ABSTRACT
Resident duty-hour restrictions demand effective communication and teamwork in patient care. The process of resident sign-out is a potential source of miscommunication and medical error. Resident sign-out was followed over a 3-month period. Residents signing out to the night coverage team were asked to identify two patient groups: (1) problem patients who were especially ill and likely to present specific clinical problems; and (2) nonproblem patients who were likely not to be at risk for a problem requiring attention. Data on adverse events collected by the night float resident were classified into three categories: a problem predicted during sign-out in a problem patient, an unpredicted problem on a problem patient, and an unpredicted problem on all other patients. Resident sign-out accurately predicted only 42 per cent of adverse events. Only one third of major adverse events were predicted at checkout. One third of events occurred in patients identified at sign-out as being in the nonproblem group. The process of transfer of care must be standardized and individual practices reviewed to prevent error. Instruction on proper transfer of care and illustrations of potential points of breakdown should be given for all levels of training.
