Recent attempts to improve surgical resident working conditions have taken many forms. We evaluated a system in which a well-trained physician extender had been hired to assist residents taking call on a busy cardiothoracic teaching service. The physician extender (''night nurse,'' NN) helped with perioperative care using well-defined protocols. The NN, who was in-house Sunday-Friday, 7 PM-7 AM, rounded with the residents at the beginning of the evening. Concerns were discussed and care plans formulated. Thereafter, all pages (except codes and extreme emergencies) were directed to the NN, freeing the residents to complete work-ups, patient assessments, or study. The NN assessed patients and initiated care plans including orders that followed either care protocols or plans previously arranged with the resident. For unanticipated concerns, the resident was notified for input and/or patient assessment. For 30 consecutive nights, the patient load and acuity were evaluated. Residents and NN kept diaries of all pages received. The residents also documented time slept and times awakened. Residents received 10 times fewer calls when the NN was available (21.8 +/- 10.5 vs 2.9 +/- 2.4) and slept an average of 2.5 hr more (135 +/- 106.1 vs 286.2 +/- 68.2 min). Care was maintained as judged by morbidity and mortality statistics. Such a system has allowed us to avoid cross coverage, thereby maintaining resident continuity of care and involvement in meaningful care plans, while providing increased time for patient evaluation, self-education, and increased uninterrupted sleep. (C) 1994 Academic Press, Inc.
