Computerized Mortality Prediction for Community-acquired Pneumonia at 117 Veterans Affairs Medical Centers

A simplified pneumonia severity index (PSI) for clinical outcome prediction in COVID-19

BACKGROUND

The Pneumonia Score Index (PSI) was developed to estimate the risk of dying within 30 days of presentation for community-acquired pneumonia patients and is a strong predictor of 30-day mortality after COVID-19. However, three of its required 20 variables (skilled nursing home, altered mental status and pleural effusion) are not discreetly available in the electronic medical record (EMR), resulting in manual chart review for these 3 factors. The goal of this study is to compare a simplified 17-factor version (PSI-17) to the original (denoted PSI-20) in terms of prediction of 30-day mortality in COVID-19.

METHODS

In this retrospective cohort study, the hospitalized patients with confirmed SARS-CoV-2 infection between 2/28/20-5/28/20 were identified to compare the predictive performance between PSI-17 and PSI-20. Correlation was assessed between PSI-17 and PSI-20, and logistic regressions were performed for 30-day mortality. The predictive abilities were compared by discrimination, calibration, and overall performance.

RESULTS

Based on 1,138 COVID-19 patients, the correlation between PSI-17 and PSI-20 was 0.95. Univariate logistic regression showed that PSI-17 had performance similar to PSI-20, based on AUC, ICI and Brier Score. After adjusting for confounding variables by multivariable logistic regression, PSI-17 and PSI-20 had AUCs (95% CI) of 0.85 (0.83-0.88) and 0.86 (0.84-0.89), respectively, indicating no significant difference in AUC at significance level of 0.05.

CONCLUSION

PSI-17 and PSI-20 are equally effective predictors of 30-day mortality in terms of several performance metrics. PSI-17 can be obtained without the manual chart review, which allows for automated risk calculations within an EMR. PSI-17 can be easily obtained and may be a comparable alternative to PSI-20.

Machine learning-based prediction of in-ICU mortality in pneumonia patients

Conventional severity-of-illness scoring systems have shown suboptimal performance for predicting in-intensive care unit (ICU) mortality in patients with severe pneumonia. This study aimed to develop and validate machine learning (ML) models for mortality prediction in patients with severe pneumonia. This retrospective study evaluated patients admitted to the ICU for severe pneumonia between January 2016 and December 2021. The predictive performance was analyzed by comparing the area under the receiver operating characteristic curve (AU-ROC) of ML models to that of conventional severity-of-illness scoring systems. Three ML models were evaluated: (1) logistic regression with L2 regularization, (2) gradient-boosted decision tree (LightGBM), and (3) multilayer perceptron (MLP). Among the 816 pneumonia patients included, 223 (27.3%) patients died. All ML models significantly outperformed the Simplified Acute Physiology Score II (AU-ROC: 0.650 [0.584-0.716] vs 0.820 [0.771-0.869] for logistic regression vs 0.827 [0.777-0.876] for LightGBM 0.838 [0.791-0.884] for MLP; P < 0.001). In the analysis for NRI, the LightGBM and MLP models showed superior reclassification compared with the logistic regression model in predicting in-ICU mortality in all length of stay in the ICU subgroups; all age subgroups; all subgroups with any APACHE II score, PaO₂/FiO₂ ratio < 200; all subgroups with or without history of respiratory disease; with or without history of CVA or dementia; treatment with mechanical ventilation, and use of inotropic agents. In conclusion, the ML models have excellent performance in predicting in-ICU mortality in patients with severe pneumonia. Moreover, this study highlights the potential advantages of selecting individual ML models for predicting in-ICU mortality in different subgroups.

Performance of Machine Learning Algorithms for Predicting Disease Activity in Inflammatory Bowel Disease

This study aimed to explore the effectiveness of predicting disease activity in patients with inflammatory bowel disease (IBD), using machine learning (ML) models. A retrospective research was undertaken on IBD patients who were admitted into the First Affiliated Hospital of Wenzhou Medical University between September 2011 and September 2019. At first, data were randomly split into a 3:1 ratio of training to test set. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to reduce the dimension of variables. These variables were used to generate seven ML algorithms, namely random forests (RFs), adaptive boosting (AdaBoost), K-nearest neighbors (KNNs), support vector machines (SVMs), naïve Bayes (NB), ridge regression, and eXtreme gradient boosting (XGBoost) to train to predict disease activity in IBD patients. SHapley Additive exPlanation (SHAP) analysis was performed to rank variable importance. A total of 876 participants with IBD, consisting of 275 ulcerative colitis (UC) and 601 Crohn's disease (CD), were retrospectively enrolled in the study. Thirty-three variables were obtained from the clinical characteristics and laboratory tests of the participants. Finally, after LASSO analysis, 11 and 5 variables were screened out to construct ML models for CD and UC, respectively. All seven ML models performed well in predicting disease activity in the CD and UC test sets. Among these ML models, SVM was more effective in predicting disease activity in the CD group, whose AUC reached 0.975, sensitivity 0.947, specificity 0.920, and accuracy 0.933. AdaBoost performed best for the UC group, with an AUC of 0.911, sensitivity 0.844, specificity 0.875, and accuracy 0.855. ML algorithms were available and capable of predicting disease activity in IBD patients. Based on clinical and laboratory variables, ML algorithms demonstrate great promise in guiding physicians' decision-making.

Trends in Illness Severity, Hospitalization, and Mortality for Community-Onset Pneumonia at 118 US Veterans Affairs Medical Centers

BACKGROUND

Deaths from pneumonia were decreasing globally prior to the COVID-19 pandemic, but it is unclear whether this was due to changes in patient populations, illness severity, diagnosis, hospitalization thresholds, or treatment. Using clinical data from the electronic health record among a national cohort of patients initially diagnosed with pneumonia, we examined temporal trends in severity of illness, hospitalization, and short- and long-term deaths.

DESIGN

Retrospective cohort PARTICIPANTS: All patients >18 years presenting to emergency departments (EDs) at 118 VA Medical Centers between 1/1/2006 and 12/31/2016 with an initial clinical diagnosis of pneumonia and confirmed by chest imaging report.

EXPOSURES

Year of encounter.

MAIN MEASURES

Hospitalization and 30-day and 90-day mortality. Illness severity was defined as the probability of each outcome predicted by machine learning predictive models using age, sex, comorbidities, vital signs, and laboratory data from encounters during years 2006-2007, and similar models trained on encounters from years 2015 to 2016. We estimated the changes in hospitalizations and 30-day and 90-day mortality between the first and the last 2 years of the study period accounted for by illness severity using time covariate decompositions with model estimates.

RESULTS

Among 196,899 encounters across the study period, hospitalization decreased from 71 to 63%, 30-day mortality 10 to 7%, 90-day mortality 16 to 12%, and 1-year mortality 29 to 24%. Comorbidity risk increased, but illness severity decreased. Decreases in illness severity accounted for 21-31% of the decrease in hospitalizations, and 45-47%, 32-24%, and 17-19% of the decrease in 30-day, 90-day, and 1-year mortality. Findings were similar among underrepresented patients and those with only hospital discharge diagnosis codes.

CONCLUSIONS

Outcomes for community-onset pneumonia have improved across the VA healthcare system after accounting for illness severity, despite an increase in cases and comorbidity burden.

Machine learning-derived prediction of in-hospital mortality in patients with severe acute respiratory infection: analysis of claims data from the German-wide Helios hospital network

BACKGROUND

Severe acute respiratory infections (SARI) are the most common infectious causes of death. Previous work regarding mortality prediction models for SARI using machine learning (ML) algorithms that can be useful for both individual risk stratification and quality of care assessment is scarce. We aimed to develop reliable models for mortality prediction in SARI patients utilizing ML algorithms and compare its performances with a classic regression analysis approach.

METHODS

Administrative data (dataset randomly split 75%/25% for model training/testing) from years 2016-2019 of 86 German Helios hospitals was retrospectively analyzed. Inpatient SARI cases were defined by ICD-codes J09-J22. Three ML algorithms were evaluated and its performance compared to generalized linear models (GLM) by computing receiver operating characteristic area under the curve (AUC) and area under the precision-recall curve (AUPRC).

RESULTS

The dataset contained 241,988 inpatient SARI cases (75 years or older: 49%; male 56.2%). In-hospital mortality was 11.6%. AUC and AUPRC in the testing dataset were 0.83 and 0.372 for GLM, 0.831 and 0.384 for random forest (RF), 0.834 and 0.382 for single layer neural network (NNET) and 0.834 and 0.389 for extreme gradient boosting (XGBoost). Statistical comparison of ROC AUCs revealed a better performance of NNET and XGBoost as compared to GLM.

CONCLUSION

ML algorithms for predicting in-hospital mortality were trained and tested on a large real-world administrative dataset of SARI patients and showed good discriminatory performances. Broad application of our models in clinical routine practice can contribute to patients' risk assessment and quality management.

Performance of Machine Learning Algorithms for Predicting Adverse Outcomes in Community-Acquired Pneumonia

Background: The ability to assess adverse outcomes in patients with community-acquired pneumonia (CAP) could improve clinical decision-making to enhance clinical practice, but the studies remain insufficient, and similarly, few machine learning (ML) models have been developed.

Objective: We aimed to explore the effectiveness of predicting adverse outcomes in CAP through ML models.

Methods: A total of 2,302 adults with CAP who were prospectively recruited between January 2012 and March 2015 across three cities in South America were extracted from DryadData. After a 70:30 training set: test set split of the data, nine ML algorithms were executed and their diagnostic accuracy was measured mainly by the area under the curve (AUC). The nine ML algorithms included decision trees, random forests, extreme gradient boosting (XGBoost), support vector machines, Naïve Bayes, K-nearest neighbors, ridge regression, logistic regression without regularization, and neural networks. The adverse outcomes included hospital admission, mortality, ICU admission, and one-year post-enrollment status.

Results: The XGBoost algorithm had the best performance in predicting hospital admission. Its AUC reached 0.921, and accuracy, precision, recall, and F1-score were better than those of other models. In the prediction of ICU admission, a model trained with the XGBoost algorithm showed the best performance with AUC 0.801. XGBoost algorithm also did a good job at predicting one-year post-enrollment status. The results of AUC, accuracy, precision, recall, and F1-score indicated the algorithm had high accuracy and precision. In addition, the best performance was seen by the neural network algorithm when predicting death (AUC 0.831).

Conclusions: ML algorithms, particularly the XGBoost algorithm, were feasible and effective in predicting adverse outcomes of CAP patients. The ML models based on available common clinical features had great potential to guide individual treatment and subsequent clinical decisions.