The two reports, both written by Kevin Volpp and colleagues at the Philadelphia Veterans Affairs Medical Center and University of Pennsylvania School of Medicine, looked at 30-day mortality rates for some 8.5 million Medicare beneficiaries and about 318,600 patients at VA acute-care hospitals who were admitted for heart attack, congestive heart failure, gastrointestinal bleeding or stroke, or for general, orthopedic or vascular surgery. Statistics for the academic years 2000-03 were compared with the academic years 2003-05 in an effort to gauge the impact of the Accreditation Council for Graduate Medical Educations requirement to limit residents to 80-hour workweeks starting in 2003.
Mortality decreases in the VA hospitals were in the 11.1% to 13.9% range for nonsurgical patients, with more improvement shown in hospitals, with higher teaching intensity (as measured by a resident-to-bed ratio). No such improvement was shown for the surgical patients. (The various changes in mortality measures for the Medicare patients were deemed statistically insignificant.)
Mortality is a crude indicator, but its an important one, said Ken Kizer, who led the VA health system from October 1994 to September 1999, and is now chairman and chief executive officer of Medsphere Systems Corp. This doesnt affirm those crying that this would cause all sorts of problems with discontinuity of care. I dont think those concerns are borne out by these data.
The fact that you dont see negative effects may be the more important finding in the long term.
Foster said that if any improvements resulted from decreasing resident work hours, there was a need for more sensitive indicators. Foster added, however, that if an indicator could be found for patient harm, then maybe a link to resident work hours could be drawn.
Some hospital executives question the value of the limits. In Denver, University of Colorado Hospital President and CEO Bruce Schroffel said the requirements have led to higher costs because of the wider employment of nurse practitioners, physician assistants, and other physician extenders, and he would be shocked if these changes contribute to an increase in quality.
Robert Dickler, senior vice president of healthcare affairs at the American Association of Medical Colleges, suggested that studies be done to see if there are links to resident work hours and changes in patient morbidity or a hospitals performance on other quality indicators. Dickler said one thing that needs to be measured is the impact the lighter workload has on resident education.