Observations of residents' work activities for 24 consecutive hours: implications for workflow redesign.

PURPOSE: To examine resident workflow as part of an institutional approach to redesigning the processes of health care delivery. METHOD: In 2003 the authors observed the workflows for 24 hours of seven residents who were at various levels of training (two each from the internal medicine, pediatrics, and obstetrics and gynecology programs, and one from general surgery) at Denver Health Medical Center, an urban, public teaching hospital. RESULTS: Although the residents spent varying proportions of their time in various activities, all had extremely fragmented workflows as they engaged in from 5.0 to 11.3 different activities per hour of nonsleeping time, many of which required only minutes to complete. All residents experienced frequent interruptions and changes in focus. The internal medicine and surgery residents spent large amounts of time traveling, covering three and six miles, respectively, during their 24-hour shifts. Three of the residents slept between one-quarter and one-third of their time on duty (one without any interruption). CONCLUSIONS: The authors suggest that fragmented workflow exists in all residency programs and that applying the same work limitations to all residents in all training programs (to reduce fatigue-related errors) may be overly restrictive. Improving these processes of care will be difficult and will likely require analytic skills and knowledge of systems engineering that most physicians do not have.