“I am in blood stepp’d in so far that, should I wade no more, Returning were as tedious as go o’er.”

– Macbeth, Act 3, Scene 4

Transfused blood would be the right ink with which to write the story of modern medical care, because that blood is the perfect metaphor for what ails that medical care. Transfused blood can indeed be, on the one hand, “the gift of life.” On the other hand, we now know transfusions can cause much more harm to some patients than we thought in the past, and have good evidence that the reasons we used to transfuse many patients are no longer valid. Despite all that we still transfuse blood unnecessarily into hundreds of thousands of Americans each year.

That we continue an unnecessary practice that costs perhaps $1 billion a year, may harm thousands of patients and causes shortages of blood for those who really must have transfusions, tells us a lot about where we are in American health care and where we need to go.

Our story begins several years ago when evidence began to emerge that transfusing some patients put them at higher risk of developing infections and complications. Other studies then discovered that many of the kinds of patients we transfused when their blood counts were somewhat low turned out to do just as well or even better if we did not transfuse them unless their blood counts got really low.

As a result of these developments, transfusion and other medical experts have been telling physicians we should stop transfusing so many patients. In the past couple of years, clear new standards for which patients really need blood transfusions and which do not have emerged and been widely disseminated to physicians and hospitals. The new standard: In general, unless a patient has complicating factors such as active bleeding or a few other specific problems, he or she does not need a transfusion unless their hemoglobin is less than 7mg/dl (remember that number).

So, class, here are the questions for today’s pop quiz:

1. Why would any patient who does not meet the new criteria for a transfusion get a transfusion?

2. Why would a physician order blood transfused into such a patient?

3. Why would a hospital allow blood to be transfused into such a patient?

4. Why would an insurance company pay for blood to be transfused into such a patient?

5. Why wouldn’t all the hospitals in a state get together and agree we are not going to transfuse blood into such patients?

The current model for changing practices in health care is primarily one of diffusion; we put new medical knowledge into the health care system and wait for it to diffuse slowly through the practices of individual physicians. If the system works right, every physician finally comes around to the new way of practice. On the downside, that process takes about 10 years to change practice, and even after 10 years some physicians will still do things the old way. On the upside, before every physician is doing things the new way, the slow process sometimes spits out those new ideas we thought were great but turned out to be wrong.

The problem is that no accelerated process exists for rapid and systematic implementation of new practice when substantial harm to patients and cost to the system results from slow transition to the new way. Hospitals and physicians rarely act collectively and systematically on patient care issues; we don’t want to tell physicians what to do even when we know what they should do and are not doing; physicians don’t want to be told what to do and practice cookbook medicine, etc. Physician discretion often works to protect patients, so we are reluctant to tinker with it even when that discretion may not be a good idea. Those are basically the answers to the quiz.

Well, I say to heck with that model in this and some other cases. The diffusion model is no longer affordable as the only model for practice change in modern health care, and it should no longer be acceptable as the only model when important issues of patient care are at stake. We ought to have a way to identify some changes in medical practice that are so important that they are systematically and broadly implemented much more quickly. Blood transfusions are ripe for demonstrating the potential of that new approach. By one estimate, such an approach in all American intensive care units would prevent transfusion of more than 1 million units of blood and save almost $1 billion. It would also prevent complications and some unnecessary deaths.

It should not take more than a couple of days in every state for a convention of blood transfusion experts and other physicians, hospital executives, health insurance company executives, and a few bureaucrats to agree on a blood transfusion protocol that would be implemented immediately. That protocol would basically say that if, on careful review, the patient does not meet the new criteria for transfusion, they do not get transfused. Period. If they do get transfused despite not meeting the criteria, their insurance company should not pay for that transfusion. Period.

You might think that kind of rapidly implemented consensus change in medical practice is impossible, but I disagree, and here is why: If you put together a convention of those same physicians and hospital executives who know the new criteria for transfusions and the reasons for those criteria, not one would allow their mothers or themselves to be transfused unless they met the new criteria. If health care professionals can do it for themselves and their families, they should be able to do it for you and yours.

Erik Steele, D.O., a physician in Bangor, is chief medical officer of Eastern Maine Healthcare Systems and is on the staff of several hospital emergency rooms in the region.