Rural Chinese Population Research Articles

Prediction of Late Hospital Arrival in Patients with Mild and Rapidly Improving Acute Ischemic Stroke in a Rural Area of China.

Among all ischemic stroke patients, more than half are mild and rapidly improving acute ischemic stroke (MaRAIS) patients. However, many MaRAIS patients do not recognize the disease early on, and thus they delay access to the treatment that would be most effective if provided earlier. This is especially true in rural areas. The aim of this study was to develop and validate a late hospital arrival risk nomogram in a rural Chinese population of patients with MaRAIS. We developed a prediction model based on a training dataset of 173 MaRAIS patients collected from September 9, 2019 to May 13, 2020. Data analyzed included demographics and disease characteristics. A least absolute shrinkage and selection operator (LASSO) regression model was used to optimize feature selection for the late hospital arrival risk model. Multivariable logistic regression analysis was applied to build a prediction model incorporating the features selected in the LASSO regression models. The discrimination, calibration, and clinical usefulness of the prediction model were assessed using the C-index, calibration plot, and decision curve analysis, respectively. Internal validation was then assessed using bootstrapping validation. Variables contained in the prediction nomogram included transportation mode, history of diabetes, knowledge of stroke symptoms, and thrombolytic therapy. The model had moderate predictive power with a C-index of 0.709 (95% confidence interval: 0.636-0.783) and good calibration. In the internal validation, the C-index reached 0.692. The risk threshold was 30-97% according to the analysis of the decision curve, and the nomogram could be applied in clinical practice. This novel nomogram, which incorporates transportation mode, history of diabetes, knowledge of stroke symptoms, and thrombolytic therapy, was conveniently applied to facilitate individual late hospital arrival risk prediction among MaRAIS patients in a rural area of Shanghai, China.

The association of cortisol/testosterone ratio and sleep quality with coronary heart disease: A case-control study in Chinese rural population.

There were limited studies that have probed into the combined effect of the cortisol/testosterone (C/T) ratio as a biomarker of stress and Pittsburgh Sleep Quality Index (PSQI) on coronary heart disease (CHD). This research aimed to explore the association of C/T ratio and PSQI with the risk of CHD in a rural Chinese population, as well as the interaction and combined effect between C/T ratio and PSQI on CHD. A case-control study was performed including 307 individuals without CHD and 307 patients drawn from Henan Rural Cohort. Logistic regression was utilized to survey the independent and joint effects of the C/T ratio and PSQI on CHD. To estimate the interaction impact of the C/T ratio and sleep quality (PSQI) on CHD, a cross-product term was introduced in the generalized linear model. Higher C/T ratio and PSQI index scores are related to increased odds ratio for CHD (Odds ratios (ORs) and 95% confidence interval (CI) were 1.17 (1.07, 1.29), p-trend<0.001; 1.16 (1.09, 1.22), respectively). The odds ratio of C/T ratio for CHD increased with increasing PSQI in women (pinteraction=0.018) and total population (pinteraction=0.033). The combined group of high C/T ratio and high PSQI had the highest risk of CHD (Total: OR=7.53, 95% CI: 4.12-13.76). The risk of CHD was associated with low testosterone levels, high C/T ratios, and high PSQI scores. Additionally, poor sleep quality aggravated the effect of high C/T ratio on coronary heart disease.

Association between barium exposed, CYP19A1 and central obesity: A cross-sectional study in rural China

Backgroundobesity is a major risk factor for many metabolic diseases such as diabetes and cardiometabolic diseases. This study aimed to evaluate the association of plasma and urinary barium concentrations, CYP19A1 gene polymorphisms, and their interaction with central obesity in a rural Chinese population. Methodsrestricted cubic spline model was used to explore the dose-response relationship between barium and the risk of developing central obesity and waist circumference; logistic regression model was used to assess the association between barium, CYP19A1 gene polymorphisms and their interaction with central obesity. Resultsthe results of the restricted cubic spline model showed that plasma barium concentration was linearly associated with the risk of developing central obesity and non-linearly associated with waist circumference. Logistic regression analysis showed that participants with Q4 plasma barium concentration exhibited a higher risk of central obesity compared to participants with Q1 barium concentration; participants carrying the rs10046-AA gene exhibited a lower risk of central obesity than those carrying the rs10046-G(GG+GA) gene; participants carrying the rs10046-GA genotype showed 1.754 times higher risk of central obesity than those carrying rs10046-GG+AA genotype. There was a significant interaction between plasma barium and CYP19A1 gene polymorphism on central obesity. Conclusionthe development of central obesity was associated with plasma barium and CYP19A1.

Association of ABCA1 R219K polymorphism and telomere length in a Chinese rural population: possible linking to systemic inflammation

Association of ABCA1 R219K polymorphism and telomere length in a Chinese rural population: possible linking to systemic inflammation

All-cause mortality risk with different metabolic abdominal obesity phenotypes: the Rural Chinese Cohort Study.

We aimed to investigate the association of metabolic obesity phenotypes with all-cause mortality risk in a rural Chinese population. This prospective cohort study enrolled 15704 Chinese adults (38·86 % men) with a median age of 51·00 (interquartile range: 41·00-60·00) at baseline (2007-2008) and followed up during 2013-2014. Obesity was defined by waist circumference (WC: ≥ 90 cm for men and ≥ 80 cm for women) or waist-to-height ratio (WHtR: ≥ 0·5). The hazard ratio (HR) and 95 % CI for the risk of all-cause mortality related to metabolic obesity phenotypes were calculated using the Cox hazards regression model. During a median follow-up of 6·01 years, 864 deaths were identified. When obesity was defined by WC, the prevalence of participants with metabolically healthy non-obesity (MHNO), metabolically healthy obesity (MHO), metabolically unhealthy non-obesity (MUNO) and metabolically unhealthy obesity (MUO) at baseline was 12·12 %, 2·80 %, 41·93 % and 43·15 %, respectively. After adjusting for age, sex, alcohol drinking, smoking, physical activity and education, the risk of all-cause mortality was higher with both MUNO (HR = 1·20, 95 % CI 1·14, 1·26) and MUO (HR = 1·20, 95 % CI 1·13, 1·27) v. MHNO, but the risk was not statistically significant with MHO (HR = 0·99, 95 % CI 0·89, 1·10). This result remained consistent when stratified by sex. Defining obesity by WHtR gave similar results. MHO does not suggest a greater risk of all-cause mortality compared to MHNO, but participants with metabolic abnormality, with or without obesity, have a higher risk of all-cause mortality. These results should be cautiously interpreted as the representation of MHO is small.

Anterior segment measurements and determinants of angle width with short, medium, and long axial lengths in a rural Chinese population

PurposeTo report the anterior segment measurements and investigate the determinants of angle width with short, medium, and long axial length (AL) in a rural Chinese population. DesignObservational, population-based, cross-sectional study. MethodsSubjects aged ≥35 years who underwent complete ocular examinations during the follow-up of the Handan Eye Study were included. Ocular data of the right eye were analyzed. Anterior segment parameters were obtained and stratified by age and sex. AL was categorized into short (<22.0 mm), medium (22.0–23.5 mm), and long (>23.5 mm) subgroups. Linear regression analyses were performed to identify the parameters associated with angle width (angle opening distance at 500 μm from the scleral spur (SS) [AOD500]). ResultsThe final analysis included 4435 subjects (58.0 [49.0, 64.0] years old, 44.1% males). Smaller AOD500 was significantly associated with female sex (P = 0.032), larger iris thickness at 750 μm from the SS (IT750) (P < 0.001), larger lens vault (LV) (P < 0.001), and smaller anterior chamber volume (ACV) (P < 0.001) in the short AL subgroup; larger sphere equivalent (SE), IT750, iris curvature (IC), and LV and smaller ACV (all P < 0.001) in the medium AL subgroup; and larger SE, IT750, IA, IC, and LV and smaller ACV (all P < 0.001) in the long AL subgroup. ConclusionsOur study provides the anterior segment parameters of a large rural Chinese population. IT750, ACV, and LV were found to be the most important factors associated with angle width.

Prevalence and morphological subtype distributions of anaemia in a Chinese rural population: the Henan Rural Cohort study.

This study aimed to evaluate the recent prevalence and the distributions of morphological subtypes of anaemia in the rural population. Anaemia was defined according to the WHO and the Chinese criteria, and the morphological subtypes of anaemia were classified based on the erythrocyte parameters. The age-standardised prevalence was calculated according to the data of the Population Census 2010 in China. A cross-sectional study in Henan Province. 33 585 subjects aged 18-79 years old. The standardised prevalence of anaemia across the WHO and the Chinese definitions was 13·63 % and 5·45 %, respectively. Regardless of which criteria was used, the standardised prevalence of anaemia was higher among women than among men and that increased with age in men, while markedly decreased after menopause in women. There were shifts in morphological patterns of anaemia using the WHO and the Chinese criteria that the standardised prevalence of microcytic anaemia was 3·74 % and 2·97 %, normocytic anaemia was 9·20 % and 2·34 %, and macrocytic anaemia was 0·75 % and 0·14 %, respectively. Besides, there were differences in the influencing factors of anaemia according to different criteria or gender. However, age, education level and renal damage were consistently significantly associated with anaemia in all participants. Anaemia may still be a serious health problem in rural China. It is necessary to reformulate prevention and management strategies to reduce the disease burden of anaemia.

Impacts of solid fuel use versus smoking on life expectancy at age 30 years in the rural and urban Chinese population: a prospective cohort study

The impact of solid fuel use on life expectancy (LE) in less-developed countries remains unclear. We aimed to evaluate the potential impact of household solid fuel use on LE in the rural and urban Chinese population, with the effect of smoking as a reference. We used data from China Kadoorie Biobank (CKB) of 484,915 participants aged 30-79 free of coronary heart disease, stroke, or cancer at baseline. Analyses were performed separately for solid fuel use for cooking, solid fuel use for heating, and smoking, with participants exposed to the other two sources excluded. Solid fuels refer to coal and wood, and clean fuels refer to electricity, gas, and central heating. We used a flexible parametric Royston-Parmar model to estimate hazard ratios of all-cause mortality and predict LE at age30. Totally, 185,077, 95,228, and 230,995 participants were included in cooking-, heating-, and smoking-related analyses, respectively. During a median follow-up of approximately 12.1 years, 12,725, 7,531, and 18,878 deaths were recorded in the respective analysis. Compared with clean fuel users who reported cooking with ventilation, participants who used solid fuels with ventilation and without ventilation had a difference in LE (95% confidence interval [CI]) at age 30 of-1.72 (-2.88,-0.57) and-2.62 (-4.16,-1.05) years for men and-1.33 (-1.85,-0.81) and-1.35 (-2.02,-0.67) years for women, respectively. The difference in LE (95% CI) for heating was-2.23 (-3.51,-0.95) years for men and-1.28 (-2.08,-0.48) years for women. In rural men, the LE reduction (95% CI) related to solid fuel use for cooking (-2.55;-4.51,-0.58) or heating (-3.26;-6.09, 0.44) was more than that related to smoking (-1.71;-2.54,-0.89). Conversely, in urban men, the LE reduction (95% CI) related to smoking (-3.06;-3.56,-2.56) was more than that related to solid fuel use for cooking (-1.28;-2.61, 0.05) and heating (-1.90;-3.16,-0.65). Similar results were observed in women but with a smaller magnitude. In this Chinese population, the harm to LE from household use of solid fuels was greater than that from smoking in rural residents. Conversely, the negative impact of smoking was greater than solid fuel use in urban residents. Our findings highlight the complexity and diversity of the factors affecting LE in less-developed populations. National Natural Science Foundation of China, National Key R&D Program of China, Kadoorie Charitable Foundation, UK Wellcome Trust.

Social trust, social capital, and subjective well-being of rural residents: micro-empirical evidence based on the Chinese General Social Survey (CGSS)

Despite a recent line of research highlighting trust as an important determinant of residents’ happiness in Western countries, empirical evidence regarding the strength of these linkages in the developing world needs to be more comprehensive and conclusive. This paper contributes to this literature by performing a deeper examination into the trust-based explanation of happiness and, particularly, exploring the mediating role of social capital in rural China, where rapid economic growth coexists with gradual and fundamental social changes. Using data from the nationally representative cross-sectional Chinese General Social Survey (CGSS) in 2012, 2013, and 2015, we indicate that trust positively affects the happiness of rural residents and social capital. Our findings support the role of social capital in mediating happiness prediction. Furthermore, additional tests suggest heterogeneous social capital for different ages and degrees of regional market competition. Specifically, the mediating effect of social capital on rural residents over 30 years old (inclusive) and living in high-market competition areas is significant. These findings provide a valuable direction for the government that producing an environment that enhances social networks and facilitates the exchange of social support holds promise for improving the life satisfaction of the rural Chinese population. Trust can significantly improve rural residents’ happiness through increasing social capital. The government should effectively manage the rural social trust crisis and help rural residents build social capital in multiple ways, improving the life satisfaction of rural residents.

Elevated remnant cholesterol increase 6-year type 2 diabetes mellitus onset risk

AimTo investigate the association of remnant cholesterol (RC) with future type 2 diabetes mellitus (T2DM) risk, and to assess the underlying impact of some recognized risk factors on it. MethodsA total of 11468 nondiabetic adults in rural China were recruited in 2007–2008 and followed up in 2013–2014. Logistic regression was used to assess the risk of incident T2DM by quartiles of baseline RC, estimating odds ratios (ORs) and 95% confidence intervals (CIs). Association of the combinations of RC and low-density lipoprotein cholesterol (LDL-C) with T2DM risk were further evaluated. ResultsMultivariable adjusted OR (95% CI) for incident T2DM associated with quartile 4 versus quartile 1 of RC was 2.72 (2.05–3.62). Per 1-standard deviation (SD) increases in RC levels was associated with a 34% higher T2DM risk. However, gender modified the specific association (Pinteraction < 0.05), with the association stronger among females. When combining the low LDL-C and low RC groups as reference, individuals with RC levels ≥ 0.56 mmol/L had more than a 2-fold T2DM risk, regardless of LDL-C level. ConclusionsElevated RC levels increase T2DM risk in rural Chinese populations. In those who cannot control their risk by lowering LDL-C levels, the goal of lipid-lowering therapy can be shifted to RC.

Health-related quality of life assessed by EQ-5D-5L and its determinants among rural adults: result from the Henan rural cohort study.

Evidence on the health-related quality of life (HRQoL) among Chinese rural populations and its influence factors were limited. This study aimed to describe the distribution of HRQoL assessed by EQ-5D-5L and its determinants among a Chinese rural population. A total of 23,510 eligible participants (9542 men and 13,968 women) from the Henan rural cohort study were included. Tobit regression and generalized linear models were performed to investigate the associations between demographic characteristics, lifestyle factors, common chronic diseases, and HRQoL. Of all respondents, the most frequently reported problem was pain/discomfort dimension (23.05%), followed by mobility (12.72%), anxiety/depression (7.77%), and usual activities (6.45%), while the least reported was the self-care dimension (3.84%). The more problems reported in all dimensions with age increased except anxiety/depression. The mean (standard deviation) utility index and VAS score were 0.954 (0.111) and 78.34 (14.80), respectively. The regression analysis indicated that increasing age, former drinkers, poor sleep quality, and suffering from common chronic diseases were significantly associated with low HRQoL, while high education level, average monthly income, and physical activity were significantly associated with high HRQoL. This study described the distribution of HRQoL and its influence factors among the Chinese rural population, which was helpful for medical institutions and policymakers to allocate medical resources and better understand HRQoL among the Chinese rural population.

Indoor air pollution from coal combustion and tobacco smoke during the periconceptional period and risk for neural tube defects in offspring in five rural counties of Shanxi Province, China, 2010–2016

Indoor air pollution may increase the risk for neural tube defects (NTDs) in Chinese rural populations. However, this association remains a subject of debate. We conducted a population-based case-control study of 222 NTD and 517 control mothers recruited between 2010 and 2016 in five rural areas in northern China. An indoor air pollution exposure evaluation index (IAPEEI) was used to evaluate mothers’ exposure to tobacco-sourced and coal-sourced indoor air pollution. Essential characteristics were collected using structured questionnaires within 10 days of delivery. We found that exposure to indoor air pollution (IAPEEI ≥ 1) can lead to 3.41 times the risk of conceiving NTD fetuses compared with the no-exposure group (IAPEEI = 0) (adjusted odds ratio and 95 % confidence interval: 3.41 [2.34–5.02]). The risk increased with increasing IAPEEI score, indicating a clear dose–response trend (P < 0.001). Using a coal stove for heating (especially in the bedroom) and passive smoking are significantly associated with an increased likelihood of NTD occurrence. Exposure to indoor air pollution is a daily reality for rural women in China, and its impact on reproductive health deserves extensive attention.

Is first pregnancy age associated with hypertension in the Chinese rural women population?

The purpose of this study was to investigate the relationship between first pregnancy age and hypertension later in the life of women from Chinese rural areas. In total, 13,493 women were enrolled in the Henan Rural Cohort study. Logistic regression and linear regression were used to evaluate the association between first pregnancy age and hypertension and blood pressure indicators [including systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP)]. The restricted cubic spline was used to examine the dose-response relationship between the first pregnancy age and hypertension or blood pressure indicators. After adjusting for potential confounders, each 1-year increase in first pregnancy age was associated with a 0.221 mmHg increase in SBP values, a 0.153 mmHg increase in DBP values, and a 0.176 mmHg decrease in MAP values (all P < 0.05). The β of SBP, DBP, and MAP showed a trend of first increasing and then decreasing with increasing first pregnancy age and there was no statistical significance after first pregnancy age beyond 33 years on SBP, DBP, and MAP, respectively. A 1-year increment in first pregnancy age was associated with a 2.9% [OR (95% CI): 1.029 (1.010, 1.048)] higher odds of prevalent hypertension. The odds of hypertension increased sharply and then eventually leveled off with an increment of first pregnancy age after adjusting for potential confounders. First pregnancy age might increase the risk of hypertension later in life and might be an independent risk factor for hypertension in women.

Association of triglyceride–glucose index and its 6-year change with risk of hypertension: A prospective cohort study

Background and aimWe aimed to investigate the association of triglyceride–glucose (TyG) index and its dynamic change with risk of hypertension in rural Chinese and, further, to explore whether the TyG index mediates the obesity-related hypertension. Methods and resultsA prospective cohort study, including 10,309 subjects without hypertension at baseline, was conducted in 2007–2008 and followed up in 2013–2014. TyG index was calculated as Ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2]. Logistic regression model was used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs). Mediation analysis was performed to examine the contribution of the TyG index to the association of obesity-hypertension incidence. During a median follow-up of 6 years, 2073 subjects developed hypertension. In multivariate logistic model adjusted for age, sex, alcohol drinking, smoking, physical activity and education, monthly income, family history of hypertension, TC, and HDL-C, the risk of hypertension was 1.14 (1.07–1.22) for per-SD increase in TyG. After additional controlling for obesity, this association was nonsignificant (1.06, 0.99–1.13) and (1.05, 0.99–1.13) for BMI and WC, respectively. Increasing trends were found for hypertension incidence as the TyG change increased, with or without adjustment for obesity (all Ptrend < 0.05). With per-SD increment in TyG change, the risks of hypertension incidence were 1.14 (1.07–1.22) for absolute TyG change, and 1.15 (1.08–1.22) for relative TyG change in multivariate logistic model; the results were significant after further adjustment for BMI or WC, respectively. The TyG index partially mediated the obesity-incident hypertension association: 6.84% for BMI and 6.68% for WC, respectively. ConclusionsElevated TyG index and its dynamic change were positively associated with risk of incident hypertension in rural Chinese population, and the TyG index may play a partially mediating role in obesity-related incident hypertension.

Association of Non-Insulin-Based Insulin Resistance Indices with Risk of Incident Prediabetes and Diabetes in a Chinese Rural Population: A 12-Year Prospective Study.

Three non-insulin-based insulin resistance (IR) indices, ie, triglyceride-glucose (TyG) index, triglyceride-to-high-density lipoprotein cholesterol (TG/HDL-C) ratio, and metabolic score for IR (METS-IR), were considered powerful and simplified alternatives for IR. However, evidence for the association between the three IR indices and incident type 2 diabetes mellitus (T2DM), especially impaired fasting glucose (IFG), remains limited. Therefore, this study aimed to explore the association among IR indices, incident IFG, and T2DM in a rural population cohort. We analyzed data from 2209 adults (aged 24-75 years) at baseline and from 1205 normoglycemic participants who were followed up. Cox proportional hazards models were used to evaluate the associations between the three indices and IFG or T2DM. Restricted cubic spline curves based on the Cox regression model were used to examine the association between baseline indices and incident T2DM. For the baseline data, logistic analyses demonstrated that the TyG index, TG/HDL-C ratio, and METS-IR had a significantly positive correlation with IFG or T2DM after multivariable adjustment. During a median follow-up of 12.17 years, 157 incident cases of IFG and 97 incident cases of T2DM were noted. The risk of T2DM, but not IFG, was strongly associated with the baseline TyG index in the adjusted model, and participants with the TyG index in the third tertile had a higher risk of developing T2DM (adjusted hazards ratio, 2.84; 95% confidence intervals, 1.26-6.37; p for trend <0.001) than those in the lowest tertile (reference). Moreover, a linear relationship was observed between the TyG index and T2DM incidence. The TG/HDL-C ratio and METS-IR had no significant relationship with the risk of IFG or T2DM. The TyG index is more useful than the TG/HDL-C ratio and METS-IR in predicting T2DM in the normoglycemic population.

Association of long-term exposure to PM2.5 constituents with glucose metabolism in Chinese rural population

BackgroundEvidence on the associations of fine particulate matter (PM2.5) constituents and glucose metabolism is limited in resource-limited areas. This study aimed to explore the associations of PM2.5 constituents with glucose metabolism in rural areas, and to further specify the most responsible constituent. MethodsA total of 38,442 adults were recruited from the Henan Rural Cohort Study during 2015–2017. Three-year averaged concentrations of PM2.5 mass and its constituents (black carbon (BC), ammonium (NH4+), nitrate (NO3−), organic matter (OM), inorganic sulfate (SO42−), soil particles (SOIL) and sea salt (SS)) were estimated by a hybrid satellite-based model. Generalized linear model was applied to explore the associations of PM2.5 mass and its constituents with type 2 diabetes mellitus (T2DM), fasting blood glucose (FBG), insulin, and HOMA-β. Proportion and residual analyses were employed to specify the most responsible constituent. ResultsThe adjusted odds ratio (OR) for T2DM associated with 1 μg/m3 increase was 1.02 for PM2.5 mass, 1.28 for BC, 1.15 for NH4+, 1.08 for NO3−, 1.10 for OM, 1.11 for SO42−, and 1.12 for SOIL. Significant associations of PM2.5 mass and its constituents with elevated FBG, decreased insulin and HOMA-β were also observed. Proportion and residual analyses indicated that BC was the most responsible constituent, in which 1 percentage increment in the proportion of BC in PM2.5 corresponded with 1.51-fold risk for T2DM, 0.17 mmol/L increase in FBG, 2.18 μU/mL decrease in insulin, and 38.26 % decrease in HOMA-β; and 1 μg/m3 increment in the PM2.5-adjusted BC corresponded with 1.59-fold risk for T2DM, 0.53 mmol/L increase in FBG, 4.79 μU/mL decrease in insulin, and 91.32 % decrease in HOMA-β. ConclusionsPM2.5 mass and its constituents (BC, NH4+, NO3−, OM, SO42−, SOIL) were associated with T2DM, increased FBG, decreased insulin and HOMA-β, of which BC was most responsible for these associations. Trial registrationThe Henan Rural Cohort Study has been registered at Chinese Clinical Trial Register (Registration number: ChiCTR-OOC-15006699). Date of registration: 06 July 2015. http://www.chictr.org.cn/showproj.aspx?proj=11375.

Identification of the Optimal Model for the Prediction of Diabetic Retinopathy in Chinese Rural Population: Handan Eye Study.

To identify an optimal model for diabetic retinopathy (DR) prediction in Chinese rural population by establishing and comparing different algorithms based on the data from Handan Eye Study (HES). Five algorithms, including multivariable logistic regression (MLR), classification and regression trees (C&RT), support vector machine (SVM), random forests (RF), and gradient boosting machine (GBM), were used to establish DR prediction models with HES data. The performance of the models was assessed based on the adjusted area under the ROC curve (AUROC), sensitivity, specificity, and accuracy. The data on 4752 subjects were used to build the DR prediction model, and among them, 198 patients were diagnosed with DR. The age of the included subjects ranged from 30 to 85 years old, with an average age of 50.9 years (SD = 3.04). The kappa coefficient of the diagnosis between the two ophthalmologists was 0.857. The MLR model revealed that blood glucose, systolic blood pressure, and body mass index were independently associated with the development of DR. The AUROC obtained by GBM (0.952), RF (0.949), and MLR (0.936) was similar and statistically larger than that of CART (0.682) and SVM (0.765). The MLR model exhibited excellent prediction performance and visible equation and thus was the optimal model for DR prediction. Therefore, the MLR model may have the potential to serve as a complementary screening tool for the early detection of DR, especially in remote and underserved areas.

Association between Dietary Protein Intake and Type 2 Diabetes Mellitus in Chinese Rural Elderly Population: A Matched Case-Control Study.

The purpose of this study was to examine the association between dietary protein intake and the risk of type 2 diabetes mellitus (T2DM) among a Chinese rural elderly population. We used the demographic and dietary information of adults over age 65 in the Henan Rural Cohort Study to identify and pair 950 T2DM patients with healthy controls in a 1 : 1 matched case-control study. Dietary data was collected through a Food Frequency Questionnaire. A multivariate logistic regression model was applied to calculate the odds ratio (OR) and 95% confidence interval (CI) of T2DM risk according to protein intake. After adjustment for confounding factors, higher intake of total protein was negatively associated with T2DM risk in the total population (extreme-tertile OR=0.75, 95% CI: 0.58-0.93) and women (extreme-tertile OR=0.84, 95% CI: 0.47-0.93). Multivariate-adjusted ORs for the risk of T2DM in the highest compared with lowest tertile of plant protein intake in the total population and in women were 0.86 (95% CI: 0.60-0.84) and 0.58 (95% CI: 0.36-0.95), respectively. Our results suggest that the protein intake, especially plant protein, has a significant association with the risk of T2DM in rural elderly populations, and the sources of protein may be also important in future guidelines.

The association of serum testosterone with dyslipidemia is mediated by obesity: the Henan Rural Cohort Study.

This study aimed to evaluate the relationships of serum testosterone with dyslipidemia and blood lipid levels and test whether obesity mediated these associations by gender in Chinese rural population. A total of 6150 subjects were finally analyzed in this study. Serum testosterone for each subject was detected by liquid chromatography equipped with tandem mass spectrometry. Logistic regression and linear regression were employed to evaluate the associations of serum testosterone with the prevalence of dyslipidemia and blood lipid levels. Mediation analysis was conducted to identify the mediation effects of obesity on the relationship between serum testosterone and dyslipidemia. After adjusting for multiple confounders, per unit change in serum ln-testosterone levels was associated with a decreased prevalent dyslipidemia in men (odds ratio (OR): 0.785, 95% confidence interval (CI) (0.708, 0.871)). Males with the levels of serum testosterone in the third or fourth quartiles had a 49.4% (OR: 0.506, 95% CI 0.398, 0.644) or 67.1% (OR: 0.329, 95% CI 0.253, 0.428) significantly lower odds of prevalence of dyslipidemia. In addition, a onefold increase in ln-testosterone was related to a 0.043mmol/L (95% CI 0.028, 0.059) increase in high-density lipoprotein cholesterol (HDL-C) in men. Results of the mediation analysis suggested that obesity played a partial role in the association of testosterone with dyslipidemia in men. These findings suggested that serum testosterone levels were negatively associated with lipid levels and prevalent dyslipidemia, and obesity mediated the effects of serum testosterone on dyslipidemia in men, implying that obesity prevention should be highlighted to decrease the prevalence of dyslipidemia related to changes in testosterone levels.

The role of dietary carotenoids in preventing the development of esophageal squamous cell carcinoma.

Background and aims: Experimental studies showed that carotenoids had anti-carcinogenic properties, but epidemiological studies on the association between dietary carotenoids and esophageal squamous cell carcinoma (ESCC) risk were limited, and the findings were inconsistent. This study aimed to investigate the roles of intake of dietary carotenoids in the development of ESCC among a rural Chinese population. Methods: A population-based case-control study was conducted in Southwest China. A total of 915 incident ESCC cases and 925 community-based controls were included. A validated food frequency questionnaire with 76-item was adopted to collect information about dietary consumption. Intake of dietary calories and each carotenoid was calculated according to the China food composition tables. Odds ratios (OR) with 95% confidence intervals (CI) were calculated by a logistic regression model, with adjustments for age, gender, body mass index, family cancer history, cigarette smoking, alcohol drinking, education, marital status, prudent pattern score, and total calories. Results: In comparison to the highest with lowest intake quartiles, intake of total carotenes (OR: 0.70, 95% CI: 0.52-0.96, Ptrend: 0.024), α-carotene (OR: 0.62, 95% CI: 0.46-0.83, Ptrend: 0.014), β-carotene (OR: 0.63, 95% CI: 0.46-0.86, P-trend: 0.005), and the sum of lutein and zeaxanthin (OR: 0.41, 95% CI: 0.29-0.56, Ptrend<0.001) was significantly associated with a decreased risk of ESCC after adjustment for confounders. Conclusions: The results indicated that a higher intake of total carotene, α-carotene, β-carotene, and the sum of lutein and zeaxanthin was associated with a lower risk of ESCC.