Interrater Reliability of the Braden and Braden Q by Skin Champion Nurses

The interrater reliability of the Braden Q skin risk assessment scale has never been reported. The purpose of the study was to assess the interrater reliability among pediatric Skin Champion (SC) nurses with the use of the Braden and Braden Q scales. The pilot study included 16 paired SC nurses. Each pair of nurses attempted to assess 8 patients using the Braden and Braden Q scales. However, patient care requirements at the time of the study assessments limited the number slightly. The actual number of assessments with the Braden scale was n = 52 and the Braden Q was n = 63. The Intra-class Correlation Coefficient (ICC) for the Braden scale was 0.894, 95% confidence interval (CI) (0.823, 0.938), which is excellent agreement. The ICC for the Braden Q was 0.726, 95% CI (0.585, 0.824), which is fair to good agreement. Among the six subcategories on the Braden scale, mobility and activity had higher agreement scores among the SC nurses. Among the seven subcategories on the Braden Q scale, mobility and sensory perception had higher agreement scores. Nutrition and friction/shear subcategories on both scales had the lowest agreement scores. Subcategories with the lowest agreement usually have the greatest measurement error. Possible sources of error include unclear definitions of scoring criteria, different clinical data pulled from different locations in the chart. Error can be reduced by clarifying the subcategory definitions and standardizing the data used for the assessment and the location of each data point in the EMR. A high interrater agreement is the goal because it provides confidence that the scale is used reliably to identify high risk patients who require additional care to prevent harmful events.