AACFS Home Page
The CFIDS/M.E. Information Page

Index to Abstracts from AACFS Conference October 1998

Abstracts of Papers Presented at
The Bi-Annual Research Conference of the
American Association for
Chronic Fatigue Syndrome (AACFS)

October 10-11, 1998 -- Cambridge, Massachusetts

Poster Session 1: Epidemiology

Co-chairs: Leonard Jason, Ph.D., and William Reeves, M.D.
October 10, 1998, 3:00 pm-3:30 pm

1. The prevalence of fatigue and bodily pain in those with chronic sinusitis
Alexander C Chester
2. Prevalence and overlap of Chronic Fatigue Syndrome and Fibromyalgia Syndrome among 100 patients with Multiple Chemical Sensitivity
Albert Donnay, Grace Ziem
3. Descriptive data on patients with Chronic Fatigue Syndrome and chronic fatigue
Evengârd B, Lindh G, Jonzon E, Sandberg A, Theorell T.
4. Measuring phases of recovery in patients with CFS
Patricia A. Fennell, Leonard A. Jason, Susan M. Klein
5. What do doctors and health workers of northeast Brazil know and do about Chronic Fatigue Syndrome?
Luis C Nacul, Fabio LR Pereira, Georgia A Jaques, Anaisa G Rarnos, Clineu JA Rodrigues, Isabel B Oliva, Nadja R Gonçalves, Roberta C Almeida
6. Clinical and pathogenetical characterization of > 200 patients of an Italian Center for CFS Study
D. Racciatti, A. Barberio, I. Vecchiet, E. Pizzigallo
7. Measuring outcomes of treatment in Chronic Fatigue Syndrome: a comparison of simple questioning vs. use of a validated outcome instrument (short-form 36)
Stanlev N. Schwartz MD, FACP, FIDSA; Rick Jones, PhD.
8. Do the 1988 and 1994 CFS case definitions identify the same illness complex?
Lana A. Tierskv, Stacy Weisberg, John DeLuca, and Benjamin H. Natelson
9. Chronic Fatigue Syndrome: a nursing perspective
Theresa L. Walker BSN RN

The prevalence of fatigue and bodily pain in those with chronic sinusitis

Author: Alexander C Chester

Objective: Recent rhinological reports have noted a high prevalence of significant chronic fatigue and bodily pain in individuals presenting with chronic sinusitis. Diminished social functioning, and limitations on physical activities also were noted more commonly in sinusitis patients. The general functioning of this group was below that of patients with congestive heart failure, angina, chronic obstructive pulmonary disease, and back pain. The following study was designed as an attempt to confirm the findings linking chronic sinusitis with both fatigue and bodily pain in a general internal medicine practice.

Methods: All 297 consecutive outpatients under the age of forty presenting to a single general medical practice for a comprehensive examination were interviewed after obtaining consent. Laboratory data included:complete blood count, erythrocyte sedimentation rate (Wintrobe method), standard 20 channel chemical analysis (SMAC), rapid plasma reagent serology, and urinalysis. For those over the age of 30, a chest roentgenogram was obtained. Symptoms recorded as present for this study must be both frequent and problematic (e.g., those with "bodily pain" had this symptom both frequently and severely enough to be troublesome). Chronic sinusitis was defined as any or all of the following chronic symptoms: facial pressure, cheek pressure, or frontal headache. Chronic unexplained fatigue was defined as fatigue for greater than one month that eludes explanation after the history, physical, and baseline laboratory tests were reviewed. Those with fatigue considered a normal response to excessive physical or mental demands, as well as mild fatigue, defined as that causing no significant impact on their sense of well being, were not considered "fatigued" for the purposes of this study.

Results: Of the 297 patients, 85 relayed a history of chronic sinusitis. When the 85 (study population) were compared to the remainder (control) there were no significant differences in physical or laboratory findings. Age, marital history, prevalence of depression, exercise patterns, and alcohol consumption were similar.

Fatigue & Chronic Sinusitis -- No. (%)




P value

Unexplained fatigue

46 (54)

21 (10)

< 001

Daytime sleepiness

38 (45)

13 (6)

< 001

Sudden onset of fatigue*




Occuring after an URI*        




*Expressed as a percentage of those in each group with chronic fatigue (e.g., 43% of those with chronic fatigue in the study group had a sudden onset of that fatigue).

Bodilv Pain & Chronic Sinusitis No. (%)




P value

Bodily pain

23 (27)

10 (5)

< 001

Upper backache                        

14 (16)

10 (5)

< 001


10 (12)

2 (1)

< 001

*Refers only to point tenderness criteria for fibromyalgia.

Conclusion: Chronic sinusitis appears to be commonly associated with fatigue and body pain in the setting of a general internal medicine practice. This association raises the possibility that chronic sinusitis might also be related to other states of unexplained fatigue and/or bodily pain associated with upper respiratory symptoms: chronic fatigue syndrome, sick building syndrome, chemical sensitivity syndrome, Gulf War Syndrome, etc. A greater awareness and interest on the part of internists is needed to define the systemic symptoms of chronic sinusitis.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Prevalence And Overlap Of Chronic Fatigue Syndrome and Fibromyalgia Syndrome Among 100 Patients with Multiple Chemical Sensitivity

Authors: Albert Donnay and Grace Ziem

Objective: To determine the prevalence and overlap of chronic fatigue syndrome (CFS) and fibromyalgia syndrome ~MS) among adults diagnosed with multiple chemical sensitivity (MCS).

Methods: One hundred new adult patients (68F and 32M) with MCS, diagnosed by history, in a private practice from January 1996 through January 1998 were evaluated upon their initial visit for CFS and FMS using diagnostic criteria of the US Center for Disease Control (1994) and the American College of Rheumatology (1990), respectively. Data were recorded on the number of criteria met for CFS and the number of FMS pressure points demonstrating pain upon manual palpation. MCS was defined simply by multiple chronic symptoms in multiple organs triggered by multiple chemicals at or below previously tolerated levels of exposure and associated with a variety of immune, neurological and other abnormalities, as previously reported by Ziem and McTamney (1997).

Results: Of the 100 MCS patients, 88% met the criteria for CFS with all the major and an average of 6.4 of the 8 minor criteria, and 49% met the criteria for FMS with pain in an average of 15.2 of the 18 pressure points. In comparison, the 12% of MCS cases who did not meet the definition of CFS reported an average 1.75 minor criteria, while the 51% who did not meet the definition of FMS reported an average 3.26 painful pressure points. Unexpectedly, the prevalence of CFS was slightly higher among men (91% v.87%), but FMS was more than twice as common among women (59% v.28%). Among women, 56% met criteria for all three disorders, and an additional 31% had both MCS and CFS. This pattern was reversed in men, with only 28% having all three, and 63% having both MCS and CFS. The diagnosis of MCS alone was relatively rare, occurring in only 10% of women and 9% of men. Even rarer was the overlap of MCS and FMS without any CFS, which was found in just 2 women and no men. While FMS was almost always accompanied by CFS (n 100% of male FMS cases and 95% of female), CFS was associated with FMS in only 31% of the male CFS cases and 65% of the female.

Conclusion: MCS seldom occurs alone. In this study of 100 adults evaluated with MCS, almost all met current diagnostic criteria for CFS and almost half also had FMS, including the majority of women but less than a third of the men. This gender difference was due entirely to FMS, which was twice as prevalent in women as men. Nevertheless, this is significantly more men than found in studies of FMS among CFS clinic patients -- such as that by Buchwald et al (1994)--in which women with FMS outnumbered men by 3 to 1, and FMS among the general population--such as that by Wolfe et al (1995)--in which this ratio was almost 7 to 1. Also never before reported is the marginally higher prevalence of CFS found among men. The finding that only 53% of those with CFS also had FMS is significantly lower than that reported in several other studies, while the 96% rate of CFS among those with FMS is much higher than previous studies of FMS populations in which approximately 50% met criteria for CFS (Buchwald et al 1987; Hudson et al 192). The overall overlaps we found between MCS-CFS and MCS-FMS are 20% greater and less, respectively, than the 69% reported for both by Buchwald and Garrity (1994), but the MCS-FMS overlap is just 6% < the 55% rate of MCS found among FMS patients by Slotkoff, Radulovic and Clauw (1997). These extensive 2- and 3-way overlaps highlight the need for physicians to screen suspected cases of CFS, FMS and MCS for all three disorders using published criteria and to report results in these terms. We strongly recommend this be made standard practice in both clinical settings and research protocols. To improve early detection, treatment and prevention, further studies are needed to determine the relative order in which the hallmark symptoms of chronic fatigue, chemical sensitivity, and fibromyalgia evolve in these overlapping cases, as well as the kinds of events (illness, injury, toxic exposure, etc.) associated with the onset of these symptoms. We present a questionnaire developed for this purpose and discuss its use in the clinical evaluation of new CFS, FMS and MCS patients.

Thanks to Carolyn Viviani for scanning in and formatting this abstract.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Descriptive data on patients with Chronic Fatigue Syndrome and chronic fatigue

Authors: Evengârd B, Lindh G, Jonzon E, Sandberg A, Theorell T.
Infectious Disease, Karolinska Institutet at Huddinge University Hospital and at Institute of Psychosocial Environment, Stockholm, Sweden.

Background: As part of a multi-disciplinary effort to find causative and/or influencing factors to Chronic fatigue syndrome, a patient population attending a university hospital-based CFS-polyclinic was characterized using questionnaires describing demographic and functional data, as well as coping strategies. The results were compared with patients with chronic fatigue to establish whether a difference exists between CFS and CF -- making CFS an entity.

Material: 239 consecutive patients fulfilling CDC-criteria for CFS, and 144 consecutive patients with CF attending a CFS-polyclinic in Stockholm responded to questionnaires including Wessely fatigue scale, Sense of Coherence (SOC), FIQ (Fibrositis impact questionnaire) and CPRS-S-A (depression scale) and NEPO (100 symptoms during the last 14 days). A follow-up after 3 years was performed as to fatigue scale and FIQ. For statistical analysis Chi-2 and One-way ANOVA have been used (SPSS 8.0).






















age (average)

38 years


40 years

duration of illness*

3.9 years


5.0 years

infectious onset***




Fatigue scale












¹ (normal 140-160 p)








CPRS-S-A (< 6)        





7.7 SD4.0


7.4 SD4.7


6.5 SD3.8


7.4 SD4.7


8.2 SD3.4


7.5 SD4.1





Conclusion: In the CFS group, there are more women as compared to the CF group, and the CF patients have a longer duration before coming to the polyclinic. More patients in the CFS group report an infectious onset. As to functional capacity, sense of coherence, and depression, there are no differences between the groups, nor in the number of symptoms reported. After three years the CFS patients report a decrease in fatigue and in functional capacity. More follow-up data will be presented.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Measuring Phases of Recovery in Patients with CFS

Authors: Patricia A. Fennell, Leonard A. Jason, and Susan M. Klein

Objective: In order to validly understand the relationship between stress, fatigue and activity it is critically important to understand the phase that a patient with Chronic Fatigue Syndrome (CFS) is in. If individuals are assessed when they are in different phases, this will complicate interpretation of findings, and is perhaps one of the reasons that there are so many inconsistencies within the CFS research literature.

Method: Recently, a four phase model has been developed by Fennell (1993). In brief, this model postulates that individuals with CFS progress through a series of phases. Phase I, the crisis phase, begins with the onset of the CFS illness. During this phase, patients feel that treatment is urgent. They have a low tolerance for ambiguity, and they feel responsible for the illness, yet have an externalized locus of control. They are generally preoccupied with thoughts pertaining to CFS, or they are in denial that they are actually suffering from the illness. During Phase II, the stabilization phase, patients recognize their symptom patterns and begin to feel some returning control over their lives. During this process, they seek alternative sources of support and treatment. They have a low tolerance for the chronicity of the illness, and continue to have a low tolerance for ambiguity. During Phase III, the existential conflict phase, patients have an increased internalized locus of control and an increased tolerance for ambiguity and chronicity. Patients become more aware of societal effects, and they begin to show compassion for themselves as they come to accept the complete loss of the pre-illness self and grieve deeply. From this grieving process comes the creative construction of meaning of their illness experience. In Phase IV, integration, individuals with CFS have successfully blended their pre and post crisis lives. They reconstruct a definition of themselves, form new friendships and rebuild former relationships. They express appreciation for themselves for having weathered the CFS experience and are finally able to accept the ambiguities of a relapsing and remitting illness.

Results: A questionnaire has been developed to tap these four phases, and it has been administered to several groups of people with CFS. Findings from this new instrument with several samples will be presented during this symposium.

Conclusion: Individuals perceive stress and cope with the illness experience differently, depending on which phase of the illness they are in. These findings have important implications for how we understand the relationships between stress, fatigue, and activity.

Thanks to Carolyn Viviani for scanning in and formatting this abstract.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

What do doctors and health workers of northeast Brazil know and do about Chronic Fatigue Syndrome?

Authors: Luis C Nacul, Fabio LR Pereira, Georgia A Jaques, Anaisa G Rarnos, Clineu JA Rodrigues, Isabel B Oliva, Nadja R Gonçalves, Roberta C Almeida
Universidade de Pernambuco, Faculdade de Ciências Médicas, Departamento de Medicina Social. Rua Arnóbio Marques 310, Recife-PE, 50100-130, Brasil.


a) To assess knowledge and practices of physicians of Recife, Northeast Brazil, in respect to chronic fatigue syndrome (CFS).
b) To assess beliefs and perceptions of community health workers (CHWs) on severe fatigue.

Methods: Data were separately obtained from physicians and CHWs. A questionnaire survey was directed to a random sample of 72 doctors working in teaching hospitals (55), and in primary health care (PHC) (17). A form containing 30 questions (mostly open), assessed their capacity of recognizing a clinical case of typical CFS, their knowledge/beliefs about diagnostic criteria, risk factors, treatment and prognosis, and their experience in diagnosing and treating patients with severe fatigue and CFS. A focus group discussion conducted with 7 CHWs explored lay-people's denominations of severe fatigue, and the CHWs perceptions on its causes and treatment. In a further stage of the research, we are currently conducting a community-based prevalence study of fatigue, chronic fatigue and CFS, in a district of Recife with a population of around 300,000 people.

Results: When prompted with a typical case history of CFS, 25 (34.7%) physicians mentioned CFS, either as the only diagnosis (5.6%) or as one of two or more possible diagnoses (29.5%). Hospital doctors gave the right diagnosis more frequently than those working in PHC (odds ratio=9.4; p=0.0l). A correct diagnosis was not associated with either time since medical graduation, or level of specialization. Personal and family problems (23.9%), "stress" (21.1%) and viral infections (19.7%) were the most frequently mentioned risk factors believed to be related to CFS. Psychotherapy (25.3%) and antidepressants (22.5%) were the most often proposed treatments. The prognosis is considered good by 65.9%. Based on doctors' experience, the median prevalence of fatigue and prolonged fatigue were estimated as 18.5% and 3.5%, respectively. CFS was considered a rare disorder, which has only been diagnosed by 20 (27.8%) of the doctors. The estimated prevalence of severe and prolonged fatigue was 0.2%. A typical CFS patient, according to doctors who have ever made this diagnosis, is a female (95.4%), young adult (87%), housewife (41.7%) with a low socio-economic background (57.1%). Only 1 doctor (1.4%) knew the diagnostic criteria. CHWs perceive severe fatigue as frequently related to "stress," other psychological problems, and to adverse life situations. Tuberculosis and infections due to unknown organisms could also explain some cases. CHWs believe that emotional support and good nutrition, possibly with vitamins supplementation, favour recovery.

Conclusions: CFS is an uncommon diagnosis in Recife, but its prevalence is possibly similar to that found in developed countries. Most doctors, especially those working in PHC, are unable to recognise a case of CFS and do not know the diagnostic criteria. Cases are believed to be often associated with psychological factors, especially "stress." Treatment emphasizes psychological support and the use of antidepressants.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Clinical and pathogenetical characterization of > 200 patients of an Italian Center for CFS Study

Authors: D. Racciatti, A. Barberio, I. Vecchiet, E. Pizzigallo
Clinic of Infectious Diseases, "G. D'Annunzio" University of Chieti, (Italy).

Objective: CFS probably is an heterogeneous disease, maybe multifactorial, or it includes different pathologies which manifest with the same symptoms. In some cases the mode of presentation of the illness implicate the exposure to chemical and/or food toxins as precipitating factors (e.g.: ciguatera poisoning, Gulf War Syndrome, etc.). In other CFS cases, the etiology is still unknown. There are various hypotheses that consider a possible involvement of multiple events in CFS onset where different precipitating factors can interact each other, even if not always all present in the single patient: latent and/or chronic viral infections, immunologic and neuroendocrine dysfiinctions, psychological, environmental and mood factors. According to this theory, we examined a consecutive sample of patients complaining of severe and chronic fatigue to assess the pathogenetic events and consecutively the appropriate therapeutic strategies for each patient.

Methods: The Clinic of Infectious Diseases of "G. D'Annunzio" University of Chieti is one of the main National Reference Centers for the CFS Study in Italy. From January 1992 to January 1998, 238 patients came to our observation: 110 of them met CDC criteria for CFS (1994), 105 did not; the other 23 patients are still under evaluation. All patients received a physical examination (including tests searching for the possible coexistence of a fibromyalgia syndrome), a psychiatric interview with several neuropsychological tests, laboratory tests (including magnesium determination on serum, and lymphocyte subsets characterization), neuroendocrine evaluation (circadian-rhythm of several hormones, buspirone challenge test), brain SPECT to assess the cerebral perfiision, and other examinations where necessary, according to the symptomatology reported by each patient (e.g. electromyographic, MR imaging, etc.).

Results: According to the preliminary results obtained from the various investigations, we subdivided our patients into different subgroups on the basis of pathogenetic events and clinical characteristics. We studied them in a comparative way to search for selective features that may allow a better characterization of each subgroup. We report the more significant data collected from this evaluation for a better understanding of the syndrome, and in particular, of its pathways in order to choose appropriate therapies for each subgroup.

Conclusion: CFS probably is an heterogeneous illness where different pathogenetic mechanisms may be implicated. So, on the basis of the relative pathways and clinico-pathological features, a classification of CFS patients into different subgroups should be considered for a better setting of each patient in order to evaluate a more appropriate therapy for each case.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Measuring outcomes of treatment in Chronic Fatigue Syndrome: a comparison of simple questioning vs. use of a validated outcome instrument (short-form 36)

Authors: Stanlev N. Schwartz MD, FACP, FIDSA; Rick Jones, PhD.
University of Oklahoma College of Medicine-Tulsa and Laureate Psychiatric Clinic and Hospital, Tulsa, OK

Objective: We hypothesized that assessing benefit of treatment of chronic fatigue syndrome (CFS) based solely on a patient's perception of improvement or decline ("I am better", "I am worse") might not provide a reliable measurement of outcome.

Methods: We performed Short Form 36 (SF36) tests (Medical Outcomes Trust) at time zero of treatment and again after approximately one year on 85 patients who met the 1994 CFS diagnostic criteria. No unique or standard treatment was used. Each patient was also asked a single self-report question (SRQ) after a year of treatment: "are you unchanged, better, much better, worse or much worse?". Each possible SRQ result was assigned a numerical value ranging from -2 for "much worse" to +2 for "much better." We compared the SRQ responses to each individual SF36 subscale and the summary SF36 scales at both time zero and time one year as well as to the changes in each scale over one year.

Results: We obtained SRQ data from 45 of the original 85 patients. Of the 45 patients, 15 reported improvement (10 "better," 5 "much better"), 11 reported worsening (7 "worse," 4 "much worse") and 19 reported no change. We found no correlation between the results of the SRQ and the physical component summary scales or the mental component summary scales. We found a positive correlation between the SRQ and the SF36 Vitality Transformed Score at one year (coefficient of correlation 0.339, p < 0.05). When analyzing improved and worsened patients separately, we could find no correlation between the direction of the SRQ result and changes in any individual SF36 subscale. Patients who reported improvement did not consistently show improvement in any of the individual scales or the summary scales. Similarly, a worsening SRQ did not correlate with negative changes on any of the subscales or summary scales.

Conclusions: Determining a benefit of therapy may be complex in patients with CFS. A simple "better-worse" query does not accurately or adequately reflect changes in physical and mental functioning as measured by a validated outcome instrument. Given that the severity of CFS is self-reported, we believe that clinicians, and especially investigators, should use validated outcome instruments such as the SF36 to determine the benefit and impact of treatment. The lack of appropriate outcomes measurement may be one reason why certain CFS treatments have appeared to yield different results in the hands of different investigators.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Do the 1988 and 1994 CFS case definitions identify the same illness complex?

Authors: Lana A. Tierskv, Stacy Weisberg, John DeLuca, and Benjamin H. Natelson

Objective: The decision to modify the original CFS case definition was based more on a perceived need to do so than on empirical data to support the proposition that the 1988 CFS case definition did not identify a distinct group of cases. If it is true that the 1988 criteria do not delineate a distinct group of cases, patients fulfilling these criteria should have the same symptom complexes as those fulfilling the relaxed 1994 case definition. Conversely, if patients fulfilling the 1988 case definition are a distinct group, their symptom pattern should differ from the profile of individuals fulfilling the 1994 CFS case definition. One purpose of this study was to evaluate these possibilities by comparing the symptom presentation of subjects diagnosed according to the 1988 CFS diagnostic criteria to the symptom complexes of subjects diagnosed according to the 1994 CFS diagnostic criteria. Differences in activity reduction and symptom onset were also investigated.

Methods: Sixty-two subjects completed a questionnaire that obtained information pertaining to CFS symptomatology as well as illness course. The data collection format of the list allowed subjects to respond if a CFS symptom was present as well as at what severity the symptom was experienced. Based on responses to the questionnaire CFS subjects were divided into two groups:

1) Subjects meeting the 1988 criteria for CFS, and
2) Subjects meeting the 1994 criteria.
Patients meeting the 1988 criteria also all met the 1994 criteria; however none of the patients meeting the 1994 criteria also met the 1988 standards. Thus, we were able to delineate two groups.

Results: When subjects who met the 1988 case criteria (n=45) were compared to those who met only the 1994 criteria (n=17), subjects in the 1988 group were found to suffer from a more severe form of the illness. Specifically, subjects in the 1988 group demonstrated more severe symptoms as well as a greater reduction in activity. The latter group also more frequently reported infectious-type symptoms as well as a sudden illness onset.

Conclusions: The 1988 and 1994 CFS case definition criteria appear to identify distinct patient groups. Given that the subjects in the 1988 group more frequently endorsed infectious symptoms as well as more frequently reported a sudden flu-like illness onset, an infectious etiology for this group is hypothesized.

Return to the Index for this Session
Return to the Main Index for the AACFS Conference

Chronic Fatigue Syndrome: a nursing perspective

Chronic Fatigue Syndrome (CFS) is a illness of uncertain etiology which is estimated to affect a large proportion of the U.S. population (Demitrack & Abbey, 1996). A CDC clinic-based study using surveillance data from four sites estimated the point prevalence of CFS to be 4.6-11.3 cases per 100,000 population (Gunn et ai.; 1993). A more recent population-based survey in San Francisco suggests that the prevalence may be a high as 0.2% of the total population (Steele et al., 1996). Because CFS has no biologic marker, medical practitioners must rely on the patient's accurate description of psychological and physical symptoms in order to pursue a diagnosis. The recovery rate of most patient's afflicted with CFS varies. Longer duration of illness seems to be correlated with poorer prognosis.

In this study, I undertook a survey to determine the outcomes of patients seen and treated in our CFS clinic. The purpose of this study was to determine which treatments and activities correlated best with a good outcome. Because of the enormous cost incurred while "looking for a diagnosis", the financial impact on the patient and the public is substantial. Even after the diagnosis is made, costs continue to mount since many patients find that their functional capacity is completely depleted in the process. This raises the question of disability. Does disability help patients recover or not? Our study suggests that contrary to popular belief, patients recover from CFS whether they are receiving disability or not.

Methods: The subjects were selected randomly from a list of patients who were evaluated for another study and found to meet CDC/NIH criteria for CFS. Subjects were contacted and administered a brief, structured interview over the telephone. Those who could not be contacted on the first phonecall, were not recontacted. Those who were available were asked to respond to a standardized interview. Information on the individual characteristics and activities of each patient and their own assessment of their health were recorded. All patients were asked whether they were better, worse, or the same since their initial clinical visit. The patient's health self-rating was an estimate on a scale from 1 to 10 of their current symptoms (1 = profoundly symptomatic and 10= asymptomatic). Median health ratings in different groups of patients were compared for significance using the Mann-Whitney U test.

Results: Forty patients were contacted by telephone. All forty patients agreed to be interviewed a mean of 14.5 months (S.D. = 7.25 mos.; range = 2 to 36 months) after their first clinic visits. There were 38 females and two males in this group. One subject was pregnant. The mean age of the subjects was 41 years (range 20 to 63 years).

The subject's state of health, activity levels, treatments, and responses to treatments are summarized in the attached tables. The distribution of subjects according to their health rating is shown in Figure 1. In this figure, subjects fell into two clusters, one with health ratings > 4.5 (Group A), the other with ratings < 3 (Group B). This separation of patients according to symptom severity is consistent with the previously reported finding by Hickie and co-workers that patients defined by the current CFS case definition constitute at least two major groups with respect to symptom range and severity (Hickey et al., 1995). In the current study, these two groups were compared with respect to several other criteria, including age, exercise, activity (work vs. Disability), sleep habits and duration, medications (allopathic and alternative), weight gain or loss, and their perception of the cause (s) of their illness. The two groups were not distinguishable by any of these criteria. Comparisons of subjects' health rating based on some of these variables are depicted in attached figures 2-5.