Predicting inpatient falls using a shift-based, pre-fall model of interdependency factors: patient, organization, and nurse.
Description
The purpose of this study was to illustrate a methodology for preventing patient falls on a medical surgical unit (MSU) in a small hospital, by applying a shift-based, centric model focusing on system interdependencies (patient, organization, nurse) and primary factors for risk for fall. The research question ask: Are there patient, organization and nurse centric factors associated with falls that can predict a patient’s risk for falling on this MSU. An exploratory, quantitative study with a retrospective review of 32 patient falls on a Medical Surgical unit was performed. As data for each patient was collected across shifts (2 shifts on the day prior to the fall, 2 shifts on the day of the fall and 2 shifts on the day after the fall), there was the potential for 192 observations. Accounting for patients that fell on either their admission shift or their discharge shift, the total number of observations was reduced to 155, of which, 84 observations were pre-fall. Data was entered into SPPS. Chi -square test of independence was performed for each of the 220 independent variables (patient, organization and nurse) with patient falls. Standard logistic regression and generalized linear mixed model logistic regression was applied to identify variables in combination that predict falls and non-falls and identified 5 predictive variables. Predictive accuracy of the predictive model was calculated using area under the curve (AUC) of the receiver operating characteristics (ROC). The pre-fall predictive model classified falls and non-falls correctly at 89.3% with a high level of predictive accuracy (area under the curve of .956) and high model quality of 91%. This study adds to the current body of knowledge to prevent inpatient falls by predicting population specific variables for falls on a MSU. The applied predictive model methodology is generalizable, although the results are not, as this was a small population in one small community hospital. With the advent of machine learning, the use of large amounts of data, such as 220 variables or more, can be used to make timely decisions to help predict nurse, patient and organization variables impacting falls.
Files
Steps to reproduce
Retrospective chart review of patient falls. Reviewed electronic health record. 220 independent variables (patient, organization and nurse) were collected manually for each patient fall. Data collection tools were creewted. Patients were de-identified. Data was entered in to SPPS software. Chi -square test of independence was performed for each of the 220 independent variables (patient, organization and nurse) with patient falls. Standard logistic regression and generalized linear mixed model logistic regression was applied to identify variables in combination that predict falls and non-falls and identified 5 predictive variables. Predictive accuracy of the predictive model was calculated using area under the curve (AUC) of the receiver operating characteristics (ROC). The pre-fall predictive model classified falls and non-falls correctly at 89.3% with a high level of predictive accuracy (area under the curve of .956) and high model quality of 91%.