Comparing the Fukuda et al. Criteria and the Canadian Case Definition for Chronic Fatigue Syndrome J of Chronic Fatigue Syndrome, Vol. 12 Issue 1, pp. 37-52, 2004 ISSN: 1057-3321 Pub Date: 10/14/2004 Leonard A. Jason PhD, DePaul University, Center for Community Research, Chicago, IL 60614, E-mail: ljason@depaul.edu Susan R. Torres-Harding PHD, DePaul University Amber Jurgens BA, DePaul University Jean Helgerson BA, DePaul University Abstract: Because the pathogenesis of Chronic Fatigue Syndrome (CFS) has yet to be determined, case definitions have relied on clinical observation in classifying signs and symptoms for diagnosis. The selection of diagnostic signs and symptoms has major implications for which individuals are diagnosed with CFS and how seriously the illness is viewed by health care providers, disability insurers and rehabilitation planners, and patients and their families and friends. Diagnostic criteria also have implications for whether research based on varying definitions can be synthesized. The current investigation examined differences between CFS as defined by Fukuda et al. (1994) and a set of criteria that has been proposed for a clinical Canadian Case definition. There were twenty-three participants who met the Canadian criteria, 12 in the CFS (Fukuda et al. (7) criteria) group and the 33 from the chronic fatigue (CF)-psychiatric group. Dependent measures included: work status, psychiatric comorbidity, symptoms, and functional impairment (measured by the Medical Outcomes Study). People meeting the Fukuda et al. and Canadian criteria were compared with people who had a chronically fatiguing illness explained by a psychiatric condition. Statistical tests used included binomial logistic regression and analysis of variance. The Canadian criteria group, in contrast to the Fukuda et al. criteria group, had more variables that statistically significantly differentiated them from the psychiatric comparison group. Overall, there were 17 symptom differences between the Canadian and CF-psychiatric group, but only 7 symptom differences between the CFS and CF-psychiatric group. The findings suggest that both the Canadian and Fukuda et al. case definitions select individuals who are statistically significantly different from psychiatric controls with chronic fatigue, with the Canadian criteria selecting cases with less psychiatric co-morbidity, more physical functional impairment, and more fatigue/weakness, neuropsychiatric, and neurological symptoms. DISCUSSION `````````````````````` This study examined differences in sociodemographic characteristics, symptom frequency, and functional impairment with individuals meeting different diagnostic criteria sets for chronic fatigue syndrome. When samples of individuals meeting the CFS criteria were compared to the Canadian case definition, findings revealed no sociodemographic or psychiatric differences between the two samples. On measures of overall fatigue and disability, the Canadian criteria did select a group of patients with more impairment in physical functioning and did report higher levels of fatigue/weakness, neuropsychiatric, infectious, cardiopulmonary, and neurological symptoms when compared to the CFpsychiatric group. The CFS group reported more disturbed sleep, rheumatological, gastrointestinal and reproductive symptoms than the CF psychiatric group. The Canadian and CFS criteria groups differed statistically on two items, with the Canadian criteria group reporting a greater need to focus on one thing at a time, and the CFS group reporting more trouble staying asleep. The overall findings suggest that the Canadian clinical criteria appear to select a more symptomatic group of individuals than the CFS criteria, and these individuals do demonstrate less current and lifetime psychiatric impairment than those selected according to the Fukuda CFS criteria. In contrast, the Fukuda CFS group was no different from the CF-psychiatric group in psychiatric impairment. Predictably, the CFpsychiatric group showed the highest frequency of current and lifetime psychiatric disorders. Overall, there were 17 symptom differences between the Canadian and CF-psychiatric group, but only seven symptom differences between the CFS and CF-psychiatric group. It is of interest that in the prior study, when the ME criteria were utilized, there were 22 symptom differences between the ME and CF-psychiatric group, and eight symptom differences between the CFS and CF-psychiatric group (18). Findings suggest that both the ME and Canadian criteria select a group of patients with more symptoms, although the ME criteria appear to identify a group with higher rates of symptoms and the Canadian criteria identify a group with higher levels of physical functional impairment and less psychiatric comorbidity. The Canadian group had statistically significant differences from the CF-psychiatric group primarily in the fatigue/weakness, neuropsychiatric and neurological areas, while the CFS group's differences were largely found in the rheumatological and gastrointestinal areas. Similar findings emerged when ME was compared with CFS (18). The ME group in that study and the Canadian criterion group in the present study were both different from the CF-Psychiatric on 10 symptoms: * five neuropsychiatric symptoms (disturbances in the eyesight, need to focus on one thing at a time, confusion or disorientation, difficulty retaining information, slow to process visual and auditory information), * two neurological symptoms (feeling weak or dizzy after standing up quickly, feeling dizzy when moving head suddenly), * two fatigue/weakness items (neck weak, back weak), * and one infectious symptom (lymph node pain). In contrast, in both studies, the CFS condition was different from the CF-psychiatric group on * two rheumatological symptoms (sinus congestion, neck pain) * and one gastrointestinal symptom (pain in the lower abdomen) * and one reproductive symptom (decreased sexual interest/ function). Given the importance of the weakness/fatigue, neurological and neuropsychiatric symptoms, it seems possible to conclude that the Canadian and ME criteria identify a more debilitating illness group than the CFS group. Cardiopulmonary and neurological abnormalities have been suggested as important symptoms to consider in patients with CFS (31-33). For example, Jason et al. (34) found several cardiopulmonary and neurological symptoms (i.e., shortness of breath, chest pain, dizziness after standing, skin sensations, general dizziness, dizzy moving the head, and alcohol intolerance) uniquely differentiated a CFS group from controls. Findings from the present study indicate that the Canadian criteria does capture many of these cardiopulmonary and neurological abnormalities, which are not currently assessed by the Fukuda et al. (7) CFS case definition. Komaroff and associates (22) have suggested that eliminating the symptoms of muscle weakness, arthralgias, and sleep disturbance would provide greater sensitivity and specificity in CFS diagnosis. In contrast, the present investigation found that general muscle weakness did differentiate the Canadian criteria group from the CF-Psychiatric group, whereas trouble staying awake differentiated the CFS group from the CF-Psychiatric group. Komaroff and associates (22) also suggested adding anorexia and nausea as minor symptoms in the CFS case definition. However, in the present study, both occurred with relatively low frequency and neither uniquely differentiated the three groups. Hartz and associates (35) also investigated the occurrence of symptoms in persons with fatigue, and recommended the inclusion of fever and chills, muscle weakness, and sensitivity to alcohol as CFS case definition symptoms. Results of the current investigation indicate that muscle weakness and sensitivity to alcohol uniquely differentiated the Canadian group from CF-Psychiatric, and muscle weakness in the Canadian criteria group occurred at multiple sites, with back, shoulders and neck being the most frequently reported form of weakness. There were no differences in the occurrence of fever and chills in this study. It had been predicted that the Canadian criteria would select patients with more disability than those with the Fukuda et al. (7) CFS criteria. This small study shows that the Canadian group had higher scores than the CF-Psychiatric group, but were only directionally worse than the Fukuda et al. (7) CFS criteria group. In combination with symptom patterns, it is possible to conclude that the Canadian group does select individuals with greater impairment, particularly given the physical composite score, fatigue/weakness, neurological and neuropsychiatric symptoms, as these symptoms can interfere with daily living and occupational performance. None of the current definitions have been empirically derived or prospectively contrasted with one another. Katon et al. (5) found that patients with CFS were indistinguishable from those with chronic fatigue not meeting the 1988 criteria. One study that has compared the 1988 criteria, British and Australian case definitions found similar laboratory abnormalities across all the definition groups (36). Studies examining sources of diagnostic unreliability have shown that subject, occasion, and information variance account for only a small portion of diagnostic reliability (20). However, criterion variance, differences in the formal inclusion and exclusion criteria used by clinicians to classify patients data into diagnostic categories, accounts for the largest proportion of diagnostic unreliability. The Fukuda et al. (7) CFS definition as well as the Canadian definition would be improved if more attention was devoted to developing operationally explicit, objective criteria and standardized interviews (37). A recent study by Linder et al. (17) used artificial neural network to classify patients with chronic fatigue (including CFS and idiopathic chronic fatigue), lupus erythematosus, and fibromyalgia, and were able to achieve a sensitivity of 95% and a specificity of 85%. Those chronic fatigue symptoms that had the highest accuracy were "acute onset of symptoms" and "sore throat," which supports the hypothesis of an infectious etiology. Results from the present investigation highlight the importance of contrasting different diagnostic criteria in order to gain a greater understanding of the syndrome now known as CFS. The findings do suggest that the Canadian criteria point to the potential utility in designating post-exertional malaise and fatigue, sleep dysfunction, pain, clinical neurocognitive, and clinical autonomic/neuroimmunoendocrine symptoms as major criteria for future attempts to define this syndrome. Unfortunately, there are no published instruments specifically designed to assess many of these symptoms objectively. It is important to carefully assess each diagnostic criterion, as the manner in which clinicians phrase their assessment questions can have serious implications. In the community epidemiology study discussed earlier (18), rates of post-exertional fatigue for individuals with CFS range from 93.8% to 40.6% depending on how the question is asked. Depending on how the question is worded, inaccurate information may be obtained, which could lead to an inaccurate diagnosis. Certainly, there were several decisions that needed to be made concerning the analysis of the data. Our decision to use the psychiatric group as a comparison was made for two reasons: it is important to attempt to differentiate the symptoms of CFS from those with fatigue due to a psychiatric explanation, and the prior study comparing the ME definition to CFS had also used a psychiatric group as a control group, and by keeping the contrast groups similar, it was possible to compare the results of the present study with that of Jason et al. (18). Some readers might feel that it was a mistake to include only 12 of the original 32 Fukuda et al. (7) cases in the CFS group. In contrast, we could have compared the 32 individuals with the Fukuda et al. (7) criteria with the 23 meeting the Canadian case definition and the 33 with CF-psychiatric, and then presented the symptom profiles for these three groups. Unfortunately, such a scheme would have produced groups that were not independent, and it would have been difficult to justify statistical approaches with some cases included within more than one group. It is worth mentioning that it is quite likely that the Fukuda et al. (7) CFS group as originally constituted would have been quite different from the CF-psychiatric group. Within the Fukuda et al. (7) CFS criteria, there are probably those individuals who are more debilitated, and those who are less so, and those in the former group are more likely to meet the Canadian case definition. There were several methodological limitations in this study. Findings from this study are preliminary and should be interpreted within the context of limitations on statistical power imposed by a small sample size, particularly with regard to the smaller CFS group. Because some differences between groups may not have been detected, more research with larger samples is necessary to replicate these findings. In addition, most measures are of self-reports (such as with the symptom questionnaire), and in future studies, there is a need to examine biological markers that might also differentiate these groups. In summary, those individuals in this study meeting the Canadian criteria appear to have more symptoms, more physical functional impairment, and less psychopathology than those in the CF-psychiatric group. In addition, the Canadian criteria identifies patients with more fatigue/ weakness, neurological and neuropsychiatric symptoms than the Fukuda CFS criteria does. NOTES ```````````` 1. Only 65.2% of the participants who were diagnosed with the Canadian case definition endorsed the item postexertional malaise, which needed to occur for 24 hours after exercise. In examining these patients, the examining physician and other data within their records clearly indicated that they had postexertional malaise. In addition, in the Canadian criteria, it is indicated that the fatigue comes in many "flavours." This Canadian case definition makes that point that lack of stamina and fatigue need to be considered when assessing this dimension. 2. Only 78.3% of the participants in the Canadian criteria group indicated that they had memory and concentration problems. However, to meet the criteria for Clinical Neurocognitive difficulties, there are many symptoms that can fulfill this criteria in addition to memory and concentration problems.