Examining the validity of pressure ulcer risk assessment scales: a replication study.

Risk assessment scales (RASs) intended to identify patients most at risk of developing pressure ulcers have been widely used for many years. Numerous studies have evaluated their predictive validity but potential bias has been inherent in the design of all. To overcome these problems a simulation study was conducted in which clinical nurses were asked to identify the degree of risk experienced by four patients employing the three RASs discussed most frequently in the literature (Norton, Braden and Waterlow Scores). These findings were compared with nurses' clinical judgment rated on a visual analogue scale. The simulations consisted of high-resolution photographs accompanied by case studies of the patients. The nurses' scores were compared to estimates of risk generated by an expert panel. Nurses' clinical judgment agreed much more closely with expert opinion than any of the RASs. A replication study was undertaken to confirm these findings. One hundred and fifteen nurses participated in replication. Again the nurses' clinical judgment matched expert opinion much more closely than the results of the RASs. Replication also drew attention to a number of methodological issues which deserve consideration when using simulation to test the effectiveness of clinical tools and the need to establish adequate measures of external validity whenever use of this method is contemplated.

Establishing the validity of pressure ulcer risk assessment scales: a novel approach using illustrated patient scenarios.

Preventing pressure ulcers is an important nursing goal and over the years a number of risk assessment scales (RASs) have been developed to expedite clinical judgement. The aim of this study was to examine the validity of the three most commonly used RASs compared to nurses' own clinical judgement. Patient simulations were presented to 236 clinical nurses. Nine hundred and forty one assessments were completed and compared to the ratings from a panel of tissue viability experts. Clinical judgement exactly matched expert opinion (69.1%) more often than assessment with any of the RASs. The Waterlow Score matched exactly in 20% of cases, the Braden Scale in 8.5% of cases and the Norton Score in 4.6% of cases. Thus none of these RASs can be considered valid, assuming that the expert panel genuinely reflected the 'gold standard' in terms of the external criterion. The implications for clinical practice are far-reaching considering the extent to which RASs are currently used to reach important clinical decisions relating to the deployment of expensive pressure-relieving aids and nursing time.