Over the past year, I have thought extensively about the concept of quality health care. I have participated in research designed to measure one aspect of quality, i.e., adherence to known quality indicators.

Examples of this version of quality include:

• Foot exam in diabetes mellitus
• Aspirin for patients with known CAD
• Colon cancer screening in patients above age 50
• ACE inhibitor (or ARB) in diabetes patients with proteinuria

To use a philosophical term, the above examples are necessary but not sufficient. Checking the boxes of the quality scorecard is useful, but does not define a high quality physician.

The competency that I worry about the most is clinical judgment. I fear that clinical judgment remains (and will remain) difficult to define and therefore even more difficult to measure.

Some commentors have suggested that competency measures are easy – just measure outcomes. Now this might work in subspecialties where one sees the same diagnosis repeatedly. This certainly would work for a cardiovascular surgeon who only does CABG procedures. This certainly would work for an interventional cardiologist.

But I am not concerned about subspecialists. I worry about specialists – general internists and family physicians. We live in a different world. We see undifferentiated patients, with undifferentiated complaints, and our job is to achieve differentiation. We live in a world of vague or common symptoms. We live in a world of causal comments that we must recognize as clues to follow.

At this risk of seeming self serving I will briefly present 3 patients in whom clinical judgment made a major difference. Now clearly I have picked patients for whom I made the right diagnosis. The point here is not the diagnosis, but rather the clinical judgment involved in getting to the answer – and the answer matters in all three cases.

#1

A 52 woman came to my office for investigation of abdominal pain. Her internist (my resident when I was an intern) asked me for a second opinion.

The patient had had severe (8/10) LUQ pain for 2 months. Extensive investigations including UGI, CT of the abdomen and colonoscopy had shown no abnormalities. Multiple laboratory tests had normal results.

She had a history of Type II Diabetes Mellitus for 15 years. On questioning, she said that she had a similar pain in the RUQ 3 years ago. At that time, testing showed no abnormalities, her doctor blamed depression and treated her with tricyclic antidepressants – with a slow resolution.

On examination, I noted that she had surface pain in a dermatomal pattern over the LUQ. She had no deep pain.

As I considered her problem, I considered the possibility of herpes zoster (shingles) with a rash. Then I wondered if diabetic neuropathy could present like this. I ordered nerve conduction tests and proved a diagnosis that I not previously known – Diabetic truncal radiculoneuropathy. For a good review of diabetic neuropathies – THE DIABETIC NEUROPATHIES: TYPES, DIAGNOSIS AND MANAGEMENT

The point! By carefully considering the patient’s complaint and physical exam – I avoided more expensive imaging studies. I made a correct (albeit rather unusual) diagnosis. This diagnosis led to appropriate treatment and the patient recovered over the next 3 months.

#2

A 57 yo executive secretary came to my office for spells. She described 7 episodes over the past 6 months – one a month until the previous month during which she had 2.

The spells caused great physical and emotional distress. She stated that she would feel ill and her pulse would go up to 150 (she recounted measuring her pulse precisely). The spells lasted approximately 15 minutes, after which she felt drained. She had had spells during the day, at work, at home and in the evening – no obvious inciting events.

After the 3rd episode she saw a cardiologist who ordered an echocardiogram and a stress test – both of which were normal – and told her that the spells were “in her head”. As she recounted the story, I could feel her frustration.

I ordered a loop monitor, which the patient wore for the next 2 months without any further spells (perhaps a therapeutic loop monitor). I was convinced that she had PSVT (paroxysmal supraventricular tachycardia), but I could not prove it.

So I told her that when her next episode occurred to immediately get to the ER and get an EKG. She lived in a very small town and was always within 5 minutes of the ER.

2 months later I received a faxed EKG confirming my suspicions. I referred her to an electrophysiologist who performed a radiofrequency ablation. Two years later she had no recurrences.

The point! The patient gave a clear history – virtually pathognomic for her eventual diagnosis. My job was to persist until we could confirm the diagnosis. However, many such patients have their symptoms dismissed as psychological.

#3

A 64 yo veteran was admitted to my service for chest pain. The housestaff presented his story on rounds, and reported that he was scheduled for a sestamibi stress test later that morning. They reported on his risk factors, and had prescribed all the correct medications.

However, when I went to the bedside, I asked the patient what was bothering him, and he immediately complained of RUQ pain. He related a 2 month history of progressive RUQ abdominal pain. He denied a relation to eating. He admitted to a brief episode of chest pain the previous day, but minimized that pain compared to the RUQ pain. On exam he had a positive Murphy’s sign.

I ordered a RUQ ultrasound for that day, and assumed that he had gallbladder disease causing his abdominal pain (and probably his “chest pain”).

Later that day, radiology called to let me know that he had a liver mass – suggesting that he needed an abdominal CT. We had that test the next day – and they again called to tell me that he had a liver abscess. Now most clinicians joke that the national plant of radiology is the hedge, and yet these radiologists did not equivocate – they called it an abscess.

Liver abscesses are rare in the US. Over the next few days the patient developed a fever and an elevated white count. As this occured around Christmas (several years ago), we struggled to get the abscess drained. Surgery wanted interventional radiology (IR) to drain the abscess, and IR said that they could not drain the abscess because of location. We persisted and eventually got surgery to do an open drainage of a large staph abscess. The patient had a successful recovery.

The point! Rather than focusing on the initial complaint and reason for admission, I listened to the patient and proceeded in a linear fashion towards the correct diagnosis. Making that diagnosis may well have saved that patient’s life.

Now I have told you my successes. Like all physicians I have failures. My clinical judgment is good, but even the best make mistakes, because clinical judgment is difficult.

I submit that clinical judgment – making the correct diagnosis, ordering the correct tests or the correct antibiotics – is the key competency for generalist physicians We must note the clues, sorting through the false alarms to find those complaints which need evaluation.

Yet how does one measure this. Perhaps clinical judgment is like pornography. As Justice Potter Stewart once said “I shall not today attempt further to define the kinds of material [pornography] . . . but I know it when I see it.” And yet this seems unsatisfactory.

I do not believe that we assess clincial judgment well, because it requires so many skills and situations to truly demonstrated. Clinical judgment is nuanced. The best physicians note subtle details or inconsistencies to reach a good judgment. I believe this is a difficult process to evaluate.

And yet, I reassert that it is the most important process for physicians. As physicians we must make judgments daily, hourly and multiple times with each patient. These judgments matter, and thus we should emphasize clinical judgment in our training and our assessment.

As the quality movement grows, I fear that it could minimize our focus on judgment. If that is true, then we may harm health care rather than improve health care. Our quest for quality should not ignore other concerns in health care.

HL Mencken (although perhaps a despicable human being) has that wonderful quote –

For every complex problem, there is a solution that is simple, neat, and wrong.

Perhaps this applies to the quality movement.