In this issue of American Family Physician, Howard and Ellerbeck¹ discuss the underutilization of therapy with beta blockers after acute myocardial infarction (MI). That beta blockers are useful after acute MI is certainly not a new concept in medicine; the evidence that beta blockers significantly reduce mortality after acute MI was produced in the Beta-Blocker Heart Attack Trial² (BHAT) in the early 1980s.

If the evidence has been around for nearly 20 years, why are so many practitioners not using beta-blocker therapy in patients who have had an MI? This is not an easy question to answer. As physicians, we are inundated with new literature, new medications and new technology, and this information overload can be quite overwhelming. However, it has been shown repeatedly that beta blockers decrease mortality after acute MI. How many ways can it be said?

How many times do we have to hear the evidence before we change our practice habits? Therapeutic trials are performed with the idea that favorable outcomes will cause health practitioners to change their practice habits. This is an easy concept: if something helps, we should do it, and if something does not help, we should not do it. The concept does not appear to be that easy in practice, however. If matters were clear-cut, for the past 20 years eligible patients would have received beta blockers after acute MI, and the evidence does not show that this is the case.³

What will it take to get us to change our ways? Studies have evaluated which interventions are the most helpful in trying to get physicians to put new evidence into practice. Obviously, though, we all learn in different ways.

One study⁴ attempted to categorize interventions according to their effectiveness in promoting behavior changes in physicians. Certain interventions were consistently effective in promoting change. These included educational outreach visits, computerized or manual reminders, interactive educational workshops and multifaceted interventions (combining two or more of the following: reminders, local consensus processes, audit and feedback reminders and marketing). The interventions with variable effectiveness included use of local opinion leaders and patient-mediated interventions. Audit and feedback and local consensus processes alone also showed variable effectiveness. The interventions that were shown to have little or no effect included educational materials and didactic educational meetings.

So where does that leave us? Many physicians receive their new information in just those ways that were shown to be less helpful: educational materials (such as clinical practice guidelines, audiovisual materials and electronic publications) and didactic educational meetings (such as lectures).⁵ I see what the studies show; however, I know that for me, repetition is the key to learning, and perhaps others learn this way too. So, for what it is worth, let me stress yet again: beta blockers significantly reduce mortality after acute myocardial infarction. Let's put this evidence into practice.