No medical professional is ethically obligated to provide futile care. However, to be useful futility must be recognized prospectively; that is, in advance of the impending demise of the patient as opposed to after the fact. We wondered what percentage of futile care in the NICU (defined as care devoted to infants who would die before discharge) was recognized to be futile at the time.

Methods: Using a prospective study design, we surveyed on a daily basis the judgments of medical caretakers (md's, rn's) regarding future survival of patients in their care. We compiled ≈6 assessments/patient/day on each day of life (DOL) for 270 infants admitted to our NICU who required mechanical ventilation. We then retrospectively identified the infants who did not survive, and matched the averaged daily prognistications of the caretakers with the ultimate fate of the infants. Recognizing that many neonatologists provide a 72-hour `trial of therapy' for neonates to `declare' themselves, we concentrated particularly on futility assessments after DOL 3.

Results: We obtained roughly 28,000 prognostications concerning 270 ventilated infants during the 11 month study period. 56 (22%) of 270 infants died; 24 (43%) of these died by DOL 3. Of doomed infants alive on DOL 4, 50% died between DOL 4 and DOL 9. 91% of these infants were predicted to die by every caretaker on each day of life (consistent, accurate prognistication of non-survival). In contrast, only 21% of infants who died after DOL 10 were predicted to die more than two days before their death. Despite comprising only 28% of all NICU non-survivors, these late-dying infants accounted for 73% of NICU bed-days devoted to infants who ultimately did not survive.

Conclusions: 1) For infants who died within the first ten days of life, futility prognistications were excellent. These early-dying infants represented roughly three-fourths of all NICU deaths. 2) For infants who died after DOL 10, futility predictions were very inaccurate, neither specific, nor sensitive, nor timely. 3) These unpredicted, late-dying infants accounted for the vast majority of NICU resources devoted to non-survivors. 4) Futility in the NICU remains unrecognized, and perhaps unrecognizable, for infants after the first ten days of life.