Due to actual standard of neonatal care bilirubin is monitored in neonates several times after birth to avoid hyperbilirubinemia. As the peak is often one or two days after discharge from hospital a more precise predication than the actual one is desirable to avoid needlessly follow-ups. The precision of an algorithm to predict bilirubin values 24-48h into the future is evaluated in this study.

Neonatal jaundice is common and mostly transient without need for intervention. To identify infants at risk of bilirubin toxicity, bilirubin is measured several times after birth and before discharge home. As the peak bilirubin concentrations are reached at days 4-6 after birth, and most infants are discharged home around 48 hours after birth, parents are often asked to return to hospital one or two days after discharge with their newborn baby to have the bilirubin value checked in order to prevent that hyperbilirubinemia requiring treatment is not missed. A precise prediction of the course of bilirubin values based on the measurements done during the initial hospital stay after birth would be desirable to avoid unnecessary follow ups (both for the families and the health care system).

This study aims to evaluate the precision of the NeoprediX B1 algorithm. This algorithm predicts bilirubin values 24-48h into the future based on preceding values and some clinical variables. Five different scenarios will be explored, differing in the type, number and timepoints of measurements used for the prediction as well as the timepoint that is to be predicted.

Inclusion Criteria:

• Newborn infant
• gestational age at birth ≥ 34+0/7 weeks

Exclusion Criteria:

• Birth weight < 1500g
• Gestational age at birth > 42+6/7 weeks
• Genetically defined syndrome
• Severe congenital malformation adversely affecting life expectancy or admission for a priori planned palliative care
• Parents not fluent in German

Newborn infants