AIMS: Among various risk assessment scales for the development of pressure ulcers in long-term care residents that have been published in the last three decades, the Braden scale is among the most tested and applied tools. The sum score of the scale implies that all items are equally important. The aim of this study is to show whether specific items are of greater significance than others and therefore have a higher clinical relevance. DESIGN: Data analysis of six pressure ulcer prevalence studies (2004-2009). METHODS: A total of 17,666 residents (response rate 79.6%) in 234 long-term care facilities participated in 6 annual point prevalence studies that were conducted from 2004 to 2009 throughout Germany. For the classification of the sample regarding pressure ulcers as a dependent variable and the Braden items as predictor variables, Chi-square Automatic Interaction Detector (CHAID) for modelling classification trees has been used. RESULTS: Pressure ulcer prevalence was 5.4% including pressure ulcer grade 1 and 3.4% for pressure ulcer grades 2-4. CHAID analysis for the classification tree provided the item 'friction and shear' as the most important predictor for pressure ulcer prevalence. On the second level, the strongest predictors were 'nutrition' and 'activity' and on the third level they were 'moisture' and 'mobility'. Residents with problems regarding 'friction and shear' and poor nutritional status present with an 18.0 (14.8) pressure ulcer prevalence which is 3-4 times higher than average. CONCLUSION: CHAID analyses have shown that all items of the Braden scale are not equally important. For residents in long-term care facilities in Germany, the existence of 'friction and shear' as a potential and especially as a manifest problem has had the strongest association with pressure ulcer prevalence.
AIMS: Among various risk assessment scales for the development of pressure ulcers in long-term care residents that have been published in the last three decades, the Braden scale is among the most tested and applied tools. The sum score of the scale implies that all items are equally important. The aim of this study is to show whether specific items are of greater significance than others and therefore have a higher clinical relevance. DESIGN: Data analysis of six pressure ulcer prevalence studies (2004-2009). METHODS: A total of 17,666 residents (response rate 79.6%) in 234 long-term care facilities participated in 6 annual point prevalence studies that were conducted from 2004 to 2009 throughout Germany. For the classification of the sample regarding pressure ulcers as a dependent variable and the Braden items as predictor variables, Chi-square Automatic Interaction Detector (CHAID) for modelling classification trees has been used. RESULTS: Pressure ulcer prevalence was 5.4% including pressure ulcer grade 1 and 3.4% for pressure ulcer grades 2-4. CHAID analysis for the classification tree provided the item 'friction and shear' as the most important predictor for pressure ulcer prevalence. On the second level, the strongest predictors were 'nutrition' and 'activity' and on the third level they were 'moisture' and 'mobility'. Residents with problems regarding 'friction and shear' and poor nutritional status present with an 18.0 (14.8) pressure ulcer prevalence which is 3-4 times higher than average. CONCLUSION: CHAID analyses have shown that all items of the Braden scale are not equally important. For residents in long-term care facilities in Germany, the existence of 'friction and shear' as a potential and especially as a manifest problem has had the strongest association with pressure ulcer prevalence.
Authors: Nancy Bergstrom; Susan D Horn; Mary Rapp; Anita Stern; Ryan Barrett; Michael Watkiss; Murray Krahn Journal: Ont Health Technol Assess Ser Date: 2014-10-01
Authors: William V Padula; Robert D Gibbons; Peter J Pronovost; Donald Hedeker; Manish K Mishra; Mary Beth F Makic; John Fp Bridges; Heidi L Wald; Robert J Valuck; Adam J Ginensky; Anthony Ursitti; Laura Ruth Venable; Ziv Epstein; David O Meltzer Journal: J Am Med Inform Assoc Date: 2017-04-01 Impact factor: 4.497