Validation of a neonatal skin color scale.

Ethnic classification does not correlate well with skin tone. As there are no neonatal skin color scales, we aimed to create and validate one of our own. After creating the scale and briefly training our staff, we conducted a prospective, observational study to assess reproducibility and correlation of each scale color with the melanin and erythema indexes and transcutaneous bilirubin. The reliability of our color scale was measured using Kappa agreement (and its 95% confidence interval) and the concordance index by comparing inter-observer classification of neonatal skin color. We also calculated inter-rater agreement with the intraclass correlation coefficient (ICC). The Kendall tau-b correlation coefficient was used to test the correlation between our color scale and the Mexameter® MX 18. We obtained data from 258 newborns. Inter-observer agreement on color assignment was 83.2%. Median melanin index was significantly different among the 4 color groups, whereas erythema index and transcutaneous bilirubin were not.Conclusions: Our proposed neonatal skin color scale correlates well with the melanin index at 24 h of life, increasing from colors 1 to 4, and the only chromophore different among our four color groups is melanin. Scale color assignment is reproducible. Therefore, it can be used to classify neonatal skin color. Further research is warranted to assess the clinical relevance of these findings. What is known: • Classifying neonates by skin color is difficult because to date there are no skin color scales available based on real skin tone regardless of ethnicity or country of origin. • Skin color differs among individuals from a given ethnic group and depends, among others, on melanin and hemoglobin. What is new: • We created a neonatal skin color scale based on real skin color. • We conducted a study to validate it, and confirmed a good inter-observer agreement in color assignment as well as a good correlation between each color in the scale and the median melanin level.