Leonard A. Jason, Karen M. Jordan, Judith A. Richman,
Alfred Rakemaker, Cheng-Fang Huang, William McCready,
Jennifer Shlaes, Caroline P. King, Dana Landis, Susan Torres,
Trina Haney-Davis, Erin L. Frankenberry
Authors: Arthur J. Hartz. MD, PhD, Evelyn M. Kuhn, PHD, Paul H. Levine, MD, Richard London, MD
Objective:The simultaneous examination of a large number of patient characteristics in a group of subjects followed for fatigue makes it possible to compare the relative importance of these characteristics as prognostic factors.
Methods: The data analyzed were from 208 subjects in a registry of persons who were 18 years old or older and had idiopathic fatigue for at least six months. All subjects completed an extensive baseline questionnaire that provided information about fatigue, demographic characteristics, medical conditions, life style, sleeping habits, psychological characteristics, and the presence of criteria for chronic fatigue syndrome (CFS). Changes in the status of fatigue from baseline to follow-up after six months were tested for an association with risk factors at baseline and changes in risk factors.
Results:Characteristics of the subjects at baseline that independently predicted less improvement m fatigue were: interrupted sleep (p <.01), the nurnber of CFS features (p=.02), younger age (p=06), and watching > eight hours of television a day (p=.07). Changes in subject characteristics that were independently associated with improvement in fatigue levels at (p < .05) were a reduction in the number of CFS somatoform symptoms (especially muscle aches), an improved ability to concentrate, and decreased depression score. Changes in the number of somatoform symptoms not used to define CES were uncorrelated with changes in the levels of fatigue.
Conclusions: These findings provide experimental evidence for the importance of the CFS criteria which were formulated from clinical impressions. The associations of changes in chronic fatigue with changes in depression score and muscle aches supports theories that chronic fatigue, depression, and fibromyalgia may have similar etiologic factors.
Authors: Leonard A. Jason, Karen M. Jordan, Judith A. Richman, Alfred Rakemaker, Cheng-Fang Huang, William McCready, Jennifer Shlaes, Caroline P. King, Dana Landis, Susan Torres, Trina Haney-Davis, Erin L. Frankenberry
Objective: Previous estimates of the prevalence of fatigue and chronic fatigue have derived largely from treated populations and have been biased by differential entrance to health care treatment linked with gender, racial/ethnic and social class status. This study involves a community-based prevalence study of prolonged fatigue and chronic fatigue. The study:
addresses the rate of prolonged fatigue and chronic fatigue in a socioeconomically and ethnically diverse sample of 28,000 adults in Chicago and
establishes the relative prevalence of prolonged fatigue and chronic fatigue across race/ethnicity, socioeconomic status and gender.
Method:The present study involved a cross-sectional screening survey of a random community sample. Univariate and multivariate statistical techniques were utilized to delineate the overall rate of prolonged fatigue and chronic fatigue in this Chicago population and its relative prevalence by gender, race/ethnicity, and social class.
Results: Prevalence rates for prolonged fatigue, ranged from 5.0% to 7.7%, whereas chronic fatigue ranged from 2.6 % to 4.1%. These data are similar to findings from a random community sample in San Francisco, which found that 6.3% of adults reported prolonged fatigue and 4.3% reported chronic fatigue (Steele et al., in press), and from a community sample in Great Britain, where 4.7% of respondents reported chronic fatigue for 50% of the time (Pawlikowska et al., 1994). These studies indicate that a substantial proportion of adults have prolonged and chronic fatigue, and these are relatively high incidence symptoms when considering the prevalence of other common chronic health conditions. In Steele et al. (1997) study of the 14,627 adults, 0.2% of the study population were classified as having CFS-like disorders, 1.8% as having idiopathic chronic fatigue, and 2.2% as having chronic fatigue explained. In the present study, lower to upper estimates were 1.4 -2.1% for CFS-like, 1.0-1.6% for idiopathic chronic fatigue, and 0.2-0.4% for chronic fatigue explained. In general, the overall rates of chronic fatigue were similar in the two studies, but the present investigation reported higher rates of CFS-like disorder, and lower rates of chronic fatigue explained. Differences in the two studies are possibly due to the exclusionary illnesses. In the present study, because those identified as possibly having CFS were medically screened in later phases of the study, individuals were [included unless] there was a clear-cut exclusionary illness. In the Steele et al. study, a medical evaluation was not part of the study, and as a consequence, those investigators might have employed more stringent exclusionary criteria. In the present study, fewer exclusionary criteria were possibly used in order to insure that all possible cases of CFS were selected, and in a later phase of the study, a medical evaluation would be used to thoroughly screen out exclusionary illnesses.
Conclusion: Findings indicated that fatigue is common in urban populations, but that prolonged fatigue and chronic fatigue occurs in only about 5.0% to 7.7% and 2.6% to 4.1%, respectively, of the sample of the population.
Authors: Karen Jordan, Ph.D., Penny Ayers, Kari Taylor, and Leonard Jason, Ph.D.
Objective: To estimate the prevalence of chronic fatigue syndrome (CFS) in children and adolescents and to describe the characteristics of children and adolescents with CFS and chronic fatigue.
Methods: As part of a larger study on the epidemiology of adult chronic fatigue syndrome that utilized a community-based sample, telephone interviews were conducted with a randomly-chosen adult household member. Questions were asked about the randomly-chosen adult's and all household children's fatigue and CFS symptoms. Follow-up interviews, with both the primary caretaker and the child, were then conducted for those children and adolescents that were reported to suffer from CFS symptoms, as well as a randomly-chosen control group. In addition, these children and adolescents will receive medical and psychological evaluations in order to allow the diagnosis of CFS; however, as these evaluations are ongoing, the data presented will focus on the data from the screening interview and the follow-up interview.
Results: Eighty children of the 3895 children ages 5 to 17 surveyed (2.1%) met criteria for CFS-like illness (i.e., met criteria for CFS based on self-report, although no medical or psychological evaluations were performed), with almost equal distribution between girls (52.5%) and boys (47.5%). The age range of these children was from 5 to 17 years, with an average of 11.6 years. Over half (45, 56.3%) were 12 and younger, whereas 35 (43.8%) were aged 13 and older. The average number of symptoms reported was 5.8. Girls were significantly more likely to suffer from frequent headaches, sore throats, and lymph node pain, whereas boys were more likely to suffer from impairment in memory and concentration and post-exertional malaise. There was no difference in the number of symptoms or frequency of individual definitional symptoms between the two age groups. In addition to prevalence data, information regarding school attendance, school performance, and other educational information will be presented. A comparison of caregiver's and child's descriptions of the illness will be presented.
Conclusion: Fatigue and symptoms of CFS do indeed strike children and adolescents. This preliminary research demonstrates that even younger children do display CFS symptoms and definable fatigue and that they are very similar to adolescents in number and frequency. Similarly, both boys and girls appear to be stricken by fatigue and CFS symptoms, although there are differences in the types of symptoms most commonly striking each gender. Further work is in progress to evaluate these children medically and psychiatrically in order to allow firm diagnoses to be made and for further description of this group.
Authors: Han K. Kan, Ph.D., Clare M. Mahan, Ph.D., Kyung Y. Lee, Ph.D., Carol A. Magee, Ph.D., Benjamin H. Natelson, M.D.
The "National Health Survey of Gulf War Veterans and Their Families" is a Congressionally-mandated health study designed to estimate and compare the prevalence of various symptoms and health conditions among Gulf veterans and those of non-Gulf veterans. In Phase I, a structured health questionnaire was mailed to a population-based sample of 15,000 Gulf veterans and 15,000 era veteran controls to obtain information on their exposures, presence of various symptoms, medical and psychological conditions, recent clinic visits and hospitalization and pregnancy outcomes. After three follow-up mailings, 15,825 veterans responded. In Phase II, telephone interviews were attempted on the 14,175 non-respondents using the same questionnaire as mailed in Phase I, with the addition of a question on reasons for the previous non-response. A total of 5,116 veterans participated in the telephone interviews, increasing the overall response rate to 70%.
The 1994 revised definition. of CFS published by the CDC was adopted to identify individuals who met CFS criteria. The prevalence of CFS for the entire Gulf veteran population was obtained from weighted estimates of individual military strata values from the samples using SUDAAN, a statistical program consisting of a family of procedures used to analyze data from complex surveys. This survey is complex in that the sampling design was stratified random sampling with unequal probabilities of selection in the various strata (gender, branch of service, reserve and national guard unit status, deployment status). An estimated 5.1% of Gulf veterans and 1.2% of Gulf era veteran controls met the criteria for CFS resulting in an adjusted rate ratio (RR) of 4.45 (95% confidence interval, 3.3-5.9). The risk of CFS was significantly higher among whites than non-whites; ground troops (Army and Marines) than non-ground troops; and those who served in the Reserve and National Guard units than regular units.
This survey is one of a few population-based studies of Gulf veterans and suggests that CFS may account for a significant portion of Gulf veterans who complained of medically unexplained constellation of symptoms.
Authors:KG Manu, Colby College, Waterville, Maine; S Pollack, St. John's University, Queens, New York; and P Manu, Albert Einstein College of Medicine, Bronx, New York
NOTE:Peter Manu is a consultant with Metlife Insurance Company. During the audience questioning, Judy Morris, M.D., asked Dr. Manu if Metlife had given him the cases, and he answered yes. She then asked how the cases were selected, and he said that he did not know. Consequently, there is no way to judge what type of biases (inadvertent or not) are contained in the data used for the study.
Objective: To determine whether diagnoses of chronic fatigue syndrome (CFS) and idiopathic chronic fatigue (ICF) are made according to the definitions agreed upon by experts and endorsed by the United States Centers of Diseases Control and Prevention.
Methods: Structured cross-sectional study of 112 occupationally-disabled patients (mean age 42 years, proportion of women = 86%, mean duration of total work disability =28 months) given the diagnosis of CFS as the cause of their severe functional impairment. The evaluation sought substantiation of the diagnosis in accord with the standard definition (Fukuda K, Straus SE, Hickie I, et al. The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med 1994:121:953-959) and its clearly delineated requirements for the characteristics of fatigue, duration, associated symptoms, initial clinical and laboratory evaluation, and medical or psychiatric exclusions. In addition, the study attempted to determine through univariate and stepwise logistic analyses whether characteristics of patients and their physicians have predictive value for a correct or incorrect diagnosis of CFS.
Results:Forty-two (37.5%) 6f the 112 patients did not meet criteria for the diagnosis of CFS. Of these 42 patients, 21 had an insufficient number of associated symptoms, 18 had not received the required initial clinical and laboratory evaluation, 8 had medical exclusions (i.e., malignancy, primary sleep disorder, connective tissue disorder, inflammatory bowel disease, and morbid obesity) and Shad psychiatric exclusions. Ten patients had more than one reason to fail the CFS definition. Patient and physician characteristics did not predict the correctness of the CFS diagnosis.
Conclusions: The diagnosis of CFS was misapplied to a substantial proportion of a population severely disabled by fatigue, primarily because the patients did not have the expected associated symptoms and because the physicians did not thoroughly pursue accepted medical or psychiatric explanations for fatigue. Although at least half of the patients given the inappropriate diagnosis of CFS may have met the ICF criteria, the distinction between CFS and ICF appeared irrelevant to patients and physicians. The CFS and ICF entities should be abolished in favor of a new construct, disabling chronic fatigue DCF), in which objective testing of cognitive and physical impairments should replace the symptom criteria of the current definition of CFS.
Update: Wichita population-based study of fatiguing illness
Authors: M. Reyes Ph.D., R. Nisenbaum, Ph.D., G. Stewart, M.A., W.C. Reeves, M.D.
NOTE: The abstract for this paper was printed before the study was finalized; the version actually presented at the conference will be uploaded soon to the CDC's Chronic Fatigue Syndrome Research website, but is not yet available there. In the meantime, those interested in reading the new conclusions can find a summary by Reuters News Service at:
Intelihealth: Chronic Fatigue Syndrome common in U.S. (Reuters, 16 October 1998).
Objective: To assess the magnitude of fatiguing illness in Wichita, Kansas, by estimating the prevalence of prolonged fatigue (> 1 month), chronic fatigue (> 6 months), and chronic fatigue syndrome (CFS).
Methods: The Centers for Disease Control and Prevention recently completed the first phase of a population-based, cross-sectional, random-digit-dial survey of residents in Wichita. Respondents were asked about their fatigue status, demographic characteristics, medical history and fatigue-related symptoms. Persons meeting the symptom and fatigue criteria of the 1994 CFS case definition were clinically evaluated. For comparison, a sample of non-fatigued persons was also clinically evaluated. Follow-up interviews of respondents will be conducted at 12-, 24-, and 36-month intervals.
Results: We identified a random sample of >: 90,000 persons. Respondents reporting > 1 month of fatigue (n=4,215) and a sample of 4,149 respondents reporting no fatigue were further interviewed about their current health status, symptoms, and medical history. Persons with suspected CFS (n=347) and a sample of non-fatigued persons (n=64) were clinically evaluated. Preliminary findings to date will be presented.
Conclusion: Analyses of these data will provide the first population-based estimate of the prevalence of fatiguing illness. We are longitudinally monitoring these respondents to study the clinical course of fatiguing illness and to identify incident cases of CFS. [See note above: The results of the completed study were reported by Reuters.]
Authors: Nancy F. Hill, Lana A. Tierskv, Vanessa R. Scavalla, Benjamin H. Natelson
Object: The present investigation examined the fluctuation of symptoms experienced by Chronic Fatigue Syndrome (CFS) patients over time. Both change in the incidence of symptoms as well as the severity of symptoms experienced were evaluated at three points in time. This study differs from those previously published in that two of the evaluations were face-to-face evaluations versus a postal questionnaire.
Methods: Twenty-three (23) patients meeting the CDC 1988 criteria for CFS (Holmes, et al., 1988), which was modified to include at least substantial symptom severity in the month prior to intake were evaluated at three points in time. At
Time 1, patients received a comprehensive history and physical evaluation.
At Time 2, (1.5 years later) individuals were sent a postal follow-up questionnaire.
Time 3 which occurred nearly 2 years after Time 2, included a physical evaluation.
Based on the physical evaluation as well as the patients' rating of minor symptoms and reported reduction in activity, patients were categorized as:
having severe CFS,
greatly improved and
Results: At Time 2, 12 of the 23 patients (52%) remained severely ill, 5 slightly improved, 5 greatly improved and 1 recovered. At Time 3, 9 subjects remained severely ill, and 3 of those who were severely ill at Time 2, improved. Out of the five subjects slightly improved at Time 2, 3 worsened again, and 2 improved further. Of the 5 that had improved at Time 2,4 remained the same and 1 worsened. The one patient who had reported recovery at Time 2, remained recovered at Time 3. Over the 3.5 years (range 2.3 to 4.7 years) between Time 1 and Time 3, 13 of the 23 patients (57%) remained severely ill, 3 slightly improved, 6 greatly improved, and 1 recovered. The presence of a psychiatric diagnosis (i.e., diagnosis of a depressive or anxiety disorder) concurrent with illness onset and mode of illness onset (sudden vs. gradual) was not predictive of positive or negative illness outcomes. Illness severity between Time 2 and Time 3 remained stable; those who improved at Time 2 tended to remain in an improved state at Time 3. Only one patient was diagnosed with an alternate illness that could explain the CFS. Depression appears to lessen over time.
Conclusion:This study shows that for the severely ill subset of CFS patients studied here, the majority show no symptom improvement and only 4% of the patients recovered. However, in the remaining 39%, improvement in CFS symptoms did occur. Further studies investigating the illness at closer and more numerous time periods is recommended.