Predicting Falls in Rehabilitation: A Comparison of Three Instruments Including Hester Davis.

PURPOSE: The aim of this study was to evaluate the ability of the Hester Davis Scale (HDS), Section GG, and facility fall risk assessment scores to predict patients who fall during inpatient rehabilitation. DESIGN: This study was an observational quality improvement project. METHODS: Nurses administered the HDS in parallel to the facility's current fall risk assessment and Section GG of the Centers for Medicare & Medicaid Services Inpatient Rehabilitation Facility Patient Assessment Instrument. Receiver operating characteristic curves were compared in 1,645 patients. Relationships of individual scale items to falls were also assessed. RESULTS: The HDS (area under the curve [AUC] = .680, 95% CI [.626, .734]), facility fall risk assessment (AUC = .688, 95% CI [.637, .740]), and Section GG scores (AUC = .687, 95% CI [.638, .735]) adequately identified patients who fell. AUCs did not significantly differ between assessments. HDS scores of ≥13, facility scores of ≥14, and Section GG scores of ≤51 resulted in the highest sensitivity/specificity balance. CONCLUSIONS: HDS, facility fall risk assessment, and Section GG scores adequately and similarly identified patients of mixed diagnoses at risk of falling in inpatient rehabilitation. CLINICAL RELEVANCE TO THE PRACTICE OF REHABILITATION NURSING: Rehabilitation nurses have several options including the HDS and Section GG to identify patients at greatest risk of falling.

Validation of the Casa Colina Fall Risk Assessment Scale in Predicting Falls in Inpatient Rehabilitation Facilities.

OBJECTIVE: The aim of this study was to assess the validity, efficacy, and generalizability of a fall risk assessment tool created specifically for inpatient rehabilitation facilities (IRFs). DESIGN: The Casa Colina Falls Risk Assessment Scale (CCFRAS) was assessed both retrospectively and prospectively on consecutive patients at three IRFs to determine the sensitivity and specificity of this tool in predicting fall risk. SETTING: The setting was in three IRFs. PARTICIPANTS: Individuals admitted to three IRFs participated in the study. MAIN OUTCOMES MEASURES: Each IRF quantified the number of falls detected for the patient population under evaluation and determined the site-specific sensitivity and specificity of the CCFRAS. RESULTS: The sensitivity and specificity of the CCFRAS ranged from 75% to 80% and from 47% to 70%, respectively, for the different IRFs. Using a logistic regression analysis, we identified the optimal CCFRAS cutoff score for identifying high-risk patients at each individual facility, thus improving the specificity to 70%-79%. CONCLUSION: Multisite evaluation of this assessment tool indicates that the CCFRAS is effective and broadly generalizable for predicting patients at high risk for falling.

Using Functional Independence Measure Subscales to Predict Falls-Rapid Assessment.

BACKGROUND: Falls remain a major issue in inpatient rehabilitation. Decreased scores on the Functional Independence Measure (FIM), given to every patient, have been shown to predict falls risk. PURPOSE: The aim of the study was to extend previous research using FIM to predict falls by using only subscales assessed earliest during admissions to indicate high risk of falls. DESIGN: Retrospective cohort study. METHODS: Two consecutive samples of patients (n1 = 1,553, n2 = 12,301) admitted to a rehabilitation hospital over 9-month and 5-year periods, respectively, were used to evaluate the predictive utility of using only a small number of FIM subscales. Subscales were selected from those assessed earliest and were related to previously published research on falls risk factors. The metric was developed using a historical data set and was validated with a second, separate group of patients. Receiver operating characteristic curves were used to evaluate predictive utility. FINDINGS: The combination of Toileting and Expression subscales yielded a comparable area under the curve to the full FIM, and both were greater than the existing method of falls risk assessment. Likelihood of falling was strongly linearly related to score on the Toileting/Expression metric. CONCLUSIONS: The sum of two FIM subscales can be used to predict which patients may fall during their stay in a rehabilitation hospital. CLINICAL RELEVANCE: The FIM scores are assessed early during a patient's stay, are required for all Medicare patients, and may be useful for simple, rapid, and accurate assignment of falls risk.