Misclassification due to body hair and seasonal variation on melanin density estimates for skin type using spectrophotometry

Recent advances have enabled quite accurate estimations of cutaneous melanin density by spectrophotometry using reflectance of light at wavelengths 400 and 420 nm. Our purpose was to assess the effect of body hair and seasonal variation at the upper inner arm and buttock on measurements of melanin density. We estimated melanin density of 104 volunteers at 3-monthly intervals over 12 months both before and after shaving. Removing body hair at the upper inner arm had no effect, but substantially reduced melanin estimates at the buttock in men. Significant seasonal variation was only observed at the upper inner arm, with highest readings in summer-autumn. In case-control studies, misclassification due to body hair at the buttock and seasonal variation at the upper inner arm could affect the observed odds ratio substantially. However, both sources of error can be reduced by careful attention to key aspects of study design. © 2002 Elsevier Science B.V. All rights reserved.