September 18, 2009

Congressional Testimony on Resident Work Hours

Testimony of Peter Lurie, M.D., M.P.H. and Hillary Peabody, M.P.H.
Health Research Group at Public Citizen
Before the Medicare Payment Advisory Commission

Thank you for the opportunity to address the Commission.  In light of today’s focus on medical education, we would like to bring to your attention an issue that has critical importance both for medical education and the overall quality of medical care provided to Medicare beneficiaries.

Last December, the Institute of Medicine (IOM) released its landmark report, Resident Duty Hours: Enhancing Sleep, Supervision, and Safety, the most comprehensive study of resident work hours conducted to date.  Research reviewed in the IOM report demonstrates conclusively that working long hours is linked with a higher incidence of medical errors among residents and that reducing those hours can, in certain circumstances, reduce medical error rates.

Among the studies reviewed in the IOM report is a randomized, controlled trial in two intensive care units of the Brigham and Women’s Hospital.[1] Interns were randomly assigned to a standard schedule or a schedule with reduced work hours (including a night float).  The first-year residents on the intervention schedule made 26 percent fewer serious medical errors and 36 percent fewer non-intercepted serious medical errors.  After reviewing all the available data on resident work hours, the IOM concluded that there is “a robust evidence base linking fatigue with decreased performance in both research laboratory and clinical settings.”[2]  

Since the publication of the IOM report, another prospective interventional (although non-randomized) study, based on data from the United Kingdom, has been published.  It indicates a similar 33 percent reduction in medical errors and a nearly statistically significant 31 percent reduction in non-intercepted serious adverse events.[3]

The IOM made three overarching recommendations: work hour adjustments, enhanced supervision, and workload reduction.  It further called for the Accreditation Council on Graduate Medical Education (ACGME) and other organizations charged with implementing these recommendations to do so “with all deliberate speed.”  Yet, ten months after its release, the report seems to be flying well under the radar.  Its recommendations have been not been systematically implemented and the ACGME gives every indication that it will resist many of the IOM recommendations.

In fact, even the inadequate voluntary guidelines instituted by the ACGME in 2003 are poorly enforced, so actual hours worked often far exceed the ACGME’s guidelines.  In the first year after the ACGME’s hours reforms were introduced, 84 percent of interns and 91 percent of teaching facilities had a work hours-related violation.[4]  The new guidelines have had only a limited impact upon actual hours worked.  Mean weekly hours worked decreased by only four hours (from 71 hours to 67 hours) and extended shift hours decreased by only two hours (from 31 hours to 29 hours).

We are also concerned that the patient’s view of this extremely important issue is not adequately reflected in the public discourse.  This is particularly important as it is the public that suffers from any medical errors that may result from long work hours.  The available evidence suggests that the public is deeply concerned about the current work hours of medical residents.  In a 2002 national public opinion poll conducted by the National Sleep Foundation, 70 percent of respondents reported that they were “somewhat likely” or “very likely” to request another doctor if they learned that their doctor had been working for 24 hours.[5] In addition, 74 percent of respondents in a 2004 Kaiser Family Foundation public opinion poll listed “stress, overwork, or fatigue of health professionals” as a “very important cause of medical errors” and 66 percent agreed that reducing the work hours of doctors to avoid fatigue would be a “very effective” way to reduce medical errors.[6] We strongly urge the Commission to take into account the concerns of consumers and to make resident work hours a central focus of efforts to improve patient safety.

In its June 2009 report, the Commission observed that “Medicare – with an enormous financial stake in health care and graduate medical education – has never specifically linked any of its direct [graduate medical education] or [indirect medical education] subsidies to promoting or fostering important goals in medical education.”[7]  Resident work hour reform presents just such an opportunity.  A reimbursement formula that takes into account compliance with the IOM report’s recommendations would further this goal.  While compensating for reductions in resident hours with the hiring of substitute personnel would not be cheap (the IOM estimates it would cost $1.7 billion dollars per year, which represents only 0.4 percent of Medicare outlays[8]), the savings in medical errors averted may well be on a scale similar to these costs.  Most importantly, unnecessary medical errors and needless patient suffering will have been averted.

---