Identification of Quality Indicator Codes Associated With Falls in Inpatient Rehabilitation Facilities

Abstract

<h3>Research Objectives</h3> To investigate if quality indicator (QI) codes are associated with patient falls in inpatient rehabilitation facilities (IRFs). <h3>Design</h3> In this retrospective cohort study, we collected data from the medical records of 1700 patients. <h3>Setting</h3> With the authorization of an institutional review board, we extracted data from 4 Utah IRFs. <h3>Participants</h3> We included the medical records of all patients over 14 years of age who were admitted and discharged from these facilities in 2020. <h3>Interventions</h3> All patients were discharged prior to data extraction, so no interventions beyond standard care were delivered. Items extracted include age, sex, race/ethnicity, diagnosis, fall status, Morse Fall Scale score, and QI codes for understanding, self-care, and mobility performance. <h3>Main Outcome Measures</h3> We calculated means, standard deviations, confidence intervals, and statistical differences between those who fell and those who did not. We explored associations between QI codes and falls using multivariaable logistic regression analysis. <h3>Results</h3> Medical records revealed 97 (5.71%) patients fell during IRF admission. Quality indicator codes for patient understanding, self-care, and mobility performance were significantly different between patients who fell and those who did not. A multivariable logistic regression model showed that low performance with understanding, walking 10 feet, and toileting was significantly associated with falls. Having an admission QI code below 4 for understanding increased patients' odds of falling by an estimated 78%. Admission QI codes below 3 for walking 10 feet and toileting indicated over 2 times greater odds of falling. Diagnosis, age, sex, and race/ethnicity were not significantly associated with falls. <h3>Conclusions</h3> Quality indicator codes for understanding, self-care, and mobility are significantly associated with falls. Further research should explore how these required codes may be used to identify patients likely to fall during admission. <h3>Author(s) Disclosures</h3> The authors have no conflicts of interest to disclose.