Circulation 36,825 • Volume 17, No. 4 • Winter 2002

AHRQ Responds: Promotes Beta-Blockade, Encourages Further Study

Kaveh G. Shojania, MD; Robert M. Wachter, MD; Kathryn M. McDonald, MM

To the Editor

The AHRQ Patient Safety Practices report, developed by the UCSF-Stanford Evidence-based Practice Center, found that the perioperative beta-blockade in appropriate patients was strongly supported by published evidence demonstrating reduced perioperative morbidity and mortality.1 Dr. Royster’s lead article in the Summer 2002 APSF Newsletter summarized this evidence and supported our report’s conclusion. In response to Dr. Royster’s prediction that this practice may evolve into a standard of care, Dr. Kleinman argues that, while existing evidence supports use of perioperative beta-blockade, sufficient questions remain to warrant a larger trial.

Perioperative beta-blockade ranked in the top category of the 70-plus patient safety practices reviewed in the AHRQ Report because evidence indicating that this practice reduces adverse events from medical care (our operational definition of patient safety practices) is much better than that of most of the other candidate safety practices reviewed. Thus, this high ranking results just as much from the paucity of evidence supporting other practices as it does from the strong, but not ironclad, evidence supporting perioperative beta blockade.

Many of Dr. Kleinman’s concerns regarding strength of evidence apply equally well to other prominent patient safety practices. For example, the major clinical trial to examine the impact of computerized order entry showed only a non-significant reduction in adverse drug events.2 Thus, the mandate that hospitals across the country adopt this multi-million dollar technology3 rests largely on one study showing an impact on a surrogate safety outcome—medications errors, rather than on compelling evidence about the clinical outcome of interest—adverse events. Similarly, fatigue’s impact on clinical performance has been on errors and other surrogate endpoints in simulated scenarios.4 Nonetheless, work hour restrictions are being vigorously pursued5 despite little hard evidence in favor of their benefit and some evidence for possible harm (e.g., due to increased handoffs).6

In the end, we agree with Dr. Kleinman that the evidence supporting perioperative beta-blockers is strong, but not as firm as one would like for a clinical practice potentially affecting millions of patients. As with all probabilistic decisions, the conclusion one draws on the basis of existing evidence depends on whether we would rather risk adopting an ineffective (or even harmful) safety practice or risk missing the opportunity to adopt an effective one. In this case, the editors of the AHRQ report and the investigators who reviewed this specific practice7 interpreted the available evidence as favoring use of perioperative beta-blockade. We did not intend to imply that this or any other practice reviewed in the report yet has sufficient supporting evidence to warrant the label standard of care. In fact, we highlighted perioperative beta-blockade as a practice for which further research would be highly beneficial.1 The balancing act facing all of us—as clinicians and advocates of patient safety—is how to promote practices that will improve safety without demanding overly stringent standards of evidence. In this case, in our judgment a reasonable compromise would be to promote use of perioperative beta-blockade (without necessarily labeling it as standard of care) but to follow-up such promotion with well-designed outcomes analyses.

Kaveh G. Shojania, MD
Robert M. Wachter, MD
Department of Medicine, University of California San Francisco

Kathryn M. McDonald, MM
Center for Primary Care and Outcomes Research, Stanford University


  1. Shojania KG, Duncan BW, McDonald KM, Wachter RM, eds. Making health care safer: a critical analysis of patient safety practices. Evidence Report/Technology Assessment No. 43 from the Agency for Healthcare Research and Quality: AHRQ Publication No. 01-E058; 2001.
  2. Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA. 1998;280:1311-1316.
  3. Milstein A, Galvin RS, Delbanco SF, Salber P, Buck Jr CR. Improving the safety of health care: the leapfrog initiative. Eff Clin Pract. 2000;3:313-316.
  4. Gaba DM, Howard SK. Fatigue among clinicians and the safety of patients. N Engl J Med. 2002;347:1249-1255.
  5. Weinstein DF. Duty hours for resident physiciansÑtough choices for teaching hospitals. N Engl J Med. 2002;347:1275-1278.
  6. Shojania KG, Duncan BW, McDonald KM, Wachter RM. Safe but sound: patient safety meets evidence-based medicine. JAMA. 2002;288:508-513.
  7. Auerbach AD, Goldman L. beta-Blockers and reduction of cardiac events in noncardiac surgery: scientific review. JAMA. 2002;287:1435-1444.