AHMF

Summary of Selected Presentations AACFS Conference Seattle Washington, Jan 26-29 2001
Ellie Stein MD

What follows is a brief summary of the presentations which in my opinion were the most interesting or important, either because they showed something new, or because they disproved assumptions which have been hindering the acceptance of CFS as a serious disorder worth studying. It is not intended to be complete, and many presentations and posters are not mentioned. The summaries are filtered through my personal memory and judgment, both of which are liable to error.

After each author's name there may be references other important papers by the author. The bibliography of these papers is at the end of the summary.

This summary is intended for consumers. In addition to summaries of the presentations there are some explanations of medical terminology and the relevance/importance of some technical papers .

Topics covered in this summary include:

Epidemiology
Autonomic Function
Biochemistry
Immunology/Infection
Genetics
The Brain
Service Delivery/Physician Attitudes
Non-Medication Treatment
Medication Treatment
Clinical Discussion
References

Part I: CDC Committee and Research Update

This pre-conference half-day hosted by the CDC was disappointing given the amount of money spent and misspent ($US 13 million misspent and 4 million missing) that the CDC is still at the stage of asking, "What is fatigue?" and, "How can we measure it?" No further comment.

Part II: Research and Clinical Presentations

Lea Steele, San Franscisco (Steele et al, 1998)
Studied American Gulf War veterans to discern whether CFS and GWI are related or the same illness. The incidence of CFS among GW vets was much greater than in the general population (7% vs. .4%) however the prevalence of CFS doesn't vary with any of the factors that predict the severity of GWI eg. branch of service, location of service and timing of deployment. Also, the symptom pattern was slightly different, with GWI having more joint pain, diarrhea, night sweats and skin rashes than CFS patients. This suggests that GWI and CFS are not identical.

Rosane Nissenbaum, CDC in Atlanta Georgia
Described the Wichita Kansas population study funded by the CDC. Population studies are important because they are free of the self selection biases which occur when studies are done on patient groups. ie. patients who go to their doctor are probably not representative of all patients with a disorder. Through telephone interviews of 33,000 randomly selected households subjects were identified who had CFS-like illness and CF not meeting the full CFS criteria. These people were followed up at one, two and three year intervals. There was only 30% stability over the first year meaning 30% of the subjects with CFS-like illness still met full criteria after 1 year. Interestingly, many people said that they were improved, but reported no change in their hours per week of work and home duties, suggesting they were not entirely well. Nevertheless, these data suggest that population derived samples have a higher rate of spontaneous recovery than clinical samples in which recovery is generally less than 20% over 4 years.

Leonard Jason, Chicago (Jason et al, 1999;Jason et al, 1997)
Described subgroups of patients from the Chicago epidemiological study which used similar methodology to the Wichita study. The overall prevalence of CFS was found to be 0.4%. However, women, minorities and the unemployed had a higher prevalence, and more severe symptoms, than men, caucasians and employed. Lifetime psychiatric history did not correlate with any severity measures. Those subjects with concurrent psychiatric disorder had greater fatigue and social disability than those without concurrent psychiatric disorder. This disproved the long-held myth that minorities and the poor don't get CFS, and that psychiatric disorder is an important variable in CFS.

Pascale DeBecker, Brussels Belgium (De Becker et al, 1998)
Reported on the largest well-studied patient group in the world. Each of 1500 patients seen at Dr. Kenny DeMeirleir's CFS clinic at the Free University of Brussels has been extensively assessed. Each patient met the 1988 and/or the 1994 diagnostic criteria and completed a severity checklist of 50 symptoms. A factor analysis was done to see which symptoms covaried, i.e., were found together more often. The beauty of this model is that it is done by computer without any bias from the researcher. They found 4 symptom clusters: general and infective symptoms, neurocognitive, musculoskeletal and psychiatric. The first three clusters were strongly associated with whether the patient met the CFS criteria and with maximum fitness on exercise testing. The psychiatric cluster didn't seem relevant either to diagnosis or fitness suggesting it is not a core aspect of CFS. This is a strong argument against CFS being a "psychosomatic" or "functional somatic" disorder. DeBecker suggested 10 additional symptoms which should be added to the CFS diagnosis based on this research.

Katherine Rowe, Melbourne Australia (Rowe, 1997)
Did a factor analysis similar to DeBecker's on 149 PWCs who became ill before the age of 18. Of the 38 symptoms used from an adult checklist, 24 were commonly endorsed in the young patients. She found a 5 factor solution ie. there were 5 clusters of symptoms: muscle pain/fatigue, neurocognitive, abdominal/head/chest pain, neurophysiological and immune/infective. Furthermore, she did structural equation modelling to see the direction of effects between symptom groups. The immune group seemed to occur first and other symptoms resulted from this. She also found three discrete groups based on severity. As an aside, Dr. Rowe reported that young patients were able to give accurate and discriminating symptom reports; they did not answer 'yes' to everything or 'no' to everything.

Stanley Schwartz
Followed up a group of CFS patients with and without depression, and a group of depressed patients, over 4 years. All patients received "conventional" treatment. He found that in both groups the depressive symptoms were much more responsive to treatment than the CFS symptoms, and that the patients with CFS and depression had a better outcome than those with depression alone.

Daniel Cukor/Lana Tiersky, New Jersey (Tiersky et al, 1997)
Tested the often proposed hypothesis that CFS is a form of somatization disorder because patients who complain of more physical symptoms also have more psychological symptoms. In this study of 145 PWCs diagnosed using the 1994 criteria there was no correlation between the number of physical symptoms and the number or presence of psychiatric diagnosis. In addition, other variables such as age of onset in CFS was different that that typically seen in somatization disorder. They conclude CFS has no relation to somatization disorder.

Patricia Soetekouw, Belgium (Soetekouw et al, 1999)
Confirms the presence of subtle but important differences between subjects with CFS and healthy controls. CFS subjects asked to stand for 5 minutes after lying down had poorer cardiac response to the stress of standing and had more symptoms during standing.

James Baraniuk (Baraniuk et al, 1998)
Studied autonomic function in the nose in an interesting way. Subjects were asked to do a hand grip exercise. Normally exercise causes increased lactic acid which activates the blood vessels in the nose to constrict so that more blood can go to the muscles and correct the acidity. When the blood vessels in the nose constrict there is less inflammation and mucous and airflow increases. This effect was missing in subjects with CFS but was present in healthy controls. Baraniuk feels the cause of much nasal congestion and dripping could be autonomic dysfunction.

Julian Stewart, NY (Stewart et al, 1999)
Argues that the term orthostatic intolerance is a misnomer because abnormalities are detectable even lying down. He is a pediatric cardiologist who studies Postural Orthostatic Tachycardia Syndrome (POTS). He hypothesizes and has some research evidence to support that POTS is due to a pooling of blood in the periphery especially the legs. This is not generally due to abnormalities of the veins but rather due to failure of arterial vasoconstriction. Capillary permeability seems normal but venous emptying may be imparied.

Patrick Englebienne, Brussels Belgium (in my personal opinion the most important paper at the conference)
Described an elegant experiment which connects the RNAse-L and channelopathy hypotheses. Englebienne works with De Merileir, De Becker and colleagues in Brussels.

Review of RNAse-L
This is an enzyme produced by white blood cells which are challenged by certain viruses and possibly also some toxic exposures. The enzyme breaks down the viral RNA and also destroys the infected cell so that the virus has nowhere to live. Suhadolnik first discovered in 1995 that persons with CFS had elevated levels of an abnormal type of this enzyme (Suhadolnik et al, 1994). Instead of the normal size 80KDa enzyme subjects with CFS show a 37 KDa enzyme. This finding has been found also by the Belgian group which now commercially tests the ratio of the 37KDa and 80KDa enzymes (De Meirleir et al, 2000). They and an independent group in the US have found that a high ratio is associated with more severe clinical symptoms and decreased exercise fitness. De Meirleir et al have recently found that the 37Kda RNAse-L is associated with incomplete cell death. This means that the cell constituents cannot be recycled for use by other cells.

Review of the channelopathy hypothesis of CFS
'Channel' is the name given to the thousands of portals which control what enters and leaves a cell. Every cell has specific environmental requirements and controls the environment through channels, each of which is exquisitely specific and sensitive. If channels are blocked or leaky, the cell will not function optimally, and may even die. Poisoning by ciguatera fish toxin is an experimental model for channel dysfunction. It kills cells by incapacitating the channel which controls sodium and potassium concentrations inside and outside of cells. Causes of channelopathy include toxins (eg. ciguatera, DDT), antibodies against the channels, and prolonged stress.

Chaudhuri suggested that CFS is a channelopathy because it shares many similarities with disorders that are known to be associated with abnormal channel function (Strickland et al, 1998). These similarities include: variability of symptom expression, either excess or loss of function, triggers such as environmental stresses (e.g. smell, food, exercise, infection), and neurologic involvement. To date there has only been indirect evidence to support this theory. Chaudhuri and Behan have shown that PWCs use more energy at rest than healthy controls . Burnett in Adelaide has shown that some PWCs have decreased total body potassium and abnormal potassium response to exercise. One of the most important channels is the one that controls sodium and potassium concentrations; this channel alone takes up to 40% of the body's energy. Therefore even a small anomaly in this channel function could represent a significant energy drain.

In the body every molecule has a regulatory system, usually something that binds to it, changing the shape and thereby turning it on or off. For RNAse-L, the 'brakes' are applied by a protein called RNAse-L inhibitor protein (RLI). Englebienne reports that RLI looks very similar to a family of channels called the ATP binding casette (ABC) family. As a result, the 37 KDa variety of RNAse-L, which is missing some of the usual regulator areas, binds by mistake to the ABC channels. These channels include (you guessed it) the sodium/potassium pump that Chaudhuri thinks is involved in CFS. Other members of the family are found in red blood cells, immune function, the uptake of tryptophan (which is necessary for sleep and mood), can cause sensitivity to toxic chemicals and pain, and are known to be involved in neurological disorders—all possible mechanisms of CFS symptoms.

Howard Urnovitz (and Paul Cheney)
Followed up on a finding in Gulf War vets of RNA fragments found in the blood. It turns out this RNA is from the gene which makes antibodies. In order for the body to make antibodies to any infection or foreign molecule it has to be a very flexible gene, able to recombine and match the shape of invaders. Perhaps because of this ability to change, these areas of DNA are also susceptible to being permanently changed by outside toxins such as infections, vaccines, radioactivity and pesticides. In other words, a number of different events could cause similar genetic mutations and thus similar symptoms. This group is hypothesizing that certain of these RNA molecules may be specific markers of CFS, but more evidence is needed.

Kevin Maher
Found that PWCs had lower levels of perforin in their natural killer (NK) cells. NK cells seek out and kill cells which are infected or cancerous. Perforin is one of the enzymes which is responsible for cell death. Mice bred not to produce perforin generally die young of infection and show increased immune activation similar to subjects with CFS. Low perforin could explain the finding by other groups that PWCs have low NK cell activity.

Eng Tan
Reports finding high levels of antibodies to tubulin, a protein found in cells. If tubulin were attacked it would have serious effects on cell function.

Kenny De Meirleir, Belgium
Informally reported on his lab's experience with PCR testing for mycoplasma. 69% of PWCs are positive for mycoplasma compared with 3% of healthy controls. RNAse-L ratios are elevated in those PWCs with mycoplasma compared with PWCs who test negative. Interestingly, 67% of a sample of people living in Bijlmer who became ill after a plane crash occurred in the area are positive for mycoplasma. They have a CFS-like syndrome plus dry skin. They hypothesize that treating mycoplasma when present will lower the infective load and help the immune system recover. However this view was strongly criticized by others who pointed out that PCR detects DNA sequences and doesn't actually indicate the presence of live virus. Therefore they caution against the risks of antibiotic treatment in the absence of evidence of active infection.

Leslie Aaron, Richard Harrell, David Lewis
An amazing twin study was reported involving 65 pairs of twins in which one twin had CFS and the other didn not. By comparing monozygous (identical) twins and dizygous (fraternal) twins, one can estimate the relative contribution of genetics and environmental causes. For example, if CFS were 100% genetic, 100% of identical twin pairs and 50% of fraternal twins (who only share 1/2 of their genes) will be the same with respect to whether they have CFS or not. Things like life experiences, infections, stress etc wouldn't make any difference. Conversely, if a disorder has no genetic components, then there will be no difference between the rates of concordance in identical and fraternal twins. In this study the concordance rate for identical twins was 55% and for fraternal twins was 19%. This suggests that both genetic and environmental factors contribute to CFS.

Roderick Mahurin, from Dedra Buchwald's group in Seattle
Did a study to assess whether there is any objective evidence to support subjective reports from PWCs that mental tasks require more energy than they did prior to illness. They used SPECT scans to measure brain blood flow at rest and during a difficult arithmatic/memory task. The blood flow pattern of the PWCs was similar to that found in healthy controls doing a more difficult task. This validation of subjective effort adds to the evidence against PWCs being overly sensitive to effort, i.e. neurotic.

Richard Gracely, from Daniel Clauw's FM research group
Pain, like fatigue, is a subjective symptom which is hard to validate. This has led to skepticism regarding whether people with CFS/FM are experiencing "real" and significant pain. If the skeptics read this study they may have to reconsider. Functional MRI measures neuronal function. f-MRI scans of the brain were done while light and then intense thumb pressure was being applied. This allows detection of which brain areas are activated with pressure/pain. The FM patients showed similar brain activation patterns shown by healthy controls but at lower pain levels. Like the previous study, this adds to the objective evidence of lower pain thresholds in FM.

Gijs Bleijenberg, The Netherlands
Completed surveys of physicians in 1993 and 1999. The number of GPs who "never" make a diagnosis of CFS has decreased from 27% to 13%. However, in practice, when asked what they would do when presented with a potential PWC, 78% refer to a specialist rather than, or before, making a diagnosis of CFS. This is complicated by the patient report that 21% of GPs and 53% of specialists don't take CFS seriously. Physicians perceive that PWCs causes problems for physicians including: the need for longer consultations, more communication problems and increased "non compliance" compared to other patients.

Anthony Komaroff (Komaroff, 2000;Komaroff & Buchwald, 1991)
In his keynote address, Komaroff reminded physicians of 4 important points:
1. the patient will tell you the diagnosis
2. just because it is not written doesn't mean it's not true, be willing to seek evidence
3. psychiatric illness is real disease, it's not an issue of moral character or strength
4. never succumb to the temptation to blame the patient by saying "there is nothing wrong with you" when unable to make a diagnosis.

Leonard Jason, De Paul University, Chicago (Jason et al, 1997;Jason et al, 1995;Jason et al, 1999)
Experimentally tested effect of the CFS label on attitudes of medical students towards PWCs by giving them an identical case example with three different diagnostic labels: CFS, ME, and 'Florence Nightingale Disease'. Students rated those with CFS as more likely to have a correct diagnosis and more likely to improve. ME was associated with being thought to result from a physical cause, and to include candidates less suitable for organ transplantation. Given a typical CFS history and three different recommended treatments, Ampligen, CBT, and coping skill training, students assumed the subjects treated with Ampligen were more likely to have been correctly diagnosed and to be more severely affected. Only 20% of medical students, 40% of family practice residents, and 90% of psychiatric residents, had ever read an article about CFS.

Jos Van der Meer, The Netherlands (Bazelmans et al, 1999;Vercoulen et al, 1997)
Did the largest yet randomized blinded study of CBT in CFS. They randomly assigned 92 PWCs to receive 16 weekly sessions of CBT, 90 to receive 11 sessions of guided group support, and 88 to be untreated. They also tested each patient with an actometer to measure their actual activity continuously for 2 weeks. This device is worn on the ankle and measures all movement. Using measures of functional impairment, fatigue and overall wellness the CBT group did significantly better than the other two groups. Yet the rates of improvement were low, 50% in the CBT group and 35% in the other groups. They found that the presence of psychiatric diagnosis did not affect outcome. The least active patients (perhaps the most ill) at the start did the least well. This study is an advance over other CBT studies to date as they actually used the 1994 case definition for CFS which requires physical symptoms to be present. However it was not reported whether any of those symptoms improved with treatment.

Pat Fennell, Albany NY (Fennell, 1995)
Proposed a model of psychosocial support based on the stage of psychological adjustment to CFS. In each of four stages—crisis, stabilization, resolution and integration—different psychological tasks are required and different therapeutic approaches are necessary. She noted that psychological support for CFS differs from the Kubler Ross model because there is generally no death or end to the symptoms; rather, the PWC and family have to learn to cope, adjust, and find new meaning. She suggested that chronic illness can be a litmus test like the tablets which the dentist gives to see where one has missed brushing. Whereas healthy people can compensate for unresolved emotional issues, chronic illness tends to diminish this ability and preexisting problems come to light.

Nancy Klimas
Reports on preliminary results from an extraordinary study probably only possible at a major institution like the CDC. Patients with tender lymph nodes and immune inflammatory shift (Th2) had their lymph nodes excised, the cells cultured and forced in vitro towards a more favorable immune profile then reinjected. In 10 of 11 cases, the clinical picture and immune profiles are related to the procedure covaried suggesting that immune profiles do have an impact on clinical symptoms. As a group, the patients reported significant improvement in brain fog, and many improved in objective tests of cognitive function.

Kittil Rammohan
Reported on a trial of Modafinil (Provigil), a drug approved for treating narcolepsy in patients with MS, who have fatigue similar to that in CFS. This drug is used for treatment of daytime sleepiness, not nighttime sleep. At 200 mg/day more patients reported improvement in the Fatigue Severity Scale than those who were untreated or who were on a higher dose. There were very few side effects, suggesting it may be feasible to try this drug in PWCs with daytime sleepiness.

Katherine Rowe, Melbourne
Reported on a long term follow-up of children and adolescents treated 7 years ago with immunoglobulin. About half were entirely recovered, many reported improvement but continued fatigu,e and 7% had a severe course. In a telephone interview the former PWCs now in their 20s uniformly reported that receiving a diagnosis was useful and that delay in diagnosis was a problem. This refutes hypotheses that giving a diagnosis begins a self-fulfilling prophesy of continued illness. It also suggests that intravenous immunoglobulin may have utility in young patients.

David Bell, Lyndonville NY (Bell et al, 1994;Streeten et al, 2000)
Reported on the 15 year follow-up of 47 children who became ill with CFS at the same time in a small community, i.e. an epidemic outbreak. 37% were completely well, 43% were functioning well but had to make compensations for their health, 11% were still ill and 9% were severely ill or worsening. Those who were most ill initially and missed the most school had the poorest outcome. Some of the subjects commented that since they had been ill since childhood they found it hard to assess if they were "completely cured" or not.

if the RNAse theory is correct and the cascade is started by the enzyme calpain then calcium channel blockers may help because they inhibit calpain
tricyclic antidepressants and flexaril, a muscle relaxant which is related, help both sleep and pain, other sleep medications such as zopiclone (Imovane) only address sleep, not pain. There was agreement that serotonin antidepressants (eg. prozac, paxil, zoloft) are not generally helpful in CFS, at least for pain relief.
there was debate about whether treating orthostatic intolerance (OI) actually improves the overall CFS syndrome, some say yes, some say no. Therefore not all clinicians would treat with fluodrocortisone even in the face of clinical OI symtpoms eg. dizziness, pain, fatigue, nausea upon standing.
some clinicians felt their patients could achieve significant improvements by addressing each of the presenting symptoms. A poster was presented by Jacob Tietelbaum in which 30 of 33 of FM patients, most of whom also met criteria for CFS, improved using this symptomatic and laboratory testing approach, compared with 12 of 33 patients in the placebo treatment group. These patients were given placebo look-alikes for all treatments. Several doctors had the more nihilistic attitude that nothing really works, and as a result offer little symptomatic treatment.
despite anecdotal reports of improvement with bovine (from cows) growth hormone (GH) in selected CFS patients with low GH levels, treatment remains controversial because of side effects.
there was agreement that replacement corticosteroid, e.g. 10 20 mg cortisol daily, is not good because there is limited symptom improvement and potentially serious side effects such as adrenal suppression.
there was considerable debate about the usefullness of elimination diets, most alterative treaters advocated it and most conventional physicians disregarded it.

Baraniuk,J.N., Clauw,D.J., & Gaumond,E. (1998) Rhinitis symptoms in chronic fatigue syndrome. Annals of Allergy, Asthma, & Immunology, 81, 359-365.

Bazelmans,E., Vercoulen,J.H., Swanink,C.M., Fennis,J.F., Galama, JM, van Weel,C., van der Meer,J.W., & Bleijenberg,G. (1999) Chronic Fatigue Syndrome and Primary Fibromyalgia Syndrome as recognized by GPs. Family Practice, 16, 602-604.

Bell,D.S., Bell,K.M., & Cheney,P.R. (1994) Primary juvenile fibromyalgia syndrome and chronic fatigue syndrome in adolescents. Clinical Infectious Diseases, 18 Suppl 1, S21-S23.

De Becker,P., Dendale,P., De Meirleir,K., Campine,I., Vandenborne,K., & Hagers,Y. (1998) Autonomic testing in patients with chronic fatigue syndrome. American Journal of Medicine, 105, 22S-26S.

De Meileir,K., Bisbal,C., Campine,I., De Becker,P., Salehzada,T., Demettre E., & Lebleu,B. (2000) A 37kDa 2-5A binding protein as a potential biochemical marker for chronic fatigue syndrome. American Journal of Medicine, 108, 99-105.

Fennell,P.A. (1995) The four progressive stages of the CFS experience: a coping tool for patients. Journal of Chronic Fatigue Syndrome, 1, 69-79.

Jason,L.A., King,C.P., Richman,J.A., Taylor,R., Torres,S.R., & Song,S. (1999) U.S. case definition of chronic fatigue syndrome: diagnostic and theoretical issues. Journal of Chronic Fatigue Syndrome, 5, 3-33.

Jason,L.A., Richman,J.A., Friedberg,F., Wagner,L., Taylor,R., & Jordan,K.M. (1997) Politics, science, and the emergence of a new disease. The case of chronic fatigue syndrome. American Psychologist, 52, 973-983.

Jason,L.A., Taylor,R., Wagner,L., Holden,J., Ferrari,J.R., Plioplys,A.V., Plioplys,S., Lipkin,D., & Papernik,M. (1995) Estimating rates of chronic fatigue syndrome from a community-based sample: a pilot study. American Journal of Community Psychology, 23, 557-568.

Komaroff,A.L. (2000) The biology of chronic fatigue syndrome. American Journal of Medicine, 108, 169-171.

Komaroff,A.L. & Buchwald,D. (1991) Symptoms and Signs of Chronic Fatigue Syndrome. Reviews of Infectious Diseases, 13(Suppl 1), S8-S11.

Rowe,K.S. (1997) Double-blind randomized controlled trial to assess the efficacy of intravenous gammaglobulin for the management of chronic fatigue syndrome in adolescents. Journal of Psychiatric Research , 31, 133-147.

Soetekouw,P.M., Lenders,J.W., Bleijenberg,G., Thien,T., & van der Meer,J.W. (1999) Autonomic function in patients with chronic fatigue syndrome. Clinical Autonomic Research, 9, 334-340.

Steele,L., Dobbins,J.G., Fukuda,K., Reyes,M., Randall,B., Koppelman,M., Reeves, & WC. (1998) The epidemiology of chronic fatigue in San Francisco. American Journal of Medicine, 105, 83S-90S.

Stewart,J.M., Gewitz,M.H., Weldon,A., Arlievsky,N., Li,K., & Munoz,J. (1999) Orthostatic intolerance in adolescent chronic fatigue syndrome. Pediatrics, 103, 116-121.

Streeten,D.H., Thomas,D., & Bell,D.S. (2000) The roles of orthostatic hypotension, orthostatic tachycardia, and subnormal erythrocyte volume in the pathogenesis of the chronic fatigue syndrome. American Journal of the Medical Sciences, 320, 1-8.

Strickland,P., Morriss,R., Wearden,A., & Deakin,B. (1998) A comparison of salivary cortisol in chronic fatigue syndrome, community depression and healthy controls. Journal of Affective Disorders, 47, 191-194.

Suhadolnik,R.J., Reichenbach,N.L., Hitzges,P., Sobol,R.W., Peterson,D.L., Henry, B, Ablashi,D.V., Muller,W.E., Schroder,H.C., & Carter,W.A. (1994) Upregulation of the 2-5A synthetase/RNase L antiviral pathway associated with chronic fatigue syndrome. Clinical Infectious Diseases, 18 Suppl 1, S96-104.

Tiersky,L.A., Johnson,S.K., Lange,G., Natelson,B.H., & DeLuca,J. (1997) Neuropsychology of chronic fatigue syndrome: a critical review. Journal of Clinical & Experimental Neuropsychology, 19, 560-586.

Vercoulen,J.H., Bazelmans,E., Swanink,C.M., Fennis,J.F., Galama,J.M., Jongen,P.J., Hommes,O., van der Meer,J.W., & Bleijenberg,G. (1997) Physical activity in chronic fatigue syndrome: assessment and its role in fatigue. Journal of Psychiatric Research, 31 , 661-673.