Rethinking the Concept of the Diagnosis

It was my first day of my first rotation of my third year of medical school. I was starting a month of inpatient pulmonary medicine. One thing was certain- I had absolutely no idea what I was doing. I arrived in the morning with a squeaky clean, perfectly pressed short white coat- its multiple pockets overflowing with pharmaceutical branded pens compiled from drug company sponsored lunches, a library of miniature medical reference handbooks, and even a tuning fork- a diagnostic tool used so infrequently in internal medicine, I was instantly mocked by the intern on service for even bringing it into the hospital.

The senior resident on service assigned me to follow a new patient admitted for respiratory distress. This 40 year old guy came in with difficulty breathing, pulmonary nodules on his chest X-ray, a patchy rash, worsening kidney function, tingling in his fingers and a seizure the previous day. The team still didn’t know exactly what was going on with him.

Will my patient figure out I am a fraud?

As I prepared to enter my first patient’s room for the first time, I paused at his door for a moment and took a focused deep breath and tried to breath out my anxiety. As if I were an actor waiting for the director to yell out “action”, I morphed into character for the part I was about to play- that of knowledgable health care professional. In reality, I knew I was a clueless imposter.

My patient had already been seen by multiple doctors and specialists and his treatment was well underway. His diagnosis however was still a mystery. As I entered his room, I could see the hope spark from his eyes that I might be the one to finally reveal why he was in the hospital. After a moment the disappointment in his voice was profound when he realized that all I was going to do was re-ask him the same generic medical history questions he had answered many times before.

There is a guilt that can accompany you on your journey as a medical student. The reality is, often, you have precious little to give a patient; you are there only to take-to steal knowledge and experience from their unfortunate situation. The only thing you have to offer back is your sympathy, which many times is just not a fair trade. So there I was playing my part as “doctor,” asking questions and doing my best impression of a physical exam. I offered up my honest sympathy, and when I was done, escaped the room releasing a similar exhalation that I entered with-this time filled with the relief that I didn’t make a complete ass of myself.

Diagnosing isn’t so hard

That evening I went home, opened Harrison’s Principles of Internal Medicine-a multi-thousand page reference textbook that most experienced physicians use a doorstop, and scanned for possible diseases that could fit my patient’s presentation. After about an hour of deep digging, I came to a confident conclusion.

The next morning, dressed again in my short white coat costume, I presented my patient to the entire medical team (attending physician, fellow, resident, interns and other students). I reviewed the full details of his hospital presentation and medical history, his physical exam, lab and radiology results, and ended with a differential diagnosis. I concluded he most likely had a disease called lymphomatoid granulomatosis.

Imagine my surprise when I looked up from my notes to see the entire medical team staring blankly at me and the resident and fellow chuckling at my conclusion.

In all my research, I somehow failed to notice the epidemiology of the disease. Not only had no one on my medical team ever seen this disease before, no one had even heard of it. It turns out lymphomatoid granulomatosis is an extremely rare disorder. There is apparently an overproduction of white blood cells that infiltrate tissues, especially in the lungs, destroying the blood vessels. The cause is unknown. The most effective treatment is unknown. In other words, this disease is effectively a list of signs and symptoms that show up together. Although it has a name, it basically is a mystery. The crazy thing was that after about five days of testing, it appeared that this guy actually had lymphomatoid granulomatosis- which remains to this day one of the rarest diseases I have ever seen.

Was I a crack progeny diagnostician inhabiting the spirit of Dr. House? Nope-I just looked at a list of signs, symptoms and lab results, opened a book, and essentially matched them up.

What exactly does it mean to have a diagnosis?

When we don’t understand the true cause of an abnormal bodily state, how do we decide that it merits a unique clinical diagnosis? If you analyze many diseases, the underlying biological cause is often not really understood or only partially understood. These maladies are often not caused by one specific malfunction, but are rather the result of a complex web of pathological interactions across organ systems-it can be hard to distinguish what was the root cause of the biological malfunction. Yet we still choose to give it a single name. It is an unquestioned tenet of modern medicine.

We love to diagnose. Why?

I recently had a patient with whom I was chatting before I was to administer her anesthesia for a routine colonoscopy. She adamantly recounted her medical history in far more detail than I really needed to know. She told me how a few years ago she developed fatigue, depression, fuzzy thinking and whole body pain. No one could diagnose her problem. A year passed. She visited specialist after specialist and as she said “none of these stupid doctors” could figure out what was wrong with her. They kept telling her everything was normal. Then she finally found someone, who in her mind was the Sherlock Holmes of medicine. This medical savant concluded she had Chronic Lyme Disease. She was so relieved to have the answer. With time she had physical and psychological therapy, a long course of antibiotics, and was prescribed antidepressants. By the time I met her, she felt so much better.

There was one complicating factor

According to mainstream medical consensus, the diagnosis of Chronic Lyme Disease does not exist. The “disease” has no clinical definition nor objective clinical findings. The infirmities are not the normal sequelae of a bout of lyme disease. These patients tend to have a constellation of certain signs and symptoms that do not fit any known disease’s accepted diagnostic criteria; In other words, they have a bunch of problems that cannot be categorized. Because these symptoms don’t fit into any known diagnostic category, the typical physician’s explanation to these patients is: “There appears to be nothing medically wrong with you”-which translated from doctor-speak means: “You are bat shit crazy!”

So what then does some enterprising healthcare professional who practices on the fringes of science and rational thought do with a problem they can’t explain? They name it. And naming it makes it real. Naming it can even make the patient feel better-without any other intervention. There is a certain psychological boost that comes from your pain being legitimized. A tangible diagnosis propels you out of crazyland and back to a place where treatment and cure are at least conceivable-for in our medical paradigm we do not feel comfortable treating something that has no name.

We name diseases we don’t understand at all

There are a number of very common “diagnoses” that have no concretely understood biological cause. (Amyotrophic Lateral Sclerosis- or “the ice bucket challenge disease”). There are also “syndromes”– which are really just a conglomeration of signs and symptoms-a description of a state of being and not an explanation of cause (Acute Respiratory Distress Syndrome). There are even diagnoses which are only given when everything else has been excluded by certain tests: “Diagnosis of exclusion.” (Fever of unknown origin- that’s a real diagnosis).

Should we even bother diagnosing?

We take for granted that establishing a specific “diagnosis” is the best course of action when approaching a sick patient. A diagnosis allows us to all speak the same language; it allows us to categorize patients so we can scientifically study disease prognosis and potential treatments. We accept that lupus exists, chronic pain exists, irritable bowel syndrome exists, but we cannot yet really explain their underlying cause. Why? It’s probable that these clinical entities are not actually a single disease at all, but rather a spectrum of complicated multifaceted biological and psychological malfunctions. Now, this is not some novel genius insight. However, we frequently don’t know how to apply this insight to actual patient care. Doctors feel comfortable putting diseases into discreet boxes. Without these classifications, we assume, practicing medicine would be chaos.

But would it? Let’s look to other medical paradigms…

Let’s imagine you go to your internist and he diagnoses you with rheumatoid arthritis. You’re skeptical-so your buddy tells you to go to a Traditional Chinese Medicine (TCM) doctor? Well you go, and the practitioner feels your pulse, looks at your tongue, and takes a detailed history. Then he explains your problem in a nomenclature that means nothing to you. For example maybe he says you have a “Qi stagnation with yin and kidney deficiency.” The practitioner may prescribe you some herbs or acupuncture to aid your body’s return to it’s normal state from it’s current off-kilter predicament. It’s a subtle difference from how we approach disease in the western world-the treatment goal is to return you to baseline health, not to eliminate a recognized interloping disease.

The western doctor has provided you a tangible diagnosis you don’t understand, whereas the TCM doctor basically gave you a description of your current body state using terms you also don’t understand. Who is correct? Now I am not trying to argue here for the merits of Traditional Chinese Medicine, but I would claim that western medicine might learn something from the way TCM conceives disease. It is important to recognize the whole body will constantly conform and adapt to a emerging novel pathology, moving away from it’s baseline status across organ system unrelated to the new disease. For example, a newly diagnosed cancer causes depression, depression changes what we eat and our activity level, our decreased activity worsens our cardiovascular health and on and on. A person’s state of health exists beyond their diagnosis-in interconnected ways that can be difficult to easily classify. Focusing on treating only the root pathological cause of illness likely blinds us to the subtle consequences of disease that may have ramifications different from what we expect. For illness does not always follow the rules and our common diagnoses are not as neatly packaged as they are in the textbook..

You see, rheumatoid arthritis is not as concrete an entity as it may seem. The underlying biological cause and downstream effects of that disease, including exacerbating or mitigation health factors, likely vary tremendously between patients- so much so that one diagnosis behaves completely different from another. Yet both are termed rheumatic arthritis. If you back your car into a trash can and cause a dent in your fender, you technically got in a “car crash”. If your car was parked in a training zone on a military base and run over by an M-1 Abrams tank, crushing it to a pulp, the car also got into a “car crash”. At it’s most basic level, what happened to your car in these two scenarios share the same name, but the underlying cause is so very different it’s almost absurd to think of them as the same thing. I would claim that we often make this mistake in medical diagnosis. Why? Mostly because we lack the means to describe disease more specifically. How do you accurately name something you don’t really understand? It’s very difficult, if not impossible.

Inaccurate diagnosis leads to inaccurate treatment

To study a novel treatment for a condition, the scientific method requires we agree on diagnostic criteria in order to categorize the disease we want to treat. We find people with that disease, subject them to a treatment, and see if it works. It’s the best we can do. But what if our first step in our medical research, the diagnosis, is inherently flawed?

Variability is so extreme in some diagnoses, as like our car crash example, that the the effectiveness of certain treatments is extremely variable. The variability among test subjects, even those who inhabit the same diagnosis, is probably far greater than we realize, dramatically affecting the measured effectiveness of certain treatment interventions. In other words, we could be ineffectually treating our tank crushed car with a treatment that would be perfect for our fender bender car (like a little touch up paint)… and we would be blind to our mistake.

As our understanding of healthy and diseased states comes more into focus, we need to evolve how we describe them. That way, maybe the next patient who has a collection of symptoms that don’t easily fit into any known diagnosis can assume her doctors will believe her problem is real even though they cannot give it a name. She won’t be desperately clinging to the hope that some pseudo-authority figure wearing a immaculately pressed short white coat and holding tuning fork will walk into her room and tell her the name of what is wrong with her. She won’t need to.