Revisiting Duty Hour Restrictions: Any Evidence That Patients Have Benefited?

Adverse events occur in approximately 3% of all hospital admissions. Around one-half of these events are attributable to preventable errors, and 10% to 14% result in death.¹ Similar rates of adverse events have been reported in obstetrics.² The 2002 National Survey of physicians, patients, and their families reported that 35% of physicians had experienced an error in their own care or that of a family member (of which 18% were serious), and 42% of the general public had experienced an error in their own care or that of a family member (of which 24% were serious).^3–5 In all, it is estimated that between 44,000 and 98,000 Americans die in hospitals each year as a result of medical errors.⁶ This is equivalent to 2 fully loaded Boeing 747s going down each and every day of the year.

In November 1999, the Institute of Medicine (IOM) published their landmark report, To Err Is Human: Building a Safer Health System,⁶ in which they called for a national effort to reduce medical errors. This publication reframed the discussion on health care in the United States, and motivated a broad array of stakeholders to engage actively in patient safety initiatives. It is unlikely that we can prevent all medical errors, and, as stated by Dr. Lucian Leape: “We need to accept the notion that error is an inevitable accompaniment of the human condition, even among conscientious professionals with high standards.”⁷

A number of factors conspire to make obstetrics a particularly high-risk area prone to medical errors. These include the fact that we work with numerous caregivers from different disciplines, have multiple handoffs, and are surrounded by high-tech equipment and powerful drugs. We are asked to make rapid decisions in an environment prone to distraction with rare but potentially serious long-term consequences for both individuals and families. And, because patient volume and flow are variable and events unpredictable, we are often required to work long hours while sleep deprived. It is this last issue that has received the most attention.

Sleep deprivation is associated with impairment in both cognitive and psychomotor skills, including an increase in surgical complication rates. Acute sleep deprivation of 24 hours results in cognitive performance impairment equivalent to a serum alcohol concentration of 0.10%, a level considered functionally (and legally) intoxicated in the United States.⁸ In light of such data and with a view to promoting patient safety and resident well-being, the Accreditation Council for Graduate Medical Education (ACGME)—the agency that accredits the majority of residency training programs in the United States—established duty hour limits nationally and across all specialties in July 2003 (Table 1).^9,10 “Substantial compliance” with these recommendations is now required for continued ACGME accreditation, with punitive measures being dealt out for programs that show themselves to be repeat offenders. In December 2008, at the request of the United States House of Representatives Committee on Energy and Commerce as part of its investigation into preventable medical errors, the IOM issued a subsequent report recommending that additional actions be taken to further reduce resident fatigue and ensure patient safety (summarized in Table 1).¹¹ They mandated that ACGME take action on all of these recommendations within 24 months. In response, a task force from ACGME has been actively soliciting input from its constituents¹² and is due to present its findings in July 2010.

Table 1.

Comparison Between the Current ACGME Duty Hour Limits and the More Stringent IOM Recommendations

	Current ACGME Duty Hour Limits9,10	2008 IOM Recommendations11
Maximum number of hours worked per week	80 h/wk, averaged over a 4-wk period	No change
Maximum number of hours worked per shift	30 h	30 h with 5 h protected sleep period between 10 PM and 8 AM
		16 h with no protected sleep period
Minimum number of hours off between shifts	10 h	10 h after day shift
		12 h after night shift
		14 h after any extended period of 30 h
Minimum time off per week	1 d/wk, averaged over a 4-wk period for a total of 4 d/mo	5 d/mo
		1 d/wk, no averaging
		1 48-h period/mo
Maximum in-house call	Every third night, averaged over a 4-wk period	Every third night, no averaging
		No more than 4 consecutive night calls
		48 h off after 3 or 4 consecutive night calls
New patients	No new patients after 24 h	No new patients after 16 h
Moonlighting	Internal (not external) moonlighting is counted against the 80-h weekly limit	Both internal and external moonlighting are counted against the 80-h weekly limit
		All duty hour limits apply also to moonlighting
Regulatory oversight	“Substantial compliance” is required for continued ACGME accreditation	Duty hour limits are an “absolute rule” that will be rigorously and rigidly enforced

ACGME, Accreditation Council for Graduate Medical Education; IOM, Institute of Medicine.

The more stringent duty hour limits proposed by the IOM have received widespread criticism. The original goals of the duty hour limits were to improve resident well-being and patient care, and it is not clear whether either of these objectives has been met. Although it is likely that the overall quality of residents’ lives has been improved with the introduction of the 80-hour work week, the data (although limited) suggest that residents do not spend their off-duty time catching up on sleep or reading, and they are no less “fatigued” or more “fit for duty” than residents were previously.¹² This is consistent with data from New York State, which has had global standards on resident duty hours for almost 20 years. Moreover, there is increasing evidence that the 80-hour work week may have had a number of unanticipated adverse effects. There are now more handovers meaning more opportunity for error, performance on national Council on Resident Education in Obstetrics and Gynecology (CREOG) examinations has not improved and appears to be falling, and graduating physicians may be less well prepared for independent practice because they have had less clinical experience and less opportunity to be given graded decision-making responsibility by their attending physicians.^12–14 In addition, the implementation of the proposed IOM duty hour limits would be enormously expensive. It is estimated that an additional $1.7 billion (which equates to 9% of the money currently spent on graduate medical education each year) would be required to cover these clinical responsibilities, which does not include the costs of the other IOM recommendations (eg, closer supervision of junior residents, more stringent monitoring of duty hours, and more frequent site visits by regulatory authorities). The abject lack of data demonstrating an improvement in short- and long-term patient outcome is particularly troublesome.

If it is still unclear whether the current duty hour limits have met their allotted goals, and possible that they even may have had a negative effect on resident education and training, why then impose even more draconian restrictions? Improvements of the magnitude envisioned by the IOM require a national commitment to well-defined objectives focused on improving patient care with clear outcome metrics and a tracking system set in place prospectively to determine whether these objectives are being met. Equal attention should be paid to other initiatives that have been shown to minimize preventable medical errors, including the implementation of electronic health records, the introduction of standardized management algorithms, meticulous peer review, and team training.^15,16 The decision of whether to impose additional restrictions on resident duty hours should be based on the prevailing evidence, and not on political expedience.