Age‐friendly environments and incident motoric cognitive risk syndrome in older adults: A prospective cohort study Authors: Hui Zhang, Zixin Hu, Chongyu Ding, Yaqian Xu, Shuai Jiang, Yi Li, Jiucun Wang, Xiaofeng Wang, Meng Hao, Xiangwei Li Published: 2026 ABSTRACT AbstractINTRODUCTIONMotoric cognitive risk syndrome (MCR) is a predementia condition that predicts multiple adverse health outcomes. Age‐friendly environments (AFE) represent a key strategy for promoting healthy aging and may reduce MCR risk. However, little is known about the association between AFE status and MCR risk.METHODSThis prospective cohort study used data from the National Health and Aging Trends Study. AFE status was assessed across eight domains based on the World Health Organization (WHO) framework. MCR was evaluated annually from 2011 to 2019. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs); restricted cubic spline regressions were used to examine dose–response relationships.RESULTSA total of 7,608 older adults (mean age: 76.81 ± 7.53 years; 57.32% female) without baseline MCR were included. Over a median follow‑up of 4 years, 801 incident MCR cases occurred. Compared with participants in the lowest quartile (Q1) of AFE scores, those in Q3 (HR 0.75, 95% CI: 0.61–0.91) and Q4 (HR 0.62, 95% CI: 0.50–0.78) had significantly lower MCR risks. Additionally, a dose–response association was observed between higher AFE scores and lower risk of incident MCR (HR 0.96, 95% CI: 0.94–0.98).DISCUSSIONAFE was associated with a decreased risk of incident MCR among older adults. These findings suggest that creating AFE may be a feasible population‑based strategy to delay pre‑dementia syndromes, thereby contributing to dementia prevention. BACKGROUND Motoric cognitive risk syndrome (MCR), a predementia syndrome, is characterized by subjective cognitive complaints and slow gait speed in individuals without dementia or mobility disability. 1 Observational epidemiological studies suggest that MCR is associated with more than a two‐fold increased risk of developing cognitive impairment and dementia, 2 , 3 as well as a pattern of brain changes characterized by significant white matter abnormalities and frontal–parietal atrophy. 4 , 5 , 6 It has been reported that approximately 25.6% of individuals with MCR may progress to dementia over a median follow‐up period of 3 years. 7 Additionally, MCR is a strong and independent predictor of adverse health outcomes, including falls, disability, and mortality. 3 Notably, there is growing evidence that MCR is common, and affects approximately 10% of older adults worldwide. 8 Given its high prevalence and associated adverse effects, identifying modifiable risk factors for MCR is increasingly important for implementing early prevention strategies. Previous studies have identified several risk factors for MCR at the individual level, such as chronic diseases, poor physical performance, unfavorable behavioral and social factors. 9 , 10 , 11 , 12 Beyond individual‑level factors, emerging evidence suggests that community environments may also play an important role in shaping cognitive aging. 13 , 14 , 15 However, most existing studies have merely examined single environmental factors in isolation. Multiple community‑level determinants coexist across the life course, and the cumulative effect of environmental conditions may be more relevant to MCR risk than any single domain. To address this gap, the World Health Organization (WHO) introduced the concept of age‑friendly environments (AFE), a multidimensional framework that encompasses multiple domains to optimize community environments for healthy aging. 16 , 17 In particular, it has been linked to a reduced risk of geriatric syndromes, such as incident frailty and sarcopenia, as well as slower progression of multimorbidity. 18 , 19 , 20 Nevertheless, no study has yet examined the association between a comprehensive, multi‑domain AFE assessment and the risk of incident MCR. Therefore, using data from a large, prospective, and population‐based cohort, we first assessed AFE status, and then investigated its association with incident MCR. We further examined the dose‑response relationships to determine whether higher AFE scores confer progressively greater protection. METHODS Study population We analyzed data from the National Health and Aging Trends Study (NHATS), a longitudinal survey of Medicare beneficiaries aged ≥ 65 years residing in the United States. 21 The total sample consisted of two subsets: subset I included participants enrolled in 2011 (Round 1, n = 8,245), and subset II comprised newly recruited individuals in 2015 (Round 5, n = 4,182). In this analysis, we used data from Round 1 (2011) to Round 9 (2019). At baseline, a total of 12,427 individuals (subset I and II) were recruited. After excluding 1,164 individuals (750 in subset I and 414 in subset II) with missing AFE information, 1,973 individuals (1,400 in subset I and 573 in subset II) with prevalent MCR, dementia, or missing MCR data at baseline, and 1,682 individuals (1,143 in subset I and 539 in subset II) who were lost to follow‑up or died before MCR assessment. A final sample of 7,608 individuals was included in this study (Figure S1). AFE Based on the original AFE framework proposed by the WHO, 16 we identified corresponding items from the NHATS questionnaire based on two criteria: (1) conceptual alignment with the WHO domain definitions, and (2) availability in the NHATS dataset. Specifically, we assessed AFE across eight domains in NHATS, 17 including social participation (nine items), outdoor space and building (two items), housing (10 items), respect and social inclusion (six items), communication and information (three items), community support and health care (seven items), transportation (one item), civic participation and employment (one item). Details on the specific NHATS items mapped onto each domain are provided in Table S1. In our analyses, each item was classified as present (scored 1) or absent (scored 0), resulting in a total derived AFE score ranging from 0 to 39, with higher scores indicating better AFE status. In addition to analyzing derived AFE score as a continuous variable, the derived AFE scores were also categorized into quartiles (Q1–Q4). MCR MCR was defined as the presence of both subjective cognitive complaints and slow gait speed in individuals without dementia or mobility disability. 1 In the NHATS, subjective cognitive complaints were based on responses to three standardized questionnaires (Table S2). Gait speed was measured using a 3‐m test in the NHATS. Slow gait speed was defined as gait speed ≥ 1 standard deviation (SD) below age‐specific (65–74 years or ≥75 years) and sex‐specific (male or female) means within this cohort. 22 , 23 The specific cutoffs for slow gait speed are presented in Table S2. Dementia was defined by the participant or a proxy that a doctor had diagnosed the participant with dementia or Alzheimer's disease. Mobility disability was determined as the inability to walk even with an assistive device, consistent with prior studies. 8 , 24 , 25 RESEARCH IN CONTEXT Systematic review: The authors searched the literature using the PubMed database and references from relevant articles. Previous studies have identified individual‑level risk factors for motoric cognitive risk syndrome (MCR), but the role of community environments remains underexplored. The age‑friendly environments (AFE) framework emphasizes multidimensional domains of community environments, and represents a key strategy for promoting healthy aging. However, its association with MCR risk is unknown. Interpretation: In this prospective and population‐based study, we found that higher AFE scores were dose‑dependently associated with a reduced risk of incident MCR. These findings suggest that promoting AFE may represent a feasible community‑based strategy to delay pre‑dementia syndromes, thereby contributing to dementia prevention. Future directions: Future studies are warranted to confirm these findings in diverse populations and to explore the causal mechanisms linking AFE to MCR. Systematic review: The authors searched the literature using the PubMed database and references from relevant articles. Previous studies have identified individual‑level risk factors for motoric cognitive risk syndrome (MCR), but the role of community environments remains underexplored. The age‑friendly environments (AFE) framework emphasizes multidimensional domains of community environments, and represents a key strategy for promoting healthy aging. However, its association with MCR risk is unknown. Systematic review: The authors searched the literature using the PubMed database and references from relevant articles. Previous studies have identified individual‑level risk factors for motoric cognitive risk syndrome (MCR), but the role of community environments remains underexplored. The age‑friendly environments (AFE) framework emphasizes multidimensional domains of community environments, and represents a key strategy for promoting healthy aging. However, its association with MCR risk is unknown. Interpretation: In this prospective and population‐based study, we found that higher AFE scores were dose‑dependently associated with a reduced risk of incident MCR. These findings suggest that promoting AFE may represent a feasible community‑based strategy to delay pre‑dementia syndromes, thereby contributing to dementia prevention. Interpretation: In this prospective and population‐based study, we found that higher AFE scores were dose‑dependently associated with a reduced risk of incident MCR. These findings suggest that promoting AFE may represent a feasible community‑based strategy to delay pre‑dementia syndromes, thereby contributing to dementia prevention. Future directions: Future studies are warranted to confirm these findings in diverse populations and to explore the causal mechanisms linking AFE to MCR. Future directions: Future studies are warranted to confirm these findings in diverse populations and to explore the causal mechanisms linking AFE to MCR. Covariates In this study, covariates included demographic characteristics, lifestyle factors, and chronic diseases. Demographic characteristics included baseline age, age groups (less than 75 vs. 75 years or older), race/ethnicity (non‐Hispanic White, non‐Hispanic Black, or Other), gender (male or female), marital status (married/living with partner, or singles), educational levels (less than high school, high school, or more than high school), and living arrangements (community or others). Lifestyle factors included body mass index (BMI), vigorous exercise (yes/no), and smoking status (ever/never). Vigorous exercise was assessed by asking participants whether they had spent any time engaging in vigorous activities in the last month (yes/no). BMI was used to divide participants into normal or underweight (< 25 kg/m2), overweight (25–30 kg/m2), and obese groups (> 30 kg/m2). Chronic diseases were obtained from self‐reports by participants or proxies, including the presence of arthritis (including osteo or rheumatoid arthritis), hypertension, diabetes mellitus, lung disease (such as emphysema, asthma, or chronic bronchitis), cancer, osteoporosis, stroke, heart attack or myocardial infarction, and heart disease (including angina, and congestive heart failure). We divided these participants into three groups based on the number of chronic diseases (0–1, 2–3, > 3). Depressive symptoms were assessed with the Patient Health Questionnaire for Depression, which comprises two items for depression. 26 Having depressive symptoms were defined using a cutoff score of 3. 26 Statistical analysis We describe data from the two samples with the mean (M) and SD or frequency (%) for continuous and categorical variables, respectively. Cox proportional hazard models were used to estimate the hazard ratio (HR) and confidence intervals (CIs) of MCR associated with AFE status after adjusting for confounders using three models. Proportional hazards assumptions for Cox models were checked using scaled Schoenfeld residuals. 27 Model 1 was unadjusted; Model 2 was adjusted for age, gender, race, educational level, and marital status; Model 3 was adjusted for model 2 + smoking status, body mass index categories, vigorous exercise, numbers of chronic disease, living arrangements, and depressive symptoms. We applied restricted cubic spline (RCS) regressions, with three knots located at the 10th, 50th, and 90th percentiles of AFE score, to estimate the possible nonlinear relationship between AFE score and incident MCR risk by adjusting for potential confounding factors. 28 The test of potential nonlinearity was assessed by the chi‐square test, comparing the model with only the linear term to the model that included the cubic spline terms. In these analyses, AFE score was analyzed as a continuous variable, and HRs and 95% CI were calculated using the median value of AFE score as a reference value. Additionally, we adopted a multidimensional framework based on the existing literature, 12 , 29 , 30 organizing MCR risk predictors into three domains: sociodemographic (age, gender, education, marital status, race), health‑related (BMI, chronic diseases, depression), and behavioral (smoking, vigorous exercise). They were intended to examine whether the association between AFE and MCR varied across key strata. These analyses were exploratory in nature and intended to generate hypotheses rather than to draw confirmatory conclusions. Subgroup analyses for living arrangements were not performed due to the limited sample size within certain strata, which precluded reliable statistical estimation. Effect modification was evaluated by including multiplicative interaction terms between AFE score and each stratification variable. Sensitivity analyses were performed to assess the robustness of results: (1) among participants living in community (n = 7,309) (2) among participants with ≥3 years of follow‐up (n = 5,144) (3) among participants with all AFE items completed (n = 6,424). A two‐sided p‐value < 0.05 was considered statistically significant for the primary analysis. To account for multiple comparisons arising from the eight domain‐specific analyses of the AFE scale, we applied a Bonferroni correction. For these secondary analyses, statistical significance was defined as p‐value < 0.006 (0.05/8). All analyses were conducted using R statistical software (version 4.4.1; www.r‐project.org). RESULTS Characteristics of the study population The characteristics of the study population were presented in Table 1. A total of 7,608 older adults (4,952 in subset I, and 2,656 in subset II) without MCR at baseline were analyzed in the pooled sample. Among them, the mean age of the individuals was 76.35 (7.53) years, and the mean AFE score was 20.91 (4.01). During a median follow‐up period of 4 years, 801 (10.53%) individuals were diagnosed as MCR. TABLE 1Characteristics of the study population at baseline.VariablesSubset ISubset IIPooled sampleSample, n (%)4952 (65.09)2656 (34.91)7608Age, years, M (SD)76.81 (7.53)75.50 (7.47)76.35 (7.53)Less than 75 years, n (%)2121 (42.83)1341 (50.49)3462 (45.50)75 years or older, n (%)2831 (57.17)1315 (49.51)4146 (54.50)Gender, n (%)Male2100 (42.41)1147 (43.19)3247 (42.68)Female2852 (57.59)1509 (56.81)4361 (57.32)Body mass index, n (%)Normal or less (< 25 kg/m2)1690 (34.13)782 (29.44)2472 (32.49)Overweight (25–30 kg/m2)1829 (36.93)997 (37.54)2826 (37.15)Obese (> 30 kg/m2)1342 (27.10)837 (31.51)2179 (28.64)Race/ethnicity, n (%)Non‐Hispanic White3658 (73.89)1927 (72.55)5585 (73.41)Non‐Hispanic Black931 (18.80)496 (18.67)1427 (18.76)Others363 (7.33)233 (8.77)596 (7.83)Marital status, n (%)Married/living with partner2613 (52.77)1401 (52.75)4014 (52.76)Single2334 (47.13)1254 (47.21)3588 (47.16)Education, n (%)Less than high school1075 (21.71)452 (17.02)1527 (20.07)High school1370 (27.67)724 (26.26)2094 (27.52)More than high school2506 (50.61)1479 (55.69)3985 (52.38)Smoking status, n (%)Never2419 (48.85)1305 (49.13)3724 (48.95)Ever2533 (51.15)1349 (50.79)3882 (51.03)Vigorous exercise, n (%)Yes1955 (39.48)1139 (42.88)3094 (40.67)No2997 (60.52)1517 (57.12)4514 (59.33)Depressive symptomsYes549 (11.09)258 (9.71)807 (10.61)No4383 (88.51)2377 (90.29)6760 (86.60)Living arrangements, n (%)Community4743 (95.78)2566 (96.61)7309 (96.07)Others209 (4.22)90 (3.39)299 (3.93)No. of chronic diseases, n (%)0–11349 (27.24)804 (30.27)2153 (28.30)2–32274 (45.92)1289 (48.53)3563 (46.83)> 31289 (26.03)520 (19.58)1809 (23.78)AFE score, M (SD)20.84 (4.04)21.04 (3.96)20.91 (4.01)AFE groups, n (%)Q1 (< 18)1017 (20.54)496 (18.67)1513 (19.89)Q2 (≥ 18, < 21)1224 (24.72)673 (25.34)1897 (24.93)Q3 (≥ 21, < 24)1389 (28.05)771 (29.03)2160 (28.39)Q4 (≥ 24)1322 (26.70)716 (26.96)2038 (26.79) Note: Missing in pooled sample: body mass index: 131 (1.72%), marital status: 6 (0.08%), education: 2 (0.03%), smoking: 2 (0.03%), no. of chronic diseases: 83 (1.09%), depressive symptoms: 41 (0.53%).Abbreviations: AFE, age‐friendly environments; M, mean; SD, standard deviation.John Wiley & Sons, Ltd. Characteristics of the study population at baseline. Note: Missing in pooled sample: body mass index: 131 (1.72%), marital status: 6 (0.08%), education: 2 (0.03%), smoking: 2 (0.03%), no. of chronic diseases: 83 (1.09%), depressive symptoms: 41 (0.53%). Abbreviations: AFE, age‐friendly environments; M, mean; SD, standard deviation. Association between AFE and incident MCR The association between AFE status and the risk of incident MCR is presented in Table 2. After adjusting for confounding factors, higher AFE score was associated with a reduced risk of incident MCR in subset I (HR 0.97, 95% CI: 0.95–0.99), subset II (HR 0.92, 95% CI: 0.89–0.96), and the pooled sample (HR 0.96, 95% CI: 0.94–0.98). Additionally, participants were divided into four groups according to AFE score quartiles. Compared with the lowest quartile (Q1), individuals in the highest quartile (Q4) had a lower risk of incident MCR in subset I (HR 0.71, 95% CI: 0.55–0.93), subset II (HR 0.43, 95% CI: 0.27–0.68), and the pooled sample (HR 0.62, 95% CI: 0.50–0.78). Furthermore, RCS regression indicated a linear association between AFE score and incident MCR risk in the pooled sample (Figure 1, p‐overall < 0.001, p‐nonlinear = 0.699), as well as subset I (Figure S2, p‐overall = 0.035, p‐nonlinear = 0.686) and subset II (Figure S3, p‐overall < 0.001, p‐nonlinear = 0.931). TABLE 2Association between AFE status and MCR risk.Model 1Model 2Model 3VariablesEvent/N (%)HR (95% CI) p‐valueHR (95% CI) p‐valueHR (95% CI) p‐value Pooled sample Per each additional score801/7608 (10.53)0.91 (0.89, 0.93)< 0.0010.80 (0.72, 0.88)< 0.0010.96 (0.94, 0.98)< 0.001CategoriesQ1230/1513 (15.20)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)Q2219/1897 (11.54)0.68 (0.56, 0.81)< 0.0010.82 (0.68, 0.98)0.0340.84 (0.69, 1.01)0.061Q3206/2160 (9.54)0.51 (0.42, 0.62)< 0.0010.71 (0.58, 0.86)< 0.0010.75 (0.61, 0.91)0.004Q4146/2038 (7.16)0.36 (0.29, 0.44)< 0.0010.58 (0.46, 0.72)< 0.0010.62 (0.50, 0.78)< 0.001 p‐Trend< 0.001< 0.001< 0.001 Subset I Per each additional score583/4952 (11.77)0.92 (0.90, 0.94)< 0.0010.96 (0.94, 0.99)0.0010.97 (0.95, 0.99)0.011CategoriesQ1163/1017 (15.64)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)Q2154/1224 (12.58)0.69 (0.55, 0.86)< 0.0010.85 (0.68, 1.07)0.1680.87 (0.70, 1.09)0.229Q3152/1389 (10.94)0.53 (0.43, 0.67)< 0.0010.76 (0.61, 0.96)0.0210.80 (0.63, 1.01)0.056Q4114/1322 (8.62)0.40 (0.31, 0.51)< 0.0010.67 (0.51, 0.86)0.0020.71 (0.55, 0.93)0.012 p‐Trend< 0.0010.0010.009 Subset II Per each additional score218/2656 (8.21)0.89 (0.86, 0.92)< 0.0010.91 (0.88, 0.95)< 0.0010.92 (0.89, 0.96)< 0.001Categories218/2656 (8.21)Q167/496 (13.51)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)1 (Ref)Q265/673 (9.66)0.65 (0.46, 0.92)0.0140.74 (0.52, 1.04)0.0840.78 (0.55, 1.10)0.156Q354/771 (7.00)0.46 (0.32, 0.65)< 0.0010.59 (0.40, 0.86)0.0060.64 (0.44, 0.93)0.0820Q432/716 (4.47)0.27 (0.18, 0.42)< 0.0010.39 (0.25, 0.61)< 0.0010.43 (0.27, 0.68)< 0.001 p‐Trend< 0.001< 0.001< 0.001 Note: Model 1: unadjusted. Model 2: adjusted for age, gender, race, educational level, marital status. Model 3: adjusted for model 2 + smoking status, body mass index categories, vigorous exercise, numbers of chronic disease, living arrangements, and depressive symptoms.Abbreviations: AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome.John Wiley & Sons, Ltd. Association between AFE status and MCR risk. Note: Model 1: unadjusted. Model 2: adjusted for age, gender, race, educational level, marital status. Model 3: adjusted for model 2 + smoking status, body mass index categories, vigorous exercise, numbers of chronic disease, living arrangements, and depressive symptoms. Abbreviations: AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome. Restricted cubic spline plot of the associations of AFE score with incident MCR. The Model is adjusted for age, gender, race, educational level, marital status, smoking status, body mass index categories, vigorous exercise, the number of chronic disease and living arrangements and depressive symptoms. AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome. We also conducted stratified analyses in the pooled sample across various subgroups according to age groups, gender, BMI category, race/ethnicity, marital status, educational levels, smoking status, vigorous exercise, and the number of chronic diseases (Figure 2). We found significant associations of increased AFE score with the reduced risk of incident MCR in most stratum. Interaction tests comparing the HRs across strata were not significant for most subgroups (p‐interaction > 0.05), except for age group (p‐interaction < 0.001) and the number of chronic diseases (p‐interaction = 0.002), suggesting that the beneficial association may be more pronounced in younger older adults and those with fewer chronic conditions. Association between AFE scores and MCR risk in subgroups. All models are adjusted for age, gender, race, educational level, marital status, smoking status, body mass index categories, vigorous exercise, the number of chronic diseases, living arrangements, and depressive symptoms. p‐Interaction was evaluated by including multiplicative interaction terms between AFE score and each stratification variable. AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome. Association between AFE domains and incident MCR We further analyzed the associations between individual AFE domains and the risk of incident MCR (Table 3). In the pooled sample, after applying this correction, higher scores in social participation (HR 0.83, 95% CI: 0.79–0.87), communication and information (HR 0.80, 95% CI: 0.73–0.88), and transportation (HR 0.64, 95% CI: 0.54–0.76) were significantly associated with a reduced risk of incident MCR. In contrast, a higher housing score was associated with an increased risk (HR 1.09, 95% CI: 1.04–1.13). No significant associations were observed for respect and social inclusion, outdoor space and buildings, community support and health services, or civic participation and employment. Similar findings were observed in subset I and subset II (Table 3). TABLE 3Association between AFE domains and MCR risk.VariablesEvent/N (%)HR (95% CI) p‐value Pooled Sample Social participation801/7608 (10.53)0.83 (0.79, 0.87)< 0.001Outdoor space and building801/7606 (10.53)1.11 (0.97, 1.25)0.123Housing801/7608 (10.53)1.09 (1.04, 1.13)< 0.001Respect and social inclusion801/7608 (10.53)0.94 (0.89, 0.98)0.010Communication and information801/7608 (10.53)0.80 (0.73, 0.88)< 0.001Community support and health care801/7603 (10.53)0.98 (0.92, 1.04)0.563Transportation787/7491 (10.51)0.64 (0.54, 0.76)< 0.001Civic participation and employment758/7302 (10.38)0.91 (0.78, 1.05)0.187 Subset I Social participation583/4952 (11.77)0.84 (0.79, 0.89)< 0.001Outdoor space and building583/4951 (11.77)1.09 (0.93, 1.26)0.287Housing583/4952 (11.77)1.11 (1.05, 1.17)< 0.001Respect and social inclusion583/4952 (11.77)0.95 (0.89, 1.01)0.082Communication and information583/4952 (11.77)0.88 (0.78, 0.98)0.021Community support and health care583/4952 (11.77)0.99 (0.92, 1.07)0.838Transportation573/4871 (11.76)0.70 (0.57, 0.86)< 0.001Civic participation and employment550/4745 (11.59)0.90 (0.76, 1.06)0.209 Subset II Social participation218/2656 (8.21)0.82 (0.75, 0.90)< 0.001Outdoor space and building218/2655 (8.21)1.13 (0.89, 1.44)0.317Housing218/2656 (8.21)1.03 (0.95, 1.12)0.466Respect and social inclusion218/2656 (8.21)0.93 (0.84, 1.02)0.129Communication and information218/2656 (8.21)0.63 (0.52, 0.76)< 0.001Community support and health care218/2651 (8.22)0.95 (0.85, 1.08)0.445Transportation214/2620 (8.17)0.53 (0.39, 0.74)< 0.001Civic participation and employment208/2557 (8.13)0.98 (0.74, 1.30)0.892 Note: All models are adjusted for age, gender, race, educational level, smoking status, marital status, body mass index categories, vigorous exercise, the number of chronic diseases, living arrangements, and depressive symptoms.Abbreviations: AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome.John Wiley & Sons, Ltd. Association between AFE domains and MCR risk. Note: All models are adjusted for age, gender, race, educational level, smoking status, marital status, body mass index categories, vigorous exercise, the number of chronic diseases, living arrangements, and depressive symptoms. Abbreviations: AFE, age‐friendly environments; CI, confidence intervals; HR, hazard ratio; MCR, motoric cognitive risk syndrome. Sensitivity analyses of the association between AFE status and incident MCR In sensitivity analyses, we reassessed the robustness of our results under the following conditions: (1) among participants living in community, (2) among participants with ≥3 years of follow‐up, (3) among participants with all AFE items completed. Similar findings were observed across all sensitivity analyses (Table S3). DISCUSSION In this prospective, population‑based, and cohort study, we assessed AFE across eight domains and examined its association with incident MCR for the first time. We found that higher AFE scores were linearly associated with a reduced risk of incident MCR, independent of socioeconomic status, lifestyle factors, and chronic diseases. This study presents the largest examination of relationship between systemic AFE assessments and incident MCR risk (a predementia syndrome), whereas previous studies have focused on individual AFE items in relation to dementia and cognitive decline. 13 , 14 For example, Luo et al. conducted a prospective study using 16,190 participants from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2015), and found that more bus lines (beta = 0.012, p‐value < 0.05) were significantly associated with slower cognitive decline. 31 Similarly, Clarke et al. reported a significant association between the presence of a public transit stop and a slower rate of cognitive decline (beta = 0.007, p‐value < 0.01) in the Chicago Health and Aging Project (1993–2011) involving 6,518 older adults. 32 These two findings suggest that accessible public transportation may be benefit for cognitive health. Our study examined a related but distinct indicator, and observed that better transportation status (assessed by private car ownership) was associated with a reduced risk of incident MCR risk (HR 0.51, 95% CI: 0.37–0.71, p‐value < 0.001). Collectively, these findings support the broader concept that improving transportation status may be a promising strategy for delaying pre‑dementia in older adults. Interestingly, better housing conditions (e.g., bathroom modifications) were paradoxically associated with a higher risk of incident MCR. The observed positive association between better housing and MCR risk is likely explained by reverse causation. Individuals with early or subclinical motor‐cognitive decline may proactively install home modifications to cope with emerging functional limitations. 33 , 34 , 35 Indeed, bathroom modifications (e.g., having grab bar, seat, and medical emergency system) are frequently reported among individuals with falls, poorer physical performance, and sensory impairment. 36 , 37 , 38 All of these conditions are need‐based adaptation, and have been recognized as established risk factors for MCR. 39 , 40 Thus, the association does not imply that better housing directly increases MCR risk; rather, it reflects that those who are already at higher risk tend to have more home modifications. Future studies with detailed measures of housing quality, lifestyle, and neighborhood characteristics are needed to further explore the mechanisms underlying this association. Additionally, in stratified analyses, we further found that the inverse association between AFE status and MCR risk was significantly stronger in individuals aged less than 75 years than in those aged 75 years or older (p‐interaction = 0.002). One possible explanation, though speculative, is that younger older adults may derive greater benefits from AFE due to higher environmental engagement and social participation, as suggested by previous studies, 41 , 42 whereas the oldest‐old may face physical or cognitive limitations that attenuate such effects. 43 , 44 , 45 , 46 Future studies with more comprehensive measures are needed to clarify the mechanisms underlying this age‐related difference. Strengths and limitations Our study has several strengths. To our knowledge, this is the first study to examine the association between AFE status and incident MCR. Our findings provide novel evidence for the potential benefit of AFE in preventing/delaying pre‐dementia syndrome. The large sample size with both two subsets enhanced the robustness and generalizability of the findings. Additionally, comprehensive stratified and sensitivity analyses across multiple subgroups all showed significant association, indicating the stability of the results. Last, the study examined eight individual AFE domains, and identifies several specific domains (e.g., transportation, social participation) that may contribute most strongly to MCR risk reduction, and then offer actionable targets for intervention. Nevertheless, limitations should be acknowledged. First, AFE status was assessed based on self‐reported information, which may introduce recall bias and misclassification. Second, due to the observational study design, causal associations between AFE status and MCR cannot be established. Third, AFE status was measured at a single time point, and subsequent temporal changes were not captured. Future studies with repeated assessments of AFE are warranted to address this limitation and enhance the reliability of findings. Fourth, although we adjusted for a wide range of covariates in our analyses, residual confounding cannot be entirely ruled out. Data on factors such as genetic predisposition, early‑life environmental exposures, and cognitive reserve were not available, and these may have influenced the observed associations. Fifth, the eight AFE domains may be intercorrelated, and the current analysis merely assigned equal weight to each domain. Hence, the relative contribution of individual domains to MCR risk reduction may be biased. Last, constrained by the data availability in NHATS, our AFE indicators were derived from existing NHATS items based on conceptual alignment with the WHO framework, rather than from a validated AFE scale. For example, private car ownership was used as a proxy for the transportation domain, reflecting access to personal mobility. However, this indicator does not capture public transit or walkability, both of which are important aspects of age‑friendly transportation. Future studies with dedicated AFE measures are needed to confirm our findings. CONCLUSIONS In this large‑scale, prospective cohort study, we examined the association of AFE status with MCR risk among older adults. We found that higher AFE scores were dose‐dependently associated with a decreased risk of incident MCR, independent of socioeconomic status, lifestyle factors, and chronic diseases. These findings suggest that creating AFE may be a feasible population‑based strategy to delay or prevent pre‑dementia syndromes. Future studies are warranted to validate these findings and explore causal mechanisms in diverse populations. AUTHOR CONTRIBUTIONS Hui Zhang and Xiangwei Li designed and conducted the research. Hui Zhang, Chongyu Ding, Yaqian Xu, Zixin Hu, Yi Li, Shuai Jiang, and Meng Hao analyzed the data and performed statistical analyses. Hui Zhang drafted the manuscript. Xiaofeng Wang, Jiucun Wang, Meng Hao and Li Xiangwei supervised the study. Hui Zhang and Xiangwei Li are primarily responsible for the final manuscript. All the authors read and approved the final manuscript. CONFLICT OF INTEREST STATEMENT The authors declare that they have no conflicts of interest. Author disclosures are available in the Supporting Information. CONSENT STATEMENT Informed consent was obtained from all participants or their proxies by the NHATS investigators. Our analysis of publicly available, deidentified data was exempt from the need for institutional review board approval. Supporting information Supporting Information Supporting Information