Go to the CFIDS/M.E. Information Page
Mary Schweitzer's Essays on CFIDS/M.E.

Presentation to the Chronic Fatigue Syndrome Advisory Committee
U.S. Department of Health and Human Services
Mary M. Schweitzer, Ph.D.
Washington, DC - July 17, 2006

Thank you for allowing me this time to speak.

For ten years I have been coming to these meetings and asking the CDC to give just one press conference to the nation, letting everyone know (patients, family, employers, doctors) how very serious the disease called Chronic Fatigue Syndrome really is. For roughly the same period of time I have tried to persuade Dr. Reeves that going to the public would do more to educate family doctors than all of the effort that goes into formal education campaigns limited to those in the medical profession willing or able to participate.

Be careful what you wish for.

In June of this year, the CDC in partnership with the CFIDS Association of America began a $4 million ad campaign to tell the public about Chronic Fatigue Syndrome. I was stunned to find it was based on the claim that 4 million Americans suffered from CFS.


[Note: Since Dr. Reeves has not even submitted the study upon which the 4 million is based, he refused to confirm this number, although he had told the CFIDS Association to go ahead with the number in the campaign. When asked a direct question about my testimony, he refused to answer. It will be a year at least until this sees publication, which means it will be a year at least until Dr. Reeves formally acknowledges it. I do not understand why he should not at least disseminate a working paper, given that he has already changed the definition that the CDC is using for CFS.]

In 1998, using the Wichita study, CDC estimated that 500,000 Americans had CFS. Eight years later, we are expected to believe the number of patients had increased eight-fold? That means that every two years, the incidence of CFS doubled?

The current best estimate - as CDC acknowledges now - would be to take the 1999 Jason estimate and use it on today's population, which would give roughtly 900,000-1 million patients. That still means an unexplained four-fold increase in the estimate.

There is only one way this could have happened, and we have really had the hints all along. Dr. Reeves, and the CDC, have abandoned the Fukuda research definition of CFS and created a new one – secretly.

The CDC’s website for information on CFS no longer lists Fukuda as the standard for research on the illness. The revised Fukuda definition remains as a clinical definition – but under publications, there is now a link to “case definitions.” They are listed in chronological order. The last (presumably the one that is now in use), is an article called “Factor analysis of symptoms among subjects with unexplained chronic fatigue: what can we learn about chronic fatigue syndrome?” by Nisenbaum, Reyes, Unger – and Reeves.

It was published in the Journal of Psychosomatic Research in 2004. The conclusion? “CFS symptomatology is a multidimensional phenomenon overlapping with other unexplained fatiguing syndromes and this must be considered in CFS research.”

What are "unexplained fatiguing syndromes"?

Many have complained that the Fukuda definition leaves researchers (and clinicians) with far too heterogeneous (mixed) a population. Instead of working towards a more homogeneous grouping, CDC has headed in the opposite direction – muddying the waters even more.

How had this happened?

In 1998, using the accepted Fukuda definition for CFS, Dr. William Reeves of the CDC announced that 500,000 Americans had CFS. One year later, Lenny Jason and others at DePaul University published a scholarly article in the Annals of Internal Medicine that gave a higher estimate: 800,000 Americans.

I have always assumed that one of the reasons Jason’s figure was 60 percent higher than Reeves’ was that Reeves had misspecified his model [i.e., statistically and scientifically it was incorrect.]

During the data collection process, Reeves omitted from the beginning any patients who had ever been diagnosed with depression, patients who were medically obese, and patients who had diabetes or thyroid problems, whether or not they were undergoing treatment.

We already knew that two of those factors – thyroid problems and obesity – correlated with having CFS. That is, not everyone with CFS had either – but the percentage of CFS patients with those problems was far greater than in the normal population. As for depression, many patients were misdiagnosed with depression before the CFS diagnosis – and others had secondary depression because of the limitations of the disease.

It would have been easy to have included those patients in the studies, simply by adding variables for the conditions. But Dr. Reeves refused – saying in 1996 that he did not want to lose “degrees of freedom” – a statistical term that does not apply to a data set – and to the extent it applies to a model, it is only a problem when you are working with a small sample, such as 50 patients. Here we were talking about thousands of patients. There was no statistical (or scientific) reason to omit variables.

Because these patients were omitted, the ensuing demographic model was misspecified – hence the difference in the results. Many patients with severe cases of CFS were left out entirely. A second problem with those particular variables was that obesity and diabetes are far more common among the poor and minority groups. By omitting patients with those conditions, instead of including them with a variable, the data set was also biased against the poor, and minorities. The DePaul study showed that those of all incomes and ethnic groups were equally subject to the disease; Reeves’ study was biased towards whites.

Eventually the CDC did change its estimate to be closer to Jason’s. A year or so ago, I noticed that 900,000 suddenly appeared on the website as the official estimate of the number of CFS patients in the U.S. However, every article Reeves has published has used the Wichita group, drawn from the misspecified model.

Dr. Reeves' more recent research has focused on creating questionnaires that could be used to diagnose CFS, as well as the intensity of the illness. [“Psychometric properties of the CDC symptom inventory for the assessment of chronic fatigue syndrome,” Nisenbaum, Heim, Jones, Unger – and Reeves. Population Health Metrics, 2005]. He never proved that they could diagnose CFS – to the contrary, using these questionnaires, he pulled in so many different populations that he began saying that there was no difference at all between Gulf War Syndrome and CFS, and many other “fatiguing illnesses.”

Not by coincidence, two of the “fatigue questionnaires” relied on similar questionnaires developed by Simon Wessely [et al] from the Journal of Psychosomatic Research [1993, 2004]

Then last April Dr. Reeves announced the results of the CDC’s genome study. The results were a bit premature – and as it turns out, from a scientific standpoint hardly robust. The scientists all remarked on the degree of heterogeneity in the population.

When Dr. Reeves gave his report on the genome study, he announced to my surprise that, “People with CFS do not have impairment in cognitive function,” and “Sleep abnormalities are not associated with CFS.”

At that moment, I knew that the majority of people in this data set did not have the same disease that I did. What was going on here?

Dr. Reeves is no longer using Fukuda. He is using his own method for establishing a population on which to base the world’s research on Chronic Fatigue Syndrome. And he has been telling us this all along – in his own way.

The new definition was announced in a backwards fashion. We know that the patients in the genome study were chosen from the Wichita study – but others also were given the questionnaires that Dr. Reeves has created to define Chronic Fatigue.

As I pointed out last fall, these same questionnaires proved to be disastrous failures in diagnosing patients with CFS or charting improvement when used in the Ampligen double-blind studies.

I have never been a fan of the Fukuda definition, mainly because cognitive dysfunction and post-exertional fatigue are the hallmarks of this disease, and it is possible to be diagnosed with CFS-Fukuda without having either. Nevertheless, there was an effort to include physical symptoms.

But Dr. Reeves told us in April that patients with CFS do not have cognitive dysfunction. (They are just tired.) Nor did they have problems with sleep. At least, not anything that would distinguish them from other patients with "chronic fatigue."

He had just announced to us all that two of the eight possible physical symptoms of CFS according to Fukuda were meaningless. And in doing so, he had told us he was now using a different definition for CFS.

And we were silent.

Thus, we come to where we started. Chronic Fatigue Syndrome, a disease name and definition created to replace the misdiagnosis of Chronic Epstein-Barr Syndrome that was first applied to the victims of the Incline Village, Nevada, outbreak, has clearly outlived any possible usefulness to scientists, doctors, or patients. There are too many ways to define this illness.

We began with the Holmes definition. Then we had the Fukuda definition. Soon there was the Wessely-Sharpe definition (roughly comparable to the CDC’s old definition of "chronic fatigue"). In Australia, there was the Lloyd-Hickey definition (that patients had once been sick and then “forgot” how to be well), and now we have the Reeves definition – as it turns out, quite similar to Wessely’s. Or perhaps we should just call it “chronic fatigue” - which is not chronic fatigue syndrome.

To use the phrase I have seen most often in referring to the Australian and English research on "CFS", Dr. Reeves is now using a biopsychosocial model of the disease.

Instead of reducing the degree of heterogeneity that was a problem in CFS-Fukuda – Reeves has dramatically increased it. No wonder the scientists in the genome study were so puzzled.

The result of the Wessely definition in England has been studies that “prove” that graded exercises (GET) and cognitive behavior therapy (CBT) “cure” chronic fatigue syndrome. Wessely and Sharpe were able to accomplish this by diluting the population pool so much that no more than 8 percent of the patients in the study actually had CFS-Fukuda – and probably fewer had Myalgic Encephalomyelitis, the real name of the disease that occurred at Incline Village, Nevada – and the disease that I have. In fact, the few patients in the study with severe cases of Myalgic Encephalomyelitis were unable to complete the study (a piece of information conveniently left out by Dr. Sharpe at the time, but later confirmed by the “25-percenters” in England).

We can expect similar results in the United States.

So let me give you just one example of the consequences of the Wessely-Reeves version of CFS.

Last year, a 32-year-old woman who had been ill with ME-CFS for years lost the ability to swallow. Because the disease is seen as psychosomatic in England, she was taken to a mental hospital where they tried to persuade her to eat (being locked in isolation was one of the methods of persuasion). She came home two weeks later, with the psychiatrists all expressing “puzzlement” as to why she would not “cooperate.” Sophia asked her mother to find a doctor who would not send her back to the psychiatrists again, but her mother could not.

Unable to swallow, Sophia finally lost her battle to live.

There was a formal inquest, and the conclusion was that Sophia died of “acute aneuric renal failure due to dehydration arising as a result of CFS.” The neuro-pathologist at the inquest reported that there was "clear inflammation of the dorsal root ganglia." The doctors at the inquest could not decide whether to say M.E. or CFS – but concluded that CFS was the “modern” term for M.E., so the conclusion was that she died of CFS.

She did die of CFS. She died of the misunderstandings caused by the biopsychosocial (or psychosomatic) school headed by Simon Wessely. Had she been diagnosed with M.E. (or even CFS-Fukuda), her inability to swallow would have been taken not as a sign of refusing to work towards her own recovery – but what it was: the dramatic result of years of severe illness. She would have been placed on a feeding tube, and she would be alive today.

Dr. Reeves already stated in the last meeting that he thought trauma in childhood could be a probable cause of CFS later in life (because it would reduce the body’s immune defenses). I hate to think what the results of that will be when school systems start applying it to parents of sick children.

But I can now see why he introduced the new Reeves definition of CFS in secret. It is not a secret any more, and we are all left with the consequences. I ask this committee to face them head-on.

I ask once again that we go back to the real disease definition – Myalgic Encephalomyelitis. We can also use the Canadian Consensus document for ME-CFS, so clinicians can learn how to diagnose and treat this puzzling illness.

Jumbling everything together into the vague useless category of “fatiguing illnesses” will not help anybody get well. All it can do is please the insurance lobbyists on K-Street, and a government wholly lacking in compassion for those less fortunate.

The job of the CDC is to understand the current threat of serious illness and communicate that threat to the citizens of the United States. Dr. Reeves has failed in that mission. I can only hope that this committee will do better.

Mary M. Schweitzer, Ph.D.
Presentation to the Chronic Fatigue Syndrome Advisory Committee
U.S. Department of Health and Human Services
July 17, 2006

Go to the CFIDS/M.E. Information Page
Mary Schweitzer's Essays on CFIDS/M.E.