Validation and reliability of visual assessment with a shade guide for tooth-color classification

This study evaluated the validity and reliability of the visual assessment of tooth color using a commercial shade guide. Ninety-two individuals were randomly selected from subjects enrolled in a randomized controlled trial comparing two formulations of carbamide peroxide. Initially, each individual had the color of his or her six maxillary anterior teeth (n=552) determined by one examiner using a digital spectrophotometer (Vita Easyshade). Then, a visual assessment was made by two calibrated examiners using a shade guide (Vitapan Classical). Reliability was evaluated either by kappa coefficient when all the colors were grouped into two categories (dark and light) or by weighted kappa when all 16 colors of the shade guide were considered. The digital spectrophotometer (Vita Easyshade) was used as the gold standard to calculate sensitivity and specificity of the visual assessment. The reliability of the visual assessment using two color categories was "substantial" (k=0.69). When all colors of the shade guide were considered, the reliability was "mild" (kw=0.33). Sensitivity and specificity of the visual assessment with respect to the gold standard was 86.9% and 81.9%, respectively. Despite its subjectivity, visual assessment of tooth color using the Vitapan Classical shade guide is a valid method, with good reliability for differentiating between dark and light colors.