Blood Transfusion Administration in Seriously Ill Patients

An Evaluation of SUPPORT Data

Published in: Journal of the American Geriatrics Society, v. 48, no. 5, suppl., May 2000, p. S39-S43

Posted on RAND.org on December 31, 1999

by Marie L. Borum, Joanne Lynn, Zhenshao Zhong

BACKGROUND: Administration of blood transfusion in seriously ill patients is highly variable. Limited data are available to guide transfusion decisions. OBJECTIVE: To explore characteristics of patients who received blood transfusions and decisions to forgo transfusions in the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). DESIGN: Prospective study of preferences, decision-making, and outcomes. SETTING: Five teaching hospitals. PARTICIPANTS: 9105 patients aged 18 years and older meeting defined diagnostic and illness severity criteria. MEASUREMENT: Data included blood transfusions, demographic characteristics, diagnoses, comorbid conditions, acute physiology score (APS), nutritional support, and functional status before hospitalization. RESULTS: A total of 2863 patients (31.4%) received blood transfusions, usually early in their hospitalization. Transfused patients were more likely male (57.3%; P = .008), with a younger mean age (56 vs 64 years), significantly higher APS (P < .001), and significantly lower 2- and 6-month survival predictions (P < .001). The patients with acute respiratory failure or multiorgan system failure with sepsis (1714; 59.9% of all patients receiving transfusions), multiorgan system failure with malignancy (480, 16.8%), and cirrhosis (248, 8.7%) were more likely to receive blood than those with other diseases. Few patients made a decision not to receive blood before (5; 0.05%) or after (126; 1.4%) study entry. Most patients with decisions to forgo transfusions also had decisions against trying resuscitation. CONCLUSIONS: Most transfusions occurred in patients with acute respiratory or multiorgan system failure. Few patients decide to forgo transfusions. Additional investigation is necessary to evaluate blood transfusion practices in seriously ill patients.

This report is part of the RAND Corporation external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.