To investigate the correlation and dose-response relationship between the triglyceride glucose-body mass index (TyG-BMI) and the incidence of stroke.

A prospective cohort study design was employed to compare differences in demographic characteristics, behavioral lifestyles, family history, and blood biochemical results between individuals who developed stroke and those who did not. All subjects were divided into four groups, Q1 to Q4, based on the quartiles of TyG-BMI in ascending order. The Cox proportional hazards regression model was used to analyze the association between TyG-BMI and stroke risk, calculating the hazard ratio (HR) and its 95% confidence interval (CI). A Restricted Cubic Spline (RCS) model was used to analyze the dose-response relationship between TyG-BMI and stroke risk. After the RCS analysis suggested a nonlinear trend, we further utilized the segmented package in R software to perform piecewise Cox regression to identify the threshold effect. The goodness of fit of the models was compared using the Davies test and the likelihood ratio test.

This study included 17 173 subjects with a median follow-up time of 19.5 years. During the follow-up period, 2 374 new stroke cases occurred, with an incidence density of 805 per 100 000 person-years. Significant differences(P<0.05) were observed between the stroke group and the non-stroke group in terms of age, education level, income, physical activity intensity, history of hypertension, family history of stroke, systolic blood pressure (SBP), diastolic blood pressure (DBP),body mass index (BMI), waist circumference (WC), fasting plasma glucose (FPG), total cholesterol (TC), triglycerides (TG), and TyG-BMI. After adjusting for confounding factors, the risk of stroke gradually increased with rising TyG-BMI levels (P<0.001). Compared to the Q1 group, the risk of new-onset stroke increased by 18.1% (HR=1.181, 95% CI: 1.037-1.346), 33.3% (HR=1.333, 95% CI: 1.171-1.517), and 57.0% (HR=1.570, 95% CI: 1.366-1.804) in the Q2, Q3, and Q4 groups, respectively. The RCS analysis showed a positive association between TyG-BMI and stroke risk (P for overall <0.05), and a non-linear dose-response relationship was identified (P for non-linearity <0.05). Piecewise Cox regression identified a threshold value of 214.3,and the likelihood ratio test confirmed that the piecewise model had a significantly better fit than the traditional linear model (P=0.037). On both sides of the threshold, each unit increase in TyG-BMI was significantly associated with an increased risk of stroke (below the threshold: HR=1.151, 95% CI: 1.032-1.325, P<0.001; above the threshold: HR=1.232, 95% CI: 1.085-1.418,P<0.001).

TyG-BMI exhibits a nonlinear dose-response relationship with the risk of stroke, with a significant threshold effect identified at 214.3. The risk of stroke increased more markedly when TyG-BMI exceeded this threshold. TyG-BMI may serve as a simple indicator for stroke risk screening in rural community populations.

To classify cross-district/county medical visits into upward flow (from low-GDP districts to high-GDP districts) and downward flow (from high-GDP districts to low-GDP districts), analyze their influencing factors and differences from overall cross-district/county visits, and provide theoretical support for optimal allocation of medical resources.

Data were obtained from the front pages of medical records of inpatients in Sichuan Province in 2022. Taking inpatients as the study subjects, two types of cross-district/county medical visits were classified. Geographic information software was used for spatial visualization, and a two-level random intercept logistic regression model was applied to analyze influencing factors from patient, institutional and regional dimensions.

In 2022, the proportion of cross-district/county hospitalization in Sichuan was 29.20%, including 23.37% upward flow and 5.83% downward flow. Maintenance chemotherapy for malignant tumors accounted for the highest proportion in overall (4.97%) and upward flow (6.67%), while schizophrenia (2.69%) ranked first in downward flow. West China Hospital of Sichuan University in Wuhou District was the top choice for overall (5.69%) and upward flow (7.10%), and Chengdu Second People’s Hospital in Jinjiang District was preferred for downward flow (7.92%). The effects of occupation type, medical insurance and other individual factors became more significant after distinguishing the two flows.

Upward and downward cross-district/county medical visits showed distinct demand differences: the former pursues high-quality medical services, while the latter mainly focuses on rehabilitation and other needs. Local governments should strengthen specialized or primary care services based on patient outflow and resource conditions, improve the rate of seeking care within the county, and promote the efficient, fair and sustainable development of the medical service system.

To reveal the complex associations among six domains of health literacy among residents in Chengdu using network analysis, and to identify differences and key nodes in health literacy networks between urban and rural residents.

Participants were recruited using a multi-stage sampling design with probability proportional to size (PPS). A total of 6 450 participants aged 15-69 years were enrolled. Health literacy and sociodemographic characteristics were assessed using the National Residents’ Health Literacy Surveillance Questionnaire (Health Statistics Form No. 115). R 4.4.1 was used to conduct statistical analysis.

Among six domains of health literacy, the possession rates of Scientific Health Perspectives, Infectious Disease Prevention and Control, Chronic Disease Prevention and Management, Safety and First Aid, Basic Healthcare, and Health Information were 61.49%, 36.39%, 38.96%, 63.45%, 39.68%, 50.32% among urban residents, and 52.73%, 35.93%, 36.00%, 52.36%, 39.07%, 38.85% among rural residents. Network comparison showed differences in network structure and global strength between urban and rural residents. The strongest correlations in both urban and rural networks were between Scientific Health Perspectives and Safety and First Aid (urban: 1.489, rural: 1.386). Additionally, the associations between Scientific Health Perspectives and Chronic Disease Prevention and Management (urban: 1.047, rural:0.752), as well as between Infectious Disease Prevention and Control and Basic Healthcare (urban: 0.997, rural: 0.495), were stronger in urban areas than in rural areas. In contrast, the correlations between Scientific Health Perspectives and Infectious Disease Prevention and Control (rural: 0.917, urban: 0.675), Scientific Health Perspectives and Basic Healthcare (rural: 1.310, urban: 0.863), Chronic Disease Prevention and Management and Basic Healthcare (rural: 1.225, urban: 0.870), and Chronic Disease Prevention and Management and Health Information (rural: 1.257, urban: 1.009) were stronger in rural areas than in urban areas. The centrality strength values for Infectious Disease Prevention and Control, Chronic Disease Prevention and Management, and Basic Healthcare differed between urban and rural networks (P<0.05).

Health literacy levels among rural residents still require comprehensive improvement, while persistent gaps remain in urban areas regarding literacy in Infectious Disease Prevention and Control, Chronic Disease Prevention and Management, and Basic Healthcare. To enhance the overall health literacy level, tailored strategies are needed for different settings.

To analyze the relationship between learning stress and emotional eating behavior among first-year medical graduate students, with the aim of providing a reference for health education for first-year medical graduate students.

A total of 1 583 first-year medical graduate students in Guizhou Province were selected using stratified cluster random sampling, and questionnaire surveys were conducted using the Emotional Eating Behavior Scale, the Postgraduate Learning Stress Questionnaire, the Social Support Scale, and the Eating Attitude Test. Taking learning stress as the independent variable, emotional eating behavior as the dependent variable, and social support and eating attitude as the mediating variables, a chained mediation model was constructed.

There were significant differences in emotional eating behavior scores among first-year medical graduate students based on gender, smoking, and family relationships (t=-3.998, P<0.001; t=2.521, P=0.012; F=4.344, P=0.013). Learning stress had a direct effect on emotional eating behavior (B=0.078). Social support and eating attitude had a chained mediation effect between learning stress and emotional eating behavior, accounting for 4.76% (B=0.008).

Learning stress can be directly associated with emotional eating behavior, and can also be indirectly associated with emotional eating behavior through the independent mediating effects or chained mediating effects of social support and eating attitude.

To systematically analyse the current status and methodological characteristics of discrete choice experiment (DCE) in colorectal cancer screening.

A scoping review was conducted following the PRISMA-ScR guidelines. Chinese and English databases were searched from inception to February 2025. Included studies were analysed in terms of attribute sources, attribute categories, attribute-setting methods, and experimental design.

A total of 40 studies were included. The number of publications showed an overall increasing trend, while relevant studies in mainland China remain lacking. Attribute development was mainly based on literature review combined with other methods. Attributes were classified into five categories, among which sensitivity, pre-screening preparation, and cost were the main preference attributes. Most studies included 4-6 attributes with 2-4 levels. In terms of experimental design, orthogonal and fractional factorial designs were most commonly used. Multinomial logit and mixed logit models were the main analytical methods. Some studies used multiple models or software for analysis.

Current DCE studies on colorectal cancer screening mainly adopt conventional designs and commonly used models, with relatively concentrated attribute settings. Future studies should strengthen local research and develop attribute systems suited to population needs, so as to provide a basis for demand-oriented screening strategies and to improve participation in primary-level screening.

To investigate the association between different obesity phenotypes and the risk of dementia in an elderly health examination population.

Based on the health examination database of the Health Management (Examination) Center of the Affiliated Hospital of Chengdu University of Traditional Chinese Medicine from 2015 to 2025, participants were categorized according to body mass index (BMI) and metabolic health status into four groups: metabolically healthy non-obese (MHNO), metabolically healthy obese (MHO), metabolically unhealthy non-obese (MUNO), and metabolically unhealthy obese(MUO). The Fine-Gray competing risk model was used to evaluate the association between different obesity phenotypes and the cumulative risk of dementia, and subdistribution hazard ratios (SHR) with their 95% confidence intervals (CI) were calculated.

A total of 42 915 participants were eventually included, consisting of 17 015 in the MHNO group, 7 520 in the MHO group, 8 480 in the MUNO group, and 9 900 in the MUO group. Fine-Gray competing risk model analysis showed that, using the MHNO group as the reference, the risk of all-cause dementia, Alzheimer’s disease (AD), and vascular dementia(VD)showed an increasing trend with the severity of metabolic abnormalities and obesity. After full adjustment for confounding factors, the SHR(95% CI) for all-cause dementia was 1.102 (1.040-1.226) in the MHO group, 1.392 (1.257-1.542) in the MUNO group, and 1.725 (1.571-1.894) in the MUO group, with similar gradients observed for AD and VD risks. The results of sensitivity analysis were consistent with the main analysis.

A significant gradient association was observed between different obesity phenotypes and dementia risk, demonstrating a sequential increase in dementia risk with the severity of metabolic abnormalities and obesity. These findings highlight the need to prioritize the MUO population in dementia prevention and implement stratified intervention strategies for different phenotypes.

To investigate the preventive effect and potential mechanism of intervention with Akkermansia muciniphila (AKK) isolates on dextran sulfate sodium (DSS)-induced acute colitis in mice.

Forty 6-8-week-old male C57BL/6J mice were randomly divided into a blank control group, a model group, an HX003 group, and an HX004 group. After 4 weeks of prophylactic intervention, acute colitis was induced by dextran sulfate sodium (DSS). The pathological changes of colon and spleen tissues in each group were observed, the serum LPS level was determined, and the mRNA transcription levels and protein expression levels of inflammatory factors IL-6, IL-10, IL-1β, and IFN-γ in mouse colon tissues were detected. The colonic contents of mice were collected, and 16S rRNA sequencing was used to analyze the structure and abundance changes of the intestinal flora.

Prophylactic intervention with AKK isolates alleviated colonic inflammation in mice to a certain extent. Serum lipopolysaccharide (LPS) levels in mice exhibited a decreasing trend (HX003, P=0.228). Both transcription and expression of IL-6 in colonic tissue showed downward trends. The transcription level of the pro- inflammatory cytokine IL-1β was significantly decreased (HX003, P<0.05), accompanied by a downward trend in its protein expression. In contrast, the transcription level of the anti-inflammatory cytokine IL-10 was markedly upregulated (HX003, P<0.001), with an increasing trend observed in its expression. Compared with the model group, the intervention groups displayed an upward trend in gut microbial diversity and a downward trend in the Firmicutes/Bacteroidetes ratio.

AKK isolates alleviate colonic inflammation in mice with acute colitis to a certain extent and improve the diversity and structure of their intestinal flora.

To analyze the weekend/holiday attenuation and delayed reporting effects of notifiable infectious diseases in Shapingba district, Chongqing from 2015 to 2024, and to provide evidence for improving the quality of infectious disease reporting in medical institutions.

Data on reported notifiable infectious disease cases in Shapingba District from 2015 to 2024 were collected from the National Notifiable Disease Surveillance System of the China Information System for Disease Control and Prevention. Statistical analyses were performed using R 4.4.2. Pearson χ² tests and Wilcoxon rank-sum tests were employed for inter-group comparisons, while negative binomial regression and linear regression were applied to assess attenuation and delayed reporting effects.

Significant weekend/holiday attenuation and delayed reporting effects were observed for notifiable infectious diseases in Shapingba district. The daily average number of reports during weekends, short holidays, and long holidays decreased by 62.15%, 60.37%, and 69.38%, respectively, compared to workdays. The attenuation effect weakened for weekends and short holidays during 2023—2024. Compared with workdays, the time from diagnosis to reporting was prolonged by an average of 2.58 hours, 2.18 hours, and 2.60 hours. The weekend/holiday reporting delay effect weakened from 2020 to 2022, but re-emerged after 2023. Tertiary healthcare institutions exhibited the most pronounced weekend/holiday attenuation and delayed reporting effects.

Significant weekend/holiday attenuation and delayed reporting effects persist in Shapingba district, particularly in tertiary healthcare institutions. Medical institutions should strengthen the management of infectious disease reporting during weekends and holidays.

To analyze the epidemiological characteristics and temporal variation trends of active surveillance for foodborne diseases in Shanxi Province from 2015 to 2024, and to provide a scientific basis for formulating precise prevention and control strategies.

Active surveillance data from 2015 to 2024 were extracted from the Shanxi Provincial Foodborne Disease Pathogenic Factor and Etiological Food Traceability Platform. Joinpoint Regression Program 4.9.1.0 software was used to analyze the long-term temporal trend of foodborne disease detection rate. Annual percentage change(APC), average annual percentage change (AAPC), monthly percentage change (MPC), and average monthly percentage change(AMPC) were adopted to analyze the annual and monthly temporal trends of detection rate. Meanwhile, the distribution and trend characteristics of detection rate by gender and age group were analyzed, and stratified analysis was conducted according to pathogen type to describe the temporal trend distribution of pathogens.

A total of 18 621 cases were monitored from 2015 to 2024, with 3 973 positive samples, and the total detection rate was 21.336%. The detection rate of foodborne diseases showed a significant upward trend from 2015 to 2024 (AAPC=9.8%, 95% CI: 3.9%-16.0%, P=0.004). The detection rate showed a significant downward trend with increasing age (AAPC=-10.0%, P<0.001). The detection rates in the 0-10 years and >70-80 years age groups showed a significant upward trend over time (AAPCs were 10.6% and 16.8%, respectively, all P<0.001). The monthly trend showed significant phased characteristics. The Joinpoint model identified two inflection points in May and September: the detection rate increased significantly from January to May (MPC=16.6%, P=0.040), decreased non-significantly from May to September (MPC=-4.4%, P=0.306), and decreased significantly from September to December (MPC=-22.5%, P=0.025). The average monthly trend of the whole stage was not statistically significant (AMPC=-2.9%, P=0.330). Among the pathogens, Salmonella and Norovirus showed a significant upward trend from 2015 to 2024, with AAPCs of 19.1%and 9.0%, respectively. Bacterial pathogens showed a significant upward trend from January to August and a significant downward trend from August to December. The detection rate of Norovirus showed a continuous and significant downward trend in monthly changes (MPC=-7.5%, P=0.013).

The detection rate of foodborne diseases in Shanxi Province showed a significant upward trend from 2015 to 2024, and there were differences in temporal trend characteristics among different genders and age groups. It is suggested to improve the monitoring system and formulate precise prevention and control measures for key populations, key places and different seasons to reduce the risk of foodborne diseases.

Uterine leiomyoma (UL) is a common benign tumor among women of reproductive age. Alanine aminotransferase (ALT) is closely linked to metabolic and endocrine dysregulation, but its causal relationship with UL remains unclear. This study used Mendelian randomization (MR) to assess the potential causal association between ALT and UL.

Using summary statistics from genome-wide association studies (GWAS) of European ancestry, we performed two-sample MR to evaluate the causal effect of ALT on UL risk. We first conducted univariable MR and then performed multivariable MVMR analyses adjusting for body mass index (BMI), blood pressure, glycemic traits, lipid profiles, sex hormone-binding globulin (SHBG), and lifestyle-related factors to estimate the independent effect of ALT. Reverse MR (UL→ALT) was applied to assess directionality. Colocalization analyses were further performed to test whether ALT and UL share causal variants at key loci.

Univariable MR showed that genetically predicted higher ALT levels were associated with an increased risk of UL (OR=1.125, P=0.012; OR=1.143, P=0.007). In MVMR, the positive effect of ALT on UL remained robust after adjustment for multiple metabolic, hormonal, and lifestyle-related factors (OR=1.110-1.139; P=0.008-0.045). Reverse MR provided no evidence supporting a causal effect of UL on ALT. Colocalization analyses identified strong evidence of shared causal variants between ALT and UL at loci including HSPA4 and ATF1 (PP.H4>0.8).

This study provides multi-layered genetic evidence that elevated ALT independently increases the risk of UL, suggesting that hepatic metabolic dysfunction may contribute to UL development through pathways involving sex hormone metabolism and metabolic inflammation. As an easily accessible clinical marker, ALT may have potential value for risk prediction and early warning of UL.