Implementing a new tuberculosis surveillance system in Zhejiang, Jilin and Ningxia: improvements, challenges and implications for China's National Health Information System

Interpretable machine learning in predicting drug-induced liver injury among tuberculosis patients: model development and validation study

BACKGROUND

The objective of this research was to create and validate an interpretable prediction model for drug-induced liver injury (DILI) during tuberculosis (TB) treatment.

METHODS

A dataset of TB patients from Ningbo City was used to develop models employing the eXtreme Gradient Boosting (XGBoost), random forest (RF), and the least absolute shrinkage and selection operator (LASSO) logistic algorithms. The model's performance was evaluated through various metrics, including the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPR) alongside the decision curve. The Shapley Additive exPlanations (SHAP) method was used to interpret the variable contributions of the superior model.

RESULTS

A total of 7,071 TB patients were identified from the regional healthcare dataset. The study cohort consisted of individuals with a median age of 47 years, 68.0% of whom were male, and 16.3% developed DILI. We utilized part of the high dimensional propensity score (HDPS) method to identify relevant variables and obtained a total of 424 variables. From these, 37 variables were selected for inclusion in a logistic model using LASSO. The dataset was then split into training and validation sets according to a 7:3 ratio. In the validation dataset, the XGBoost model displayed improved overall performance, with an AUROC of 0.89, an AUPR of 0.75, an F1 score of 0.57, and a Brier score of 0.07. Both SHAP analysis and XGBoost model highlighted the contribution of baseline liver-related ailments such as DILI, drug-induced hepatitis (DIH), and fatty liver disease (FLD). Age, alanine transaminase (ALT), and total bilirubin (Tbil) were also linked to DILI status.

CONCLUSION

XGBoost demonstrates improved predictive performance compared to RF and LASSO logistic in this study. Moreover, the introduction of the SHAP method enhances the clinical understanding and potential application of the model. For further research, external validation and more detailed feature integration are necessary.

Analysis of drug resistance in pulmonary tuberculosis patients with positive sputum tuberculosis culture in Northeast China

Objective: The objective of this study is to determine the drug resistance status of pulmonary tuberculosis patients in Jilin Province. Methods: A retrospective survey was conducted on 395 sputum culture TB-positive patients admitted to the tuberculosis hospital in Jilin Province in 2019. Sputum samples were cultured in acidic Roche medium. Drug sensitivity testing was conducted using the proportional method. Sensitivity was reported if the percentage of drug resistance was less than 1%, and resistance was reported if the percentage was ≥1%. Statistical analysis was performed using SPSS 22.0. Results: 395 tuberculosis patients with positive sputum tuberculosis culture were included in the study, with 102 being initially treated and 293 being retreated. The study population consisted of 283 males and 112 females. Sex, age, nationality, occupation, marital status, diabetes comorbidity, initial treatment, normal health status, BCG vaccine vaccination, smoking, and alcohol consumption were considered as factors that may affect the rate of multidrug resistance. And only the history of treatment (initial treatment) was associated with multidrug resistance (p = 0.032). This indicates that retreatment is the most significant risk factor for the occurrence of multidrug resistance in tuberculosis. The multidrug resistance rate in retreated patients is 3.764 times higher than that in initially treated patients. Conclusion: The prevalence of multidrug-resistant is higher in retreated patients compared to initially treated patients in the study population. Multidrug resistance is only associated with the treatment history (initial retreatment) and not with other factors.

Evaluation of the multidrug-resistant tuberculosis surveillance system in Maputo City, Mozambique in the period 2017-2018

Introduction

multidrug-resistant tuberculosis (MDR-TB) remains a public health problem worldwide. In Mozambique, cases of MDR-TB have increased annually. In 2018, 1,206 cases were reported, as compared to 943 cases in 2017. The aim of this study was to assess the surveillance system for multidrug-resistant tuberculosis in Maputo City.

Methods

an extract from the national database was considered for a cut-out of the City of Maputo in the period 2017-2018; the study was conducted per the guidelines of the Centers for Disease Control and Prevention, where the description of the system was carried out, and evaluation of the attributes. Each attribute was evaluated according to the established criteria and parameters.

Results

the surveillance system is based on the collection of data in health centers. Four hundred and six cases of MDR-TB were notified, of which 56.8% (231/406) were male and 95.9% (386/406) were ≥15 years. The system was complex with 4 levels of information transmission. With regard to flexibility, there was no changing the variables in the database. Acceptability was good. The quality of the data was regular with discrepancy of data of 14.5%. The system was considered stable as there was no system interruption. Timeliness with case notification monthly. The system sensitivity was 72.9%, the positive predictive value (PPV) was 2.3% and regarding utility the system has fulfilled its objectives.

Conclusion

the system was not flexible, the data quality was regular, had moderate sensitivity and low positive predictive value. Continuous assessment of data and scale up the diagnosis for the detection of cases of MDR-TB is recommended.