Descriptive Analysis of Supervised Falls Occurring During Physical Therapy Sessions in Adult Inpatient Rehabilitation.

Abstract

Importance: Falls are commonly reported incidents that affect the safety of patients during inpatient hospitalization. Inpatient rehabilitation (IR) units report the highest fall rates when compared with other hospital units. Falls commonly result in patient injury and elevate episode costs. There is a dearth of information regarding characteristics of patients who fall during physical therapy sessions, which represents a unique subset of falls.

Objective: The aim of the study was to describe the nature of falls, characteristics of fallers, and characteristics of therapists who experienced patient falls, which occurred during physical therapy sessions in IR.

Design: This was an observational study which included a retrospective analysis of medical records.

Setting: This study examined falls occurring within 2 IR departments at a large hospital system located in an urban setting in the United States.

Participants: This study involved patients receiving adult IR with diagnoses including, but not limited to, stroke, traumatic brain injury, and spinal cord injury.

Exposures: This study examined characteristics of patients who fell as compared with patients who did not fall, quantified the conditions surrounding falls, and described physical therapists who experienced patient falls.

Main outcomes and measures: Mann-Whitney U tests, chi square tests, and binomial logistic regression analyses were performed to compare characteristics of faller and non-faller groups.

Results: Among the 6238 unique patient admissions, a total of 40 falls were identified. The rate of falling was 0.43 falls per 1000 patient days. The majority of falls occurred because of buckling (47.5%) and during gait training (40.0%). Falls most often occurred close to discharge (mode = 6 days prior). Fallers were younger than nonfallers [exponential power of B, ie, Exp(B), = 1.02; 95% CI = 1.01-1.04]. Diagnoses representing the largest proportion of fallers included brain dysfunction/stroke (30.0%) and spinal cord injury/peripheral nerve injury (30.0%). Fallers had comorbid diabetes mellitus type 2 [Exp(B) = 2.70; 95% CI = 1.45-5.04] and received renal dialysis [Exp(B) = 3.23; 95% CI = 1.14-9.17] in a higher proportion than nonfallers. Fallers were often high functioning, the majority receiving at most minimal assistance (72.5%). Falls most often occurred with therapists who had 1 to 2 years of experience (27.5% of falls).

Conclusions: The rate of falls during therapy was lower than the rate of falls previously reported in similar settings. Therapists should exercise caution when managing younger patients and patients with certain diagnoses. Therapists should screen for buckling risk when prescribing higher-risk activities. Therapists with various levels of experience should receive fall prevention training.

Clinical relevance: Therapists can use frequently occurring patient characteristics to screen for falling and use additional precautionary measures, particularly for patients with specific diagnoses, at higher risk for knee buckling, and of a younger age.