ABSTRACT
ABSTRACT
Background: Myalgic encephalomyelitis/ chronic fatigue syndrome (ME/CFS) remains undiagnosed in up to 91% of patients. Recently, the United States-based Institute of Medicine (IOM) developed new diagnostic criteria, naming it systemic exertion intolerance disease (SEID).
Purpose: We examined how subjects fit SEID criteria and existing ME/CFS case definitions early in their illness.
Methods: A total of 131 subjects fitting 1994 Fukuda CFS criteria at the time of study recruitment completed a survey of symptoms they experienced during their first 6 months of illness. Symptoms were drawn from SEID and existing criteria (1994 Fukuda, 2003 Canadian Consensus Criteria (CCC), and 2011 Myalgic Encephalomyelitis-International Consensus Criteria (ME-ICC)). We calculated and compared the number/percentage of subjects fitting single or combinations of case definitions and the number/percentage of subjects with SEID experiencing orthostatic intolerance (OI) and/or cognitive impairment.
Results: At 6 months of illness, SEID criteria identified 72% of all subjects, similar to when Fukuda criteria (79%) or the CCC (71%) were used, whereas the ME-ICC selected for a significantly lower percentage (61%, p < .001). When severity/frequency thresholds were added to the Fukuda criteria, CCC and ME-ICC, the percentage of these subjects also fitting SEID criteria increased to 93%, 97%, and 95%. Eighty-seven percent of SEID subjects endorsed cognitive impairment and 92%, OI; 79% experienced both symptoms.
Conclusions: SEID criteria categorize a similar percentage of subjects as Fukuda criteria early in the course of ME/CFS and contain the majority of subjects identified using other criteria while requiring fewer symptoms. The advantage of SEID may be in its ease of use.
Introduction
Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a chronic, complex, disabling medical condition that is estimated to affect 836,000 to 2.5 million people in the United States, more than multiple sclerosis or systemic lupus erythematosus [1–3]. People of all ages, both sexes, and different ethnic and socioeconomic backgrounds can be afflicted; up to 69% of patients are unable to work and 25% experience periods of being homebound and/or bedridden [4–6]. Despite ME/CFS’s prevalence and severity, in large community-based epidemiology studies, 84–91% of subjects [7,8] identified by researchers with ME/CFS have not been diagnosed by their physicians and those diagnosed often relate many months or years of visiting multiple clinicians [9] before arriving at an answer. Obtaining a diagnosis is reported as ‘the single most helpful event’ by 90% of patients affected by ME/CFS, allowing them to be less anxious or fearful, formulate ways to cope with their symptoms, and communicate with healthcare professionals and others [10].
One barrier to diagnosis may be the current case definition for ME/CFS. The Fukuda case definition [11] requires severe function-limiting fatigue of at least 6 months’ duration accompanied by a minimum of four out of eight secondary symptoms: multi-joint pain, muscle pain, sore throat, unrefreshing sleep, new headaches, problems with concentration and memory, tender cervical/ axillary lymph nodes, and post-exertional malaise (PEM) (prolonged, sometimes delayed, exacerbation of a patient’s baseline symptoms after physical or cognitive activity). Because most of these symptoms are commonly encountered in medical practices [12], clinicians are advised to form a differential diagnosis and rule out other potential causes first. Many medical conditions or situations (e.g. aging, overwork, work-related stress) can lead to these symptoms, causing some clinicians to hesitate diagnosing patients with ME/CFS.
To better specify illness criteria, experienced researchers and clinicians constructed alternative clinical case definitions in 2003 (Canadian Consensus Criteria (CCC)) [13] and 2011 (Myalgic Encephalomyelitis-International Consensus Criteria (ME-ICC)) [14]. These definitions require PEM, not fatigue, as the hallmark symptom of ME/CFS. Additional neurological and other symptoms observed in ME/CFS patients are also required. By comparison, Fukuda’s secondary symptoms are optional rather than required (any 4 or more out of 8 symptoms are deemed appropriate), meaning that study subjects can fit Fukuda via 163 different combinations of symptoms [15]. These symptom combinations may not be equivalent to one another; yet, in most studies, subjects have been analyzed as one large group rather than subgrouped by symptoms or other characteristics. Nevertheless, the Fukuda criteria remain the definition overwhelmingly used by United States-based and international clinicians and researchers.
Consequently, in 2013, the US Department of Health and Human Services, following a recommendation made by the US Chronic Fatigue Syndrome Advisory Committee [16], requested that the IOM develop evidence-based diagnostic criteria for use in clinical settings. The IOM report, ‘Beyond Myalgic Encephalomyelitis/ Chronic Fatigue Syndrome: Redefining an Illness’, released in February 2015, recommended new diagnostic criteria attached to a new name, systemic exertion intolerance disease (SEID) [17]. Patients must have three core symptoms (PEM, unrefreshing sleep, and substantial decrease in function accompanied by fatigue) and either neurocognitive symptoms or OI. Functional disability must have lasted for a minimum of 6 months and three symptoms (PEM, unrefreshing sleep, and cognitive impairment) must be of at least moderate intensity and present for 50% or more of that time.
The symptoms chosen are strikingly similar to the four core symptoms (fatigue, PEM, neurocognitive issues, and unrefreshing sleep) identified empirically from ME/CFS studies [18] to help distinguish patients from controls.
Since the release of the report, questions have been raised concerning how many Fukuda-, ME-ICC-, or CCC-diagnosed patients will fit the new criteria. As the new criteria were specifically constructed to facilitate earlier and more accurate diagnosis, we were also interested in how it and other criteria would perform during the beginning stages of ME/CFS. This paper attempts to answer those questions by examining a database of subjects initially selected by Fukuda criteria and assessing whether and how they fit SEID, ME-ICC, and CCC criteria at 6 months of illness.
Materials and methods
Participants
From March 2010 to August 2011, we recruited 200 ME/CFS subjects residing in the San Francisco Bay area as part of our Stanford University GEISD (Genetic Expression and Immune System Dynamics) study. These 200 subjects were re-contacted from January 2013 to July 2013 and asked if they wished to participate in our survey. This project was reviewed and approved by the Stanford University Institutional Review Board. Informed written consent was obtained and subjects were given an individualized secured hyperlink to access the survey. A paper version of the survey was also offered in case of technical or cognitive difficulties. All subjects were at least 14 years old, non-pregnant, able to speak English, and had not been exposed to more than 2 weeks of antimicrobial use recently. Subjects were screened for Fukuda 1994 CFS criteria using a standardized telephone interview (see Supplementary Materials). Subjects were excluded if they reported one of the following: (1) alternative medical or psychiatric conditions that could explain their ME/CFS symptoms, (2) substance abuse issues in the last year (not including nicotine/ caffeine), (3) an influenza vaccination within the past 4 weeks, or (4) certain immunological conditions.
Survey materials and scoring
In 2012, we used the Research Electronic Data Capture (REDCap) web application ((http://project-redcap.org/) to design an online survey incorporating the 54-item symptom scale (Supplementary Materials, Figure 2) of the DePaul Symptom Questionnaire (DSQ) [19,20]. (http://condor.depaul.edu/ljason/cfs/) The DSQ was devised to elicit most symptoms covered by the Fukuda, CCC, and ME-ICC criteria and is currently being used by the US Centers for Disease Control and Prevention for their Multi-Site Clinical Assessment of CFS study [21]. Each item is linked to one symptom; subjects assess the severity and frequency of each symptom over the last 6 months using a 0–4 score, with 0 meaning absence of a symptom and higher scores denoting higher intensity or more frequent occurrence. For this study, we modified the time period of the question stem to cover the first rather than the last 6 months of a subject’s illness. Although the DSQ has not yet been tested using this time period, we chose to use a modified version of it because of its inclusion of most symptoms covered in different criteria and so that we might compare our results to those of prior studies that had employed the DSQ.
To evaluate areas not covered by the DSQ but present in at least one of the four sets of criteria (substantial reduction in function, OI, problems with infections, and temperature intolerance; see Supplementary Materials, Table 1), we assessed ‘Yes’ and ‘No’ responses to five other questions in the survey (Supplementary Materials, Non-DSQ items).
After completion, subjects submitted the survey electronically or mailed the survey back to staff. Responses from the paper version were then entered manually into REDCap by study staff.
Data were extracted from REDCap and exported into Microsoft Excel 2013 files. For all case definitions, subjects answering, after onset of their illness that they had reduced or would, if they could, reduce their activities to 50% or less of their pre-illness levels were considered substantially functionally impaired. Subjects answering ‘Yes’ to non-DSQ items were counted as endorsing the corresponding symptom. Otherwise, operationalization of each case definition is described in the Supplementary Materials, Table 1. Where a specific case definition did not indicate the number of symptoms needed to qualify for a category (e.g. ‘sleep dysfunction’ in the CCC), only one symptom within that category was required to screen positive for that category.
For SEID criteria, where ‘profound fatigue’ and unrefreshing sleep, PEM, and cognitive difficulties of moderate intensity must exist for 50% or more of the time to qualify, we counted these symptoms as significant only when they were rated by subjects as equal to two or more on both the DSQ intensity and frequency scales (‘2/2’ scoring), equivalent to at least ‘moderate intensity’ occurring ‘about half the time’. For Fukuda criteria, where some published versions state ‘fatigue’ while others state ‘severe fatigue’, we have operationalized it in two ways: (1) requiring mere presence of fatigue, equivalent to at least a ‘1/1’ score (i.e. ‘mild/ a little of the time’) and (2) requiring a ‘2/2’ score to denote ‘severe fatigue’ (designated ‘Fukuda-severe fatigue’). For the CCC and ME-ICC, severity and frequency of symptoms are not specified in the original publications so we have evaluated symptoms using both the ‘1/1’ and ‘2/2’ scoring methods.
Data analysis
Initially, all subjects were classified by the four sets of criteria. Next, taking only the subjects who qualified for Fukuda or Fukuda-severe fatigue at 6 months of illness, we explored how these subjects fit other criteria. Venn diagrams illustrating the number of subjects fitting single and different combinations of criteria (Supplementary Materials, Figures 3–8) were generated using web-based applications provided by Chris Seidel (http://www.pangloss.com/seidel/Protocols/venn4.cgi) and the Department of Bioinformatics and Evolutionary Genomics at the University of Gent, Belgium (http://bioinformatics.psb.ugent.be/webtools/Venn/). Since the new SEID criteria compared to currently available criteria were of greatest interest, our analysis focused on how subjects fit pairs rather than triads, quartets, or quintets of criteria.
Table 1. Number and percentage of subjects fulfilling different criteria at 6 months of illness using 1/1 scoring.
Table 2. Number and percentage of subjects fulfilling different criteria at 6 months of illness using 2/2 scoring.
Table 3. Number and percentage of subjects, who, at 6 months of illness, fit Fukuda and other criteria.
Table 4. Number and percentage of subjects, who, at 6 months of illness, fit Fukuda-severe fatigue and other criteria.
Table 5. Number and percentage of subjects who fulfill either CCC or ME-ICC criteria and SEID criteria at 6 months of illness.
Using the web-based statistical application VassarStats (http://vassarstats.net/), we tested for significant differences between proportions of subjects fitting pairs of criteria using McNemar’s test. Correcting for multiple comparisons using Bonferroni’s method, we deemed a two-tailed p-value of equal to or less than 0.005 significant.
We also evaluated how the presence of OI and cognitive impairment contributed to subjects qualifying for SEID by calculating the number and percentage of subjects fitting SEID criteria endorsing one or both of these symptoms.
Results
A total of 150 out of the original 200 subjects (75%) responded to the survey. Based on screening interview questions (data not shown), respondents were more likely than non-respondents to self-identify as Caucasian (94% vs. 84%, p = .04) and endorse tender lymph nodes or PEM (67% vs. 48%, p < .001; 98% vs. 92%, p = .04). There were no significant differences between responders and non-responders in mean age, duration of illness, fatigue scores (using the Fatigue Severity Scale (FSS) and Multidimensional Fatigue Inventory (MFI-20), and self-assessed cognitive/physical functioning as a percentage of pre-illness functioning, or prevalence of women, viral onset, and other Fukuda criteria symptoms (data not shown).
Of these 150, 131 subjects (87%) had enough information available to conduct the analyses in this study. These subjects were predominantly female (80%), were a mean 51.3 ± 12.5 (standard deviation) years of age (range: 19.9–74.6), and had been sick for a mean of 15.4 ± 10.3 years. Their mean total MFI-20 score was 74.1 ± 13.6 (0–100 range with higher scores designating more fatigue) and their mean FSS score was 6.0 ± 1.1 (0–7 range, with higher scores denoting more function-limiting fatigue). There were no statistically significant differences between the survey completers and non-completers with regard to percentage of women (84% for non-completers), mean age (54.2 ± 11.6 years), mean duration of illness (13.5 ± 8.3), mean MFI-20 score (70.8 ± 13.8) and mean FSS score (5.5 ± 1.4).
At 6 months of illness, the four sets of criteria (Fukuda, ME-ICC, CCC, and SEID), operationalized in different ways, captured from 45% to 79% of all subjects ( and ). SEID criteria covered approximately the same percentage of subjects (72%) as either version of Fukuda (fatigue presence only = 79%/ ‘severe fatigue’ required = 72%) and the 1/1 version of the CCC (71%) despite mandating fewer symptoms and requiring higher severity and frequency thresholds. Since the CCC scored in a 2/2 manner (‘CCC 2/2’) and the ME-ICC criteria necessitated more symptoms, a significantly lower percentage of subjects, 56% and 61% (p < .001), respectively, conformed to them compared to Fukuda criteria (72–79%).
Restricting the analysis to subjects who qualified for the Fukuda 1994 case definition at 6 months of illness, the other sets of criteria selected for 56–93% of these subjects ( and ). Out of 104 subjects fitting Fukuda criteria, SEID encompassed a similar percentage of subjects (84%) as the highest-performing criteria, CCC 1/1 (89%) (). Compared to CCC 1/1, CCC 2/2 and both versions of the ME-ICC covered a significantly lower percentage of subjects, ranging from 56% to 74% (p ≤ .004). When the 94 subjects fulfilling Fukuda-severe fatigue criteria were examined, the same percentage of subjects (93%) fitted SEID and CCC 1/1 (). In contrast, compared to both SEID and CCC 1/1, the CCC 2/2 and both versions of the ME-ICC covered a significantly lower percentage of subjects, from 61% to 76% (p ≤ .002). Interestingly, when the fatigue component of Fukuda criteria did not count unless it attained certain thresholds, the percentage of subjects fitting both other criteria and Fukuda-severe fatigue criteria increased rather than decreased. The largest increase was seen with SEID criteria, which covered 84% of subjects fitting Fukuda versus 93% fitting Fukuda-severe fatigue.
As expected, when severity/ frequency thresholds were put in place, the percentage of subjects fulfilling CCC, or ME-ICC criteria fell compared to when only presence of symptoms were required. For example, 89% of subjects qualifying for Fukuda criteria also fulfilled CCC scored in a 1/1 manner, but this decreased to 70% when CCC was scored in a 2/2 manner ().
The opposite pattern emerged though when we examined how subjects fitting CCC or ME-ICC criteria fulfilled SEID criteria. Generally, SEID covered the overwhelming majority, from 87% to 97%, of CCC- and ME-ICC-qualifying subjects (). Stricter symptom thresholds for the CCC and ME-ICC in this situation yielded a higher percentage of subjects qualifying for both criteria even as the absolute number of CCC or ME-ICC subjects fell, with the highest figure of 97% referring to subjects fitting SEID and CCC 2/2 criteria.
When the four sets of criteria, as originally written (i.e. with symptom frequency and severity thresholds required only for SEID criteria), were applied, 64 subjects (49%) qualified for all four case definitions in the first 6 months; 18 subjects (14%) qualified for none of the criteria and 4 subjects (3%) qualified for Fukuda but none of the other criteria (Supplementary Materials, Figure 3). When severity and frequency thresholds were attached to all criteria (i.e. also mandating ‘2/2’ criteria for symptoms associated with the ME-ICC and CCC, ‘2/2’ level of fatigue for Fukuda criteria), 50 subjects (38%) fit all four definitions; 28 (21%), none of the criteria; and 5 (4%), Fukuda-severe fatigue but none of the others (Supplementary Materials, Figure 4).
The majority of subjects (79%) fulfilling SEID criteria experienced both OI and cognitive impairment; 13% endorsed only OI while 8% endorsed only cognitive impairment. The prevalence of cognitive impairment was 87%, whereas that of OI was 92%.
Discussion
These findings may allay concerns that implementation of SEID criteria will exclude many subjects that fit any of the existing criteria. Despite requiring fewer symptoms and a higher degree of symptom severity and frequency, at 6 months of illness, SEID criteria identify a similar, high percentage of subjects as Fukuda or the other high-performing alternative criteria, CCC 1/1 ( and ). When assessed for overlap with Fukuda or other criteria (–), SEID criteria cover 84%–97% of subjects selected using other criteria. Furthermore, the rise in the percentage of subjects SEID covers when severity and frequency thresholds are instituted for the Fukuda criteria, CCC, and ME-ICC suggests that SEID may select for a sicker subset of subjects. In contrast, applying the 2/2 version of the CCC or either version of the ME-ICC excluded significantly more subjects (–).
Overall, our results are consistent with other studies where recruited subjects carried a clinician diagnosis of ME/CFS. Jason et al. found that SEID criteria described a similarly high percentage of subjects as Fukuda criteria (88% vs. 92%) and that subjects meeting SEID criteria often met Fukuda criteria and the CCC as well (98% and 85%, respectively) [22]. Furthermore, Jason showed that 99.1% of subjects fitting CCC under the 2/2 level of symptoms also qualified for the SEID criteria [23], similar to our figure of 97%. The low-to-moderate percentages of subjects meeting CCC and ME-ICC in our study reflect those of other studies, where 52.5–77.2% of subjects diagnosed with ME/CFS met the CCC and 48.9–62% met the ME-ICC [22–26].
In a separate study [27], Jason et al. found that while SEID criteria identified 75–100% of subjects fitting Fukuda criteria and very few healthy controls, 24–47% of subjects with other illnesses (multiple sclerosis, major depression, and lupus) endorsed SEID criteria symptoms. Consequently, it was concluded that SEID criteria are sensitive but not specific. However, we would advise caution with this interpretation as the lower number of items and less distinctive symptoms used to represent PEM in Jason’s study do not reflect the SEID description of PEM. Furthermore, diagnosis of the medical conditions used for comparison were likely not based solely on patient report of symptoms but also through confirmatory testing and response to treatment.
Similarly, discrepancies in the prevalence of OI between our and Jason’s studies could be due to the less common symptoms (e.g. shortness of breath) they chose to represent OI and their institution of minimum severity/frequency thresholds for OI when the SEID criteria do not obligate such parameters[1]. OI affected 92% of our subjects, similar to the 80% prevalence recorded by Lapp et al. [28] through tilt table testing, and permitted an additional 13% our of subjects, who did not have cognitive impairment, to qualify for SEID. In contrast, 67% of Jason’s subjects [23] were afflicted by OI and OI symptoms only contributed an additional 2% of subjects being diagnosed with SEID. OI is important to recognize because it can be confirmed objectively and, in some cases, can be treated effectively, leading to higher function and quality of life [29,30].
Strengths of this study include use of a well-known symptom questionnaire and the addition of survey questions permitting comparison of all four different case definitions. Our querying of symptoms within the first 6 months of illness rather than later in the course of the illness is a unique feature. Very few studies [31,32] have explored symptomatology and none have assessed how case definitions perform during the early stages of ME/CFS.
Limitations
Limitations of this study include convenience sampling; the inability to confirm the diagnosis or symptoms of subjects with objective methods; utilization of a time period in the DSQ which has not yet been tested; imperfect operationalization of criteria (e.g. DSQ items may not correspond well to some criteria symptoms); and potential recall issues due to the retrospective nature of the survey. Since the rate of diagnosis of ME/CFS is so low and the current diagnostic process complicated or prolonged, recruiting subjects via convenience samples or screening questionnaires is not uncommon and sometimes logistically necessary. Patient recall can be affected by both forgetting and adding symptoms and features to their original clinical presentation. However, being afflicted by ME/CFS is often a life-changing event. Some studies suggest the more salient an event to a person’s life, the more likely they are to recount it accurately [33]. Using an untested time period in the DSQ might affect results but allowed us to compare our results to some degree with prior studies that had also assessed the overlap of different case definitions. Finally, results might have been different had we employed different inclusion criteria, for example, enrolling all subjects with a clinician diagnosis of ME/CFS, regardless of case definition used. Nevertheless, despite these drawbacks, our results reflect and reinforce those of prior studies.
Typical of many ME/CFS studies, this study’s subjects were primarily middle-aged, self-identified Caucasian women who had been sick for over a decade and were being seen at a tertiary-level clinic. Studies have consistently demonstrated that more women are afflicted by ME/CFS than men at a roughly 2–3:1 ratio, suggesting that this may be an intrinsic feature of the illness rather than an artifact of our clinical setting [34]. On the other hand, some studies suggest that ME/CFS prevalence and severity may be increased in non-Caucasian (Hispanic-Americans, African-Americans, and United Kingdom residents of Black and minority ethnic backgrounds) and economically poorer groups [35,36]. Thus, our results may not be as generalizable to patients who are male, younger, and of certain ethnic minority backgrounds; have been sick for a short period of time; or are seen in a community-based setting.
None of the four sets of criteria, operationalized in any way, initially identified more than 80% of all study subjects. This is not surprising as 30–80% of subjects from community-based studies report a gradual rather than acute onset of the illness [7,37]. Although most studies do not define the terms ‘acute’ or ‘gradual,’ in two different samples recruited with Fukuda criteria, 43% and 64% experienced illness onset occurring over a period of greater than 6 months, suggesting that some Fukuda-related symptoms may not develop until later [38]. Furthermore, we found that 14% and 21% of all subjects did not qualify for any of the four criteria during the first 6 months. Consequently, while current criteria are useful, there need to be further studies investigating which symptoms, features, and/or objective tests might allow for early, accurate diagnosis of more patients.
Conclusions
Using existing data, SEID criteria appear to perform similarly to the most employed criteria, that is, Fukuda, and yet also cover a large percentage of those qualifying for other criteria. These results may help to allay concerns that application of the new criteria will exclude many patients diagnosed via any of the current criteria. Since some subjects did not fit SEID criteria, we recommend that other criteria be retained for the time being and not discarded yet. However, neither the CCC nor ME-ICC should be solely adopted for clinical diagnosis. Our data indicate that many subjects fitting Fukuda criteria would be missed, especially if symptom frequency/ severity thresholds are established as suggested by Jason et al. [39] Furthermore, the SEID criteria appear to be more user-friendly as a fewer number of symptoms are needed for diagnosis. The complexity of the CCC and ME-ICC are exemplified by the convoluted algorithms (Supplementary Materials, Table 1) that had to be constructed to classify subjects. These criteria would be even more difficult to apply in a busy clinical setting. Lastly, the number of categories and the diverse symptoms classified under each CCC or ME-ICC category do not solve the problem of heterogeneity introduced by the polythetic nature of Fukuda criteria. Instead, permitting even more categories and symptoms will lead to increased heterogeneity. Future studies should attempt to validate the SEID criteria and address further how SEID relates to the other definitions, for example, whether it is an entirely separate disease from, a subgroup of, or the severe end of subjects fitting Fukuda criteria.
| Criteria | Number of subjects fulfilling criteria at 6 months | Number of subjects fulfilling Fukuda at study recruitment | Percentage of subjects fulfilling criteriaa |
|---|---|---|---|
| Fukuda | 104 | 131 | 791,2,3 |
| Fukuda-severe fatigue | 94 | 131 | 721 |
| CCCb | 93 | 131 | 712 |
| ME-ICCb | 80 | 131 | 613 |
| SEID | 94 | 131 | 72 |
Note: Fukuda: Fukuda 1994 criteria; Fukuda-severe fatigue: Fukuda 1994 criteria with fatigue required to be of at least moderate severity and occurring at least 50% of the time; CCC: Canadian Consensus Criteria; ME-ICC: Myalgic Encephalomyelitis – International Consensus Criteria; SEID: systemic exertion intolerance disease.
aPercentages were calculated by taking the number of subjects fulfilling a specific criterion and dividing it by 131. Pairs of categories exhibiting significantly different percentages are designated by matching superscript numbers: 1Fukuda vs. Fukuda-severe fatigue (p = .002), 2Fukuda vs. CCC (p = .001), 3Fukuda vs. ME-ICC (p < .001).
bFor 1/1 scoring, CCC or ME-ICC symptoms merely needed to be present.
| Criteria | Number of subjects fulfilling criteria at 6 months | Number of subjects fulfilling Fukuda at study recruitment | Percentage of subjects fulfilling criteriaa |
|---|---|---|---|
| Fukuda | 104 | 131 | 792,5 |
| Fukuda-severe fatigue | 94 | 131 | 723,6 |
| CCCb | 73 | 131 | 561,2,3,4 |
| ME-ICCb | 59 | 131 | 451,5,6,7 |
| SEID | 94 | 131 | 724,7 |
Note: Fukuda: Fukuda 1994 criteria; Fukuda-severe fatigue: Fukuda 1994 criteria with fatigue required to be of at least moderate severity and occurring at least 50% of the time; CCC: Canadian Consensus Criteria; ME-ICC: Myalgic encephalomyelitis – International Consensus Criteria; SEID: systemic exertion intolerance disease.
aPercentages were calculated by taking the number of subjects fulfilling a specific criterion and dividing it by 131. Pairs of categories exhibiting significantly different percentages are designated by matching superscript numbers, all at p < .001: 1CCC vs. ME-ICC, 2Fukuda vs. CCC, 3Fukuda-severe fatigue vs. CCC, 4SEID vs. CCC, 5Fukuda vs. ME-ICC, 6Fukuda-severe fatigue vs. ME-ICC, 7SEID vs. ME-ICC. The percentage of subjects fitting ME-ICC and CCC assessed by 2/2 scoring also differed significantly from the percentages of subjects fitting these same definitions when 1/1 scoring was employed.
bFor 2/2 scoring, CCC or ME-ICC symptoms needed to be of at least moderate intensity and occurring at least 50% of the time to count.
| CCC and ME-ICC 1/1 scoringa | CCC and ME-ICC 2/2 scoringb | |||||
|---|---|---|---|---|---|---|
| Combinations of Criteriac | Number of subjects fulfilling criteria combination at 6 months | Number of subjects fulfilling Fukuda at 6 months | Percentage of subjects fulfilling criteria combinationse,f | Number of subjects fulfilling criteria combination at 6 months | Number of subjects fulfilling Fukuda at 6 months | Percentage of subjects fulfilling criteria combinatione,f |
| Fukuda and CCC | 93 | 104d | 891,4,5 | 73 | 104 | 702,5 |
| Fukuda and ME-ICC | 77 | 104 | 741,6 | 58 | 104 | 563,4.6 |
| Fukuda and SEID | 87 | 104 | 842,3 | – | – | – |
Note: Fukuda: Fukuda 1994 criteria; CCC: Canadian Consensus Criteria; ME-ICC: Myalgic encephalomyelitis – International Consensus Criteria; SEID = systemic exertion intolerance disease.
aFor 1/1 scoring, CCC or ME-ICC symptoms merely needed to be present.
bFor 2/2 scoring, CCC or ME-ICC symptoms needed to be of at least moderate intensity and occurring at least 50% of the time to count.
cSubjects fulfill both criteria at 6 months of illness.
d104 out of the total of 131 subjects fitting Fukuda criteria at the time of study recruitment also qualified for Fukuda criteria at 6 months of illness.
eThe percentage refers to the number of subjects fitting combinations of criteria divided by 104. E.g. ‘Fukuda and CCC (scored in a 1/1 manner)’ percentage is calculated from 93 divided by 104.
fCategories exhibiting significantly different percentages are designated by matching superscript number, preceded by corresponding p-values.
p = .004: 1Fukuda and CCC 1/1 vs. Fukuda and ME-ICC 1/1.
p = .001: 2Fukuda and SEID vs. Fukuda and CCC 2-2.
p < .001: 3Fukuda and SEID vs. Fukuda and ME-ICC 2/2; 4Fukuda and CCC 1/1 vs. Fukuda and ME-ICC 2/2; 5Fukuda and CCC 1/1 vs. Fukuda and CCC 2/2; 6Fukuda and ME-ICC 1/1 vs. Fukuda and ME-ICC 2/2.
| CCC and ME-ICC 1/1 scoringa | CCC and ME-ICC 2/2 scoringb | |||||
|---|---|---|---|---|---|---|
| Combinations of Criteriac | Number of subjects fulfilling criteria combinations at 6 months | Number of subjects fulfilling Fukuda-severe fatigue at 6 months | Percentage of subjects fulfilling criteria combinationse,f | Number of subjects fulfilling criteria combinations at 6 months | Number of subjects fulfilling Fukuda-severe fatigue at 6 months | Percentage of subjects fulfilling criteria combinationse,f |
| Fukuda-severe fatigue and CCC | 87 | 94d | 932,5,6 | 73 | 94 | 781,5 |
| Fukuda-severe fatigue and ME-ICC | 71 | 94 | 762,3,7 | 57 | 94 | 614,6,7 |
| Fukuda-severe fatigue and SEID | 87 | 94 | 931,3,4 | – | – | – |
Note: Fukuda: Fukuda 1994 criteria; Fukuda-severe fatigue: Fukuda 1994 criteria with fatigue required to be of at least moderate severity and occurring at least 50% of the time; CCC: Canadian Consensus Criteria; ME-ICC = Myalgic encephalomyelitis – International Consensus Criteria; SEID: systemic exertion intolerance disease
aFor 1/1 scoring, CCC or ME-ICC symptoms merely needed to be present.
bFor 2/2 scoring, CCC or ME-ICC symptoms needed to be of at least moderate intensity and occurring at least 50% of the time to count.
cSubjects fulfill both criteria at 6 months of illness.
d94 out of the total of 131 subjects fitting Fukuda criteria at the time of study recruitment also qualified for Fukuda-severe fatigue criteria at 6 months of illness.
eThe percentage refers to the number of subjects fitting combinations of criteria divided by 94. E.g. ‘Fukuda-severe fatigue and CCC (scored in a 1/1 manner)’ percentage is calculated from 87 divided by 94.
fCategories exhibiting significantly different percentages are designated by matching superscript number, preceded by corresponding p-values.
p = .002: 1Fukuda-severe fatigue and SEID vs. Fukuda-severe fatigue and CCC 2/2.
p = .001: 2Fukuda-severe fatigue CCC 1/1 and vs. Fukuda-severe fatigue and ME-ICC 1/1.
p < .001: 3Fukuda-severe fatigue and SEID vs. Fukuda-severe fatigue and ME-ICC 1/1; 4Fukuda-severe fatigue and SEID vs. Fukuda-severe fatigue and ME-ICC 2/2; 5Fukuda-severe fatigue and CCC 1/1 vs. Fukuda-severe fatigue and CCC 2/2; 6Fukuda-severe fatigue and CCC 1/1 vs. Fukuda-severe fatigue and ME-ICC 2/2; 7Fukuda-severe fatigue and ME-ICC 1/1 vs. Fukuda-severe fatigue and ME-ICC 2/2
| CCC and ME-ICC 1/1 scoringa | CCC and ME-ICC 2/2 Scoringb | |||||
|---|---|---|---|---|---|---|
| Combinations of Criteriac | Number of subjects fulfilling criteria combinations at 6 months | Number of subjects used to calculate percentaged | Percentage of subjects fulfilling criteria combinationse | Number of subjects fulfilling criteria combinations at 6 months | Number of subjects used to calculate percentage | Percentage of subjects fulfilling criteria combinations |
| CCC and SEID | 81 | 93 | 87 | 71 | 73 | 97 |
| ME-ICC and SEID | 70 | 80 | 88 | 56 | 59 | 95 |
Note: CCC: Canadian Consensus Criteria; ME-ICC = Myalgic encephalomyelitis – International Consensus Criteria; SEID = systemic exertion intolerance disease.
aFor 1/1 scoring, CCC or ME-ICC symptoms merely needed to be present.
bFor 2/2 scoring, CCC or ME-ICC symptoms needed to be of at least moderate intensity and occurring at least 50% of the time to count.
cSubjects fulfill both criteria at 6 months of illness.
dThis number varies depending on the number of subjects fulfilling CCC and ME-ICC criteria, scored in a 1/1 or 2/2 manner, at 6 months. For example, the number ‘93’ refers to the number of subjects who fulfilled CCC criteria scored in a 1/1 manner at 6 months of illness.
eThe percentage refers to the number of subjects fitting the combination of criteria divided by the first criteria. e.g. ‘CCC (scored in a 1/1 manner) and SEID’ percentage is calculated from 81 divided by 93.
Supplementary_Materials.docx
Download MS Word (2284 KB)Acknowledgements
All authors were involved in the design of the study and writing of the manuscript; JLN and IV recruited subjects and collected the data, LC analyzed the data, and JGM and LC interpreted the results. The authors thank Dr Leonard Jason and Madison Sunnquist of DePaul University for their preliminary analysis of the data, Luciana Mendiola for her editorial expertise, and Donn Garvert, MS for his guidance on statistical analysis.
Disclosure statement
Dr. Chu co-authored the 2015 Institute of Medicine report, ‘Beyond myalgic encephalomyelitis/ chronic fatigue syndrome: redefining an illness.’