“What is a diagnosis?”
A Little Bit of Philosophy of Modern Medicine
We seem to have many budding (or maybe former) philosophers among the readers of Sensible Medicine. A couple of weeks ago, it was Plato; today, it is a dive into epistemology.
Adam Cifu
Have you ever asked your student or intern, “What is a diagnosis?”
I often pose this question to the students and interns who rotate through inpatient pediatrics with me. Understanding the answer to this question affects how doctors approach solving problems. It will make novice notes more organized and easier for experienced clinicians to understand.
Often, trainees tell me that a diagnosis is simply a description of “a constellation of symptoms.” More than once, I have been told that a diagnosis is a billing code. I sometimes have to do quite a bit of steering before we arrive at the idea that, when possible, a diagnosis must offer a coherent explanation for signs and symptoms. Facts cannot speak for themselves. We have to make sense of discrete facts and organize them within a structured explanatory model.
Learning that diagnoses must explain rather than simply describe illness changes how trainees write their assessments. The quality of notes improve and trainees understand what it means to move from a “reporter” to a “synthesizer”. They begin to make arguments in their assessments, rather than merely summarize the patient’s clinical data and hospital course.
After asking trainees to define what it means to make a diagnosis, I often ask how modern medicine differs fundamentally from non-modern medicine.1 Folk medicine, or pre-modern medicine, also attempts to explain signs and symptoms. “Doctor, why am I sick?” is a question that has always preoccupied people.
Most trainees respond to my second question by saying that modern medicine is “evidence-based.” Because I can be irritating, I’ll ask, “What do we mean by evidence?”
I am grateful for Vinay Prasad and Adam Cifu’s distinction between reductive and empirical evidence in Ending Medical Reversal. I typically teach that evidence is a tether that ties the explanations for patients’ signs and symptoms to the real world. Once an explanation is proposed, you can go to the real world and gather information that can prove the explanation as correct or incorrect.
For example, if the explanation for a fever, cough, and shortness of breath in an otherwise healthy adolescent were an imbalance of “qi” or “yellow bile humor,” what could possibly be collected in the real world to inform one that such an explanation is right or wrong? If the explanation is pneumonia, defined as a bacterial infection of the alveoli and terminal airways, one can observe an elevated white count, decreased oxygen saturation, crackles on exam, and chest X-ray findings. These findings are what Dr. Prasad calls reductive evidence. The treatment that then follows typically depends on what he calls empirical evidence. Large, controlled clinical trials have found that, if the patient is not too sick, treating with high-dose oral amoxicillin for 5 days yields optimal results in most otherwise healthy adolescents.
We generally require building a cumulative body of evidence from multiple lines of testing to increase confidence in our explanation. The probability that evidence gathered from multiple lines of testing would converge on a single explanatory hypothesis becomes vanishingly low if the explanation is incorrect and arrived at by chance. Importantly, requiring evidence leads to epistemic humility. There always remains a possibility that your explanations are wrong or that the treatments you prescribe are ineffective. If they are, you’re going to find out. Being rigorous about evidence is the only way for medicine to detect errors and institute corrections. By grounding medicine in evidence, we make it self-corrective.
Modern medicine also demands honesty about both ignorance and uncertainty. A good hallmark of a trustworthy doctor is her ability to say "I don't know" or "the scientific community doesn't know yet." This crucial point probably does not get enough emphasis in medical education. A proper philosophy of modern medicine mandates that we-- both collectively and individually-- must never pretend to know things we don’t know.
This is why I think that some basic training in philosophy would help a medical trainee understand why what we call conventional medicine is superior to any other non-science-based system of medicine. A doctor of modern medicine must have no compunction in acknowledging that requiring explanations for signs and symptoms to be tethered to reality is objectively better than being unaccountable to reality.
This brings me to my point that trainees ought to understand that modern medicine must prioritize truth above all. Truths are statements about reality, and evidence is what ties our concepts about diseases to reality. The purpose of medicine is to uphold individual human dignity by addressing illness, saving lives, and relieving suffering. Without a grounding in reality and a commitment to truth, we cannot fulfill this purpose.
Medical institutions must build trust with their communities by publicly declaring that truth-seeking is their highest priority. By making truth-seeking in medical training the highest goal, trainees gain a powerful arsenal against indoctrination and dogma within a hierarchical educational structure rife with power differentials and peer pressure. As physicians, we must vigorously defend the philosophy that underlies our profession and be unapologetic in affirming that we have the privilege of serving the best medical tradition ever devised.
Dr. Erica Li is a pediatrician in Washington State. She went to medical school at UC Davis and trained in general pediatrics in Los Angeles. She currently teaches medical students and interns. Certified in Pediatric Hospital Medicine by the American Board of Pediatrics, Dr. Li is technically a subspecialist who identifies as a generalist.
As an aside, I almost never use the term “Western medicine,” because it implies there is an equally valid and parallel "Eastern" or “Native” counterpart. The difference between so-called Western and non-Western medicine is in epistemology, not in geography or culture.
 | A guest post by
|
Insightful essay.
I'll offer a different angle that is relevant to us all: When is a diagnosis not a disease? Alternatively, are diagnoses always 'diseases'. For example, One may argue that red hair is a diagnosis, but it surely is not a disease. Constitutional short stature is not a disease but some parents may demand treatment. How about Gilbert syndrome (mild hyperbilirubinemia)? --a lab finding with no clinical relevance to the affected individual, though an ill-informed clinician may decide to 'work it up', which itself may be cause harm. At one time homosexuality was a disease, and then by fiat it wasn't. The American College of Cardiology's can double the population of hypertensives by stroke of pen by redefining hypertension. Drug manufacturers won't rest until every one of us carries a diagnosis of hypercholesterolemia and is taking their pills.
How about a positive covid test in an otherwise asymptomatic individual? In many jurisdictions, these souls had serious limits on their civi liberties based on the evidence-free assertion that they 'might' spread covid. And contrary to evidence, public health authorities considered vaccinated individuals as if they wore a magic cape that prevented the spread of covid. Compare this to the 30 percent of the general public who walk around with staph aureus in their noses. Like those with a positive covid test they 'might' spread staph infection to another, right? Outside of high school wrestlers, this risk is extremely low and not worth our concern. Common sense reigns, and we don't treat staph carriers as infected pariahs.
The point of the above examples, among many more, is the influence of sociology on medicine. The sociocultural norms of society subject medicine to massive biases that easily result in net harm.
I always thought that the old medical CPC challenge of the revered expert making the diagnosis was antiquated and silly. But the challenge of "making the diagnosis" seems to be carried to the extreme in these days of coding correctly to get the maximum payment.
I have had my diagnostic coups. During my internship in 1964 I worked up a patient on the woman's ward who explained to me about how she one by one gained weight, started to get increased blood sugar, then blood pressure, then some increased hair on her lip, stretch marks on her abdomen and bruises on her arms. My sole diagnosis on my handwritten H&P was Cushing's Syndrome and wouldn't you know she turned out to have a pituitary tumor. My resident and attending were flabbergasted.
I also recall in my practice a long-time patient who on one of her visits sat across from my desk and kept pushing up her eyelids with her finger. She was getting myasthenia gravis and I guessed it, the only time I ever saw it. I was quite proud of myself, although there was no resident to impress that time.
But by and large I never viewed what I did for my patients to be making a diagnosis and applying the corresponding treatment. My view is that medical practice is a service relationship that starts with an individual with a unique medical concern. Each one of us has attained a certain body of medical knowledge and what we do is to take that knowledge base, modified by our experience and personal human skills, and we use the resources available to us and our colleagues in our community and apply it to the problem of that individual person sitting in front of us who comes to us seeking help and advice.
When I was still in practice and dutifully coded my bills for the third-party payors I always used a single broad multipurpose diagnosis. These days as I understand it that doesn't fly if you want to make the best income. Too bad. We should be paid by how well we satisfy the patient.