Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

Hospital Systems and Quality of Care

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Complication rates may not reflect hospital quality of care for cardiac bypass surgery patients

Patient complication rates should not be used to judge how well hospitals care for patients undergoing coronary artery bypass graft (CABG) surgery until more is known about how CABG complication rates reflect care quality, according to a study supported by the Agency for Health Care Policy and Research (HS06560). The study found that many hospital characteristics that are sometimes thought to be associated with higher quality of care were associated with expected or lower-than-expected death rates but also with higher complication rates.

This is surprising, notes Jeffrey H. Silber, M.D., Ph.D., of the University of Pennsylvania. Logically, there should be more complications in hospitals with worse outcomes, and those hospitals generally should provide worse quality of care. Dr. Silber explains that this apparent paradox may be due to the fact that hospital deaths are clear-cut events and are reported consistently, whereas complications are not so easy to identify, agree upon, or consistently report.

Dr. Silber and coinvestigators obtained data on 16,673 patients undergoing CABG procedures at 57 hospitals across the Nation and data from the 1991 American Hospital Association Annual Survey. They compared the expected number of deaths, complications, and deaths following complications (failure to rescue) based on patients' clinical conditions with actual complication rates of CABG patients at participating hospitals. They also studied hospital characteristics often used as indicators of quality of care that were associated with each outcome.

Results showed that of all the patients studied, 43 percent had complications, 4 percent died, and 10 percent died as a result of complications. The three hospital characteristics associated with the highest complication rates were the presence of a magnetic resonance imaging facility, an approved residency training program, and a bone marrow transplant (BMT) unit, characteristics usually associated with high quality of care. These findings raise serious questions about the usefulness of complication rates as an indicator of hospital quality of care for patients undergoing CABG surgery, conclude the researchers.

Details are in "Evaluation of the complication rate as a measure of quality of care in coronary artery bypass graft surgery," by Dr. Silber, Paul R. Rosenbaum, Ph.D., J. Sanford Schwartz, M.D., and others, in the Journal of the American Medical Association 274(4), pp. 317-323, 1995.

Greater ICU use does not necessarily translate into fewer hospital deaths

The more technology-intensive nature of American versus Canadian medical care is evident in a recent study which shows that the number of intensive care unit (ICU) days per million people in western Massachusetts is two to three times that of patients in Alberta, Canada. The main reason for the difference is the greater proportion of U.S. hospitalized patients treated in the ICU rather than a difference in hospital admission rates or length of ICU stay. Surprisingly, this greater use of ICUs does not appear to lead to a lower hospital death rate, according to a study supported in part by the Agency for Health Care Policy and Research (HS06026).

Led by Stanley Lemeshow, Ph.D., of the University of Massachusetts, the researchers used hospital discharge data to compare differences in ICU use and hospital death rates among 50,030 patients admitted to hospitals in Alberta and western Massachusetts during 1990 and 1991. ICU use in Massachusetts was significantly higher for all patient groups, except those undergoing coronary artery bypass surgery, where ICU use was similar for Alberta and Massachusetts patients. The hospital death rate in western Massachusetts was similar to, or higher than, that of Alberta.

Evidence indicates that Alberta, which has fewer ICU beds per hospital bed and per capita than western Massachusetts, reserves the ICU for sicker patients. For instance, a much higher proportion of Alberta ICU patients received mechanical ventilation; conversely, a greater proportion of Massachusetts ICU patients undergoing elective surgery were less severely ill than similar Alberta ICU patients. Compared with patients in western Massachusetts, patients in Alberta tended to receive surgery earlier in the day and, if necessary, were kept longer in the recovery room until their condition stabilized and they could be transferred to regular beds on a surgical floor. These same patients in Massachusetts would be more likely to be transferred to the ICU after surgery until their condition stabilized. Also, some categories of patients in Alberta were likely to be treated for extended periods in the emergency department, in contrast to similar Massachusetts patients, who were rapidly transferred from the emergency department to the ICU.

For more information, see " A comparison of intensive care unit utilization in Alberta and western Massachusetts," by John Rapoport, Ph.D., Daniel Teres, M.D., F.C.C.M., Robert Barnett, M.B.B.S., and others, in Critical Care Medicine 23(8), pp. 1336-1346, 1995.

Nearly three-quarters of ICU patients in acute renal failure requiring dialysis die in the hospital

About 70 percent of critically ill patients with acute renal failure (ARF) requiring dialysis die in the hospital. Although dialysis can be lifesaving in some circumstances, it may be unhelpful in others, such as this subgroup of ARF patients, explains Glenn M. Chertow, M.D., of Brigham and Women's Hospital, in this report of a study supported by the Agency for Health Care Policy and Research (HS07118).

He and other researchers developed a model to predict in-hospital death among these critically ill ARF patients, which they derived from evaluating the medical records of 132 ICU patients with ARF who required dialysis. Using patient medical history, clinical condition, and laboratory results, the model identified 24 percent of high-risk patients who died, without misclassifying any survivors. Overall, 70 percent of the patients died during hospitalization, 63 percent of them within 30 days of beginning dialysis.

According to the model, the need for mechanical ventilation, malignancy, and nonrespiratory organ system failure were independently associated with in-hospital death. According to the researchers, this model may be useful for identifying a fraction of those patients who will die regardless of dialysis and for whom a less invasive approach might be appropriate, without denying renal dialysis to patients who might benefit.

Details are in "Prognostic stratification in critically ill patients with acute renal failure requiring dialysis," by Dr. Chertow, Cindy L. Christiansen, Ph.D., Paul D. Cleary, Ph.D., and others, in the Archives of Internal Medicine 155, pp. 1505-1511, 1995.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care