Monday, February 1, 2010

Competency and medical care

A radiologist in Saskatchewan is again in the news over the weekend for what is becoming a tired and frustratingly common reason: a question of past competence (Click here). We have seen this before in other medical specialties, for example the stories involving pathologists in both Newfoundland and Ontario in recent years.

I have a couple of comments about this:

First, there has been a tendency, particular with this case, to misdirect the conversation into one involving foreign trained doctors. I have had the opportunity to work with many foreign medical grads; the vast, vast majority are as well trained, caring and competent as any produced by Canadian medical schools. We shouldn't let closet racism enter the discussion here. We need to be able to deal with competency issues in physicians, and that's what this is. Full stop.

Second, the error rate from this particular radiologist has been reported in the CBC stories to be at about 2%, and some have asked whether it is fair to label someone who is 98% "right" as being incompetent. Just how significant is a 2% error rate anyway?

The real answer of course is that it depends upon both the nature of the error and the patient flow rate. For example, if a radiologist sees 50 radiographs per day (not unreasonable) a 2% error rate means that each day, one patient would be misdiagnosed.

However, not all errors are problematic. The radiologist might deliberately err on the side of caution, for example, and order extra "unnecessary" tests to be certain of the patient's condition. Also, there is no information to tell us what the baseline disagreement rate (the "inter-rater reliability") between radiologists in this population is. Is 2% high? Average? Or low?

On the other hand, if only one in ten of the errors have a risk of seriously harming a patient, then only 10 working days would be required to produce a clinical disaster. That's two patients a month, on average. Clearly, we need to know how often bad things happen, not what the rate of error is, in order to interpret this information.

The CBC article quotes the provincial college as stating that a worrisome number of Dr. Tsatsi's errors had the potential to do serious harm to patients. While that implies a degree of review, it doesn't leave the rest of us with much comfort regarding process. It would be far better (and somewhat braver) for the provincial college to quote relative risk in this case: that a patient reviewed by the doctor under investigation had, say, twice the risk, or four times the risk of harm than a patient whose x-ray was reviewed by another.

Questioning the competency of medical care is never wrong; individual physicians should be prepared to open up and declare their experience and comfort level in the management of any case. Having said that, inadequate reporting of these cases and poor presentation of the facts by the medical establishment just muddies the waters, adds to a sense of vague unease in the population, and ultimately serves no one. Give us the numbers, folks.