Global, regional, and national time trends in mortality for stroke, 1990-2019: An age-period-cohort analysis for the global burden of disease 2019 study and implications for stroke prevention

Explainable machine learning for predicting neurological outcome in hemorrhagic and ischemic stroke patients in critical care

Aim

The objective of this study is to develop accurate machine learning (ML) models for predicting the neurological status at hospital discharge of critically ill patients with hemorrhagic and ischemic stroke and identify the risk factors associated with the neurological outcome of stroke, thereby providing healthcare professionals with enhanced clinical decision-making guidance.

Materials and methods

Data of stroke patients were extracted from the eICU Collaborative Research Database (eICU-CRD) for training and testing sets and the Medical Information Mart for Intensive Care IV (MIMIC IV) database for external validation. Four machine learning models, namely gradient boosting classifier (GBC), logistic regression (LR), multi-layer perceptron (MLP), and random forest (RF), were used for prediction of neurological outcome. Furthermore, shapley additive explanations (SHAP) algorithm was applied to explain models visually.

Results

A total of 1,216 hemorrhagic stroke patients and 954 ischemic stroke patients from eICU-CRD and 921 hemorrhagic stroke patients 902 ischemic stroke patients from MIMIC IV were included in this study. In the hemorrhagic stroke cohort, the LR model achieved the highest area under curve (AUC) of 0.887 in the test cohort, while in the ischemic stroke cohort, the RF model demonstrated the best performance with an AUC of 0.867 in the test cohort. Further analysis of risk factors was conducted using SHAP analysis and the results of this study were converted into an online prediction tool.

Conclusion

ML models are reliable tools for predicting hemorrhagic and ischemic stroke neurological outcome and have the potential to improve critical care of stroke patients. The summarized risk factors obtained from SHAP enable a more nuanced understanding of the reasoning behind prediction outcomes and the optimization of the treatment strategy.

Secondary Prevention via Case Managers in Stroke Patients: A Cost-Effectiveness Analysis of Claims Data from German Statutory Health Insurance Providers

Strokes remain a leading cause of death and disability worldwide. The STROKE OWL study evaluated a novel case management approach for patients with stroke (modified Rankin Scale 0-4) or transient ischemic attack (TIA) who received support across healthcare settings and secondary prevention training from case managers for one year. The primary aim of this quasi-experimental study was a reduction in stroke recurrence. Here, we report the results of a health economic analysis of the STROKE OWL study, conducted in accordance with CHEERS guidelines. The calculations were based on claims data of cooperating statutory health insurance companies. In addition to a regression analysis for cost comparison, the incremental cost-effectiveness ratio was determined, and a probabilistic sensitivity analysis was carried out. In total, 1167 patients per group were included in the analysis. The intervention group incurred 32.3% higher direct costs (p < 0.001) than the control group. With a difference of EUR 1384.78 (95% CI: [1.2384-1.4143], p < 0.0001) and a 5.32% increase in hazards for the intervention group (HR = 1.0532, 95% CI: [0.7869-1.4096], p = 0.7274) resulting in an ICER of EUR 260.30, we found that the case management intervention dominated in the total stroke population, even for an arbitrarily high willingness to pay. In the TIA subgroup, however, the intervention was cost-effective even for a low willingness to pay. Our results are limited by small samples for both TIA and severe stroke patients and by claims data heterogeneity for some cost components, which had to be excluded from the analysis. Future research should investigate the cost-effectiveness of case management interventions for both severe stroke and TIA populations using appropriate data.

Big Data-Enabled Analysis of Factors Affecting Medical Expenditure in the Cerebral Infarction of a Developing City in Western China

Purpose

Cerebral infarction (CI) has been one of the leading causes of death in China since 2017, and controlling the medical expenses of this disease is an urgent issue for the Chinese government. This study aims to explore the important factors that affect the hospitalization expenses of CI patients and to provide a scientific basis for establishing a reasonable reimbursement mechanism and hospitalization expense standard for CI patients.

Methods

Data from 109,314 inpatients from the Healthcare Security Administration of Chengdu in western China from January 2016 to December 2018 were utilized. Descriptive statistical analysis was used for variable characteristic analysis. The Mann-Whitney test and Kruskal-Wallis test were used for single-factor analysis, and multiple linear stepwise regression was used for single-factor analysis and multiple-factor analysis.

Results

This study found that the average direct economic burden of CI in Chengdu was approximately 10,569 Chinese yuan (CNY), about 1450 US dollars, the average length of stay (LOS) was 14.47 days, the indirect economic burden was approximately 2817 CNY, and the total economic burden was 13,386 CNY for a CI inpatient. Gender, insurance type, grade of medical institution, the level of payment type, age, LOS, and complications and comorbidities (CCs) are the most important factors affecting CI medical costs.

Conclusion

Citizens should improve their lifestyle habits to reduce disease risk to avoid the associated medical and economic burdens. Hospitals should improve their medical technology to decrease the LOS and reduce direct medical costs. The government should actively promote the hierarchical diagnosis and treatment policy to reduce the waste of medical resources caused by low-acuity patients going to high-level hospitals for treatment. The National Healthcare Security Administration should optimize the medical insurance payment method and establish a corresponding mechanism to reduce the occurrence of excessive medical treatments such as overuse.