Risk Factors for Falls in Hospital In-Patients: A Prospective Nested Case Control Study

Background: Patient falls are considered a challenge to the patient’s safety in hospitals, which, in addition to increasing the length of stay and costs, may also result in severe injuries or even the death of the patient. This study aims to investigate the associations between risk factors among fallers in comparison with the control group. Methods: A prospective nested case control study was performed on 185 patients who fell and 1141 controls were matched with the patients at risk of fall in the same ward and during the same time. This study was conducted in a university educational hospital in Tehran with 800 beds during a 9-month period. The data included demographics, comorbidities, admission details, types of medication, clinical conditions, and activities before or during the fall. The data was collected from clinical records, hospital information system, error reporting system and observations, and the interviews with the fallers, their families and care givers (physicians, nurses, etc). Data analysis was conducted through time-based matching using a multi-level analysis. Results: In a multilevel model including patient-related, medication, and care-related variables, the factors that were significantly associated with an increased risk of patient falls included: longer length of stay (odds ratio [OR] = 1.01; CI=0.32 to 0.73), using chemotherapy drugs, sedatives, anticonvulsants, benzodiazepines, and angiotensin-converting enzyme (ACE) inhibitors, visual acuity (OR=6.93; CI=4.22 to 11.38), balance condition (OR= 6.41; CI=4.51 to 9.11), manual transfer aid (OR=8.47; CI=5.65 to 12.69), urinary incontinence (OR= 8.47, CI= 5.65 to 12.69), and cancer (OR=2.86, CI=1.84-4.44). These factors were found to be associating with more odds for a falling accident among patients. Several characteristics such as fall history (OR=0.48; CI= 1.003 to 1.02), poly-pharmacy (OR=1.37, CI=00.85 to 2.2), stroke (OR=0.94, CI= 0.44 to 2.02), and nurse to patient ratio (incidence rate ratio=1.01, CI=0.01 to 0.03) were not significantly associated with falling in hospitals. Conclusion: It seems that a combination of both patient-related factors and history of medication should be considered. Moreover, modifiable clinical characteristics of patients such as vision improvement, provision of manual transfer aid, diabetes control, regular toilet program, and drug modification should be considered during the formulation of interventions.