Most physicians are incompetent.

There; I said it.

If you read howardisms on a regular basis, you might have already suspected this. Lots of evidence indicates that substandard or inappropriate care occurs in most patient-physician encounters (up to 80% of encounters). The wrong diagnoses are made far too often. Suboptimal or incorrect therapies are too commonly employed. Many (if not most) surgeries performed across specialties are unnecessary. Far too much invasive and noninvasive testing is performed. Non-factual advice is doled out by the ton.

Before you get too preachy, let me give an example of what I am talking about.

I saw a patient recently whose details I will change slightly. She is typical of patients that I see almost daily. She presented as an ED follow-up re abdominal pain. She had complained to her gynecologist of similar abdominal pain a few years ago. Because she also once had an abnormal pap smear and she was done having children, then, naturally, the other gynecologist performed a vaginal hysterectomy. This predictably did not improve the pain. She bounced around from one doctor to the next, often seeing nurse practitioners or physician assistants who persisted with the previous (and incorrect) diagnoses.

Her pain tends to come and go. Recently, the pain became so severe that she presented to the ED. While there she had a “rainbow panel” of mostly irrelevant labs (CBC, CMP, UA, pregnancy test, etc. – yes, a pregnancy test even though she had had a hysterectomy). When these unnecessary labs were all normal, she, of course, had a CT scan. When this was also normal, she had a pelvic ultrasound (after all, she is a woman). Finally, the US revealed a 1.5 centimeter physiologic cyst and she was given Norco and told to follow-up with a gynecologist (me!).

I must admit, when I interview a new patient, I start from the very beginning and assume that everyone who saw her before me was wrong. You should try it; it’s a great de-biasing strategy. The sad thing is though, most of the time it seems to be true. The intermittent pain she complained of for years was a crampy, spasmodic pain mostly behind her umbilicus but it moved around quite a bit. It was accompanied by constipation, with only 1-2 bowel movements per week. There were some dietary triggers. She had been under more stress recently. The Norco, like other narcotics prescribed before, tended to make the pain worse. She had nausea with the pain as well.

I treated her for constipation-dominant irritable bowel syndrome. I ordered no tests. Am I some great diagnostician? No. I took a history and listened to the patient. I did not simply carry on with the previous (incorrect) diagnoses she had been given. Why had she had an unnecessary surgery and thousands of dollars of other tests and treatments? The answer is complicated I think, but we must consider basic competency.

What competencies were necessary to treat this patient? We could break it up in a variety of ways, but how about this for starters:

• Patient interviewing/interrogation
• Construction of a differential diagnosis
• Choice of appropriate therapeutics

I know these sound like the most basic of skills for physicians, but obviously one or more of them was not employed by numerous previous providers who saw this patient; and this story is repeated every day among almost every patient I see. These are basic competencies. But physicians and other providers are not being trained in a competency-based environment and that must change in order for these bad outcomes to change.

It’s difficult to design and implement a competency-based curriculum and most educators only give lip-service to the idea. Currently, our educational system emphasizes performance on standardized tests. This means that we select a certain number out of a group and reject a certain number out of a group. That group can be fantastic or horrible – It doesn’t matter; we still will select the top performers of the group.

If you are in the top x% of MCAT takers, you can probably get into medical school. If you’re in the top 85-90% of medical school matriculants, you can pass the licensing exams. If you are in the top 98% of those students, you can enter a residency program, and if you are in the top 90+% of residents, you can become board-certified in your specialty. This means we are selecting board-certified physicians based on their position on a bell-shaped curve, not their competency. What’s more, a physician doesn’t have to be competent at every task, just most of them. You only need to be able to answer, say, 70% of the questions and you are good. But when competency is lacking in even one domain, the whole system breaks down. You can’t just be good at taking at patient interrogation and choosing therapeutics if you are not good at making a differential diagnosis. Competency requires being good at each step so that failure down the line doesn’t occur.

You can’t reliably judge competency by a statistical distribution.

We must assess the knowledge, skills, and attitudes (KSA) of a person to evaluate competency. KSA is a framework for both teaching and assessing competency. Some literature incorrectly reports that the ‘A’ stands for abilities. But your ability is a product of your knowledge, skills, and attitudes. You need all three domains to be competent. ‘A’ stands for attitude.

Let’s think about a real example: hysterectomy. According to the overwhelming majority of scientific evidence, vaginal hysterectomy is the way that most hysterectomies should be performed. It is easy, quick, cost-effective, and associated with the best patient outcomes by any standard with which it’s measured. It is appropriate for over 90% of all women who undergo hysterectomies for benign indications. Yet, vaginal hysterectomy is performed only about 10% of the time. This discrepancy reflects a gap in the competency of practicing gynecologists. The lack of competency could be in any of the three KSA domains. You need all three – knowledge, skills, and attitude – to be truly competent at a task.

Let me show you why.

Knowledge. Knowledge is the low-hanging fruit of medical education and evaluation. It’s easy to teach and easy to assess. Give a student a lecture or a book to read and then give an exam and you can teach and assess knowledge fairly easily. Knowledge is what medical education focuses on almost exclusively. You can read my book about vaginal hysterectomy and have the knowledge of how to do one; I can ask you questions or give you an exam and assess your knowledge and retention of knowledge of how to do one. But just because you do that and pass the test doesn’t mean you are competent to do a vaginal hysterectomy. Knowledge doesn’t equal skill.

Skills. Acquisition of skills in medical training is more complicated than acquisition of knowledge. Students (and residents) need careful training and adequate repetition to master individual skills; how much repetition varies by the skill and the learner. Educators are often not that good at either teachings skills or evaluating mastery of a skill, let alone customizing a course of instruction for a particular student. A surgery like vaginal hysterectomy is actually composed of dozens of individual skills. All must be mastered to be competent (I discuss these in detail in my book). If even one skill is substandard, the whole surgery suffers. The most difficult part of teaching something like vaginal hysterectomy is breaking it down into its many component skills and then ensuring that the learner has mastered each. Skills are not just physical acts. A skill might be interpreting a lab test or constructing a differential diagnosis. A competency-based curriculum must break down a given task into individual skills and then encourage and evaluate mastery of each.

Attitude. Just because you know that vaginal hysterectomy is a great thing and possess the skills to do it doesn’t mean that your patient will actually end up with one. Attitude is the least taught and least evaluated part of competency but perhaps the most important element. Attitude really does determine in the end what you do and how well you do it. So what it attitude? Think of it this way: there are plenty of gynecologists who know that vaginal hysterectomy is the best approach and many who possess at least the basic competency to do one yet they choose not to. What attitudes might contribute to this? Here are some possibilities:

• Placing the needs of the physician (financial and otherwise) above the needs of the patient.
• Believing, in general, that newer is better.
• Being susceptible to clever marketing.
• Discounting the role of systematic evidence in individual decision making.
• A fear of change or trying new things.

These are all attitudes. There are many, many more attitudes which may contribute to the lack of vaginal hysterectomies, but you get the idea. The wrong attitude can easily trump knowledge and skill. Indeed, attitude is actually more important than either skill or knowledge. For example, a physician may not believe that he possesses the skills to perform a vaginal hysterectomy. In that situation, he has only two choices if he has the attitude that the patient’s needs are more important that the physician’s needs: either take action to improve his skills or refer the patient to someone else. But how often does either actually happen? Almost never. The attitude of the physician stands in the way of optimal patient care.

Another injurious attitude is the practice of defensive medicine. If fear of litigation is your main motivator (rather than delivering the best care possible to patients), then that attitude will drive you to make lots of uninformed and non-evidence based care decisions and ultimately cause patient harm.

Were all of the previous providers who saw my patient lacking in the knowledge or skills necessary to interrogate the patient, make a differential diagnosis, and select an appropriate intervention? I seriously doubt it. I would argue that attitude was the leading cause of the incompetent care. The original gynecologist was motivated to make money by doing a hysterectomy; this was his cognitive disposition to respond (or bias) that led him to interpret the fact set incorrectly. Later, midlevel providers who saw the patient likely believed that they were not in a position to question the diagnosis of the specialist; this, too, is an attitude that biased them into not considering alternate diagnoses and treatments. The emergency room physician was worried about liability and had the mindset that he must exclude serious, potentially life-threading illnesses, so he ordered too much testing and missed the forest for the trees. This is also an attitude (and a bias). The need of the physician (protect myself from a potential lawsuit) outweighed the need of the patient (get an accurate diagnosis and not waste time and money doing it).

How do we teach and test attitude? It’s tough. I believe it starts with modeling good behavior and not being afraid to assess a provider’s attitude when looking for reasons why a deviation from the standard of care has occurred. What are some attitudes that we should encourage? You’ve probably heard some of these before but maybe not thought of them in this context before:

• First, do no harm. Every intervention has harm associated with it; only undertake it if you can clearly delineate why the benefits of the action outweighs its risks and the risk of inaction (e.g., don’t do a hysterectomy for IBS).
• The patient always comes first. The concerns and interests of the patient always, always take precedent over the concerns and interests of the physician (e.g., if the patient needs a hysterectomy and you can’t do it vaginally, refer her to someone else who can).
• Treat all patients equally regardless of their illness, how well you know them, or their ability to pay. Treatment choices should not be based upon type of insurance (e.g., the need for hysterectomy should not be based upon the willingness of the insurance company to pay for it).
• Respect evidence not anecdote. Evidence from large controlled trials will always trump the individual experiences of a particular physician (e.g., just because you believe laparoscopic hysterectomy is better than vaginal hysterectomy is not a reason to do one).
• Listen and be open-minded. Your patients have the keys to diagnosis, but you must let them talk and ask questions in the right way (e.g., you only know about the constipation if you let her tell you and you haven’t already decided what her problem is before you even ask).

Here are some other attitudes for you to ponder from Malvinder Parmar: