Prediction of early unplanned intensive care unit readmission in a UK tertiary care hospital: a cross-sectional machine learning approach

Abstract

ObjectivesUnplanned readmissions to the intensive care unit (ICU) are highly undesirable, increasing variance in care, making resource planning difficult and potentially increasing length of stay and mortality in some settings. Identifying patients who are likely to suffer unplanned ICU readmission could reduce the frequency of this adverse event.SettingA single academic, tertiary care hospital in the UK.ParticipantsA set of 3326 ICU episodes collected between October 2014 and August 2016. All records were of patients who visited an ICU at some point during their stay. We excluded patients who were ≤16 years of age; visited ICUs other than the general and neurosciences ICU; were missing crucial electronic patient record measurements; or had indeterminate ICU discharge outcomes or very early or extremely late discharge times. After exclusion, 2018 outcome-labelled episodes remained.Primary and secondary outcome measuresArea under the receiver operating characteristic curve (AUROC) for prediction of unplanned ICU readmission or in-hospital death within 48 hours of first ICU discharge.ResultsIn 10-fold cross-validation, an ensemble predictor was trained on data from both the target hospital and the Medical Information Mart for Intensive Care (MIMIC-III) database and tested on the target hospital’s data. This predictor discriminated between patients with the unplanned ICU readmission or death outcome and those without this outcome, attaining mean AUROC of 0.7095 (SE 0.0260), superior to the purpose-built Stability and Workload Index for Transfer (SWIFT) score (AUROC=0.6082, SE 0.0249; p=0.014, pairwise t-test).ConclusionsDespite the inherent difficulties, we demonstrate that a novel machine learning algorithm based on transfer learning could achieve good discrimination, over and above that of the treating clinicians or the value added by the SWIFT score. Accurate prediction of unplanned readmission could be used to target resources more efficiently.

Design and implementation of a deep recurrent model for prediction of readmission in urgent care using electronic health records

“…In this research, the event logs were created using CHARTTIME attributes, as this is the best match to the time of actual measurement. All the patient data in the MIMIC-III database has been de-identified and all dates have been randomly shifted to the future so that dates are internally consistent for the same patient but inconsistent across patients [8], [9].…”

“…Negative cases, on the contrast, are those that the patient do not need ICU readmission. Specifically, patients who were transferred or discharged from ICU and did not return and are still alive within the next 30 days are considered to be negative cases [8]. In the MIMIC Dataset, we had the following instances that contributed to the positive cases: i) patients were transferred to low-level wards from ICU, but returned to ICU again (3,555 records); ii) patients were transferred to low-level wards from ICU, and died later(1,974 records); iii) patients were discharged, but returned to the ICU within the next 30 days (3,205 records); iv) patients were discharged, and died within the next 30 days (2,556 records).…”

Readmissions to General ICUs in a Geographic Area of Poland Are Seemingly Associated with Better Outcomes

“…Readmission to an intensive care unit (ICU) is perceived to be associated with high mortality rates. As a result, ICU readmission rates are currently recognized as ICU performance metrics and quality improvement targets [1][2][3]. This relationship, however, is not observed in crowded emergency departments, where patients who were discharged and then readmitted during a return visit had lower in-hospital mortality and ICU admission rates [4].…”

Machine learning algorithm for early mortality prediction in patients with advanced penile cancer

“…While there does not exist an intuitive method for risk stratification of patients with penile cancer, there is strong evidence of the potential of machine learning for stratification of patients in other diseases such as heart failure [7][8][9][10], kidney disease [11], and critical care [12][13][14][15][16][17]. Furthermore, machine learning has been shown to be effective for readmission prediction [17][18][19], drug adverse event prediction [20]. While such a method does not exist, we posit that a machine learning-based method can be useful for clinical decision support in the management of penile cancer.…”