Uric acid influences the relationship between unsafe neighborhood and digestive diseases among middle-aged or older adults: a longitudinal cohort study from the China Health and Retirement Longitudinal Study (CHARLS) Authors: Jiayin Du, Shuting Yang, Chao Qu, Lindi Cai, Xuetao Yuan, Junxiang Zhang, Xuqi Li, Tianli Shen Published: 2026 ABSTRACT AbstractBackgroundAdverse childhood experiences (ACEs) are related to an increased risk of gastrointestinal disorders. However, whether uric acid (UA) affects the development of digestive diseases and its role in the link between ACEs and digestive diseases have not been fully studied.MethodsAll data for this longitudinal cohort study came from the China Health and Retirement Longitudinal Study (CHARLS). Cox proportional hazards regression model was applied to assess the relationship between ACEs or UA and digestive diseases and the mediating effect of UA between ACEs and digestive diseases. Lastly, subgroup analyses were performed by demographic characteristics and health status grouping.ResultsThe study included 3666 volunteers, of whom 617 had digestive disorders. Using Cox proportional hazards regression model, we analyzed the association between ACE and digestive diseases, and found that household mental illness, and living in an unsafe neighborhood were all obviously associated with disease incidence. Studies have also demonstrated an association between UA and digestive disorders. In the mediation analysis, we found that unsafe neighborhood had a negative indirect effect on digestive system diseases through UA, and this mediating effect was shown to have comparable results in the subgroup analyses.ConclusionsWe found evidence that ACEs and UA levels are associated with the development of digestive diseases, and that UA levels influence the relationship between one of the ACEs, namely unsafe neighborhood and digestive disease. Further research should address the underlying mechanisms and interventions to prevent childhood adversities and their devastating effects on health.Graphical abstract Abbreviations: CHARLS, China Health and Retirement Longitudinal Study; DAG, directed acyclic graph; ACE, adverse childhood experience; UA, uric acid; BMI, body mass index; HPA, hypothalamic–pituitary–adrenal. Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28154-x.Keywords: CHARLES, Adverse childhood experiences, Uric acid, Digestive disease, Middle-aged and older adults Introduction Adverse childhood experiences (ACEs) are stressful or traumatic events that occur during childhood. In previous studies, ACEs have been assessed using different scales. One study included emotional abuse, physical abuse, sexual abuse, battered mother, mental illness in household, household substance abuse, parental separation or divorce, and incarcerated household members [1], while another study reported seven types of negative experiences [2]. At present, two assessment methods are widely used internationally—the Childhood Trauma Questionnaire Short Form (CTQ-SF) and the Adverse Childhood Experiences International Questionnaire (ACE-IQ)—and internal consistency and correlation between the two have been demonstrated [3]. Combining the evaluation of ACEs from different studies and questionnaire information from the China Health and Retirement Longitudinal Study (CHARLS), we provided a definition of ACEs that comprises 7 traditional items, 3 community-level items, and 2 new items [4]. At least 50% of American adults are estimated to have experienced one or more ACEs before the age of 18 [5]. Importantly, a growing body of research has demonstrated the adverse effects of ACEs on lifelong health outcomes [6]. Many studies have indicated that exposure to ACEs is linked to both mental and physical health issues in adulthood, including autism [7], Alzheimer’s disease [8], opioid abuse [9], and chronic diseases [4]. In terms of ACEs and chronic diseases, Li et al. proposed that ACEs are associated with chronic lung disease, with the effect being mediated by depression [10], while Pape et al. focused on the association between ACEs and asthma [11]. In the study of Lin et al., the cumulative ACE score correlates with increased odds of having various chronic diseases, among which the correlation with diseases of the digestive system is significant. Despite this finding, the relationship between ACE and digestive system diseases has not received further attention, and no other variables have been introduced [4]. Digestive diseases encompass diseases of the digestive tract and digestive glands and have been reported to lead to millions of health care visits and thousands of deaths in the United States each year, resulting in billions of dollars in health costs [12]. Common digestive disorders include gastroesophageal reflux, irritable bowel syndrome, intestinal infections, and colon polyps [13]. These diseases tend to share many risk factors, such as age, sex, alcohol consumption and smoking, with ACEs also influencing the risk of developing a disorder [14, 15]. Uric acid (UA) is a metabolic product of the human body, and regulation of its level is influenced by various factors, such as dietary structure and lifestyle [16]. Studies have reported that adverse experiences during childhood can also increase uric acid levels, leading to metabolic abnormalities [17]. Additionally, its role has long been controversial. On the one hand, high uric acid levels can become a potential risk factor for many diseases, including hypertension and cardiovascular diseases [18, 19], and on the other hand, the beneficial effects of UA on the human body have been confirmed [20]. Many studies have shown a broad association between UA and digestive disorders. Wu et al. reported that UA plays a key regulatory role in the evolution of colitis [21] and Liu et al. described a causal relationship between UA and intestinal bacteria using Mendelian randomization analysis [22]. We observed that UA may function as a potential link in the association between ACEs and digestive system diseases. However, no studies have linked UA to a wide range of digestive disorders or performed statistical analyses to investigate the interaction relationships among ACEs, UA, and digestive system diseases. In this longitudinal cohort study conducted in China, we used the CHARLS database to explore the relationship between ACE exposure or UA and digestive diseases and their causal relationship. Methods Study population The study involved volunteers from CHARLS, which covers approximately 450 villages or resident communities across 28 provinces in China and is designed to acquire a large range of high-quality microdata on middle-aged and older adults. The contents of this database have been reported in detail elsewhere [23]. In simple terms, the study began collecting questionnaires from participants in 2011, and since then, participants have been followed up every two years and new participants have been recruited; thus far, four follow-up visits have been conducted. Previously enrolled participants were surveyed in 2014 about their life histories, including childhood experiences, and only the 2011 and 2015 surveys included blood test data. In this longitudinal study, participant tracking was conducted using unique identifiers across four waves of data collection, spanning from the baseline year of 2011 to the final follow-up in 2018. There were 17,708 participants in the 2011 baseline study and 20,452 participants in the 2014 life history survey. Through systematic exclusion criteria, we retained 3,666 eligible participants after removing: (1) cases with incomplete ACE data; (2) individuals with pre-existing digestive disorders at baseline or missing disease status; (3) subjects lost to follow-up during the 7-year observation period; (4) participants with missing blood test parameters; and (5) individuals aged < 45 years or with undocumented age information. We obtained informed consent from each interviewee, and CHARLS has gained ethical clearance from the Institutional Review Board of Peking University. Assessment of ACEs We extracted 12 ACEs from the CHARLS database: physical abuse, emotional neglect, household mental illness, household substance abuse, domestic violence, incarcerated household member, parental separation or divorce, bullying, unsafe neighborhood, parental death, sibling death, and parental disability. The detailed methods of data collection have been reported previously [4]. ACEs are presented as multicategorical variables and divided into five groups based on the cumulative ACE score: 0, 1, 2, 3, and 4 or higher. Assessment of digestive disease In this study, the occurrence of digestive diseases was classified as a binary variable based on the response to a question in the CHARLS questionnaire, “Have you been diagnosed with stomach or other digestive disease (except for tumor or cancer) by a doctor?” By tracing respondent records across all data from 2011, 2013, 2015 and 2018, the outcomes to this question were classified as “1” or “0” depending on whether or not, respectively, the respondents reported that they were told by their doctor that they had this condition. Covariates We included a number of potential confounding factors as covariates in this study: age, sex, residence (rural or urban), education status (“elementary school or below” or “middle school or above”), marital status (“single, divorced, and widowed” or “married but living without a spouse” or “married and living with a spouse”), smoking status (“smoker” or “non-smoker”), drinking status (“drink more than once a month” or “drink less than once a month” or “non-drinker”), body mass index (BMI), hypertension (“yes” or “no”), diabetes (“yes” or “no”), and selective blood test indicators [24]. BMI was calculated by dividing weight by the square of height in international units. Hypertension was diagnosed using a dual-criterion system: (1) objective measurement defined as an average systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg, calculated from three consecutive readings obtained during a single visit using an Omron HEM-7200 digital sphygmomanometer following CHARLS standardized protocols, with physiologically implausible values (SBP > 300 mmHg or DBP > 200 mmHg) excluded prior to averaging; or (2) subjective confirmation via participant self-report of physician-diagnosed hypertension [25]. Diabetes mellitus was diagnosed according to the American Diabetes Association criteria, requiring the fulfillment of at least one of the following conditions: Self-reported physician-diagnosed diabetes, or fasting blood glucose (FBG) ≥ 7.0 mmol/L, or random plasma glucose ≥ 11.1 mmol/L, or glycated hemoglobin (HbA1c) ≥ 6.5% [26]. Blood test data were obtained directly from the CHARLS database. Statistical analysis We proceeded descriptive analysis, with continuous variables depicted by the mean ± standard deviation (SD) and categorical variables described by count (percentage). The differences in continuous and categorical variables between participants with and without a diagnosis with digestive disease were tested using the nonparametric rank sum test and chi-squared test, respectively. We utilized the Cox proportional hazards regression model to calculate the hazard ratio (HR) and 95% confidence interval (CI) for the association between ACE, UA, and digestive system diseases, and analyzed the trend relationship between the cumulative ACE score, an ordered multiple categorical variables, and presence of digestive disease to evaluate potential dose–response correlations. In the study, we established three models by continuously adjusting the confounding factors, in order to gradually reveal these relationships. Model 1 was a crude model without any adjustments; in model 2, we adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, and diabetes; and in model 3, we adjusted for the levels of low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) based on the results from model 2. We incorporated ACEs that demonstrate significant associations with digestive diseases, into our analytical model to assess mediating effects using the fully adjusted model. In addition, participants were stratified into subgroups by age, gender, residence, educational status, marital status, BMI, smoking status, drinking status, hypertension, diabetes, and UA levels, and then mediation analyses were conducted within each subgroup to explore potential effect modifications. For grouping, cutoff values were set at 65 years for age, 25 kg/m² for BMI, and 7 mg/dl for UA [27–29]. All statistical analyses were performed employing the R software version 4.3.3, and missing value for several covariates were interpolated using the “mice” package. All P-values were two-sided, and statistical significance was set at P < 0.05. Results Descriptive statistics The study participants were 3666 adults aged 45 years and older. The flowchart in Fig. 1 explains the selection process. A summary of the participants’ basic characteristics was given in Table 1. The average age was 61.26 years (SD, 8.35), with 1751 men (47.76%) and 1915 women (52.24%). A total of 617 (16.83%) participants were diagnosed with digestive diseases according to our definitions, with younger age, female sex, lower BMI, and lower educational attainment being more highly associated with a diagnosis. More than four in five of the participants (80.14%) reported at least one adverse event, with 29.60% reporting 1 ACE, 25.07% reporting 2 ACEs, 15.58% reporting 3 ACEs, and 9.90% reporting 4 + ACEs. Of the reported experiences, the occurrence rates of physical abuse, emotional neglect, and sibling death exceeded 25% (27.11%, 31.12%, and 28.83%, respectively). Flow Chart of the Screening Process for the Selection of Eligible Participants in CHARLS 2011–2018. Abbreviations: CHARLS, the China Health and Retirement Longitudinal Study; ACE, adverse childhood experience Table 1Basic characteristics of the participantsCharacteristicsaTotal (n = 3666)Non-digestive disease (n = 3049)Digestive disease (n = 617)P-valueAge, years61.26 ± 8.3561.19 ± 8.3761.61 ± 8.240.025BMI, kg/m223.96 ± 3.8124.00 ± 3.7723.81 ± 4.000.023Sex< 0.001 Male1751 (47.76)1505 (49.36)246 (39.87) Female1915 (52.24)1544 (50.64)371 (60.13)Residence0.073 Rural3306 (90.18)2737 (89.77)312 (10.23) Urban360 (9.82)312 (10.23)48 (7.78)Marital status0.678 Married and living with a spouse3120 (85.11)2602 (85.34)518 (83.95) Married but living without a spouse154 (4.20)126 (4.13)28 (4.54) Single, divorced, and windowed392 (10.69)321 (10.53)71 (11.51)Education Status0.001 Elementary school or blow2365 (64.51)1931 (63.33)434 (70.34) Middle school or above1301 (35.49)1118 (36.67)183 (29.66)Smoking Status0.267 Non-smoker1628 (44.41)1367 (44.83)261 (42.30) Smoker2038 (55.59)1682 (55.17)356 (57.70)Drinking Status0.131 Non-drinker2218 (60.50)1823 (59.79)395 (64.02) Drink less than once a month431 (11.76)368 (12.07)63 (10.21) Drink more than once a month1017 (27.74)858 (28.14)159 (25.77) Low density lipoproteincholesterol level, mg/dl115.47 ± 34.77115.26 ± 34.32116.50 ± 36.910.003 High density lipoproteincholesterol level, mg/dl50.32 ± 14.9950.07 ± 14.9351.52 ± 15.230.038 Total Cholesterol (mg/dl)192.32 ± 37.66192.05 ± 37.44193.65 ± 38.740.011 Triglyceride level, mg/dl134.92 ± 99.25136.13 ± 102.15128.95 ± 83.270.019 C-reactive protein level, mg/l2.48 ± 6.012.40 ± 5.272.86 ± 8.770.022 Uric acid level, mg/dl4.45 ± 1.244.48 ± 1.254.29 ± 1.190.058Hypertension0.546 No2241 (61.13)1871 (61.36)370 (59.97) Yes1425 (38.87)1178 (38.64)247 (40.03)Diabetes0.439 No3066 (83.63)2543 (83.40)523 (84.76) Yes600 (16.37)506 (16.60)94 (15.24)Physical abuse0.474 No2672 (72.89)2230 (73.14)442 (71.64) Yes994 (27.11)819 (26.86)175 (28.36)Emotional neglect0.116 No2525 (68.88)2117 (69.43)408 (66.13) Yes1141 (31.12)932 (30.57)209 (33.87)Householdsubstance abuse0.831 No3376 (92.09)2806 (92.03)570 (92.38) Yes290 (7.91)243 (7.97)47 (7.62)Household mentalillness< 0.001 No3256 (88.82)2732 (89.60)524 (84.93) Yes410 (11.18)317 (10.40)93 (15.07)Domestic violence0.300 No3374 (92.03)2813 (92.26)561 (90.92) Yes292 (7.97)236 (7.74)56 (9.08)Incarceratedhousehold member0.629 No3658 (99.78)3043 (99.80)615 (99.68) Yes8 (0.22)6 (0.20)2 (0.32)Parental separationor divorce0.897 No3650 (99.56)3035 (99.54)615 (99.68) Yes16 (0.44)14 (0.46)2 (0.32)Unsafeneighborhood0.053 No3382 (92.25)2825 (92.65)557 (90.28) Yes284 (7.75)224 (7.35)60 (9.72)Bullying0.801 No3199 (87.26)2663 (87.34)536 (86.87) Yes467 (12.74)386 (12.66)81 (13.13)Parental death0.648 No3076 (83.91)2554 (522)522 (84.60) Yes590 (16.09)495 (16.23)95 (15.40)Sibling death0.086 No2609 (71.17)2188 (71.76)421 (68.23) Yes1057 (28.83)861 (28.24)196 (31.77)Parental disability0.422 No2928 (79.87)2443 (80.12)485 (78.61) Yes738 (20.13)606 (19.88)132 (21.39)Number of ACEs0.005 0728 (19.86)632 (20.73)96 (15.56) 11085 (29.60)908 (29.78)177 (28.69) 2919 (25.07)740 (24.27)179 (29.01) 3571 (15.58)479 (15.71)92 (14.91) ≥ 4363 (9.90)290 (9.51)73 (11.83)Abbreviations: ACE adverse childhood experience, BMI body mass indexaFor continuous variables, numbers represent the mean ± standard deviation, and for categorical variables, numbers represent count (%) Basic characteristics of the participants Low density lipoprotein cholesterol level, mg/dl High density lipoprotein cholesterol level, mg/dl Household substance abuse Household mental illness Incarcerated household member Parental separation or divorce Unsafe neighborhood Abbreviations: ACE adverse childhood experience, BMI body mass index aFor continuous variables, numbers represent the mean ± standard deviation, and for categorical variables, numbers represent count (%) Association between ACE, UA and digestive disease Table 2 presents the data regarding the association between single ACEs and occurrence of digestive disease. As shown in Table 2, household mental illness and living in an unsafe neighborhood correlated with reports of digestive disorders, and the association persisted under all models. Furthermore, there is a significant dose-response relationship between the cumulative ACE score and digestive diseases. (Table 3). And, Table 4 demonstrates a robust correlation between UA and digestive system diseases. The HR (95% CI) before and after adjusting for confounding factors were 0.89 (0.84–0.93), 0.94 (0.89–0.99), and 0.94 (0.89–0.99), respectively. Additionally, correlation analysis across varying UA levels revealed that at normal UA concentrations, UA remained significantly associated with the outcome variable (Model 3, p = 0.034). In contrast, at elevated UA levels, this association did not reach statistical significance. Table 2Association between ACE and digestive disease in adulthoodACEHR (95% CI)P-valuePhysical abuse Model 1a1.07 (0.90–1.28)0.424 Model 2b1.09 (0.92–1.31)0.317 Model 3c1.10 (0.92–1.32)0.230Emotional neglect Model 11.16 (0.98–1.37)0.090 Model 21.15 (0.97–1.36)0.106 Model 31.14 (0.97–1.35)0.122Householdsubstance abuse Model 10.97 (0.72–1.30)0.829 Model 20.97 (0.72–1.31)0.845 Model 30.97 (0.72–1.31)0.845Household mentalIllness Model 11.47 (1.18–1.84)< 0.001*** Model 21.47 (1.18–1.83)< 0.001*** Model 31.45 (1.17–1.81)< 0.001***Domestic violence Model 11.19 (0.90–1.56)0.220 Model 21.19 (0.91–1.57)0.211 Model 31.20 (0.91–1.58)0.195Incarceratedhousehold member Model 11.49 (0.37–5.99)0.571 Model 21.47 (0.37–5.91)0.587 Model 31.45 (0.36–5.84)0.598Parental separationor divorce Model 10.73 (0.18–2.90)0.649 Model 20.74 (0.18–2.97)0.672 Model 30.73 (0.18–2.94)0.662Unsafeneighborhood Model 11.32 (1.01–1.73)0.040* Model 21.32 (1.01–1.72)0.044* Model 31.31 (1.00–1.70)0.050*Bullying Model 11.04 (0.82–1.31)0.748 Model 21.05 (0.83–1.33)0.689 Model 31.06 (0.84–1.34)0.639Parental death Model 10.95 (0.76–1.18)0.630 Model 20.94 (0.75–1.17)0.555 Model 30.94 (0.75–1.17)0.551Sibling death Model 11.15 (0.97–1.37)0.097 Model 21.14 (0.97–1.36)0.121 Model 31.14 (0.96–1.35)0.141Parental disability Model 11.10 (0.90–1.33)0.356 Model 21.10 (0.91–1.33)0.345 Model 31.10 (0.91–1.33)0.327Abbreviations: ACE adverse childhood experience, HR hazard ratio, 95% CI 95% confidence intervalsa Model 1 was the crude modelb Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetesc Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels*P < 0.05; **P < 0.01; ***P < 0.001 Association between ACE and digestive disease in adulthood Household substance abuse Household mental Illness Incarcerated household member Parental separation or divorce Unsafe neighborhood Abbreviations: ACE adverse childhood experience, HR hazard ratio, 95% CI 95% confidence intervals a Model 1 was the crude model b Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetes c Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels *P < 0.05; **P < 0.01; ***P < 0.001 Table 3Association between number of ACEs and reports of digestive disease in adulthoodModelsHR (95% CI) by No. of ACEsP value for trend0123≥4Model 1a1[Reference]b1.26 (0.98–1.62)1.53 (1.19–1.96)1.25 (0.94–1.67)1.60 (1.18–2.18)0.004**Model 2c1[Reference]1.27 (0.99–1.63)1.53 (1.19–1.96)1.27 (0.95–1.69)1.59 (1.17–2.16)0.005**Model 3d1[Reference]1.26 (0.98–1.61)1.52 (1.18–1.95)1.26 (0.95–1.68)1.56 (1.14–2.12)0.007**Abbreviations: ACE adverse childhood experience, HR hazard ratio, 95% CI 95% confidence intervalsa Model 1 was the crude modelb Reference: No ACE exposurec Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetesd Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels*P < 0.05; **P < 0.01; ***P < 0.001 Association between number of ACEs and reports of digestive disease in adulthood Abbreviations: ACE adverse childhood experience, HR hazard ratio, 95% CI 95% confidence intervals a Model 1 was the crude model b Reference: No ACE exposure c Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetes d Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels *P < 0.05; **P < 0.01; ***P < 0.001 Table 4Association between UA and digestive disease in adulthoodUAHR (95% CI)P-valueTotal Model 1a0.89 (0.84–0.93)< 0.001*** Model 2b0.94 (0.89–0.99)0.015* Model 3c0.94 (0.89–0.99)0.017*UA ≤ 7 Model 10.88 (0.83–0.93)< 0.001*** Model 20.94 (0.88–0.99)0.037* Model 30.94 (0.88–0.99)0.034*UA > 7 Model 10.76 (0.42–1.36)0.352 Model 20.69 (0.37–1.31)0.257 Model 30.62 (0.33–1.16)0.134Abbreviations: UA uric acid, HR hazard ratio, 95% CI 95% confidence intervalsa Model 1 was the crude modelb Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetesc Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels*P < 0.05; **P < 0.01; ***P < 0.001 Association between UA and digestive disease in adulthood Abbreviations: UA uric acid, HR hazard ratio, 95% CI 95% confidence intervals a Model 1 was the crude model b Model 2 was adjusted for age, sex, marital status, educational status, residence, body mass index (BMI), smoking and drinking status, hypertension, diabetes c Model 3 was adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels *P < 0.05; **P < 0.01; ***P < 0.001 The mediating effect of UA in the relationship between ACE and digestive system diseases When examining the mediating effect of UA between ACE, including household mental illness and unsafe neighborhood, as well as the cumulative scores of ACE, and digestive diseases, we found that only in the unsafe neighborhood model UA had a significant mediating effect, with the mediation proportion being − 9.74%, as shown in Fig. 2. After conducting subgroup analyses, it was found that the data of the main strata defined by age, gender, residence, educational status, marital status, BMI, smoking status, hypertension, and diabetes, showed comparable results, and the mediating percentage ranged from − 35.20% to 82.01%, as shown in Fig. 3. However, in the models of household mental illness and the cumulative scores of ACE there was no significant mediating effect and this was not observed in the subsequent subgroup analyses either as shown in Fig S1, S2. Pathway Diagrams of Unsafe Neighborhood, UA and Digestive Diseases. Abbreviations: ACE, adverse childhood experience; UA, uric acid. Adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels. *P < 0.05; **P < 0.01; ***P < 0.001 Mediating Effect of UA in the Association Between Unsafe Neighborhood with Digestive Diseases. Abbreviations: ACE, adverse childhood experience; 95% confidence intervals. Adjusted for age, sex, marital status, educational status, residence, BMI, smoking and drinking status, hypertension, diabetes, and low density lipoprotein cholesterol (LDL), high density lipoprotein cholesterol (HDL), total cholesterol (CHO), triglyceride (TG), C-reactive protein (CRP) levels. *P < 0.05; **P < 0.01; ***P < 0.001 Discussion ACEs have been a hot topic of research for a long time and their profound impact on people’s later health is acknowledged. Although strong associations between cumulative ACE scores and various chronic disorders have been found [30], as well as significant relationships between specific ACEs and chronic diseases as a whole [4], a detailed classification of ACEs and digestive diseases has not been reported. This is the first study to assess the association between singular ACEs and digestive disease, and to examine the mediating role of UA in this association. In summary, we identified the following relationships: First, we found robust associations between household mental illness, unsafe neighborhood, and cumulative ACE scores with prevalence of digestive disease. We also found that UA had a protective effect against digestive disease, and this protective effect only occurs when the UA level is normal, but not when it is high. Among all the ACEs we studied, UA only had a negative mediating effect between unsafe neighborhood and digestive diseases. Moreover, in the subgroup analysis, the mediating role of UA within the model of unsafe neighborhood has yielded consistent findings. Specifically, within the subgroups of individuals aged between 45 and 64, the male demographic, those residing in rural areas, the married individuals cohabiting with their spouses, those with a primary school education or less, individuals with a BMI below 25, non-smokers, those diagnosed with hypertension, and diabetics, UA exhibited a notable mediating influence. Conversely, this effect was not observed in other subgroups. In the next sections, we discuss some of the reasons behind these findings. It has been reported that ACEs may be related to increased transcription levels of inflammatory genes in old age [31]. The inflammatory reaction has a complicated regulatory mechanism, affected by epigenetic mechanisms and transcription factors, and its competence in later life may be determined by bioreprogramming in response to early life environments [32]. Another study proposed a biological embedding model, suggesting that stress in childhood may amplify the pro-inflammatory tendencies of macrophages, leading to increased susceptibility to chronic diseases in older individuals [33]. The hypothalamic–pituitary–adrenal (HPA) axis is an important chain of neuroendocrine regulation in the human body, in which related hormones and neurotransmitters regulate the function of many organs throughout the body, including those of the digestive system [34]. One study has shown that childhood trauma may induce inflammatory immune system changes and neuroplastic dysfunction [35], which will undoubtedly adversely affect the HPA axis. The effect of UA levels on human health has long been relatively controversial. On the one hand, an elevated level of UA may have harmful effects on the body and present as gout, diabetes, kidney disease, hypertension, or obesity, among others [19, 36, 37]. On the other hand, some studies have suggested that increased UA levels have positive effects on antioxidant, anti-inflammatory, and neuroprotective processes, cognitive function, and intelligence [38–40]. Our study suggests that the dual role of UA may be related to its own level regulation. Within the normal range, the anti-inflammatory and antioxidant effects of UA may play a dominant role. In the subsequent mediating effect, we found that unsafe neighborhood had a positive direct effect on digestive system diseases, as well as a negative indirect effect on digestive system diseases through UA. It confirms the complex multiple mechanisms by which unsafe neighborhood influences digestive system diseases. In existing studies, the core of this mediating relationship may be related to the differences in adverse DNA methylation, which leads to an increase in metabolic risks, but many recent studies have revealed multiple mechanisms by which DNA methylation increases or decreases the risks of many adverse outcomes [41]. The specific mechanisms related to this in our study still require further verification through experiments. On the other hand, it may be that chronic stress brought by ACE leads to an increase in cortisol and abnormal regulation of the HPA axis, thereby causing dysregulation of the body’s homeostasis [42]. ACE is related to the increase of many inflammatory and metabolic markers, including UA, which is consistent with our results [17, 43]. Combined with the protective effect of UA on digestive system diseases mentioned above, it can be verified that unsafe neighborhood causes the prevention of digestive system diseases by increasing UA levels, and this may be because the adverse environment stimulates the body to produce some compensatory physiological responses. At the same time, the positive direct effect of unsafe neighborhood on digestive system diseases indicates that there are other potential intermediate factors playing a role between the two. Through this study, we call for attention to the influence of ACEs on the physical and psychological health and for early screening of chronic diseases, especially digestive diseases. Humanistic care should be prioritized for children and adolescents, especially those who already have ACEs, to prevent or intervene to ameliorate the impact of traumatic events on future health outcomes. Our study has several advantages. First, we conducted a longitudinal cohort study and performed a mediation analysis. This approach allowed our research to establish a more rigorous causal relationship. Second, in addition to analyzing the trend association between the ACE cumulative score and digestive disease occurrence, we studied each ACE independently to explore the relationship between specific conditions and disease; our goal was to evaluate internal associations and potential relationships with other variables. Third, the data were adjusted twice; initially for sociodemographic characteristics and lifestyle habits only, and then for part of the blood test results to yield more stable results. Fourth, we included subgroup analyses by age, gender, residence, educational status, marital status, BMI, smoking status, drinking status, hypertension, diabetes, and UA levels to reveal variations in different groups. Despite these strengths, our study has several limitations. First, the data about ACEs were collected based on the respondents’ recall, which may have introduced bias, even though an explanation regarding data reliability has been provided previously [44]. Second, the intensity and frequency of ACEs should be quantitatively evaluated in future studies to increase accuracy. In addition, while this study adjusted for key confounders including demographic characteristics, lifestyle behaviors, clinical comorbidities, and laboratory biomarkers, the potential for residual or unmeasured confounding bias cannot be entirely excluded. Such bias may introduce systematic error in the effect size estimation. Lastly, the questionnaire on digestive diseases collected through the CHARLS database does not explicitly list the specific diseases, which precluded the possibility of definitive inference regarding the causal relationship between exposure variables and specific digestive system disorders. Conclusion Our study shows a significant relationship between household mental illness, unsafe neighborhood, cumulative ACE scores and UA with prevalence of digestive disease and supports that UA may statistically account for part of the association between unsafe neighborhood and digestive system diseases. This conclusion aims to emphasize the importance of paying attention to ACE and UA levels in older populations to predict and intervene in the occurrence of digestive diseases. Supplementary Information Supplementary Material 1. Acknowledgements This study was analyzed using the data provided by the China Health and Retirement Longitudinal Study. The authors would like to thank the CHARLES participants and the staff members for their contribution to data collection and for making the data publicly available. Abbreviations Adverse childhood experience The China Health and Retirement Longitudinal Study Uric acid Body mass index (calculated as weight in kilograms divided by height in meters squared) Systolic blood pressure Diastolic blood pressure Fasting blood glucose Glycated hemoglobin Odds ratio 95% confidence intervals Standard deviation Low density lipoprotein cholesterol High density lipoprotein cholesterol Cholesterol Triglycerides C-reactive protein Hypothalamic–pituitary–adrenal Authors’ contributions JD and SY conceived and designed the study and drafted the manuscript. JD and CQ extracted the data and prepared the figures and graphical abstract. JD, LC, XY, and JZ performed the statistical analyses and interpreted the data. XL and TS supervised the study and critically revised the manuscript for important intellectual content. All authors reviewed and approved the final manuscript. Funding This work was supported by the National Natural Science Foundation of China (No.82500611 and No.82470537), Leading Talent Project of the Shaanxi Provincial High-Level Health Talent Cultivation Program (2025-QT520) and the Fundamental Research Funds for the Central Universities (No.xtr062025009). Data availability The datasets generated and analysed during the current study are available in the China Health and Retirement Longitudinal Study, https://charls.pku.edu.cn/. Declarations Ethics approval and consent to participate The study protocol was approved by the Ethical Review Committee of Peking University (approval number: IRB00001052 − 11015). All research procedures were conducted in accordance with the ethical guidelines of the Declaration of Helsinki. Written informed consent was obtained from all individual participants included in the 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. Jiayin Du and Shuting Yang contributed equally to this work and should be considered co-first authors. Contributor Information Xuqi Li, Email: lixuqi@163.com. Tianli Shen, Email: shentianli1994@stu.xjtu.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets generated and analysed during the current study are available in the China Health and Retirement Longitudinal Study, https://charls.pku.edu.cn/.