To establish a model for predicting the risk of visual disability in patients with diabetic retinopathy and verify it externally.

A total of 383 DR Patients who received ophthalmology treatment in a tertiary eye hospital in Anhui Province from April to December 2022 were conveniently selected as the modeling group to construct a visual disability risk prediction model and test the prediction effect. It was convenient to select 165 patients with diabetic retinopathy from January 2023 to April 2023 in this hospital as the verification group for external verification.

The incidence of visual disability in the modeling group was 51.70%. Gender, living style, whether suffering from other chronic diseases, regular revisit, DR Stage, number of diseased eyes, intraocular pressure value and family history of diabetes were the influencing factors(P< 0.05). The final regression equation was as follows: Logit(P) =－8.825+0.797×sex－0.874×residence style+1.504×whether you have other chronic diseases－0.871×whether you have regular follow－up visits +0.743×DR Stage+1.250×number of diseased eyes+0.166×intraocular pressure value +0.920×family history of diabetes. The Hosmer-Lemeshow test of the modeling group showed that x²=12.861, P=0.117, area under ROC curve was 0.838, 95% CI(0.795, 0.882), sensitivity was 0.737, specificity was 0.809. The Hosmer-Lemeshow test of the verification group showed that x²=15.141, P=0.056, area under ROC curve was 0.785, 95%CI was 0.704 - 0.866, sensitivity was 0.795, specificity was 0.727, accuracy was 75.76%.

The prediction effect of this model is good, and it can provide reference for clinical evaluation of the risk of visual disability in diabetic retinopathy patients.

To evaluate the association between abnormal renal function indicators (SUA and eGFR) and non-alcoholic fatty liver disease (NAFLD).

Based on 2017-2021 Beijing Health Management Cohort, the cross-lagged panel model was used to investigate the causal temporal relationship between abnormal SUA and glomerular filtration rate and the onset of NAFLD.

(1) Increased SUA and NAFLD: ① Path coefficients from baseline SUA to follow-up HSI were statistically significant in both the general population (β=0.018, 95% CI: 0.003-0.032) and BMI≥24 kg/m² (β=0.051, 95% CI: 0.032-0.070), but not the other way around. ② Path coefficients from baseline SUA to follow-up HSI (β=0.048, 95% CI: 0.028-0.068) and baseline HSI to follow-up SUA (β=0.023, 95% CI: 0.005-0.041) were statistically significant in BMI<24 kg/m². (2) eGFR abnormalities and NAFLD: ① Path coefficients from baseline HSI to follow-up eGFR were statistically significant in both the general population (β=0.024, 95% CI: 0.012-0.036) and BMI≥24 kg/m² population (β=0.035, 95% CI: 0.018-0.052), but not the other way around. ② In BMI<24 kg/m², the path coefficients from baseline eGFR to follow-up HSI and baseline HSI to follow-up eGFR were not statistically significant.

In the general population and BMI≥24 kg/m², the abnormal SUA is earlier than the incidence of NAFLD, and the incidence of NAFLD may affect the subsequent glomerular filtration rate. In people with normal BMI, the onset of NAFLD is associated with elevated SUA, but not with abnormal glomerular filtration rate.

Observational studies have proposed a link between frailty and chronic obstructive pulmonary disease. However, the causal relationship between the two diseases needs further investigation.

The study data were drawn from the GWAS dataset, in which the frailty data contained 175 226 samples, by selecting SNPs closely related to frailty as instrumental variables. Two-sample MR was applied to assess causality between diseases. Inverse variance weighting (IVW) methods were used as the main analyses. MR-Egger intercepts, MR-PRESSO and funnel plots were also used to detect horizontal multi-effects, and sensitivity analyses were performed using the “leave-one-out” and Cochran’s Q tests simultaneously.

IVW analysis showed that frailty genetic susceptibility increased the risk of COPD, with odds ratios (OR) of 1.935 (95% CI:1.178-13.179; P=0.009). No horizontal pleiotropism was observed in MR-Egger intercept (P=0.757), MR-PRESSO and funnel plots detection. In the test of heterogeneity (P=0.952), no heterogeneity was observed in the funnel plot. The “leave-one-out” did not reveal a single SNP with a biased effect on the instrumental variable.

The study findings suggest a possible positive causal relationship between frailty and increased risk of developing COPD.

To understand the changes of height and growth retardation of students in nutrition improvement plan areas from 2012 to 2021 since the implementation of "Nutrition Improvement Plan for Students receiving Compulsory Education in Rural areas" in 2011, so as to provide a theoretical basis for better improvement of students’ health status and further decision-making.

The stratified cluster sampling method was used to evaluate nutritional status of the students in the monitored schools in Hebei Province by physical examination, monitoring their height and calculating the growth retardation rate.

In 2021, the average height of male and female students in the same age group increased by 3.62cm and 2.65cm on average compared with 2012. The height increase of boys and girls in the same age group was statistically different (all P<0.01). From 2012 to 2021, the peak of average height increase for boys was between the ages of 12 and 13, with an added value of 5.7cm, while that of girls was between the age of 11 and 12, with an added value of 4.3cm. From 2012 to 2021, the growth retarding rates of boys and girls decreased year by year, and the Chi-square linear trend test showed that the difference was statistically significant (Chi-square linear trend test all P<0.005).

From 2012 to 2021, the average height of students aged 6 to 15 years old in the areas under the nutrition improvement plan of rural students in our province increased. However, since the baseline data in 2012 was higher than the national level in rural areas, the height has tended to be stable in recent years, so the average growth rate of students is lower than the national level, and the growth retardation rate of students shows a decreasing trend year by year, and the growth retardation rate is lower than the national average level.

To explore the relationship among self-perceived burden, family resilience and quality of life in patients with ovarian cancer, and to analyze the potential role of family resilience between self-perceived burden and quality of life.

Ovarian cancer patients hospitalized in the oncology departments of two tertiary hospitals in Sichuan Province from May 2022 to January 2023 were selected as the research subjects, and the general information questionnaire, family resilience assessment scale (FRAS), self-perceived burden scale of cancer patients (SPBS-CP), and ovarian cancer quality of life scale (FACT-O) were used to conduct questionnaire surveys. SPSS 22.0 was used for descriptive analysis, correlation analysis and multiple linear regression analysis. PROCESS 3.4 was used to analyze the mediating effect of family resilience between self-perceived burden and quality of life, and the Bootstrap method was used to test the mediating effect.

The family resilience score of ovarian cancer patients was (140.61±15.82), self-perceived burden was (59.46±19.95), and quality of life was (91.52±32.01). There was a positive correlation between family resilience and quality of life in patients with ovarian cancer (r=0.464, P<0.01), and the self-perceived burden was negatively correlated with family resilience and quality of life (r=-0.385, -0.439, P<0.01). Family resilience had a partial mediating effect between ovarian cancer patients’ self-perceived burden and quality of life, and the mediating effect accounted for 30.35% of the total effect.

The self-perceived burden of ovarian cancer patients is at a moderate level, family resilience is at a lower middle level, and the quality of life needs to be improved. It is recommended that healthcare professionals improve the quality of life by enhancing family resilience and reducing the burden of self-perception through health education methods such as enhancing family support, establishing coping beliefs, and promoting family communication.

To construct the Bayesian network model of H-type hypertension in middle-aged and elderly people, and to explore the influencing factors of H-type hypertension and the network relationship between factors, and the strength of each influencing factor on H-type hypertension.

A total of 1 119 middle-aged and elderly people who underwent physical examination in the hospital health management center from May 2022 to April 2023 were selected as the research objects and relevant data were collected. Univariate logistic regression analysis and multivariate logistic regression analysis models were used for preliminary screening of variables, “bnlearn” Bayesian network software package was used for model construction, and Netica software was used for model inference.

Logistic regression analysis model was used to screen the variables, such as age, gender, education level, smoking, drinking, body mass index(BMI), fasting blood glucose, etc. FBG, total cholesterol(TC), triglyceride(TG), high density lipoprotein cholesterol((HDL-C), low density lipoprotein cholesterol(LDL-C) and uric acid(UA) were included in the Bayesian network model. A Bayesian network model of H-type hypertension related factors in middle-aged and elderly people with 13 nodes and 16 directed edges was constructed by using 12 selected variables as network nodes. Age, FBG, TG, HDL-C and BMI were directly related to H-type hypertension, while gender, smoking, drinking, educational level, TC, LDL-C and UA were indirectly related to H-type hypertension. When the age was≥60years old, FBG≥6.85 mmol/L, BMI≥24.83 kg/m², HDL-C≥1.02 mmol/L, TG<1.6 mmol/L, the risk of H-type hypertension reached 0.565.

The Bayesian network model reveals the direct and indirect factors and correlation strength of H-type hypertension in middle-aged and elderly people, clarifies the complex network relationship between factors, and provides a scientific basis for early prevention of H-type hypertension in middle-aged and elderly people.

To measure the health human resource demand of maternal and child health institutions in China, and to analyze the spatial and temporal characteristics of maternal and child health human resource allocation.

The data of health personnel in maternal and child health care institutions from 2018 to 2022 were used, the health human resource density index was used to analyze the change trend of the allocation level of maternal and child health human resources in China from 2017 to 2021, and the actual demand, shortage and proportion of health personnel in maternal and child health institutions in 2021 were calculated. At the same time, global and local spatial autocorrelation was used to explore the spatial characteristics of the allocation level of maternal and child health human resources.

From 2017 to 2021, the density index of health personnel in maternal and child health institutions in China increased year by year, but the LISA diagram showed that the allocation of human resources among provinces was uneven, showing a ’east-west’ differentiation distribution. In 2021, the number of practicing(assistant) physicians per thousand population, registered nurses per thousand population, pharmacists per thousand population and technicians per thousand population in China were 11.30, 14.91, 1.31 and 2.73, respectively, and the ratio of doctors to nurses was 1∶1.32. The Global Moran index and Local Moran index of each health human resource index were greater than 0.

The number of health human resources in maternal and child health institutions in China is increasing, but the inter-regional allocation is not balanced. The internal structure of resource allocation needs to be optimized, and there is obvious spatial aggregation of the same attribute in resource allocation.

To explore the correlation between visceral fat index(CVAI) and the incidence of diabetes in Chinese elderly population.

The data were obtained from the China Health and Retirement Longitudinal Survey(CHARLS) 2011 and 2015 waves. A total of 2 295 elderly people aged ≥60 years were included. Binary logistic regression was used to analyze the correlation between CVAI and the incidence of diabetes. The restricted cubic spline(RCS) method was used to test the dose-response relationship. Subgroup analysis was performed according to BMI.

333(14.50%) of the 2295 elderly participants developed diabetes during the 4-year follow-up period. CVAI were divided by quartiles. There were statistically significant differences in baseline FBG, HBA1c, SBP, DBP, TG, TC, HDL-C, LDL-C, UA, gender, smoking status, drinking status, hypertension history and follow-up incident diabetes among groups(P<0.05). Compared with the group with lowest quartile of CVAI Q1, the OR(95% CI) in the group the highest quartile of CVAI Q4 was 1.96(1.37-2.83). The restricted cubic splines model results showed that the higher CVAI accompanied by higher risk of incident diabetes in the elderly. Subgroup analysis showed that CVAI was independently associated with incident diabetes in both non-overweight and overweight/obese groups.

There is a positive association between CVAI the risk of diabetes mellitus in the elderly population. The control of visceral fat in the elderly is helpful to prevent the occurrence of diabetes.

To explore the impact mechanism of peer relationship networks in children and adolescents on overweight and obesity, and to provide some empirical basis for the health management and intervention of overweight and obesity in children and adolescents.

Based on a sampling survey data on overweight and obesity among children and adolescents aged 7-17 in Nanchong City of Sichuan Province, social network analysis, regression analysis, and Bayesian network models were used to analyze the specific impact mechanism of peer relationship networks on overweight and obesity.

Logistic regression showed that high personal popularity(OR=0.001, 95% CI=0-0.026) and high personal activity(OR=0.084, 95% CI=0.009-0.790) were associated with reduced detection of overweight and obesity in children and adolescents, with a relatively stronger correlation between personal activity and obesity. In addition, high household income(OR=19.237, 95% CI=3.799-97.403), high-energy dietary patterns(OR=21.660, 95% CI=1.600-292.904), sitting still for more than 8 hours(OR=10.395, 95% CI=2.013-53.687) increased the risk of overweight and obesity. Long exercise time(OR=0.085, 95% CI=0.019-0.378), both parents have attended college(OR=0.023, 95% CI=0.003-0.169), and mental health(OR=0.030, 95% CI=0.006-0.147) Children and adolescents with high sleep quality(OR=0.006, 95% CI=0.001-0.045) had a lower risk of developing overweight and obesity. The BIC score of the Bayesian network model was -3 954.8. When both individual activity and popularity were set at the 0:1:0 level, the detection probability of overweight and obesity in children and adolescents was 0.126, which was at the lowest level. Adjusting for other intervention variables under the recommended appropriate social network level(i.e. moderate popularity and moderate activity), with the best effect on controlling exercise duration, the likelihood of overweight and obesity in children and adolescents will decrease to 0.044. If social network indicators, dietary habits, exercise duration, psychological and sleep conditions were controlled simultaneously, the detection probability of overweight and obesity was 0, which will achieve the best expected prevention of overweight and obesity detection.

Strengthening social networks, changing lifestyle habits, and improving dietary quality may be important links in the prevention and treatment of obesity in children and adolescents. Therefore, it is necessary to strengthen the responsibility of government departments to accelerate the construction of a comprehensive overweight and obesity prevention and control system.

To use disability-adjusted life years(DALYs) to assess unequal differences in the burden of disease due to diabetes across global socioeconomic groups.

Number of DALYs due to diabetes, age-standardized DALYs rates from the GBD database, 1990-2019, and national Human Development Index(HDI) data from the Human Development Report 2021-2022 were collected. The relationship between age-standardized DALY rates and HDI was analyzed to understand socioeconomic differences in diabetes disease burden. Calculation of Annual Percentage Change(APC) and Average Annual Percentage Change(AAPC) to assess temporal trends in the burden of diabetes disease during recent decades.

From 1990-2019, the number of DALYs due to diabetes globally increased from 28.3 million to 69.9 million, an increase of 146.85%, with the fastest growth(206.49%) in Lower middle-income countries. The age-standardized DALYs rates were negatively correlated with HDI(r=-0.480, P<0.001). Countries with medium HDI exhibit higher DALYs rates due to diabetes mellitus. From 1990-2019, the average annual increase in the rate of age-standardized DALYs was 0.78%(AAPC=0.78%, 95% CI=0.74% to 0.81%), and the average annual increase in the age-standardized mortality rate was 0.46%(AAPC=0.46%, 95% CI=0.41%～0.51%).

The global burden of diabetes disease has increased substantially over the last few decades, with the fastest growth in Lower middle-income countries. The age-standardized burden was higher in developing countries. Although a slower increase in the burden of diabetes was observed, the persistence of the increase in the burden of disease suggests that more diabetes prevention programs and health-care services should be made available to developing countries.