Rainbow at shoreline

The Massachusetts ME/CFS & FM Association, a 501(c)3 founded in 1985, exists to meet the needs of patients with ME (Myalgic Encephalomyelitis), CFS (Chronic Fatigue Syndrome) or FM (Fibromyalgia), their families and loved ones. The Massachusetts ME/CFS & FM Association works to educate health-care providers and the general public regarding these severely-disabling physical illnesses. We also support patients and their families and advocate for more effective treatment and research.

Videocast of CFSAC Meeting of May 2010

The Chronic Fatigue Syndrome Advisory Committee (CFSAC), representing a number of major federal health agencies, provided study updates and new recommendations at a one-day meeting on May 10, 2010.  This was also an opportunity to become acquainted with Dr. Elizabeth Unger, the new Acting Chief of the Chronic Viral Diseases Branch (CVDB), at the CDC.

Read more: Videocast of CFSAC Meeting of May 2010


Kenneth Casanova's letter to DSM-V Committee of the APA

The introductory explanation text of CSSD in the Draft unfortunately lacks the requisite scientific rigor and specificity for medical and psychiatric differential diagnosis.

The CSSD diagnostic criteria in many instances would reasonably diagnose a percentage of patients: such patients would be abnormally concerned/preoccupied with actual medical symptoms, over-interpretation of bodily sensations, or the somatic projection of ideational content—to the point where such processes become pathological. The example of the true hypochondriac or the patient who easily somatizes feelings would validate a portion of the CSSD definition.

However, at the same time, the CSSD criteria is so broad that it draws no clear boundary between the patient responding within normal expectations to an actual medical condition, and patients who are pathologically misapprehending or excessively concerned. By unscientifically conflating two major groups of patients, the draft criteria must result in a substantial number of cases in which reasonable and appropriate patient responses to actual physical illness are falsely psychologized. Such a lack of diagnostic clarity creates an amorphous and contradictory criteria for misdiagnosis—with severe consequences for patient suffering and possible medical malpractice.


Specifically flawed CSSD diagnosis

The essence of CSSD is to have one severe physical symptom or multiple physical symptoms that are chronic (at least 6 months) and about which an individual either has misapprehended as to its causation or is excessively concerned about or preoccupied with (beyond a realistic viewpoint).

Following the critique above as to the difficulty with the criteria:

A person may be fully diagnosed with only the following elements of the definition:

(A.) Multiple somatic symptoms or one severe symptom that have been

(B.) chronic and persistent for at least six months, and

(C.) create a high level of health anxiety and establish a central role in the patient's life for health concerns.

Can anyone doubt that such a minimal definition could theoretically diagnose anything from true hypochondriasis to severe rheumatoid arthritis, medication resistance epilepsy, to the pain of severe radiculopathy, to drug resistant pelvic inflammatory disease, to a brain tumor, to Lou Gehrig's disease, to neurofibromatosis, and to many other chronic illnesses. Can such an unscientific and medically questionable diagnostic criteria be contemplated?

Another example may be the early stages of MS: In its early stages MS is difficult to diagnose—in fact decades ago, many physicians believed MS was a psychiatric syndrome.


Special problems with physiologically-induced pain disorders

A very serious red flag is raised in the actual criteria: " XXX.3 Pain disorder. This classification is reserved for individuals presenting predominantly with pain complaints who also have many of the features described under criterion B." Criterion B requires that two of five conditions be met. B's conditions would be fulfilled if the patient experienced a "high level of health-related anxiety" and that "health concerns assume a central role in their lives". Medicine currently has come to realize that pain itself can no longer be relegated to the periphery of clinical concern and should no longer be waved off as a "mere symptom"—but should be fully investigated.

Pain is a legitimate medical symptom and is now recognized at the "fifth vital sign" to be evaluated by physicians, along with blood pressure, pulse, respiratory rate and temperature. Untreated pain can become very detrimental to a person's health as well as very disabling. Pain, in the joints, sore throat (chronic mononucleosis, a known medical diagnosis), and other conditions are all too often psychiatrically dismissed. The misleading nature of the diagnosis of pain in the draft CSSD definition can have serious consequences.

Clearly the CSSD disorder is not only theoretically flawed and disparate, but as a practical methodology, it is a potential minefield for medical and psychiatric practices and the patients seeking their assistance.


Language in the Introduction and subsections of CSSD definition is open to misinterpretation

The Draft explanation of Somatic Symptom Disorders both in its Introduction and subsections clearly demonstrates the lack of precision and the resulting conflation of two disparate medical phenomenon. In the explanation of Complex Somatic Symptom Disorder some selection of text will show the difficulty:

"The hallmark of this disorder is disproportionate or maladaptive response to somatic symptoms or concerns." Obviously, the word "disproportionate" is a matter of degree or "portion". The determination of degree cannot be entirely objective, and in cases of actual medical conditions, normal patient response varies across a wide range of factors, including personal, economic, occupational, family, etc. circumstances.

"Patients typically experience distress and a high level of functional impairment." Such a statement is perfectly consistent with a number of medically understood illnesses, and therefore in the problematic context of the CSSD criteria can be disorienting and misleading. "Sometimes the symptoms accompany diagnosed general medical disorders..."

"There may be a high level of health care utilization..." No experienced specialist or general physician is unaware of cases in which patients have had to see five or more doctors before receiving an accurate diagnosis—especially with the more difficult to diagnose illnesses. Endocrine, hematological, circulatory, occult pulmonary conditions come to mind.

"In severe cases, they may adopt a sick role." Now the concept of the "sick role" may infrequently constitute a distinctly categorical "role-type" that is pathological and somewhat separable from a real physiological illness. However, in many chronic illnesses, whose symptoms wax and wane in severity—it would be more accurate to say that the person is chronically sick. Undoubtedly, different individuals or even the same individual will adapt or respond variously, with an attitude of courage, hopefulness, worry, or even despair in different times or circumstances. However, to label such common variations as a "sick role" can often be too superficial and facile—a false engagement in type-casting. To be sure, many patients who are chronically ill need intelligent counseling in coping and in modulation of their attitude and emotions. Hopelessness can creep in and assistance is needed—but to label as a psychiatric disorder a normal spectra of physical disorder with emotional and mental sequelae is a distortion. Again, in some cases the viewpoint is accurate, but in too many others a distortion with consequences.

"Some patients feel that their medical assessment and treatment have been inadequate." In some cases, this statement reflects an adequate further description of a psychiatric problem. In other cases, the statement demonstrates a failing in the criteria.

Again, the dual nature of the criteria is reflected in the following wording: "Patients with this diagnosis typically have multiple, current, somatic symptoms that are distressing...
The symptoms may or may not be associated with a known medical condition. Symptoms may be specific...or relatively non-specific (e.g., fatigue or multiple symptoms.)
"

Note: The classification or facile diversion of fatigue to the psychological realm can be a very medically dangerous undertaking. A multitude of serious medical, and currently poorly understood biological conditions, manifest fatigue as an early and chronic symptom.

"... Such patients often manifest a poorer health-related quality of life than patients with other medical disorders and comparable symptoms." Unfortunately this statement represents perhaps the nadir of scientific thinking in the entire statement, and therefore puts in relief the lack of rigor which proceeds and follows it. Yes, patients with one medical disorder will often have a poorer quality of life than those with another medical disorder.

In the Introduction to this section in the Draft, there is some clarity in attempting to set a line between the pathological and normal response to medical illness: "Having somatic symptoms of unclear etiology is not in itself sufficient to make this diagnosis. Some patients, for instance with irritable bowel syndrome or fibromyalgia would not necessarily qualify for a somatic symptom disorder diagnosis." But looking underneath the text raises questions: is the diagnosis of fibromyalgia itself uncertain; or alternatively, is the question the addition of CSSM to some cases of fibromyalgia? By what criteria would CSSM be added: what is a within the range of normal varying responses to a chronic illness, and which responses would add a psychiatric diagnosis? The criteria leave these questions open.


Incorporation changes from DSM-IV in CSSD

Another major aspect of the new CSSD criteria is its departure from the various qualifying distinctions contained within the several previous diagnostic categories it replaces.

These previous diagnoses, to be obliterated and incorporated within the more diffuse CSSD, are:

  • Somatization disorder
  • Undifferentiated Somatoform disorder
  • Hypochondriasis
  • Pain Disorder Associated with Both Psychological Factors and a General Medical Condition
  • Pain Disorder Associated with Psychological Factors.

A major question is: what is lost, if anything, in the "lumping" of the older conditions? Moreover, what, if anything, that is lost provided more rigorous procedures for making more accurate diagnoses—or at least less inaccurate?

Somatization disorder requires a history of many physical complaints before the age of 30. The new CSSD throws this qualification overboard. Why the change? Has the historical finding, which has counted as a distinct marker, evaporated?

A second major change is that fatigue, a symptom highlighted in the statement about CSSD, is specifically stated as not a symptom found in somatization disorder. This issue of fatigue directly impacts the differential diagnosis between the proposed CSSD definition and the physiological, multi-systemic illness of CFS—also known in Europe as myalgic encephalopathy or myalgic encephalomyelitis. Somatization disorder would be hard to confuse with CFS, for instance sleep disorder and decreased concentration are not physical symptoms included in the diagnosis of somatization disorder. Also in somatization disorder, head, joint and possible muscle pain are the only stated symptoms in common with those of CFS/ME. Yet the CSSD criteria, with a psychological not a medical interpretation will provide a diagnosis of CFS/ME.

Eliminating distinctions of somatization disorder negates distinctions that must have taken years to discriminate.

A second diagnostic criteria transformed/lumped into CSSM is Undifferentiated Somatoform Disorder. This diagnosis instead of simply relying upon multiple somatic symptoms (CSSD) actually group specific symptoms necessary for diagnostic fulfillment.

These include: "One or more physical complaints (e.g., fatigue, loss of appetite, gastrointestinal or urinary complaints) which either 1) the symptoms cannot be fully explained by a known general medical condition, or 2) when there is a medical condition are excessive in relation to the condition." In this condition, as in CSSD, much of the differential diagnosis depends on the interpretation of the individual physician—whether medical or psychological or both.

Yet the new CSSD definition simply widens further the amount of undifferentiated territory. Is negation of diagnostic detail supportable, and again what are the practical consequences?


Issues related to possible coordination of DSM-V with CDC publication of ICD-10-CM and WHO ICD-11

There is discussion that: "The APA [the American Psychological Assn., the sponsor of the new DSM-V] has already worked with the CMS [U.S. Centers for Medicare and Medicaid Services] and CDC to develop a common structure for the currently in-use DSM-IV and the mental disorders section of the ICD-10-CM."

The ICD-10-CM, overseen by the CDC, will be a revised coding system in the U.S. for all diseases and conditions. This coding system includes disorder names, logical groupings of disorders and code numbers. The new ICD-10-CM will contain codes for all Medicare and Medicaid claims reporting. The ICD-10-CM is scheduled to be published Oct. 1, 2013.

Currently the coding system in the U.S. is the ICD-9-CM. The U.S. Coding System is separate from the international coding system under the auspices of the World Health Organization. The WHO system is currently the ICD-10. WHO will be revising its system in 2014 to the ICD-11.

Concerns about confusion of CSSD with CFS/ME and fibromyalgia in DSM-V, ICD-10-CM and ICD-11

CFS/ME and fibromyalgia medical researchers, patients and patient organizations are rightly concerned that flawed CSSD definition will adversely affect CFS/ME and fibromyalgia research and clinical care through the application of DSM-V; as well as any coordination of the CSSD diagnosis with the coding of CFS/ME and FM in the ICD-10-CM and ICD 11.

The real issues for CFS/ME and FM are two.

If the CSSM diagnosis appears in either of the new ICD codings, there are two possibilities. First, by itself, the new CSSM diagnosis would be more confused with CFS/ME and FM than any of the DSM-IV diagnostic categories.

Second, what would be the influence of the diagnostic category of CSSD in the direct categorization of CFS/ME and FM in both the new WHO definitions and the U.S. definitions. The categorization of CFS/ME and FM could be directly applied to the CSSD definition; or alternately be redefined, detrimentally, in other WHO ICD or CDC ICD categories. Certainly, cooperation of the APA, WHO and CDC is expected and very useful—except when a flawed category is shared.

Moreover, the direct categorization of CFS/ME historically both in the WHO ICD-10 and the CDC ICD-9-CM (both current) must be noted. WHO has been clearly the more medically progressive and accurate. The ICD-10 since 1990 has listed CFS/ME under G93.3 "neurological disorders". During the same period of time through to the present, the CDC has instead listed CFS under R53.82 under the general category of Symptoms, Signs and Ill-Defined Conditions as Chronic Fatigue Syndrome (780.71). Many efforts have been made to get the CDC to reassign CFS/ME to the neurological section, but the CDC has resisted. Under the U.S. system CFS/ME has been listed as a vague syndrome as opposed to a defined disease entity, thereby undermining its medical credibility.

How will the APA's new definition of CSSD—which could misdiagnose CFS/ME—influence the CDC's publication of the new U.S. ICD-10-CM?

CFS/ME and FM patients and organizations sincerely hope that the APA will be mindful of the detrimental effects that the flawed CSSD category could have on the ICD codings.

All new ICD-10-CM coding categories will be mandatory for reimbursement for Medicare and Medicaid and are also widely used by private insurance companies. A flawed classification of CFS/ME and FM in any of the new systems—DSM-V, ICD-10-CM or ICD-11—will have both medical system access consequences, as well as diagnostic ramifications, that could place greater focus on CFS/ME and FM as psychiatric disorders as opposed to a medical/biological disorders.


Conclusion

CFS/ME and FM through intense medical research over the past 20 years have been demonstrated to be complex, multi-systemic, biological illnesses. The illnesses follow from initial infectious or toxic triggers and involve dysregulation of the multiple body systems, including the immune, nervous, endorcrine, cardio-vascular systems. Certain viruses have long been implicated. Genetic and genomic factors are being elucidated.

Dr. Anthony Komaroff, Professor of Medicine at Harvard Medical School and long-time researcher in the field has said the following: "there are now 4,000 published studies that show underlying abnormalities in patients with this illness [CFS/ME]. It is not an illness that people simply can imagine that they have and it's not a psychological illness. In my view, that debate, which has been wage for more than twenty years, should now be over."

The diagnosis of CSSD is flawed in and of itself, in its application to a variety of medical illnesses and specifically to CFS/ME and FM.

Sincerely,

Kenneth Casanova

Dr. Alan Gurwitt's letter to DSM-V Committee of the APA

The Board of Directors of the Massachusetts Chronic Fatigue and Immune Dysfunction Syndrome/Myalgic Encephalopathy and Fibromyalgia Association (Mass. CFIDS/ME &FM) has reviewed your proposed revisions of DSM-IV destined for DSM-V. Our particular focus has been on your newly proposed category "Complex Somatic Symptom Disorder". Our concern is not so much about the fact that you wanted to simplify terms for somatoform disorders but about the criteria cited and the potential misuse of the category.

We share the concern, heard from individuals and organizations around the world, that this new category might be too readily used to include chronic fatigue syndrome ("CFS", also now known as ME or myalgic encephalopathy) and fibromyalgia as if they are forms of a psychological disorder. Most simply, they are not now psychologically caused illnesses nor have they ever been so. This concern doesn't come out of the blue. It is based on two and a half decades of mistaken and harmful misdiagnoses based on totally inappropriate and harmful misconceptions of what CFS and fibromyalgia are and what they are caused by.

Our organization, celebrating our 25th year of existence helping patients with these illnesses, has seen firsthand the terrible toll exacted by the trail of misdiagnoses. Patients are tainted, dismissed, not properly treated, and often referred to equally misinformed mental health clinicians.

In our view, the key problem is not so much the diagnostic nomenclature as it is the very wrong conceptions on the part of psychiatrists and non-psychiatric physicians of the very nature and likely causes of both CFS/ME and FM. There may be psychological disturbances following the onset; these are important but they are secondary.

Over the past fifteen years, increasing numbers of researchers from around the world have pinpointed the likely biological causes, the complex pathophysiology that follow from the initial infectious or toxic triggers, and the interacting and dysfunctional multiple body systems (immune, central and peripheral nervous systems, endocrine, cellular [mitochondrial], etc) involved. Genetic and genomic factors are being elucidated. Certain viruses have long been implicated. Most recently a retrovirus, XMRV, has been implicated and is actively being studied in several centers. While no widely accepted biomarkers are currently available, many key researchers believe that it will not be long before one or more biomarkers will be found. While there is no definitive cure as yet, forms of treatment have been developed over the years that can alleviate many of the symptoms. Here is where mental health clinicians can help; if there are serious secondary psychological symptoms, certain therapies can help.

Dr. Anthony Komaroff, Professor of Medicine at Harvard Medical School and a long-time researcher in the field has said the following:
"...there are now over 4,000 published studies that show underlying biomedical abnormalities in patients with this illness. It's not an illness that people can simply imagine that they have and it ‘s not a psychological illness. In my view, that debate, which has waged for twenty years, should now be over". Four years later there is even more evidence for Dr. Komaroff's assertion.

The bottom line is that CFS and fibromyalgia are not psychological illnesses. It is then essential that the American Psychiatric Association vigorously help educate graduate and resident psychiatrists on this essential truth. While the wording and criteria for "Complex Somatic Symptom Disorder" will matter so as to avoid confusing chronic fatigue syndrome and fibromyalgia with a somatoform disorder it will be new understanding of the biological nature, proper diagnostic techniques, and appropriate treatments of these illnesses that will matter most.

Alan Gurwitt, M.D.

President, Massachusetts CFIDS/ME & FM Association
(Retired adult and child psychiatrist, Distinguished Fellow of the American Psychiatric Association)
4/15/10

Notice about names

The Massachusetts ME/CFS & FM Association would like to clarify the use of the various acronyms for Chronic Fatigue Syndrome (CFS), Chronic Fatigue & Immune Dysfunction Syndrome (CFIDS) and  Myalgic Encephalomyelitis (ME) on this site. When we generate our own articles on the illness, we will refer to it as ME/CFS, the term now generally used in the United States. When we are reporting on someone else’s report, we will use the term they use. The National Institutes of Health (NIH) and other federal agencies, including the CDC, are currently using ME/CFS. 

Massachusetts ME/CFS & FM Association changed its name in July, 2018, to reflect this consensus.