The effect of changing practice on fall prevention in a rehabilitative hospital: The hospital injury prevention study

OBJECTIVE: To determine whether a change in practice to introduce a multidisciplinary fall-prevention program can reduce falls and injury in nonacute patients in a rehabilitation hospital. DESIGN: A quasi-experimental study. SETTING: Three geriatric wards with a similar design, equipment, staffing levels, and skill mix. PARTICIPANTS: Eight hundred twenty-five consecutive patients. INTERVENTION: The patients' fall-risk status was assessed using the Downton Score. Current practice was maintained on the two control wards (n=550). On the experimental ward (n=275), a fall-prevention program was introduced. A multidisciplinary team met weekly specifically to discuss patients' fall risk and formulate a targeted plan. Patients at risk were identified using wristbands; risk factors were corrected or environmental changes made to enhance safety. MEASUREMENTS: Primary outcomes were number of fallers, recurrent fallers, total falls, patients sustaining injury, and falls per occupied bed days. Secondary outcomes were place of discharge and mortality. RESULTS: Patients were matched for age and risk status. Control wards had proportionally more fallers (20.2% vs 14.2%: P=.033), patients sustaining injury (8.2% vs 4%: P=.025), and total number of falls (170 vs 72: P=.045). These results did not remain significant after controlling for differing length of stay. There was no reduction in recurrent fallers (6.4% vs 4.7%: P=.43) and no effect on place of discharge (home discharges; 57.5% vs 60.7%: P=.41) or mortality (15.3% vs 13.8%: P=.60). CONCLUSION: This study shows that falls might be reduced in a multidisciplinary fall-prevention program, but the results are not definitive because of the borderline significance achieved and the variable length of stay. More research on fall prevention in hospital is required, particularly as to what interventions, if any, are effective at reducing falls in this group of patients.