Instrumental variable-based high-dimensional mediation analysis with unmeasured confounders for survival data in the observational epigenetic study.

Background: High dimensional mediation analysis is frequently conducted to explore the role of epigenetic modifiers between exposure and health outcome. However, the issue of high dimensional mediation analysis with unmeasured confounders for survival analysis in observational study has not been well solved. Methods: In this study, we proposed an instrumental variable based approach for high dimensional mediation analysis with unmeasured confounders in survival analysis for epigenetic study. We used the Sobel's test, the Joint test, and the Bootstrap method to test the mediation effect. A comprehensive simulation study was conducted to decide the best test strategy. An empirical study based on DNA methylation data of lung cancer patients was conducted to illustrate the performance of the proposed method. Results: Simulation study suggested that the proposed method performed well in the identifying mediating factors. The estimation of the mediation effect by the proposed approach is also reliable with less bias compared with the classical approach. In the empirical study, we identified two DNA methylation signatures including cg21926276 and cg26387355 with a mediation effect of 0.226 (95%CI: 0.108-0.344) and 0.158 (95%CI: 0.065-0.251) between smoking and lung cancer using the proposed approach. Conclusion: The proposed method obtained good performance in simulation and empirical studies, it could be an effective statistical tool for high dimensional mediation analysis.

Explainable Machine Learning with Pairwise Interactions for Predicting Conversion from Mild Cognitive Impairment to Alzheimer's Disease Utilizing Multi-Modalities Data.

BACKGROUND: Predicting cognition decline in patients with mild cognitive impairment (MCI) is crucial for identifying high-risk individuals and implementing effective management. To improve predicting MCI-to-AD conversion, it is necessary to consider various factors using explainable machine learning (XAI) models which provide interpretability while maintaining predictive accuracy. This study used the Explainable Boosting Machine (EBM) model with multimodal features to predict the conversion of MCI to AD during different follow-up periods while providing interpretability. METHODS: This retrospective case-control study is conducted with data obtained from the ADNI database, with records of 1042 MCI patients from 2006 to 2022 included. The exposures included in this study were MRI biomarkers, cognitive scores, demographics, and clinical features. The main outcome was AD conversion from aMCI during follow-up. The EBM model was utilized to predict aMCI converting to AD based on three feature combinations, obtaining interpretability while ensuring accuracy. Meanwhile, the interaction effect was considered in the model. The three feature combinations were compared in different follow-up periods with accuracy, sensitivity, specificity, and AUC-ROC. The global and local explanations are displayed by importance ranking and feature interpretability plots. RESULTS: The five-years prediction accuracy reached 85% (AUC = 0.92) using both cognitive scores and MRI markers. Apart from accuracies, we obtained features' importance in different follow-up periods. In early stage of AD, the MRI markers play a major role, while for middle-term, the cognitive scores are more important. Feature risk scoring plots demonstrated insightful nonlinear interactive associations between selected factors and outcome. In one-year prediction, lower right inferior temporal volume (<9000) is significantly associated with AD conversion. For two-year prediction, low left inferior temporal thickness (<2) is most critical. For three-year prediction, higher FAQ scores (>4) is the most important. During four-year prediction, APOE4 is the most critical. For five-year prediction, lower right entorhinal volume (<1000) is the most critical feature. CONCLUSIONS: The established glass-box model EBMs with multimodal features demonstrated a superior ability with detailed interpretability in predicting AD conversion from MCI. Multi features with significant importance were identified. Further study may be of significance to determine whether the established prediction tool would improve clinical management for AD patients.

Predicting Six-Month Re-Admission Risk in Heart Failure Patients Using Multiple Machine Learning Methods: A Study Based on the Chinese Heart Failure Population Database.

Since most patients with heart failure are re-admitted to the hospital, accurately identifying the risk of re-admission of patients with heart failure is important for clinical decision making and management. This study plans to develop an interpretable predictive model based on a Chinese population for predicting six-month re-admission rates in heart failure patients. Research data were obtained from the PhysioNet portal. To ensure robustness, we used three approaches for variable selection. Six different machine learning models were estimated based on selected variables. The ROC curve, prediction accuracy, sensitivity, and specificity were used to evaluate the performance of the established models. In addition, we visualized the optimized model with a nomogram. In all, 2002 patients with heart failure were included in this study. Of these, 773 patients experienced re-admission and a six-month re-admission incidence of 38.61%. Based on evaluation metrics, the logistic regression model performed best in the validation cohort, with an AUC of 0.634 (95%CI: 0.599-0.646) and an accuracy of 0.652. A nomogram was also generated. The established prediction model has good discrimination ability in predicting. Our findings are helpful and could provide useful information for the allocation of healthcare resources and for improving the quality of survival of heart failure patients.

Healthy life expectancy for 202 countries up to 2030: Projections with a Bayesian model ensemble.

Background: Healthy life expectancy (HLE) projections are required for optimising social and health service management in the future. Existing studies on the topic were usually conducted by selecting a single model for analysis. We thus aimed to use an ensembled model to project the future HLE for 202 countries/region. Methods: We obtained data on age-sex-specific HLE and the sociodemographic index (SDI) level of 202 countries from 1990 to 2019 from the Global Burden of Disease (GBD) database and used a probabilistic Bayesian model comprised of 21 forecasting models to predict their HLE in 2030. Results: In general, HLE is projected to increase in all 202 countries, with the least probability of 82.4% for women and 81.0% for men. Most of the countries with the lowest projected HLE would be located in Africa. Women in Singapore have the highest projected HLE in 2030, with a 94.5% probability of higher than 75.2 years, which is the highest HLE in 2019 across countries. Maldives, Kuwait, and China are projected to have a probability of 49.3%, 41.2% and 31.6% to be the new entries of the top ten countries with the highest HLE for females compared with 2019. Men in Singapore are projected to have the highest HLE at birth in 2030, with a 93.4% probability of higher than 75.2 years. Peru and Maldives have a probability of 48.7% and 35.3% being new top ten countries in male's HLE. The female advantage in HLE will shrink by 2030 in 117 countries, especially in most of the high SDI and European countries. Conclusions: HLE will likely continue to increase in most countries and regions worldwide in the future. More attention needs to be paid to combatting obesity, chronic diseases, and specific infectious diseases, especially in African and some Pacific Island countries. Although gender gaps may not be fully bridged, HLE could partially mitigate and even eliminate them through economic development and improvements in health care.

Cardiovascular Disease Mortality Attributable to Low Whole-Grain Intake in CHINA: An Age-Period-Cohort and Joinpoint Analysis.

Cardiovascular disease (CVD) is the leading cause of death worldwide. Low whole-grain intake is found to be one of the most important risk factors for cardiovascular disease development and progression. In this study, we focused on exploring the long-term trends of low whole-grain intake attributed to cardiovascular disease mortality in China during 1990-2019 and relative gender differences. Study data were obtained from the Global Burden of Disease (GBD) 2019 study. We used the age-period-cohort model to estimate the adjusted effect of age, period, and cohorts. Annual and average annual percentage changes were estimated by joinpoint regression analysis. We observed an increasing trend with a net drift of 1.208% for males and 0.483% for males per year. The longitudinal age curve suggested that the attributed rate increased for both genders. Period and cohort effects all suggested that the risk for males showed an increased trend that was higher than that of females. Our findings suggest that males and senior-aged people were at a higher risk of cardiovascular disease mortality attributed to low whole-grain intake. Effective strategies are needed to enhance people's health consciousness, and increasing whole-grain intake may achieve a better preventive effect for cardiovascular disease.

The long-term trend of Parkinson's disease incidence and mortality in China and a Bayesian projection from 2020 to 2030.

Background: Parkinson's disease is a disabling degenerative disease of the central nervous system that occurs mainly in elderly people. The changes in the incidence and mortality of Parkinson's disease at the national level in China over the past three decades have not been fully explored. Methods: Research data were obtained from the Global Burden of Disease 2019 study. The trends of crude and age-standardized incidence and mortality rates by gender of Parkinson's disease in China were analyzed with the age-period-cohort model and the Joinpoint regression analysis. The effects of age, time period, and birth cohort on the incidence and mortality of Parkinson's disease were estimated. The gender- and age-specific incidence and mortality rates of Parkinson's disease from 2020 to 2030 were projected using the Bayesian age-period-cohort model with integrated nested Laplace approximations. Results: From 1990 to 2019, the annual percentage change of the age-standardized incidence rate was 0.8% (95% CI: 0.7%-0.8%) for males and 0.2% (95% CI, 0.2-0.2%) for females. And the age-standardized mortality rate for males was 2.9% (95% CI: 2.6%-3.2%) and 1.8% (95% CI: 1.5%-2.1%) for females. The results of the age-period-cohort analysis suggested that the risk and burden of Parkinson's disease continued to increase for the last several decades. Projection analysis suggested that the overall Parkinson's disease incidence will continue to increase for the next decades. It was projected that China would have 4.787 million Parkinson's patients by the year 2030, however, the mortality of Parkinson's disease for both genders in China may keep decreasing. Conclusion: Though the mortality risk may decrease, Parkinson's disease continues to become more common for both genders in China, especially in the senior-aged population. The burden associated with Parkinson's disease would continue to grow. Urgent interventions should be implemented to reduce the burden of Parkinson's disease in China.

Long-Time Trend of Colorectal Cancer Mortality Attributable to High Processed Meat Intake in China and a Bayesian Projection from 2020 to 2030: A Model-Based Study.

Colorectal cancer is among the leading causes of cancer worldwide. Processed meat was known to be positively associated with a higher risk of gastrointestinal cancer. This study focused on the long-time trends of colorectal cancer mortality attributable to high processed meat intake in China from 1990 to 2019 and the projection for the next decade based on data obtained from the Global Burden of Disease 2019 study. We used an age-period-cohort model to fit the long-time trend. The joinpoint model was conducted to estimate the average and annual change of the attributable mortality. The Bayesian age-period-cohort model was used to project the crude attributable mortality from 2020 to 2030. An upward trend in colorectal cancer mortality attributable to high processed meat intake was observed for both sexes in China from 1990 to 2019, with an overall net drift of 4.009% for males and 2.491% for females per year. Projection analysis suggested that the burden of colorectal cancer incidence and mortality would still be high. Our findings suggested that colorectal cancer death attributable to high processed meat intake is still high in China, and elderly males were at higher risk. Gradually decreasing the intake of processed meat could be an effective way to reduce colorectal cancer mortality.

Stroke mortality attributable to high red meat intake in China and South Korea: An age-period-cohort and joinpoint analysis.

The high intake of red meat is well recognized as a major health concern worldwide. It has been recognized as a risk factor for several non-communicable chronic diseases, including stroke. However, previously published studies have not performed a comprehensive analysis of the long-time trend of stroke mortality attributable to high red meat intake in China and South Korea, two countries with similar dietary patterns and changing trends. Therefore, this study aimed to reveal the influence of age, time period, and birth cohort on long-term trends of stroke mortality attributable to high red meat intake and relative gender differences in China and South Korea. Data were obtained from the Global Burden of Disease 2019 database. The age-period-cohort model was used to estimate the effect of age, time period, and birth cohort. The average and annual percent changes were estimated using the joinpoint regression analysis. Results indicated that the overall attributable age-standardized mortality rates of stroke in China decreased by 1.0% (P < 0.05) for female and 0.1% (P > 0.05) for male individuals, compared with a decrease of 4.9% for female and 3.7% for male individuals in South Korea (both P < 0.05). Age-period-cohort analysis revealed that the attributable stroke mortality decreased along with the time period, and increased along with age. Significant gender differences were observed, male individuals in both countries were at higher risk than their female counterparts, especially in China. Joinpoint analysis suggested that the attributable stroke mortality for both genders in South Korea and female individuals in China showed a decreasing trend, while it is stable for male individuals in China. Although prominent reductions were observed during the past decades, the attributable stroke mortality risk in China and South Korea is still high. Our findings indicate that controlling the intake of red meat may be a cost-effective strategy to reduce stroke mortality risk and the corresponding disease burden, especially for Chinese male individuals.