Risk Assessment and Fall Prevention: Practice Development in Action
Dempsey, Jennifer, Contemporary Nurse : a Journal for the Australian Nursing Profession
BACKGROUND AND LITERATURE REVIEW
Australia's ageing population has placed extreme pressure on the acute health system as attempts to meet demands caused by chronic illness and aged related adverse events such as falls are experienced (Grbich 1996; Hill et al. 2000; NSW Health 2003; Mooney & Scotton 1999). The Central Coast has a larger percentage of the population (45% more) aged greater than 80 years (4.5%) compared to the NSW state average (3.1%). The hospital's admissions profile is even more heavily skewed to the elderly, where approximately 40% of all patients admitted from the Emergency Department are aged over 70 years of age. This demand is predicted to increase as the group that has the most falls, older women, will increase twofold in 50 years (Moller 2000). Therefore, greater resources will be required to provide acute and long-term care for fallers (NSW Health 2001).
A recommended strategy to assist health services cope with increased demand for acute services is rigorous research that may be applied to prevention practice (Evans et al. 1998; Hill et al. 2000). Nurses have demonstrated continuing research interest in identifying the patient at risk to fall however research conducted in the past has been criticised for its lack of rigour (Evans et al. 1998; Hill et al. 2000; Morse 1993). Many published studies have focused on the identification of the characteristics of patients who have fallen such as age, gender, mobility and mental status but have lacked power or historical controls to adjust for underlying trends in falls rates (Oliver, Hopper & seed 2000).
In a search for the evidence, a systematic review of randomised controlled trials examining multi-factorial interventions aimed at falls prevention concluded that the most efficacious intervention was a risk assessment and management program (Chang et al. 2004). As part of such programs, many assessment tools have been created with the aim of predicting falls and implementing preventative measures. Myers (2003) also exposed the lack of rigour in research design in a systematic review of 47 articles related to fall risk assessment where only 17 studies discussed reliability and validity of the instrument developed. The ability of instruments to discriminate between levels of risk was also questioned with sensitivity ranging from 54-91% and specificity ranging from 27-78%. Another major limitation was the lack of testing in clinical areas outside of the area in which the instrument was developed (Myers 2003).
More recently, a number of studies have considered fall risk assessment and support the contention that assessment is an important component of prevention programs. Unfortunately, several Australian studies evaluated prevention programs without including details of the assessment instrument itself (Hathaway et al. 2001; Mitchell & Jones 1996). Reference is made to the development and implementation; however, no information is given on reliability or validity of instruments. Other studies have adopted assessment tools developed elsewhere. Prior to commencing a prevention research initiative, Donoghue et al. (2003) reviewed patients who fell to validate components of a risk assessment instrument that was based on an instrument developed in Victoria. This study investigated the evidence on the characteristics of fallers rather than the instrument itself and aimed to test an abbreviated version in the future. Information on reliability or validity of the original instrument was not included. Therefore, recommendation to include these instruments in clinical practice is contrary to the principle that assessment information should be evidenced based (Antrobus & Kitson 1999; French 2005; Ingersoll 2000; Kitson 1997; Pearson 2003; Schneider, Elliott et al. 2003;Winch, Creedy & Chaboyer 2002).
Although many assessment tools have been created with the aim of predicting falls and implementing preventative measures, according to Evans et al. …