Objectives a) To quantify the use of do-not-resuscitate orders in a tertiary care children's hospital; and b) to characterize the circumstances in which such orders are written. Design Retrospective chart review. Setting University teaching hospital. Patients All inpatients who died in an urban children's hospital over a 1-yr period of time. Interventions None. Measurements and Main Results The hospital records of 54 of 69 patients who died were reviewed. Eighty-two percent of patient deaths occurred in the ICU; 13% of patient deaths occurred in the operating room, and 5% occurred in a medical ward. Other findings included the following: 25 (46%) of 54 patients died after attempted cardiopulmonary resuscitation; 13 (24%) patients were brain dead; and 16 (30%) died with a do-not-resuscitate order in effect. Age was associated with resuscitation status: do-not-resuscitate orders were written for five (21%) of 22 infants (p < .002). Fifteen of 16 do-not-resuscitate orders were written for patients who were in the ICU, where they remained until death. Findings in patients when the do-not-resuscitate orders were written were as follows: 15 ICU patients were receiving mechanical ventilation; 14 (95%) of 15 were receiving inotropic agents; 12 (80%) of 15 patients were receiving narcotic analgesics; and one (6%) patient was being dialyzed. At least one therapeutic modality was withdrawn in 7 (44%) of 16 patients. Do-not-resuscitate orders followed documented conferences with physicians and family members in 13 (81%) of 16 cases. These discussions were initiated by physicians in 12 (92%) of 13 cases. Conclusions Do-not-resuscitate orders in diatric patients are written more often in than younger hospitalized children who Most do-not-resuscitate orders are written patients who are receiving aggressive med therapy in the ICU. (Crit Care Med 1 21:52–55)