Accepting Critically Ill Transfer Patients

Summaries for Patients are a service provided by Annals to help patients better understand the complicated and often mystifying language of modern medicine.

Summaries for Patients are presented for informational purposes only. These summaries are not a substitute for advice from your own medical provider. If you have questions about this material, or need medical advice about your own health or situation, please contact your physician. The summaries may be reproduced for not-for-profit educational purposes only. Any other uses must be approved by the American College of Physicians.

The summary below is from the full report titled "Accepting Critically Ill Transfer Patients: Adverse Effect on a Referral Center's Outcome and Benchmark Measures." It is in the 3 June 2003 issue of Annals of Internal Medicine (volume 138, pages 882-890). The authors are A.L. Rosenberg, T.P. Hofer, C. Strachan, C.M. Watts, and R.A. Hayward.

What is the problem and what is known about it so far?

To make wise choices about their health care, people often want to know about the quality and efficiency of the care being dispensed by individual physicians, groups of physicians, hospitals, and health systems. Various organizations, like health insurance companies, and regulatory agencies, like the Centers for Medicare & Medicaid Services (formerly the Health Care Financing Administration), regularly issue "report cards" on the quality of health care and the performance of those who provide it. These reports typically use a method called benchmarking, an approach that measures performance on standard quality indicators like patient outcomes (such as mortality rates), processes (such as tests administered), or costs of care (such as charges for surgical procedures). This information is then compared with a standard.

Why did the researchers do this particular study?

To find out if benchmarking unfairly penalized medical centers that accept critically ill patients transferred from other medical centers. Currently, benchmarking efforts frequently do not account for this factor, even though transfer patients often stay longer in the hospital and die at a higher rate than other patients.

Who was studied?

4208 patients admitted to the medical intensive care unit at a university hospital between January 1, 1994, and April 1, 1998. These included patients with diagnoses such as chronic obstructive pulmonary disease, acute respiratory distress syndrome, respiratory failure, pneumonia, gastrointestinal bleeding, sepsis, stroke, liver failure, and heart failure.

How was the study done?

The researchers collected information on these patients from an electronic clinical database. These data included length of stay in the intensive care unit; death; and admission source, such as direct admission from the emergency department, admission from a non–intensive care nursing floor of the hospital, or admission from another hospital. The researchers analyzed the data to determine whether admission source made a difference in quality-of-care factors, such as length of stay and death. They then evaluated the possible difference between a hospital that receives about a quarter of its medical intensive care unit patients as transfers and one that receives no transfer patients.

What did the researchers find?

Transferred patients were likely to be sicker than nontransferred patients. They were often admitted with conditions that complicated their care—such as severe infections and upper gastrointestinal bleeding—and increased the length of hospital stay and risk for death. Thus, two hospitals with the same efficiency and quality but a different percentage of transfer patients would perform differently in a benchmarking program that did not account for admission source. The hospital with more transfers would have a much higher mortality rate.

What were the limitations of the study?

The results may not be generalizable because the data were derived from only one hospital. Also, the results from this strictly medical intensive care unit may not be generalizable to other types of intensive care units, such as general surgical units. In addition, the researchers did not have information on the reason that the patients were transferred.

What are the implications of the study?

Because current benchmarking programs do not account for admission source, hospitals that accept transfer patients may seem to be less efficient and have lower quality than hospitals that do not. This may deter hospitals from taking patients who may benefit from transfer. The authors suggest that benchmarking and profiling efforts recognize and account for this phenomenon.