OBJECTIVES: The aims of this study were to examine and compare the predictive validity of 4 risk assessment scales used for the prediction of pressure ulcer (PrU) development and to identify risk factors.
DESIGN: Cross-sectional descriptive study.
SETTING: A general hospital in Sweden.
PARTICIPANTS: Patients (all aged ≥18 years) admitted to medical, surgical, orthopedic, oncology, and rehabilitation wards. Of 412 patients available, a total of 346 patients participated in the study.
METHOD: Data were collected using the Swedish version of the European Pressure Ulcer Advisory Panel minimum data set as well as the Norton, Modified Norton, Braden, and Risk Assessment Pressure Sore (Ulcer) (RAPS) scales. The predictive validity was estimated by measuring sensitivity, specificity, positive predictive value, and negative predictive value. Multiple logistic regression analysis was used to determine risk factors associated with PrUs.
RESULTS: The RAPS scale reached best balance between sensitivity and specificity at the recommended cutoff level of ≤29, followed by the Braden scale and the Norton scale at recommended cutoff levels ≤18 versus ≤16, respectively. The modified Norton scale also reached an acceptable balance between sensitivity and specificity but at the cutoff level of ≤23, which is a higher cutoff level than recommended. General physical condition, physical activity, moisture, friction, and shear emerged as significant risk factors.
CONCLUSIONS: The results support that the recommended cutoff levels of the RAPS, Norton, and Braden scales are valid in a general hospital setting. However, the recommended cutoff level of the modified Norton scale (≤20) has to be increased when used in this care context.
Ulrika Källman, RN, is a PhD student, Department of Medicine and Health, Faculty of Health Sciences, Linköping University, Linköping, Sweden, and Registered Nurse, Department of Dermatology, Södra Älsvorgs Hospital, Borås Sweden. Margareta Lindgren, PhD, RN, is Associate Professor, Department of Medicine and Health, Faculty of Health Sciences, Linköping University.
Acknowledgments: The authors thank the study participants for their patience and cooperation and also the partnering organizations and their senior nurse leaders and staff for all the help given during the research project. Additional thanks are extended to Södra Älvsborg Hospital, Borås, and Linköping University, Linköping, Sweden. The authors have disclosed that they have no financial relationships related to this article. Submitted October 12, 2012; accepted in revised form March 8, 2013.