Implementation fidelity of a multifactorial in-hospital fall prevention program and its association with unit systems factors: a single center, cross-sectional study

Abstract

Background: Falls are a common, costly global public health burden. In hospitals, multifactorial fall prevention programs have proved effective in reducing falls' incidence; however, translating those programs accurately into daily clinical practice remains challenging. This study's aim was to identify ward-level system factors associated with implementation fidelity to a multifactorial fall prevention program (StuPA) targeting hospitalized adult patients in an acute care setting.

Methods: This retrospective cross-sectional study used administrative data on 11,827 patients admitted between July and December 2019 to 19 acute care wards at the University Hospital Basel, Switzerland, as well as data on the StuPA implementation evaluation survey conducted in April 2019. Data were analysed using descriptive statistics, Pearson's coefficients and linear regression modelling for variables of interest.

Results: The patient sample had an average age of 68 years and a median length of stay of 8.4 (IQR: 2.1) days. The mean care dependency score was 35.4 points (ePA-AC scale: from 10 points (totally dependent) to 40 points (totally independent)); the mean number of transfers per patient -(e.g., change of room, admission, discharge) was 2.6 (range: 2.4- 2.8). Overall, 336 patients (2.8%) experienced at least one fall, resulting in a rate of 5.1 falls per 1'000 patient days. The median inter-ward StuPA implementation fidelity was 80.6% (range: 63.9-91.7%). We found the mean number of inpatient transfers during hospitalisation and the mean ward-level patient care dependency to be statistically significant predictors of StuPA implementation fidelity.

Conclusion: Wards with higher care dependency and patient transfer levels showed higher implementation fidelity to the fall prevention program. Therefore, we assume that patients with the highest fall prevention needs received greater exposure to the program. For the StuPA fall prevention program, our results suggest a need for implementation strategies contextually adapted to the specific characteristics of the target wards and patients.

Keywords: Accidental falls; Fall prevention program; Implementation outcomes; Implementation science; Patient safety; Preventive health services; Routinely collected health data.

Fig. 1

a Number of transfers of a patient case and StuPA fidelity. b Care dependency and StuPA fidelity. c Care dependency and the “Clinical Practice” subgroup’s StuPA fidelity scores. a-cThe StuPA “Clinical Practice” subgroup includes only items associated with direct clinical practice. Each dot represents one of the 19 wards. Mean Care dependency [10–40 points] is scored on a continuum: 10 points = totally dependent; 40 points = totally independent. These points are shown as means of each ward. Each chart’s grey line represents a simple linear regression of the two variables

Fig. 2

a-d Explorative plots of StuPA fidelity and ward characteristics. Each dot represents one of the 19 test wards. For the Injury Severity Score [0–3 points], 0 points = no fall-related injuries occurred: 3 points = every fall resulted in severe fall-related injuries

Fig. 3

a-d Explorative plots of StuPA variables V8 and V9 and ward characteristics fall risk and injury severity. Each dot represents one of the 19 test wards. For the Injury Severity Score [0–3 points], 0 points = no fall-related injuries occurred: 3 points = every fall resulted in severe fall-related injuries

Fig. 4

Distribution of answers for individual StuPA variables