The correlation of transcutaneous bilirubin measurements with serum bilirubin concentrations is not good enough to allow for accurate prediction of the serum values. To impose the jaundice meter's potential clinical usefulness, we evaluated 344 paired jaundice meter-serum bilirubin measurements in 125 infants, using new guidelines from the marketing company which were designed to identify which infants require serum bilirubin determinations rather than to predict the actual bilirubin values. Use of the new guidelines correctly assessed the need for serum determinations in most infants, but false positives and, more importantly, false negatives (missed high serum values) did occur.