Jump to content

Dr. Hafeez Diwan's Blog

  • entries
    9
  • comments
    15
  • views
    2,785

The Texas Sharpshooter Fallacy and Dermatopathologists - you don’t have to be in Texas to suffer from it


Dr. Hafeez Diwan

1,014 views

 Share

Although I am in Houston, Texas, I cannot make any exclusive claims to the Texas Sharpshooter Fallacy. This fallacy is one which I believe we dermatopathologists (as well as others) should be aware of.

What exactly is this fallacy? Where do sharpshooters come into this, and why specifically in Texas? The name, if Wikipedia is to be believed, comes from the story of a Texan who shot holes in the side of a barn and then painted a bull’s eye around a group of them, giving rise to the mistaken impression that this Texan was a sharpshooter. Unlike this fellow who did this on purpose, we, that is to say human beings, do this quite unthinkingly. We draw conclusions that are incorrect based on parts of data that seems to fit. Here is an example:

We have hundreds of dreams. Sometimes, something we dream may resemble something that happens later on, say the next day. We may then believe that the dream is prophetic; but in doing so, we discount the thousands of dreams when nothing in our dreams comes to pass. I once dreamt, in the mid 1980s that the Soviet army had invaded the empty lot behind our house in Pakistan. This they never did, and so the dream did not truly foretell a future happening. But this did not stop me from concluding that I was something like Nostrodamus when I dreamt of robbers breaking into our house and robbing us at gunpoint – this, they actually did, but in Pakistan, where I grew up, such occurrences were (and sadly continue to be) commonplace, and so this dream should be viewed in the context of the hundreds of dreams that were just…dreams. This is the Texas sharpshooter fallacy in action: we take some data that appears to fit with a preconception or bias or seemingly uncanny coincidental impression that we have, and draw a conclusion that is invalid.

In the diagnostic arena, which is the focus of this blog, let me tell you about something that happened to me recently. I saw the case of an elderly gentleman with a large erythematous plaque on half his face and scalp, which grew from a small reddish lesion to its present size. When I looked at the slides, I saw necrosis and very atypical-appearing vessels dissecting through the dermis. I have had some bad experiences with angiosarcomas in the past, and so the moment I laid eyes on the slide, without any history, I wondered about these vessels. I said aloud to the person who was showing me this case, “I think this could be angiosarcoma.” At this, my colleague gave me the history, and the location and the age seemed to go along with a diagnosis of angiosarcoma. I showed the case to other experienced dermatopathologists, who similarly gasped and drew the same conclusion. However, we all noted, but didn’t give enough importance to the necrosis and bacteria in the specimen. I don’t know about you, but I haven’t seen too many angiosarcomas with necrosis and bacteria. I thought that this was probably a superinfection or the lesion may have been traumatized. As you can appreciate, my friends and I had fallen into the Texas sharpshooter fallacy. We had ignored the entire context and had focused on a portion of it, and had come to an erroneous conclusion.

Nevertheless, I didn’t completely behave like Rambo (diagnostically speaking). I said something like, angiosarcoma can’t be excluded, please treat with antibiotics and re-biopsy residual erythematous areas. As it turns out, antibiotic treatment led to miraculous clearing up of the angiosarcoma – if I were a really advanced case of the Texas-sharpshooter-fallacy syndrome, I suppose I could have at that time reported the first ever case of angiosarcoma successfully treated with antibiotics!

So what appeared to be an angiosarcoma was reactive change around an infection. I shudder to think what might have happened if the patient had received treatment for his “angiosarcoma”-that-wasn’t.

I have started noticing this fallacy more and more. Being aware of it is helpful, like a vaccine that may help inoculate us against it. Even so, I have no doubt that I will fall victim to it the next time I sign out (in a short while), but I won’t go down without a fight – a fight against my inner Texas sharpshooter.
 Share

2 Comments


Recommended Comments

Dr. Phillip McKee

Posted

Great story. It is a bit like HIV/AIDS patient's biopsies. Once you make a diagnosis, look carefully for the 2, 3 or 4 diseases that are also present in the specimen.
Link to comment
×
×
  • Create New...