A new tool for the evaluation of the rehabilitation outcomes in older persons: a machine learning model to predict functional status 1 year ahead

Will-to-Live and Self-Rated Health in Older Hospitalized Patients Are Not Predictive for Short-Term Mortality

Background: Self-assessed will-to-live and self-rated health are associated with long-term survival in community-dwelling older persons but have not been examined in frailer older patients in relation to short-term prognosis. The aim was to explore whether will-to-live and self-rated health are predictive for six-month mortality and can guide ceiling of treatment decisions in hospitalized patients in an acute geriatric ward. We included the Surprise Question as reference, being a well-established clinical tool for short-term prognostication. Methods: This multicentric prospective study included patients of 75 years and older admitted at acute geriatric wards of two Belgian hospitals. Will-to-live and self-rated health were scored on a Likert scale (0-5, 0-4) and assessed by junior geriatricians. The senior geriatricians answered the Surprise Question for clinical judgment of prognosis. Receiver-operator characteristic (ROC) curves were constructed to determine diagnostic accuracy. For time-dependent analysis, Cox regression was performed with adjustment for age and gender. Results: Of 93 included patients in the study, 69 were still alive after six months and 24 died, resulting in a six-month mortality of 26%. The mean age was 86 years (range 75-100), 67% of the patients were women. Median will-to-live and self-rated health were 3 (moderate and good). Both will-to-live and self-rated health were not predictive for six-month mortality (area under the ROC curve [AUC] 0.496, p = 0.951 for will-to-live; 0.447, p = 0.442 for self-rated health) as opposed to Surprise Question (AUC 0.793, p < 0.001). After correction for sex and age, the hazard ratio of six-month mortality was 0.92 for will-to-live (p = 0.667), 0.86 for self-rated health (p = 0.548), and 10.28 for Surprise Question (p < 0.001). Conclusion: Will-to-live and self-rated health are not predictive for six-month mortality in patients admitted to the acute geriatric ward, unlike prognostic tools such as Surprise Question. Clinical Trial Registration Number: B670202100792.

Prediction of emergence from prolonged disorders of consciousness from measures within the UK rehabilitation outcomes collaborative database: a multicentre analysis using machine learning

PURPOSE

Predicting emergence from prolonged disorders of consciousness (PDOC) is important for planning care and treatment. We used machine learning to examine which variables from routine clinical data on admission to specialist rehabilitation units best predict emergence by discharge.

MATERIALS AND METHODS

A multicentre national cohort analysis of prospectively collected clinical data from the UK Rehabilitation Outcomes (UKROC) database 2010-2018. Patients (n = 1170) were operationally defined as "still in PDOC" or "emerged" by their total UK Functional Assessment Measure (FIM + FAM) discharge score. Variables included: Age, aetiology, length of stay, time since onset, and all items of the Neurological Impairment Scale, Rehabilitation Complexity Scale, Northwick Park Dependency Scale, and the Patient Categorisation Tool. After filtering, prediction of emergence was explored using four techniques: binary logistic regression, linear discriminant analysis, artificial neural networks, and rule induction.

RESULTS

Triangulation through these techniques consistently identified characteristics associated with emergence from PDOC. More severe motor impairment, complex disability, medical and behavioural instability, and anoxic aetiology were predictive of non-emergence, whereas those with less severe motor impairment, agitated behaviour and complex disability were predictive of emergence.

CONCLUSIONS

This initial exploration demonstrates the potential opportunities to enhance prediction of outcome using machine learning techniques to explore routinely collected clinical data. Implications for rehabilitationPredicting emergence from prolonged disorders of consciousness is important for planning care and treatment.Few evidence-based criteria exist for aiding clinical decision-making and existing criteria are mostly based upon acute admission data.Whilst acknowledging the limitations of using proxy data for diagnosis of emergence, this study suggests that key items from the UKROC dataset, routinely collected on admission to specialist rehabilitation some months post injury, may help to predict those patients who are more (or less) likely to regain consciousness.Machine learning can help to enhance our understanding of the best predictors of outcome and thus assist with clinical decision-making in PDOC.

A model for predicting physical function upon discharge of hospitalized older adults in Taiwan-a machine learning approach based on both electronic health records and comprehensive geriatric assessment

Background

Predicting physical function upon discharge among hospitalized older adults is important. This study has aimed to develop a prediction model of physical function upon discharge through use of a machine learning algorithm using electronic health records (EHRs) and comprehensive geriatrics assessments (CGAs) among hospitalized older adults in Taiwan.

Methods

Data was retrieved from the clinical database of a tertiary medical center in central Taiwan. Older adults admitted to the acute geriatric unit during the period from January 2012 to December 2018 were included for analysis, while those with missing data were excluded. From data of the EHRs and CGAs, a total of 52 clinical features were input for model building. We used 3 different machine learning algorithms, XGBoost, random forest and logistic regression.

Results

In total, 1,755 older adults were included in final analysis, with a mean age of 80.68 years. For linear models on physical function upon discharge, the accuracy of prediction was 87% for XGBoost, 85% for random forest, and 32% for logistic regression. For classification models on physical function upon discharge, the accuracy for random forest, logistic regression and XGBoost were 94, 92 and 92%, respectively. The auROC reached 98% for XGBoost and random forest, while logistic regression had an auROC of 97%. The top 3 features of importance were activity of daily living (ADL) at baseline, ADL during admission, and mini nutritional status (MNA) during admission.

Conclusion

The results showed that physical function upon discharge among hospitalized older adults can be predicted accurately during admission through use of a machine learning model with data taken from EHRs and CGAs.