Thursday, August 21, 2008

Professor Schweitzer explains Lies, Damn Lies and Statistics

In the interests of being forewarned ...

I have twice in the last few weeks heard a reference in the media to someone having "fatigue syndrome."

Why would that be of concern?

In ICD-10, which is the next set of WHO codes for diseases (the U.S. is the only industrialized nation still using ICD-9), both CFS and Myalgic Encephalomyelitis are categorized in the neurological chapter at F93.3  [The WHO schedule of codes is particularly important in the U.S. because although it was designed to be able to match diseases from one nation to another, in the U.S. it is primarily used for insurance reimbursement - those are the tiny numbers your doctor's receptionist puts on your records for billing.]

ICD-10's inclusion of CFS in a code that had belonged to M.E. since the late 1960s left Simon Wessely, a British psychiatrist who is convinced that "CFS" is psychosomatic, without an acceptable psychological code. So he adopted the term "fatigue syndrome," which is formally coded in "neuroses" at G48.0, with the formal name "neurasthenia" (a term not included any more in U.S. psychological textbooks, but originally coined to refer to the "nervous condition" supposedly common among women in the Victorian Era - goes with "swooning.")

The U.S. CDC has been slipping away from the Fukuda definition by revising the WAY CFS is diagnosed - not by looking for four of the eight symptoms listed in Fukuda, but by using a questionnaire that was supposedly demonstrated to be effective by "correctly" diagnosing CFS patients in a two-day Wichita hospital stay in 2003.

However, the article describing that two-day hospital stay clearly states (in the tables - not so clearly in the prose), that of 58 patients who were diagnosed with CFS in the Wichita Surveillance Study according to the Fukuda guidelines, only 6 still could be diagnosed with CFS using those guidelines. Another 4 could be added, but that would mean they only had 10.

[See < http://www.biomedcentral.com/1741-7015/3/19 >, "Chronic fatigue syndrome-a clinically empirical approach to its definition and study"]

So they rediagnosed the whole bunch, using the new questionnaires.  They ended up with 43 "CFS" patients.  But it was a tautology - the Wichita two-day hospital stay "proved" that the questionnaires identified "CFS" by using those same questionnaires to diagnose patients in the first place. 

The new studies to emerge from Atlanta used the questionnaires from the Wichita hospital stay.  Thus, they do not diagnose CFS-Fukuda - they diagnose what can only be called CFS-new definition or perhaps CFS-Reeves. 

Another study using the new questionnaires "proved" that there was no difference between patients who complained of "chronic fatigue" and patients who had "CFS" (new criteria) with regard to two critical symptoms in the Fukuda definition - cognitive dysfunction and sleep disruption.

Furthermore, the CFS-new definition patients did not have NMH/POTS.

The questionnaires do not exclude patients with depression.  Technically, an international study had recommended that patients who had a bout of depression that had resolved SEVEN YEARS BEFORE THE ONSET OF CFS could be included, but in the Wichita study, that little requirement that there be seven years BEFORE the onset of CFS wasn't mentioned. So the new questionnaires also include some patients with depression.

The result?  Estimates of "CFS" that match Simon Wessely's - 4-7 times as many as would be predicted using the Lenny Jason-DePaul study that was published in 1999. 

Why would that be a problem?  If the CDC wanted to "test" whether a particular disease or condition or biomarker was associated with "CFS", they could use their questionnaires to come up with a heterogeneous mishmash of patients who all have "fatigue," but not what was named "CFS" back in 1988.  They have already done so with an article that insisted NMH/POTS was not linked to CFS. 

And now we are hearing about "fatigue syndrome."  That would complete the transition to the British psychological school of CFS.  [The British psychiatric version is where "cognitive behaviour therapy [CBT]" and "graded exercise therapy [GET]" got their start - although, intriguingly, the studies that were supposed to prove the efficacy of both turned out not to be very robust, even though the data set was "cooked" - statistician-ese for a data set that is put together to get the results you want.]

So keep a lookout on and watch out for the phrase "fatigue syndrome."

And keep pushing politicians to focus on BIOLOGICAL TESTING, OBJECTIVE BIOLOGICAL MARKERS (not questionnaires), and TREATMENTS.

Mary Schweitzer

--

ERRATA
I apparently had a brain hiccup and mixed up the WHO ICD-10 codes for "chronic fatigue syndrome" and "fatigue syndrome."

So here they are corrected:

ICD-10:

G93.3  (not F93.3) is where M.E. and postviral fatigue syndrome are listed, and if you have a copy of the index (it is not on WHO's website but I have a FAX of it at home) you will see that "chronic fatigue syndrome" is coded there, too.

F48.0 is "neurasthenia," and "fatigue syndrome" is coded there.

I had never heard the phrase "fatigue syndrome" used in the U.S. before, so two references in one months caught my eye.

The main problem would be that "fatigue syndrome" would code as a neurosis and also as "neurasthenia", which is the old-fashioned "nervous condition" that women used to be diagnosed with a hundred years ago.

Sorry about the error. 

Mary Schweitzer

* * *

There's a previous post in this blog about not necessarily trusting research results, and Mary's analysis here is along the same lines.

If you are doing your research on patients who have depression, then you're not going to find information useful in treating post-viral CFS.  And, unfortunately, too many researchers have an ulterior motive, and the patients in their CFS studies have everything *but* True CFS.

In selecting which CFS research to post here, I try to limit it to those studies that are researching what Mary and I have, what Doctors Cheney/Bell/Peterson consider CFS, and not those that are loaded up with patients who are simply fatigued from overwork or depressed.  Sure, there are those who claim that their research "proves" exercise cures CFS, but I can tell you from 21 years of experience with this disease that exercise makes my symptoms worse.  Even at my best, I no longer had the stamina I did before, and by orders of magnitude that can't be explained away with mere aging (and certainly not being just 3 months older than I was before I had the virus, or one week older than when I relapsed).  On the other hand, I lived with a depressive, and saw first-hand that exercise can produce energy/euphoria in someone who has depression.  So, I can say with some certainty that if exercise makes you feel better, you don't have what I have ... even if your doctor *did* tell you it's CFS.

We are moving toward a biomarker for the disease.  People who actually have depression who were misdiagnosed with CFS (and vice versa) will finally get the right diagnosis and the right treatment.  And all the psychologizers who've made a living saying CFS patients are "just depressed" will be proven wrong and watch their reputations fall in shambles.

Meanwhile, remember that not every research study purporting to study CFS actually studies what Mary and I have.  Read critically and skeptically, especially the patient selection criteria.

No comments: