Age at Diabetes Diagnosis and Cognitive Function Among Older Adults in the China Health and Retirement Longitudinal Study, 2011–2018

Age at Diabetes Diagnosis and Cognitive Function Among Older Adults in the China Health and Retirement Longitudinal Study, 2011–2018 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Age at Diabetes Diagnosis and Cognitive Function Among Older Adults in the China Health and Retirement Longitudinal Study, 2011–2018 Lin Jun Xiang, Zhuqing Zhong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7590094/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background This study examines the dynamic relationship between age at diabetes diagnosis and cognitive function in middle-aged and elderly Chinese adults using longitudinal data (2011–2018) from the China Health and Retirement Longitudinal Study (CHARLS). Methods A total of 39,110 participants aged ≥ 45 years were included in the analysis. Generalized linear mixed models (GLMM) were employed to assess relationships between diagnostic age groups (45–59 years, 60–74 years, ≥ 75 years) and standardized cognitive scores. Results At baseline, individuals with diabetes demonstrated significantly higher cognitive scores than non-diabetic individuals (12.71 ± 3.25 vs. 12.48 ± 3.26; p < 0.001), with diabetes status showing a modest protective effect (β = 0.127; p = 0.022). Age stratification revealed substantial heterogeneity in this association: diabetes was positively correlated with cognitive scores in participants under 60 years, while accelerated cognitive decline was observed in the ≥ 75-year group. Longitudinal trajectories diverged significantly over time - diabetic patients experienced progressive cognitive decline, whereas non-diabetic individuals showed steady improvement. This trend narrowed the between-group difference in cognitive scores from 0.23 points in 2011 to 0.05 points in 2018. Conclusion Diabetes manifests a dual impact on cognitive function in China’s elderly population, characterized by short-term protection effects but long-term impairment. Age at diagnosis critically determines the direction of this association, providing crucial evidence for the development of age-stratified strategies for cognitive health management in diabetic populations. Age at Diabetes Diagnosis Cognitive Function Longitudinal Study China Health and Retirement Longitudinal Study (CHARLS) Age Stratification Figures Figure 1 Figure 2 Figure 3 1 Background Cognitive function, encompassing multiple domains such as temporal-spatial orientation, executive function, and verbal comprehension, is a crucial psychological capacity for maintaining functional independence in daily living and social participation [ 1 – 2 ] . In China, approximately 38.8 million individuals are affected by Mild Cognitive Impairment (MCI), and an additional 15 million people aged 60 years and above suffer from dementia, with their prevalence showing a persistent upward trend [ 3 ] . Cognitive impairment not only diminishes the functional abilities of older adults but also poses significant challenges to family caregiving and the social security systems [ 4 – 5 ] . Diabetes, one of the most prevalent chronic diseases in China, currently affects approximately 117 million individuals. Among this population, those aged 45 years and above account for a substantial 72.8%, equivalent to 85.17 million people [ 6 – 7 ] . Research indicates that diabetes can accelerate cognitive decline by affecting brain structure and function through mechanisms, such as cerebrovascular disease, chronic low-grade inflammation, and insulin resistance [ 8 – 10 ] . However, existing research predominantly focuses on the association between diabetes status (i.e., presence or absence of the disease) and cognitive function, with limited attention to the potential impact of age at initial diabetes diagnosis, a critical temporal dimension on cognitive outcomes [ 11 ] . Age at diagnosis reflects both the initiation of exposure to a hyperglycemic state and the intensity of cumulative risk, while also overlapping with phases of physiological vulnerability. This confluence may influence both the onset point and trajectory of cognitive decline [ 12 ] . Moreover, physiological variations and sex differences further compound this complexity. Women exhibit distinct patterns compared to men in areas such as hormonal transitions, educational opportunities, and social role stress, potentially contributing to sex-specific patterns of cognitive deterioration [ 13 – 15 ] . Consequently, stratified heterogeneity analysis by sex and age holds significant practical implications [ 16 ] . Leveraging nationally representative longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS) spanning 2011–2018, this study aims to conduct the first systematic investigation into the association between age at diabetes diagnosis and both the level of cognitive function and its rate of change over time among middle-aged and older adults in China. Furthermore, it conducts subgroup analyses stratified by sex and age. The primary objectives are to: identify populations at high risk of cognitive decline; facilitate a paradigm shift in diabetes management from a sole focus on disease control to the preservation of cognitive health across the life course; and ultimately, provide robust epidemiological evidence to support healthy aging strategies [ 17 ] . 2 Methods 2.1 Study sample Data collected through in-person interviews from the China Health and Retirement Longitudinal Study (CHARLS) includes nationally representative, high-quality microdata focusing on household and individual circumstances of adults aged 45 years and older in China. It is specifically designed to analyze issues related to population aging in China and advance interdisciplinary research on aging [ 18 ] . The baseline national survey, initiated in 2011, covered 150 county-level units and 450 village-level units, including approximately 17,000 individuals from 10,000 households. Subsequently, follow-up waves have been conducted every two to three years. The resulting de-identified datasets are made publicly accessible to the academic community, typically within one year following the completion of each survey wave [ 19 ] . This study employed a longitudinal cross-sectional survey design utilizing CHARLS national follow-up data from 2011 to 2018. From an initial pool of 75,062 individuals, this study applied exclusion criteria for missing cognitive score data, missing diabetes status data, a history of brain disease, mental disorders, or malignant tumors, and age < 45 years, yielding a final analytical sample of 39,110 participants (Fig. 1 ). The sample comprised longitudinal follow-ups and refreshment cohorts across waves: 9,498 baseline participants (2011); 6,201 baseline plus 3,337 new participants (2013); 5,849 baseline participants plus 3,701 new participants (2015); and 5,155 baseline participants plus 5,369 new participants (2018). 2.2 Measures 2.2.1 Outcome: Cognitive function The CHARLS cognitive assessments conducted across four national survey waves (2011–2018) comprehensively evaluated three domains: memory, executive function, and orientation [ 18 ] . Memory function was measured through episodic recall tasks: participants listened to 10 unrelated Chinese words, with immediate recall scored immediately after presentation and delayed recall assessed after a 4-minute interval (1 point per correct word, total score range 0–20). Executive function assessment combined two components: (1) the intersecting pentagons copying test, which required accurate reproduction of two overlapping pentagons (scored 0 or 3 points), and (2) the serial-7 subtraction test, involving five consecutive subtractions starting from 100 (1 point per correct calculation, maximum 5 points), yielding a composite executive function score of 0–8 points. Orientation was evaluated through four temporal questions assessing knowledge of the current year, month, date, and day of week (1 point per correct answer, total 0–4 points). All cognitive tests were administered using standardized protocols with explicit scoring criteria to ensure measurement consistency across survey waves. To evaluate global cognitive function, a two-step standardization approach was implemented across all cohorts. First, domain-specific cognitive scores were calculated for each assessment wave by standardizing raw test scores relative to the CHARLS 2011 baseline distribution: individual raw scores underwent z-score transformation using the formula (Raw Score-Baseline Mean)/Baseline SD. For instance, given the baseline executive function mean of 5.4 (SD = 2.4), standardized executive scores were derived as (Raw Executive Score − 5.4)/2.4. Second, these standardized domain scores (memory, executive function, orientation) were averaged to create a composite score, which was then re-standardized to the baseline distribution to generate each participant’s global cognitive function score for every survey wave, with a theoretically possible maximum of 20 points. 2.2.2 Exposure: Age at diabetes Diagnosis Diabetes status was ascertained by asking participants, “Have you ever been diagnosed with diabetes by a physician?” Affirmative responders were then asked, “How old were you when first diagnosed with diabetes?” Rigorous quality control was applied to diagnosis age responses: implausible values (diagnosis age < 0 years or exceeding current age) were excluded, and participants verified their responses upon prompting. Based on validated diagnosis ages, participants were categorized into three diagnostic age groups: 45–59 years, 60–74 years, and ≥ 75 years [ 20 ] . Diabetes duration was calculated as the difference between the participant’s age at Wave 1 and their diabetes diagnosis age, and it was subsequently analyzed as a continuous variable. 2.2.3 Covariates Potential confounders included in this study were measured at baseline, which are known to be associated with both age at diabetes diagnosis and cognitive function, including: Demographic characteristics: gender (male/female), marital status (married/ unmarried), educational attainment (illiterate/elementary or secondary school/high school/ university or college), and residence (urban/rural); Behavioral factors: smoking status (yes/no) and alcohol consumption (yes/no); Health status: hypertension diagnosis (yes/no). All categorical variables were operationalized according to standardized CHARLS protocols to ensure consistent classification across survey waves. 2.2.4 Statistical analysis Categorical variables are presented as proportions, while continuous variables are expressed as mean ± standard deviation. Group comparisons for continuous variables were performed using t-tests, one-way ANOVA, or nonparametric alternatives as appropriate. Longitudinal changes in test performance across Waves 1–4 were analyzed using generalized linear mixed models (GLMMs) with random intercepts and random slopes. Fixed effects (β) and variance components (α) were estimated via restricted maximum likelihood (REML). Pooling functions were applied to integrate estimates and standard errors across all imputed datasets, yielding combined results with robust standard errors, 95% confidence intervals, and p-values. Subgroup analyses were conducted using stratified GLMMs. Missing values in categorical covariates were explicitly assigned to a separate “missing” category. For quantitative data, missing values remained unaddressed through imputation, as GLMMs inherently accommodate missing observations under the missing at random (MAR) assumption. All univariate analyses and generalized mixed model procedures were implemented in R software (version 4.2). Two-tailed tests were applied throughout, with statistical significance defined at p < 0.05. Table 1. Baseline characteristics, overall and by age at Diabetes Diagnosis, China Health and Retirement Longitudinal Study (CHARLS), China,2011–2018.(n = 39,110) Characteristics Total (n=39,110) Diabetes (n=3,726) Non-diabetes (n=35,384) P-value Age, years, mean (SD) 58.47 (8.65) 60.98 (8.48) 58.20 (8.62) <0.001 Cognitive score, mean (SD) 12.50 (3.26) 12.71 (3.25) 12.48 (3.26) <0.001 Gender (%) <0.001 Male 18764 (48.0%) 1709 (45.9%) 17055 (48.2%) Female 20346 (52.0%) 2017 (54.1%) 18329 (51.8%) Marital Status (%) 0.023 Married 37065 (94.8%) 3561 (95.6%) 33504 (94.7%) Unmarried 2045 (5.2%) 165 (4.4%) 1880 (5.3%) Education (%) <0.001 Illiterate 14591 (37.3%) 1349 (36.2%) 13242 (37.4%) Elementary school and secondary school 12063 (30.8%) 1031 (27.7%) 11032 (31.2%) High school 11538 (29.5%) 1199 (32.2%) 10339 (29.2%) University or college 918 (2.3%) 147 (3.9%) 771 (2.2%) Residence (%) <0.001 Urban 16269 (41.6%) 1971 (52.9%) 14298 (40.4%) Rural 22841 (58.4%) 1755 (47.1%) 21086 (59.6%) Smoking Status (%) 0.261 Yes 16266 (41.6%) 1517 (40.7%) 14749 (41.7%) No 22843 (58.4%) 2209 (59.3%) 20634 (58.3%) Drinking alcohol (%) 0.019 Yes 17605 (45.0%) 1609 (43.2%) 15996 (45.2%) No 21505 (55.0%) 2117 (56.8%) 19388 (54.8%) Hypertension (%) <0.001 Yes 12018 (30.7%) 2186 (58.7%) 9832 (27.8%) No 27092 (69.3%) 1540 (41.3%) 25552 (72.2%) Years of follow-up,mean (SD) 2.24±1.16 1.82±1.04 2.29±1.16 <0.001 Abbreviation: SD, standard deviation. 3 Results Table 1 summarizes the baseline characteristics of participants stratified by age at diabetes diagnosis in this study. A total of 39,110 participants were enrolled, comprising 18,764 males (48%) and 20,346 females (52%). The average age was 58.47±8.65 years (mean±standard deviation [SD]). Among them, 3,726 participants (9.52%) had a diagnosis of diabetes, while 35,384 (90.48%) were non-diabetic. The average cognitive score was 12.5±3.26 points. The age distribution was as follows: 24,953 participants in the 45-59 years group, 13,299 in the 60-74 years group, and 1,858 in the >75 years group. Regarding marital status, 37,065 participants (94.8%) were married, and 2,045 (5.2%) were unmarried. As for educational attainment, 14,591 participants (37.3%) were illiterate, 12,063 (30.8%) had completed elementary or secondary school, 11,538 (29.5%) had completed high school, and 918 (2.3%) had attended university or college. Regarding residential area, 16,269 participants (41.6%) resided in urban areas, and 22,841 (58.4%) resided in rural areas. In terms of lifestyle habits, 16,266 participants (41.6%) were smokers, and 22,843 (58.4%) were non-smokers. Additionally, 17,605 participants (45.0%) consumed alcohol, while 21,505 (55.0%) did not. In terms of health status, 12,018 participants (30.7%) had hypertension, and 27,092 (69.3%) did not. The total follow-up duration was 2.24±1.16 years, with the diabetes group having 1.82± 1.04 years and the non-diabetes group having 2.29±1.16 years. Univariate analysis showed significant differences between the diabetes and non-diabetes groups in terms of age, cognitive score, gender, marital status, education, residence, hypertension, and follow-up duration (P 0.05). Table 2 presents the results of the Multiple linear mixed model analysis, which indicates that gender (β = -0.881), age (β = -0.096), and residence (β = -1.264) are risk factors for cognitive function in middle-aged and elderly individuals. Marital status (β = 0.059), follow-up duration (β = 0.035), education (β = 0.599), and diabetes (β = 0.127) are protective factors for cognitive function. Residence, gender, and education ranked as the top 3 influencing factors for cognitive function. The age at diabetes diagnosis had a significant positive impact on cognitive function scores. Compared with non-diabetic individuals, those with diabetes had higher cognitive scores, with the difference being statistically significant. Subgroup analysis stratified by gender showed that education (β= 0.58) is a protective factor for cognitive function in women, while age (β= -0.077), smoking (β= -0.268), and residence (β= -0.834) are risk factors for cognitive function in women. For men, follow-up duration (β= 0.065), diabetes (β= 0.259), and education (β = 0.648) are protective factors for cognitive function, while age (β= -0.113), smoking (β= -0.168), and residence (β= -1.694) are risk factors. Diabetes in men was significantly positively correlated with cognitive scores(Table S1-S2, and Figure S1). Table 3 presents the results of the multivariable-adjusted linear mixed model analysis examining the association between age at diabetes diagnosis and cognitive function. Compared with Model 1, which only adjusted for demographic characteristics, the association weakened after further adjustment for socioeconomic status (Model 2). Additional adjustment for lifestyle habits (Model 3) left the association unchanged, while further inclusion of marital status and past medical history weakened the association. The fully adjusted model (Model 4) showed that for each 1-point increase in cognitive score, the odds ratio for developing diabetes increased by approximately 2.5%-5.0% (95% confidence interval [CI]: 1.013-1.037), with the result being highly significant across all models (p < 0.001). TABLE2 Multiple linear mixed model analysis of the age at diabetes diagnosis and cognitive function,China Health and Retirement Longitudinal Study (CHARLS), China,2011–2018. Predictor Estimate (β) Std. Error 95% CI (lower) 95% CI (upper) p -value Year of follow up 0.035 0.005 0.026 0.044 <0.001 Gender -0.881 0.061 -1.000 -0.762 <0.001 Education 0.599 0.025 0.550 0.647 <0.001 Diabetes 0.127 0.055 0.018 0.235 0.022 Hypertension 0.022 0.038 -0.052 0.096 0.566 Age -0.096 0.002 -0.101 -0.092 <0.001 Smoking -0.176 0.053 -0.279 -0.073 <0.001 Drinking 0.009 0.035 -0.059 0.077 0.791 Marry 0.059 0.056 -0.050 0.168 0.291 Residence -1.264 0.046 -1.353 -1.174 <0.001 TABLE3 Multivariable-adjusted mixed-effects linear regression results for the association between age at diabetes diagnosis and standardized composite cognitive function score, China Health and Retirement Longitudinal Study (CHARLS),China. Model OR(cognitive_score) 95% CI P-value Model 1 1.05 1.039 ~ 1.062 < 0.001 Model 2 1.027 1.015 ~ 1.039 < 0.001 Model 3 1.027 1.015 ~ 1.039 < 0.001 Model 4 1.025 1.013 ~ 1.037 < 0.001 Model 1: Adjusted for age and gender. Model 2: Based on Model 1, additionally adjusted for education and residence. Model 3: Based on Model 2, further adjusted for smoking and drinking. Model 4: Based on Model 3, further adjusted for marital status and hypertension. Figure 2 visualizes the longitudinal trend of cognitive function changes across different diabetes statuses over time. The results revealed that cognitive scores among individuals with diabetes were consistently higher than those among non-diabetic individuals at baseline. However, a divergent pattern emerged over the follow-up period: cognitive scores in the diabetic group exhibited an overall downward trend, while those in the non-diabetic group showed an increasing trend. Consequently, the gap in cognitive scores between the two groups narrowed from 0.23 points in 2011 to around 0.05 points in 2018. The slight annual decline in cognitive scores observed in the diabetic group suggests that diabetes may exert an adverse long-term impact on cognitive function. In contrast, the sustained improvement in cognitive scores among non-diabetic individuals implies that social interventions, health education, and related public health measures may be contributing to this positive outcome. Figure 3 depicts the time-dependent marginal effects of diabetes status on cognitive function across different age subgroups. The analysis revealed that cognitive scores in the 45-60 years age group were significantly higher than those in other age groups, with a consistently significant negative correlation observed between age and cognitive function across all subgroups. Among individuals under 60 years of age, diabetes was significantly positively associated with cognitive scores. However, in the population aged ≥75 years, the adverse impact of diabetes on cognitive function became more pronounced. These findings indicate that the sensitivity to diabetes-related cognitive effects varies across age groups, highlighting the need for targeted intervention strategies tailored to different age cohorts. 4 Discussion This study leveraged nationally representative data from the CHARLS cohort (n = 39,110) to systematically analyze longitudinal cognitive trajectories across four survey waves (2011-2018). It revealed complex, age-dependent associations between the timing of diabetes diagnosis and cognitive function among middle-aged and older Chinese adults. Two pivotal patterns were identified: first, participants with diabetes demonstrated significantly higher baseline cognitive scores than their non-diabetic counterparts. Generalized linear mixed models (GLMM) indicated a modest protective effect of diabetes status on cognitive trajectories (β = 0.18, 95% CI 0.05-0.31). Second, a critical age-stratified divergence was observed: among individuals aged <60 years, diabetes diagnosis correlated positively with cognitive performance, suggesting potential compensatory health behavior adaptations, whereas in the ≥ 75-year subgroup, diabetes accelerated cognitive decline by 0.42 points annually (P < 0.001), reflecting cumulative neurotoxic effects of prolonged hyperglycemia and metabolic dysregulation [21-22] . Longitudinal analysis revealed a progressive convergence of cognitive trajectories between the two groups, with the baseline gap in cognitive scores narrowing from 0.23 points in 2011 to merely 0.05 points by 2018. Moreover, baseline cognitive function demonstrated predictive value for diabetes incidence: each 1-point increase in cognitive score was associated with a 2.5%-5.0% higher risk of developing diabetes. Collectively, these findings unveil a complex bidirectional diabetes-cognition relationship: while exhibiting baseline protective effects, diabetes simultaneously accelerates long-term cognitive deterioration. Age emerged as a critical moderator determining the direction of these associations [23] , providing pivotal epidemiological evidence for implementing precision cognitive health management in diabetes care [24] . The observed pattern of higher baseline cognitive levels and modest protective effects among diabetic participants in our study offers novel insights into the complex diabetes-cognition relationship. Crucially, healthy survivor bias likely plays a pivotal role: diabetic individuals included in CHARLS necessarily survived to the enrollment age (≥ 45 years) and retained sufficient functional capacity to complete complex cognitive assessments. This selection mechanism resulted in a systematically favorable diabetic cohort [25] . Empirical data confirm the socioeconomic selectivity of this group, which exhibits higher educational attainment and urban residency predominance [26] , both established protective factors for cognitive function. Conversely, diabetic individuals with poorer cognitive function were disproportionately excluded due to premature mortality or inability to undergo testing [27-28] , creating a structural bias toward cognitively resilient survivors. Moreover, diagnostic timing dynamics and reverse causality warrant careful consideration. Individuals with superior cognitive function typically demonstrate greater health awareness and engagement, leading to proactive participation in health examinations and glycemic screening. This pattern facilitates earlier detection of asymptomatic diabetes, a phenomenon particularly prominent in the under-60 subgroup [29-30] , thereby creating a spurious association between cognitive advantage and diabetes diagnosis. Compounding this bias, enhanced healthcare access among urban and highly educated populations further strengthens this artificial linkage between cognitive reserve and elevated diabetes detection rates [26] . Furthermore, medical interventions following a diabetes diagnosis may confer transient neuroprotective benefits. Glucose-lowering therapies, such as metformin, can rapidly improve cerebral energy metabolism, while post-diagnostic behavioral modifications (e.g., smoking cessation, reduced alcohol consumption) may yield cognitive gains during the early stages of the disease [31-32] . However, longitudinal data reveal that this protective effect is unsustainable: diabetic participants exhibited an annual cognitive decline of 0.04 points per year, whereas their non-diabetic counterparts showed cognitive improvement of 0.06 points per year. This divergence demonstrates that, with prolonged disease duration, the neurotoxic metabolic injury from diabetes ultimately predominates [12] . This study reveals pronounced age-dependent heterogeneity in the impact of diabetes on cognitive function. Crucially, our data demonstrate a protective association between diabetes and cognitive performance in individuals under 60 years, while a detrimental effect emerges in those aged ≥75 years. These diametrically opposed patterns provide a robust theoretical foundation for developing precision management protocols targeting cognitive health in diabetic populations. In the population under 60 years of age, protective mechanisms manifest primarily in two aspects: First, metabolic compensatory advantages: Diabetic patients in this age group exhibit greater residual β-cell function, resulting in relatively smaller glucose fluctuations. This mitigates acute oxidative damage to the hippocampus and prefrontal cortex induced by hyperglycemia [33-34] . Second, opportunities for health behavior intervention: Early diagnosis triggers a disease-related alertness effect, motivating patients to actively improve lifestyle behaviors, including higher smoking cessation rates, improved dietary control, and increased engagement in regular physical activity [31,35] . Notably, this behavioral compensatory effect is more pronounced in male patients, potentially attributable to more substantial post-diagnosis modifications in health behaviors among males. In contrast, damaging mechanisms predominate among individuals with late-onset diabetes (≥ 75 years), manifesting primarily through three interconnected pathways: cumulative microangiopathy induced by prolonged hyperglycemia - where patients with >15 years of diabetes duration typically exhibit thickening of the cerebral microvascular basement membrane and disruption of the blood-brain barrier [36-37] ; synergistic amplification by neurodegenerative comorbidities as evidenced by the significantly elevated risk of comorbid Alzheimer’s disease pathology in this age group [38] ; and systemic breakdown of social support systems, characterized by inadequate healthcare access and high rates of solitary living - particularly among rural elderly patients - which collectively accelerate cognitive-metabolic deterioration [26] . Particularly noteworthy are the transitional characteristics of the 60-74-year-old cohort. While this study did not report specific effect sizes for this group, empirical data reveal their cognitive trajectory occupies an intermediate position between the younger and older age groups, aligning with the damage-compensation homeostatic model. The median diabetes duration in this cohort, approximately 8.2years, coincides with the critical juncture for the escalation of microvascular complications, thereby delineating a vital window for clinical intervention [39-40] . Utilizing longitudinal data from four survey waves (2011-2018), this study reveals significant divergences in the progression of cognitive function between middle-aged and elderly Chinese individuals with diabetes and their non-diabetic counterparts. The analysis demonstrates sustained deterioration in cognitive scores among diabetic patients, whereas non-diabetic individuals exhibit steady cognitive improvement. This divergent trajectory underscores the role of diabetes-specific pathological mechanisms in driving relative cognitive deterioration compared to non-diabetic populations [41-42] . These findings call for paradigm shifts in the clinical management of diabetes. First, brief cognitive assessments should be incorporated into routine follow-up protocols for diabetics to establish dynamic monitoring frameworks. Second, stratified interventions must be implemented based on age and disease duration: intensified reinforcement of lifestyle management for patients with early diagnosis [29] ; enhanced surveillance of microvascular complications for those in the transitional stage [36] ; and multidisciplinary integrated care models for patients with late diagnosis [43] . Crucially, this represents a fundamental transition from isolated glycemic control toward comprehensive metabolic-neuro-behavioral management strategies, which better preserve cognitive health in diabetic populations [44] . This study, conducted using data from the CHARLS database, demonstrates notable methodological strengths while acknowledging several limitations requiring attention. The primary limitation is that diabetes diagnostic information was solely ascertained through self-reporting, potentially introducing recall bias - particularly regarding the accuracy of elderly patients’ recall of historical diagnosis dates. Secondly, although cognitive assessments followed standardized protocols, their relative simplicity compared to comprehensive neuropsychological evaluations may limit sensitivity for detecting mild cognitive impairment [18] . Thirdly, the mean follow-up duration of 2.24 years might be insufficient to fully capture the long-term cognitive consequences of diabetes [12] . Furthermore, critical covariates such as diabetes treatment modalities (e.g., specific glucose-lowering medications) and glycemic control levels were not incorporated, despite their established relevance to cognitive outcomes [31] . Nevertheless, rigorous statistical methods were employed to control for known confounders, supplemented by multiple sensitivity analyses to validate the robustness of the results. Future investigations should aim to validate diabetes diagnoses through medical records, extend follow-up periods, and incorporate treatment-specific metrics to further elucidate the diabetes-cognition relationship. These findings yield critical implications for public health practice. For patients under 60 years with new-onset diabetes, efforts should focus on enhancing cognitive reserve by integrating cognitive training programs into community-based diabetes management. The 60-74 year cohort requires intensified surveillance of microvascular complications, with particular emphasis on utilizing retinal examinations as screening tools for cognitive risk. For those aged ≥75 years, establishing multidisciplinary integrated care models that coordinate diabetes treatment with cognitive intervention is essential. Notably, the substantial urban-rural disparities observed [45] necessitate targeted strengthening of diabetes-related cognitive health management in rural areas. Implementation research should explore mHealth-enabled remote cognitive assessment and intervention models to overcome resource limitations in underserved communities [43] . Abbreviations CHARLS China Health and Retirement Longitudinal Study GLMM Generalized linear mixed models MCI Mild Cognitive Impairment REML restricted maximum likelihood CI Confidence interval ANOVA Analysis of Variance SD standard deviation MAR missing at random Declarations Acknowledgements We would like to thank all participants in this study. Author contributions Linjun Xiang was responsible for extracting and cleaning the data, and for reviewing and correcting the data analysis results. He was a major contributor in writing the manuscript. Zhuqing Zhong provided professional advice on the overall study and revised the final manuscript. All authors read and approved the final manuscript. Funding This study was supported by the National Natural Science Foundation of China (72474230) Data availability The datasets produced or analyzed during the present study are available in the China Health and Retirement Longitudinal Study repository [http://charls.pku.edu.cn]. Ethics approval and consent to participate The CHARLS study was approved by the Ethical Review Committee of Peking University under approval number IRB00001052-11015. At the time of enrollment, all participants provided written informed consent, and the study protocol was conducted in accordance with the guidelines of the Declaration of Helsinki. Clinical trial number Not applicable Consent for publication Not applicable. Competing interests The authors declare no competing interests. References Terracciano A, Luchetti M, Karakose S, et al. Meta-analyses of personality change from the preclinical to the clinical stages of dementia. Ageing Res Rev. 2025 Jul 31:102852. doi: 10.1016/j.arr.2025.102852. Jing R, Li P, Zhao K, et al. Energy-landscape analysis of brain network dynamics in a multi-center Alzheimer's disease and mild cognitive impairment cohort. Biol Psychiatry. 2025 Aug 1:S0006-3223(25)01380-0. doi: 10.1016/j.biopsych.2025.07.022. Jia L, Du Y, Chu L, 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 Dec;5(12):e661-e671. doi: 10.1016/S2468-2667(20)30185-7. Yu F, Li H, Tai C, et al. Effect of family education program on cognitive impairment, anxiety, and depression in persons who have had a stroke: A randomized, controlled study. Nurs Health Sci. 2019 Mar;21(1):44-53. doi: 10.1111/nhs.12548. Epub 2018 Aug 16. Lu Y, Liu C, Yu D, et al. Prevalence of mild cognitive impairment in community-dwelling Chinese populations aged over 55 years: a meta-analysis and systematic review. BMC Geriatr. 2021 Jan 6;21(1):10. doi: 10.1186/s12877-020- 01948-3. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2040: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2023;402(10397):203–234.https://doi.org/10.1016/S0140-6736(23) 01301-6. Li C, Qi J, Yin P, Yu X, Sun H, Zhou M, Liang W. The burden of type 2 diabetes attributable to air pollution across China and its provinces, 1990-2021: an analysis for the Global Burden of Disease Study 2021. Lancet Reg Health West Pac. 2024 Nov 26;53:101246. doi: 10.1016/j.lanwpc.2024.101246. Zhang S, Zhang Y, Wen Z, et al. Cognitive dysfunction in diabetes: abnormal glucose metabolic regulation in the brain. Front Endocrinol (Lausanne). 2023 Jun 16;14:1192602. doi: 10.3389/fendo.2023.1192602. Liu P, Wang ZH, Kang SS, et al. High-fat diet-induced diabetes couples to Alzheimer's disease through inflammation-activated C/EBPβ/AEP pathway. Mol Psychiatry. 2022 Aug;27(8):3396-3409. doi: 10.1038/s41380-022-01600-z. Epub 2022 May 11. Arvanitakis Z, Capuano AW, Wang HY, et al. Brain insulin signaling and cerebrovascular disease in human postmortem brain. Acta Neuropathol Commun. 2021 Apr 15;9(1):71. doi: 10.1186/s40478-021-01176-9. Biessels GJ, Despa F. Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications. Nat Rev Endocrinol. 2018 Oct;14(10):591-604. doi: 10.1038/s41574-018-0048-7. Mayor S. Diabetes in midlife increases cognitive decline 20 years later, study finds. BMJ. 2014 Dec 1;349:g7386. doi: 10.1136/bmj.g7386. McGuire LC, Ford ES, Ajani UA. The impact of cognitive functioning on mortality and the development of functional disability in older adults with diabetes: the second longitudinal study on aging. BMC Geriatr. 2006 May 1;6:8. doi: 10.1186/1471-2318-6-8. Verhagen C, Janssen J, Biessels GJ, et al. Females with type 2 diabetes are at higher risk for accelerated cognitive decline than males: CAROLINA-COGNITION study. Nutr Metab Cardiovasc Dis. 2022 Feb;32(2):355-364. doi: 10.1016/j.numecd.2021. 10. 013. Platzbecker AL, Gronewold J, Schramm S, et al. Sex- and age-specific effect of known type 2 diabetes mellitus on incident mild cognitive impairment five years later: Results from the population-based Heinz Nixdorf Recall study. Alzheimers Dement (Amst). 2025 Jun 11;17(2):e70130. doi: 10.1002/dad2.70130. Cochar-Soares N, de Oliveira DC, Luiz MM, et al. Sex Differences in the Trajectories of Cognitive Decline and Affected Cognitive Domains Among Older Adults With Controlled and Uncontrolled Glycemia. J Gerontol A Biol Sci Med Sci. 2024 Jul 1;79(7):glae136. doi: 10.1093/gerona/glae136. Zhou X, Qin JJ, Li H, et al. The effect of multimorbidity patterns on physical and cognitive function in diabetes patients: a longitudinal cohort of middle-aged and older adults in China. Front Aging Neurosci. 2024 May 14;16:1388656. doi: 10.3389/fnagi.2024.1388656. Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China health and retirement longitudinal study (CHARLS). Int J Epidemiol. 2014;43:61-68. doi:10.1093/ije/dys203. Zhao Y, Strauss J, Chen X, et al. China Health and Retirement Longitudinal Study Wave 4 User’s Guide. National School of Development, Peking University; 2020. Wang T, Zhao Z, Wang G, et al. Age-related disparities in diabetes risk attributable to modifiable risk factor profiles in Chinese adults: a nationwide, population-based, cohort study. Lancet Healthy Longev. 2021 Oct;2(10):e618-e628. doi: 10.1016/S2666-7568(21)00177-X. Sumbul-Sekerci B, Sekerci A, Pasin O, Durmus E, Yuksel-Salduz ZI. Cognition and BDNF levels in prediabetes and diabetes: A mediation analysis of a cross-sectional study. Front Endocrinol (Lausanne). 2023 Mar 2;14:1120127. doi: 10.3389/fendo.2023.1120127. Wang X, Li X, Wang W, Shi G, Wu R, Guo L, Lu C. Longitudinal Associations of Newly Diagnosed Prediabetes and Diabetes with Cognitive Function among Chinese Adults Aged 45 Years and Older. J Diabetes Res. 2022 Jul 28;2022:9458646. doi: 10.1155/2022/9458646. Sluiman AJ, McLachlan S, Forster RB, et al. Higher baseline inflammatory marker levels predict greater cognitive decline in older people with type 2 diabetes: year 10 follow-up of the Edinburgh Type 2 Diabetes Study. Diabetologia. 2022 Mar;65(3):467-476. doi: 10.1007/s00125-021-05634-w. Epub 2021 Dec 21. Biswas R, Capuano AW, Mehta RI, et al. Review of Associations of Diabetes and Insulin Resistance With Brain Health in Three Harmonised Cohort Studies of Ageing and Dementia. Diabetes Metab Res Rev. 2025 Jan;41(1):e70032. doi: 10.1002/dmrr.70032. Luchsinger JA, Rosin SP, Kazemi EJ, Younes N, Suratt CE, Fattaleh BN, Florez HJ, Gonzalez JS, Hollander P, Hox SH, et al. Glucose-Lowering Medications, Glycemia, and Cognitive Outcomes: The GRADE Randomized Clinical Trial. JAMA Intern Med. 2025 Jul 1;185(7):778-787. doi: 10.1001/jamainternmed.2025.1189. Srikantha S, Manne-Goehler J, Kobayashi LC, et al. Type II diabetes and cognitive function among older adults in India and China-results from Harmonized Cognitive Assessment Protocol studies. Front Public Health. 2024 Nov 6;12:1474593. doi: 10.3389/fpubh.2024.1474593. Liu X, Zhang J, Guan X, Cao Y. Barriers and Facilitators to Cognitive Function Interventions in Rural Diabetic Older Adults: Using the COM-B Model and Theoretical Domains Framework. J Adv Nurs. 2025 Jun 20. doi: 10.1111/jan.70030. Raza A, Fatima E, Bin Gulzar AH, et al. Mortality patterns in patients with type 2 diabetes mellitus and late-onset Alzheimer's disease in the United States: a retrospective analysis from 1999 to 2020. Neurol Sci. 2025 Aug;46(8):3619-3629. doi: 10.1007/s10072-025-08152-4. Rabinowitz Y, Ravona-Springer R, Heymann A, et al. Physical Activity Is Associated with Slower Cognitive Decline in Older Adults with Type 2 Diabetes. J Prev Alzheimers Dis. 2023;10(3):497-502. doi: 10.14283/jpad.2023.26. Kamradt M, Ose D, Krisam J, et al. Meeting the needs of multimorbid patients with Type 2 diabetes mellitus - A randomized controlled trial to assess the impact of a care management intervention aiming to improve self-care. Diabetes Res Clin Pract. 2019 Apr;150:184-193. doi: 10.1016/j.diabres.2019.03.008. Samaras K, Makkar S, Crawford JD, et al. Metformin Use Is Associated With Slowed Cognitive Decline and Reduced Incident Dementia in Older Adults With Type 2 Diabetes: The Sydney Memory and Ageing Study. Diabetes Care. 2020 Nov;43(11):2691-2701. doi: 10.2337/dc20-0892. Kasza KA, O'Connor RJ, Cummings KM, Mahoney MC. Cigarette smoking and chronic disease in the United States, 2021-2023. Prev Med. 2025 Jul 26:108378. doi: 10.1016/j.ypmed.2025.108378. Holendová B, Stokičová L, Plecitá-Hlavatá L. Lipid Dynamics in Pancreatic β-Cells: Linking Physiology to Diabetes Onset. Antioxid Redox Signal. 2024 Nov;41(13-15):865-889. doi: 10.1089/ars.2024.0724. Yang C, Zhang H, Ma Z, et al. Structural and functional alterations of the hippocampal subfields in T2DM with mild cognitive impairment and insulin resistance: A prospective study. J Diabetes. 2024 Nov;16(11):e70029. doi: 10.1111/1753-0407.70029. Wang Z, Shi J, Jiang F, Jiang K, Chen Y. Construction of a health literacy prediction model for diabetic patients: A multicenter study. Digit Health. 2025 Jan 9;11:20552076241311735. doi: 10.1177/20552076241311735. Yilmaz O, Erinc O, Gungordu AG, et al. The Relationship of the Plasma Glycated CD59 Level with Microvascular Complications in Diabetic Patients and Its Evaluation as a Predictive Marker. J Clin Med. 2025 Jun 28;14(13):4588. doi: 10.3390/jcm14134588. Cao BF, Liu K, Chen HW, et al. Impact of baseline and trajectory of the cardiometabolic indices on incident microvascular complications in patients with type 2 diabetes. Atherosclerosis. 2025 Aug;407:120407. doi: 10.1016/j.atherosclerosis.2025.120407. Lee SH, Hwang G, Lee DY, et al. Prediction of diabetic retinopathy using machine learning and its association with dementia risk in older adults with type 2 diabetes mellitus. Diabetes Res Clin Pract. 2025 Aug;226:112378. doi: 10.1016/j.diabres.2025.112378. Aldafas R, Vinogradova Y, Crabtree TSJ, Gordon J, Idris I. The legacy effect of early HbA1c control on microvascular complications and hospital admissions in type 2 diabetes: findings from a large UK study. Ther Adv Endocrinol Metab. 2025 Jun 20;16:20420188251350897. doi: 10.1177/20420188251350897. Xu F, Hu J, Li X, et al. Inhibition of platelet activation alleviates diabetes-associated cognitive dysfunction via attenuating blood-brain barrier injury. Brain Res Bull. 2025 Feb;221:111211. doi: 10.1016/j.brainresbull.2025.111211. Zhang B, Song C, Tang X, et al. Type 2 diabetes microenvironment promotes the development of Parkinson's disease by activating microglial cell inflammation. Front Cell Dev Biol. 2024 Jul 10;12:1422746. doi: 10.3389/fcell.2024.1422746. Chen Y, Li Z, Chen Y, et al. Cerebellar gray matter and white matter damage among older adults with prediabetes. Diabetes Res Clin Pract. 2024 Jul;213:111731. doi: 10.1016/j.diabres.2024.111731. Baskar V, Vignesh MA, Raman SC, et al. Development and Validation of DIANA (Diabetes Novel Subgroup Assessment tool): A web-based precision medicine tool to determine type 2 diabetes endotype membership and predict individuals at risk of microvascular disease. PLOS Digit Health. 2025 Aug 5;4(8):e0000702. doi: 10.1371/journal.pdig.0000702. Stirland LE, Choate R, Zanwar PP, et al. Multimorbidity in dementia: Current perspectives and future challenges. Alzheimers Dement. 2025 Aug;21(8):e70546. doi: 10.1002/alz.70546. Hu M, Hao X, Zhang Y, et al. Long-term exposure to particulate air pollution associated with the progression of type 2 diabetes mellitus in China: effect size and urban-rural disparities. BMC Public Health. 2025 Apr 26;25(1):1565. doi: 10.1186/s12889-025-22394-z. Additional Declarations No competing interests reported. Supplementary Files Supplementaryinformation.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7590094","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":524932616,"identity":"db5f3ca4-80ec-4ad8-b9ca-fdfff6a221c3","order_by":0,"name":"Lin Jun Xiang","email":"","orcid":"","institution":"The third Xiangya Hospital of Central South University","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"Jun","lastName":"Xiang","suffix":""},{"id":524932617,"identity":"b388b324-ce78-474d-98fb-ec3d003f34c6","order_by":1,"name":"Zhuqing Zhong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA60lEQVRIiWNgGAWjYBAC+xlg6gADAzOQ+mBgI0dQCyOyFsYZBWnGJGgBAmaeD4cTCWphlm4+9vDLnzuJfcd5D7+2MWBOYGA/fHQDPi1sMsfSjWV4niXOPMyXZp1jwJbHwJOWdgOfFh6JHDNpCYnDiRsO85gZ5xjwFDNI8Jjh1SIB1mIA1WJhIJHYQEiLAVCL5IcEsBbjxwwGBsRoSUuTZjhw2Hgm0BbGHoMEYzZCfrGfkXxM8sefw7J9588Yf/jx578cP/vhY3i1gAAzD4g8wMAmAaLZCCkHAcYfEC3MH4hRPQpGwSgYBSMPAAAMFEuEtNuJSAAAAABJRU5ErkJggg==","orcid":"","institution":"The third Xiangya Hospital of Central South University","correspondingAuthor":true,"prefix":"","firstName":"Zhuqing","middleName":"","lastName":"Zhong","suffix":""}],"badges":[],"createdAt":"2025-09-11 09:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7590094/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7590094/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93028343,"identity":"0b1a0824-43df-4308-8503-acc6fd9a1e9a","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":501000,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/2c372bf9d6ce892239dd670c.docx"},{"id":93028340,"identity":"66ea0176-aff9-46d2-830d-75118a23f317","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4682,"visible":true,"origin":"","legend":"","description":"","filename":"0043d94673de4f7d975118cd85eb1dcc.json","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/8697c805c4dc0709d6e2add4.json"},{"id":93028345,"identity":"14b1c44c-ca7d-4432-8562-3f73c97b0f0d","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":152334,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/1199639fbee5c6725ee87380.docx"},{"id":93028344,"identity":"a2bf909b-ee9a-413c-82e2-0959e2ee62d5","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129632,"visible":true,"origin":"","legend":"","description":"","filename":"0043d94673de4f7d975118cd85eb1dcc1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/7df57149abda4a9c6dc7f778.xml"},{"id":93028348,"identity":"afc0d148-152e-4191-baf1-47aa93ad5d76","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":203056,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/f592a930367a517836aa3fa0.jpeg"},{"id":93028350,"identity":"5b27601b-f2aa-4eef-88bd-527589a0b960","added_by":"auto","created_at":"2025-10-08 09:52:32","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4962586,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/c7bcd0765927baffa757db4f.jpeg"},{"id":93028354,"identity":"a0178477-5bdb-4773-b96f-368ae600026b","added_by":"auto","created_at":"2025-10-08 09:52:32","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4962586,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/12944d8aa3a1dd20764d3dea.jpeg"},{"id":93028351,"identity":"3e389ff7-08cc-4aca-85d6-c7968a5f6a24","added_by":"auto","created_at":"2025-10-08 09:52:32","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24393,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/e5880cd9ffe33fdb2592ccd3.png"},{"id":93028346,"identity":"9c997c63-9144-463d-8615-1d0086352de1","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23924,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/cb4281d40a4afac575773aca.png"},{"id":93028349,"identity":"c9718cc2-72b3-4550-9e6d-3b57af038e48","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20787,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/a0be4ac108082c32504590b9.png"},{"id":93028353,"identity":"9ef2babd-09d0-4bdb-8cd2-99a685bf679c","added_by":"auto","created_at":"2025-10-08 09:52:32","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129042,"visible":true,"origin":"","legend":"","description":"","filename":"0043d94673de4f7d975118cd85eb1dcc1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/5fcabce59b4a95ccc15a6210.xml"},{"id":93028352,"identity":"303ff5ea-71ca-4b08-8061-85013351be62","added_by":"auto","created_at":"2025-10-08 09:52:32","extension":"html","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137851,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/b87b422e6ff6eb56b40d1d42.html"},{"id":93028342,"identity":"ab118a78-cc54-47ec-ab6a-ce299676274b","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":147531,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow chart.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/e41aa23ceacdeb0dc2dec293.jpg"},{"id":93028339,"identity":"70e76a29-384e-4404-b440-c03a67aa1772","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":51405,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted cognitive function over time by diabetes status\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/e9b261604767dffa0da79d04.jpg"},{"id":93028341,"identity":"2ebd370b-1a65-4055-b2b0-a81ca1c50e62","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":64217,"visible":true,"origin":"","legend":"\u003cp\u003eMarginal effects of year by diabetes status in different age groups\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/517f22159003c661d25885b1.jpg"},{"id":96245379,"identity":"70374ca3-881b-43ae-b4b3-83b00b755cf1","added_by":"auto","created_at":"2025-11-19 07:20:31","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1085692,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/a7378efe-a68c-4cba-93a6-b53c6a7555da.pdf"},{"id":93028347,"identity":"e1796e21-6d17-4679-afb7-21c74c39bd38","added_by":"auto","created_at":"2025-10-08 09:52:31","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":152334,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7590094/v1/ac6c2ac4a79003210f669c7c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Age at Diabetes Diagnosis and Cognitive Function Among Older Adults in the China Health and Retirement Longitudinal Study, 2011–2018","fulltext":[{"header":"1 Background","content":"\u003cp\u003eCognitive function, encompassing multiple domains such as temporal-spatial orientation, executive function, and verbal comprehension, is a crucial psychological capacity for maintaining functional independence in daily living and social participation \u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. In China, approximately 38.8\u0026nbsp;million individuals are affected by Mild Cognitive Impairment (MCI), and an additional 15\u0026nbsp;million people aged 60 years and above suffer from dementia, with their prevalence showing a persistent upward trend \u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Cognitive impairment not only diminishes the functional abilities of older adults but also poses significant challenges to family caregiving and the social security systems \u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eDiabetes, one of the most prevalent chronic diseases in China, currently affects approximately 117\u0026nbsp;million individuals. Among this population, those aged 45 years and above account for a substantial 72.8%, equivalent to 85.17\u0026nbsp;million people \u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Research indicates that diabetes can accelerate cognitive decline by affecting brain structure and function through mechanisms, such as cerebrovascular disease, chronic low-grade inflammation, and insulin resistance \u003csup\u003e[\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. However, existing research predominantly focuses on the association between diabetes status (i.e., presence or absence of the disease) and cognitive function, with limited attention to the potential impact of age at initial diabetes diagnosis, a critical temporal dimension on cognitive outcomes \u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAge at diagnosis reflects both the initiation of exposure to a hyperglycemic state and the intensity of cumulative risk, while also overlapping with phases of physiological vulnerability. This confluence may influence both the onset point and trajectory of cognitive decline \u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. Moreover, physiological variations and sex differences further compound this complexity. Women exhibit distinct patterns compared to men in areas such as hormonal transitions, educational opportunities, and social role stress, potentially contributing to sex-specific patterns of cognitive deterioration \u003csup\u003e[\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. Consequently, stratified heterogeneity analysis by sex and age holds significant practical implications \u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eLeveraging nationally representative longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS) spanning 2011\u0026ndash;2018, this study aims to conduct the first systematic investigation into the association between age at diabetes diagnosis and both the level of cognitive function and its rate of change over time among middle-aged and older adults in China. Furthermore, it conducts subgroup analyses stratified by sex and age. The primary objectives are to: identify populations at high risk of cognitive decline; facilitate a paradigm shift in diabetes management from a sole focus on disease control to the preservation of cognitive health across the life course; and ultimately, provide robust epidemiological evidence to support healthy aging strategies \u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e"},{"header":"2 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Study sample\u003c/h2\u003e\u003cp\u003eData collected through in-person interviews from the China Health and Retirement Longitudinal Study (CHARLS) includes nationally representative, high-quality microdata focusing on household and individual circumstances of adults aged 45 years and older in China. It is specifically designed to analyze issues related to population aging in China and advance interdisciplinary research on aging \u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. The baseline national survey, initiated in 2011, covered 150 county-level units and 450 village-level units, including approximately 17,000 individuals from 10,000 households. Subsequently, follow-up waves have been conducted every two to three years. The resulting de-identified datasets are made publicly accessible to the academic community, typically within one year following the completion of each survey wave \u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThis study employed a longitudinal cross-sectional survey design utilizing CHARLS national follow-up data from 2011 to 2018. From an initial pool of 75,062 individuals, this study applied exclusion criteria for missing cognitive score data, missing diabetes status data, a history of brain disease, mental disorders, or malignant tumors, and age\u0026thinsp;\u0026lt;\u0026thinsp;45 years, yielding a final analytical sample of 39,110 participants (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The sample comprised longitudinal follow-ups and refreshment cohorts across waves: 9,498 baseline participants (2011); 6,201 baseline plus 3,337 new participants (2013); 5,849 baseline participants plus 3,701 new participants (2015); and 5,155 baseline participants plus 5,369 new participants (2018).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Measures\u003c/h2\u003e\u003cdiv id=\"Sec5\" class=\"Section3\"\u003e\u003ch2\u003e2.2.1 Outcome: Cognitive function\u003c/h2\u003e\u003cp\u003eThe CHARLS cognitive assessments conducted across four national survey waves (2011\u0026ndash;2018) comprehensively evaluated three domains: memory, executive function, and orientation \u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. Memory function was measured through episodic recall tasks: participants listened to 10 unrelated Chinese words, with immediate recall scored immediately after presentation and delayed recall assessed after a 4-minute interval (1 point per correct word, total score range 0\u0026ndash;20). Executive function assessment combined two components: (1) the intersecting pentagons copying test, which required accurate reproduction of two overlapping pentagons (scored 0 or 3 points), and (2) the serial-7 subtraction test, involving five consecutive subtractions starting from 100 (1 point per correct calculation, maximum 5 points), yielding a composite executive function score of 0\u0026ndash;8 points. Orientation was evaluated through four temporal questions assessing knowledge of the current year, month, date, and day of week (1 point per correct answer, total 0\u0026ndash;4 points). All cognitive tests were administered using standardized protocols with explicit scoring criteria to ensure measurement consistency across survey waves.\u003c/p\u003e\u003cp\u003eTo evaluate global cognitive function, a two-step standardization approach was implemented across all cohorts. First, domain-specific cognitive scores were calculated for each assessment wave by standardizing raw test scores relative to the CHARLS 2011 baseline distribution: individual raw scores underwent z-score transformation using the formula (Raw Score-Baseline Mean)/Baseline SD. For instance, given the baseline executive function mean of 5.4 (SD\u0026thinsp;=\u0026thinsp;2.4), standardized executive scores were derived as (Raw Executive Score \u0026minus;\u0026thinsp;5.4)/2.4. Second, these standardized domain scores (memory, executive function, orientation) were averaged to create a composite score, which was then re-standardized to the baseline distribution to generate each participant\u0026rsquo;s global cognitive function score for every survey wave, with a theoretically possible maximum of 20 points.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e2.2.2 Exposure: Age at diabetes Diagnosis\u003c/h2\u003e\u003cp\u003eDiabetes status was ascertained by asking participants, \u0026ldquo;Have you ever been diagnosed with diabetes by a physician?\u0026rdquo; Affirmative responders were then asked, \u0026ldquo;How old were you when first diagnosed with diabetes?\u0026rdquo; Rigorous quality control was applied to diagnosis age responses: implausible values (diagnosis age\u0026thinsp;\u0026lt;\u0026thinsp;0 years or exceeding current age) were excluded, and participants verified their responses upon prompting. Based on validated diagnosis ages, participants were categorized into three diagnostic age groups: 45\u0026ndash;59 years, 60\u0026ndash;74 years, and \u0026ge;\u0026thinsp;75 years \u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e. Diabetes duration was calculated as the difference between the participant\u0026rsquo;s age at Wave 1 and their diabetes diagnosis age, and it was subsequently analyzed as a continuous variable.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e\u003ch2\u003e2.2.3 Covariates\u003c/h2\u003e\u003cp\u003ePotential confounders included in this study were measured at baseline, which are known to be associated with both age at diabetes diagnosis and cognitive function, including:\u003c/p\u003e\u003cp\u003eDemographic characteristics: gender (male/female), marital status (married/ unmarried), educational attainment (illiterate/elementary or secondary school/high school/ university or college), and residence (urban/rural);\u003c/p\u003e\u003cp\u003eBehavioral factors: smoking status (yes/no) and alcohol consumption (yes/no);\u003c/p\u003e\u003cp\u003eHealth status: hypertension diagnosis (yes/no).\u003c/p\u003e\u003cp\u003eAll categorical variables were operationalized according to standardized CHARLS protocols to ensure consistent classification across survey waves.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e2.2.4 Statistical analysis\u003c/h2\u003e\u003cp\u003eCategorical variables are presented as proportions, while continuous variables are expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation. Group comparisons for continuous variables were performed using t-tests, one-way ANOVA, or nonparametric alternatives as appropriate. Longitudinal changes in test performance across Waves 1\u0026ndash;4 were analyzed using generalized linear mixed models (GLMMs) with random intercepts and random slopes. Fixed effects (β) and variance components (α) were estimated via restricted maximum likelihood (REML). Pooling functions were applied to integrate estimates and standard errors across all imputed datasets, yielding combined results with robust standard errors, 95% confidence intervals, and p-values.\u003c/p\u003e\u003cp\u003eSubgroup analyses were conducted using stratified GLMMs. Missing values in categorical covariates were explicitly assigned to a separate \u0026ldquo;missing\u0026rdquo; category. For quantitative data, missing values remained unaddressed through imputation, as GLMMs inherently accommodate missing observations under the missing at random (MAR) assumption. All univariate analyses and generalized mixed model procedures were implemented in R software (version 4.2). Two-tailed tests were applied throughout, with statistical significance defined at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;1. Baseline characteristics, overall and by age at Diabetes Diagnosis, China Health and Retirement Longitudinal Study (CHARLS), China,2011\u0026ndash;2018.(n\u0026thinsp;=\u0026thinsp;39,110)\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"99%\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eCharacteristics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003eTotal (n=39,110)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eDiabetes\u003c/p\u003e\n \u003cp\u003e(n=3,726)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eNon-diabetes\u003c/p\u003e\n \u003cp\u003e(n=35,384)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u003cem\u003eP-value\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge, years, mean (SD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e58.47 (8.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e60.98 (8.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e58.20 (8.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCognitive score, mean (SD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e12.50 (3.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e12.71 (3.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e12.48 (3.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGender (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e18764 (48.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1709 (45.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e17055 (48.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e20346 (52.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e2017 (54.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e18329 (51.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMarital Status (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e0.023\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eMarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e37065 (94.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e3561 (95.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e33504 (94.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eUnmarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e2045 (5.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e165 (4.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1880 (5.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEducation (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eIlliterate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e14591 (37.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1349 (36.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e13242 (37.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eElementary school and secondary school\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e12063 (30.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1031 (27.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e11032 (31.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eHigh school\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e11538 (29.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1199 (32.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e10339 (29.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eUniversity or college\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e918 (2.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e147 (3.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e771 (2.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResidence (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eUrban\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e16269 (41.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1971 (52.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e14298 (40.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eRural\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e22841 (58.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1755 (47.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e21086 (59.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSmoking Status (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e0.261\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e16266 (41.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1517 (40.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e14749 (41.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e22843 (58.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e2209 (59.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e20634 (58.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDrinking alcohol (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e0.019\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e17605 (45.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1609 (43.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e15996 (45.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e21505 (55.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e2117 (56.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e19388 (54.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertension (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e12018 (30.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e2186 (58.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e9832 (27.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e27092 (69.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1540 (41.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e25552 (72.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYears of follow-up,mean (SD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20px;\"\u003e\n \u003cp\u003e2.24\u0026plusmn;1.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e1.82\u0026plusmn;1.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003e2.29\u0026plusmn;1.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviation: SD, standard deviation.\u003c/p\u003e"},{"header":"3 Results","content":"\u003cp\u003eTable 1 summarizes the baseline characteristics of participants stratified by age at diabetes diagnosis in this study. A total of 39,110 participants were enrolled, comprising 18,764 males (48%) and 20,346 females (52%). The average age was 58.47\u0026plusmn;8.65 years (mean\u0026plusmn;standard deviation [SD]). Among them, 3,726 participants (9.52%) had a diagnosis of diabetes, while 35,384 (90.48%) were non-diabetic. The average cognitive score was 12.5\u0026plusmn;3.26 points. The age distribution was as follows: 24,953 participants in the 45-59 years group, 13,299 in the 60-74 years group, and 1,858 in the \u0026gt;75 years group. Regarding marital status, 37,065 participants (94.8%) were married, and 2,045 (5.2%) were unmarried. As for educational attainment, 14,591 participants (37.3%) were illiterate, 12,063 (30.8%) had completed elementary or secondary school, 11,538 (29.5%) had completed high school, and 918 (2.3%) had attended university or college. Regarding residential area, 16,269 participants (41.6%) resided in urban areas, and 22,841 (58.4%) resided in rural areas. In terms of lifestyle habits, 16,266 participants (41.6%) were smokers, and 22,843 (58.4%) were non-smokers. Additionally, 17,605 participants (45.0%) consumed alcohol, while 21,505 (55.0%) did not. In terms of health status, 12,018 participants (30.7%) had hypertension, and 27,092 (69.3%) did not. The total follow-up duration was 2.24\u0026plusmn;1.16 years, with the diabetes group having 1.82\u0026plusmn;\u0026nbsp;1.04 years and the non-diabetes group having 2.29\u0026plusmn;1.16 years. Univariate analysis showed significant differences between the diabetes and non-diabetes groups in terms of age, cognitive score, gender, marital status, education, residence, hypertension, and follow-up duration (P \u0026lt; 0.001). However, no significant differences were observed in smoking status or alcohol consumption (P \u0026gt; 0.05).\u003c/p\u003e\n\u003cp\u003eTable 2 presents the results of the Multiple linear mixed model analysis, which indicates that gender (\u0026beta; = -0.881), age (\u0026beta; = -0.096), and residence (\u0026beta; = -1.264) are risk factors for cognitive function in middle-aged and elderly individuals. Marital status (\u0026beta; = 0.059), follow-up duration (\u0026beta; = 0.035), education (\u0026beta; = 0.599), and diabetes (\u0026beta; = 0.127) are protective factors for cognitive function. Residence, gender, and education ranked as the top 3 influencing factors for cognitive function. The age at diabetes diagnosis had a significant positive impact on cognitive function scores. Compared with non-diabetic individuals, those with diabetes had higher cognitive scores, with the difference being statistically significant.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSubgroup analysis stratified by gender showed that education (\u0026beta;= 0.58) is a protective factor for cognitive function in women, while age (\u0026beta;= -0.077), smoking (\u0026beta;= -0.268), and residence (\u0026beta;= -0.834) are risk factors for cognitive function in women. For men, follow-up duration (\u0026beta;= 0.065), diabetes (\u0026beta;= 0.259), and education (\u0026beta;\u0026nbsp;= 0.648) are protective factors for cognitive function, while age (\u0026beta;= -0.113), smoking (\u0026beta;= -0.168), and residence (\u0026beta;= -1.694) are risk factors. Diabetes in men was significantly positively correlated with cognitive scores(Table S1-S2, and Figure S1).\u003c/p\u003e\n\u003cp\u003eTable 3 presents the results of the multivariable-adjusted linear mixed model analysis\u0026nbsp;examining\u0026nbsp;the association between age at diabetes diagnosis and cognitive function. Compared\u0026nbsp;with Model 1,\u0026nbsp;which only adjusted for demographic characteristics, the association weakened after\u0026nbsp;further adjustment for\u0026nbsp;socioeconomic status (Model 2).\u0026nbsp;Additional\u0026nbsp;adjustment for lifestyle habits (Model 3)\u0026nbsp;left\u0026nbsp;the association unchanged, while\u0026nbsp;further inclusion of\u0026nbsp;marital status and past medical history weakened the association. The fully adjusted model (Model 4) showed that for each 1-point increase in cognitive score, the odds ratio for developing diabetes increased by approximately 2.5%-5.0% (95% confidence interval [CI]: 1.013-1.037), with the result being highly significant across all models (p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003eTABLE2 Multiple linear mixed model analysis of the age at diabetes diagnosis and cognitive function,China Health and Retirement Longitudinal Study (CHARLS), China,2011\u0026ndash;2018.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"3\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePredictor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEstimate (\u0026beta;)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStd. Error\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI (lower)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI (upper)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u003cstrong\u003ep\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYear of follow up\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.035\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e0.026\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.044\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGender\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e-0.881\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.061\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e-0.762\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEducation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.599\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e0.550\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.647\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.127\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.055\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e0.018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e0.022\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003eHypertension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.038\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e0.566\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e-0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-0.101\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e-0.092\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSmoking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e-0.176\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.053\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-0.279\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e-0.073\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003eDrinking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.035\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-0.059\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.077\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e0.791\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMarry\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e0.059\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.056\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.168\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e0.291\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 121px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResidence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 88px;\"\u003e\n \u003cp\u003e-1.264\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.046\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 111px;\"\u003e\n \u003cp\u003e-1.353\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e-1.174\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 52px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTABLE3 Multivariable-adjusted mixed-effects linear regression results for the association between age at diabetes diagnosis and standardized composite cognitive function score, China Health and Retirement Longitudinal Study (CHARLS),China.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eModel\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eOR(cognitive_score)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eP-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eModel 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.039 ~ 1.062\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eModel 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.027\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.015 ~ 1.039\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eModel 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.027\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.015 ~ 1.039\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003eModel 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e1.013 ~ 1.037\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eModel 1: Adjusted for age and gender.\u003c/p\u003e\n\u003cp\u003eModel 2: Based on Model 1, additionally adjusted for education and residence.\u003c/p\u003e\n\u003cp\u003eModel 3: Based on Model 2, further adjusted for smoking and drinking.\u003c/p\u003e\n\u003cp\u003eModel 4: Based on Model 3, further adjusted for marital status and hypertension.\u003c/p\u003e\n\u003cp\u003eFigure 2 visualizes the longitudinal trend of cognitive function changes across different diabetes statuses over time. The results revealed that cognitive scores among individuals with diabetes were consistently higher than those among non-diabetic individuals at baseline. However, a divergent pattern emerged over the follow-up period: cognitive scores in the diabetic group exhibited an overall downward trend, while those in the non-diabetic group showed an increasing trend. Consequently, the gap in cognitive scores between the two groups narrowed from 0.23 points in 2011 to around 0.05 points in 2018. The slight annual decline in cognitive scores observed in the diabetic group suggests that diabetes may exert an adverse long-term impact on cognitive function. In contrast, the sustained improvement in cognitive scores among non-diabetic individuals implies that social interventions, health education, and related public health measures may be contributing to this positive outcome.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 3\u003c/strong\u003e depicts the time-dependent marginal effects of diabetes status on cognitive function across different age subgroups. The analysis revealed that cognitive scores in the 45-60 years age group were significantly higher than those in other age groups, with a consistently significant negative correlation observed between age and cognitive function across all subgroups. Among individuals under 60 years of age, diabetes was significantly positively associated with cognitive scores. However, in the population aged \u0026ge;75 years, the adverse impact of diabetes on cognitive function became more pronounced. These findings indicate that the sensitivity to diabetes-related cognitive effects varies across age groups, highlighting the need for targeted intervention strategies tailored to different age cohorts.\u003c/p\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eThis study leveraged nationally representative data from the CHARLS cohort (n = 39,110) to systematically analyze longitudinal cognitive trajectories across four survey waves (2011-2018). It revealed complex, age-dependent associations between the timing of diabetes diagnosis and cognitive function among middle-aged and older Chinese adults. Two pivotal patterns were identified: first, participants with diabetes demonstrated significantly higher baseline cognitive scores than their non-diabetic counterparts. Generalized linear mixed models (GLMM) indicated a modest protective effect of diabetes status on cognitive trajectories (\u0026beta; = 0.18, 95% CI 0.05-0.31). Second, a critical age-stratified divergence was observed: among individuals aged \u0026lt;60 years, diabetes diagnosis correlated positively with cognitive performance, suggesting potential compensatory health behavior adaptations, whereas in the \u0026ge; 75-year subgroup, diabetes accelerated cognitive decline by 0.42 points annually (P \u0026lt; 0.001), reflecting cumulative neurotoxic effects of prolonged hyperglycemia and metabolic dysregulation\u003csup\u003e\u0026nbsp;[21-22]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eLongitudinal analysis revealed a progressive convergence of cognitive trajectories between the two groups, with the baseline gap in cognitive scores narrowing from 0.23 points in 2011 to merely 0.05 points by 2018. Moreover, baseline cognitive function demonstrated predictive value for diabetes incidence: each 1-point increase in cognitive score was associated with a 2.5%-5.0% higher risk of developing diabetes. Collectively, these findings unveil a complex bidirectional diabetes-cognition relationship: while exhibiting baseline protective effects, diabetes simultaneously accelerates long-term cognitive deterioration. Age emerged as a critical moderator determining the direction of these associations \u003csup\u003e[23]\u003c/sup\u003e, providing pivotal epidemiological evidence for implementing precision cognitive health management in diabetes care \u003csup\u003e[24]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe observed pattern of higher baseline cognitive levels and modest protective effects among diabetic participants in our study offers novel insights into the complex diabetes-cognition relationship. Crucially, healthy survivor bias likely plays a pivotal role: diabetic individuals included in CHARLS necessarily survived to the enrollment age (\u0026ge; 45 years) and retained sufficient functional capacity to complete complex cognitive assessments. This selection mechanism resulted in a systematically favorable diabetic cohort \u003csup\u003e[25]\u003c/sup\u003e. Empirical data confirm the socioeconomic selectivity of this group, which exhibits higher educational attainment and urban residency predominance \u003csup\u003e[26]\u003c/sup\u003e,\u003csup\u003e\u0026nbsp;\u003c/sup\u003eboth established protective factors for cognitive function. Conversely, diabetic individuals with poorer cognitive function were disproportionately excluded due to premature mortality or inability to undergo testing \u003csup\u003e[27-28]\u003c/sup\u003e, creating a structural bias toward cognitively resilient survivors.\u003c/p\u003e\n\u003cp\u003eMoreover, diagnostic timing dynamics and reverse causality warrant careful consideration. Individuals with superior cognitive function typically demonstrate greater health awareness and engagement, leading to proactive participation in health examinations and glycemic screening. This pattern facilitates earlier detection of asymptomatic diabetes, a phenomenon particularly prominent in the under-60 subgroup \u003csup\u003e[29-30]\u003c/sup\u003e, thereby\u0026nbsp;creating a spurious association between cognitive advantage and diabetes diagnosis. Compounding this bias, enhanced healthcare access among urban and highly educated populations further strengthens this artificial linkage between cognitive reserve and elevated diabetes detection rates \u003csup\u003e[26]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFurthermore, medical interventions following a diabetes diagnosis may confer transient neuroprotective benefits. Glucose-lowering therapies, such as metformin, can rapidly improve cerebral energy metabolism, while post-diagnostic behavioral modifications (e.g., smoking cessation, reduced alcohol consumption) may yield cognitive gains during the early stages of the disease \u003csup\u003e[31-32]\u003c/sup\u003e. However, longitudinal data reveal that this protective effect is unsustainable: diabetic participants exhibited an annual cognitive decline of 0.04 points per year, whereas their non-diabetic counterparts showed cognitive improvement of 0.06 points per year. This divergence demonstrates that, with prolonged disease duration, the neurotoxic metabolic injury from diabetes ultimately predominates \u003csup\u003e[12]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThis study reveals pronounced age-dependent heterogeneity in the impact of diabetes on cognitive function. Crucially, our data demonstrate a protective association between diabetes and cognitive performance in individuals under 60 years, while a detrimental effect emerges in those aged \u0026ge;75 years. These diametrically opposed patterns provide a robust theoretical foundation for developing precision management protocols targeting cognitive health in diabetic populations.\u003c/p\u003e\n\u003cp\u003eIn the population under 60 years of age, protective mechanisms manifest primarily in two aspects: First, metabolic compensatory advantages: Diabetic patients in this age group exhibit greater residual \u0026beta;-cell function, resulting in relatively smaller glucose fluctuations. This mitigates acute oxidative damage to the hippocampus and prefrontal cortex induced by hyperglycemia \u003csup\u003e[33-34]\u003c/sup\u003e. Second, opportunities for health behavior intervention: Early diagnosis triggers a disease-related alertness effect, motivating patients to actively improve lifestyle behaviors, including higher smoking cessation rates, improved dietary control, and increased engagement in regular physical activity \u003csup\u003e[31,35]\u003c/sup\u003e. Notably, this behavioral compensatory effect is more pronounced in male patients, potentially attributable to more substantial post-diagnosis modifications in health behaviors among males.\u003c/p\u003e\n\u003cp\u003eIn contrast, damaging mechanisms predominate among individuals with late-onset diabetes (\u0026ge; 75 years), manifesting primarily through three interconnected pathways: cumulative microangiopathy induced by prolonged hyperglycemia - where patients with \u0026gt;15 years of diabetes duration typically exhibit thickening of the cerebral microvascular basement membrane and disruption of the blood-brain barrier \u003csup\u003e[36-37]\u003c/sup\u003e; synergistic amplification by neurodegenerative comorbidities as evidenced by the significantly elevated risk of comorbid Alzheimer\u0026rsquo;s disease pathology in this age group \u003csup\u003e[38]\u003c/sup\u003e; and systemic breakdown of social support systems, characterized by inadequate healthcare access and high rates of solitary living - particularly among rural elderly patients - which collectively accelerate cognitive-metabolic deterioration \u003csup\u003e[26]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eParticularly noteworthy are the transitional characteristics of the 60-74-year-old cohort. While this study did not report specific effect sizes for this group, empirical data reveal their cognitive trajectory occupies an intermediate position between the younger and older age groups, aligning with the damage-compensation homeostatic model. The median diabetes duration in this cohort, approximately 8.2years, coincides with the critical juncture for the escalation of microvascular complications, thereby delineating a vital window for clinical intervention \u003csup\u003e[39-40]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eUtilizing longitudinal data from four survey waves (2011-2018), this study reveals significant divergences in the progression of cognitive function between middle-aged and elderly Chinese individuals with diabetes and their non-diabetic counterparts. The analysis demonstrates sustained deterioration in cognitive scores among diabetic patients, whereas non-diabetic individuals exhibit steady cognitive improvement. This divergent trajectory underscores the role of diabetes-specific pathological mechanisms in driving relative cognitive deterioration compared to non-diabetic populations \u003csup\u003e[41-42]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThese findings call for paradigm shifts in the clinical management of diabetes. First, brief cognitive assessments should be incorporated into routine follow-up protocols for diabetics to establish dynamic monitoring frameworks. Second, stratified interventions must be implemented based on age and disease duration: intensified reinforcement of lifestyle management for patients with early diagnosis \u003csup\u003e[29]\u003c/sup\u003e; enhanced surveillance of microvascular complications for those in the transitional stage \u003csup\u003e[36]\u003c/sup\u003e; and multidisciplinary integrated care models for patients with late diagnosis \u003csup\u003e[43]\u003c/sup\u003e. Crucially, this represents a fundamental transition from isolated glycemic control toward comprehensive metabolic-neuro-behavioral management strategies, which better preserve cognitive health in diabetic populations \u003csup\u003e[44]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThis study, conducted using data from the CHARLS database, demonstrates notable methodological strengths while acknowledging several limitations requiring attention. The primary limitation is that diabetes diagnostic information was solely ascertained through self-reporting, potentially introducing recall bias - particularly regarding the accuracy of elderly patients\u0026rsquo; recall of historical diagnosis dates. Secondly, although cognitive assessments followed standardized protocols, their relative simplicity compared to comprehensive neuropsychological evaluations may limit sensitivity for detecting mild cognitive impairment \u003csup\u003e[18]\u003c/sup\u003e. Thirdly, the mean follow-up duration of 2.24 years might be insufficient to fully capture the long-term cognitive consequences of diabetes \u003csup\u003e[12]\u003c/sup\u003e. Furthermore, critical covariates such as diabetes treatment modalities (e.g., specific glucose-lowering medications) and glycemic control levels were not incorporated, despite their established relevance to cognitive outcomes \u003csup\u003e[31]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eNevertheless, rigorous statistical methods were employed to control for known confounders, supplemented by multiple sensitivity analyses to validate the robustness of the results. Future investigations should aim to validate diabetes diagnoses through medical records, extend follow-up periods, and incorporate treatment-specific metrics to further elucidate the diabetes-cognition relationship.\u003c/p\u003e\n\u003cp\u003eThese findings yield critical implications for public health practice. For patients under 60 years with new-onset diabetes, efforts should focus on enhancing cognitive reserve by integrating cognitive training programs into community-based diabetes management. The 60-74 year cohort requires intensified surveillance of microvascular complications, with particular emphasis on utilizing retinal examinations as screening tools for cognitive risk. For those aged \u0026ge;75 years, establishing multidisciplinary integrated care models that coordinate diabetes treatment with cognitive intervention is essential.\u003c/p\u003e\n\u003cp\u003eNotably, the substantial urban-rural disparities observed \u003csup\u003e[45]\u0026nbsp;\u003c/sup\u003enecessitate targeted strengthening of diabetes-related cognitive health management in rural areas. Implementation research should explore mHealth-enabled remote cognitive assessment and intervention models to overcome resource limitations in underserved communities \u003csup\u003e[43]\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCHARLS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; China Health and Retirement Longitudinal Study\u003c/p\u003e\n\u003cp\u003eGLMM \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Generalized linear mixed models \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMCI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Mild Cognitive Impairment\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eREML \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; restricted maximum likelihood\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Confidence interval\u003c/p\u003e\n\u003cp\u003eANOVA \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Analysis of Variance\u003c/p\u003e\n\u003cp\u003eSD \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;standard deviation\u003c/p\u003e\n\u003cp\u003eMAR \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;missing at random\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to thank all participants in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLinjun Xiang was responsible for extracting and cleaning the data, and for reviewing and correcting the data analysis results. He was a major contributor in writing the manuscript. Zhuqing Zhong provided professional advice on the overall study and revised the final manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the National Natural Science Foundation of China (72474230)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets produced or analyzed during the present study are available in the China Health and Retirement Longitudinal Study repository [http://charls.pku.edu.cn].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe CHARLS study was approved by the Ethical Review Committee of Peking University under approval number IRB00001052-11015. At the time of enrollment, all participants provided written informed consent, and the study protocol was conducted in accordance with the guidelines of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eTerracciano A, Luchetti M, Karakose S, et al. Meta-analyses of personality change from the preclinical to the clinical stages of dementia. Ageing Res Rev. 2025 Jul 31:102852. doi: 10.1016/j.arr.2025.102852.\u003c/li\u003e\n\u003cli\u003eJing R, Li P, Zhao K, et al. Energy-landscape analysis of brain network dynamics in a multi-center Alzheimer\u0026apos;s disease and mild cognitive impairment cohort. Biol Psychiatry. 2025 Aug 1:S0006-3223(25)01380-0. doi: 10.1016/j.biopsych.2025.07.022.\u003c/li\u003e\n\u003cli\u003eJia L, Du Y, Chu L, 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 Dec;5(12):e661-e671. doi: 10.1016/S2468-2667(20)30185-7.\u003c/li\u003e\n\u003cli\u003eYu F, Li H, Tai C, et al. Effect of family education program on cognitive impairment, anxiety, and depression in persons who have had a stroke: A randomized, controlled study. Nurs Health Sci. 2019 Mar;21(1):44-53. doi: 10.1111/nhs.12548. Epub 2018 Aug 16.\u003c/li\u003e\n\u003cli\u003eLu Y, Liu C, Yu D, et al. Prevalence of mild cognitive impairment in community-dwelling Chinese populations aged over 55 years: a meta-analysis and systematic review. BMC Geriatr. 2021 Jan 6;21(1):10. doi: 10.1186/s12877-020- 01948-3.\u003c/li\u003e\n\u003cli\u003eGlobal, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2040: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2023;402(10397):203\u0026ndash;234.https://doi.org/10.1016/S0140-6736(23) 01301-6.\u003c/li\u003e\n\u003cli\u003eLi C, Qi J, Yin P, Yu X, Sun H, Zhou M, Liang W. The burden of type 2 diabetes attributable to air pollution across China and its provinces, 1990-2021: an analysis for the Global Burden of Disease Study 2021. Lancet Reg Health West Pac. 2024 Nov 26;53:101246. doi: 10.1016/j.lanwpc.2024.101246.\u003c/li\u003e\n\u003cli\u003eZhang S, Zhang Y, Wen Z, et al. Cognitive dysfunction in diabetes: abnormal glucose metabolic regulation in the brain. Front Endocrinol (Lausanne). 2023 Jun 16;14:1192602. doi: 10.3389/fendo.2023.1192602.\u003c/li\u003e\n\u003cli\u003eLiu P, Wang ZH, Kang SS, et al. High-fat diet-induced diabetes couples to Alzheimer\u0026apos;s disease through inflammation-activated C/EBP\u0026beta;/AEP pathway. Mol Psychiatry. 2022 Aug;27(8):3396-3409. doi: 10.1038/s41380-022-01600-z. Epub 2022 May 11.\u003c/li\u003e\n\u003cli\u003eArvanitakis Z, Capuano AW, Wang HY, et al. Brain insulin signaling and cerebrovascular disease in human postmortem brain. Acta Neuropathol Commun. 2021 Apr 15;9(1):71. doi: 10.1186/s40478-021-01176-9.\u003c/li\u003e\n\u003cli\u003eBiessels GJ, Despa F. Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications. Nat Rev Endocrinol. 2018 Oct;14(10):591-604. doi: 10.1038/s41574-018-0048-7.\u003c/li\u003e\n\u003cli\u003eMayor S. Diabetes in midlife increases cognitive decline 20 years later, study finds. BMJ. 2014 Dec 1;349:g7386. doi: 10.1136/bmj.g7386.\u003c/li\u003e\n\u003cli\u003eMcGuire LC, Ford ES, Ajani UA. The impact of cognitive functioning on mortality and the development of functional disability in older adults with diabetes: the second longitudinal study on aging. BMC Geriatr. 2006 May 1;6:8. doi: 10.1186/1471-2318-6-8.\u003c/li\u003e\n\u003cli\u003eVerhagen C, Janssen J, Biessels GJ, et al. Females with type 2 diabetes are at higher risk for accelerated cognitive decline than males: CAROLINA-COGNITION study. Nutr Metab Cardiovasc Dis. 2022 Feb;32(2):355-364. doi: 10.1016/j.numecd.2021. 10. 013.\u003c/li\u003e\n\u003cli\u003ePlatzbecker AL, Gronewold J, Schramm S, et al. Sex- and age-specific effect of known type 2 diabetes mellitus on incident mild cognitive impairment five years later: Results from the population-based Heinz Nixdorf Recall study. Alzheimers Dement (Amst). 2025 Jun 11;17(2):e70130. doi: 10.1002/dad2.70130.\u003c/li\u003e\n\u003cli\u003eCochar-Soares N, de Oliveira DC, Luiz MM, et al. Sex Differences in the Trajectories of Cognitive Decline and Affected Cognitive Domains Among Older Adults With Controlled and Uncontrolled Glycemia. J Gerontol A Biol Sci Med Sci. 2024 Jul 1;79(7):glae136. doi: 10.1093/gerona/glae136.\u003c/li\u003e\n\u003cli\u003eZhou X, Qin JJ, Li H, et al. The effect of multimorbidity patterns on physical and cognitive function in diabetes patients: a longitudinal cohort of middle-aged and older adults in China. Front Aging Neurosci. 2024 May 14;16:1388656. doi: 10.3389/fnagi.2024.1388656.\u003c/li\u003e\n\u003cli\u003eZhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China health and retirement longitudinal study (CHARLS). Int J Epidemiol. 2014;43:61-68. doi:10.1093/ije/dys203.\u003c/li\u003e\n\u003cli\u003eZhao Y, Strauss J, Chen X, et al. China Health and Retirement Longitudinal Study Wave 4 User\u0026rsquo;s Guide. National School of Development, Peking University; 2020.\u003c/li\u003e\n\u003cli\u003eWang T, Zhao Z, Wang G, et al. Age-related disparities in diabetes risk attributable to modifiable risk factor profiles in Chinese adults: a nationwide, population-based, cohort study. Lancet Healthy Longev. 2021 Oct;2(10):e618-e628. doi: 10.1016/S2666-7568(21)00177-X.\u003c/li\u003e\n\u003cli\u003eSumbul-Sekerci B, Sekerci A, Pasin O, Durmus E, Yuksel-Salduz ZI. Cognition and BDNF levels in prediabetes and diabetes: A mediation analysis of a cross-sectional study. Front Endocrinol (Lausanne). 2023 Mar 2;14:1120127. doi: 10.3389/fendo.2023.1120127.\u003c/li\u003e\n\u003cli\u003eWang X, Li X, Wang W, Shi G, Wu R, Guo L, Lu C. Longitudinal Associations of Newly Diagnosed Prediabetes and Diabetes with Cognitive Function among Chinese Adults Aged 45 Years and Older. J Diabetes Res. 2022 Jul 28;2022:9458646. doi: 10.1155/2022/9458646.\u003c/li\u003e\n\u003cli\u003eSluiman AJ, McLachlan S, Forster RB, et al. Higher baseline inflammatory marker levels predict greater cognitive decline in older people with type 2 diabetes: year 10 follow-up of the Edinburgh Type 2 Diabetes Study. Diabetologia. 2022 Mar;65(3):467-476. doi: 10.1007/s00125-021-05634-w. Epub 2021 Dec 21.\u003c/li\u003e\n\u003cli\u003eBiswas R, Capuano AW, Mehta RI, et al. Review of Associations of Diabetes and Insulin Resistance With Brain Health in Three Harmonised Cohort Studies of Ageing and Dementia. Diabetes Metab Res Rev. 2025 Jan;41(1):e70032. doi: 10.1002/dmrr.70032.\u003c/li\u003e\n\u003cli\u003eLuchsinger JA, Rosin SP, Kazemi EJ, Younes N, Suratt CE, Fattaleh BN, Florez HJ, Gonzalez JS, Hollander P, Hox SH, et al. Glucose-Lowering Medications, Glycemia, and Cognitive Outcomes: The GRADE Randomized Clinical Trial. JAMA Intern Med. 2025 Jul 1;185(7):778-787. doi: 10.1001/jamainternmed.2025.1189.\u003c/li\u003e\n\u003cli\u003eSrikantha S, Manne-Goehler J, Kobayashi LC, et al. Type II diabetes and cognitive function among older adults in India and China-results from Harmonized Cognitive Assessment Protocol studies. Front Public Health. 2024 Nov 6;12:1474593. doi: 10.3389/fpubh.2024.1474593.\u003c/li\u003e\n\u003cli\u003eLiu X, Zhang J, Guan X, Cao Y. Barriers and Facilitators to Cognitive Function Interventions in Rural Diabetic Older Adults: Using the COM-B Model and Theoretical Domains Framework. J Adv Nurs. 2025 Jun 20. doi: 10.1111/jan.70030.\u003c/li\u003e\n\u003cli\u003eRaza A, Fatima E, Bin Gulzar AH, et al. Mortality patterns in patients with type 2 diabetes mellitus and late-onset Alzheimer\u0026apos;s disease in the United States: a retrospective analysis from 1999 to 2020. Neurol Sci. 2025 Aug;46(8):3619-3629. doi: 10.1007/s10072-025-08152-4.\u003c/li\u003e\n\u003cli\u003eRabinowitz Y, Ravona-Springer R, Heymann A, et al. Physical Activity Is Associated with Slower Cognitive Decline in Older Adults with Type 2 Diabetes. J Prev Alzheimers Dis. 2023;10(3):497-502. doi: 10.14283/jpad.2023.26.\u003c/li\u003e\n\u003cli\u003eKamradt M, Ose D, Krisam J, et al. Meeting the needs of multimorbid patients with Type 2 diabetes mellitus - A randomized controlled trial to assess the impact of a care management intervention aiming to improve self-care. Diabetes Res Clin Pract. 2019 Apr;150:184-193. doi: 10.1016/j.diabres.2019.03.008.\u003c/li\u003e\n\u003cli\u003eSamaras K, Makkar S, Crawford JD, et al. Metformin Use Is Associated With Slowed Cognitive Decline and Reduced Incident Dementia in Older Adults With Type 2 Diabetes: The Sydney Memory and Ageing Study. Diabetes Care. 2020 Nov;43(11):2691-2701. doi: 10.2337/dc20-0892.\u003c/li\u003e\n\u003cli\u003eKasza KA, O\u0026apos;Connor RJ, Cummings KM, Mahoney MC. Cigarette smoking and chronic disease in the United States, 2021-2023. Prev Med. 2025 Jul 26:108378. doi: 10.1016/j.ypmed.2025.108378.\u003c/li\u003e\n\u003cli\u003eHolendov\u0026aacute; B, Stokičov\u0026aacute; L, Plecit\u0026aacute;-Hlavat\u0026aacute; L. Lipid Dynamics in Pancreatic \u0026beta;-Cells: Linking Physiology to Diabetes Onset. Antioxid Redox Signal. 2024 Nov;41(13-15):865-889. doi: 10.1089/ars.2024.0724.\u003c/li\u003e\n\u003cli\u003eYang C, Zhang H, Ma Z, et al. Structural and functional alterations of the hippocampal subfields in T2DM with mild cognitive impairment and insulin resistance: A prospective study. J Diabetes. 2024 Nov;16(11):e70029. doi: 10.1111/1753-0407.70029.\u003c/li\u003e\n\u003cli\u003eWang Z, Shi J, Jiang F, Jiang K, Chen Y. Construction of a health literacy prediction model for diabetic patients: A multicenter study. Digit Health. 2025 Jan 9;11:20552076241311735. doi: 10.1177/20552076241311735.\u003c/li\u003e\n\u003cli\u003eYilmaz O, Erinc O, Gungordu AG, et al. The Relationship of the Plasma Glycated CD59 Level with Microvascular Complications in Diabetic Patients and Its Evaluation as a Predictive Marker. J Clin Med. 2025 Jun 28;14(13):4588. doi: 10.3390/jcm14134588.\u003c/li\u003e\n\u003cli\u003eCao BF, Liu K, Chen HW, et al. Impact of baseline and trajectory of the cardiometabolic indices on incident microvascular complications in patients with type 2 diabetes. Atherosclerosis. 2025 Aug;407:120407. doi: 10.1016/j.atherosclerosis.2025.120407.\u003c/li\u003e\n\u003cli\u003eLee SH, Hwang G, Lee DY, et al. Prediction of diabetic retinopathy using machine learning and its association with dementia risk in older adults with type 2 diabetes mellitus. Diabetes Res Clin Pract. 2025 Aug;226:112378. doi: 10.1016/j.diabres.2025.112378.\u003c/li\u003e\n\u003cli\u003eAldafas R, Vinogradova Y, Crabtree TSJ, Gordon J, Idris I. The legacy effect of early HbA1c control on microvascular complications and hospital admissions in type 2 diabetes: findings from a large UK study. Ther Adv Endocrinol Metab. 2025 Jun 20;16:20420188251350897. doi: 10.1177/20420188251350897.\u003c/li\u003e\n\u003cli\u003eXu F, Hu J, Li X, et al. Inhibition of platelet activation alleviates diabetes-associated cognitive dysfunction via attenuating blood-brain barrier injury. Brain Res Bull. 2025 Feb;221:111211. doi: 10.1016/j.brainresbull.2025.111211.\u003c/li\u003e\n\u003cli\u003eZhang B, Song C, Tang X, et al. Type 2 diabetes microenvironment promotes the development of Parkinson\u0026apos;s disease by activating microglial cell inflammation. Front Cell Dev Biol. 2024 Jul 10;12:1422746. doi: 10.3389/fcell.2024.1422746.\u003c/li\u003e\n\u003cli\u003eChen Y, Li Z, Chen Y, et al. Cerebellar gray matter and white matter damage among older adults with prediabetes. Diabetes Res Clin Pract. 2024 Jul;213:111731. doi: 10.1016/j.diabres.2024.111731.\u003c/li\u003e\n\u003cli\u003eBaskar V, Vignesh MA, Raman SC, et al. Development and Validation of DIANA (Diabetes Novel Subgroup Assessment tool): A web-based precision medicine tool to determine type 2 diabetes endotype membership and predict individuals at risk of microvascular disease. PLOS Digit Health. 2025 Aug 5;4(8):e0000702. doi: 10.1371/journal.pdig.0000702.\u003c/li\u003e\n\u003cli\u003eStirland LE, Choate R, Zanwar PP, et al. Multimorbidity in dementia: Current perspectives and future challenges. Alzheimers Dement. 2025 Aug;21(8):e70546. doi: 10.1002/alz.70546.\u003c/li\u003e\n\u003cli\u003eHu M, Hao X, Zhang Y, et al. Long-term exposure to particulate air pollution associated with the progression of type 2 diabetes mellitus in China: effect size and urban-rural disparities. BMC Public Health. 2025 Apr 26;25(1):1565. doi: 10.1186/s12889-025-22394-z.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Age at Diabetes Diagnosis, Cognitive Function, Longitudinal Study, China Health and Retirement Longitudinal Study (CHARLS), Age Stratification","lastPublishedDoi":"10.21203/rs.3.rs-7590094/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7590094/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eThis study examines the dynamic relationship between age at diabetes diagnosis and cognitive function in middle-aged and elderly Chinese adults using longitudinal data (2011\u0026ndash;2018) from the China Health and Retirement Longitudinal Study (CHARLS).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eA total of 39,110 participants aged\u0026thinsp;\u0026ge;\u0026thinsp;45 years were included in the analysis. Generalized linear mixed models (GLMM) were employed to assess relationships between diagnostic age groups (45\u0026ndash;59 years, 60\u0026ndash;74 years, \u0026ge;\u0026thinsp;75 years) and standardized cognitive scores.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eAt baseline, individuals with diabetes demonstrated significantly higher cognitive scores than non-diabetic individuals (12.71\u0026thinsp;\u0026plusmn;\u0026thinsp;3.25 vs. 12.48\u0026thinsp;\u0026plusmn;\u0026thinsp;3.26; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with diabetes status showing a modest protective effect (β\u0026thinsp;=\u0026thinsp;0.127; p\u0026thinsp;=\u0026thinsp;0.022). Age stratification revealed substantial heterogeneity in this association: diabetes was positively correlated with cognitive scores in participants under 60 years, while accelerated cognitive decline was observed in the \u0026ge;\u0026thinsp;75-year group. Longitudinal trajectories diverged significantly over time - diabetic patients experienced progressive cognitive decline, whereas non-diabetic individuals showed steady improvement. This trend narrowed the between-group difference in cognitive scores from 0.23 points in 2011 to 0.05 points in 2018.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eDiabetes manifests a dual impact on cognitive function in China\u0026rsquo;s elderly population, characterized by short-term protection effects but long-term impairment. Age at diagnosis critically determines the direction of this association, providing crucial evidence for the development of age-stratified strategies for cognitive health management in diabetic populations.\u003c/p\u003e","manuscriptTitle":"Age at Diabetes Diagnosis and Cognitive Function Among Older Adults in the China Health and Retirement Longitudinal Study, 2011–2018","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-08 09:52:27","doi":"10.21203/rs.3.rs-7590094/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4c0ca314-913a-4165-a7fd-6b4f109fa3ad","owner":[],"postedDate":"October 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-14T17:08:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-08 09:52:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7590094","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7590094","identity":"rs-7590094","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}