I had a recent conversation with an old friend about her elderly father that encapsulates a lot of what is both great and terribly wrong with health care in America today.

Here are the basic facts: the man is in his mid-80s, retired from teaching school, and is active and vigorous, living in the community; he is cognitively intact. He has a history of coronary disease and had an intracoronary stent placed some years back. He is asymptomatic on a typical “cocktail” of meds including aspirin, a statin, and an ACE inhibitor. Over the summer, he had a routine follow-up visit with his cardiologist, who detected a carotid bruit. After a duplex sonogram and a CT angiogram, a high-grade unilateral internal carotid stenosis was identified, and carotid endarterectomy surgery was recommended. My friend called me to see if I could recommend a surgeon in the city where she and her father both live.

It will come as no surprise to those who know me that I asked about the specific advice her father had been given and the process by which he apparently decided to go ahead with the surgery. Both the cardiologist and the vascular surgeon to whom he was referred pitched surgery as pretty much a no-brainer.  Although his daughter had questions and reservations, the patient did not really see this as a “decision” to be made. His doctors had recommended a course of action and he was going to do as they said. When I pressed for details about what had been presented as risks and benefits, here is what I learned they told him: Surgery would require a 1 to 2 day hospital stay; risk of major complication was about 1 percent; risk of stroke without surgery was 15 percent over five years. Seemed like they were implying that his chances were 15 times better with surgery than without — who wouldn’t go for that?

OK, so here’s what I think is good about this. Assuming that he had appropriate indications for his prior stent (big “if”), then this man has clearly benefited from modern cardiovascular care. He is asymptomatic, and his medical therapy has likely added years of event-free survival. He is, in many ways, a counter-example to the prolonged disability anticipated by Zeke Emanuel, which I wrote about recently.  I certainly would not want to “check out” at 75 if I could be active, cognitively intact and symptom-free at 85.

So what’s so bad?

First, there was no attempt made to engage the patient in shared decision-making. Even if the evidence strongly favors a particular course of action, it is a truism that the outcomes of interest to the patient are not necessarily the outcomes that interest the physician, and neither set necessarily includes the outcomes for which reliable evidence exists. For example, if a patient is thinking “avoidance of disability” while the doctor is thinking “survival” and the evidence is unreliable with regards to either, then the conversation can’t possibly go well. Put another way, to say that one treatment is “better” than another without agreeing what better means to the patient is to head irretrievably down the wrong path.

Second, there is the citation of evidence that, on the face of it, can’t possibly be accurate. Even if the surgeon were accurately quoting his own track record for this type of surgery, it is not credible that he has sufficient experience in 85-year-old men with coronary artery disease (and whatever particular set of other co-morbidities my friend’s father has) to formulate an accurate assessment of risk for this patient. Likewise, the quoted risk of stroke in the absence of surgery is likely an extrapolation from a mix of observational and interventional studies in populations that probably included few if any 85 year olds. Of course, physicians must make recommendations in the absence of perfect data all the time. The point is not that this uncertainty is avoidable; it is that the uncertainty should be appreciated by physicians and disclosed to patients.

The third problem is one that I have also written about:  the innumeracy that is common among patients and physicians. Even if the numbers quoted were right (that is, they were the best point estimates of the risk and benefit for a population of similar patients facing a similar decision), the framing makes a big difference.

Would he have agreed to surgery if the same data had been presented in a different way? What if he had been told that he had an 85 percent chance of making it to 90 years old without a stroke? Or if he were told that the risk of stroke was 3 percent per year? How about if he were told that he had a greater than 80 percent chance of making it to the end of his predicted lifespan 91 for a man his age without surgery or a stroke?

Finally, the ugly. I can’t help but wonder how much of this whole cascade — from the routine and likely unnecessary follow-up visit with the cardiologist, to the multi-modality imaging, to the inevitable recommendation for invasive intervention — was driven by the unholy alignment of fee-for-service reimbursement with the well cultivated (and totally wrong) belief among our patients that more care is always better, and facilitated by Medicare shielding the patient from the cost of care.

The good, the bad and the ugly of American medicine. What do you think?

Ira Nash is a cardiologist who blogs at Auscultation.