 80.89) | |
| Yes | 79( 6.99) | 0( 0.00) | 79(19.11) | |
| Mets | 0.01 | |||
| No | 543(47.98) | 346(51.67) | 197(41.57) | |
| Yes | 612(52.02) | 347(48.33) | 265(58.43) | |
| Prediabetes | 0.58 | |||
| No | 418(38.97) | 246(39.80) | 172(37.52) | |
| Yes | 737(61.03) | 447(60.20) | 290(62.48) |
Data are expressed as mean (Standard error) and numbers (weighted percentage) as appropriate. All estimates were weighted to be nationally representative
Abbreviations: HbA1C Glycosylated Hemoglobin, Type A1C,HDL High-density lipoprotein cholesterol, TC Total cholesterol, LDL Low-density lipoprotein cholesterol, SII Systemic immune-inflammation index, CERAD-WL Consortium to Establish a Registry for Alzheimer’s Disease - Word List, CERAD-DR Consortium to Establish a Registry for Alzheimer’s Disease - Delayed Recall, DSST Digit Symbol Substitution Test
Correlation between cognitive function and all-cause mortality
The four test scores all showed an inverse correlation with the risk of mortality in the unadjusted model (Table 2). After adjusting for demographic factors in Model 2, this association remained significant. Model 3 was adjusted for demographic factors, relevant biochemical relevant indicators, and smoking status, in which cognitive function was still associated with all-cause mortality [with quartile 1 as reference, the hazard ratio (HR) with 95% CI in quartile 4 were 0.28 (0.13,0.59), 0.30 (0.13,0.70), 0.36 (0.18,0.70), and 0.38 (0.17,0.83) for the CERAD-WL and CERAD-DR, Animal fluency test, and DSST scores, respectively; all P < 0.05]. Multiplicative interaction between CKM group and cognitive function quartiles were statistically significant for all four cognitive tests (all P for interaction < 0.0001; Table 3), indicating that CKM severity modifies the strength of the relationship between cognitive function and mortality. Restricted cubic spline (RCS) confirmed the negative linear correlation (Fig. 2). Kaplan-Meier curves showed that individuals in the Q4 of the four cognitive tests had a lower risk of all-cause mortality (Fig. 3). In non-advanced CKM group, good cognitive function indicated a longer lifespan among older adults. Specifically, the animal fluency test scores interacted with different groups and significantly affected all-cause mortality (Pinteraction = 0.02). Except for DSST, higher test scores were associated with better clinical outcomes in the advanced CKM group (P < 0.05, Table 3). The RCS plot showed the correlation between the cognitive scores of US older adults and their all-cause mortality in both groups (Figure S1).
Table 2.
Association between cognitive function and all-cause mortality in older adults
| Variables | Unadjusted | Model 1 | Model 2 | |||
|---|---|---|---|---|---|---|
| CERAD-WL | HR (95%CI) | P value | HR (95%CI) | P value | HR (95%CI) | P value |
| Q1 (0–15) | (Reference) | (Reference) | (Reference) | |||
| Q2 (16–19) | 0.47(0.28,0.78) | 0.003 | 0.71(0.41,1.22) | 0.21 | 0.71(0.43,1.19) | 0.19 |
| Q3 (20–22) | 0.21(0.13,0.36) | < 0.001 | 0.40(0.22,0.73) | 0.003 | 0.35(0.17,0.73) | 0.01 |
| Q4 (23–30) | 0.11(0.06,0.22) | < 0.001 | 0.26(0.13,0.54) | < 0.001 | 0.28(0.13,0.59) | < 0.001 |
| Continuous | 0.85(0.83,0.88) | < 0.001 | 0.91(0.87,0.95) | < 0.001 | 0.91(0.87,0.96) | < 0.001 |
| CERAD-DR | HR (95%CI) | P value | HR (95%CI) | P value | HR (95%CI) | P value |
| Q1 (0–4) | (Reference) | (Reference) | (Reference) | |||
| Q2 (5–5) | 0.49(0.31,0.78) | 0.003 | 0.64(0.38,1.10) | 0.1 | 0.68(0.38,1.21) | 0.19 |
| Q3 (6–7) | 0.24(0.15,0.41) | < 0.001 | 0.41(0.23,0.73) | 0.002 | 0.44(0.24,0.82) | 0.01 |
| Q4 (8–10) | 0.15(0.07,0.31) | < 0.001 | 0.32(0.14,0.73) | 0.01 | 0.30(0.13,0.70) | 0.01 |
| Continuous | 0.71(0.67,0.75) | < 0.001 | 0.81(0.73,0.89) | < 0.001 | 0.82(0.73,0.92) | < 0.001 |
| AFT | HR (95%CI) | P value | HR (95%CI) | P value | HR (95%CI) | P value |
| Q1 (3–12) | (Reference) | (Reference) | (Reference) | |||
| Q2 (13–15) | 0.79(0.56,1.13) | 0.2 | 0.76(0.51,1.12) | 0.16 | 0.79(0.56,1.11) | 0.18 |
| Q3 (16–19) | 0.62(0.40,0.96) | 0.03 | 0.77(0.51,1.18) | 0.24 | 0.87(0.55,1.38) | 0.55 |
| Q4 (20–36) | 0.23(0.12,0.44) | < 0.001 | 0.41(0.21,0.83) | 0.01 | 0.36(0.18,0.70) | 0.002 |
| Continuous | 0.90(0.88,0.93) | < 0.001 | 0.94(0.90,0.97) | < 0.001 | 0.94(0.91,0.97) | < 0.001 |
| DSST | HR (95%CI) | P value | HR (95%CI) | P value | HR (95%CI) | P value |
| Q1 (4–33) | (Reference) | (Reference) | (Reference) | |||
| Q2 (34–46) | 0.74(0.51,1.07) | 0.11 | 0.79(0.52,1.20) | 0.26 | 0.79(0.54,1.17) | 0.24 |
| Q3 (47–57) | 0.48(0.29,0.80) | 0.005 | 0.67(0.40,1.11) | 0.12 | 0.61(0.35,1.07) | 0.08 |
| Q4 (58–105) | 0.16(0.09,0.26) | < 0.001 | 0.38(0.19,0.76) | 0.01 | 0.38(0.17,0.83) | 0.01 |
| Continuous | 0.96(0.95,0.97) | < 0.001 | 0.98(0.96,0.99) | < 0.001 | 0.97(0.96,0.99) | < 0.001 |
CERAD-WL Consortium to Establish a Registry for Alzheimer’s Disease-Word Learning, CERAD-DR Consortium to Establish a Registry for Alzheimer’s Disease-Delayed Recall, AF Animal Fluency, DSST Digit Symbol Substitution Test
Model 1: unadjusted model
Model 2: Adjusted for age, sex, and ethnicity
Model 3: Model 2 + Educational level + Total cholesterol + Uric acid+ Smoking status + SII
Table 3.
Association between cognitive function and all-cause mortality in older adults with non-advanced or advanced CKM
| Variables | Non-advanced CKM | Advanced CKM | ||
|---|---|---|---|---|
| CERAD-WL | Number of deaths | Hazard ratio (95% CI) | Number of deaths | Hazard ratio (95% CI) |
| Q1 (0–15) | 16 (15.4%) | (Reference) | 57 (41.3%) | (Reference) |
| Q2 (16–19) | 15 (8.2%) | 0.506(0.185, 1.386) | 42 (30.4%) | 0.735(0.455,1.185) |
| Q3 (20–22) | 14 (7.2%) | 0.184(0.045, 0.759) | 27 (25.5%) | 0.413(0.174,0.980) |
| Q4 (23–30) | 8 (3.8%) | 0.132(0.042, 0.416) | 18 (22.5%) | 0.423(0.140,1.276) |
| P for trend | 0.001 | 0.027 | ||
| Continuous | 0.872(0.808, 0.941) | 0.926(0.876,0.980) | ||
| P for interaction CKM group | < 0.001 | |||
| CERAD-DR | Number of deaths | Hazard ratio (95% CI) | Number of deaths | Hazard ratio (95% CI) |
| Q1 (0–4) | 18 (15.3%) | (Reference) | 66 (45.5%) | (Reference) |
| Q2 (5–5) | 7 (6.9%) | 0.321(0.086, 1.209) | 22 (31.9%) | 0.937(0.510,1.722) |
| Q3 (6–7) | 15 (6.5%) | 0.400(0.169, 0.948) | 43 (28.7%) | 0.416(0.188,0.920) |
| Q4 (8–10) | 13 (5.4%) | 0.247(0.048, 1.262) | 13 (13.3%) | 0.331(0.136,0.803) |
| P for trend | 0.042 | 0.004 | ||
| Continuous | 0.728(0.565, 0.937) | 0.848(0.764,0.942) | ||
| P for interaction CKM group | < 0.001 | |||
| AFT | Number of deaths | Hazard ratio (95% CI) | Number of deaths | Hazard ratio (95% CI) |
| Q1 (3–12) | 13 (9.9%) | (Reference) | 46 (38.0%) | (Reference) |
| Q2 (13–15) | 12 (8.6%) | 1.549(0.722, 3.324) | 31 (30.4%) | 0.588(0.371,0.933) |
| Q3 (16–19) | 20 (10.7%) | 1.187(0.528, 2.669) | 46 (34.1%) | 0.744(0.451,1.228) |
| Q4 (20–36) | 8 (3.4%) | 0.172(0.053, 0.555) | 21 (20.2%) | 0.485(0.246,0.958) |
| P for trend | 0.012 | 0.056 | ||
| Continuous | 0.911(0.862, 0.963) | 0.947(0.912,0.984) | ||
| P for interaction CKM group | < 0.001 | |||
| DSST | Number of deaths | Hazard ratio (95% CI) | Number of deaths | Hazard ratio (95% CI) |
| Q1 (4–33) | 16 (13.7%) | (Reference) | 58 (37.2%) | (Reference) |
| Q2 (34–46) | 12 (7.9%) | 0.354(0.175, 0.717) | 50 (33.6%) | 1.032(0.634,1.679) |
| Q3 (47–57) | 15 (8.2%) | 0.280(0.115, 0.682) | 24 (24.2%) | 0.792(0.355,1.771) |
| Q4 (58–105) | 10 (4.1%) | 0.115(0.039, 0.334) | 12 (20.7%) | 0.820(0.299,2.252) |
| P for trend | < 0.001 | 0.567 | ||
| Continuous | 0.959(0.937, 0.982) | 0.984(0.962,1.007) | ||
| P for interaction CKM group | < 0.001 | |||
P for interaction with CKM group was obtained from survey-weighted Cox models including multiplicative interaction terms between CKM group (non-advanced vs. advanced) and cognitive function quartiles. Abbreviations see Table 2
Fig. 2.
RCS of the correlation between cognitive function and the risk of all-cause mortality
Fig. 3.
Kaplan-Meier curves of the relationship between cumulative mortality and quartiles of score
Subgroup analysis
We also investigated the association between the four cognitive tests and all-cause mortality across different subgroups. Subgroup analysis showed significant interactions between the CERAD-WL score, DSST score and smoking status subgroups and the risk of mortality (Fig. 4). CERAD-DR also interacted with BMI subgroups (all Pinteraction < 0.05). After adjusting for confounding factors, the inverse correlation remained significant among individuals with either non-advanced CKM syndrome or advanced CKM syndrome (all HR < 1). The subgroup analyses, except for DSST, showed significant trends in both CKM groups even after FDR correction (Table S2). DSST may be associated with a lower risk of all-cause mortality, but this result has not yet reached statistical significance.
Fig. 4.
Subgroup analysis and forest plot of cognitive function scores and mortality risk in older adults with CKM Syndrome
Correlation between CKM severity and all-cause mortality
When we compared mortality risk between CKM groups within the same levels of cognitive function, advanced CKM remained associated with higher all-cause mortality across most strata (Table S3). Across tertiles of CERAD-WL, the multivariable-adjusted HRs for advanced versus non-advanced CKM indicated a consistently elevated risk of death for advanced CKM irrespective of word-learning performance. Similar patterns were observed for AFT and DSST, although the magnitude of association varied by cognitive test. For CERAD-DR, advanced CKM was associated with higher mortality in the low and middle tertiles, whereas the association was attenuated in the highest tertile. Overall, these stratified analyses support an independent contribution of CKM severity to mortality beyond differences in cognitive function.
Correlation between CKM severity and cognitive function
At baseline, cognitive performance was worse among participants with advanced CKM than among those with non-advanced CKM (Table 1). Compared with non-advanced CKM (Table S4), participants with advanced CKM had a 2.24-point lower mean CERAD-WL score (β = −2.24; 95% CI: −2.74, − 1.74), a 1.03-point lower CERAD-DR score (β = −1.03; 95% CI: −1.28, − 0.77), a 1.92-point lower animal fluency score (β = −1.92; 95% CI: −2.56, − 1.28), and a 12.49-point lower DSST score (β = −12.49; 95% CI: −14.35, − 10.62; all P < 0.0001). These findings indicate that greater CKM severity is clearly associated with poorer cognitive function.
Discussion
This nationally representative population cohort study included a total of 1155 older adults, all of whom had complete cognitive test records and data on CKM-associated components. We found that cognitive function scores were negatively associated with the risks of all-cause mortality among older individuals with CKM syndrome. The non-advanced CKM group exhibited better cognitive function than the advanced CKM group. These findings indicate the potential application of the four test scores in determining the risk of all-cause mortality among older individuals with CKM syndrome. Additionally, we identified an intriguing phenomenon in which non-obese and non-smoking statuses interacted with cognitive scores. These traits were associated with protection against all-cause mortality. Our study could contribute to extending the life expectancy of older adults, guiding health education among older individuals, and raising awareness of cognitive function.
Cognitive decline gradually emerges during the progression of CKM syndrome. The heart and the brain are both affected by aging, smoking, BMI, metabolic status, and low-grade chronic inflammation [17–19]. Therefore, for older individuals with CKM syndrome, brain health should also be paid sufficient attention. In this study, compared to participants with non-advanced CKM, individuals with advanced CKM exhibited higher average age, BMI, waist circumference, fasting blood glucose levels, triglyceride levels, and SII levels. They also were more likely to be smokers and less likely to have higher education. The quartiles of CERAD-WL, CERAD-DR, animal fluency test, and DSST scores were calculated to measure the distribution characteristics of cognitive scores and their association with all-cause mortality in older individuals with CKM syndrome. Consistent with previous studies [20, 21], we observed that after adjusting for confounding factors, the group with the highest cognitive scores (Q4) had a significantly lower risk of mortality compared to Q1. Moreover, cognitive scores were significantly and negatively correlated with all-cause mortality in both advanced and non-advanced CKM. These findings suggest that cumulative exposure to the risk factors of the CKM syndrome can lead to cognitive decline in brain health. By testing interaction terms between CKM group and cognitive function quartiles, we found that CKM severity significantly altered the strength of the cognitive–mortality association. In general, lower cognitive performance was linked to higher mortality in both non-advanced and advanced CKM, but the relative risk gradients across cognitive categories differed by CKM stage and cognitive test. One possible explanation is that individuals with advanced CKM already have a very high baseline risk due to extensive cardiometabolic and vascular damage, so the incremental effect of cognitive deficits on mortality may be smaller or more variable than in those with non-advanced CKM. Alternatively, differences in underlying pathophysiology captured by specific cognitive domains may interact differently with CKM-related vascular and metabolic insults. Future studies are needed to replicate these patterns and to elucidate the mechanisms through which CKM progression shapes the prognostic significance of cognitive impairment.
CERAD-WL, CERAD-DR, animal fluency test, and DSST are commonly used to assess cognitive function, particularly among older adults. A cross-sectional study conducted by Casagrande et al. investigated the association between cognitive function and diabetes or prediabetes. The study found that compared to individuals with normal blood glucose levels, those with hyperglycemia had poorer executive function and processing speed, indicating that CERAD-WL, CERAD-DR, and DSST scores were all significantly lower [22]. Additionally, Song et al. further explored the relationship between cognitive function and inflammation. They found that a higher dietary inflammatory index among older individuals was associated with lower scores on the animal fluency test and DSST [23]. Cognitive tests play a crucial role in assessing cognitive function among older adults.
Previous studies have found that cognitive impairment may be associated with a higher risk of mortality [24–26]. Cognitive decline in older adults may also be associated with neuroinflammation. Activation of the NLRP3 inflammasome triggers the release of pro-inflammatory cytokines, leading to neuronal damage, induction of apoptosis and pyroptosis, increased oxidative stress, and impaired mitophagy [27]. NLRP3 inflammasome finally activates neuroinflammation and results in cognitive decline. Elevated levels of inflammatory factors can contribute to sarcopenia, which is associated with poor cardiovascular health and all-cause mortality [28–30]. Chronic low-grade inflammation state (CLIS) plays a pivotal role in mediating cardiovascular and cerebrovascular diseases in older populations, affecting not only vascular health but also contributing to cognitive decline [31, 32]. Kipinoinen et al. revealed that CLIS in midlife is linked to cognitive decline a decade later, particularly impaired verbal fluency and word learning [33]. Naomi et al. reported that obesity is a CLIS that may contribute to brain atrophy and cognitive impairment through oxidative stress and inflammation [34]. Jin et al. found that early intervention for improving lifestyle and cardiometabolic health can suppress inflammation and delay cognitive decline in older adults [35]. Therefore, early identification and reversal of CLIS are critical for preventing and managing CKM syndrome. In Table 3, the associations between cognitive function quartiles and mortality did not always follow a perfectly monotonic dose–response pattern within each CKM group. Several factors may account for these apparent inconsistencies. First, categorizing continuous cognitive scores into quartiles inevitably introduces misclassification and loss of information. Individuals near the cut points may be assigned to different quartiles despite having similar levels of cognition, which can blur the underlying gradient and produce non-monotonic patterns. This is supported by our continuous analyses, in which lower cognitive scores were consistently associated with higher mortality risk across all tests and CKM groups. Second, different cognitive tools capture distinct domains, such as memory, language fluency, processing speed and executive function, each of which may relate to CKM-related vascular and metabolic injury in non-identical ways. It is therefore plausible that some domains exhibit plateau effects, whereas others show a more linear trend. Finally, among participants with advanced CKM, the very high baseline risk conferred by severe cardiometabolic and vascular disease may attenuate the relative impact of intermediate levels of cognitive impairment, resulting in weaker gradients across quartiles. Taken together, we interpret these irregularities as reflecting categorization of continuous measures, and genuine domain-specific heterogeneity, rather than as evidence against an overall inverse relationship between cognition and mortality.
Strengths and limitations
The strengths of this study include: [1] This study was conducted based on the nationally representative NHANES database and, for the first time, explores the potential link between cognitive function and all-cause mortality in the CKM population [2]. A comprehensive cognitive function assessment was conducted using four tests to minimize the bias associated with reliance on a single test.
This study has some limitations [1]. Since NHANES cognitive tests are only conducted on individuals ≥ 60 years, other age groups were not included in this study, limiting the generalizability of our conclusions to these populations [2]. Because cognitive tests and CKM parameters were only available for a subset of age-eligible participants, our analytic sample differed from those who were excluded due to missing data. Compared with excluded individuals, included participants were slightly younger but had more adverse cardiometabolic profiles and a higher prevalence of hypertension, ASCVD, and heart failure. These differences suggest that our findings are most generalizable to older adults with relatively high CKM burden and more complete clinical assessments, and the absolute risk of mortality may not fully reflect that of all older adults with CKM-related conditions. Nevertheless, the biological mechanisms linking impaired cognitive function with mortality are unlikely to be fundamentally different, and our results should be interpreted in this context of potential selection bias. The NHANES database lacks data on atrial fibrillation and peripheral vascular disease, which may have falsely decreased the number of advanced CKM in this study [3]. The non-significant findings of this study regarding the association between DSST and all-cause mortality may be attributed to the small sample size, potential residual confounding [4]. The study did not measure or analyze frailty, which may have limited the comprehensiveness of the discussion on factors affecting cognitive function and mortality in CKM syndrome.
Conclusions
Higher CERAD-WL, CERAD-DR, and animal fluency test scores are associated with lower risks of all-cause mortality risk in older adults with advanced or non-advanced CKM syndrome. These findings suggest the potential use of comprehensive cognitive assessment in identifying mortality risk among older adults with CKM syndrome and highlight the importance of early cognitive screening and intervention in this population.
Supplementary Information
Acknowledgements
The authors express their gratitude to the participants and staff of NHANES for their invaluable contributions to this study. The authors thank SciDraw (https://scidraw.io/) for providing the images of the article. Additionally, the authors extend their sincere thanks to the editors and reviewers for their thoughtful and constructive feedback, which has significantly enhanced the quality of this research.
Abbreviations
- BMI
Body Mass Index
- CERAD-DR
Consortium to Establish a Registry for Alzheimer’s Disease - Delayed Recall
- CERAD-WL
Consortium to Establish a Registry for Alzheimer’s Disease - Word List
- CKD
Chronic Kidney Disease
- CKD-EPI
Chronic Kidney Disease Epidemiology Collaboration
- CKM
Cardiovascular-Kidney-Metabolic
- CVD
Cardiovascular Disease
- DSST
Digit Symbol Substitution Test
- eGFR
Estimated Glomerular Filtration Rate
- HbA1c
Hemoglobin A1c
- HR
Hazard Ratio
- HDL-C
High-Density Lipoprotein Cholesterol
- LDL-C
Low-Density Lipoprotein Cholesterol
- NDI
National Death Index
- NHANES
National Health and Nutrition Examination Survey
- PREVENT
Predicting Risk of Cardiovascular Disease Events
- Q1-Q4
Quartiles 1 to 4
- SII
Systemic Immune-Inflammation Index
- SDOH
Social Determinants of Health
- TC
Total Cholesterol
- US
United States
- WBC
White Blood Cell count
Author contributions
Yan Yan conceptualized the study and served as the corresponding author. Zhaorong Lin was responsible for drafting the manuscript and conducting the data analysis. Ai Wang and Maosen Lin collected data and provided a comprehensive revision of key intellectual content and assisted in data interpretation. Weifeng Guo and Jiasheng Yin reviewed and validated the data.
Funding
This research was funded by the National Natural Science Foundation of China, grant number (Grant number: 82070463) and Grant H2025-049 from Zhongshan Hospital, Fudan University.
Data availability
The database used and/or analyzed for this study available from the corresponding author on reasonable request (Yan Yan; Email: [yanyan_zhongshan@sina.com](mailto: yanyan_zhongshan@sina.com) ; yan.yan@zs-hospital.sh.cn).
Declarations
Ethics approval and consent to participate
Our study was conducted in strict accordance with the principles of the Declaration of Helsinki. NHANES received approval from the National Center for Health Statistics Research Ethics Review Board, and all participants provided written informed consent.
Consent for publication
Not Applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Qiu Y, Li G, Wang X, Zheng L, Wang C, Wang C, et al. Prevalence of cognitive frailty among community-dwelling older adults: A systematic review and meta-analysis. Int J Nurs Stud. 2022;125:104112. [DOI] [PubMed] [Google Scholar]
- 2.Li W, Shen C, Kong W, Zhou X, Fan H, Zhang Y, et al. Association between the triglyceride glucose-body mass index and future cardiovascular disease risk in a population with Cardiovascular-Kidney-Metabolic syndrome stage 0–3: a nationwide prospective cohort study. Cardiovasc Diabetol. 2024;23(1):292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhu R, Wang R, He J, Wang L, Chen H, Niu X, et al. Prevalence of Cardiovascular-Kidney-Metabolic syndrome stages by social determinants of health. JAMA Netw Open. 2024;7(11):e2445309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gao C, Gao S, Zhao R, Shen P, Zhu X, Yang Y, et al. Association between systemic immune-inflammation index and cardiovascular-kidney-metabolic syndrome. Sci Rep. 2024;14(1):19151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Nowak N, De Looze C, O’Halloran A, Kenny RA, Sexton DJ. The association between kidney function, cognitive function, and structural brain abnormalities in community-dwelling individuals aged 50 + is mediated by age and biomarkers of cardiovascular disease. Cardiovascular Res. 2023;119(11):2106–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.van Nieuwkerk AC, Delewi R, Wolters FJ, Muller M, Daemen M, Biessels GJ. Cognitive impairment in patients with cardiac disease: implications for clinical practice. Stroke. 2023;54(8):2181–91. [DOI] [PubMed] [Google Scholar]
- 7.Pignalosa FC, Desiderio A, Mirra P, Nigro C, Perruolo G, Ulianich L, et al. Diabetes and cognitive impairment: A role for glucotoxicity and dopaminergic dysfunction. Int J Mol Sci. 2021;22(22):12366. 10.3390/ijms222212366. [DOI] [PMC free article] [PubMed]
- 8.Kim JH, Chon D. Association between cognitive Impairment, vascular disease and All-Cause mortality. J Nutr Health Aging. 2018;22(7):790–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Nader MM, Cosarderelioglu C, Miao E, Whitson H, Xue QL, Grodstein F, et al. Navigating and diagnosing cognitive frailty in research and clinical domains. Nat Aging. 2023;3(11):1325–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Li N, Li Y, Cui L, Shu R, Song H, Wang J, et al. Association between different stages of cardiovascular-kidney-metabolic syndrome and the risk of all-cause mortality. Atherosclerosis. 2024;397:118585. [DOI] [PubMed] [Google Scholar]
- 11.Lv X, Li W, Ma Y, Chen H, Zeng Y, Yu X, et al. Cognitive decline and mortality among community-dwelling Chinese older people. BMC Med. 2019;17(1):63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Li H, Zheng Y, Li Q, Wang M. Cognitive Function, Healthy Lifestyle, and All-Cause mortality among Chinese older adults: A longitudinal prospective study. Nutrients. 2024;16(9):1297. 10.3390/nu16091297. [DOI] [PMC free article] [PubMed]
- 13.Ndumele CE, Neeland IJ, Tuttle KR, Chow SL, Mathew RO, Khan SS, et al. A synopsis of the evidence for the science and clinical management of Cardiovascular-Kidney-Metabolic (CKM) syndrome: A scientific statement from the American heart association. Circulation. 2023;148(20):1636–64. [DOI] [PubMed] [Google Scholar]
- 14.Aggarwal R, Ostrominski JW, Vaduganathan M. Prevalence of Cardiovascular-Kidney-Metabolic syndrome stages in US Adults, 2011–2020. JAMA. 2024;331(21):1858–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sakal C, Li T, Li J, Yang C, Li X. Association between sleep efficiency variability and cognition among older adults: Cross-Sectional accelerometer study. JMIR Aging. 2024;7:e54353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Li Y, Qiu L, Zhang C. Healthy dietary pattern improves cognitive function in elderly persons with periodontitis: A Cross-Sectional study of NHANES. Int Dent J. 2025;75(2):545–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Legdeur N, Heymans MW, Comijs HC, Huisman M, Maier AB, Visser PJ. Age dependency of risk factors for cognitive decline. BMC Geriatr. 2018;18(1):187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Xu X, Xu Y, Shi R. Association between obesity, physical activity, and cognitive decline in Chinese middle and old-aged adults: a mediation analysis. BMC Geriatr. 2024;24(1):54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jia L, Du Y, Chu L, Zhang Z, Li F, Lyu D, et al. Prevalence, risk factors, and management of dementia and mild cognitive impairment in adults aged 60 years or older in china: a cross-sectional study. Lancet Public Health. 2020;5(12):e661–71. [DOI] [PubMed] [Google Scholar]
- 20.Zipperer MB, Churilla JR, Stapleton JN, Richardson MR. Joint effect of cognitive function and C-reactive protein on all-cause mortality risk: 1999–2002 NHANES. Ann Epidemiol. 2022;74:111–7. [DOI] [PubMed] [Google Scholar]
- 21.Zhang C, Liu Y, Zeng L, Luo X, Fan G, Shi H, et al. Combined associations of cognitive impairment and psychological resilience with all-cause mortality in community-dwelling older adults. J Affect Disord. 2024;351:962–70. [DOI] [PubMed] [Google Scholar]
- 22.Casagrande SS, Lee C, Stoeckel LE, Menke A, Cowie CC. Cognitive function among older adults with diabetes and prediabetes, NHANES 2011–2014. Diabetes Res Clin Pract. 2021;178:108939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Song W, Feng Y, Gong Z, Tian C. The association between dietary inflammatory index and cognitive performance in older adults aged 60 years and older. Front Nutr. 2022;9:748000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Tamosiunas A, Sapranaviciute-Zabazlajeva L, Luksiene D, Virviciute D, Bobak M. Cognitive function and mortality: results from Kaunas HAPIEE study 2006–2017. Int J Environ Res Public Health. 2020;17(7):2397. 10.3390/ijerph17072397. [DOI] [PMC free article] [PubMed]
- 25.An J, Li H, Tang Z, Zheng D, Guo J, Liu Y, et al. Cognitive impairment and risk of All-Cause and cardiovascular disease mortality over 20-Year Follow-up: results from the BLSA. J Am Heart Association. 2018;7(15):e008252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Li S, He X, Wu L, Tang X, Ouyang Y, Jing W, et al. The association of cognitive function and its changes with all-cause mortality among community-dwelling older adults. Front Aging Neurosci. 2024;16:1419235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Xu Y, Yang Y, Chen X, Jiang D, Zhang F, Guo Y, et al. NLRP3 inflammasome in cognitive impairment and Pharmacological properties of its inhibitors. Translational Neurodegeneration. 2023;12(1):49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Jimenez-Gutierrez GE, Martíínez-Góómez LE, Martíínez-Armenta C, Pineda C, Martíínez-Nava GA, Lopez-Reyes A. Molecular Mechanisms of Inflammation in Sarcopenia: Diagnosis and Therapeutic Update. Cells. 2022;11(15):2359. 10.3390/cells11152359. [DOI] [PMC free article] [PubMed]
- 29.Liu X, Wang Y, Wang Z, Li L, Yang H, Liu J, et al. Association between sarcopenia-related traits and cardiovascular diseases: a bi-directional Mendelian randomization study. Front Endocrinol. 2023;14:1237971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Benz E, Pinel A, Guillet C, Capel F, Pereira B, De Antonio M, et al. Sarcopenia and sarcopenic obesity and mortality among older people. JAMA Netw Open. 2024;7(3):e243604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Liberale L, Badimon L, Montecucco F, Lüscher TF, Libby P, Camici GG. Inflammation, Aging, and cardiovascular disease: JACC review topic of the week. J Am Coll Cardiol. 2022;79(8):837–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chiva-Blanch G, Vilella-Figuerola A, Padró T, Formiga F, Ferrer A, Badimon L. Functional and cognitive decline is associated with increased endothelial cell inflammation and platelet activation: liquid biopsy of microvesicles in Community- dwelling octogenarians. Front Cell Dev Biology. 2021;9:716435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kipinoinen T, Toppala S, Rinne JO, Viitanen MH, Jula AM, Ekblad LL. Association of midlife inflammatory markers with cognitive performance at 10-Year Follow-up. Neurology. 2022;99(20):e2294–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Naomi R, Teoh SH, Embong H, Balan SS, Othman F, Bahari H, et al. The role of oxidative stress and inflammation in obesity and its impact on cognitive Impairments-A narrative review. Antioxid (Basel Switzerland). 2023;12(5):1071. 10.3390/antiox12051071. [DOI] [PMC free article] [PubMed]
- 35.Jin Y, Liang J, Hong C, Liang R, Luo Y. Cardiometabolic multimorbidity, lifestyle behaviours, and cognitive function: a multicohort study. Lancet Healthy Longev. 2023;4(6):e265–73. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The database used and/or analyzed for this study available from the corresponding author on reasonable request (Yan Yan; Email: [yanyan_zhongshan@sina.com](mailto: yanyan_zhongshan@sina.com) ; yan.yan@zs-hospital.sh.cn).