Machine learning algorithms to predict colistin-induced nephrotoxicity from electronic health records in patients with multidrug-resistant Gram-negative infection.

OBJECTIVES: Colistin-induced nephrotoxicity prolongs hospitalisation and increases mortality. The study aimed to construct machine learning models to predict colistin-induced nephrotoxicity in patients with multidrug-resistant Gram-negative infection. METHODS: Patients receiving colistin from three hospitals in the Clinical Research Database were included. Data were divided into a derivation cohort (2011-2017) and a temporal validation cohort (2018-2020). Fifteen machine learning models were established by categorical boosting, light gradient boosting machine and random forest. Classifier performances were compared by the sensitivity, F1 score, Matthews correlation coefficient (MCC), area under the receiver operating characteristic (AUROC) curve, and area under the precision-recall curve (AUPRC). SHapley Additive exPlanations plots were drawn to understand feature importance and interactions. RESULTS: The study included 1392 patients, with 360 (36.4%) and 165 (40.9%) experiencing nephrotoxicity in the derivation and temporal validation cohorts, respectively. The categorical boosting with oversampling achieved the highest performance with a sensitivity of 0.860, an F1 score of 0.740, an MCC of 0.533, an AUROC curve of 0.823, and an AUPRC of 0.737. The feature importance demonstrated that the days of colistin use, cumulative dose, daily dose, latest C-reactive protein, and baseline haemoglobin were the most important risk factors, especially for vulnerable patients. A cutoff colistin dose of 4.0 mg/kg body weight/d was identified for patients at higher risk of nephrotoxicity. CONCLUSIONS: Machine learning techniques can be an early identification tool to predict colistin-induced nephrotoxicity. The observed interactions suggest a modification in dose adjustment guidelines. Future geographic and prospective validation studies are warranted to strengthen the real-world applicability.

Is use of a long-term proton pump inhibitor or histamine-2 receptor antagonist a risk factor for iron-deficiency anaemia in Taiwan? A neglected clinical drug-drug interaction.

BACKGROUND: Proton pump inhibitors (PPIs) and histamine-2 receptor (H2) antagonists change the gastric pH and reduce the intestinal absorption of nonheme iron. Case reports and case-control studies have demonstrated that absorption of iron is affected by gastric acidity, but the clinical importance of these drug-drug interactions has remained uncertain. OBJECTIVES: The present case-control study employed 2 million longitudinal claims in 2011-2018 in the Taiwan National Health Insurance Research Database to investigate the impact of PPIs/H2 antagonists on the occurrence of iron-deficiency anaemia (IDA). METHODS: The present study retrospectively compared exposure to PPIs/H2 antagonists for 1 year among 5,326 cases with IDA and 21,304 matched controls. The postdiagnosis prescribing pattern was also calculated to understand current practice. RESULTS: Long-term (≥2 month) use of PPIs/H2 antagonists resulted in a higher risk of developing IDA than noncontinuous use/nonuse of those drugs (adjusted odds ratio [aOR] = 2.36, 95% confidence interval [CI] = 1.94-2.86, P < 0.001). There were significant changes in the postdiagnosis prescribing patterns of PPIs/H2 antagonists. The risk of developing IDA remained significant in the female subgroup (aOR = 2.16, 95% CI = 1.73-2.70, P < 0.001) and was even more prominent in those aged ≥ 50 years (aOR = 2.68, 95% CI = 1.94-3.70, P < 0.05). CONCLUSIONS: This study found that long-term use of PPIs/H2 antagonists increased the risk of developing IDA, and there was strong evidence of prescription pattern adjustments postdiagnosis. Physicians and pharmacists should be aware of this risk when patients are expected to take or have been taking PPIs/H2 antagonists for the long term.

Proton pump inhibitors (PPIs) and histamine-2 receptor (H2) antagonists, 2 kinds of gastric suppressants commonly used for gastroesophageal reflux disease, decrease iron absorption in the gut and thus increase the risk of developing iron-deficiency anaemia (IDA). We constructed a retrospective matched case-control study within the Taiwan National Health Insurance Research Database. The longer period of PPIs/H2 antagonists used, the higher risk of IDA was, with the highest risk in female elderly groups (adjusted odds ratio = 2.68 in females aged ≥ 50). PPI users had a higher risk than H2 antagonist users during the 1-year follow-up. The prescription patterns postdiagnosis of IDA witnessed considerable drops for both groups, with less than a 10th of original users remaining the usages (1.72% and 9.85% taking PPIs and H2 antagonists within 90 days after receiving a diagnosis, respectively). Physicians and pharmacists should be aware of the risk of developing IDA in patients currently undergoing or expected to take long-term gastric acid suppressants.

Explainable Machine Learning Techniques To Predict Amiodarone-Induced Thyroid Dysfunction Risk: Multicenter, Retrospective Study With External Validation.

BACKGROUND: Machine learning offers new solutions for predicting life-threatening, unpredictable amiodarone-induced thyroid dysfunction. Traditional regression approaches for adverse-effect prediction without time-series consideration of features have yielded suboptimal predictions. Machine learning algorithms with multiple data sets at different time points may generate better performance in predicting adverse effects. OBJECTIVE: We aimed to develop and validate machine learning models for forecasting individualized amiodarone-induced thyroid dysfunction risk and to optimize a machine learning-based risk stratification scheme with a resampling method and readjustment of the clinically derived decision thresholds. METHODS: This study developed machine learning models using multicenter, delinked electronic health records. It included patients receiving amiodarone from January 2013 to December 2017. The training set was composed of data from Taipei Medical University Hospital and Wan Fang Hospital, while data from Taipei Medical University Shuang Ho Hospital were used as the external test set. The study collected stationary features at baseline and dynamic features at the first, second, third, sixth, ninth, 12th, 15th, 18th, and 21st months after amiodarone initiation. We used 16 machine learning models, including extreme gradient boosting, adaptive boosting, k-nearest neighbor, and logistic regression models, along with an original resampling method and 3 other resampling methods, including oversampling with the borderline-synthesized minority oversampling technique, undersampling-edited nearest neighbor, and over- and undersampling hybrid methods. The model performance was compared based on accuracy; Precision, recall, F1-score, geometric mean, area under the curve of the receiver operating characteristic curve (AUROC), and the area under the precision-recall curve (AUPRC). Feature importance was determined by the best model. The decision threshold was readjusted to identify the best cutoff value and a Kaplan-Meier survival analysis was performed. RESULTS: The training set contained 4075 patients from Taipei Medical University Hospital and Wan Fang Hospital, of whom 583 (14.3%) developed amiodarone-induced thyroid dysfunction, while the external test set included 2422 patients from Taipei Medical University Shuang Ho Hospital, of whom 275 (11.4%) developed amiodarone-induced thyroid dysfunction. The extreme gradient boosting oversampling machine learning model demonstrated the best predictive outcomes among all 16 models. The accuracy; Precision, recall, F1-score, G-mean, AUPRC, and AUROC were 0.923, 0.632, 0.756, 0.688, 0.845, 0.751, and 0.934, respectively. After readjusting the cutoff, the best value was 0.627, and the F1-score reached 0.699. The best threshold was able to classify 286 of 2422 patients (11.8%) as high-risk subjects, among which 275 were true-positive patients in the testing set. A shorter treatment duration; higher levels of thyroid-stimulating hormone and high-density lipoprotein cholesterol; and lower levels of free thyroxin, alkaline phosphatase, and low-density lipoprotein were the most important features. CONCLUSIONS: Machine learning models combined with resampling methods can predict amiodarone-induced thyroid dysfunction and serve as a support tool for individualized risk prediction and clinical decision support.