Practicing Medicine
How patient stories inform medical knowledge
We were watching a historical mini-series on television the other night, a story covering several decades over the span of eight episodes. One of the characters developed a cough. He took great pains to hide it from his family, but minutes later they all noticed that he was coughing up blood. At that point, I turned to my wife and said, “he'll be dead shortly”. Sure enough, he took to his bed, a doctor dropped by to confirm the poor prognosis, and by the end of the episode he was dead and buried.
The point is not that I’m a brilliant diagnostician. It’s fairly easy to spot overused plot devices (or “tropes”) including, as in this case, the “incurable cough of death” and “blood from the mouth”. In historical dramas, these are both shorthand for tuberculosis. Movie-goers will recall similar harbingers of doom for Satine in Moulin Rouge and Fantine in Les Miserables, among others. In the show we were watching, the diagnosis was never named. The plot required that we get the poor man out of the way quickly so that his son could take over the family business and prove that he was just like his father, also a trope!
(As an aside, and we’ll return to this thought later, there’s no way that his cough was going to turn out to be irrelevant! Yet another trope, the “Law of Conservation of Detail”, tells us that, in the time-limited context of a TV show, any plot point introduced early in the story and passed off as insignificant will inevitably turn out to be very important.)
Strangely, all of that got me thinking about the practice of medicine and its relationship to language, conversations and stories, not just when things go well, but also when they don’t. I’ll explore these topics over the course of a few posts.
To start with, the symptom described in plain language as “coughing up blood” has a corresponding medical term, “hemoptysis”. While some complain that medical terminology derived from Greek and/or Latin sounds impenetrable or even “elitist”, in the end it's far more efficient to use one precisely defined medical word to describe a symptom, rather than several plain language words together in a phrase. This is true for any specialized body of knowledge. It’s easier, for example, for a mathematician to talk about “triangles” than it is to refer to “closed polygons having three sides, three angles and three vertices” (with several of those words requiring further translation to get to plain language). Hence, a significant portion of medical education is dedicated to learning the “shorthand” of medical terminology, to support more efficient medical documentation and communication.
In terms of the diagnostic process, it is unusual for a single symptom, like hemoptysis, to completely define the diagnosis and prognosis of the patient. In the rare circumstances where this does occur, we use the fancy medical term “pathognomonic” to describe the linkage, one example being the “pill-rolling tremor” considered to be pathognomonic for Parkinson’s Disease. Outside of historical dramas, hemoptysis is not pathognomonic for tuberculosis.
More commonly, to make a diagnosis the doctor follows a time-honored process, a skill they spend years learning in medical school and honing during their practicing lifetime.
This process starts with an enquiry about the main symptom, otherwise known as the “presenting complaint”, and later includes other symptoms. The complete “history” includes details about the circumstances in which the symptoms occur, including what makes things better and what makes them worse, which other symptoms might occur at the same time, etc. In effect, the patient tells their personal disease story in plain language and the doctor listens and seeks clarification where needed. Then, to support their internal thought processes, the doctor translates and edits the patient story into a comprehensive medical story, called the “history of the present illness”, using medical terminology.
Because physicians have pooled and shared their observations while treating millions (or even billions) of patients over the centuries, they’ve found that many of these patient stories are stereotypical. “Bottom-up” (or inductive) reasoning allowed them to figure out that groups of patients with similar stories are often suffering from the same underlying problem. Further scientific observation and investigation revealed the underlying anatomical and physiological abnormalities, which were then categorized as diseases. In the end, disease names become a form of shorthand for the entire story. For example, when discussing a 45-year-old patient with my colleague, I can convey quite a lot of useful information simply by saying that the patient developed Type 1 Diabetes at the age of 12, even though the specific details will vary from patient to patient.
With all that medical knowledge in hand (or “on tap”, because you can look up what you don’t know), doctors apply deductive (or “top-down”) reasoning to make sense of individual patient stories. The presenting complaint and history of the present illness help the doctor decide which diseases seem more likely and which others are less likely. This isn’t always a conscious process involving deep reasoning, it can be just a matter of intuitively recognizing something they’ve seen before. For example, a zoologist quickly recognizes a striped horse-like creature as a zebra, although closer examination might later reveal that it’s a horse on which someone has painted stripes. Similarly, doctors quickly conclude that an older adult experiencing crushing pain in the middle of the chest, radiating to the jaw and/or left arm, associated with sweating and nausea could be having a myocardial infarction (or “heart attack”), and then they start to look for a blockage in one of the arteries supplying blood to the heart muscle. Deeper thought could disclose other explanations, but in some cases it’s best to deal with the life-threatening possibilities first, pending further information like test reports.
When making a diagnosis, doctors actively consider what further information might be helpful. This could be more stories, including some based on the observations of other people, like family members or bystanders. It might mean more “objective” information, such as the physical examination, X-rays, or lab tests. All of these add to the medical story, progressively firming up or even changing the “working diagnosis”.
In the end, the doctor can tell the patient a related (overlapping) story, preferably in plain language, explaining the probable cause of their problem, the nature of any further investigations and treatments, and the possible outcome (the “prognosis”). This story makes sense of the patient history (the past), explains where things stand at present, and presents some options and predictions for the future, including tests, treatments, and possible outcomes.
Before we leave this topic, it is worth noting that context matters, another thing that doctors learn through their training and experience.
It’s not simply a matter of the patient and their story interacting with the doctor and their knowledge. The diagnostic possibilities for any given patient reflect where and when the patient is seen.
Just as African savannah hoofbeats probably come from zebras while western plains hoofbeats suggest horses, hemoptysis means different things in different places. In India or Bangladesh (or in Europe in the 1800’s) it may well be tuberculosis, whereas in the Western world in 2023 it is more likely lung cancer. Some diseases follow seasonal patterns. Influenza, for example, is more common in winter, so the patient with a fever and cough in summer likely has some other problem. The doctor’s place of work is relevant. Patients with chest pain are more likely to be suffering from a heart attack if they are seen first in the emergency department rather than the family doctor’s office. (It’s not that you never see heart attacks in the office, it’s just that the patients sort themselves out, to some degree, based on factors like severity of symptoms, heart attack awareness campaigns, ease of getting seen by a doctor, etc.).
Precisely because context matters and disease patterns vary, doctors who move from one place to another will face a learning curve before they “catch up” to the locally trained doctors. This works both ways. The doctor from Canada who moves to India has as much to learn as the doctor moving the other way. The doctor who spent years working in emergency will have things to learn when they transition to office-based family practice, as will the experienced family doctor who starts to work in the emergency department.
Even without moving from one place to another, doctors must be lifelong learners. Things change over time; new diseases come along, we learn to think about old diseases in new ways, new treatments come along and old treatments fall out of favour.
And that’s the way that it works, at least in theory.
In practice, medicine started off as an art hundreds of years ago when little was known about the inner workings of the body. It has gained scientific credibility over time. Modern medicine aspires to be a body of scientifically validated knowledge and techniques, resting on a solid foundation of basic sciences including human anatomy, physiology, molecular biology, biochemistry, pathology, etc.
However, we do know that we don’t know everything. It’s not uncommon for individual patients to have symptoms “not yet diagnosed”, a fancy term meaning “we don’t know”. Medical research is ongoing, new discoveries are published every day. The published clinical practice guidelines always point out gaps in our knowledge (“opportunities for research”). Deciding what to do with patients who fall into those gaps is still more art than science.
Beyond the medical lingo and methods, the art and the science, the practice of medicine represents a specific way of thinking about disease. Clearly, it’s far more complicated than “coughing is bad” (or, even worse, “coughing up blood means that you are going to die”). While it works well much of the time, there are many ways that it can go wrong.
In my next post, I’ll explore how much is “art”, how much is “science”, and how we find the right balance.