CMOs: Evidence-Based Medicine's Best Advocates

Many physicians have resisted evidence-based medicine guidelines over the years—sometimes for good reason. Some still refer to evidence-based medicine by the epithet "cookbook medicine," saying that because each patient is different, there's no way to prospectively determine what particular interventions will work best for a particular patient with a particular malady or group of maladies.

Hospital administration, in the past, has not been much help. Halfhearted attempts to force physicians to adopt evidence-based medicine practices have been hampered by the fact that reimbursement has rarely been at stake.

Furthermore, attempts to push through such changes have often come from administrators. Doctors don't tell them how to do their jobs (don't they?) so an administrator shouldn't tell him how to do his.

So it wasn't necessarily the message but the messenger.  I'm finding that physician resistance to implementing such guidelines is fading in the face of

• Better and more thorough clinical research
• Electronic medical records that incorporate that research in the form of reminders and accessible research
• A younger cadre of physicians who are no longer resistant to computer or other technological assistance in diagnosis
• A reordering of the responsibilities and accountabilities of the chief medical officer

That last factor is a particularly important catalyst to change.

I got to thinking about evidence-based medicine upon reading news of some important research from Johns Hopkins Hospital. Reported by my colleague Cheryl Clark earlier this week, the research surrounding blood transfusions in surgery could provide an interesting test case to determine how difficult it remains to incorporate evidence-based medicine protocols and change current surgical practices.

According to the story, the evidence boils down to this:  

Current research says transfusions for most surgeries should not be initiated until the patient's hemoglobin level—normally 12 to 14—has dropped to 6 or 7 grams per deciliter (g/dl).  A level of 7 or 8 is considered safe. But the recommendations of three specialty societies that guide current practice leave the trigger point in question. "They say that if a patient's hemoglobin level is less than 7 g/dl, then the patient would benefit from a blood transfusion. But if it's greater than 10, they would not benefit. But they don't say what should be done if the level is between 7 and 10," Steven M. Frank, MD, leader of the study, said during an interview. Thus, many surgeons initiate transfusion when levels are at 10, while others start at 9 or 10 or 11.  Additionally, surgeons vary in the target point at which they stop transfusing. Some stopping at 11 or 12, even though they could stop at 10.

Essentially, the researchers found that by lowering hemoglobin threshold to 8 (instead of) 10, they used 66% fewer blood transfusions and they had no different outcomes between the two groups in length of stay, heart attack, stroke, death, and even the ability to walk.

The logical conclusion: If there's no benefit from giving extra blood, then all that's left is the additional cost and the potential for harm by giving more blood than is needed.

Certainly the research must be evaluated, but this is something that seems to make so much sense that it should change practices rather quickly, at least for physicians who are made aware of the news.

I'm not writing to pillory physicians and accuse the entire group of being unwilling to change. The approach to changing clinical practices has been uneven and I understand resistance to hearing from a non-physician that a physician should change the way he or she practices medicine given the physician's investment in a long academic career and apprenticeship (through residency).

Leadership means delegating this responsibility to the physicians themselves, and it's why a strong CMO role is essential.

You can bet that as more research like this is published, showing that doing more of something provides no additional benefit, insurers and the Centers for Medicare & Medicaid Services will likely be watching closely, evaluating, and figuring out ways to incent hospitals and surgeons to adopt practice guidelines incorporating such findings—barring any extenuating circumstances.

In some cases, the penalties for noncompliance might be financial. But long before financial penalties for deviation from evidence-based medicine protocols, some hospitals and physician groups have been phenomenally successful in adopting them.

In fact, that's how many hospitals have successfully integrated evidence-based medicine protocols—by finding ways to have the physicians police themselves. And that is the challenge of leadership.

Just because new research is published and vetted is not sufficient to engineer change. Change can only come from a respected CMO who has the confidence of leaders on the surgical team. If he or she can convince them through research that certain practices are better for the patient and that the physicians will be evaluated on how closely they adhere to them, they'll change.

Peer pressure works.