The danger of grading doctors…

It goes without saying that caregivers are not interchangeable. Quality matters. What else is there other than our health?

From the day I began as a doctor, the absence of a legitimate meritocracy has been a source of inflammation. In 1996, when I started private practice, referrals depended too much on old-boy networks. In 2014, the situation is worse. Now, referrals depend almost exclusively on who employs whom. I could be a wizard of catheter ablation, but referring doctors who are employed by competing systems will not send me patients. They might sneak their mother in, but their patients go to the electrophysiologist employed by their employer.

Speaking frankly, this stinks. Always has. Always will. A ‘good’ doctor should be rewarded with referrals and compensation. I wish it could be this way, but I am doubtful that it can.

In many ways judging doctors is like judging teachers. It’s really hard to do from the outside.

Here in the hospital and medical community, you come to know the good doctors. You see their work and then you see their patients. From the inside, thoughtfulness is easy to discern from shooting from the hip. From the outside, though, judging doctors entails looking at many competing measures. For instance, good bedside manner does not equate to technical skills–and vice versa. Box-checking and guideline adherence–especially the latter–are far from useful quality measures. Even procedural outcomes are tough to judge, as the complexity of people are hard to codify before procedures. And the list goes on.

Another simplistic way to judge caregiver quality is testing. Good doctors should pass tests and maintain certification. On the surface this makes sense.

Digging deeper into the merits of board certification, however, reveals that it is not so simple. Dr. Wes Fisher has led the way in exposing the challenges of testing doctors. His most recent take on the current-day expansion of medical certification is worth a read.

I’m conflicted about testing doctors.

On the one hand, there’s potential to do some good. Let’s call this my pro-testing side:

It starts with the fact that doctors share great responsibility for our current crisis of overtreatment and inefficiency. Where do you think all those deaths from medical errors come from? It’s not just faulty implementation of treatment. The root cause of many medical errors is that (good-intentioned) therapy was initiated unnecessarily. And it’s not fair to say doctors intervene too much because of the fee-for-service system. That’s just a partial answer. The main reason doctors intervene so much is intervention bias. We have been taught, and have come to believe–to the point of hard-wiring neural connections–that action is better than inaction.

Alas, the problem is that modern medical interventions are not like wearing a seat belt. Whenever we order a scan or blood test, or prescribe a medicine, or a procedure, there are real and quantifiable dangers that act to mitigate benefit. That shadow on a CT scan can be a real problem.

It’s possible that more intense medical education might lesson the do-something-doctor bias. Medical education, if done well, would force us to collectively face the actual science behind our beliefs. Such a clear-eyed view of absolute benefits and harms would go a long way to stemming intervention bias. Lessons 1-5 could promote the number needed to treat (NNT) concept. What I want medical education to do is foster the give-peace-a-chance philosophy.

Lest you think this argument is hyperbole, I’ll use a looming a public health disaster to make my point. One of the most common and devastating medical errors is healthcare-related bacterial infection, due in large part to antibiotic resistance. This is on us. It’s as we say, iatrogenic. We got lazy; we ignored the science; we let our intervention bias run amok. We gave Z-packs to patients who “went to the doctor to get something” for their viral illness.

Make no mistake, this sort of stuff happens because of good intentions. Doctors are good people. We aim to do the best we can. We take action so that we can help people. The problem is that we have come to over-estimate benefits while under-estimating harms. Too often, we view it as a knock on us when a patient’s disease progresses. In these cases, which are the majority, as most people are not blessed with sudden painless death, the ‘quality doctor’ would offer care rather than intervention. They are not the same thing.

Besides teaching us how to interpret science, another thing medical education could do is help us recognize the limits of human longevity. One of the greatest mistakes doctors make–me included–is treating death as optional. Oh my, that one is baddie.

That’s my pro-testing side.

My contra-testing side is stronger. Mostly because it stems from realism.

The problem with board certification in its current state is that it too often teaches mainstream dogma. It perpetuates the ‘healthcare’ machine. It leans to belief rather than skepticism.

If we have learned one thing in 2013, it’s that expert guidelines, often written by authors with financial ties to industry, are flawed. Take the cholesterol guidelines. The new recommendations don’t come from any new data. We had known this data for years, but yet, until late 2013, the go-to document recommended treating to surrogate markers. Does that mean those of us who practiced risk-based statin therapy were wrong for years, but now we are right? How much hubris does it take to hold out adherence to such documents as measures of quality? And the AF treatment guidelines: Dronedarone, really?

The most concrete example of this point comes from a sample question from the American College of Cardiology. The question concerned an elderly patient with valvular heart disease. Possible answers included surgery, valve squishing procedures and medicines. Nowhere was the right answer–the choice to discuss all options with the patient, including therapy directed only at controlling symptoms.

In his enthusiastic post, Dr. Fisher rightly emphasizes the compete lack of evidence that board certification (in its current form) improves patient outcomes. I believe it’s worth studying; not because we need to confirm benefit, but rather, so that we can exclude harm.

You see how testing is working out for our education system.

JMM

3 comments

  1. Well put, John. Every day, doctors are left to fend for themselves when attempting to keep up with innovations in medicine. We see a patient, don’t know the answer, then turn to the skill that got us where we are today: study. We find the answer or a possible solution to the vexing problem. Or we use our judgment and creativity when an answer doesn’t exist. Sometimes we find answers of our own and share them with the world or our colleagues.

    To suggest that “ivory tower” thinkers can test such a skill is fool’s play. It takes actually doing medicine to learn how to do medicine, for it involves SO much more than regurgitating information or performing pre-canned practice improvement modules.

    Medicine involves caring for real, live people and doing it often.

    The really scary thing is, the way things are constructed now (that enrich the test-makers), data-entry personnel and professional test-takers will be the ones caring for you and me in the future instead of actual doctors.

  2. Can we step back and look at this from a wider angle? How does a patient find a ‘good’ doctor? There are some bad apples out there. How can a patient know the person they are seeing is someone they can trust?

    Answer this, and you’ll understand the drive to measure stuff (even meaningless stuff) and put more certifications (even meaningless certifications) after each name.

    In the ‘good ‘ol days’ you knew your primary care doc. Everyone in your social circle did, and you could trust their recommendation because they cared about their reputation. Your doc was your doc, not the one who happened to have an opening that day.

    Those days are gone. Practices are larger, providers are interchangeable, and a patient with a complicated problem might see 4-5 different docs in a 24 hr period. The patient doesn’t know anything bout any of them (even their names – it’s harder than you think when you mind is flooded with 100 details about a problem you’ve never encountered before).

    Health care administrators want to build patient confidence, but not in individual doctors. Administrators want to build confidence in the their brand. I don’t get marketing material on individual physicians, but I do get 2-3 glossy magazines a month (unsolicited, and we seldom use non-routine care, but we live in a zip code that pays at attractive rates) from one of the local health care mega-brands. Look at our cancer center, did you know about our new piece of equipment that no one else has?, have you seen our new expansion project?

    For someone who is casually familiar with marketing and branding, it’s clearly not reliable signaling (is valet parking a proxy of physician competency?) and borderline sickening.

    I can tell you this, though. The branding works. When I run into someone who has recently had major medical problems, I ask about their experience. They NEVER name a particular doctor. They name a mega-brand. I went to XZY Memorial. I went to QQQ Med. Others nod in agreement and say “My cousin went there. They’re great!” The branding works.

    How does this tie into certification? In addition to marketing to patients, those admins are selling their brands to large patient pools. You’ve heard how Walmart has agreed to send insured employees to the Cleveland Clinic for heart procedures. That’s a glimpse into the future – just the beginning. The Cleveland Clinic didn’t go to Walmart and say ‘we have Dr. X, he’s been doing this for years. He’s they guy I’d send my own mothers to.’ They said we have X number of people certified in this, that, and the other. Collectively we perform Y procedures a year with a complication rate of Z. Our docs are among the most certified of any mega brand out there.

    It’s marketing. Since patients can no longer trust their Doctor (see what Dr. John said about referrals), we’re being asked to trust the brand. Benefits administrators (insurance people) are contracting with brands before the patients even know they will need care. The brands boost their sales pitch with measurements and certifications.

    The trajectory is clear from 30,000 ft. It’s probably not clear down in the trenches. Honestly, it’s puzzling to me how so many smart people can miss it (ie. Doctors).

    Maybe the death of small practices was inevitable, but the trend of selling small practices to mega-groups has started the ball rolling downhill. Once you sell, you’re no longer the brand. Patients don’t come to see YOU, they come because of the brand (or their insurance company contracted with the brand long ago). You’re an interchangeable part and you will be treated (and micromanaged, measured, and marketed) as such.

    If you still doubt what I’m saying, try to buy an individual insurance policy on the new federal marketplace (if the website works today). You will see policies from the traditional big name insurance players (BCBS, Aetna, etc..). You should also see policies that ONLY cover services within one of the hospital mega-brands (if you’re in a populated county. If you don’t see one, tell the website you live in a major city). The hospital brand policies are usually cheaper. Guess which kind of policy will become more popular with benefit administrators going forward?

    Back up a bit and we’ll tie it all together – how to you build and market a mega hospital brand? You market measurable outcomes, staff certifications, and cost efficiencies. What if it has a negative impact on doctors or patient care? We don’t sell doctors and we don’t sell patient care. We now sell the brand.

    Interesting times…..

Comments are closed.