Development and economic assessment of machine learning models to predict glycosylated hemoglobin in type 2 diabetes.

Background: Glycosylated hemoglobin (HbA1c) is recommended for diagnosing and monitoring type 2 diabetes. However, the monitoring frequency in real-world applications has not yet reached the recommended frequency in the guidelines. Developing machine learning models to screen patients with poor glycemic control in patients with T2D could optimize management and decrease medical service costs. Methods: This study was carried out on patients with T2D who were examined for HbA1c at the Sichuan Provincial People's Hospital from April 2018 to December 2019. Characteristics were extracted from interviews and electronic medical records. The data (excluded FBG or included FBG) were randomly divided into a training dataset and a test dataset with a radio of 8:2 after data pre-processing. Four imputing methods, four screening methods, and six machine learning algorithms were used to optimize data and develop models. Models were compared on the basis of predictive performance metrics, especially on the model benefit (MB, a confusion matrix combined with economic burden associated with therapeutic inertia). The contributions of features were interpreted using SHapley Additive exPlanation (SHAP). Finally, we validated the sample size on the best model. Results: The study included 980 patients with T2D, of whom 513 (52.3%) were defined as positive (need to perform the HbA1c test). The results indicated that the model trained in the data (included FBG) presented better forecast performance than the models that excluded the FBG value. The best model used modified random forest as the imputation method, ElasticNet as the feature screening method, and the LightGBM algorithms and had the best performance. The MB, AUC, and AUPRC of the best model, among a total of 192 trained models, were 43475.750 (¥), 0.972, 0.944, and 0.974, respectively. The FBG values, previous HbA1c values, having a rational and reasonable diet, health status scores, type of manufacturers of metformin, interval of measurement, EQ-5D scores, occupational status, and age were the most significant contributors to the prediction model. Conclusion: We found that MB could be an indicator to evaluate the model prediction performance. The proposed model performed well in identifying patients with T2D who need to undergo the HbA1c test and could help improve individualized T2D management.

Burden of gastroesophageal reflux disease in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of disease study 2019.

INTRODUCTION: For effective preventive strategies against GORD (gastro-esophageal reflux disease), we assessed the GORD burden from 1990 to 2019. METHODS: The burden of GORD between 1990 and 2019 was evaluated globally, regionally, and nationally. Using ASIR (age-standardized incidence), ASYLDs (age-standardized years lived with disabilitys), we compared them to the GBD world population per 100,000. The estimates were based on 95% uncertainty intervals (UIs). The AAPC (average annual percent change) in incidence, YLDs, along with prevalence rates with associated 95% CIs were estimated. RESULTS: Data to estimate the burden of GORD are scarce till now. The global ASIR of GORD in 2019 was 3792.79 per 100,000, an increase AAPC of 0.112% from 1990. The prevalence of GORD increased with a AAPC of 0.096% to 9574.45 per 100,000. Global ASYLDs in 2019 was 73.63, an increase AAPC of 0.105% from 1990. The GORD burden varies greatly depending on the development level and geographical location. USA demonstrated the most obvious decreasing trend in burden of GORD, while Sweden had an increasing trend. That the increase in GORD YLDs was mediated primarily by the growth and aging of population, was revealed by decomposition analyses. There was an inverse relationship between SDI (socio-demographic index) and GORD-burden. Frontier analyses revealed significant scope of improvement in the status of development at all levels. CONCLUSION: GORD is a public health challenge, especially in Latin America. Some SDI quintiles had declining rates, while some countries experienced increased rates. Thus, resources should be allocated for preventative measures based on country-specific estimates.

Disease burden of primary gallbladder and biliary tract cancers associated with body mass index in 195 countries and territories, 1990-2017: A systematic analysis for the Global Burden of Disease Study 2017.

OBJECTIVE: To investigate the incidence of primary gallbladder and biliary tract cancer, mortality and disability-adjusted life years (DALY) of the global burden from 1990 to 2017. METHODS: Data of 195 countries and territories from 1990 to 2017 were extracted from the Global Health Data Exchange. The age-standardized incidence rate (ASIR) and estimated annual percentage change (EAPC) were employed to quantify trends in the incidence of primary gallbladder and biliary tract cancer. The age-standardized death rate (ASDR), age-standardized DALY and their corresponding EAPC were used to evaluate mortality trends. RESULTS: The global incidence of primary gallbladder and biliary tract cancer rose by 75.9% from 119 900 cases in 1990 to 210 900 cases in 2017. The highest ASIR was observed in Chile (10.8 per 100 000 in 2017), followed by Japan and South Korea. Regions with the highest social development index (SDI) quintile also had the highest death cases associated with primary gallbladder and biliary tract cancer in 2017 (60 100, 95% UI 55 800-62 700). A high body mass index (BMI) was found to be closely related to age-standardized deaths and age-standardized DALY in most of the regions analyzed. CONCLUSIONS: Primary gallbladder and biliary tract cancer remains a serious threat to global public health, especially in high-SDI countries. The ASDR and age-standardized DALY decreased from 1990 to 2017. A high BMI may be associated with this cancer burden.