Inter-rater reliability and Waterlow’s pressure ulcer risk assessment tool

Abstract

To ascertain whether a lack of inter-rater reliability with the original Waterlow (1996) pressure ulcer risk assessment scale is due to different perceptions of patients by nurses or different interpretations of Waterlow as a tool. A sample of 110 qualified nurses, who used the Waterlow pressure ulcer risk assessment scale in their daily work and were delegates at five study days, were given a case study and an uncompleted copy of the tool. They were asked to complete a risk assessment for the patient. The risk assessment score obtained by delegates was analysed using the Wilcoxon Signed Rank Test to measure the null hypothesis that there is no significant difference between the median of the nurses' scores and the patient's actual or 'gold standard' score. Nurses tend to over-rate (n=72, 65 per cent) rather than under-rate (n=25, 23 per cent) the patient's risk of developing a pressure ulcer. Only 13 of the 110 nurses (12 per cent) accurately rated the patient's score as 18. The Wilcoxon Test rejected the null hypothesis that there was no difference in the risk scores arrived at by individual nurses and the patient's actual score, that is, there is a significant difference between the scores obtained by the nurses in the study and the gold standard score. The results show poor inter-rater reliability when using the Waterlow pressure ulcer risk assessment scale. Part of the problem is that nurses are not using the tool in the way it was intended.