Chronic fatigue syndrome (CFS) is an incapacitating illness defined by disabling chronic fatigue and characteristic accompanying symptoms [1]. CFS has no confirmatory physical signs or characteristic laboratory abnormalities and its etiology and pathophysiology remain unknown [2]. Many of the problems searching for markers of CFS reflect ambiguities of definition. Recently, an International CFS Study Group identified areas of vagueness in the case definition and proposed revisions in its utilization [3]. The Group recommended that validated standardized instruments be used to evaluate the functional impairment, fatigue, and accompanying symptoms associated with CFS. They recommended that future research utilize the Medical Outcomes Survey Short-Form-36 (SF-36) [4] to assess functional impairment and that the Checklist Individual Strength (CIS) [5] or Multidimensional Fatigue Inventory (MFI) [6] be used to measure general dimensions of fatigue. The Group was unaware of standardized and validated instruments that assessed the CFS symptom complex and suggested that investigators consider using a symptom checklist developed by the US Centers for Disease Control and Prevention (CDC) for use in population-based surveys. The CDC Symptom Inventory assesses the full range of CFS associated symptoms, has been used in several population-based studies (i.e., comparative data are available), and is publicly available. However, the Symptom Inventory has not been formally validated. The objective of the present study was to validate the Symptom Inventory as a questionnaire for evaluation of CFS associated symptoms.

Two hundred twenty-seven people participated in the 2-day clinical evaluation and participation rates (64 to 78%) were similar among the categories (p = .26). Five of the 227 were excluded because of incomplete psychiatric interviews, 29 because of exclusionary medical conditions, 3 because of psychiatric conditions, and 26 because of current major depression disorder with melancholic features, resulting in 164 subjects for analysis. Twenty-four subjects with prior fatigue (7 CFS and 17 ISF) were classified as in remission. There were no differences in age, body mass index, or sex between the classification groups (Table 1).

This study showed that the CDC CFS Symptom Inventory is a reliable and valid instrument for assessing symptoms associated with CFS. Many studies conducted in tertiary care settings have used ad hoc (non-validated) questionnaires to assess the frequency or intensity of CFS-defining symptoms. In addition to lack of validation, one unanswered question from such studies concerns what is worse – a severe symptom that occurs sporadically or a clinically minimal symptom occurring every day. The Symptom Inventory obviates this problem because it is scored as a product-term of intensity and frequency and better represents the variance of each symptom.

Both the Total and the Case Definition scores effectively assessed initial study classification as CFS, ISF, and never fatigued. They also showed excellent psychometric properties, including good internal consistency and validity. One might question the possibility of circular logic – defining an illness by its symptoms then assessing psychometric properties of a scale that measures the same symptoms. Unfortunately, as yet, CFS has no confirmatory physical signs or characteristic laboratory abnormalities [2]; so, in lieu of a 'gold standard', it is defined by disabling chronic fatigue and characteristic accompanying symptoms [1,3]. For initial classification in this study, we defined CFS by literally applying criteria of the 1994 case definition [1]. We classified subjects as CFS who stated that they had been fatigued for at least 6-months; that the fatigue severely affected their occupational, educational, social, or recreational activities; who endorsed the presence of at least 4 CFS defining symptoms; and, who had no exclusionary medical or psychiatric conditions. Subjects who had medically/psychiatrically unexplained chronic fatigue not fulfilling all these criteria were considered ISF. This operational classification notwithstanding, CFS represents a multi-faceted illness. Fatigue is a complex construct and we employed 2 standardized and validated instruments (the MFI and Chalder Fatigue Scale) to evaluate its various dimensions (e.g., physical fatigue, mental fatigue). Similarly, the impairment associated with CFS is not unidimensional; so we utilized the SF-36 to quantify the 8 major dimensions of function and wellbeing. Finally, CFS includes a characteristic symptom complex and we used the Symptom Inventory to evaluate the intensity and frequency of accompanying symptoms. The Total, Case Definition, and Short Form scores all had good convergent validity as determined by correlations with the 3 multidimensional measures of fatigue and impairment. The Symptom Inventory scores also discriminated between subgroups originally classified as CFS, ISF or not fatigued and this is a measure of the instrument's practicability for assessing fatiguing illness.

As noted above, there is no objective test to unequivocally diagnose CFS so it is premature to evaluate sensitivity or specificity of the various Symptom Inventory scores. Rather, our objective was to evaluate the Inventory's psychometric properties as baseline for its use in future studies. Studies of the clinical characteristics of CFS and other unexplained fatiguing illnesses should utilize the Symptom Inventory in conjunction with other instruments that assess different dimensions and consequences of fatigue (e.g., the MFI, SF-36) [3,10]. Beside the Symptom Inventory Total and the Case Definition scores, the Symptom Inventory Short Form score, consisting of only 6 symptoms (unusual fatigue after exertion, unrefreshing sleep, muscle aches, sleeping problems, memory and concentration problems), appears to be an economic and precise screening measurement for the current status of fatiguing illnesses.

The Symptom Inventory includes 10 symptoms that are not used to define CFS. Although not considered in the case definition, these symptoms are commonly reported by chronically ill people and have proven useful for stratification during analysis of descriptive and case control studies. In addition, the International CFS Study Group recommended additional research to further develop the CFS case definition and such research must assess a comprehensive range of symptoms. Finally, analytic studies of CFS must consider somatization disorders (both as confounders and comorbid conditions); the full range of symptoms in the Inventory is needed to categorize such disorders.

At least one important limitation must be considered. Although study subjects were recruited from the community and do not reflect the strong biases inherent of clinic patient populations, they did not represent the general population; rather they comprised a sample of people with and without unexplained fatiguing illnesses. Thus, the excellent psychometric properties of the Symptom Inventory cannot be generalized to the general population. To further validate and evaluate the Symptom Inventory, additional testing in a larger population-based sample not stratified by fatigue is required. There is also a need to determine the test-retest-reliability and stability of the Symptom Inventory. The present study provides preliminary results to encourage researchers to administer the Symptom Inventory along with other standardized questionnaires measuring fatigue and functional impairment in studies of CFS and other fatiguing illnesses.

The author(s) declare that they have no competing interests.

DW conceived of the scale to summarize symptom impact, had primary responsibility for statistical analysis and wrote the manuscript; RN collaborated in study design, in data analysis and writing the manuscript; CH was instrumental in the conception and design of the study, participated in fieldwork, collaborated in analysis and interpretation of the data, and writing the manuscript; JFJ collaborated in the clinical study and in preparation of the manuscript; ERU was instrumental in the conception and design of the study, collaborated in interpretation of the data, and writing the manuscript; WCR conceived of the study, served as principal investigator throughout its execution and collaborated in writing the manuscript.