Objective Newborn encephalopathy is a common cause of ethical dilemmas in the neonatal intensive care unit. In the setting of poor prognosis life-sustaining treatment (LST) is sometimes withdrawn. Prognostic assessments performed soon after birth are less accurate than those performed later. However, if decisions are delayed infants with severe impairment may survive even if LST is withdrawn, or may die only if artificial feeding is withheld. This study aimed to use decision theory to help manage uncertainty in the timing of prognostic tests in newborn encephalopathy.
Methods A model of expected value was generated to compare early prognostic testing with late prognostic testing. One-way and two-way sensitivity analyses were performed for variables including test characteristics, the pre-test probability of poor outcome and the values placed upon different outcomes.
Results If the pre-test incidence of poor outcome is low the model suggests deferral of testing and decision-making. The model was less sensitive to test characteristics and the value placed upon poor outcome. Early testing is favoured if there is negative value attached to late withdrawal of LST, or if more than a small proportion of infants with poor outcome survive following late decision-making.
Conclusions This model may provide general guidance for decisions, although it is not likely to be useful in specific cases. It highlights the need to improve our ability to prognosticate for infants with newborn encephalopathy. Improvements in palliative care for newborn infants with predicted severe impairment may help resolve the problem of decision-making in the face of uncertainty.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.