Yesterday, we looked at how real science works; today, in a repost from my old blog, we look at some really bad science. –PalMD
I’ve been meaning to touch on “Morgellons disease” (a form of delusional parasitosis) for a while, but haven’t figured out how to approach it. Thankfully, others have. In the first referenced discussion, a paper was cited. This paper was such a great example of how not to approach medical science that I just had to address it in detail, section by section…
The authors argue for a newly described illness which they call “Morgellons”. It resembles in all ways except name delusional parasitosis, a condition where a person is falsely convinced that they have parasites in their skin. In general, if you wish to describe a possible new illness, you should start by coming up with a useful operational definition. This doesn’t happen. Instead we get anecdotes.
It begins with a single practitioner’s experiences. It’s hard to overstate how problematic this is. Early in the HIV epidemic, solo practitioners in a few cities found unusual diseases, communicated with each other, and discovered the AIDS epidemic—the difference was that the discovered real diseases with shared characteristics.
This physician saw some patients who shared similar characteristics, and, rather than picking the most likely diagnosis, went straight for the zebra. A more apt metaphor might be a unicorn—she made her diagnosis based on a “research foundation” devoted to an illness made up out of whole cloth.
The next mistake is in linking the new “disease” with an account from a 17th century observer. The observations are old, non-specific, and based on no current medical science. He described an odd constellation of symptoms, and the Morgellons Movement adopted it. There is no way to validate 400 year old observations.
After co-opting a 400 year-old name, one of the authors started a foundation which “began accepting registrations from people with symptoms of this unrecognized disease.” They lay out no clear case definition, and change the characteristics of the disease to fit the patient. “…[I]t soon became evident that other symptoms within this patient group, such as disabling fatigue, life-altering cognitive decline, joint pain, and mood disorders…” sometimes accompanied the skin symptoms. Shifting the goalposts in order to make your case definition more inclusive is not great science.
This section of the paper is devoted to a long list of symptoms which once again fails to give a case definition. In this paper, which purports to report an important emerging disease, it is disturbing that half-way through the paper, no disease has yet been defined. Perhaps one of the worst sins of this section is the confusion of correlation and causation based on a false premise. The authors assume (or beg the question) that Morgellons exists. Based on that assumption they assert that:
[t]he high incidence of psychopathology, which appears to be directly attributable to this disease, confounds the clinical picture for these patients as the seek validation for an insidious dermatologic condition that defies logic, while sometimes exhibiting obvious symptoms of mental illness.
Might they have put the cart before the, er, unicorn? If someone acts mentally ill, and has a bizarre set of illogical symptoms, why not put the blame where it belongs? To take a disease of the mind, and simply assert that it is a disease of the body, will help no one.
The next assertion is truly horrible. “It appears that the putative underlying infectious disease, which has been unrecognized and untreated, can cause psychopathology in many patients.”
Holy crap! Now it’s an infectious disease?!? Based on what? And it’s an infectious disease that affects the skin and central nervous system? Maybe it’s a variant of syphilis! How can these few clinicians (OK, only one is a clinician) have stumbled onto something so important and end up ignored?
Epidemiology and Transmission
OK…time for more unfounded assertions. “The total number of registrations on the Morgellons Research Foundation website is presently 2200, which is believed to be a fraction of the actual number of cases.” Believed by whom?
Then of course they run into the same problem they’ve had from the beginning: “There is some evidence to suggest that skin lesions and fibers may not be readily apparent in all individuals with this disease… .” Then what defines the disease? How does one track the epidemiology of a disease with a name and no definition?
And here is the real coup de grace.
Skin biopsies of patients with Morgellons disease typically reveal nonspecific pathology or an inflammatory process with no observable pathogens…In general, pathologists look for signs of known diseases and, thus, may miss clues of Morgellons disease in biopsies.
OK. I can’t go on. It’s not just that a mainstream journal would publish such crap. It’s that it makes my head hurt. But I will form a logical hypothesis that my headache has something to do with reading, sitting at the computer, and being frustrated. It seems unreasonable to posit that it is due to some unknown infectious agent that can be neither measured nor defined.
In the discussion section of the paper, the authors invoke Vienese physician Ignaz Semmelweis, a physician criticized as a crank in his own time, but later lauded as a hero. Robert Park said, “to wear the mantle of Galileo it is not enough that you be persecuted by an unkind establishment, you must also be right.” This paper is not right—it’s not even wrong.