Hickam Tricked 'em with the 'ol Red Herring Ruse

One of my favorite Charlie Munger quotes goes something like this:

"Warren, over the years you and I have appreciated the great value in learning from mistakes; Another thing we realized is that it's a lot more pleasurable to learn from other people's mistakes."

In light of that sage advice, here's another recent case report invoking Hickam's dictum. In this case, a 38-year-old man from Pakistan who has sex with men presents with uveitis and a rash on his palms and soles. He has a positive interferon-gamma release assay. This caused diagnostic confusion which resulted in the unnecessary invocation of Hickam's dictum. There is, after all, a simple explanation for the two findings, as I explicated in an online comment:

"This is an interesting case. Neither Ockham's razor, nor Hickam's dictum, nor Crabtree’s bludgeon need be invoked to sort it out.

The annual incidence of syphilis in men who have sex with men in the USA is a whopping 309 cases per 100,000 persons per year - almost as common as community acquired pneumonia in unselected adults. Compare that to 2.9 syphilis cases/100,000 person-years in men who have sex only with women. A risk factor that increases the incidence of disease over 100-fold should point even the novice diagnostician straight to the diagnosis. Combined with a rash on the palms and soles? The diagnosis is all but made. Positive RPR? We’re finished here! For guidance on using incidence/disease frequency as a surrogate for disease probability, see: https://pubmed.ncbi.nlm.nih.gov/34854514/

Next, IGRA (e.g., quantiferon Gold) is NOT a diagnostic test for active tuberculosis infection (indeed, active infection is a cause of false negative IGRA), it is used clinically as a marker of prior tuberculosis exposure. In Pakistan, the annual incidence of tuberculosis is 230 cases per 100,000 persons per year, so it is hardly surprising that this patient has a positive quantiferon-gold. This is a red herring rather than an instantiation of Hickam’s dictum. Likewise, the dermatologist consultation was a false negative test for secondary syphilis. It is unbelievable that they attributed this rash to psoriasis, but this highlights that we should not put too much credence into any single piece of data or consultation if it flies in the face of the totality of the evidence. Similarly, angiotensin converting enzyme (ACE) levels are notoriously both insensitive and non-specific.

In the end, there is a simple unifying (and unsurprising) diagnosis in this case: syphilitic rash and uveitis in a man who has sex with men. For an explication of Ockhams razor and Hickams dictum (and the limits of their inferential powers), see: https://pubmed.ncbi.nlm.nih.gov/39467949/

Interestingly, Hickam started in the Boston hospital system, went to Emory, then to Duke, and finally to Indiana. Though not formally specialty trained, he did pioneering work in ophthalmology research, as well as cardiopulmonary research. An entire edition of the Archives of Internal medicine was dedicated to him after his death by reprinting seven of his original papers – the entire issue of the journal."

Hickam's Victim: The Unwary Diagnostician

In this case report, the authors describe a woman, some 82 years of age who had immigrated from Korea 30 years prior to presenting with syncope and dehydration, which was soon discovered to be Salmonella enterica subspecies typhi. Despite urine culture growing gram negative rods, 

"CT of chest, abdomen, and pelvis without contrast was ordered to better characterize the source of her sepsis."

Ignoring that we already have a source for her sepsis (indeed the most common source: urosepsis), and that gram negative sepsis almost never originates from the chest, we will move on after noting that they found multiple cavitary lesions in the upper lobes of the lungs. The authors state that they tried to link the cavitary lung disease with the urinary tract infection (staghorn calculi were found too!) and Salmonella bacteremia, because of an inclination towards Ockham's razor, but alas learned that the chest disease was tuberculosis (who knew!?) so they invoke Hickam's dictum. I rather think that this is a case of Crabtree's bludgeon which states that:

"No set of mutually inconsistent observations can exist for which some human intellect canno t conceive a coherent explanation, however complicated."

Alas, much about the epidemiology and natural history of both Salmonella and tuberculosis in this case report was neglected, so I penned this letter to the editor, below. I did not receive confirmation of my submission, and this journal's website is a bit inchoate, so I assume my letter was lost....

Added on 10/6/2025: Actually, my letter was not lost, as I received an acceptance letter today, but only after the invoice charging me $1530.00 USD as a publication fee for a letter to the editor! But let me tell you, the reviews, wherever they came from, were over the moon with my "clinical acumen" and "insightful analysis".

I replied saying nevermind, I can't afford that charge to publish a *letter* They replied saying "how much can you afford?" I offered $200 and it was immediately accepted. Never overlook the value of flattery, or the cold hard dollar. 

So I will link the letter, after it is published. In the meantime, comment below if you have your own insights about the case.

Epistemic Arrogance. Case Records of the Massachusetts General Hospital. Diagnosis is Stochastic, NOT Deterministic

This post was spurred by my reading the July 24^th NEJM case records, “A Man with Cough, Dyspnea, and Hypoxemia.”  When I was still using twitter, I mused that the discussants have too high a rate of getting the right diagnosis. This implies that there is no stochastic or aleatory uncertainty in diagnosis; that if you have the requisite information and reasoning process, you can connect the dots, every single time, to the final diagnosis. In this view, there is no randomness to diagnosis, no inherent and irreducible probabilistic (or stochastic, or aleatory) uncertainty. Of course, this is pure codswallop: stochastic uncertainty is what makes diagnosis so hard.

You cannot say that the result in any case was not possible with the information presented, but with a population of cases you can make inferences about whether the percentage of correct discussant diagnoses is improbable. The fact that the discussant gets the right diagnosis almost 100% of the time (I have tabulated the % correct for over 10 years of these cases – it rounds to 100%) in the CRMGH (Case Records of the Massachusetts General Hospital) is like finding too little variability in a research study (or  studies; or Bernie Madoff’s annual returns) and concluding that the results (returns) were manipulated. We use our well-justified first principles assumption of stochasticity, find that there is not enough variability, and conclude that something is being manipulated to remove the aleatory uncertainty. (I use stochastic, aleatory, and probabilistic herein as interchangeable adjectives for uncertainty.) And so must it be in the CRMGH. How the aleatory uncertainty is being removed is anybody’s guess and any insider’s knowledge, I suppose. Possibilities include both implicit and explicit cues. An implicit cue may be the selection of, say, a specialist in heritable cardiomyopathies as the discussant in a case where the final diagnosis is a genetic cardiomyopathy. Explicit cues may be that somebody tells the discussant the final diagnosis at the outset or tells them to revise their discussion if they get it wrong in draft. Hospitals are not closed systems either. People hear if there was a case of Creutzfeld-Jacob diagnosed recently, and the director of the microbiology lab knows if a particular rare species was recently cultured. Tales of rare diagnoses spread, as they are the currency of clinical acumen. I can’t say that any of these mechanisms -- from subtle queueing to explicit cheating -- are occurring. I only know that the near 100% hit rate means that randomness has been removed from an inherently random process.