Falls prevention revisited: a call for a new approach

Abstract

Background: Patient falls constitute a major threat to health services' ability to provide care. Previous studies confirm that nurses can identify patients at risk and that a preventative programme can reduce the rate of falls but few studies have been evaluated over time.

Aims and objectives: A study was undertaken to test a Falls Prevention Programme in an acute medical area that was re-evaluated 5 years later to determine if the effects were sustainable.

Design: The design included two groups of patients admitted before and after the programme. Variables such as staffing, equipment, environment and routines were controlled. However, because of ethical approval constraints, some variables such as age, mental status, mobility and gender were not.

Methods: The programme included a risk assessment tool, a choice of interventions, a graphic that alerted others to 'at risk patients' and simple patient and staff education. Data were collected using incident forms and a formula was used to calculate a rate of falls. A non-paired t-test compared rates and anova examined the relationship of age, gender, mobility and mental status on the incidence of falls. Control graphs determined the stability of the process.

Results: The falls rate was significantly reduced. Control graphs demonstrate that the process achieved greater control with less variation. In the next 5 years the falls rate increased to preprogramme levels and control graphs demonstrated that the process was no longer controlled. Compliance with the programme had deteriorated.

Conclusions: The practice review considered skill mix, patient activity and acuity but provided no definitive answers to explain non-compliance. The implications to nursing are discussed.

Relevance to clinical practice: Clinicians are called to conduct more rigorous research into falls prevention but it may be more useful to direct research towards examining nursing work and increasing nurse autonomy in falls prevention.