Background: Concerned about the impact of house officer (HO) fatigue on education and the quality of patient care, the ACGME instituted work restrictions for HOs effective July 1, 2003. Proponents believe HO fatigue contributes to in-hospital errors. Opponents argue that decreasing HO work hours will increase errors due to patient handoffs and HO cross-coverage. We surveyed internal medicine faculty and HOs to understand clinician perceptions of the impact of the ACGME regulations.
Methods: We created a written survey instrument based on a literature review, expert opinion, and focus groups of HOs, staff internists, and nurses. The survey asked respondents to recall types of errors and contributing factors that occurred in the 3 months prior to work-hour restrictions,and to predict how the restrictions would affect patient care and HO education. We administered the survey in July and August 2003 via email and in person to PGY2/3 medical HOs and medical ward and ICU attending physicians at Beth Israel Deaconess Medical Center. Responses were scored on the Likert scale. We calculated the percentage of respondents who agreed/strongly agreed or disagreed/strongly disagreed with statements and used the rank sum test to compare HO and attending physician responses.
Results: We received completed surveys from 81 of 95 HOs and 40 of 104 attending physicians, including all 11 staff hospitalists who, together, accounted for approximately 65% of inpatient medical admissions. HOs were more likely than attendings to attribute errors to high census (70% vs. 22%, p=0.001) and fatigue (52% vs. 38%, p=0.02); HOs were less likely than attendings to attribute errors to cross-coverage (53% vs. 79%, p<0.001) and lack of experience or knowledge (54% vs. 68%, p=0.03).
Fifty-two percent of HOs and 22% of attending physicians agreed that fatigue contributed to HO errors in the previous 3 months, but only 20% of HOs and 10% of attendings predicted that fatigue would contribute to errors after implementing work-hour restrictions. Despite this expected reduction in HO fatigue, a majority of HOs and attendings disagreed that quality of care (55% and 69%) and continuity of care (89% and 89%) would improve. Ninetyfive percent of HOs and 98% of attendings predicted that in the new system errors would occur as a result of cross-coverage, compared with 53% and 79%, respectively, at baseline. Half the HOs (55%) and attendings (50%) believed that HO errors would increase overall after the change.
Conclusion: Clinicians were skeptical that ACGME work-hour restrictions would improve care or decrease errors. Instead, many HOs and attendings predicted that the new regulations would change the underlying cause of error from fatigue to cross-coverage, and that the total number of errors would increase. If academic medical centers and their patients are to reap the intended benefits of work-hour restrictions, residency directors will need to develop and implement skill-building initiatives focused on cross-coverage.
Dr. Kripalani can be contacted at email@example.com.