Interobserver agreement, sensitivity, and specificity of a "healed" chronic wound

Abstract

The primary end point of most wound healing studies is the clinically subjective evaluation of whether a wound is healed. Recently the Wound Healing Society proposed a definition of an acceptably healed wound. With the use of color transparencies of wounds with known clinical outcomes, the investigation presented here evaluated the reliability and validity of this definition as interpreted by 16 untrained observers. Interobserver reliability (Kappa statistic) was 0.68 (95% confidence interval: 0.66, 0.70), sensitivity 0.84 (confidence interval: 0.81, 0.90), specificity 0.92 (confidence interval: 0.88, 0.97), and positive predictive value 0.93 (confidence interval: 0.89, 0.97). We conclude that reliability and validity of judgments of wound healing are good using this definition of a healed wound, permitting its use in clinical investigations and routine clinical care.