Prediction of intensive care units length of stay: a concise review

Machine Learning for Benchmarking Critical Care Outcomes

OBJECTIVES

Enhancing critical care efficacy involves evaluating and improving system functioning. Benchmarking, a retrospective comparison of results against standards, aids risk-adjusted assessment and helps healthcare providers identify areas for improvement based on observed and predicted outcomes. The last two decades have seen the development of several models using machine learning (ML) for clinical outcome prediction. ML is a field of artificial intelligence focused on creating algorithms that enable computers to learn from and make predictions or decisions based on data. This narrative review centers on key discoveries and outcomes to aid clinicians and researchers in selecting the optimal methodology for critical care benchmarking using ML.

METHODS

We used PubMed to search the literature from 2003 to 2023 regarding predictive models utilizing ML for mortality (592 articles), length of stay (143 articles), or mechanical ventilation (195 articles). We supplemented the PubMed search with Google Scholar, making sure relevant articles were included. Given the narrative style, papers in the cohort were manually curated for a comprehensive reader perspective.

RESULTS

Our report presents comparative results for benchmarked outcomes and emphasizes advancements in feature types, preprocessing, model selection, and validation. It showcases instances where ML effectively tackled critical care outcome-prediction challenges, including nonlinear relationships, class imbalances, missing data, and documentation variability, leading to enhanced results.

CONCLUSIONS

Although ML has provided novel tools to improve the benchmarking of critical care outcomes, areas that require further research include class imbalance, fairness, improved calibration, generalizability, and long-term validation of published models.

Data-driven methodology to predict the ICU length of stay: A multicentre study of 99,492 admissions in 109 Brazilian units

PURPOSE

The length of stay (LoS) is one of the most used metrics for resource use in Intensive Care Units (ICU). We propose a structured data-driven methodology to predict the ICU length of stay and the risk of prolonged stay, and its application in a large multicenter Brazilian ICU database.

METHODS

Demographic data, comorbidities, complications, laboratory data, and primary and secondary diagnosis were prospectively collected and retrospectively analysed by a data-driven methodology, which includes eight different machine learning models and a stacking model. The study setting included 109 mixed-type ICUs from 38 Brazilian hospitals and the external validation was performed by 93 medical-surgical ICUs of 55 hospitals in Brazil.

RESULTS

A cohort of 99,492 adult ICU admissions were included from the 01^st of January to the 31^st of December 2019. The stacking model combining Random Forests and Linear Regression presented the best results to predict ICU length of stay (RMSE = 3.82; MAE = 2.52; R² = 0.36). The prediction model for the risk of long stay were accurate to early identify prolonged stay patients (Brier Score = 0.04, AUC = 0.87, PPV = 0.83, NPV = 0.95).

CONCLUSION

The data-driven methodology to predict ICU length of stay and the risk of long-stay proved accurate in a large multicentre cohort of general ICU patients. The proposed models are helpful to predict the individual length of stay and to early identify patients with high risk of prolonged stay.

Hospital Length of Stay and 30-Day Mortality Prediction in Stroke: A Machine Learning Analysis of 17,000 ICU Admissions in Brazil

BACKGROUND

Hospital length of stay and mortality are associated with resource use and clinical severity, respectively, in patients admitted to the intensive care unit (ICU) with acute stroke. We proposed a structured data-driven methodology to develop length of stay and 30-day mortality prediction models in a large multicenter Brazilian ICU cohort.

METHODS

We analyzed data from 130 ICUs from 43 Brazilian hospitals. All consecutive adult patients admitted with stroke (ischemic or nontraumatic hemorrhagic) to the ICU from January 2011 to December 2020 were included. Demographic data, comorbidities, acute disease characteristics, organ support, and laboratory data were retrospectively analyzed by a data-driven methodology, which included seven different types of machine learning models applied to training and test sets of data. The best performing models, based on discrimination and calibration measures, are reported as the main results. Outcomes were hospital length of stay and 30-day in-hospital mortality.

RESULTS

Of 17,115 ICU admissions for stroke, 16,592 adult patients (13,258 ischemic and 3334 hemorrhagic) were analyzed; 4298 (26%) patients had a prolonged hospital length of stay (> 14 days), and 30-day mortality was 8% (n = 1392). Prolonged hospital length of stay was best predicted by the random forests model (Brier score = 0.17, area under the curve = 0.73, positive predictive value = 0.61, negative predictive value = 0.78). Mortality prediction also yielded the best discrimination and calibration through random forests (Brier score = 0.05, area under the curve = 0.90, positive predictive value = 0.66, negative predictive value = 0.94). Among the 20 strongest contributor variables in both models were (1) premorbid conditions (e.g., functional impairment), (2) multiple organ dysfunction parameters (e.g., hypotension, mechanical ventilation), and (3) acute neurological aspects of stroke (e.g., Glasgow coma scale score on admission, stroke type).

CONCLUSIONS

Hospital length of stay and 30-day mortality of patients admitted to the ICU with stroke were accurately predicted through machine learning methods, even in the absence of stroke-specific data, such as the National Institutes of Health Stroke Scale score or neuroimaging findings. The proposed methods using general intensive care databases may be used for resource use allocation planning and performance assessment of ICUs treating stroke. More detailed acute neurological and management data, as well as long-term functional outcomes, may improve the accuracy and applicability of future machine-learning-based prediction algorithms.