Identification of important factors in an inpatient fall risk prediction model to improve the quality of care using EHR and electronic administrative data: A machine-learning approach

Abstract

Background: Inpatient falls, many resulting in injury or death, are a serious problem in hospital settings. Existing falls risk assessment tools, such as the Morse Fall Scale, give a risk score based on a set of factors, but don’t necessarily signal which factors are most important for predicting falls. Artificial intelligence (AI) methods provide an opportunity to improve predictive performance while also identifying the most important risk factors associated with hospital-acquired falls. We can glean insight into these risk factors by applying classification tree, bagging, random forest, and adaptive boosting methods applied to Electronic Health Record (EHR) data. Objective: The purpose of this study was to use tree-based machine learning methods to determine the most important predictors of inpatient falls, while also validating each via cross-validation. Materials and methods: A case-control study was designed using EHR and electronic administrative data collected between January 1, 2013 to October 31, 2013 in 14 medical surgical units. The data contained 38 predictor variables which comprised of patient characteristics, admission information, assessment information, clinical data, and organizational characteristics. Classification tree, bagging, random forest, and adaptive boosting methods were used to identify the most important factors of inpatient fall-risk through variable importance measures. Sensitivity, specificity, and area under the ROC curve were computed via ten-fold cross validation and compared via pairwise t-tests. These methods were also compared to a univariate logistic regression of the Morse Fall Scale total score. Results: In terms of AUROC, bagging (0.89), random forest (0.90), and boosting (0.89) all outperformed the Morse Fall Scale (0.86) and the classification tree (0.85), but no differences were measured between bagging, random forest, and adaptive boosting, at a p-value of 0.05. History of Falls, Age, Morse Fall Scale total score, quality of gait, unit type, mental status, and number of high fall risk increasing drugs (FRIDs) were considered the most important features for predicting inpatient fall risk. Conclusions: Machine learning methods have the potential to identify the most relevant and novel factors for the detection of hospitalized patients at risk of falling, which would improve the quality of patient care, and to more fully support healthcare provider and organizational leadership decision-making. Nurses would be able to enhance their judgement to caring for patients at risk for falls. Our study may also serve as a reference for the development of AI-based prediction models of other iatrogenic conditions. To our knowledge, this is the first study to report the importance of patient, clinical, and organizational features based on the use of AI approaches.

Clinical Impact of an Analytic Tool for Predicting the Fall Risk in Inpatients: Controlled Interrupted Time Series

“…The increased adoption of electronic health record (EHR) systems over the past decade has stimulated the development of predictive fall risk models using machine learning techniques, which are reported to exhibit better predictive performance than the existing fall risk assessment tools alone [ 15 - 18 ]. However, most of these models have not been validated in multiple settings, and their implementation is restricted by their use of aggregated data by hospital admission rather than by patient-days.…”

Data Science Trends Relevant to Nursing Practice: A Rapid Review of the 2020 Literature

“…For the topic of falls, we identified 24 studies that met inclusion criteria. Of these studies, eight used a retrospective cohort design 40 41 42 43 44 45 46 47 ; seven used a prospective cohort design 48 49 50 51 52 53 54 ; six were secondary analyses of research data obtained from prospective, retrospective, and cross-sectional studies 55 56 57 58 59 60 ; one used mixed methods wherein data from a public dataset were used in conjunction with measurements collected from sensors 61 ; and one was a meta-analysis of prospective cohort and observational studies. 62 Ten of the studies used health records as a source of data but in two of these studies, 44 47 it was not clear whether the records were electronic when they were obtained.…”

“…Several of the studies, including two of the secondary analyses, incorporated data from mobility and gait sensors. 48 49 51 53 55 60 61 Registries and administrative datasets were used in eight studies, 40 41 42 43 45 46 50 56 while questionnaires or surveys were a source of data for four studies. 49 51 57 60 With the exception of one study that employed sensor data from 17-year-old persons, 55 all study participants were community dwelling, inpatient, and outpatient adults.…”

Potential uses of AI for perioperative nursing handoffs: a qualitative study

“…A parallel literature explores the role of AI-based prediction of adverse events of specific relevance to nursing tasks. For example, identification of patients with high risk of pressure ulcers [35, 36] or falls [37] can trigger clinical decision support for nursing interventions. The related clinical decision support literature focuses on recommending specific actions, medication dosing, and documentation reminders.…”