Nonlinear Association of Total Physical Activity with Cognitive Impairment in Older Chinese Adults: A Cross-Sectional Analysis of CHARLS

Nonlinear Association of Total Physical Activity with Cognitive Impairment in Older Chinese Adults: A Cross-Sectional Analysis of CHARLS | 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 Nonlinear Association of Total Physical Activity with Cognitive Impairment in Older Chinese Adults: A Cross-Sectional Analysis of CHARLS Yongheng Zhao, Gaixia Hou, Lizhen Ning, Zhuangzhuang Guo, Xuefeng Xi, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7763220/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background Cognitive impairment (CI) is a major challenge in China’s rapidly aging population. Although physical activity (PA) is generally considered beneficial, the dose–response relationship—particularly at high exposure levels—remains uncertain. Objective To characterize the association between total physical activity (TPA, MET-min/week) and CI in older Chinese adults, with emphasis on nonlinearity and subgroup heterogeneity. Methods We analyzed adults aged ≥ 60 years in the 2018 wave of the China Health and Retirement Longitudinal Study (CHARLS; n = 5,952). TPA was derived from the IPAQ-LF and examined as quartiles and as a log-transformed, z-standardized continuous variable. CI was defined as a validated composite score < 10. Associations were estimated using multivariable logistic regression with robust (HC3) SEs; dose–response was modeled with restricted cubic splines (RCS) and corroborated using a segmented (piecewise) logistic model. Prespecified covariates included sociodemographic, behavioral, and health factors. Sensitivity analyses included survey-weighted models, multiple imputation for missing covariates, alternative TPA/CI operationalizations, and FDR-controlled interaction tests. Results Relative to Q1 (< 1,732.5 MET-min/week), Q2 and Q3 showed no association with CI (OR 0.98, 95% CI 0.78–1.22; OR 1.19, 95% CI 0.96–1.49). Q4 (≥ 9,198.0 MET-min/week) was associated with higher CI odds (OR 1.41, 95% CI 1.11–1.78). Continuous log-TPA (per + 1 SD) was not associated with CI. RCS indicated a reverse J-shaped dose–response, with risk elevation confined to very high TPA; the segmented model located the inflection near the upper exposure tail (≈ 9,200 MET-min/week). Patterns were directionally stronger among women, younger-old adults, urban residents, and those with lower education or depressive symptoms, but no multiplicative interactions were significant after FDR correction. Findings were robust to weighting, multiple imputation, and alternative exposure/outcome definitions. Conclusions In this national sample of older Chinese adults, TPA exhibited a reverse J-shaped relationship with CI: low-to-moderate activity was not associated with increased risk, whereas very high TPA was linked to higher odds of impairment. Results support individualized, context-specific PA guidance that recognizes potential upper-limit risks in late life and should be verified in longitudinal and interventional studies. Cognitive impairment Physical activity Older adults Dose–response relationship China Figures Figure 1 Figure 2 Figure 3 Figure 4 1 Introduction Cognitive impairment (CI) is increasingly recognized as a critical public health challenge in China, especially amid rapid population aging. Recent national surveys estimate that approximately 15–20% of older Chinese adults are affected by mild cognitive impairment (MCI), and about 6% live with dementia, reflecting a substantial and growing burden nationwide [ 1 – 3 ]. The etiology of CI is multifactorial, with strong links to modifiable cardiometabolic risk factors such as hypertension, diabetes, obesity, and depression [ 3 – 6 ]. Although age-standardized stroke incidence has declined in recent decades, the absolute burden of cerebrovascular disease and associated cognitive sequelae continues to rise, driven by demographic transitions and expanding life expectancy [ 7 ]. According to the Global Burden of Disease (GBD) Study, China ranks among the countries with the highest absolute number of stroke- and dementia-related disability-adjusted life years (DALYs) globally [ 8 – 11 ]. These trends highlight the urgent need for scalable, low-cost, and preventive strategies to mitigate cognitive decline. Physical activity (PA) is one of the most promising and modifiable protective factors for brain health. Experimental and neuroimaging studies demonstrate that PA promotes neurogenesis, preserves brain volume, enhances cerebrovascular function, and improves cognitive performance in older adults [ 12 – 17 ]. The Lancet Commission on Dementia Prevention, Intervention, and Care identified PA as a key target for dementia prevention [ 18 ]. International guidelines, such as those issued by the WHO, recommend that older adults perform at least 150 minutes of moderate-intensity activity per week (equivalent to ~ 600 MET-min/week) to maintain overall health [ 19 , 20 ]. Nevertheless, the dose–response relationship between PA and cognitive outcomes remains uncertain. Some studies suggest that moderate PA levels yield maximal cognitive benefits, while excessively high levels may provide diminishing or even adverse effects [ 21 – 23 ]. Moreover, subgroup analyses indicate that the relationship may differ by sex, education, and comorbid status, and cultural or structural factors (e.g., rural–urban differences, education-based disparities) may further moderate these associations in middle-income settings like China [ 24 , 25 , 25 , 26 ]. Against this background, we investigated the association between total physical activity (TPA, measured in MET-minutes/week) and cognitive impairment in a nationally representative sample of older Chinese adults using data from the China Health and Retirement Longitudinal Study (CHARLS). The main analysis focused on the 2018 wave, which provided the most complete covariate data, while multiwave data (2011–2020) were used for robustness checks. We specifically examined both linear and non-linear dose–response patterns and evaluated potential effect modification across demographic and clinical subgroups. These findings aim to inform evidence-based PA recommendations to support cognitive health among older Chinese adults. 2. Methods 2.1. Study Design and Population This study adopted a cross-sectional design using data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative cohort of Chinese adults aged 45 years and above. CHARLS employed a multistage, stratified, probability-proportional-to-size (PPS) sampling method covering 28 provinces, 150 counties, and 450 communities [ 27 ]. The baseline survey was conducted in 2011, with follow-ups in 2013, 2015, 2018, and 2020. Data were collected via computer-assisted personal interviews (CAPI). For the main analysis, we selected participants aged ≥ 60 years from the 2018 wave (n = 5,952). This wave was chosen because it provided the most complete covariate information and represented the latest pre-pandemic survey, ensuring both data quality and comparability. For robustness checks, we extended the analysis to all available waves (2011–2020). In this multiwave dataset, a total of 9,774 unique participants contributed 29,636 person-wave observations. Because repeated observations are nested within individuals, the multiwave analyses accounted for within-person correlation (see Section 2.5 ). Exclusion criteria were limited to missing values in the primary exposure (total physical activity, TPA) or the outcome (total cognition score); missing covariates were handled via multiple imputation (Section 2.5 ). Cognitive status was defined as binary (CI_group): cognitively normal (score ≥ 10) vs. cognitively impaired (score < 10). A detailed flow diagram of participant selection is shown in Fig. 1 . Analyses adhered to STROBE recommendations for observational studies. 2.2. Assessment of Physical Activity Total physical activity (TPA) was assessed using the CHARLS implementation of the International Physical Activity Questionnaire long form (IPAQ-LF). The instrument captured occupational, transportation, and leisure-time activities; domestic/household activities were not elicited. Participants reported weekly time spent in walking, moderate, and vigorous activities. Standard metabolic equivalent task (MET) values were applied: 3.3 for walking, 4.0 for moderate, and 8.0 for vigorous activities [ 28 ]. Weekly MET-minutes were summed across all domains. Following IPAQ processing guidelines, bouts 180 minutes/day within an activity category set to 180 minutes/day) prior to all analyses. TPA was analyzed both categorically and continuously. • Categorical: Participants were divided into quartiles after 1%/99% winsorization used only to set cut-points: Q1 (< 1,732.5 MET-min/week), Q2 (1,732.5–4,158.0), Q3 (4,158.0–9,198.0), Q4 (≥ 9,198.0). Continuous models used the truncated (non-winsorized) values. • Continuous: TPA was log-transformed and z-standardized (per + 1 SD). In supplementary analyses, domain-specific PA (e.g., occupational vs. leisure-time) was examined to probe heterogeneity across activity types. 2.3. Assessment of Cognitive Function Cognitive function was evaluated via a composite score combining three domains: (1) episodic memory (immediate and delayed recall, 0–10), (2) orientation & attention / calculation via TICS-10 and serial subtraction (0–10), and (3) visuospatial ability via a figure-drawing task (0–1) [ 29 , 30 ]. The total cognition score thus ranged from 0 to 21. In line with prior CHARLS-based studies, cognitive impairment was defined as a score < 10 [ 30 ]. 2.4. Covariates Covariates were selected based on theoretical relevance and prior evidence, including sociodemographic factors (age, sex, education, marital status, residence) and health-related measures (number of chronic conditions, depressive symptoms assessed via CES-D10, instrumental activities of daily living, and digestive comorbidity) [ 31 ]. A directed acyclic graph (DAG) informed a minimal sufficient adjustment set; an extended set was additionally reported to assess robustness. ‘Low’ education was defined as primary school or below and ‘high’ education as middle school or above; depressive symptoms were defined as CES-D10 ≥ 10. ‘Digestive comorbidity’ was coded from self-reported physician diagnoses recorded in CHARLS (e.g., chronic gastritis, peptic ulcer, fatty liver, hepatitis). Coding details: smoking and alcohol drinking were coded as current vs. non-current; marital status as married vs. not married; residence as rural vs. urban. The number of chronic conditions was the count of self-reported physician-diagnosed conditions. IADL was analyzed as a continuous scale. All covariates were entered into multivariable models. Variance inflation factors (VIFs) were checked and < 2 in all models, indicating no evidence of problematic multicollinearity [ 32 ]. 2.5. Statistical Analysis All analyses were performed in R (version 4.5.0). For the main analysis, the 2018 dataset was treated as cross-sectional. For robustness, multiwave models (2011–2020) were estimated with wave fixed effects, and standard errors were clustered by individual to account for within-person correlation. Primary models were estimated without survey weights; weighted sensitivity analyses using CHARLS-provided weights yielded materially similar inferences. Group differences in baseline characteristics were tested using t-tests (continuous, normal), Mann–Whitney U tests (continuous, non-normal), and χ² tests (categorical). Effect sizes were reported as Cohen’s d (continuous) and Cramér’s V (categorical). To estimate associations between TPA and cognitive impairment, logistic regression models with HC3 robust standard errors were fitted [ 33 ]: Model 1: unadjusted; Model 2: adjusted for age, sex, education; Model 3: fully adjusted for all covariates (marital status, residence, chronic conditions, CES-D10, IADL, digestive comorbidity, smoking, drinking); Model 4: continuous log-transformed TPA per + 1SD, with full covariate adjustment. Dose–response relationships were modeled with restricted cubic splines (RCS) using the rms package, with knots at the 5th, 35th, 65th, and 95th percentiles of TPA [ 34 ]. The median of Q1 served as the reference. Nonlinearity was tested by comparing the spline-augmented model against its linear-term counterpart via likelihood-ratio tests. To quantify a clinically actionable threshold, we additionally fit a piecewise (segmented) logistic regression with a single breakpoint in log-TPA (R package segmented), reporting the breakpoint estimate and 95% CI as well as pre- and post-breakpoint slopes. Breakpoint initialization was explored over multiple plausible starting values; estimates were stable across initializations. Model performance was evaluated via AUC, Brier score, and calibration slope/intercept. Influence diagnostics (DFBETAs, Cook’s distance) were inspected to rule out domination by single observations. To gauge sensitivity to unmeasured confounding, E-values were computed for the main odds ratios (point estimates and lower confidence limits). Missing data were addressed using multiple imputation by chained equations (m = 20; 20 iterations; mice package) under a missing-at-random assumption. Imputation models included the exposure and outcome; predictive mean matching was used for continuous variables and logistic/multinomial models for binary/categorical variables. Estimates were combined using Rubin’s rules. Complete-case analyses were reported in parallel as a robustness check. Sensitivity analyses further varied the operationalization of the outcome (a stricter age-/education-adjusted impairment threshold) and the exposure (untruncated TPA and alternative RCS knot placements) to assess definition dependence. 2.6. Stratified and Interaction Analyses Stratified analyses were conducted by sex, age group (60–69, 70–79, ≥ 80), education (low vs. high), urban/rural residence, depressive symptoms (CES-D10 < 10 vs. ≥10), and digestive comorbidity. Interaction terms (TPA × modifier) were tested in fully adjusted logistic models. Wald χ² tests were used to assess significance. Multiple testing was controlled using the Benjamini–Hochberg false discovery rate [ 35 ]. Subgroup-specific RCS curves were also fitted using the same percentile-based knot strategy; for sparse subgroups (e.g., ≥ 80 years), the number of knots was reduced (e.g., three) to mitigate overfitting and boundary artifacts. 2.7. Ethical considerations Ethical approval for the China Health and Retirement Longitudinal Study (CHARLS) was granted by the Biomedical Ethics Review Committee of Peking University (IRB00001052-11015). All participants provided written informed consent at each wave. This study analyzed de-identified, public-use CHARLS data and involved no direct contact with human subjects; therefore, no new ethics approval was required. The study adhered to the principles of the Declaration of Helsinki and complied with the CHARLS data use agreement. 3 Results Main Analysis: Associations Between Total Physical Activity and Cognitive Impairment Baseline characteristics across quartiles of total physical activity (TPA) are presented in Table 1 . Participants with higher TPA levels were younger, more likely to be female, and less likely to reside in urban areas. They also reported fewer limitations in instrumental activities of daily living (IADL) and slightly fewer chronic conditions. Standardized mean differences (SMDs) indicated notable imbalance for age (0.48), IADL limitations (0.49), and urban residence (0.37), while other characteristics were generally well balanced (all SMDs < 0.25). Multivariable logistic regression results are summarized in Table 2 and visualized in Fig. 2 . Compared with the lowest activity quartile (Q1, < 1,732.5 MET-min/week), Q2 (1,732.5–4,158.0 MET-min/week) was not significantly associated with cognitive impairment (OR = 0.98, 95% CI: 0.78–1.22, p = 0.829). Q3 (4,158.0–9,198.0 MET-min/week) also showed no statistically significant difference (OR = 1.19, 95% CI: 0.96–1.49, p = 0.117). In contrast, participants in Q4 (≥ 9,198.0 MET-min/week) had significantly higher odds of cognitive impairment (OR = 1.41, 95% CI: 1.11–1.78, p = 0.005). When modeled as a continuous variable, log-transformed TPA per 1-SD increase was not significantly associated with cognitive impairment (OR = 1.02, 95% CI: 0.93–1.12, p = 0.609). These associations remained after adjustment for sociodemographic, lifestyle, and health-related covariates. Overall, the quartile-based analysis suggests a potential non-linear pattern, with increased odds of cognitive impairment observed only at very high levels of physical activity (Q4). This potential non-linearity was further evaluated using restricted cubic splines in the subsequent analysis. Detailed sensitivity and supplementary analyses are provided in Supplementary Tables S1–S3. Table 1 Baseline characteristics of participants by quartiles of total physical activity (CHARLS 2018, age ≥ 60 years, N = 5,952) Characteristic Overall (N = 5952) Q1 (< 1,732.5) Q2 (1,732.5–4,158.0) Q3 (4,158.0–9,198.0) Q4 (≥ 9,198.0) SMD (vs Q1) Continuous variables, mean (SD) Age, years 67.93 ± 5.97 69.35 ± 6.72 68.36 ± 6.09 67.46 ± 5.65 66.53 ± 4.99 0.477 Sleep duration, h 6.15 ± 1.88 6.07 ± 1.99 6.18 ± 1.79 6.15 ± 1.87 6.19 ± 1.92 0.062 CES-D10 score 7.86 ± 6.21 8.57 ± 6.40 7.18 ± 5.89 7.81 ± 6.22 8.26 ± 6.39 0.226 Chronic conditions (count) 0.86 ± 0.34 0.88 ± 0.33 0.88 ± 0.32 0.86 ± 0.35 0.84 ± 0.37 0.116 IADL limitations 0.42 ± 0.95 0.74 ± 1.27 0.37 ± 0.89 0.36 ± 0.88 0.25 ± 0.66 0.486 Categorical variables, n (%) Female 3440 (57.8) 667 (55.2) 1030 (55.2) 915 (56.3) 828 (66.0) 0.311 Male 2512 (42.2) 541 (44.8) 835 (44.8) 709 (43.7) 427 (34.0) 0.311 Education: Primary or below 1926 (32.4) 397 (32.9) 485 (26.0) 559 (34.4) 485 (38.6) 0.213 Education: Middle school 1774 (29.8) 378 (31.3) 560 (30.0) 473 (29.1) 363 (28.9) 0.213 Education: High school 1377 (23.1) 266 (22.0) 479 (25.7) 357 (22.0) 275 (21.9) 0.213 Education: College+ 875 (14.7) 167 (13.8) 341 (18.3) 235 (14.5) 132 (10.5) 0.213 Married 5004 (84.1) 979 (81.0) 1552 (83.2) 1349 (83.1) 1124 (89.6) 0.340 Not married 948 (15.9) 229 (19.0) 313 (16.8) 275 (16.9) 131 (10.4) 0.340 Urban residence 2718 (45.7) 554 (45.9) 1097 (58.8) 713 (43.9) 354 (28.2) 0.366 Ever smoker 3053 (51.3) 632 (52.3) 928 (49.8) 786 (48.4) 707 (56.3) 0.081 Ever drinker 3150 (52.9) 578 (47.8) 974 (52.2) 849 (52.3) 749 (59.7) 0.237 Digestive comorbidity 1884 (31.7) 377 (31.2) 576 (30.9) 494 (30.4) 437 (34.8) 0.077 Table 2 Association Between Total Physical Activity (TPA) and Cognitive Impairment (CI) Exposure OR (95% CI) p-value Quartiles Q1 (ref) 1.00 — Q2 vs Q1 0.98 (0.78–1.22) 0.829 Q3 vs Q1 1.19 (0.96–1.49) 0.117 Q4 vs Q1 1.41 (1.11–1.78) 0.005 Continuous (log TPA per + 1 SD) 1.02 (0.93–1.12) 0.609 Note: Logistic regression with HC3 robust SEs; adjusted for age, sex, education, marital status, urban/rural, smoking, drinking, sleep duration, CES-D10, chronic conditions, IADL, and digestive comorbidity. Model n = 5,952 individuals (CHARLS 2018, age ≥ 60). PA quartiles defined after 1%/99% winsorization with cutpoints: Q1 < 1,732.5; Q2 < 4,158.0; Q3 < 9,198.0; Q4 ≥ 9,198.0 MET-min/week. Reference group = Q1 Dose–Response Relationship Between Total Physical Activity and Cognitive Impairment Baseline-adjusted models mutually including TPA and water intake are presented in Table 3 . Compared with the lowest activity quartile (Q1), participants in Q2 and Q3 showed no significant differences in cognitive impairment, whereas those in Q4 (≥ 9,198 MET-min/week) exhibited significantly higher odds (OR = 1.39, 95% CI: 1.10–1.76, p = 0.006). In contrast, water intake was not significantly associated with cognitive impairment (per 1-SD increase: OR = 0.95, 95% CI: 0.88–1.03, p = 0.207), and no evidence of interaction between TPA and water intake was observed (p-interaction = 0.389). The restricted cubic spline (RCS) analysis further confirmed a non-linear dose–response relationship between TPA and cognitive impairment (Fig. 3 ). At low-to-moderate levels of TPA, the predicted probability of cognitive impairment remained comparable to the reference (Q1). However, beyond approximately 9,000 MET-min/week, the curve showed a progressive increase in risk, consistent with the quartile-based findings (p for trend = 0.002). Although confidence intervals widened at higher levels due to smaller sample sizes, the overall pattern indicated a reverse J-shaped association. Table 3 Associations of Total Physical Activity (TPA) and Water Intake with Cognitive Impairment Exposure OR (95% CI) p-value N TPA quartiles (mutually adjusted with water) Q1 (ref) 1.00 — 5947 Q2 vs Q1 0.97 (0.78–1.22) 0.817 5947 Q3 vs Q1 1.19 (0.95–1.48) 0.131 5947 Q4 vs Q1 1.39 (1.10–1.76) 0.006 5947 Water intake (continuous, per 1-SD increase) 0.95 (0.88–1.03) 0.207 5947 Trend tests Water intake (per 1-SD) — 0.207 5947 PA quartiles (ordinal 1–4) — 0.002 5947 Interaction (PA × Water, Wald test) — 0.389 5947 Logistic regression with HC3 robust SEs, mutually adjusted for TPA and water intake. Models were adjusted for age, sex, education, marital status, urban/rural residence, smoking (ever), drinking (ever), sleep duration, CES-D10 score, chronic conditions, IADL, and digestive comorbidity. PA quartiles were defined after 1%/99% winsorization with unified cutpoints: Q1 < 1,732.5; Q2 < 4,158.0; Q3 < 9,198.0 MET-min/week. Reference = Q1. Water intake modeled as standardized continuous variable (per 1-SD increase). N may differ slightly from Table 2 due to missing water intake data. Subgroup Analyses Subgroup analyses are presented in Table 4 . Overall, the associations between total physical activity (TPA) and cognitive impairment were generally consistent across strata, with no statistically significant interactions detected (all p-interaction > 0.10). However, several subgroups showed notable patterns. The association between very high TPA (Q4 vs Q1) and increased odds of cognitive impairment appeared stronger among females (OR = 1.52, 95% CI: 1.08–2.15, p = 0.016), younger participants aged 60–69 years (OR = 1.70, 95% CI: 1.23–2.34, p = 0.001), urban residents (OR = 1.89, 95% CI: 1.21–2.95, p = 0.005), and those with lower educational attainment (OR = 1.63, 95% CI: 1.21–2.21, p = 0.001). Similarly, elevated odds were observed among participants with higher depressive symptoms (CES-D10 ≥ 10; OR = 1.54, 95% CI: 1.09–2.19, p = 0.015) and those with digestive comorbidities (OR = 1.71, 95% CI: 1.14–2.57, p = 0.009). In contrast, associations were weaker and not statistically significant in males, older adults (≥ 70 years), rural residents, and those with higher education levels. Extended subgroup results are provided in Supplementary Table S4, with detailed interaction tests summarized in Supplementary Table S5. Multiple testing corrections using the Benjamini–Hochberg method are reported in Supplementary Table S6, all of which remained non-significant. Table 4 Subgroup Analyses of the Association Between Total Physical Activity (TPA) and Cognitive Impairment Modifier Stratum OR (Q4 vs Q1) 95% CI p-value OR (per 1-SD) 95% CI p-value p-interaction n Sex Male 1.35 0.96–1.88 0.083 1.00 0.89–1.13 0.982 0.854 2512 Female 1.52 1.08–2.15 0.016 1.07 0.93–1.23 0.360 3440 Age group 60–69 1.70 1.23–2.34 0.001 1.15 1.00–1.33 0.058 0.108 3901 70–79 1.18 0.77–1.80 0.450 0.91 0.79–1.05 0.199 1753 ≥ 80 1.22 0.21–7.17 0.827 1.07 0.74–1.55 0.717 298 Urban/Rural Rural 1.28 0.97–1.69 0.085 1.00 0.90–1.12 0.972 0.775 3234 Urban 1.89 1.21–2.95 0.005 1.10 0.92–1.31 0.305 2718 Education Low 1.63 1.21–2.21 0.001 1.10 0.98–1.23 0.117 0.155 1926 High 1.17 0.79–1.74 0.430 0.95 0.82–1.09 0.444 4026 CES-D10 < 10 1.33 0.96–1.82 0.083 1.01 0.89–1.15 0.906 0.862 3956 ≥ 10 1.54 1.09–2.19 0.015 1.05 0.92–1.19 0.451 1996 Digestive comorbidity No 1.29 0.97–1.73 0.082 0.99 0.88–1.10 0.820 0.197 4068 Yes 1.71 1.14–2.57 0.009 1.12 0.95–1.32 0.170 1884 Logistic regression with HC3 robust SEs, stratified by subgroup. Adjusted for age, sex, education, marital status, urban/rural residence, smoking, drinking, sleep duration, CES-D10 score, chronic conditions, IADL, and digestive comorbidity. PA quartiles defined after 1%/99% winsorization (Q1 < 1,732.5; Q2 < 4,158.0; Q3 < 9,198.0 MET-min/week). Continuous exposure modeled as log-transformed TPA per 1-SD. Interaction p values from robust Wald tests comparing models with and without interaction terms. Robustness and Methodological Checks A series of robustness analyses were conducted to evaluate the stability of the main findings (Supplementary Tables S7–S9). Sensitivity checks using a stricter cognitive impairment threshold ( ≤ − 1.5 SD) and excluding participants aged > 85 years yielded results consistent with the primary models. In both cases, participants in the highest TPA quartile remained at significantly greater odds of cognitive impairment (OR = 1.37, 95% CI: 1.06–1.77, p = 0.016; and OR = 1.40, 95% CI: 1.10–1.77, p = 0.005, respectively). Alternative specifications of physical activity—using raw, truncated, and log-transformed values—produced comparable estimates, indicating that the observed associations were not driven by scaling choices. Missing data patterns and sample flow are summarized in Supplementary Table S8. Of the 11,045 participants aged ≥ 60 years in 2018, complete data were available for 5,952 individuals, which comprised the analytic sample. Comparisons of included versus excluded participants showed substantial imbalances for several covariates, with standardized mean differences (SMDs) of 1.11 for education, 0.55 for gender, 0.55 for IADL, 0.44 for marital status, 0.43 for age, and 0.38 for urban residence; only 3 of 12 comparisons had SMD < 0.25. These patterns indicate that some selection bias cannot be ruled out, although the analytic sample still captures a broad segment of the source population. Model diagnostics are reported in Supplementary Table S9. The main regression model demonstrated good discrimination (AUC = 0.797) and adequate calibration (Brier score = 0.111; calibration intercept ≈ 0; slope ≈ 1). Multicollinearity diagnostics confirmed that all covariates had generalized variance inflation factors (GVIFs) < 2, indicating no evidence of problematic collinearity. Taken together, these robustness and methodological checks support the reliability of the main findings and strengthen confidence in the observed non-linear association between TPA and cognitive impairment. Subgroup and Interaction Analyses Subgroup analyses are presented in Supplementary Tables S4–S6. The association between TPA and cognitive impairment was generally consistent across sex, age group, education level, and urban/rural residence. For example, participants in the highest TPA quartile (Q4) exhibited increased odds of cognitive impairment compared with Q1 in both men and women, although the estimates were not statistically different between subgroups. No significant effect modification was observed (all FDR-adjusted p-values > 0.05), indicating that the reverse J-shaped association was stable across major demographic and health strata. Additional robustness analyses are summarized in Supplementary Tables S10–S12. Using a stricter cognitive impairment threshold (≤–1.5 SD; Table S10), participants in Q4 remained at significantly greater odds of impairment compared with Q1 (OR = 1.37, 95% CI: 1.06–1.77, p = 0.016). Excluding participants older than 85 years (Table S11) yielded nearly identical results (OR = 1.40, 95% CI: 1.10–1.77, p = 0.005), suggesting that the findings were not driven by the oldest age groups. When alternative handling of TPA was applied (Table S12), the overall pattern was unchanged. Using raw quartiles, Q4 was associated with higher odds of cognitive impairment (OR = 1.39, 95% CI: 1.10–1.76, p = 0.006), and with winsorized quartiles, the estimate was similar (OR = 1.41, 95% CI: 1.11–1.78, p = 0.005). In contrast, continuous specifications of TPA per 1-SD increase, whether raw (OR = 1.01, 95% CI: 0.92–1.11, p = 0.841) or log-transformed (OR = 1.02, 95% CI: 0.93–1.12, p = 0.609), were not statistically significant. Taken together, these subgroup and sensitivity analyses confirm that the elevated risk of cognitive impairment at very high levels of physical activity was consistent and robust across population strata and analytic specifications. Missing Data and Multiwave Analyses Robustness checks addressing missing data and repeated measurements are summarized in Supplementary Tables S13–S15b. Results were highly consistent across analytic approaches. First, odds ratios obtained using HC3 robust versus cluster-robust (ID) standard errors were identical (Supplementary Table S13), indicating stable inference across variance estimators. Second, multiple imputation of missing covariates (m = 20; Supplementary Table S14) yielded estimates closely aligned with the complete-case analysis: Q2 vs Q1 and Q3 vs Q1 remained non-significant, whereas Q4 vs Q1 remained significantly elevated; continuous specifications per + 1 SD of log-transformed TPA were not significant. Extending the analysis across all available CHARLS waves without BMI adjustment (2011–2020; Supplementary Table S15), per-wave models and a pooled model with wave fixed effects showed the same qualitative pattern: Q4 was consistently associated with higher odds of cognitive impairment relative to Q1, while Q2 and Q3 showed no clear differences. The pooled estimates were confirmed under both HC3 and cluster-robust (ID) variance settings, and all exported confidence intervals and p-values passed QC checks (Supplementary Table S15b). Together, these supplementary analyses suggest that the observed non-linear (reverse J-shaped) association—i.e., elevated risk confined to very high levels of TPA—is unlikely to be explained by variance specification, missing data bias, or instability across survey waves. 4 Discussion This study, based on a large, nationally representative sample of older Chinese adults from the CHARLS 2018 survey, identified a nonlinear association between total physical activity (TPA) and cognitive impairment (CI). Specifically, participants in the highest quartile of TPA (≥ 9,198 MET-min/week) demonstrated significantly increased odds of CI, while low-to-moderate activity levels were not associated with substantial differences in cognitive risk. Restricted cubic spline models confirmed a reverse J-shaped dose–response curve, whereby very high volumes of physical activity were linked to elevated risk. These findings remained robust across multiple sensitivity analyses, including alternative CI thresholds, exclusion of the oldest age groups, different parameterizations of TPA, and multiple imputation for missing data. Subgroup analyses suggested that the association was more pronounced among women, individuals with lower educational attainment, urban residents, and those with higher depressive symptoms or digestive comorbidities, although no statistically significant interactions were detected. Importantly, this pattern should be interpreted as a qualitative dose–response trend rather than a precise prescriptive threshold, given the cross-sectional design and exposure measurement constraints. The reverse J-shaped relationship observed in this study suggests that the cognitive benefits of physical activity may plateau or even reverse at very high levels of energy expenditure. The inflection point identified here (around 9,000 MET-min/week) is markedly higher than current global physical activity recommendations, which typically advocate 500–1,000 MET-min/week for general health benefits [ 36 , 37 ]. This discrepancy likely reflects population-specific differences in lifestyle and occupational demands: in China, high levels of TPA often arise from physically demanding work such as subsistence farming or manual caregiving rather than structured leisure-time exercise [ 38 – 41 ]. Such activities may lack the restorative and psychosocial benefits of recreational exercise and could instead contribute to chronic fatigue, oxidative stress, sleep disruption, and impaired recovery, all of which have been implicated in neurocognitive decline [ 42 – 45 ]. Moreover, in CHARLS the IPAQ-LF did not elicit domestic/household activities, which may undercount total exposure in older adults; accordingly, the apparent upper-tail elevation should be viewed as approximate rather than definitive. These findings are consistent with emerging evidence from other populations. For example, studies in European cohorts have reported that occupational physical activity may not provide the same protective effects against cognitive decline as leisure-time activity, and in some cases has been associated with increased dementia risk [ 46 , 47 ]. This suggests that not only the volume but also the context of physical activity is critical for cognitive health. Future longitudinal studies should therefore distinguish occupational from leisure-time domains and incorporate device-based measures to reduce exposure misclassification. Although interaction tests were not statistically significant, subgroup trends provide additional insights. The association between very high TPA and increased cognitive risk appeared stronger among women, those aged 60–69 years, and individuals with low educational attainment or elevated depressive symptoms. These observations align with the cognitive reserve hypothesis, which posits that individuals with fewer structural or psychosocial resources may be more vulnerable to adverse effects of excessive physical stress [ 48 – 50 ]. Similarly, participants with higher depressive symptom scores experienced stronger associations, suggesting that psychosocial well-being may modify the cognitive impact of physical activity. As none of the multiplicative interactions survived false-discovery-rate control, these subgroup patterns should be considered hypothesis-generating and warrant confirmation in adequately powered longitudinal analyses. The hypothesized mechanistic pathways linking total physical activity and cognitive impairment are summarized in Fig. 4 . The diagram illustrates the hypothesized reverse J-shaped relationship observed in the CHARLS 2018 cohort (≥ 60 years). Light-to-moderate levels of total physical activity (TPA) are associated with cognitive protection and healthy aging, whereas extreme TPA (≥ 9,000 MET-min/week), often occupational in nature, may contribute to fatigue, oxidative stress, sleep disruption, and impaired recovery. These physiological burdens may increase neuroinflammation and cortisol levels, thereby elevating the risk of cognitive impairment. Subgroup vulnerabilities (women, low education, depressive symptoms, urban residents) may further modify these associations. The figure is intended as a conceptual model to integrate empirical findings with plausible biological and psychosocial mechanisms, rather than as direct causal proof. The thresholds depicted are illustrative and should not be used as prescriptive cut-points. From a public health perspective, these results refine the current understanding of the physical activity–cognition relationship. While insufficient activity remains a well-established risk factor for multiple chronic conditions, our findings suggest that excessive physical activity—particularly when occupationally driven—may also carry cognitive risks. Thus, guidelines for older adults should not only emphasize minimum thresholds but also caution against extreme volumes of activity without adequate rest or psychosocial support. In rapidly aging societies such as China, interventions should promote accessible, sustainable forms of moderate activity—such as walking, tai chi, or light calisthenics—that balance physical stimulation with recovery and social engagement [ 51 , 52 ]. Consistent with this, any numeric “upper limits” inferred from cross-sectional data should be avoided in practice and replaced by individualized counseling that accounts for context (occupational vs leisure), comorbidities, and recovery. The observed associations also underscore the need for integrating physical activity promotion with broader health strategies that address mental health, comorbidities, and socioeconomic disparities. For example, community-based programs that combine exercise with psychological support and health literacy may yield synergistic benefits for cognitive aging [ 53 , 54 ]. Such precision public health approaches are aligned with global strategies, including the WHO’s Global Action Plan on Physical Activity[ 55 , 56 ]. Key strengths of this study include the use of a large, nationally representative dataset, standardized cognitive assessments, and rigorous analytic approaches including restricted cubic splines and multiple sensitivity checks. However, several limitations warrant consideration. First, the cross-sectional design precludes causal inference, and reverse causality (whereby cognitive decline reduces physical activity) cannot be excluded. Second, physical activity was captured via interval-based self-report items from the IPAQ-LF and then converted to continuous MET-min/week using standard coefficients; this procedure may systematically over- or under-estimate actual energy expenditure, especially in older adults. Prior validation studies indicate that IPAQ-LF may overestimate activity volume, particularly in agricultural or caregiving populations, which could partly explain the apparent risk elevation in the highest quartile. In addition, the CHARLS version of IPAQ-LF did not include domestic or household activities, which are common in older adults, leading to potential undercounting in some groups. Accordingly, the observed reverse J-shaped curve should be interpreted as reflecting overall dose–response trends rather than precise thresholds. Third, cognitive impairment was defined using a validated composite cut-off of < 10 (range 0–21), consistent with prior CHARLS-based validations [ 29 , 30 ], but dichotomization may mask subclinical variation and reduce sensitivity relative to clinical diagnoses. Fourth, notable differences between included and excluded participants (e.g., older age, lower education, greater IADL limitations among those excluded) raise the possibility of selection bias; multiple imputation and robustness checks mitigate but cannot eliminate this concern. Fifth, although we computed E-values for key contrasts—for example, the OR of 1.41 for Q4 vs Q1 corresponds to an E-value of 2.04 (lower CI: 1.39)—residual confounding remains possible and causal claims are unwarranted. Finally, unmeasured factors such as diet, genetics (e.g., APOE-ε4), or social engagement may still influence the observed relationship [ 57 , 58 ]. Future longitudinal studies incorporating accelerometry, domain-specific exposure (occupational vs. leisure), and refined cognitive phenotyping are needed to validate these findings and to better define safe upper bounds for physical activity in older adults. In summary, this study demonstrates a reverse J-shaped association between total physical activity and cognitive impairment in older Chinese adults, with significantly increased risk observed only at very high levels of activity. These results highlight the complexity of the physical activity–cognition relationship, suggesting that more is not always better. Public health strategies should emphasize moderation, context, and individual tailoring to optimize cognitive outcomes in aging populations. Future longitudinal and interventional research should incorporate accelerometry, distinguish activity domains, and assess recovery/sleep to establish causal pathways and inform implementable guidance for cognitive health promotion. 5 Conclusion This study, using nationally representative data from older Chinese adults, identified a reverse J-shaped association between total physical activity and cognitive impairment. Compared with low-to-moderate activity levels, which were not linked to substantially higher cognitive risk, very high activity (≥ 9,198 MET-min/week) was consistently associated with increased odds of impairment. These findings challenge the prevailing assumption of linear benefits from physical activity and suggest the existence of an optimal range that balances health benefits without inducing potential physiological or psychological strain. From a public health perspective, the results highlight the need to refine physical activity recommendations for aging populations, promoting moderate and sustainable engagement while avoiding excessive volumes often associated with occupational demands. Future longitudinal and interventional studies with objective exposure measures and refined cognitive outcomes are warranted to establish causality and guide evidence-based policy and practice in cognitive health promotion. Declarations Ethics approval and consent to participate This study is a secondary analysis of de-identified data from the China Health and Retirement Longitudinal Study (CHARLS). The original CHARLS protocols were approved by the Institutional Review Board of Peking University (household survey: IRB00001052-11015; biomarker collection: IRB00001052-11014), and written informed consent was obtained from all participants at each wave. No new data were collected for the present analysis; per the CHARLS data-use agreement and local institutional policies, no additional ethics approval was required. All procedures were performed in accordance with the Declaration of Helsinki. Consent for publication Not applicable. The manuscript contains no identifiable personal data (images, videos, or individual details). Availability of data and materials The public-use CHARLS datasets analysed in this study are available to qualified researchers upon registration and data-use approval at the CHARLS repository (http://charls.pku.edu.cn/en). All summary data supporting the findings are included in this article and its supplementary files. Additional materials (e.g., analysis code) are available from the corresponding author upon reasonable request. Competing interests The authors declare no competing interests. Funding This work was supported by the National Key R&D Program of China (Grant No. 2023YFF1104400) and the Heilongjiang Provincial Department of Education Key Research Project (Grant No. 14532D009). The funders had no role in the study design; data collection, analysis, or interpretation; writing of the report; or the decision to submit the article for publication. Authors’ contributions YZ and GH: conceptualization; methodology; formal analysis; visualization; writing—original draft. LN: data curation; investigation; validation; visualization. ZG: investigation; visualization; writing—review & editing. XX: conceptualization; supervision; project administration; writing—review & editing; correspondence; guarantor of the work. LL: supervision; project administration; writing—review & editing; correspondence. All authors contributed to interpretation of results, critically revised the manuscript for important intellectual content, and approved the final version for submission. YZ and GH contributed equally and share first authorship. Acknowledgements We gratefully acknowledge the CHARLS team and staff for providing access to the survey data and for their continued efforts in study design, data collection, and management. The interpretations and conclusions presented here are solely those of the authors and do not represent the views of CHARLS or its institutions. Authors’ information (optional) Not applicable. Clinical trial registration Clinical trial number: not applicable. References Qin F, Luo M, Xiong Y, Zhang N, Dai Y, Kuang W, et al. Prevalence and associated factors of cognitive impairment among the elderly population: A nationwide cross-sectional study in China. Front Public Health. 2022;10. https://doi.org/10.3389/fpubh.2022.1032666 . Jia L, Du Y, Chu L, Zhang Z, Li F, Lyu D, et al. Prevalence, risk factors, and management of dementia and mild cognitive impairment in adults aged 60 years or older in China: a cross-sectional study. Lancet Public Health. 2020;5:e661–71. https://doi.org/10.1016/S2468-2667(20)30185-7 . Jiang B, Liu Q, Li J-P, Lin S-N, Wan H-J, Yu Z-W, et al. Prevalence and risk factors for dementia and mild cognitive impairment among older people in Southeast China: a community-based study. BMC Geriatr. 2024;24:466. https://doi.org/10.1186/s12877-024-05054-6 . Zhao X, Xu X, Yan Y, Lipnicki DM, Pang T, Crawford JD, et al. Independent and joint associations of cardiometabolic multimorbidity and depression on cognitive function: findings from multi-regional cohorts and generalisation from community to clinic. Lancet Reg Health – Western Pac. 2024;51. https://doi.org/10.1016/j.lanwpc.2024.101198 . Wu Q, Ailshire JA, Kim JK, Crimmins EM. The Association Between Cardiometabolic Risk and Cognitive Function Among Older Americans and Chinese. J Gerontol Biol Sci Med Sci. 2024;79:glae116. https://doi.org/10.1093/gerona/glae116 . Randväli M, Toomsoo T, Šteinmiller J. The Main Risk Factors in Type 2 Diabetes for Cognitive Dysfunction, Depression, and Psychosocial Problems: A Systematic Review. Diabetology. 2024;5:40–59. https://doi.org/10.3390/diabetology5010004 . Sun T, Chen S, Wu K, Sun M, Zhang X, You C. Trends in Incidence and Mortality of Stroke in China From 1990 to 2019. Front Neurol. 2021;12. https://doi.org/10.3389/fneur.2021.759221 . Zhang C, Yang X, Wan D, Ma Q, Yin P, Zhou M, et al. Burden of neurological disorders in China and its provinces, 1990–2021: Findings from the global burden of disease study 2021. Med. 2025;6:100692. https://doi.org/10.1016/j.medj.2025.100692 . Ma Q, Li R, Wang L, Yin P, Wang Y, Yan C, et al. Temporal trend and attributable risk factors of stroke burden in China, 1990–2019: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2021;6:e897–906. https://doi.org/10.1016/S2468-2667(21)00228-0 . Li X, Jin T, Bai C, Wang X, Zhang H, Zhang T. Analysis of Stroke Burden in China from 1990 to 2021 and Projections for the Next Decade. Neuroepidemiology. 2024;1–12. https://doi.org/10.1159/000542487 . Li R, Qi J, Yang Y, Wu Y, Yin P, Zhou M, et al. Disease Burden and Attributable Risk Factors of Alzheimer’s Disease and Dementia in China from 1990 to 2019. J Prev Alzheimer’s Disease. 2022;9:306–14. https://doi.org/10.14283/jpad.2021.69 . Erickson KI, Voss MW, Prakash RS, Basak C, Szabo A, Chaddock L, et al. Exercise training increases size of hippocampus and improves memory. Proc Natl Acad Sci U S A. 2011;108:3017–22. https://doi.org/10.1073/pnas.1015950108 . Bliss ES, Wong RH, Howe PR, Mills DE. Benefits of exercise training on cerebrovascular and cognitive function in ageing. J Cereb Blood Flow Metab. 2021;41:447–70. https://doi.org/10.1177/0271678X20957807 . Silva NCBS, Barha CK, Erickson KI, Kramer AF, Liu-Ambrose T. Physical exercise, cognition, and brain health in aging. Trends Neurosci. 2024;47:402–17. https://doi.org/10.1016/j.tins.2024.04.004 . Kleinloog JPD, Mensink RP, Ivanov D, Adam JJ, Uludağ K, Joris PJ. Aerobic Exercise Training Improves Cerebral Blood Flow and Executive Function: A Randomized, Controlled Cross-Over Trial in Sedentary Older Men. Front Aging Neurosci. 2019;11. https://doi.org/10.3389/fnagi.2019.00333 . Tari AR, Walker TL, Huuha AM, Sando SB, Wisloff U. Neuroprotective mechanisms of exercise and the importance of fitness for healthy brain ageing. Lancet. 2025;405:1093–118. https://doi.org/10.1016/S0140-6736(25)00184-9 . Ahlskog JE, Geda YE, Graff-Radford NR, Petersen RC. Physical Exercise as a Preventive or Disease-Modifying Treatment of Dementia and Brain Aging. Mayo Clinic Proceedings. 2011;86:876–84. https://doi.org/10.4065/mcp.2011.0252 Dementia prevention, intervention. and care 2024. https://www.thelancet.com/commissions-do/dementia-prevention-intervention-and-care?utm_source=chatgpt.com . Accessed 29 Sept 2025. Bull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54:1451–62. https://doi.org/10.1136/bjsports-2020-102955 . Physical activity. https://www.who.int/initiatives/behealthy/physical-activity . Accessed 29 Sept 2025. Gallardo-Gómez D, del Pozo-Cruz J, Noetel M, Álvarez-Barbosa F, Alfonso-Rosa RM, del Pozo Cruz B. Optimal dose and type of exercise to improve cognitive function in older adults: A systematic review and bayesian model-based network meta-analysis of RCTs. Ageing Res Rev. 2022;76:101591. https://doi.org/10.1016/j.arr.2022.101591 . Afanador-Restrepo DF, Casanova-Correa A, Martín-Ojeda RI, Aibar-Almazán A, González-Martín AM, Hita-Contreras F, et al. Dose-response relationship of high-intensity training on global cognition in older adults with mild cognitive impairment or dementia: a systematic review with meta-analysis - the ACHIEVE-Study. Eur Rev Aging Phys Activity. 2024;21:23. https://doi.org/10.1186/s11556-024-00358-3 . Li Y, Fu C, Song H, Zhang Z, Liu T. Prolonged moderate to vigorous physical activity may lead to a decline in cognitive performance: a Mendelian randomization study. Front Aging Neurosci. 2024;16. https://doi.org/10.3389/fnagi.2024.1403464 . Chen X, Yang Q, Liu Z. The impact of physical activities on cognitive function among older adult populations: evidence from China. Front Public Health. 2025;13:1562747. https://doi.org/10.3389/fpubh.2025.1562747 . Wang X, Hu P, Ai Y, Zhou S, Li Y, Zhou P, et al. Dual group-based trajectories of physical activity and cognitive function in aged over 55: a nationally representative cohort study. Front Public Health. 2024;12. https://doi.org/10.3389/fpubh.2024.1450167 . Wu Z, Zhang H, Miao X, Li H, Pan H, Zhou D, et al. High-intensity physical activity is not associated with better cognition in the elder: evidence from the China Health and Retirement Longitudinal Study. Alzheimer’s Res Therapy. 2021;13:182. https://doi.org/10.1186/s13195-021-00923-3 . Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort Profile: The China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43:61–8. https://doi.org/10.1093/ije/dys203 . Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–95. https://doi.org/10.1249/01.MSS.0000078924.61453.FB . Guo S, Zheng X-Y. New evidence of trends in cognitive function among middle-aged and older adults in China, 2011–2018: an age-period-cohort analysis. BMC Geriatr. 2023;23:498. https://doi.org/10.1186/s12877-023-04166-9 . Zhou R, Li J, Chen M. The Association Between Cognitive Impairment and Subsequent Falls Among Older Adults: Evidence From the China Health and Retirement Longitudinal Study. Front Public Health. 2022;10. https://doi.org/10.3389/fpubh.2022.900315 . Zhang W, O’Brien N, Forrest JI, Salters KA, Patterson TL, Montaner JSG, et al. Validating a Shortened Depression Scale (10 Item CES-D) among HIV-Positive People in British Columbia, Canada. PLoS ONE. 2012;7:e40793. https://doi.org/10.1371/journal.pone.0040793 . O’brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Qual Quant. 2007;41:673–90. https://doi.org/10.1007/s11135-006-9018-6 . MacKinnon JG, White H. Some heteroskedasticity-consistent covariance matrix estimators with improved finite sample properties. J Econ. 1985;29:305–25. https://doi.org/10.1016/0304-4076(85)90158-7 . Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Cham: Springer International Publishing; 2015. https://doi.org/10.1007/978-3-319-19425-7 . Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J Royal Stat Soc Ser B (Methodological). 1995;57:289–300. RECOMMENDATIONS. In. WHO Guidelines on Physical Activity and Sedentary Behaviour. World Health Organization; 2020. O’Keefe JH, O’Keefe EL, Lavie CJ. The Goldilocks Zone for Exercise: Not Too Little, Not Too Much. Mo Med. 2018;115:98–105. Luo M, Gupta N, Holtermann A, Stamatakis E, Ding D. Revisiting the physical activity paradox in a Chinese context: Occupational physical activity and mortality in 142,302 urban working adults from the China Kadoorie Biobank study. Lancet Reg Health West Pac. 2022;23:100457. https://doi.org/10.1016/j.lanwpc.2022.100457 . Du L, Hong F, Luo P, Wang Z, Zeng Q, Guan H, et al. The relationship between occupational physical activity and dyslipidaemia in farmers with varying working modes in southwest China: the China multi-ethnic cohort study. BMC Public Health. 2022;22:840. https://doi.org/10.1186/s12889-022-13266-x . Ng SW, Norton EC, Popkin BM. Why have physical activity levels declined among Chinese adults? Findings from the 1991–2006 China health and nutrition surveys. Soc Sci Med. 2009;68:1305–14. https://doi.org/10.1016/j.socscimed.2009.01.035 . Li J, Zhang X, Zhang M, Wang L, Yin P, Li C, et al. Urban-rural differences in the association between occupational physical activity and mortality in Chinese working population: evidence from a nationwide cohort study. Lancet Reg Health – Western Pac. 2024;46. https://doi.org/10.1016/j.lanwpc.2024.101083 . Abd-Elfattah HM, Abdelazeim FH, Elshennawy S. Physical and cognitive consequences of fatigue: A review. J Adv Res. 2015;6:351–8. https://doi.org/10.1016/j.jare.2015.01.011 . Simioni C, Zauli G, Martelli AM, Vitale M, Sacchetti G, Gonelli A, et al. Oxidative stress: role of physical exercise and antioxidant nutraceuticals in adulthood and aging. Oncotarget. 2018;9:17181–98. https://doi.org/10.18632/oncotarget.24729 . Militello R, Luti S, Gamberi T, Pellegrino A, Modesti A, Modesti PA. Physical Activity and Oxidative Stress in Aging. Antioxidants. 2024;13:557. https://doi.org/10.3390/antiox13050557 . Wood E, Hall KH, Tate W. Role of mitochondria, oxidative stress and the response to antioxidants in myalgic encephalomyelitis/chronic fatigue syndrome: A possible approach to SARS-CoV-2 ‘long-haulers’? Chronic Dis Translational Med. 2021;7:14–26. https://doi.org/10.1016/j.cdtm.2020.11.002 . Zotcheva E, Bratsberg B, Strand BH, Jugessur A, Engdahl BL, Bowen C, et al. Trajectories of occupational physical activity and risk of later-life mild cognitive impairment and dementia: the HUNT4 70 + study. Lancet Reg Health – Europe. 2023;34. https://doi.org/10.1016/j.lanepe.2023.100721 . Nabe-Nielsen K, Garde AH, Gyntelberg F, Hansen ÅM, Prescott E, Laursen P, et al. Effects of Occupational and Leisure-Time Physical Activity on the Risk of Dementia: Results From the Copenhagen City Heart Study. Scand J Med Sci Sports. 2024;34:e14730. https://doi.org/10.1111/sms.14730 . Stern Y. Cognitive reserve in ageing and Alzheimer’s disease. Lancet Neurol. 2012;11:1006–12. https://doi.org/10.1016/S1474-4422(12)70191-6 . Pettigrew C, Soldan A. Defining Cognitive Reserve and Implications for Cognitive Aging. Curr Neurol Neurosci Rep. 2019;19:1. https://doi.org/10.1007/s11910-019-0917-z . Nogueira J, Gerardo B, Santana I, Simões MR, Freitas S. The Assessment of Cognitive Reserve: A Systematic Review of the Most Used Quantitative Measurement Methods of Cognitive Reserve for Aging. Front Psychol. 2022;13. https://doi.org/10.3389/fpsyg.2022.847186 . Potdevin FJ, Femery VG, Decatoire A, Bosquet L, Coello Y, Moretto P. Using effect size to quantify plantar pressure asymmetry of gait of nondisabled adults and patients with hemiparesis. J Rehabil Res Dev. 2007;44:347–54. https://doi.org/10.1682/jrrd.2006.07.0077 . American College of Sports Medicine, Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, Minson CT, Nigg CR, et al. American College of Sports Medicine position stand. Exercise and physical activity for older adults. Med Sci Sports Exerc. 2009;41:1510–30. https://doi.org/10.1249/MSS.0b013e3181a0c95c . Ngandu T, Lehtisalo J, Solomon A, Levälahti E, Ahtiluoto S, Antikainen R, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet. 2015;385:2255–63. https://doi.org/10.1016/S0140-6736(15)60461-5 . Kivipelto M, Solomon A, Ahtiluoto S, Ngandu T, Lehtisalo J, Antikainen R, et al. The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER): study design and progress. Alzheimers Dement. 2013;9:657–65. https://doi.org/10.1016/j.jalz.2012.09.012 . WHO’s work on the UN Decade of Healthy Ageing. (2021–2030). https://www.who.int/initiatives/decade-of-healthy-ageing . Accessed 29 Sept 2025. Global action plan on physical activity. 2018–2030: more active people for a healthier world. https://www.who.int/publications/i/item/9789241514187 . Accessed 29 Sept 2025. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer’s Disease in Late Onset Families. Science. 1993;261:921–3. https://doi.org/10.1126/science.8346443 . Liu C-C, Kanekiyo T, Xu H, Bu G. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol. 2013;9:106–18. https://doi.org/10.1038/nrneurol.2012.263 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryTablesS1S15.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 10 Nov, 2025 Editor invited by journal 13 Oct, 2025 Editor assigned by journal 10 Oct, 2025 Submission checks completed at journal 10 Oct, 2025 First submitted to journal 01 Oct, 2025 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-7763220","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":547169858,"identity":"5584d3a1-f2b3-491f-b6c4-83225dc7b3f3","order_by":0,"name":"Yongheng Zhao","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Yongheng","middleName":"","lastName":"Zhao","suffix":""},{"id":547169859,"identity":"1cf1e3e4-0abc-4001-a40f-fe7273e5ff53","order_by":1,"name":"Gaixia Hou","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Gaixia","middleName":"","lastName":"Hou","suffix":""},{"id":547169861,"identity":"9111ea1e-4eec-450f-8d0d-55b4f3192520","order_by":2,"name":"Lizhen Ning","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Lizhen","middleName":"","lastName":"Ning","suffix":""},{"id":547169863,"identity":"ada56a12-def7-4b7f-96ac-6af0826f7ab3","order_by":3,"name":"Zhuangzhuang Guo","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Zhuangzhuang","middleName":"","lastName":"Guo","suffix":""},{"id":547169865,"identity":"78647e46-6294-4cae-ba21-2836a95628d8","order_by":4,"name":"Xuefeng Xi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYJCCD0BsZ9/MfPDBhwoGHmJ0MM4AEskG7G3JhjPOkKCFcQPPGTNpzjYi1BvcSD7Y8HFHLbO5RIKxMeO8Ohlz9gOMHz7m4NYiOSMtsXHmmeN8ljMSEh8XbjvMY9mTwCw5cxtuLfwSOeaPeduOMTPcSDhsPHPbAR6DAwlszLx4tLBJ5H9s/tt2jLHhRmKbNO+cOh6D8w/wawHawtjM2FbDuOHMYTZp3gZmHoMbBGyR7Hlm2NjbdiBZsr2N2XDGscNALQ+b8frF4Hjyw4afbXV2/Mz8Hx98qKmzNziffPDDRzxaoOAwMoexgaB6IKgjRtEoGAWjYBSMVAAAjVJX/PSn4csAAAAASUVORK5CYII=","orcid":"","institution":"Henan University","correspondingAuthor":true,"prefix":"","firstName":"Xuefeng","middleName":"","lastName":"Xi","suffix":""},{"id":547169868,"identity":"6b2dc477-0610-4ff0-9fb4-c0408331d589","order_by":5,"name":"Limeng Liu","email":"","orcid":"","institution":"Mudanjiang Normal University","correspondingAuthor":false,"prefix":"","firstName":"Limeng","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-10-02 02:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7763220/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7763220/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96305875,"identity":"ad0b5e84-1115-4c46-a16e-b92fdfe67414","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1482339,"visible":true,"origin":"","legend":"","description":"","filename":"10.9TotalPhysicalActivityandCognitiveImpairmentinOlderChineseAdultsEvidenceforaNonlinearAssociation1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/b06e9c09ff89eea8c93b74e5.docx"},{"id":96364684,"identity":"f7f33634-460e-44f9-b265-4364c63613c6","added_by":"auto","created_at":"2025-11-20 10:09:33","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8967,"visible":true,"origin":"","legend":"","description":"","filename":"ec9d9fa2b5584940b72f002578024da7.json","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/ede7a3a2a311e227272f23ce.json"},{"id":96305874,"identity":"87b398e7-ea98-4f16-8601-0d2e1585226a","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41936,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTablesS1S15.docx","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/f5ec22b25e031828101a7376.docx"},{"id":96305884,"identity":"2402c75c-d242-4cdc-82a8-f6d7ea7d8b08","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175992,"visible":true,"origin":"","legend":"","description":"","filename":"ec9d9fa2b5584940b72f002578024da71enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/9ad35612e57bcf4347f318fc.xml"},{"id":96364014,"identity":"84a07a79-66e6-455f-8482-8754daae2d3a","added_by":"auto","created_at":"2025-11-20 10:08:46","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":774320,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/c8ba5fd52ca485c20a2ecea7.jpeg"},{"id":96305878,"identity":"f813cbc4-a161-4e82-9005-a3a59e7bb66e","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120504,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/84ca11a6838835d09078d4c6.jpeg"},{"id":96365087,"identity":"efebb4ac-9f67-40a1-b255-7a38318a255d","added_by":"auto","created_at":"2025-11-20 10:09:58","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120232,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/d868bb0112703030a1b8cf16.jpeg"},{"id":96305882,"identity":"cb9842ac-0146-4b58-aabf-232ffd7a13a4","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":341562,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/db13ae30a28594a721ba3a6c.jpeg"},{"id":96305879,"identity":"8fcd2d8c-6f02-41a3-83d7-5f7fab856f2d","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":236487,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/1253d0ea9e43189326cea9a4.png"},{"id":96365893,"identity":"f27f70f3-1224-4a57-b40c-00102e86d012","added_by":"auto","created_at":"2025-11-20 10:10:55","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32464,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/faaadf307b9d8be70b8e721e.png"},{"id":96366682,"identity":"6c905940-5d05-4b9d-8f75-c7deb439392f","added_by":"auto","created_at":"2025-11-20 10:11:45","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103702,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/74cec39f0270c1bf4b9e944e.png"},{"id":96305886,"identity":"8f9419ed-488d-4014-bfb2-48ff64cf10fc","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":122576,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/0585d4b3c7875f895bc61727.png"},{"id":96305888,"identity":"aa8104a1-4b4b-44f0-b590-bc9989c0459c","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":176881,"visible":true,"origin":"","legend":"","description":"","filename":"ec9d9fa2b5584940b72f002578024da71structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/aae94654614db62f3ba71686.xml"},{"id":96305887,"identity":"7be25309-9a7d-4528-ab7f-6203abe39fbc","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":188572,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/b8bd7dfb0922175ce27efd1e.html"},{"id":96365150,"identity":"17ff2a34-972c-4786-b1e7-552ff0869d20","added_by":"auto","created_at":"2025-11-20 10:10:03","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":774320,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of participant selection for the 2018 wave analysis in CHARLS.\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/3d6317aafea2e331beacc77d.jpeg"},{"id":96305870,"identity":"b8a05e93-d979-40e4-9b0c-e86fcfd2250d","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":120504,"visible":true,"origin":"","legend":"\u003cp\u003eAdjusted odds ratios (ORs) for cognitive impairment across total physical activity (TPA) quartiles and per 1-SD increase in log-TPA. Models adjust for age, sex, education, marital status, residence (urban/rural), smoking, drinking, sleep duration, CES-D10, chronic conditions, IADL, and digestive comorbidity. Reference: Q1 (\u0026lt;1,732.5 MET-min/week). *p \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/397c0c2a7e8c6580333c7bd0.jpeg"},{"id":96305871,"identity":"b8972d40-21c3-4c60-b628-39e6edcb1be7","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":120232,"visible":true,"origin":"","legend":"\u003cp\u003eRestricted cubic spline regression of total physical activity (TPA, MET-min/week; winsorized at the 1st and 99th percentiles) and predicted probability of cognitive impairment. The solid line shows the estimated probability, with shaded 95% CIs. ertical dashed lines mark TPA quartiles (\u0026lt;1,732.5; 1,732.5–4,158.0; 4,158.0–9,198.0; ≥9,198.0 MET-min/week). Models adjusted as in Table 2; reference = Q1.\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/e017e61dabf21d2144ce8c2b.jpeg"},{"id":96305877,"identity":"458c2ec5-edb8-46b1-b5eb-d32555caf421","added_by":"auto","created_at":"2025-11-19 15:21:03","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":341562,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConceptual mechanistic pathways linking total physical activity and cognitive impairment among older Chinese adults.\u003c/strong\u003e\u003cbr\u003e\nThe diagram illustrates the hypothesized reverse J-shaped relationship observed in the CHARLS 2018 cohort (≥60 years). Light-to-moderate levels of total physical activity (TPA) are associated with cognitive protection and healthy aging, whereas extreme TPA (≥9,000 MET-min/week), often occupational in nature, may contribute to fatigue, oxidative stress, sleep disruption, and impaired recovery. These physiological burdens may increase neuroinflammation and cortisol levels, thereby elevating the risk of cognitive impairment. Subgroup vulnerabilities (women, low education, depressive symptoms, urban residents) may further modify these associations. The figure is intended as a conceptual model to integrate empirical findings with plausible biological and psychosocial mechanisms, rather than as direct causal proof. The thresholds depicted are illustrative and should not be used as prescriptive cut-points.\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/2fea76b803fd61528350e0ee.jpeg"},{"id":96369439,"identity":"9da9c6e2-6688-4acb-849c-7beb08ea5b7e","added_by":"auto","created_at":"2025-11-20 10:20:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2578435,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/2ddea057-1bda-4cfc-a56b-688441d92607.pdf"},{"id":96365873,"identity":"65b18836-78be-4a01-b02d-a184efb4df8e","added_by":"auto","created_at":"2025-11-20 10:10:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":41936,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTablesS1S15.docx","url":"https://assets-eu.researchsquare.com/files/rs-7763220/v1/40b0da0ac6fcfc9559d314e5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Nonlinear Association of Total Physical Activity with Cognitive Impairment in Older Chinese Adults: A Cross-Sectional Analysis of CHARLS","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eCognitive impairment (CI) is increasingly recognized as a critical public health challenge in China, especially amid rapid population aging. Recent national surveys estimate that approximately 15\u0026ndash;20% of older Chinese adults are affected by mild cognitive impairment (MCI), and about 6% live with dementia, reflecting a substantial and growing burden nationwide [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The etiology of CI is multifactorial, with strong links to modifiable cardiometabolic risk factors such as hypertension, diabetes, obesity, and depression [\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Although age-standardized stroke incidence has declined in recent decades, the absolute burden of cerebrovascular disease and associated cognitive sequelae continues to rise, driven by demographic transitions and expanding life expectancy [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. According to the Global Burden of Disease (GBD) Study, China ranks among the countries with the highest absolute number of stroke- and dementia-related disability-adjusted life years (DALYs) globally [\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. These trends highlight the urgent need for scalable, low-cost, and preventive strategies to mitigate cognitive decline.\u003c/p\u003e\u003cp\u003ePhysical activity (PA) is one of the most promising and modifiable protective factors for brain health. Experimental and neuroimaging studies demonstrate that PA promotes neurogenesis, preserves brain volume, enhances cerebrovascular function, and improves cognitive performance in older adults [\u003cspan additionalcitationids=\"CR13 CR14 CR15 CR16\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The Lancet Commission on Dementia Prevention, Intervention, and Care identified PA as a key target for dementia prevention [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. International guidelines, such as those issued by the WHO, recommend that older adults perform at least 150 minutes of moderate-intensity activity per week (equivalent to ~\u0026thinsp;600 MET-min/week) to maintain overall health [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNevertheless, the dose\u0026ndash;response relationship between PA and cognitive outcomes remains uncertain. Some studies suggest that moderate PA levels yield maximal cognitive benefits, while excessively high levels may provide diminishing or even adverse effects [\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Moreover, subgroup analyses indicate that the relationship may differ by sex, education, and comorbid status, and cultural or structural factors (e.g., rural\u0026ndash;urban differences, education-based disparities) may further moderate these associations in middle-income settings like China [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAgainst this background, we investigated the association between total physical activity (TPA, measured in MET-minutes/week) and cognitive impairment in a nationally representative sample of older Chinese adults using data from the China Health and Retirement Longitudinal Study (CHARLS). The main analysis focused on the 2018 wave, which provided the most complete covariate data, while multiwave data (2011\u0026ndash;2020) were used for robustness checks. We specifically examined both linear and non-linear dose\u0026ndash;response patterns and evaluated potential effect modification across demographic and clinical subgroups. These findings aim to inform evidence-based PA recommendations to support cognitive health among older Chinese adults.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Study Design and Population\u003c/h2\u003e\u003cp\u003eThis study adopted a cross-sectional design using data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative cohort of Chinese adults aged 45 years and above. CHARLS employed a multistage, stratified, probability-proportional-to-size (PPS) sampling method covering 28 provinces, 150 counties, and 450 communities [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The baseline survey was conducted in 2011, with follow-ups in 2013, 2015, 2018, and 2020. Data were collected via computer-assisted personal interviews (CAPI).\u003c/p\u003e\u003cp\u003eFor the main analysis, we selected participants aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years from the 2018 wave (n\u0026thinsp;=\u0026thinsp;5,952). This wave was chosen because it provided the most complete covariate information and represented the latest pre-pandemic survey, ensuring both data quality and comparability.\u003c/p\u003e\u003cp\u003eFor robustness checks, we extended the analysis to all available waves (2011\u0026ndash;2020). In this multiwave dataset, a total of 9,774 unique participants contributed 29,636 person-wave observations. Because repeated observations are nested within individuals, the multiwave analyses accounted for within-person correlation (see Section \u003cspan refid=\"Sec7\" class=\"InternalRef\"\u003e2.5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eExclusion criteria were limited to missing values in the primary exposure (total physical activity, TPA) or the outcome (total cognition score); missing covariates were handled via multiple imputation (Section \u003cspan refid=\"Sec7\" class=\"InternalRef\"\u003e2.5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCognitive status was defined as binary (CI_group): cognitively normal (score\u0026thinsp;\u0026ge;\u0026thinsp;10) vs. cognitively impaired (score\u0026thinsp;\u0026lt;\u0026thinsp;10). A detailed flow diagram of participant selection is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Analyses adhered to STROBE recommendations for observational studies.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Assessment of Physical Activity\u003c/h2\u003e\u003cp\u003eTotal physical activity (TPA) was assessed using the CHARLS implementation of the International Physical Activity Questionnaire long form (IPAQ-LF). The instrument captured occupational, transportation, and leisure-time activities; domestic/household activities were not elicited. Participants reported weekly time spent in walking, moderate, and vigorous activities. Standard metabolic equivalent task (MET) values were applied: 3.3 for walking, 4.0 for moderate, and 8.0 for vigorous activities [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Weekly MET-minutes were summed across all domains.\u003c/p\u003e\u003cp\u003eFollowing IPAQ processing guidelines, bouts\u0026thinsp;\u0026lt;\u0026thinsp;10 minutes were recoded to zero and extreme durations were truncated (e.g., \u0026gt;\u0026thinsp;180 minutes/day within an activity category set to 180 minutes/day) prior to all analyses.\u003c/p\u003e\u003cp\u003eTPA was analyzed both categorically and continuously.\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u0026bull;\u0026emsp;Categorical: Participants were divided into quartiles after 1%/99% winsorization used only to set cut-points: Q1 (\u0026lt;\u0026thinsp;1,732.5 MET-min/week), Q2 (1,732.5\u0026ndash;4,158.0), Q3 (4,158.0\u0026ndash;9,198.0), Q4 (\u0026ge;\u0026thinsp;9,198.0). Continuous models used the truncated (non-winsorized) values.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u0026bull;\u0026emsp;Continuous: TPA was log-transformed and z-standardized (per +\u0026thinsp;1 SD).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eIn supplementary analyses, domain-specific PA (e.g., occupational vs. leisure-time) was examined to probe heterogeneity across activity types.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Assessment of Cognitive Function\u003c/h2\u003e\u003cp\u003eCognitive function was evaluated via a composite score combining three domains: (1) episodic memory (immediate and delayed recall, 0\u0026ndash;10), (2) orientation \u0026amp; attention / calculation via TICS-10 and serial subtraction (0\u0026ndash;10), and (3) visuospatial ability via a figure-drawing task (0\u0026ndash;1) [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. The total cognition score thus ranged from 0 to 21. In line with prior CHARLS-based studies, cognitive impairment was defined as a score\u0026thinsp;\u0026lt;\u0026thinsp;10 [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Covariates\u003c/h2\u003e\u003cp\u003eCovariates were selected based on theoretical relevance and prior evidence, including sociodemographic factors (age, sex, education, marital status, residence) and health-related measures (number of chronic conditions, depressive symptoms assessed via CES-D10, instrumental activities of daily living, and digestive comorbidity) [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. A directed acyclic graph (DAG) informed a minimal sufficient adjustment set; an extended set was additionally reported to assess robustness. \u0026lsquo;Low\u0026rsquo; education was defined as primary school or below and \u0026lsquo;high\u0026rsquo; education as middle school or above; depressive symptoms were defined as CES-D10\u0026thinsp;\u0026ge;\u0026thinsp;10. \u0026lsquo;Digestive comorbidity\u0026rsquo; was coded from self-reported physician diagnoses recorded in CHARLS (e.g., chronic gastritis, peptic ulcer, fatty liver, hepatitis). Coding details: smoking and alcohol drinking were coded as current vs. non-current; marital status as married vs. not married; residence as rural vs. urban. The number of chronic conditions was the count of self-reported physician-diagnosed conditions. IADL was analyzed as a continuous scale. All covariates were entered into multivariable models. Variance inflation factors (VIFs) were checked and \u0026lt;\u0026thinsp;2 in all models, indicating no evidence of problematic multicollinearity [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5. Statistical Analysis\u003c/h2\u003e\u003cp\u003eAll analyses were performed in R (version 4.5.0). For the main analysis, the 2018 dataset was treated as cross-sectional. For robustness, multiwave models (2011\u0026ndash;2020) were estimated with wave fixed effects, and standard errors were clustered by individual to account for within-person correlation. Primary models were estimated without survey weights; weighted sensitivity analyses using CHARLS-provided weights yielded materially similar inferences.\u003c/p\u003e\u003cp\u003eGroup differences in baseline characteristics were tested using t-tests (continuous, normal), Mann\u0026ndash;Whitney U tests (continuous, non-normal), and χ\u0026sup2; tests (categorical). Effect sizes were reported as Cohen\u0026rsquo;s d (continuous) and Cram\u0026eacute;r\u0026rsquo;s V (categorical).\u003c/p\u003e\u003cp\u003eTo estimate associations between TPA and cognitive impairment, logistic regression models with HC3 robust standard errors were fitted [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eModel 1: unadjusted;\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eModel 2: adjusted for age, sex, education;\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eModel 3: fully adjusted for all covariates (marital status, residence, chronic conditions, CES-D10, IADL, digestive comorbidity, smoking, drinking);\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eModel 4: continuous log-transformed TPA per +\u0026thinsp;1SD, with full covariate adjustment.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eDose\u0026ndash;response relationships were modeled with restricted cubic splines (RCS) using the rms package, with knots at the 5th, 35th, 65th, and 95th percentiles of TPA [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. The median of Q1 served as the reference. Nonlinearity was tested by comparing the spline-augmented model against its linear-term counterpart via likelihood-ratio tests.\u003c/p\u003e\u003cp\u003eTo quantify a clinically actionable threshold, we additionally fit a piecewise (segmented) logistic regression with a single breakpoint in log-TPA (R package segmented), reporting the breakpoint estimate and 95% CI as well as pre- and post-breakpoint slopes. Breakpoint initialization was explored over multiple plausible starting values; estimates were stable across initializations.\u003c/p\u003e\u003cp\u003eModel performance was evaluated via AUC, Brier score, and calibration slope/intercept. Influence diagnostics (DFBETAs, Cook\u0026rsquo;s distance) were inspected to rule out domination by single observations. To gauge sensitivity to unmeasured confounding, E-values were computed for the main odds ratios (point estimates and lower confidence limits).\u003c/p\u003e\u003cp\u003eMissing data were addressed using multiple imputation by chained equations (m\u0026thinsp;=\u0026thinsp;20; 20 iterations; mice package) under a missing-at-random assumption. Imputation models included the exposure and outcome; predictive mean matching was used for continuous variables and logistic/multinomial models for binary/categorical variables. Estimates were combined using Rubin\u0026rsquo;s rules. Complete-case analyses were reported in parallel as a robustness check.\u003c/p\u003e\u003cp\u003eSensitivity analyses further varied the operationalization of the outcome (a stricter age-/education-adjusted impairment threshold) and the exposure (untruncated TPA and alternative RCS knot placements) to assess definition dependence.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6. Stratified and Interaction Analyses\u003c/h2\u003e\u003cp\u003eStratified analyses were conducted by sex, age group (60\u0026ndash;69, 70\u0026ndash;79, \u0026ge;\u0026thinsp;80), education (low vs. high), urban/rural residence, depressive symptoms (CES-D10\u0026thinsp;\u0026lt;\u0026thinsp;10 vs. \u0026ge;10), and digestive comorbidity.\u003c/p\u003e\u003cp\u003eInteraction terms (TPA \u0026times; modifier) were tested in fully adjusted logistic models. Wald χ\u0026sup2; tests were used to assess significance. Multiple testing was controlled using the Benjamini\u0026ndash;Hochberg false discovery rate [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Subgroup-specific RCS curves were also fitted using the same percentile-based knot strategy; for sparse subgroups (e.g., \u0026ge;\u0026thinsp;80 years), the number of knots was reduced (e.g., three) to mitigate overfitting and boundary artifacts.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7. Ethical considerations\u003c/h2\u003e\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003cp\u003e for the China Health and Retirement Longitudinal Study (CHARLS) was granted by the Biomedical Ethics Review Committee of Peking University (IRB00001052-11015). All participants provided written informed consent at each wave. This study analyzed de-identified, public-use CHARLS data and involved no direct contact with human subjects; therefore, no new ethics approval was required. The study adhered to the principles of the Declaration of Helsinki and complied with the CHARLS data use agreement.\u003c/p\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cp\u003eMain Analysis: Associations Between Total Physical Activity and Cognitive Impairment\u003c/p\u003e\u003cp\u003eBaseline characteristics across quartiles of total physical activity (TPA) are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Participants with higher TPA levels were younger, more likely to be female, and less likely to reside in urban areas. They also reported fewer limitations in instrumental activities of daily living (IADL) and slightly fewer chronic conditions. Standardized mean differences (SMDs) indicated notable imbalance for age (0.48), IADL limitations (0.49), and urban residence (0.37), while other characteristics were generally well balanced (all SMDs\u0026thinsp;\u0026lt;\u0026thinsp;0.25).\u003c/p\u003e\u003cp\u003eMultivariable logistic regression results are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Compared with the lowest activity quartile (Q1, \u0026lt;\u0026thinsp;1,732.5 MET-min/week), Q2 (1,732.5\u0026ndash;4,158.0 MET-min/week) was not significantly associated with cognitive impairment (OR\u0026thinsp;=\u0026thinsp;0.98, 95% CI: 0.78\u0026ndash;1.22, p\u0026thinsp;=\u0026thinsp;0.829). Q3 (4,158.0\u0026ndash;9,198.0 MET-min/week) also showed no statistically significant difference (OR\u0026thinsp;=\u0026thinsp;1.19, 95% CI: 0.96\u0026ndash;1.49, p\u0026thinsp;=\u0026thinsp;0.117). In contrast, participants in Q4 (\u0026ge;\u0026thinsp;9,198.0 MET-min/week) had significantly higher odds of cognitive impairment (OR\u0026thinsp;=\u0026thinsp;1.41, 95% CI: 1.11\u0026ndash;1.78, p\u0026thinsp;=\u0026thinsp;0.005). When modeled as a continuous variable, log-transformed TPA per 1-SD increase was not significantly associated with cognitive impairment (OR\u0026thinsp;=\u0026thinsp;1.02, 95% CI: 0.93\u0026ndash;1.12, p\u0026thinsp;=\u0026thinsp;0.609).\u003c/p\u003e\u003cp\u003eThese associations remained after adjustment for sociodemographic, lifestyle, and health-related covariates. Overall, the quartile-based analysis suggests a potential non-linear pattern, with increased odds of cognitive impairment observed only at very high levels of physical activity (Q4). This potential non-linearity was further evaluated using restricted cubic splines in the subsequent analysis. Detailed sensitivity and supplementary analyses are provided in Supplementary Tables S1\u0026ndash;S3.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBaseline characteristics of participants by quartiles of total physical activity (CHARLS 2018, age\u0026thinsp;\u0026ge;\u0026thinsp;60 years, N\u0026thinsp;=\u0026thinsp;5,952)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCharacteristic\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOverall (N\u0026thinsp;=\u0026thinsp;5952)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eQ1 (\u0026lt;\u0026thinsp;1,732.5)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eQ2 (1,732.5\u0026ndash;4,158.0)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eQ3 (4,158.0\u0026ndash;9,198.0)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eQ4 (\u0026ge;\u0026thinsp;9,198.0)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSMD (vs Q1)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eContinuous variables, mean (SD)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge, years\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e67.93\u0026thinsp;\u0026plusmn;\u0026thinsp;5.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e69.35\u0026thinsp;\u0026plusmn;\u0026thinsp;6.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e68.36\u0026thinsp;\u0026plusmn;\u0026thinsp;6.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e67.46\u0026thinsp;\u0026plusmn;\u0026thinsp;5.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e66.53\u0026thinsp;\u0026plusmn;\u0026thinsp;4.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.477\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSleep duration, h\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6.15\u0026thinsp;\u0026plusmn;\u0026thinsp;1.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.07\u0026thinsp;\u0026plusmn;\u0026thinsp;1.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.18\u0026thinsp;\u0026plusmn;\u0026thinsp;1.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6.15\u0026thinsp;\u0026plusmn;\u0026thinsp;1.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6.19\u0026thinsp;\u0026plusmn;\u0026thinsp;1.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.062\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCES-D10 score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.86\u0026thinsp;\u0026plusmn;\u0026thinsp;6.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.57\u0026thinsp;\u0026plusmn;\u0026thinsp;6.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.18\u0026thinsp;\u0026plusmn;\u0026thinsp;5.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7.81\u0026thinsp;\u0026plusmn;\u0026thinsp;6.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8.26\u0026thinsp;\u0026plusmn;\u0026thinsp;6.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.226\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChronic conditions (count)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.86\u0026thinsp;\u0026plusmn;\u0026thinsp;0.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.86\u0026thinsp;\u0026plusmn;\u0026thinsp;0.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.84\u0026thinsp;\u0026plusmn;\u0026thinsp;0.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.116\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIADL limitations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.42\u0026thinsp;\u0026plusmn;\u0026thinsp;0.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.74\u0026thinsp;\u0026plusmn;\u0026thinsp;1.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.37\u0026thinsp;\u0026plusmn;\u0026thinsp;0.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.36\u0026thinsp;\u0026plusmn;\u0026thinsp;0.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.25\u0026thinsp;\u0026plusmn;\u0026thinsp;0.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.486\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eCategorical variables, n (%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3440 (57.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e667 (55.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1030 (55.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e915 (56.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e828 (66.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.311\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2512 (42.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e541 (44.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e835 (44.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e709 (43.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e427 (34.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.311\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation: Primary or below\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1926 (32.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e397 (32.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e485 (26.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e559 (34.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e485 (38.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.213\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation: Middle school\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1774 (29.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e378 (31.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e560 (30.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e473 (29.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e363 (28.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.213\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation: High school\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1377 (23.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e266 (22.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e479 (25.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e357 (22.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e275 (21.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.213\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation: College+\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e875 (14.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e167 (13.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e341 (18.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e235 (14.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e132 (10.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.213\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMarried\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5004 (84.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e979 (81.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1552 (83.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1349 (83.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1124 (89.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.340\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNot married\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e948 (15.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e229 (19.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e313 (16.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e275 (16.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e131 (10.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.340\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrban residence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2718 (45.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e554 (45.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1097 (58.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e713 (43.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e354 (28.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.366\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEver smoker\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3053 (51.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e632 (52.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e928 (49.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e786 (48.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e707 (56.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.081\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEver drinker\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3150 (52.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e578 (47.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e974 (52.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e849 (52.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e749 (59.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.237\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDigestive comorbidity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1884 (31.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e377 (31.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e576 (30.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e494 (30.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e437 (34.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.077\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eAssociation Between Total Physical Activity (TPA) and Cognitive Impairment (CI)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExposure\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQuartiles\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ1 (ref)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026mdash;\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ2 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.98 (0.78\u0026ndash;1.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.829\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ3 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.19 (0.96\u0026ndash;1.49)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.117\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ4 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.41 (1.11\u0026ndash;1.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.005\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eContinuous (log TPA per +\u0026thinsp;1 SD)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.02 (0.93\u0026ndash;1.12)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.609\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003eNote: Logistic regression with HC3 robust SEs; adjusted for age, sex, education, marital status, urban/rural, smoking, drinking, sleep duration, CES-D10, chronic conditions, IADL, and digestive comorbidity. Model n\u0026thinsp;=\u0026thinsp;5,952 individuals (CHARLS 2018, age\u0026thinsp;\u0026ge;\u0026thinsp;60). PA quartiles defined after 1%/99% winsorization with cutpoints: Q1\u0026thinsp;\u0026lt;\u0026thinsp;1,732.5; Q2\u0026thinsp;\u0026lt;\u0026thinsp;4,158.0; Q3\u0026thinsp;\u0026lt;\u0026thinsp;9,198.0; Q4\u0026thinsp;\u0026ge;\u0026thinsp;9,198.0 MET-min/week. Reference group\u0026thinsp;=\u0026thinsp;Q1\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eDose\u0026ndash;Response Relationship Between Total Physical Activity and Cognitive Impairment\u003c/p\u003e\u003cp\u003eBaseline-adjusted models mutually including TPA and water intake are presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Compared with the lowest activity quartile (Q1), participants in Q2 and Q3 showed no significant differences in cognitive impairment, whereas those in Q4 (\u0026ge;\u0026thinsp;9,198 MET-min/week) exhibited significantly higher odds (OR\u0026thinsp;=\u0026thinsp;1.39, 95% CI: 1.10\u0026ndash;1.76, p\u0026thinsp;=\u0026thinsp;0.006). In contrast, water intake was not significantly associated with cognitive impairment (per 1-SD increase: OR\u0026thinsp;=\u0026thinsp;0.95, 95% CI: 0.88\u0026ndash;1.03, p\u0026thinsp;=\u0026thinsp;0.207), and no evidence of interaction between TPA and water intake was observed (p-interaction\u0026thinsp;=\u0026thinsp;0.389).\u003c/p\u003e\u003cp\u003eThe restricted cubic spline (RCS) analysis further confirmed a non-linear dose\u0026ndash;response relationship between TPA and cognitive impairment (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). At low-to-moderate levels of TPA, the predicted probability of cognitive impairment remained comparable to the reference (Q1). However, beyond approximately 9,000 MET-min/week, the curve showed a progressive increase in risk, consistent with the quartile-based findings (p for trend\u0026thinsp;=\u0026thinsp;0.002). Although confidence intervals widened at higher levels due to smaller sample sizes, the overall pattern indicated a reverse J-shaped association.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eAssociations of Total Physical Activity (TPA) and Water Intake with Cognitive Impairment\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExposure\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTPA quartiles (mutually adjusted with water)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ1 (ref)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026mdash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ2 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.97 (0.78\u0026ndash;1.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.817\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ3 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.19 (0.95\u0026ndash;1.48)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.131\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQ4 vs Q1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.39 (1.10\u0026ndash;1.76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.006\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eWater intake (continuous, per 1-SD increase)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.95 (0.88\u0026ndash;1.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.207\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTrend tests\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWater intake (per 1-SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026mdash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.207\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePA quartiles (ordinal 1\u0026ndash;4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026mdash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eInteraction (PA \u0026times; Water, Wald test)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026mdash;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.389\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5947\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eLogistic regression with HC3 robust SEs, mutually adjusted for TPA and water intake. Models were adjusted for age, sex, education, marital status, urban/rural residence, smoking (ever), drinking (ever), sleep duration, CES-D10 score, chronic conditions, IADL, and digestive comorbidity. PA quartiles were defined after 1%/99% winsorization with unified cutpoints: Q1\u0026thinsp;\u0026lt;\u0026thinsp;1,732.5; Q2\u0026thinsp;\u0026lt;\u0026thinsp;4,158.0; Q3\u0026thinsp;\u0026lt;\u0026thinsp;9,198.0 MET-min/week. Reference\u0026thinsp;=\u0026thinsp;Q1. Water intake modeled as standardized continuous variable (per 1-SD increase). N may differ slightly from Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e due to missing water intake data.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSubgroup Analyses\u003c/p\u003e\u003cp\u003eSubgroup analyses are presented in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Overall, the associations between total physical activity (TPA) and cognitive impairment were generally consistent across strata, with no statistically significant interactions detected (all p-interaction\u0026thinsp;\u0026gt;\u0026thinsp;0.10). However, several subgroups showed notable patterns. The association between very high TPA (Q4 vs Q1) and increased odds of cognitive impairment appeared stronger among females (OR\u0026thinsp;=\u0026thinsp;1.52, 95% CI: 1.08\u0026ndash;2.15, p\u0026thinsp;=\u0026thinsp;0.016), younger participants aged 60\u0026ndash;69 years (OR\u0026thinsp;=\u0026thinsp;1.70, 95% CI: 1.23\u0026ndash;2.34, p\u0026thinsp;=\u0026thinsp;0.001), urban residents (OR\u0026thinsp;=\u0026thinsp;1.89, 95% CI: 1.21\u0026ndash;2.95, p\u0026thinsp;=\u0026thinsp;0.005), and those with lower educational attainment (OR\u0026thinsp;=\u0026thinsp;1.63, 95% CI: 1.21\u0026ndash;2.21, p\u0026thinsp;=\u0026thinsp;0.001). Similarly, elevated odds were observed among participants with higher depressive symptoms (CES-D10\u0026thinsp;\u0026ge;\u0026thinsp;10; OR\u0026thinsp;=\u0026thinsp;1.54, 95% CI: 1.09\u0026ndash;2.19, p\u0026thinsp;=\u0026thinsp;0.015) and those with digestive comorbidities (OR\u0026thinsp;=\u0026thinsp;1.71, 95% CI: 1.14\u0026ndash;2.57, p\u0026thinsp;=\u0026thinsp;0.009). In contrast, associations were weaker and not statistically significant in males, older adults (\u0026ge;\u0026thinsp;70 years), rural residents, and those with higher education levels.\u003c/p\u003e\u003cp\u003eExtended subgroup results are provided in Supplementary Table S4, with detailed interaction tests summarized in Supplementary Table S5. Multiple testing corrections using the Benjamini\u0026ndash;Hochberg method are reported in Supplementary Table S6, all of which remained non-significant.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSubgroup Analyses of the Association Between Total Physical Activity (TPA) and Cognitive Impairment\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"10\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModifier\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStratum\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eOR (Q4 vs Q1)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOR (per 1-SD)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003ep-interaction\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e\u003cp\u003en\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSex\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.96\u0026ndash;1.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.083\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.89\u0026ndash;1.13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.982\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.854\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e2512\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.08\u0026ndash;2.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.93\u0026ndash;1.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.360\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e3440\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAge group\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e60\u0026ndash;69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.23\u0026ndash;2.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.00\u0026ndash;1.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.058\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.108\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e3901\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e70\u0026ndash;79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.77\u0026ndash;1.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.450\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.79\u0026ndash;1.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.199\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e1753\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026ge;\u0026thinsp;80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.21\u0026ndash;7.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.827\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.74\u0026ndash;1.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.717\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e298\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eUrban/Rural\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRural\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.97\u0026ndash;1.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.085\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.90\u0026ndash;1.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.972\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.775\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e3234\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUrban\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.21\u0026ndash;2.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.005\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.92\u0026ndash;1.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.305\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e2718\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eEducation\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.21\u0026ndash;2.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.98\u0026ndash;1.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.155\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e1926\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.79\u0026ndash;1.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.430\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.82\u0026ndash;1.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.444\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e4026\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eCES-D10\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.96\u0026ndash;1.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.083\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.89\u0026ndash;1.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.906\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.862\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e3956\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026ge;\u0026thinsp;10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.09\u0026ndash;2.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.92\u0026ndash;1.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.451\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e1996\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eDigestive comorbidity\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.97\u0026ndash;1.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.082\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.88\u0026ndash;1.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.197\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e4068\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.14\u0026ndash;2.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.95\u0026ndash;1.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.170\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e1884\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eLogistic regression with HC3 robust SEs, stratified by subgroup. Adjusted for age, sex, education, marital status, urban/rural residence, smoking, drinking, sleep duration, CES-D10 score, chronic conditions, IADL, and digestive comorbidity. PA quartiles defined after 1%/99% winsorization (Q1\u0026thinsp;\u0026lt;\u0026thinsp;1,732.5; Q2\u0026thinsp;\u0026lt;\u0026thinsp;4,158.0; Q3\u0026thinsp;\u0026lt;\u0026thinsp;9,198.0 MET-min/week). Continuous exposure modeled as log-transformed TPA per 1-SD. Interaction p values from robust Wald tests comparing models with and without interaction terms.\u003c/p\u003e\u003cp\u003eRobustness and Methodological Checks\u003c/p\u003e\u003cp\u003eA series of robustness analyses were conducted to evaluate the stability of the main findings (Supplementary Tables S7\u0026ndash;S9). Sensitivity checks using a stricter cognitive impairment threshold (\u0026thinsp;\u0026le;\u0026thinsp;\u0026minus;\u0026thinsp;1.5 SD) and excluding participants aged\u0026thinsp;\u0026gt;\u0026thinsp;85 years yielded results consistent with the primary models. In both cases, participants in the highest TPA quartile remained at significantly greater odds of cognitive impairment (OR\u0026thinsp;=\u0026thinsp;1.37, 95% CI: 1.06\u0026ndash;1.77, p\u0026thinsp;=\u0026thinsp;0.016; and OR\u0026thinsp;=\u0026thinsp;1.40, 95% CI: 1.10\u0026ndash;1.77, p\u0026thinsp;=\u0026thinsp;0.005, respectively). Alternative specifications of physical activity\u0026mdash;using raw, truncated, and log-transformed values\u0026mdash;produced comparable estimates, indicating that the observed associations were not driven by scaling choices.\u003c/p\u003e\u003cp\u003eMissing data patterns and sample flow are summarized in Supplementary Table S8. Of the 11,045 participants aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years in 2018, complete data were available for 5,952 individuals, which comprised the analytic sample. Comparisons of included versus excluded participants showed substantial imbalances for several covariates, with standardized mean differences (SMDs) of 1.11 for education, 0.55 for gender, 0.55 for IADL, 0.44 for marital status, 0.43 for age, and 0.38 for urban residence; only 3 of 12 comparisons had SMD\u0026thinsp;\u0026lt;\u0026thinsp;0.25. These patterns indicate that some selection bias cannot be ruled out, although the analytic sample still captures a broad segment of the source population.\u003c/p\u003e\u003cp\u003eModel diagnostics are reported in Supplementary Table S9. The main regression model demonstrated good discrimination (AUC\u0026thinsp;=\u0026thinsp;0.797) and adequate calibration (Brier score\u0026thinsp;=\u0026thinsp;0.111; calibration intercept\u0026thinsp;\u0026asymp;\u0026thinsp;0; slope\u0026thinsp;\u0026asymp;\u0026thinsp;1). Multicollinearity diagnostics confirmed that all covariates had generalized variance inflation factors (GVIFs)\u0026thinsp;\u0026lt;\u0026thinsp;2, indicating no evidence of problematic collinearity.\u003c/p\u003e\u003cp\u003eTaken together, these robustness and methodological checks support the reliability of the main findings and strengthen confidence in the observed non-linear association between TPA and cognitive impairment.\u003c/p\u003e\u003cp\u003eSubgroup and Interaction Analyses\u003c/p\u003e\u003cp\u003eSubgroup analyses are presented in Supplementary Tables S4\u0026ndash;S6. The association between TPA and cognitive impairment was generally consistent across sex, age group, education level, and urban/rural residence. For example, participants in the highest TPA quartile (Q4) exhibited increased odds of cognitive impairment compared with Q1 in both men and women, although the estimates were not statistically different between subgroups. No significant effect modification was observed (all FDR-adjusted p-values\u0026thinsp;\u0026gt;\u0026thinsp;0.05), indicating that the reverse J-shaped association was stable across major demographic and health strata.\u003c/p\u003e\u003cp\u003eAdditional robustness analyses are summarized in Supplementary Tables S10\u0026ndash;S12. Using a stricter cognitive impairment threshold (\u0026le;\u0026ndash;1.5 SD; Table S10), participants in Q4 remained at significantly greater odds of impairment compared with Q1 (OR\u0026thinsp;=\u0026thinsp;1.37, 95% CI: 1.06\u0026ndash;1.77, p\u0026thinsp;=\u0026thinsp;0.016). Excluding participants older than 85 years (Table S11) yielded nearly identical results (OR\u0026thinsp;=\u0026thinsp;1.40, 95% CI: 1.10\u0026ndash;1.77, p\u0026thinsp;=\u0026thinsp;0.005), suggesting that the findings were not driven by the oldest age groups.\u003c/p\u003e\u003cp\u003eWhen alternative handling of TPA was applied (Table S12), the overall pattern was unchanged. Using raw quartiles, Q4 was associated with higher odds of cognitive impairment (OR\u0026thinsp;=\u0026thinsp;1.39, 95% CI: 1.10\u0026ndash;1.76, p\u0026thinsp;=\u0026thinsp;0.006), and with winsorized quartiles, the estimate was similar (OR\u0026thinsp;=\u0026thinsp;1.41, 95% CI: 1.11\u0026ndash;1.78, p\u0026thinsp;=\u0026thinsp;0.005). In contrast, continuous specifications of TPA per 1-SD increase, whether raw (OR\u0026thinsp;=\u0026thinsp;1.01, 95% CI: 0.92\u0026ndash;1.11, p\u0026thinsp;=\u0026thinsp;0.841) or log-transformed (OR\u0026thinsp;=\u0026thinsp;1.02, 95% CI: 0.93\u0026ndash;1.12, p\u0026thinsp;=\u0026thinsp;0.609), were not statistically significant.\u003c/p\u003e\u003cp\u003eTaken together, these subgroup and sensitivity analyses confirm that the elevated risk of cognitive impairment at very high levels of physical activity was consistent and robust across population strata and analytic specifications.\u003c/p\u003e\u003cp\u003eMissing Data and Multiwave Analyses\u003c/p\u003e\u003cp\u003eRobustness checks addressing missing data and repeated measurements are summarized in Supplementary Tables S13\u0026ndash;S15b. Results were highly consistent across analytic approaches. First, odds ratios obtained using HC3 robust versus cluster-robust (ID) standard errors were identical (Supplementary Table S13), indicating stable inference across variance estimators. Second, multiple imputation of missing covariates (m\u0026thinsp;=\u0026thinsp;20; Supplementary Table S14) yielded estimates closely aligned with the complete-case analysis: Q2 vs Q1 and Q3 vs Q1 remained non-significant, whereas Q4 vs Q1 remained significantly elevated; continuous specifications per +\u0026thinsp;1 SD of log-transformed TPA were not significant.\u003c/p\u003e\u003cp\u003eExtending the analysis across all available CHARLS waves without BMI adjustment (2011\u0026ndash;2020; Supplementary Table S15), per-wave models and a pooled model with wave fixed effects showed the same qualitative pattern: Q4 was consistently associated with higher odds of cognitive impairment relative to Q1, while Q2 and Q3 showed no clear differences. The pooled estimates were confirmed under both HC3 and cluster-robust (ID) variance settings, and all exported confidence intervals and p-values passed QC checks (Supplementary Table S15b).\u003c/p\u003e\u003cp\u003eTogether, these supplementary analyses suggest that the observed non-linear (reverse J-shaped) association\u0026mdash;i.e., elevated risk confined to very high levels of TPA\u0026mdash;is unlikely to be explained by variance specification, missing data bias, or instability across survey waves.\u003c/p\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eThis study, based on a large, nationally representative sample of older Chinese adults from the CHARLS 2018 survey, identified a nonlinear association between total physical activity (TPA) and cognitive impairment (CI). Specifically, participants in the highest quartile of TPA (\u0026ge;\u0026thinsp;9,198 MET-min/week) demonstrated significantly increased odds of CI, while low-to-moderate activity levels were not associated with substantial differences in cognitive risk. Restricted cubic spline models confirmed a reverse J-shaped dose\u0026ndash;response curve, whereby very high volumes of physical activity were linked to elevated risk. These findings remained robust across multiple sensitivity analyses, including alternative CI thresholds, exclusion of the oldest age groups, different parameterizations of TPA, and multiple imputation for missing data. Subgroup analyses suggested that the association was more pronounced among women, individuals with lower educational attainment, urban residents, and those with higher depressive symptoms or digestive comorbidities, although no statistically significant interactions were detected. Importantly, this pattern should be interpreted as a qualitative dose\u0026ndash;response trend rather than a precise prescriptive threshold, given the cross-sectional design and exposure measurement constraints.\u003c/p\u003e\u003cp\u003eThe reverse J-shaped relationship observed in this study suggests that the cognitive benefits of physical activity may plateau or even reverse at very high levels of energy expenditure. The inflection point identified here (around 9,000 MET-min/week) is markedly higher than current global physical activity recommendations, which typically advocate 500\u0026ndash;1,000 MET-min/week for general health benefits [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. This discrepancy likely reflects population-specific differences in lifestyle and occupational demands: in China, high levels of TPA often arise from physically demanding work such as subsistence farming or manual caregiving rather than structured leisure-time exercise [\u003cspan additionalcitationids=\"CR39 CR40\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Such activities may lack the restorative and psychosocial benefits of recreational exercise and could instead contribute to chronic fatigue, oxidative stress, sleep disruption, and impaired recovery, all of which have been implicated in neurocognitive decline [\u003cspan additionalcitationids=\"CR43 CR44\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Moreover, in CHARLS the IPAQ-LF did not elicit domestic/household activities, which may undercount total exposure in older adults; accordingly, the apparent upper-tail elevation should be viewed as approximate rather than definitive.\u003c/p\u003e\u003cp\u003eThese findings are consistent with emerging evidence from other populations. For example, studies in European cohorts have reported that occupational physical activity may not provide the same protective effects against cognitive decline as leisure-time activity, and in some cases has been associated with increased dementia risk [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. This suggests that not only the volume but also the \u003cem\u003econtext\u003c/em\u003e of physical activity is critical for cognitive health. Future longitudinal studies should therefore distinguish occupational from leisure-time domains and incorporate device-based measures to reduce exposure misclassification.\u003c/p\u003e\u003cp\u003eAlthough interaction tests were not statistically significant, subgroup trends provide additional insights. The association between very high TPA and increased cognitive risk appeared stronger among women, those aged 60\u0026ndash;69 years, and individuals with low educational attainment or elevated depressive symptoms. These observations align with the cognitive reserve hypothesis, which posits that individuals with fewer structural or psychosocial resources may be more vulnerable to adverse effects of excessive physical stress [\u003cspan additionalcitationids=\"CR49\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. Similarly, participants with higher depressive symptom scores experienced stronger associations, suggesting that psychosocial well-being may modify the cognitive impact of physical activity. As none of the multiplicative interactions survived false-discovery-rate control, these subgroup patterns should be considered hypothesis-generating and warrant confirmation in adequately powered longitudinal analyses. The hypothesized mechanistic pathways linking total physical activity and cognitive impairment are summarized in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe diagram illustrates the hypothesized reverse J-shaped relationship observed in the CHARLS 2018 cohort (\u0026ge;\u0026thinsp;60 years). Light-to-moderate levels of total physical activity (TPA) are associated with cognitive protection and healthy aging, whereas extreme TPA (\u0026ge;\u0026thinsp;9,000 MET-min/week), often occupational in nature, may contribute to fatigue, oxidative stress, sleep disruption, and impaired recovery. These physiological burdens may increase neuroinflammation and cortisol levels, thereby elevating the risk of cognitive impairment. Subgroup vulnerabilities (women, low education, depressive symptoms, urban residents) may further modify these associations. The figure is intended as a conceptual model to integrate empirical findings with plausible biological and psychosocial mechanisms, rather than as direct causal proof. The thresholds depicted are illustrative and should not be used as prescriptive cut-points.\u003c/p\u003e\u003cp\u003eFrom a public health perspective, these results refine the current understanding of the physical activity\u0026ndash;cognition relationship. While insufficient activity remains a well-established risk factor for multiple chronic conditions, our findings suggest that \u003cem\u003eexcessive\u003c/em\u003e physical activity\u0026mdash;particularly when occupationally driven\u0026mdash;may also carry cognitive risks. Thus, guidelines for older adults should not only emphasize minimum thresholds but also caution against extreme volumes of activity without adequate rest or psychosocial support. In rapidly aging societies such as China, interventions should promote accessible, sustainable forms of moderate activity\u0026mdash;such as walking, tai chi, or light calisthenics\u0026mdash;that balance physical stimulation with recovery and social engagement [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. Consistent with this, any numeric \u0026ldquo;upper limits\u0026rdquo; inferred from cross-sectional data should be avoided in practice and replaced by individualized counseling that accounts for context (occupational vs leisure), comorbidities, and recovery.\u003c/p\u003e\u003cp\u003eThe observed associations also underscore the need for integrating physical activity promotion with broader health strategies that address mental health, comorbidities, and socioeconomic disparities. For example, community-based programs that combine exercise with psychological support and health literacy may yield synergistic benefits for cognitive aging [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Such precision public health approaches are aligned with global strategies, including the WHO\u0026rsquo;s Global Action Plan on Physical Activity[\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eKey strengths of this study include the use of a large, nationally representative dataset, standardized cognitive assessments, and rigorous analytic approaches including restricted cubic splines and multiple sensitivity checks. However, several limitations warrant consideration. First, the cross-sectional design precludes causal inference, and reverse causality (whereby cognitive decline reduces physical activity) cannot be excluded. Second, physical activity was captured via interval-based self-report items from the IPAQ-LF and then converted to continuous MET-min/week using standard coefficients; this procedure may systematically over- or under-estimate actual energy expenditure, especially in older adults. Prior validation studies indicate that IPAQ-LF may overestimate activity volume, particularly in agricultural or caregiving populations, which could partly explain the apparent risk elevation in the highest quartile. In addition, the CHARLS version of IPAQ-LF did not include domestic or household activities, which are common in older adults, leading to potential undercounting in some groups. Accordingly, the observed reverse J-shaped curve should be interpreted as reflecting overall dose\u0026ndash;response trends rather than precise thresholds. Third, cognitive impairment was defined using a validated composite cut-off of \u0026lt;\u0026thinsp;10 (range 0\u0026ndash;21), consistent with prior CHARLS-based validations [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], but dichotomization may mask subclinical variation and reduce sensitivity relative to clinical diagnoses. Fourth, notable differences between included and excluded participants (e.g., older age, lower education, greater IADL limitations among those excluded) raise the possibility of selection bias; multiple imputation and robustness checks mitigate but cannot eliminate this concern. Fifth, although we computed E-values for key contrasts\u0026mdash;for example, the OR of 1.41 for Q4 vs Q1 corresponds to an E-value of 2.04 (lower CI: 1.39)\u0026mdash;residual confounding remains possible and causal claims are unwarranted. Finally, unmeasured factors such as diet, genetics (e.g., APOE-ε4), or social engagement may still influence the observed relationship [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. Future longitudinal studies incorporating accelerometry, domain-specific exposure (occupational vs. leisure), and refined cognitive phenotyping are needed to validate these findings and to better define safe upper bounds for physical activity in older adults.\u003c/p\u003e\u003cp\u003eIn summary, this study demonstrates a reverse J-shaped association between total physical activity and cognitive impairment in older Chinese adults, with significantly increased risk observed only at very high levels of activity. These results highlight the complexity of the physical activity\u0026ndash;cognition relationship, suggesting that more is not always better. Public health strategies should emphasize moderation, context, and individual tailoring to optimize cognitive outcomes in aging populations. Future longitudinal and interventional research should incorporate accelerometry, distinguish activity domains, and assess recovery/sleep to establish causal pathways and inform implementable guidance for cognitive health promotion.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study, using nationally representative data from older Chinese adults, identified a reverse J-shaped association between total physical activity and cognitive impairment. Compared with low-to-moderate activity levels, which were not linked to substantially higher cognitive risk, very high activity (\u0026ge;\u0026thinsp;9,198 MET-min/week) was consistently associated with increased odds of impairment.\u003c/p\u003e\u003cp\u003eThese findings challenge the prevailing assumption of linear benefits from physical activity and suggest the existence of an optimal range that balances health benefits without inducing potential physiological or psychological strain.\u003c/p\u003e\u003cp\u003eFrom a public health perspective, the results highlight the need to refine physical activity recommendations for aging populations, promoting moderate and sustainable engagement while avoiding excessive volumes often associated with occupational demands. Future longitudinal and interventional studies with objective exposure measures and refined cognitive outcomes are warranted to establish causality and guide evidence-based policy and practice in cognitive health promotion.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is a secondary analysis of de-identified data from the China Health and Retirement Longitudinal Study (CHARLS). The original CHARLS protocols were approved by the Institutional Review Board of Peking University (household survey: IRB00001052-11015; biomarker collection: IRB00001052-11014), and written informed consent was obtained from all participants at each wave. No new data were collected for the present analysis; per the CHARLS data-use agreement and local institutional policies, no additional ethics approval was required. All procedures were performed in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. The manuscript contains no identifiable personal data (images, videos, or individual details).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe public-use CHARLS datasets analysed in this study are available to qualified researchers upon registration and data-use approval at the CHARLS repository (http://charls.pku.edu.cn/en). All summary data supporting the findings are included in this article and its supplementary files. Additional materials (e.g., analysis code) are available from the corresponding author upon reasonable request.\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\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Key R\u0026amp;D Program of China (Grant No. 2023YFF1104400) and the Heilongjiang Provincial Department of Education Key Research Project (Grant No. 14532D009). The funders had no role in the study design; data collection, analysis, or interpretation; writing of the report; or the decision to submit the article for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYZ and GH: conceptualization; methodology; formal analysis; visualization; writing\u0026mdash;original draft.\u003c/p\u003e\n\u003cp\u003eLN: data curation; investigation; validation; visualization.\u003c/p\u003e\n\u003cp\u003eZG: investigation; visualization; writing\u0026mdash;review \u0026amp; editing.\u003c/p\u003e\n\u003cp\u003eXX: conceptualization; supervision; project administration; writing\u0026mdash;review \u0026amp; editing; correspondence; guarantor of the work.\u003c/p\u003e\n\u003cp\u003eLL: supervision; project administration; writing\u0026mdash;review \u0026amp; editing; correspondence.\u003c/p\u003e\n\u003cp\u003eAll authors contributed to interpretation of results, critically revised the manuscript for important intellectual content, and approved the final version for submission. YZ and GH contributed equally and share first authorship.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge the CHARLS team and staff for providing access to the survey data and for their continued efforts in study design, data collection, and management. The interpretations and conclusions presented here are solely those of the authors and do not represent the views of CHARLS or its institutions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; information (optional)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial registration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClinical trial number: not applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eQin F, Luo M, Xiong Y, Zhang N, Dai Y, Kuang W, et al. Prevalence and associated factors of cognitive impairment among the elderly population: A nationwide cross-sectional study in China. Front Public Health. 2022;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2022.1032666\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2022.1032666\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJia L, Du Y, Chu L, Zhang Z, Li F, Lyu D, et al. Prevalence, risk factors, and management of dementia and mild cognitive impairment in adults aged 60 years or older in China: a cross-sectional study. Lancet Public Health. 2020;5:e661\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S2468-2667(20)30185-7\u003c/span\u003e\u003cspan address=\"10.1016/S2468-2667(20)30185-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJiang B, Liu Q, Li J-P, Lin S-N, Wan H-J, Yu Z-W, et al. Prevalence and risk factors for dementia and mild cognitive impairment among older people in Southeast China: a community-based study. BMC Geriatr. 2024;24:466. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12877-024-05054-6\u003c/span\u003e\u003cspan address=\"10.1186/s12877-024-05054-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao X, Xu X, Yan Y, Lipnicki DM, Pang T, Crawford JD, et al. Independent and joint associations of cardiometabolic multimorbidity and depression on cognitive function: findings from multi-regional cohorts and generalisation from community to clinic. Lancet Reg Health \u0026ndash; Western Pac. 2024;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanwpc.2024.101198\u003c/span\u003e\u003cspan address=\"10.1016/j.lanwpc.2024.101198\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu Q, Ailshire JA, Kim JK, Crimmins EM. The Association Between Cardiometabolic Risk and Cognitive Function Among Older Americans and Chinese. J Gerontol Biol Sci Med Sci. 2024;79:glae116. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/gerona/glae116\u003c/span\u003e\u003cspan address=\"10.1093/gerona/glae116\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRandv\u0026auml;li M, Toomsoo T, Šteinmiller J. The Main Risk Factors in Type 2 Diabetes for Cognitive Dysfunction, Depression, and Psychosocial Problems: A Systematic Review. Diabetology. 2024;5:40\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/diabetology5010004\u003c/span\u003e\u003cspan address=\"10.3390/diabetology5010004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun T, Chen S, Wu K, Sun M, Zhang X, You C. Trends in Incidence and Mortality of Stroke in China From 1990 to 2019. Front Neurol. 2021;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fneur.2021.759221\u003c/span\u003e\u003cspan address=\"10.3389/fneur.2021.759221\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang C, Yang X, Wan D, Ma Q, Yin P, Zhou M, et al. Burden of neurological disorders in China and its provinces, 1990\u0026ndash;2021: Findings from the global burden of disease study 2021. Med. 2025;6:100692. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.medj.2025.100692\u003c/span\u003e\u003cspan address=\"10.1016/j.medj.2025.100692\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMa Q, Li R, Wang L, Yin P, Wang Y, Yan C, et al. Temporal trend and attributable risk factors of stroke burden in China, 1990\u0026ndash;2019: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2021;6:e897\u0026ndash;906. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S2468-2667(21)00228-0\u003c/span\u003e\u003cspan address=\"10.1016/S2468-2667(21)00228-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Jin T, Bai C, Wang X, Zhang H, Zhang T. Analysis of Stroke Burden in China from 1990 to 2021 and Projections for the Next Decade. Neuroepidemiology. 2024;1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1159/000542487\u003c/span\u003e\u003cspan address=\"10.1159/000542487\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi R, Qi J, Yang Y, Wu Y, Yin P, Zhou M, et al. Disease Burden and Attributable Risk Factors of Alzheimer\u0026rsquo;s Disease and Dementia in China from 1990 to 2019. J Prev Alzheimer\u0026rsquo;s Disease. 2022;9:306\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.14283/jpad.2021.69\u003c/span\u003e\u003cspan address=\"10.14283/jpad.2021.69\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eErickson KI, Voss MW, Prakash RS, Basak C, Szabo A, Chaddock L, et al. Exercise training increases size of hippocampus and improves memory. Proc Natl Acad Sci U S A. 2011;108:3017\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.1015950108\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1015950108\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBliss ES, Wong RH, Howe PR, Mills DE. Benefits of exercise training on cerebrovascular and cognitive function in ageing. J Cereb Blood Flow Metab. 2021;41:447\u0026ndash;70. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0271678X20957807\u003c/span\u003e\u003cspan address=\"10.1177/0271678X20957807\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSilva NCBS, Barha CK, Erickson KI, Kramer AF, Liu-Ambrose T. Physical exercise, cognition, and brain health in aging. Trends Neurosci. 2024;47:402\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tins.2024.04.004\u003c/span\u003e\u003cspan address=\"10.1016/j.tins.2024.04.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKleinloog JPD, Mensink RP, Ivanov D, Adam JJ, Uludağ K, Joris PJ. Aerobic Exercise Training Improves Cerebral Blood Flow and Executive Function: A Randomized, Controlled Cross-Over Trial in Sedentary Older Men. Front Aging Neurosci. 2019;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnagi.2019.00333\u003c/span\u003e\u003cspan address=\"10.3389/fnagi.2019.00333\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTari AR, Walker TL, Huuha AM, Sando SB, Wisloff U. Neuroprotective mechanisms of exercise and the importance of fitness for healthy brain ageing. Lancet. 2025;405:1093\u0026ndash;118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(25)00184-9\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(25)00184-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAhlskog JE, Geda YE, Graff-Radford NR, Petersen RC. Physical Exercise as a Preventive or Disease-Modifying Treatment of Dementia and Brain Aging. Mayo Clinic Proceedings. 2011;86:876\u0026ndash;84. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4065/mcp.2011.0252\u003c/span\u003e\u003cspan address=\"10.4065/mcp.2011.0252\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDementia prevention, intervention. and care 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.thelancet.com/commissions-do/dementia-prevention-intervention-and-care?utm_source=chatgpt.com\u003c/span\u003e\u003cspan address=\"https://www.thelancet.com/commissions-do/dementia-prevention-intervention-and-care?utm_source=chatgpt.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 29 Sept 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54:1451\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bjsports-2020-102955\u003c/span\u003e\u003cspan address=\"10.1136/bjsports-2020-102955\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePhysical activity. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/initiatives/behealthy/physical-activity\u003c/span\u003e\u003cspan address=\"https://www.who.int/initiatives/behealthy/physical-activity\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 29 Sept 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGallardo-G\u0026oacute;mez D, del Pozo-Cruz J, Noetel M, \u0026Aacute;lvarez-Barbosa F, Alfonso-Rosa RM, del Pozo Cruz B. Optimal dose and type of exercise to improve cognitive function in older adults: A systematic review and bayesian model-based network meta-analysis of RCTs. Ageing Res Rev. 2022;76:101591. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.arr.2022.101591\u003c/span\u003e\u003cspan address=\"10.1016/j.arr.2022.101591\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAfanador-Restrepo DF, Casanova-Correa A, Mart\u0026iacute;n-Ojeda RI, Aibar-Almaz\u0026aacute;n A, Gonz\u0026aacute;lez-Mart\u0026iacute;n AM, Hita-Contreras F, et al. Dose-response relationship of high-intensity training on global cognition in older adults with mild cognitive impairment or dementia: a systematic review with meta-analysis - the ACHIEVE-Study. Eur Rev Aging Phys Activity. 2024;21:23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s11556-024-00358-3\u003c/span\u003e\u003cspan address=\"10.1186/s11556-024-00358-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi Y, Fu C, Song H, Zhang Z, Liu T. Prolonged moderate to vigorous physical activity may lead to a decline in cognitive performance: a Mendelian randomization study. Front Aging Neurosci. 2024;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnagi.2024.1403464\u003c/span\u003e\u003cspan address=\"10.3389/fnagi.2024.1403464\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen X, Yang Q, Liu Z. The impact of physical activities on cognitive function among older adult populations: evidence from China. Front Public Health. 2025;13:1562747. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2025.1562747\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2025.1562747\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang X, Hu P, Ai Y, Zhou S, Li Y, Zhou P, et al. Dual group-based trajectories of physical activity and cognitive function in aged over 55: a nationally representative cohort study. Front Public Health. 2024;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2024.1450167\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2024.1450167\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu Z, Zhang H, Miao X, Li H, Pan H, Zhou D, et al. High-intensity physical activity is not associated with better cognition in the elder: evidence from the China Health and Retirement Longitudinal Study. Alzheimer\u0026rsquo;s Res Therapy. 2021;13:182. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13195-021-00923-3\u003c/span\u003e\u003cspan address=\"10.1186/s13195-021-00923-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\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\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ije/dys203\u003c/span\u003e\u003cspan address=\"10.1093/ije/dys203\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCraig CL, Marshall AL, Sj\u0026ouml;str\u0026ouml;m M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1249/01.MSS.0000078924.61453.FB\u003c/span\u003e\u003cspan address=\"10.1249/01.MSS.0000078924.61453.FB\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo S, Zheng X-Y. New evidence of trends in cognitive function among middle-aged and older adults in China, 2011\u0026ndash;2018: an age-period-cohort analysis. BMC Geriatr. 2023;23:498. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12877-023-04166-9\u003c/span\u003e\u003cspan address=\"10.1186/s12877-023-04166-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou R, Li J, Chen M. The Association Between Cognitive Impairment and Subsequent Falls Among Older Adults: Evidence From the China Health and Retirement Longitudinal Study. Front Public Health. 2022;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2022.900315\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2022.900315\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang W, O\u0026rsquo;Brien N, Forrest JI, Salters KA, Patterson TL, Montaner JSG, et al. Validating a Shortened Depression Scale (10 Item CES-D) among HIV-Positive People in British Columbia, Canada. PLoS ONE. 2012;7:e40793. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0040793\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0040793\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eO\u0026rsquo;brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Qual Quant. 2007;41:673\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11135-006-9018-6\u003c/span\u003e\u003cspan address=\"10.1007/s11135-006-9018-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMacKinnon JG, White H. Some heteroskedasticity-consistent covariance matrix estimators with improved finite sample properties. J Econ. 1985;29:305\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0304-4076(85)90158-7\u003c/span\u003e\u003cspan address=\"10.1016/0304-4076(85)90158-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHarrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Cham: Springer International Publishing; 2015. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-319-19425-7\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-19425-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBenjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J Royal Stat Soc Ser B (Methodological). 1995;57:289\u0026ndash;300.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRECOMMENDATIONS. In. WHO Guidelines on Physical Activity and Sedentary Behaviour. World Health Organization; 2020.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eO\u0026rsquo;Keefe JH, O\u0026rsquo;Keefe EL, Lavie CJ. The Goldilocks Zone for Exercise: Not Too Little, Not Too Much. Mo Med. 2018;115:98\u0026ndash;105.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLuo M, Gupta N, Holtermann A, Stamatakis E, Ding D. Revisiting the physical activity paradox in a Chinese context: Occupational physical activity and mortality in 142,302 urban working adults from the China Kadoorie Biobank study. Lancet Reg Health West Pac. 2022;23:100457. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanwpc.2022.100457\u003c/span\u003e\u003cspan address=\"10.1016/j.lanwpc.2022.100457\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDu L, Hong F, Luo P, Wang Z, Zeng Q, Guan H, et al. The relationship between occupational physical activity and dyslipidaemia in farmers with varying working modes in southwest China: the China multi-ethnic cohort study. BMC Public Health. 2022;22:840. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12889-022-13266-x\u003c/span\u003e\u003cspan address=\"10.1186/s12889-022-13266-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNg SW, Norton EC, Popkin BM. Why have physical activity levels declined among Chinese adults? Findings from the 1991\u0026ndash;2006 China health and nutrition surveys. Soc Sci Med. 2009;68:1305\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.socscimed.2009.01.035\u003c/span\u003e\u003cspan address=\"10.1016/j.socscimed.2009.01.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi J, Zhang X, Zhang M, Wang L, Yin P, Li C, et al. Urban-rural differences in the association between occupational physical activity and mortality in Chinese working population: evidence from a nationwide cohort study. Lancet Reg Health \u0026ndash; Western Pac. 2024;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanwpc.2024.101083\u003c/span\u003e\u003cspan address=\"10.1016/j.lanwpc.2024.101083\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbd-Elfattah HM, Abdelazeim FH, Elshennawy S. Physical and cognitive consequences of fatigue: A review. J Adv Res. 2015;6:351\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jare.2015.01.011\u003c/span\u003e\u003cspan address=\"10.1016/j.jare.2015.01.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSimioni C, Zauli G, Martelli AM, Vitale M, Sacchetti G, Gonelli A, et al. Oxidative stress: role of physical exercise and antioxidant nutraceuticals in adulthood and aging. Oncotarget. 2018;9:17181\u0026ndash;98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18632/oncotarget.24729\u003c/span\u003e\u003cspan address=\"10.18632/oncotarget.24729\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMilitello R, Luti S, Gamberi T, Pellegrino A, Modesti A, Modesti PA. Physical Activity and Oxidative Stress in Aging. Antioxidants. 2024;13:557. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/antiox13050557\u003c/span\u003e\u003cspan address=\"10.3390/antiox13050557\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWood E, Hall KH, Tate W. Role of mitochondria, oxidative stress and the response to antioxidants in myalgic encephalomyelitis/chronic fatigue syndrome: A possible approach to SARS-CoV-2 \u0026lsquo;long-haulers\u0026rsquo;? Chronic Dis Translational Med. 2021;7:14\u0026ndash;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cdtm.2020.11.002\u003c/span\u003e\u003cspan address=\"10.1016/j.cdtm.2020.11.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZotcheva E, Bratsberg B, Strand BH, Jugessur A, Engdahl BL, Bowen C, et al. Trajectories of occupational physical activity and risk of later-life mild cognitive impairment and dementia: the HUNT4 70\u0026thinsp;+\u0026thinsp;study. Lancet Reg Health \u0026ndash; Europe. 2023;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanepe.2023.100721\u003c/span\u003e\u003cspan address=\"10.1016/j.lanepe.2023.100721\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNabe-Nielsen K, Garde AH, Gyntelberg F, Hansen \u0026Aring;M, Prescott E, Laursen P, et al. Effects of Occupational and Leisure-Time Physical Activity on the Risk of Dementia: Results From the Copenhagen City Heart Study. Scand J Med Sci Sports. 2024;34:e14730. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/sms.14730\u003c/span\u003e\u003cspan address=\"10.1111/sms.14730\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStern Y. Cognitive reserve in ageing and Alzheimer\u0026rsquo;s disease. Lancet Neurol. 2012;11:1006\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S1474-4422(12)70191-6\u003c/span\u003e\u003cspan address=\"10.1016/S1474-4422(12)70191-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePettigrew C, Soldan A. Defining Cognitive Reserve and Implications for Cognitive Aging. Curr Neurol Neurosci Rep. 2019;19:1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11910-019-0917-z\u003c/span\u003e\u003cspan address=\"10.1007/s11910-019-0917-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNogueira J, Gerardo B, Santana I, Sim\u0026otilde;es MR, Freitas S. The Assessment of Cognitive Reserve: A Systematic Review of the Most Used Quantitative Measurement Methods of Cognitive Reserve for Aging. Front Psychol. 2022;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2022.847186\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2022.847186\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePotdevin FJ, Femery VG, Decatoire A, Bosquet L, Coello Y, Moretto P. Using effect size to quantify plantar pressure asymmetry of gait of nondisabled adults and patients with hemiparesis. J Rehabil Res Dev. 2007;44:347\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1682/jrrd.2006.07.0077\u003c/span\u003e\u003cspan address=\"10.1682/jrrd.2006.07.0077\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmerican College of Sports Medicine, Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, Minson CT, Nigg CR, et al. American College of Sports Medicine position stand. Exercise and physical activity for older adults. Med Sci Sports Exerc. 2009;41:1510\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1249/MSS.0b013e3181a0c95c\u003c/span\u003e\u003cspan address=\"10.1249/MSS.0b013e3181a0c95c\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNgandu T, Lehtisalo J, Solomon A, Lev\u0026auml;lahti E, Ahtiluoto S, Antikainen R, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet. 2015;385:2255\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(15)60461-5\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(15)60461-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKivipelto M, Solomon A, Ahtiluoto S, Ngandu T, Lehtisalo J, Antikainen R, et al. The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER): study design and progress. Alzheimers Dement. 2013;9:657\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jalz.2012.09.012\u003c/span\u003e\u003cspan address=\"10.1016/j.jalz.2012.09.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWHO\u0026rsquo;s work on the UN Decade of Healthy Ageing. (2021\u0026ndash;2030). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/initiatives/decade-of-healthy-ageing\u003c/span\u003e\u003cspan address=\"https://www.who.int/initiatives/decade-of-healthy-ageing\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 29 Sept 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGlobal action plan on physical activity. 2018\u0026ndash;2030: more active people for a healthier world. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/publications/i/item/9789241514187\u003c/span\u003e\u003cspan address=\"https://www.who.int/publications/i/item/9789241514187\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 29 Sept 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCorder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer\u0026rsquo;s Disease in Late Onset Families. Science. 1993;261:921\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.8346443\u003c/span\u003e\u003cspan address=\"10.1126/science.8346443\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu C-C, Kanekiyo T, Xu H, Bu G. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol. 2013;9:106\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nrneurol.2012.263\u003c/span\u003e\u003cspan address=\"10.1038/nrneurol.2012.263\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Cognitive impairment, Physical activity, Older adults, Dose–response relationship, China","lastPublishedDoi":"10.21203/rs.3.rs-7763220/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7763220/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eCognitive impairment (CI) is a major challenge in China\u0026rsquo;s rapidly aging population. Although physical activity (PA) is generally considered beneficial, the dose\u0026ndash;response relationship\u0026mdash;particularly at high exposure levels\u0026mdash;remains uncertain.\u003c/p\u003e\u003ch2\u003eObjective\u003c/h2\u003e\u003cp\u003eTo characterize the association between total physical activity (TPA, MET-min/week) and CI in older Chinese adults, with emphasis on nonlinearity and subgroup heterogeneity.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe analyzed adults aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years in the 2018 wave of the China Health and Retirement Longitudinal Study (CHARLS; n\u0026thinsp;=\u0026thinsp;5,952). TPA was derived from the IPAQ-LF and examined as quartiles and as a log-transformed, z-standardized continuous variable. CI was defined as a validated composite score\u0026thinsp;\u0026lt;\u0026thinsp;10. Associations were estimated using multivariable logistic regression with robust (HC3) SEs; dose\u0026ndash;response was modeled with restricted cubic splines (RCS) and corroborated using a segmented (piecewise) logistic model. Prespecified covariates included sociodemographic, behavioral, and health factors. Sensitivity analyses included survey-weighted models, multiple imputation for missing covariates, alternative TPA/CI operationalizations, and FDR-controlled interaction tests.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eRelative to Q1 (\u0026lt;\u0026thinsp;1,732.5 MET-min/week), Q2 and Q3 showed no association with CI (OR 0.98, 95% CI 0.78\u0026ndash;1.22; OR 1.19, 95% CI 0.96\u0026ndash;1.49). Q4 (\u0026ge;\u0026thinsp;9,198.0 MET-min/week) was associated with higher CI odds (OR 1.41, 95% CI 1.11\u0026ndash;1.78). Continuous log-TPA (per +\u0026thinsp;1 SD) was not associated with CI. RCS indicated a reverse J-shaped dose\u0026ndash;response, with risk elevation confined to very high TPA; the segmented model located the inflection near the upper exposure tail (\u0026asymp;\u0026thinsp;9,200 MET-min/week). Patterns were directionally stronger among women, younger-old adults, urban residents, and those with lower education or depressive symptoms, but no multiplicative interactions were significant after FDR correction. Findings were robust to weighting, multiple imputation, and alternative exposure/outcome definitions.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eIn this national sample of older Chinese adults, TPA exhibited a reverse J-shaped relationship with CI: low-to-moderate activity was not associated with increased risk, whereas very high TPA was linked to higher odds of impairment. Results support individualized, context-specific PA guidance that recognizes potential upper-limit risks in late life and should be verified in longitudinal and interventional studies.\u003c/p\u003e","manuscriptTitle":"Nonlinear Association of Total Physical Activity with Cognitive Impairment in Older Chinese Adults: A Cross-Sectional Analysis of CHARLS","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-19 15:20:58","doi":"10.21203/rs.3.rs-7763220/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2025-11-10T11:19:39+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-13T08:34:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-10T06:56:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-10T06:52:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Geriatrics","date":"2025-10-02T02:42:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1116cce4-8033-44a5-a573-f4e1f8334e36","owner":[],"postedDate":"November 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-11-19T15:20:58+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-19 15:20:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7763220","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7763220","identity":"rs-7763220","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}