Can we reliably automate clinical prognostic modelling? A retrospective cohort study for ICU triage prediction of in-hospital mortality of COVID-19 patients in the Netherlands

Development and evaluation of regression tree models for predicting in-hospital mortality of a national registry of COVID-19 patients over six pandemic surges

BACKGROUND

Objective prognostic information is essential for good clinical decision making. In case of unknown diseases, scarcity of evidence and limited tacit knowledge prevent obtaining this information. Prediction models can be useful, but need to be not only evaluated on how well they predict, but also how stable these models are under fast changing circumstances with respect to development of the disease and the corresponding clinical response. This study aims to provide interpretable and actionable insights, particularly for clinicians. We developed and evaluated two regression tree predictive models for in-hospital mortality of COVID-19 patient at admission and 24 hours (24 h) after admission, using a national registry. We performed a retrospective analysis of observational routinely collected data.

METHODS

Two regression tree models were developed for admission and 24 h after admission. The complexity of the trees was managed via cross validation to prevent overfitting. The predictive ability of the model was assessed via bootstrapping using the Area under the Receiver-Operating-Characteristic curve, Brier score and calibration curves. The tree models were assessed on the stability of their probabilities and predictive ability, on the selected variables, and compared to a full-fledged logistic regression model that uses variable selection and variable transformations using splines. Participants included COVID-19 patients from all ICUs participating in the Dutch National Intensive Care Evaluation (NICE) registry, who were admitted at the ICU between February 27, 2020, and November 23, 2021. From the NICE registry, we included concerned demographic data, minimum and maximum values of physiological data in the first 24 h of ICU admission and diagnoses (reason for admission as well as comorbidities) for model development. The main outcome measure was in-hospital mortality. We additionally analysed the Length-of-Stay (LoS) per patient subgroup per survival status.

RESULTS

A total of 13,369 confirmed COVID-19 patients from 70 ICUs were included (with mortality rate of 28%). The optimism-corrected AUROC of the admission tree (with seven paths) was 0.72 (95% CI: 0.71-0.74) and of the 24 h tree (with 11 paths) was 0.74 (0.74-0.77). Both regression trees yielded good calibration and variable selection for both trees was stable. Patient subgroups comprising the tree paths had comparable survival probabilities as the full-fledged logistic regression model, survival probabilities were stable over six COVID-19 surges, and subgroups were shown to have added predictive value over the individual patient variables.

CONCLUSIONS

We developed and evaluated regression trees, which operate at par with a carefully crafted logistic regression model. The trees consist of homogenous subgroups of patients that are described by simple interpretable constraints on patient characteristics thereby facilitating shared decision-making.

Prognostic models of in-hospital mortality of intensive care patients using neural representation of unstructured text: A systematic review and critical appraisal

OBJECTIVE

To review and critically appraise published and preprint reports of prognostic models of in-hospital mortality of patients in the intensive-care unit (ICU) based on neural representations (embeddings) of clinical notes.

METHODS

PubMed and arXiv were searched up to August 1, 2022. At least two reviewers independently selected the studies that developed a prognostic model of in-hospital mortality of intensive-care patients using free-text represented as embeddings and extracted data using the CHARMS checklist. Risk of bias was assessed using PROBAST. Reporting on the model was assessed with the TRIPOD guideline. To assess the machine learning components that were used in the models, we present a new descriptive framework based on different techniques to represent text and provide predictions from text. The study protocol was registered in the PROSPERO database (CRD42022354602).

RESULTS

Eighteen studies out of 2,825 were included. All studies used the publicly-available MIMIC dataset. Context-independent word embeddings are widely used. Model discrimination was provided by all studies (AUROC 0.75-0.96), but measures of calibration were scarce. Seven studies used both structural clinical variables and notes. Model discrimination improved when adding clinical notes to variables. None of the models was externally validated and often a simple train/test split was used for internal validation. Our critical appraisal demonstrated a high risk of bias in all studies and concerns regarding their applicability in clinical practice.

CONCLUSION

All studies used a neural architecture for prediction and were based on one publicly available dataset. Clinical notes were reported to improve predictive performance when used in addition to only clinical variables. Most studies had methodological, reporting, and applicability issues. We recommend reporting both model discrimination and calibration, using additional data sources, and using more robust evaluation strategies, including prospective and external validation. Finally, sharing data and code is encouraged to improve study reproducibility.

Assess and validate predictive performance of models for in-hospital mortality in COVID-19 patients: A retrospective cohort study in the Netherlands comparing the value of registry data with high-granular electronic health records

PURPOSE

To assess, validate and compare the predictive performance of models for in-hospital mortality of COVID-19 patients admitted to the intensive care unit (ICU) over two different waves of infections. Our models were built with high-granular Electronic Health Records (EHR) data versus less-granular registry data.

METHODS

Observational study of all COVID-19 patients admitted to 19 Dutch ICUs participating in both the national quality registry National Intensive Care Evaluation (NICE) and the EHR-based Dutch Data Warehouse (hereafter EHR). Multiple models were developed on data from the first 24 h of ICU admissions from February to June 2020 (first COVID-19 wave) and validated on prospective patients admitted to the same ICUs between July and December 2020 (second COVID-19 wave). We assessed model discrimination, calibration, and the degree of relatedness between development and validation population. Coefficients were used to identify relevant risk factors.

RESULTS

A total of 1533 patients from the EHR and 1563 from the registry were included. With high granular EHR data, the average AUROC was 0.69 (standard deviation of 0.05) for the internal validation, and the AUROC was 0.75 for the temporal validation. The registry model achieved an average AUROC of 0.76 (standard deviation of 0.05) in the internal validation and 0.77 in the temporal validation. In the EHR data, age, and respiratory-system related variables were the most important risk factors identified. In the NICE registry data, age and chronic respiratory insufficiency were the most important risk factors.

CONCLUSION

In our study, prognostic models built on less-granular but readily-available registry data had similar performance to models built on high-granular EHR data and showed similar transportability to a prospective COVID-19 population. Future research is needed to verify whether this finding can be confirmed for upcoming waves.