[Epidemiology and prediction of the risk of falling in patients in acute care settings: Analysis of routine data from a university hospital]

Background: Falls are a relevant issue of inpatient treatment. Epidemiological analysis concerning incidence, risk factors for falls and the quality of risk assessments are missing.

Methods: In a routine data-based cross-sectional study all patients hospitalized in the University Hospital Dresden, Germany, during 2012 and 2013 were analyzed according to fall incidence and risk factors (items of Dresden fall risk assessment [Dresden-FRA], age, sex, severity of disease, and length of stay. Fall risk and associated risk factors were determined using descriptive methods and logistic regression models. In addition, the quality of the fall risk assessment was evaluated.

Results: The risk of falling during the observation period was 1.6 %. Differences exist between the medical disciplines (risk of falling 0.1 % to 9.8 %). The fall rate was 2.9/1,000 days of treatment. Higher age, longer inpatient stay, and a higher level of disease severity were significant predictors of falls. Less serious consequences of falls were more frequently documented (84 %) than consequences of falls requiring treatment (16 %). All nine items of the Dresden-FRA were significant risk factors. The sensitivity and specificity of the Dresden-FRA were 69.9 % and 71.4 %, respectively.

Conclusion: Patients at an older age and with more severe disease have a higher risk of falling. The validated assessment of the individual risk of falling is an important tool of medical quality management and risk management. The present study can help to raise awareness about the quality of risk assessments, to further improve fall risk assessment, and to increase patient safety in acute care.