A review of evidence on the effects of worker fatigue on patient safety is included in the Agency for Healthcare Research and Quality’s (AHRQ) report Making Health Care Safer: A Critical Analysis of Patient Safety Practice (Jha et al. 2001). Consistent with the above evidence, the report notes that sleep deprivation leads to decreased alertness and poor performance on standardized testing, and that shift workers in particular experience disturbances in their circadian rhythms and tend to perform less well on reasoning and nonstimulating tasks.

Evidence on Nurse Work Hours and the Commission of Errors

Researchers conducting the evidence review presented in the AHRQ report cited above (Jha et al., 2001) were unable to locate research that could help identify the specific numbers of hours worked by health care personnel, including nurses, beyond which patient safety is threatened. These researchers noted inconsistent research findings with respect to ideal shift length for enhanced worker performance. The report suggests that while multiple studies have sought to document the impact of fatigue on the performance of medical personnel, these studies have been limited by poor design or outcomes that did not correlate well with medical error.

This situation has been improved by a 2002 study funded as part of AHRQ’s initiative to examine the effects of working conditions on patient safety. This study documented the work patterns of a sample of hospital staff nurses randomly selected from the membership of the American Nurses Association (ANA). The sample frame consisted of full-time hospital staff nurses (unit based, not working through a temporary agency) with no administrative or educational responsibilities. The study measured the effects of nurse work hours on patient safety by (1) documenting the total scheduled and unscheduled hours worked by nurses; (2) describing the nature of nurses’ overtime work hours in terms of what proportion of hours worked were overtime hours, how often nurses worked overtime, and whether overtime was voluntary or mandatory; and (3) determining whether there was an association between errors and near-errors and the numbers and types of hours worked by the nurses.

Study participants recorded information about their scheduled work hours, actual work hours, errors, and near-errors daily in a diary for 28 days. Nurses were also asked to describe all errors and near-errors. The researchers then categorized each error or near-error by type (e.g., medication administration, procedural, transcription) based on the nurse’s description. Error rates per hour were calculated according to the number of errors and hours worked, adjusting for multiple work shifts for the same nurse. The associations between error rates and both overtime and scheduled work



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement