Relationship between adverse childhood experiences and motoric cognitive risk in Chinese older adults: a nationwide study Authors: Yan Wu, Yang Yang Published: 2026 ABSTRACT AbstractBackgroundWith the rapid advancement of aging, cognitive impairment has become an important factor contributing to the increase in medical costs for older adults. Motor Cognitive Risk (MCR) is a pre-dementia syndrome. Early identification and intervention of MCR’s risk factors may reduce the incidence of dementia. However, previous studies mainly focused on the contemporaneous risk factors of MCR and paid less attention to the long-term impact of adverse childhood experiences (ACEs) on health outcomes in later life. Thus, this study assesses the relationship between ACEs and MCR among Chinese older adults.MethodsWe adopted data from the 2014 Life History Survey and the 2015 follow-up wave of the China Health and Retirement Longitudinal Study (CHARLS). The study excluded participants who were under the age of 60 and had dementia and mobility impairments. ACEs were evaluated by three dimensions and 8 items. MCR was assessed by two single-item questions. Logistic regression was used to estimate the relationship between ACEs and MCR.ResultsThe final sample included 4,937 older adults. After adjusting for covariates, childhood neglect was related with higher odds of MCR (OR = 1.382, 95% CI: 1.099–1.738). Each additional ACEs were associated with a 16.5% higher odds of MCR (OR = 1.165, 95% CI: 1.038–1.307). In addition, participants who reported three ACEs were linked to a higher odds of MCR than those with no ACEs (OR = 2.050, 95% CI: 1.136–3.699).ConclusionThese findings suggested that childhood neglect and exposure to multiple ACEs were association with MCR in old age. The results highlight early adversity has positively relationship on MCR of older adults, which underscore the important of early ACEs screening and interventions across the lifespan.Keywords: Chinese older adults, Adverse childhood experiences, Motoric cognitive risk Text box 1. Contributions to literature• This study indicates the importance to distinguish the effects of child neglect from child abuse on cognitive decline of older adults.• Highlights the importance of improving the socioeconomic conditions of older adults to buffer the long-term health consequences caused by adverse childhood experiences.• Provides a simple and evidence-based tool of cumulative risk threshold (three adverse childhood experiences) that may be used to identify older adults at an increased risk of cognitive decline in clinical and community setting. Introduction As population is aging worldwide, health problems are increasingly prevalent among older adults. Cognitive impairment represents a major contributor to disability and rising medical costs among older adults. There were currently more than 50 million patients with cognitive impairment in worldwide [1]. In China, the number of people with dementia has reached 15.07 million in 2020 [2], and by 2030, the social and economic costs of dementia is expected to reach 114.2 billion US dollars [3]. These trends underscore the critical need for early and intervene in high-risk groups for cognitive impairment. Studies have showed that motoric cognitive risk (MCR) has become as a reliable predictor of progressive dementia [4], which refers to subjective memory complaint and slow walking speed that occur in the absence of mobility disorders and dementia [5]. Identifying and intervening the risk factors for MCR are the key to reducing the incidence of MCR. While previous research has primarily focused on contemporaneous risk factors of MCR chronic diseases [6–8]. However, the life course theory and the cumulative inequality theory suggested that early-life hardships increased the risk of developing health problems in later life [9]. This theoretical foundation has spurred growing advocacy for early-life interventions to promote healthy aging [10]. Adverse childhood experiences (ACEs) are predictors of poor health outcomes in later life [11], including disadvantaged socioeconomic status in childhood, childhood abuse, and childhood neglect [12]. Adverse childhood experiences have been widely proven to be association with the development of general cognitive function [13] in old age. However, prior studies overlooked ACEs’ potential influence on MCR, which captured the coordinated decline of both cognitive and motor functions. Fortunately, recent evidences have indicated a potential connection between ACEs and MCR. From a biological perspective, ACEs frequently stimulate the individual’s stress response system, delay the development of hippocampus, amygdala, prefrontal cortex and prefrontal limbic circuits, leading to reduced corresponding executive functions. Additionally, the reduction of brain/cognitive reserves and the damaging effect of apolipoprotein E on cells can all contribute to cognitive dysfunction [14, 15]. Long-term stress during childhood can impair the individual’s physiological control system for responding to the environment, leading to accelerate declines in motor functions [16]. From a socioeconomic perspective, individuals with ACEs are limited their access to healthcare and healthy food, leading to health disadvantages and cognitive dysfunction in old age [17]. Therefore, based on the above hypotheses, this study aims to explore the association between ACEs and the risk of MCR through the the China Health and Retirement Longitudinal Study (CHARLS) data. Method Participants We adopted data from the CHARLS, which employed a stratified sampling design to survey participants aged 45 and over in China, ensuring broad geographical and demographic representation. The study adopted data from the 2014 Life History Survey and f. The analytic sample was restricted to participants aged 60 or older. Those with dementia or disability were excluded, as well as data with missing values for ACEs, cognitive risks related to physical activity, and covariates. Finally, 4,937 samples were included. The data screening process is shown in (Fig. 1). Before this study began, all participants indicated informed consent. CHARLS was approved by the Biomedical Ethics Review Committee of Peking University (IRB00001052-11015). Flowchart for screening included participants Measure ACEs Based on prior researches [18–20], this study defined ACEs from three dimensions: disadvantaged socioeconomic status in childhood, childhood abuse, and childhood neglect. Table 1 showed the questions and response for the assessment of ACEs. Disadvantaged socioeconomic status in childhood was assessed by four items. Each of the four items was scored either 0 or 1 point according to the scoring criteria in (Table 1). A total score ranging from 0 to 4 was counted by the sum of the four items’ scores. Based on previous studies and the characteristics of the CHARLS dataset [20], a score of 0–3 was coded as 0, indicating that the participant did not have a disadvantaged socioeconomic status in childhood, and a score of 4 was coded as 1, indicating that the participant have a disadvantaged socioeconomic status in childhood. The total score of ACEs was the sum of three domain scores. The total score ranges from 0 to 3, with higher scores suggesting greater exposure. There have two reasons for adopting this domain-based cumulative score. First, it is consistent with with previous research findings that highlight three dimensions of ACEs, thereby reducing the multiple associations among different elements. Second, It provides a simple, clinically explainable metric for identifying older adults with risk of ACEs. Table 1The questions and response for the assessment of ACEsDimensionsQuestionnaire itemsDisadvantaged socioeconomic status in childhood(Responses were categorized as 0 = the total score is 0–3; 1 = the total score is 4)What was the highest level of education your mother received? (Responses were categorized as 0 = Non-illiterates; 1 = Illiterates)What was the highest level of education your father received? (Responses were categorized as 0 = Non-illiterates; 1 = Illiterates)Before you were 17 years old, how was your family’s economic situation compared to that of ordinary families in your community/village? (Responses were categorized as 0 = Much better than them, same as them and a little better than them; 1 = a little worse than them, much worse than them)Before you were 17 years old, was there a period when your family couldn’t have enough to eat? (Responses were categorized as 0 = No; 1 = Yes)Childhood abuse(Responses were categorized as 0 = the total score is 0; 1 = the total score is 1–2)When you were a child, did you get hit by your female caregiver? (Responses were categorized as 0 = rarely or never; 1 = often or sometimes)When you were a child, did your male caregiver hit you? (Responses were categorized as 0 = rarely or never; 1 = often or sometimes)Childhood neglect(Responses were categorized as 0 = the total score is 0; 1 = the total score is 1–2)Did your female caregivers often express their love for you when you were a child? (Responses were categorized as 0 = often, sometimes or rarely; 1 = never)Did your female caregiver spend a lot of energy taking care of you when you were a child? (Responses were categorized as 0 = a lot, some or a little; 1 = Not at all) The questions and response for the assessment of ACEs Disadvantaged socioeconomic status in childhood (Responses were categorized as 0 = the total score is 0–3; 1 = the total score is 4) Childhood abuse (Responses were categorized as 0 = the total score is 0; 1 = the total score is 1–2) Childhood neglect (Responses were categorized as 0 = the total score is 0; 1 = the total score is 1–2) MCR According to established criteria [21–23], MCR was defined as subjective memory complaint and slow walking speed that occur in the absence of mobility disorders and dementia. Subjective cognitive complaint was assessed by the single question, “How do you feel about your current memory?” If participants responded “Fair” or “Poor”, it indicated the presence of subjective cognitive complaints for participants [21, 23]. The participants walked at a normal speed on a 2.5 m surface twice and the walking time was measured twice and averaged. Walking speed was the distance divided by the time. Slow walking speed was defined as a walking speed that was ≥ 1 standard deviation (SD) or more below the age- and gender-specific mean [23]. Age groups were divided into < 75 years and ≥ 75 years. The thresholds are as follows: 0.639 m/s (for < 75 years males) and 0.491 m/s (for ≥ 75 years males), as well as 0.568 m/s (for < 75 years females) and 0.400 m/s (for ≥ 75 years females). Covariates Covariates included sociodemographic and health-related factors. Sociodemographic variables were gender (male/female), age, and marital status (married/cohabiting vs. single/separated/widowed). Health-related covariates included body mass index (BMI) and chronic diseases history regarding hypertension, dyslipidemia, diabetes, malignant tumor, chronic pulmonary disease, liver disease, heart disease, stroke, kidney disease, digestive system disease, emotional and mental disorder, arthritis or rheumatism, asthma (yes/no). Data analysis Data were extracted from the CHARLS dataset using R and analyzed using SPSS 25.0. Participants with complete data on all variables were included in the final analysis. Multicollinearity among independent variables was estimated by variance inflation factors (VIF). Categorical variables were described using frequencies and percentages [N (%)], and Continuous variables were described using mean and standard deviation (M ± SD). The comparisons in MCR between different groups are analyzed by χ² test or t-test. Logistic regression was used to analyze the odds ratio (OR) and 95% confidence interval (95% CI) of the association between ACEs and MCR. All data analysis was unweighted for the following two reasons. Frist, the 2014 Life History Survey has no sampling weights. Although the 2015 follow-up wave has cross-sectional weights, applying them only to the outcome would create inconsistent weighting across exposure and outcome. Second, this approach was justified for associational research [24–25] and aligns with prior CHARLS life history studies [26–27]. Results Descriptive information of sample characteristics As shown in Table 2, 14.4%, 26.8%, and 22.2% of the total population reported having experienced childhood neglect, childhood abuse, and disadvantaged socioeconomic status in childhood. 50.4%, 37.2%, 10.9%, and 1.5% of the total population reported had been exposed to at 0, 1, 2, and 3 ACEs. 87.7%, 13.4%, and 11.8% of the total population had subjective memory complaints, slow gait speed, and MCR. The majority of the participants were male (50.6%) and had spouses (82.1%). The three most prevalent chronic diseases of the participants were arthritis or rheumatism (66.8%), hypertension (68.3%), and digestive system diseases (78.8%). The BMI of the participants was 23.70 ± 9.846 and their age was 68.12 ± 6.227. Compared with older adults without MCR, those with MCR were more likely to experience childhood neglect (χ² = 9.095, p = 0.003) and have a higher cumulative ACEs score (χ² = 9.638, p = 0.022). They were older (t =-3.700, p < 0.001), female (χ² = 5.871, p = 0.015), and had higher prevalence of hypertension (χ² = 22.200, p<0.001), diabetes (χ² = 4.884, p = 0.027), stroke (χ² = 52.541, p<0.001), kidney disease (χ² = 4.442, p = 0.035), and arthritis or rheumatism (χ² = 11.505, p = 0.001). They were also more likely to have a risk of MCR. As shown in Table 3, the VIF values of all independent variables were all less than 2, indicating that there was no multicollinearity among these variables. Table 2Descriptive information of sample characteristicsVariablesOptionsTotalMCRχ²/tP valueNoYesChildhood neglectNo4227 (85.60)37514769.0950.003Yes710 (14.40)602108Childhood abuseNo3612 (73.20)31974151.4880.223Yes1325 (26.80)1156169Disadvantaged socioeconomic status in childhoodNo3839 (77.80)33964431.3880.239Yes1098(22.20)957141ACEs02488(50.40)22192699.6380.02211839(37.20)16122272536(10.90)46373374(1.50)5915Subjective congitive complaintsNo608(12.30)6080--Yes4329(87.70)3745584Slow gait speedNo4273(86.60)42730--Yes664(13.40)80584MCRNo4353(88.20)---Yes584(11.80)---Age-68.12 ± 6.22768.00 ± 6.21569.01 ± 6.248-3.700<0.001BMI-23.70 ± 9.84623.64 ± 8.95224.12 ± 14.912-1.0980.272GenderMale2498(50.60)22302685.8710.015Female2439(49.40)2123316Marital statusHave spouses4053(82.10)35844691.4380.231No spouses884(17.90)769115HypertensionNo3371(68.30)302234922.200<0.001Yes1566(31.70)1331235DyslipidemiaNo4314(87.40)38065080.0940.760Yes623(12.60)54776DiabetesNo4537(91.90)40145234.8840.027Yes400(8.10)33961Malignant tumorNo4890(99.00)43105800.5010.479Yes47(1.00)434Chronic pulmonary diseaseNo4373(88.60)38635101.0180.313Yes564(11.40)49074Liver diseaseNo4781(96.80)42115701.2590.262Yes156(3.20)14214Heart diseaseNo4151(84.10)36764753.7250.054Yes786(15.90)677109StrokeNo4837(98.00)428854952.541<0.001Yes100(2.00)6535Kidney diseaseNo4629(93.80)40935364.4420.035Yes308(6.20)26048Digestive system diseaseNo3889(78.80)34474423.7760.052Yes1048(21.20)906142Emotional and mental disorderNo4893(99.10)43165770.7090.400Yes44(0.90)377Arthritis or rheumatismNo3299(66.80)294535411.5050.001Yes1638(33.20)1408230AsthmaNo4694(95.10)41355590.5820.446Yes243(4.90)21825Data were described by frequency (proportion) or mean±standard deviation. The comparisons in MCR between different groups are analyzed by χ² test or t-testAbbreviations: ACEs Adverse childhood experiences, BMI Body mass index, MCR Motor cognitive risk Descriptive information of sample characteristics Data were described by frequency (proportion) or mean±standard deviation. The comparisons in MCR between different groups are analyzed by χ² test or t-test Abbreviations: ACEs Adverse childhood experiences, BMI Body mass index, MCR Motor cognitive risk Table 3Multicollinearity diagnostics of variablesVariablesToleranceVIFACEs0.9851.016Gender0.9341.071Age0.8921.121Marital status0.8891.124BMI0.9771.024Hypertension0.8931.120Dyslipidemia0.8711.148Diabetes0.9191.088Malignant tumor0.9971.003Chronic pulmonary disease0.7841.276Liver disease0.9651.036Heart disease0.8761.142Stroke0.9741.026Kidney disease0.9381.066Digestive system disease0.9171.090Emotional and mental disorder0.9891.011Arthritis or rheumatism0.9251.082Asthma0.8041.244Abbreviations: VIF Variance inflation factors, ACEs Adverse childhood experiences, BMI Body mass index Multicollinearity diagnostics of variables Abbreviations: VIF Variance inflation factors, ACEs Adverse childhood experiences, BMI Body mass index Association between the types of ACEs and MCR Before controlling for covariates (gender, age, marital status, BMI, chronic disease history), the three dimensions of ACEs (disadvantaged socioeconomic status in childhood, childhood abuse, and childhood neglect) were used as independent variables, and MCR was used as the dependent variable for regression analysis. Older adults who experienced childhood neglect had about 1.4 times the risk of being diagnosed with MCR compared to those who did not (model 1: OR = 1.414, 95% CI: 1.128–1.772), even after adjusting for covariates (model2: OR = 1.382, 95% CI: 1.099–1.738). Disadvantaged socioeconomic status in childhood and childhood abuse were not associated with MCR. Table 4 showed the above results. Table 4Relationship between the types of ACEs and MCRACEsModel 1Model 2OR (95%CI) P OR (95%CI) P Disadvantaged socioeconomic status in childhood1.129(0.922–1.383)0.2391.151(0.885–1.338)0.424Childhood abuse1.126(0.930–1.363)0.2231.143(0.940–1.391)0.181Childhood neglect1.414(1.128–1.772)0.0031.382(1.099–1.738)0.006Model 2 has been adjusted for the covariates (gender, age, marital status, body mass Index, chronic disease history)Abbreviations: ACEs Adverse childhood experiences, MCR Motoric cognitive risk, OR Odds ratios, CI Confidence Interval Relationship between the types of ACEs and MCR Model 2 has been adjusted for the covariates (gender, age, marital status, body mass Index, chronic disease history) Abbreviations: ACEs Adverse childhood experiences, MCR Motoric cognitive risk, OR Odds ratios, CI Confidence Interval The association between the exposure intensity of ACEs and MCR Before adjusting for covariates (gender, age, marital status, BMI, chronic disease history), with the total score of ACEs as a continuous variable, for every additional ACE, the odds of MCR were higher by 17.8% (model1: OR = 1.178, 95% CI: 1.052–1.320), even after adjusting for covariates (model2: OR = 1.165, 95% CI: 1.038–1.307). Compared with those who had not experienced ACEs, the risk of older adults being diagnosed with MCR who had experienced three ACEs increased (model1: OR = 2.097, 95% CI: 1.173–3.748), and this was still the case even after adjusting for the covariates (model2: OR = 2.050, 95% CI: 1.136–3.699). However, only about 1.5% of participants (n = 74) experienced three ACEs (see Table 2), this result should be interpreted with caution due to limited statistical power and potential instability. Detailed results were shown in (Table 5). Table 5Association between the exposure intensity of ACEs and MCRACEsModel 1Model 2OR (95%CI) P OR (95%CI) P ACEs1.178(1.052–1.320)0.0051.165(1.038–1.307)0.0101111.162(0.963–1.402)0.1181.151(0.951–1.393)0.1521.301(0.986–1.716)0.0631.264(0.952–1.677)0.10532.097(1.173–3.748)0.0122.050(1.136–3.699)0.017Model 2 was adjusted for the covariates (gender, age, marital status, body mass Index, chronic disease history)Abbreviations: ACEs Adverse childhood experiences, MCR Motoric cognitive risk, OR Odds ratios, CI Confidence Interval Association between the exposure intensity of ACEs and MCR Model 2 was adjusted for the covariates (gender, age, marital status, body mass Index, chronic disease history) Abbreviations: ACEs Adverse childhood experiences, MCR Motoric cognitive risk, OR Odds ratios, CI Confidence Interval Discussion This study confirmed the relationship between ACEs and MCR in a national sample of Chinese older adults. The results supported the study hypothesis, indicating that childhood neglect and a higher cumulative burden of ACEs were associated with a increased risk of MCR. These findings emphasized the potential value of screening for adversity at an early stage and the importance of improving socioeconomic conditions throughout life to mitigate its long-term health consequences. The results showed that childhood neglect was linked to MCR. Previous studies have indicated that individuals who experienced childhood neglect were found to have lower cognitive resource utilization [28]. According to the cumulative disadvantage theory [29], the disadvantages that individuals encounter in early life can continue to accumulate as they progress through life. Based on this, it can be speculated that cognitive impairments during childhood accumulate over time and be associated with cognitive functions in old age. Therefore, policies and social services should strengthen support for families at risk of childhood neglect, establish early identification and intervention mechanisms, such as conducting regular screenings through schools, communities, and medical institutions, and providing educational publicity, psychological assistance, and child development resources for families with neglect behaviors, These measures may be reduce the potential harm of childhood neglect at its source. In contrast, disadvantaged socioeconomic status in childhood and childhood abuse were not related to the occurrence of MCR in this study. Three possible reasons may account for these results. First, previous study indicated that the negative impact of childhood socioeconomic disadvantage on health may be buffered by a good living standard and medical care in old age. The adverse circumstances of childhood did not have to be irreversible [30]. Individuals can obtain resources and opportunities through a higher social economic status in adulthood to reduce the negative effects of the early adverse situations [31]. This suggests that improving the social economic conditions in adulthood may effectively alleviate ACEs’ negative effects. Therefore, public policies should strive to create more fair opportunities for social mobility, strengthen lifelong education, vocational training, and the construction of social security systems, helping individuals to compensate for the early disadvantages. Second, the retrospective self-reporting assessment of ACEs may have recall bias and social expectation deviation. Moreover, influenced by the traditional concept of “beating produces filial piety”, older adults do not consider parental beating as physical abuse [32]. These reasons may weaken the influence of ACEs on MCR. Therefore, future research should adopt objective, multi-source and cultural adapted measurement tools for ACEs. Third, the results may be an illusion caused by the selective mortality [33]. Those frail individuals with sever ACEs may have died before entering old age, while those who survived were physically strong older adults, which may weaken the negative impact of ACEs on MCR. This also reinforces the necessity of conducting health promotion and intervention throughout the entire life stage. This study also revealed the dose-response relationship between the intensity of ACEs exposure and the risk of MCR. Individuals with three ACEs had a significantly higher risk than those without such experiences, which supports the use of a cumulative risk model for ACEs screening. In practical applications, individuals with high ACEs scores can be prioritized for early assessment and intervention in community and clinical settings. However, the small sample size of three ACEs group (n = 74) limited the precision of this estimate, and resulted in statistical instability. Future studies needed to adopted large sample size or meta analysis to validate the findings. Limitations Although this study has significant findings, some limitations still need to be considered. Firstly, The results of this study were sample-dependent, thereby may limit the generalizability of these findings. On the one hand, the self-report results of ACEs might be influenced by the recall bias and social expectation bias, which might reduce the accuracy of measurement results of ACEs. On the other hand, the cutoffs for slow gait speed was obtained from a specific sample rather than validated measurement, which might result in these thresholds not being applicable to other populations with different demographic characteristics. Therefore, the sample-dependent of these findings highlight the need for future studies to use objective assessment for ACEs and validated measurement for MCR. Second, the cross-sectional design precludes causal inference. Prospective life-course studies tracking individuals from childhood are needed to clarify these associations. Thirdly, while key covariates were adjusted for, residual confounding by unmeasured factors remains possible. Finally, although a three-domain scores ranging from 0 to 3 was adopted to simplify the measurement of ACE, it inevitably resulted in loss of variance compared to a complete item-level cumulative score. Future studies could adopt a more comprehensive ACEs scoring system to capture dose-response patterns across a wider range of exposures. Conclusion This study indicated that childhood neglect and exposure to three types of ACEs were association with a higher odds of MCR in old age. These findings demonstrated that early screening for ACEs and creating a favorable socioeconomic environment throughout one’s life had a potential for promoting cognitive and motor health in older adults. Acknowledgements Not applicable. Abbreviations Motoric cognitive risk Adverse childhood experiences Odds ratio 95% Confidence interval China Health and Retirement Longitudinal Study Body mass index Standard deviation Variance inflation factors Authors' contributions Yan Wu: conceptualization, data curation, methodology, formal analysis, visualization, writing, reviewing and editing. Yang Yang: conceptualization, data curation, methodology, supervision, writing, reviewing, and editing. Funding This work was supported by the General Project of Nanjing Medical University Science and Technology Development Fund in 2024 [grant number NMUB20240216]. Data availability The dataset supporting the conclusions of this article is available in the CHARLS repository at http://charls.pku.edu.cn/en. Declarations Ethics approval and consent to participate The study was conducted in accordance with the principles of the Helsinki Declaration and approved by the Biomedical Ethics Review Committee of Peking University (IRB00001052-11015). All participants gave their informed consent in this study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yan Wu and Yang Yang contributed equally to this work and are co-first authors. References Associated Data Data Availability Statement The dataset supporting the conclusions of this article is available in the CHARLS repository at http://charls.pku.edu.cn/en.