Abstract
Background
Myalgic encephalomyelitis (ME) and chronic fatigue syndrome (CFS) have case definitions with varying criteria, but almost all criteria require an individual to have a substantial reduction in activity level. Unfortunately, a consensus has not been reached regarding what constitutes substantial reductions. One measure that has been used to measure substantial reduction is the Medical Outcomes Study Short Form-36 Health Survey (SF-36).[1]
Purpose
The current study examined the relationship between the SF-36, a measure of current functioning, and a self-report measure of the percent reduction in hours spent on activities.
Results
Findings indicated that select subscales of the SF-36 accurately measure significant reductions in functioning. Further, this measure significantly differentiates patients from controls.
Conclusion
Determining what constitutes a significant reduction in activity is difficult because it is subjective to the individual. However, certain subscales of the SF-36 could provide a uniform way to accurately measure and define substantial reductions in functioning.
Keywords: myalgic encephalomyelitis, chronic fatigue syndrome, substantial reduction, functioning
Myalgic encephalomyelitis (ME) and chronic fatigue syndrome (CFS) are characterized by debilitating fatigue, cognitive impairment, immune dysfunction, severe exhaustion after exertion, and substantial reduction in functioning.[2] There have been over 20 case definitions developed for this illness, each with varying diagnostic criteria leading to difficulty defining the illness and identifying patients correctly.[3] A commonality among the most widely used case definitions is the requirement to have a substantial reduction in activity. Almost everyone experiences fatigue at some point in their life resulting in reduced functioning; approximately 20% of men and 30% of women in the general population report frequent tiredness.[4] It is therefore critical to have an operationalized definition of substantial reduction to differentiate patients from healthy controls with fatigue.
Of the case definitions that require a substantial reduction in functioning, many do not explicitly state how the reduction should be operationalized. For example, the Fukuda et al. [2] criteria state that there must be a substantial reduction in social, occupational, and family activities. However, no formal criteria or specific instructions are given to define what ‘substantial’ means or how to measure it. The Canadian criteria [5] propose that one must have at least a 50% reduction in pre-illness activities, but offers no clear indication of how to measure this reduction.
The ME-ICC [6] criteria use varying levels of reduction in functioning to determine the severity of impairment. The criteria state that one must have substantial reductions from pre-illness activity levels: ‘mild’ defined as a 50% reduction, ‘moderate’ defined as being unable to function outside of the home, and ‘severe’ defined as bed-ridden and unable to take care of one's own basic needs. Carruthers et al. [7] created a primer for medical practitioners to use for diagnosing ME in children and adults that extended the ME-ICC criteria by creating severity subtypes: ‘mild’ defined as one who meets all other criteria and is experiencing a significant reduction in activity, ‘moderate’ defined as a 50% reduction, ‘severe’ defined as unable to function outside of the home, and ‘very severe’ defined as bed-ridden and unable to take care of one's own basic needs.
Some attempts have been made to provide more specific, measurable guidelines for operationalizing the substantial reduction in activity criteria. Recently, the Institutes of Medicine (IOM) [8] released a case definition for Systemic Exertion Intolerance Disease, a proposed name for ME and CFS by the committee. The case definition included a substantial decrease in functioning. The IOM operationalized this as severe personal and household limitations, career loss, being housebound, or having reduced social activities. It is unclear what constitutes severe limitations, and patients could have a variety in what type of activities are increased or decreased. Further, career loss is subjective to those that were working prior to illness onset. Finally, a patient may not be predominantly housebound yet still have a substantial reduction in functioning.
Several studies have used the Medical Outcomes Study Short Form-36 Health Survey (SF-36), created by Ware and Sherbourne,[1] to attempt to operationalize this diagnostic criterion. The SF-36 is a self-report measure of an individual's current functioning and disability, which has shown good discriminant validity among subscales and internal consistency.[9] The SF-36 consists of eight subscales: Physical Functioning, Role-Physical, Bodily Pain, General Health Functioning, Vitality, Social Functioning, Role-Emotional, and Mental Health Functioning. Higher scores indicate better functioning, while lower scores indicate greater disability. Reeves et al. [10] define substantial reductions in activity using the SF-36 and cut-off scores from at or below the 25th percentile of the general population in the U.S. on at least one of the following: Physical Functioning (≤ 70), Role-Physical (≤ 50), Social Functioning (≤ 75), and Role-Emotional (≤ 66.7). Jason et al. [11] examined the inclusion of the Role-Emotional subscale as it made it possible for an individual to meet the substantial reduction in activity criterion without displaying any reduction in physical functioning. To demonstrate why this is problematic, Jason et al. [11] found that almost all of those with clinical depression would meet the disability criterion for ME and CFS, in part due to the inclusion of the Role-Emotional subscale. Further, a study by Jason et al. [12] found that the Role-Emotional subscale had the lowest level of sensitivity and specificity, indicating that the Role-Emotional subscale was least effective in differentiating patients from controls.
Additionally, Jason et al. [13] compiled an updated version of the Canadian Consensus Criteria, and recommended the use of certain subscales of the SF-36 to measure substantial reduction in activity. According to this modified case definition, individuals must score at or below specified cut-off scores on at least two of the following three SF-36 subscales: Vitality, Social Functioning, and Role-Physical. This suggestion was supported by Jason et al. [12] who comprehensively reviewed published studies that incorporated the use of the SF-36. They found that the greatest differences between patient and control mean scores on the SF-36 were on the Role-Physical, Social Functioning, and Vitality subscales. To further explore the ability of the SF-36 subscales to accurately reflect reductions in activity, they also used a Receiver Operating Curve (ROC) analysis with a combined patient sample of a tertiary care and community-based populations compared to a healthy control group. Vitality (≤ 35.0), Social Functioning (≤ 62.5), and Role-Physical (≤ 50.0) emerged as the subscales with the highest sensitivity and specificity. This result indicated that using these three subscales of the SF-36 as an indicator of substantial reduction in activities would be permit optimal differentiation between patients with ME and CFS and individuals with exclusionary conditions.
One limitation of Jason et al.'s [12] study is that patient functioning was compared to controls, rather than to their own premorbid functioning level. Further, the SF-36 is a measure of current functioning. As patients could have begun with widely different pre-illness functioning, it is important to assess differences in functioning for each patient according to their own prior level of function as opposed to comparing patient levels to those of healthy controls.
In a later study, Schafer et al. [14] assessed the relationship between the SF-36 and the hours patients spent on different activities (work, social, family, and household) pre-illness and currently (i.e., living with a current diagnosis of ME or CFS). Schafer et al. found that there was a significant, positive correlation between current work and the SF-36 subscales: Vitality, Social Functioning, and Role-Physical. This result demonstrates that patients were able to spend more hours on work when their functioning levels were higher in these subscales, suggesting that current work hours may provide a helpful indicator of disability. However, this study did not examine the relationship between the SF-36 and a reduction in functioning, rather only the current and premorbid hours spent on different activities separately. Another limitation is that hours spent on activities were assessed according to type of activity, and does not include a cumulative total of hours spent on all activities. It is important to consider the total hours spent on all activities as patients may increase or decrease hours spent on types of activities in very different ways. For example, a patient may not be able to work due to the severity of symptoms. In doing so, they may rely more on family, thus decreasing their hours spent on work and increasing their hours spent on family activities. In order to capture a holistic view of patient activity levels, it is therefore important to consider the total hours spent on all activities.
The present study aimed to assess the relationship between the current functioning level on the SF-36 subscales and the percent reduction in activity levels from pre-illness to current functioning. The percent reduction was calculated using self-report, recall of pre-illness and current hours spent on activities, separate from the SF-36. The percent reduction in functioning was then compared in a correlation to current functioning by means of the SF-36 subscales. The purpose of this study was to demonstrate that the percent reduction was correlated to subscales of the SF-36, suggesting that assessing current functioning is comparable to assessing reductions in functioning. This would then suggest that it is acceptable to compare patients’ current functioning to the functioning of the general population to measure a reduction in activity. Further, these results assess the replicability of Jason et al.'s [12] findings that a measure of current scores in the Role-Physical, Social Functioning, and Vitality subscales in ME and CFS patients could be used to operationalize significant reductions in functioning. If replicable, the findings would support the use of these three subscales to assess a patient's functioning for the various ME and CFS case definitions.
Methods
Research participants
This study used a combined sample drawing from different past studies in multiple geographic locations. Samples were assessed individually and then as a combined sample as results across samples were comparable. Participants with complete data (N = 368) were included. The recruitment process and original study design for each participant group is described below. The combined sample was 83.1% female and 16.9% male. Almost all participants identified as Caucasian (98.6%); 0.5% were Asian or Pacific Islander, and 0.9% indicated ‘Other’ regarding their race. For current work status, 65.3% reported being on disability, 9.8% reported they were working part-time, 3.3% reported they were working full-time, and the remaining 21.6% were not currently employed (student, homemaker, retired, or unemployed). In terms of education, 22.9% of participants had a high school degree, 32.3% a standard college degree, 28.2% graduate or professional degree, and the remaining had not completed a high school degree. The mean age was 47.2 years (SD = 12.8).
Due to the discourse surrounding the name of the illness, patients are described with varying names of diagnoses in each sample. However for this paper we are considering all patients diagnosed with CFS, ME, or ME/CFS to be included within the group of ME and CFS.
Norway 1 sample
Participants from suburbs of Oslo, Norway, were recruited to participate in a CFS self-management trial program. Individuals were recruited through brochures and personal communication at CFS patient organizations, CFS education programs, and healthcare professionals. Patients were required to be at least 18 years of age with a CFS diagnosis made by a physician or medical specialist. Participants were required to be physically able to attend the self-management program. The study gained approval from the Regional Committee for Medical Research Ethics (Health Region North) and the Privacy Ombudsman for Research at Oslo University Hospital. Participants completed an informed consent that provided permission to request a confirmation of their CFS diagnosis.
This sample was 86.2% female and 13.8% male. All participants identified as Caucasian. For current work status, 85.3% were on disability, 8.6% were working part-time, one participant was working full-time, and the remaining 5.2% were not currently employed (student, homemaker, retired, or unemployed). For education, 43.9% of participants had a high school degree, 38.6% a standard college degree, 13.2% graduate or professional degree, and the remaining 4.3% had not completed a high school degree. The mean age was 43.4 years (SD = 11.6).
Norway 2 sample
Participants were recruited from an inpatient medical facility and an outpatient CFS/ME Center. Participants were required to be between 18 and 65 years of age. Individuals with a suspected diagnosis were given a comprehensive medical history interview and medical examination conducted by experienced consultant physicians and a psychologist to rule out exclusionary conditions. An informed consent and other study measures were completed by hard copy. The project gained approval from the Privacy Ombudsman for research at Oslo University Hospital.
This sample was 70.4% female and 29.6% male. Almost all identified as Caucasian (96.3%); one participant identified as Asian or Pacific Islander. For current work status, 77.8 % of participants were on disability, 14.8% were working part-time, and the remaining 7.4% indicated they were a student. Regarding education, 40.7% of participants had a high school degree, 29.6% a standard college degree, 18.5% graduate or professional degree, and the remaining 11.2% had not completed a high school degree. The mean age was 39.0 years (SD = 11.9).
Newcastle sample
Individuals suspected of a CFS diagnosis were referred for a medical assessment at the Newcastle-upon-Tyne Royal Infirmary clinic. Participants were given a comprehensive medical history and medical examination by an experienced physician. An informed consent and other measures were completed by hard copy.
This sample was 82.7% female and 17.3% male. All of the participants identified as Caucasian. For current work status, 39.2% of participants were on disability, 19.6% were working part-time, 9.8% were working full-time, and the remaining 31.4% were not currently employed (student, homemaker, retired, or unemployed). For education, 16.0% of participants had a high school degree, 26.0% attended college for at least one year, 24.0% a standard college degree, 20.0% graduate or professional degree, and the remaining 14.0% had not completed a high school degree. The mean age was 45.0 years (SD = 15.2).
DePaul sample
Individuals were recruited through re-contacting participants from previous studies, visits to support groups, and postings on online forums. Participants were required to be at least 18 years of age and have a self-reported, current diagnosis of CFS, ME, or ME/CFS. Participants could complete the surveys and informed consent online, on a hard-copy, or verbally over the telephone. For all of the different methods of completion, participants were also given the opportunity to complete the surveys at home or in person at DePaul University. The first 100 participants to complete the survey received a $5.00 Amazon.com gift card. This study followed all DePaul University Institutional Review Board protocol.
This sample was 83.1% female and 16.9% male. Almost all participants identified as Caucasian (97.7%); one participant selected Asian or Pacific Islander, and the remaining 1.7% selected ‘Other’ regarding race. For current work status, 57.6% of participants were on disability, 7.0% were working part-time, 3.4% were working full-time, and the remaining 32.0% were not currently employed (student, homemaker, retired, or unemployed). Regarding education, 8.2% of participants had a high school degree or a general education diploma, 18.7% attended college for at least one year, 31.0% a standard college degree, and 42.1% graduate or professional degree. The mean age was 51.8 years (SD = 11.4).
Control group
The control group was obtained as part of an adult epidemiological study that was recruited between September 1995 and May 1997 through phone calls.[15] A stratified random sample of neighborhoods in Chicago was utilized. The first stage included a brief phone screening to assess if the individual was suffering from extreme fatigue, tiredness or exhaustion. Control participants were those who screened negative (were not suffering from prolonged fatigue) and were given a medical examination to determine that they were without exclusionary conditions and were healthy.
This sample was 48.8% female and 51.2% male. In terms of race and ethnicity, 58.1% were Caucasian, 27.9% African American, 7.0% Latino, and the remaining were of another ethnicity. For current work status, 61.7% were working full-time, 8.6% were working part-time, 10.6% were students, 8.5% were retired, and 10.6% were unemployed/homemaker. For education, 10.6% of participants had a high school degree or a general education diploma, 21.3% attended college for at least one year, 38.3% a standard college degree, 21.3% graduate or professional degree, and the remaining 8.5% had not completed a high school degree. The mean age was 40.4 years (SD = 14.6).
Measures
Participants completed the SF-36 [1], a measure of their current functioning. The SF-36 consists of eight subscales: Physical Functioning, Role-Physical, Bodily Pain, General Health Functioning, Vitality, Social Functioning, Role-Emotional, and Mental Health Functioning. Participants answered questions on a 1-3 Likert scale, which was then converted to a 100-point scale. Higher scores indicate higher functioning, while lower scores indicate greater disability. Previous studies have shown adequate internal consistency and discriminant validity among subscales.[9] Buchwald et al. [16] found that the SF-36 had good internal reliability and convergent validity in a sample of individuals with CFS.
The DePaul Symptom Questionnaire (DSQ) [13] includes questions that assess functioning according to hours spent on different activities. The following questions were assessed in the current study: “In the past 4 weeks, how many hours per week have you spent doing: household related activities, social/recreational related activities, family related activities, and work related activities?” and “Before your fatigue/energy related illness, approximately how many hours did you used to spend on: household related activities, social/recreational related activities, family related activities, and work related activities?” Participants provided the number of hours per week they spend on each type of activity and recall the number of hours per week they spent prior to illness onset. The DSQ has shown good test-retest reliability among patients and control groups.[17] The DSQ is available in the library of Research Electronic Data Capture (REDCap),[18] hosted at DePaul University: https://redcap.is.depaul.edu/surveys/?s=tRxytSPVVw.
Statistical analysis
A correlational analysis was used to assess the strength of the relationship between the SF-36 and the current number of hours patients spend on activities as well as the percent reduction in the number of hours patients spend on activities. Percent reduction was calculated by first finding the difference in hours. This was done by subtracting current hours from pre-illness hours. The difference was then divided by the number of pre-illness hours, and that total was multiplied by 100. For example, if the number of pre-illness hours was 45 and the current number of hours was 16, the percent reduction would be [(45-16)/45]*100 = 64.4%. A percent reduction was calculated for each of the following categories: social, house, work, family and total activities. Higher scores in the SF-36 subscales indicate higher current functioning. Therefore a positive correlation with current hours spent on different activities indicates that higher current functioning is correlated with a greater number of hours spent on different types of activities. Alternatively, negative correlations between a SF-36 subscale score and the percent reduction in hours indicates that lower current functioning is correlated with a greater percent decrease in hours.
Multiple regression analyses using the enter method were used to assess the predictive power of the SF-36 subscales on current and percent reduction in functioning. Using the enter method, SPSS assesses the significance of each subscale in the order in which the subscales are entered into the software. Therefore, the order in which subscales are entered into the analysis could potentially impact the results. For this reason, the physically-oriented subscales were separated from the mental, emotional, and psychological subscales. The order of subscales within each analysis was varied across different trials with no differences in significance observed, thus issues of order within each analysis were found to be negligible.
A Receiver Operating Curve (ROC) analysis was used to assess which of the SF-36 subscales best discriminated between patients with ME and CFS and healthy controls. The ROC analysis used a plot of sensitivity versus 1-specificity for scores on the SF-36. The area under the curve (AUC) assessed the degree to which each subscale was able to discriminate between patients and controls. The closer the AUC is to 1, the better discriminatory power that SF-36 subscale has.
Results
Reallocation of hours
Figure 1 presents the differences in the ways patients reduced hours in various activities. This figure demonstrates that it is important to focus on the total hours spent on different activities, or percent reduction in total hours, as there is variation in the ways patients allocate hours spent on different activities.
Figure 1.
Average percent reduction in hours spent on different activities.
Correlational analyses
Table 1 presents the results of a correlational analysis between the SF-36 subscales and different measures of hours spent on activities, specifically current hours spent on activities and the percent reduction in hours spent on different activities. The current sum of hours spent on activities is significantly correlated with Physical Functioning [r = 0.306, N = 368, p = 0.000], Role-Physical [r = 0.196, N = 368, p = 0.000], Bodily Pain [r = 0.161, N = 368, p = 0.002], General Health Functioning [r = 0.181, N = 368, p = 0.000], Vitality [r = 0.184, N = 368, p = 0.000], and Social Functioning [r = 0.403, N = 368, p = 0.000]. The percent reduction in the sum of hours is significantly correlated with Physical Functioning [r = −0.375, N = 368, p = 0.000], Role-Physical [r = −0.280, N = 368, p = 0.000], Bodily Pain [r = −0.216, N = 368, p = 0.000], General Health Functioning [r = −0.188, N = 368, p = 0.000], Vitality [r = −0.228, N = 368, p = 0.000], and Social Functioning [r = −0.489, N = 368, p = 0.000].
Table 1.
Correlation between number of hours spent on different activities and different measures of reduction in activity (N=368).
| Physical Functioning | Role-Physical | Bodily Pain | General Health | Vitality | Social Functioning | Role-Emotional | Mental Health | |
|---|---|---|---|---|---|---|---|---|
| Sum of Current Hours | .31** | .20** | .16** | .18** | .18** | .40** | −.02 | .05 |
| Social | .12* | .00 | .08 | .08 | .12* | .18** | −.03 | .03 |
| Work | .30** | .30** | .18** | .13** | .15** | .35** | .01 | −.02 |
| Family | .05 | .03 | −.03 | .08 | .04 | .16** | .03 | .09 |
| House | .16** | −.01 | .09 | .12* | .08 | .17** | −.07 | .05 |
| Percent Reduction in Sum of Hours | −.38** | −.28** | −.22** | −.19** | −.23** | −.49** | .01 | −.03 |
| Social | −.09* | −.01 | −.14** | −.04 | −.01 | −.16** | .00 | −.05 |
| Work | −.27** | −.26** | −.15** | −.13* | −.20** | −.34** | −.03 | .02 |
| Family | .00 | .00 | −.06 | −.01 | −.01 | −.08 | −.02 | −.06 |
| House | −.22** | −.08 | −.19** | −.06 | −.08 | −.27** | .02 | −.03 |
p < 0.05.
p < 0.01.
Multiple regression analyses
Tables 2 and 3 present the results of multiple regression analyses to assess which of the SF-36 subscales correlate with the sum of current hours spent on activities. Table 2 presents results for the physically derived SF-36 subscales. In assessing physical subscales, a significant regression equation was found: F(4,363) = 11.47, p < 0.001, adj. R2 = 0.10. Table 3 presents results for the mental, emotional, and psychological SF-36 subscales. In assessing psychological subscales, a significant regression equation was found: F(4,363) = 18.04, p < 0.001, adj. R2 = 0.16. Physical Functioning, Role-Physical, and Social Functioning all significantly correlated (p ≤ 0.01) with the current sum of hours.
Table 2.
Multiple regression results for current hours spent on activities by SF-36 physical health subscales.
| df | MS | F | |
|---|---|---|---|
| Regression | 4 | 2910.5 | 11.5** |
| Residual | 363 | 253.7 | |
| Total | 367 | ||
| Model | Beta | ||
| Physical Functioning | .18** | ||
| Role-Physical | .11 | ||
| Bodily Pain | .03 | ||
| General Health | .06 | ||
F(4,363) = 11.47, p < 0.005, adj. R2 = 0.10.
p < 0.01.
Table 3.
Multiple regression results for current hours spent on activities by SF-36 mental, emotional, and social health subscales.
| df | MS | F | |
|---|---|---|---|
| Regression | 4 | 4299.9 | 18.0** |
| Residual | 363 | 238.4 | |
| Total | 367 | ||
| Model | Beta | ||
| Social Functioning | .29** | ||
| Vitality | .04 | ||
| Role-Emotional | −.02 | ||
| Mental Health Functioning | .02 | ||
F(4,363) = 18.04, p < 0.005, adj. R2 = 0.16.
p < 0.01.
Tables 4 and 5 present the results of a multiple regression analyses to assess which of the SF-36 subscales are significantly correlated with the percent reduction in the sum of hours spent on activities. The physically-related subscales were assessed separately from the mental, emotional, and psychological subscales and are presented in tables 4 and 5 respectively. In assessing physical subscales, a significant regression equation was found: F(4,363) = 20.02, p < 0.001, adj. R2 = 0.17. In assessing psychological subscales, a significant regression equation was found: F(4,363) = 29.29, p < 0.001, adj. R2 = 0.24. The Physical Functioning, Role-Physical, and Social Functioning were found to significantly (p ≤ 0.01) correlate with the percent reduction in the sum of hours spent on activities.
Table 4.
Multiple regression results for percent reduction in hours spent on activities by SF-36 physical health subscales.
| df | MS | F | |
|---|---|---|---|
| Regression | 4 | 7967.1 | 20.0** |
| Residual | 363 | 397.9 | |
| Total | 367 | ||
| Model | Beta | ||
| Physical Functioning | −.29** | ||
| Role-Physical | −.24** | ||
| Bodily Pain | −.06 | ||
| General Health | −.04 | ||
F(4,363) = 20.02, p < 0.005, adj. R2 = 0.17.
p < 0.01.
Table 5.
Multiple regression results for percent reduction in hours spent on activities by SF-36 mental, emotional, and social health subscales.
| df | MS | F | |
|---|---|---|---|
| Regression | 4 | 10754.4 | 29.3** |
| Residual | 363 | 367.2 | |
| Total | 367 | ||
| Model | Beta | ||
| Social Functioning | −.46** | ||
| Vitality | −.08 | ||
| Role-Emotional | .01 | ||
| Mental Health Functioning | .04 | ||
F(4,363) = 29.29, p < 0.005, adj. R2 = 0.24.
p < 0.01.
ROC analysis
Table 6 presents the results of the ROC analysis comparing patients to controls according to the SF-36 subscales. Five subscales had an AUC value at or above 0.90 including General Health Functioning (AUC = 0.93), Social Functioning (AUC = 0.93), Vitality (AUC = 0.91), Physical Functioning (AUC = 0.91), and Role-Physical (AUC = 0.90).
Table 6.
ROC analysis of combined sample (N=368) versus epi-study controls (N=47) according to SF-36 subscale.
| AUC | Standard Error | Confidence Interval | |
|---|---|---|---|
| General Health Functioning | 0.93 | 0.02 | 0.89 - 0.98 |
| Social Functioning | 0.93 | 0.02 | 0.88 - 0.97 |
| Vitality | 0.91 | 0.03 | 0.86 - 0.97 |
| Physical Functioning | 0.91 | 0.03 | 0.85 - 0.97 |
| Role-Physical | 0.90 | 0.03 | 0.84 - 0.96 |
| Bodily Pain | 0.86 | 0.04 | 0.79 - 0.93 |
| Mental Health Functioning | 0.59 | 0.05 | 0.50 - 0.68 |
| Role-Emotional | 0.47 | 0.05 | 0.38 - 0.56 |
Discussion
It is has been a challenge to determine what constitutes a significant reduction in functioning because it requires a determination of what is considered average, or “normal,” functioning and an assessment of how far below that level one's activity level needs to decrease to be considered significantly reduced. Such an assessment is especially difficult as no two individuals will have the same pre-illness activity level of functioning. Several of the ME and CFS case definitions require a reduction in functioning rather than simply documenting low current functioning. The present study sought to determine if assessing current functioning using a validated measure of disability, the SF-36, would produce the same results as assessing the percent reduction in activity levels from an individual's pre-illness level functioning.
The SF-36 is a measure of current functioning, therefore, significant correlations with the sum of current hours spent on activities was expected. The current study compared patients to their own recalled pre-illness functioning rather than assessing current patient functioning as significantly different from the current functioning of a control group. This is important as the current operationalization of significant reductions in functioning is the SF-36, specifically the Role-Physical, Social Functioning, and Vitality subscales. The percent reduction in the sum of hours spent on activities was significantly correlated with the Role-Physical, Social Functioning, and Vitality subscales. These results suggest that assessing current functioning is comparable to assessing percent reduction in functioning.
Currently, Role-Physical, Social Functioning, and Vitality are the SF-36 subscales used to measure significant reductions in functioning in several studies.[8,19,20,21,22] In the present study, it was found that Role-Physical and Social Functioning were significantly correlated to both the current hours and percent reduction in hours spent on different activities. These findings demonstrate that as current hours decrease and percent reduction in hours increases, the scores on the Role-Physical and Social Functioning scales will decrease, indicating worse functioning. As these subscales were not only significantly predictive of current hours of activity but also the percent reduction in hours, these results further support the use of these two subscales in assessing reductions in functioning.
As Vitality was not found to be significant in predicting current or percent reduction in hours, an ROC analysis was run to compare results to Jason et al.'s [12] ROC analysis findings that originated the use of Role-Physical, Social Functioning, and Vitality to measure a reduction in functioning. The current patient group was compared to controls.[15] Similar to previous findings, Vitality was in the top three subscales of the SF-36 in differentiating patients from healthy controls. This finding indicates that the vitality subscale would adequately identify patients since their scores are significantly lower than those of healthy controls, indicating worse functioning. Social Functioning and Role-Physical, also, both had an AUC value at or above 0.90 which, again, further supports the use of these subscales to measure reductions in functioning.
This study has some limitations. First, participants were asked to recall hours spent on different activities prior to illness onset, so the findings are subject to recall-bias. One of the core symptoms of ME and CFS is neurocognitive deficits, which could result in patients having difficulty remembering their pre-illness level functioning.[23] However, Jason and Evans [23] found that recalling post exertional malaise symptoms, such as functioning level, was significantly reliable. While the findings are subject to recall bias, the sample used for this study was recruited post-illness onset, therefore pre-illness activity data was not available and recalled activity data was utilized.
Additionally, the patient sample was composed of a combination of different recruitment methods and illness severity levels from multiple geographic locations. As the sample drew from different locations, diagnoses were self-reported or made by different doctors according to different case definitions; there was not a singular case definition participants were required to meet. There were also several demographic differences between British and U.S. samples.[24] However, the heterogeneity of the patient sample may allow us to generalize our findings to a wide range of countries and settings. Further, when samples were analyzed individually, results were comparable to the combined sample.
There are some further demographic differences that pose a limitation to this study. The control group was more ethnically diverse than the patient sample. There was also a greater proportion of women to men for the patient sample compared to the control sample. Previous studies [15,25,26,27] have found that ME and CFS are more prevalent in women than men, thus a greater proportion of women in the patient sample is to be expected. Future studies should utilize a cross-cultural sample so as to assess cultural differences that may impact symptomatology.
It is important to define and operationalize what it means to have a significant reduction in activity level as it is included in almost every case definition for ME and CFS. However, because of varying ways to define significant reduction, it is evident that this criteria is subjective and therefore challenging to define. Further, it is difficult to specify a generalized definition of what it means to have significantly reduced activity given the variability in individuals’ experiences of this illness.
It could be stigmatizing for patients to have to prove an adequate reduction in activity level in order to have their illness legitimized by professionals. Defining what it means to have a substantial reduction in activity level could draw more attention to the debilitating nature of this illness. This definition could also accurately differentiate normal fatigue from those with severe functioning limitations due to ME and CFS. However, requiring this criterion for a diagnosis with ME and CFS implies that patients must prove their impairment to medical professionals. Further, patients may not meet the specifications for a substantial reduction in activity, yet may still be significantly impaired. Without meeting the substantial reduction criterion, patients lack a diagnosis and are thereby denied proper treatment. Future studies should assess more sensitive or qualitative assessments for a substantial reduction in activity level.
Acknowledgments
Funding
Funding was provided by National Institutes for Allergy and Infectious Diseases [Grant No. AI105781] and the Eunice Kennedy Shriver National Institute of Child Health and Human Development [Grant No. HD072208].
Contributor Information
Taylor Thorpe, Center for Community Research, DePaul University, Chicago, IL USA
Stephanie McManimen, Center for Community Research, DePaul University, Chicago, IL USA
Kristen Gleason, Center for Community Research, DePaul University, Chicago, IL USA
Jamie Stoothoff, Center for Community Research, DePaul University, Chicago, IL USA
Julia L. Newton, Institute for Ageing and Health, Newcastle University, Newcastle upon Tyne, UK
Elin Bolle Strand, Oslo University Hospital, Oslo, Norway.
Leonard A. Jason, Center for Community Research, DePaul University, Chicago, IL USA.
References
- 1.Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36): conceptual framework and item selection. Med Care. 1992;30:473–483. [PubMed] [Google Scholar]
- 2.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. doi: 10.7326/0003-4819-121-12-199412150-00009. [DOI] [PubMed] [Google Scholar]
- 3.Brurberg KG, Fønhus MS, Larun L, et al. Case definitions for chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME): a systematic review. BMJ Open. 2014;4(2):e003973. doi: 10.1136/bmjopen-2013-003973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hjermstad MJ, Fayers PM, Bjordal K, et al. Health-related quality of life in the general Norwegian population assessed by European Organization for Research and Treatment of Cancer Core Quality-of-Life Questionnaire: The QLQ-C30 (+3). J Clin Oncol. 1995;16:1188–1196. doi: 10.1200/JCO.1998.16.3.1188. [DOI] [PubMed] [Google Scholar]
- 5.Carruthers BM, Jain AK, De Meirleir KL, et al. Myalgic encephalomyelitis/chronic fatigue syndrome: clinical working case definition, diagnostic and treatment protocols (Canadian case definition). J Chronic Fatigue Syndr. 2003;11(1):7–115. [Google Scholar]
- 6.Carruthers BM, van de Sande MI, De Meirleir KL, et al. Myalgic encephalomyelitis: international consensus criteria. J Intern Med. 2011;270(4):327–338. doi: 10.1111/j.1365-2796.2011.02428.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Carruthers BM, van de Sande MI, De Meirleir KL, et al. Myalgic encephalomyelitis – adult & paediatric: international consensus primer for medical practitioners. The National Library of Canada Cataloguing-in-Publication Data; Calgary: 2012. [Google Scholar]
- 8.IOM (Institute of Medicine) Beyond myalgic encephalomyelitis/chronic fatigue syndrome: redefining an illness. The National Academies Press; Washington, DC: 2015. [PubMed] [Google Scholar]
- 9.McHorney CA, Ware JE, Lu AW, et al. The MOS 36-item short-form health survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care. 1994;32:40–66. doi: 10.1097/00005650-199401000-00004. [DOI] [PubMed] [Google Scholar]
- 10.Reeves WC, Wagner D, Risenbaum R, et al. Chronic fatigue syndrome – a clinically empirical approach to its definition and study. BMC Med. 2005:3–19. doi: 10.1186/1741-7015-3-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jason LA, Najar N, Porter N, et al. Evaluating centres for disease control's empirical chronic fatigue syndrome case definition. J Disabil Policy Stud. 2008;20:93–100. [Google Scholar]
- 12.Jason LA, Brown M, Evans M, et al. Measuring substantial reductions in functioning in patients with chronic fatigue syndrome. Disabil Rehabil. 2011;33(7):589–598. doi: 10.3109/09638288.2010.503256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jason LA, Evans M, Porter N, et al. The development of a revised Canadian myalgic encephalomyelitis chronic fatigue syndrome case definition. Am J Biochem Biotechnol. 2010;6(2):120–135. [Google Scholar]
- 14.Schafer C, Evans M, Jason LA, et al. Measuring substantial reductions in activity. J Prev Interv Community. 2015;43(1):5–19. doi: 10.1080/10852352.2014.973242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jason LA, Richman JA, Rademaker AW, et al. A community-based study of chronic fatigue syndrome. Arch Intern Med. 1999;159(18):2129–2137. doi: 10.1001/archinte.159.18.2129. [DOI] [PubMed] [Google Scholar]
- 16.Buchwald D, Pearlman T, Umali J, et al. Functional status in patients with chronic fatigue syndrome, other fatiguing illnesses, and healthy individuals. Am J Med. 1996;101(4):364–370. doi: 10.1016/S0002-9343(96)00234-3. [DOI] [PubMed] [Google Scholar]
- 17.Jason LA, So S, Brown AA, et al. Test-retest reliability of the DePaul Symptom Questionnaire. Fatigue: Biomed Health Behav. 2015;3:16–32. doi: 10.1080/21641846.2014.978110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jason LA, McManimen S, Sunnquist M, et al. Case definitions integrating empiric and consensus perspectives. Fatigue: Biomed Health Behav. 2016;4(1):1–23. doi: 10.1080/21641846.2015.1124520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Strand EB, Lillestøl K, Jason LA, et al. Comparing the DePaul Symptom Questionnaire with physician assessments: a preliminary study. Fatigue: Biomed Health Behav. 2016;4(1):56–62. [Google Scholar]
- 21.IOM (Institute of Medicine) Redefining an illness: report guide for clinicians. The National Academies Press; Washington, DC: 2015. [Google Scholar]
- 22.Jason LA, Skendrovic B, Furst J, et al. Data mining: comparing empiric CFS to the Canadian ME/CFS case definition. J Clin Psychol. 2011;68(1):41–49. doi: 10.1002/jclp.20827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Evans M, Jason LA. Effects of time frame on the recall reliability of CFS symptoms. Eval Health Prof. 2013;38(3):367–381. doi: 10.1177/0163278713497014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zdunek M, Jason LA, Evans M, et al. A cross cultural comparison of disability and symptomatology associated with CFS. Int J Psychol Behav Sci. 2015;5(2):98–107. doi: 10.5923/j.ijpbs.20150502.07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wessely S, Chalder T, Hirsch S, et al. The prevalence of fatigue and chronic fatigue syndrome: a prospective primary care study. 1997;87:1449–1455. doi: 10.2105/ajph.87.9.1449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Buchwald D, Pearlman T, Kith P, et al. Gender differences in patients with chronic fatigue syndrome. J Gen Intern Med. 1994;9:397–401. doi: 10.1007/BF02629522. [DOI] [PubMed] [Google Scholar]
- 27.Reyes M, Nisenbaum R, Hoaglin DC, et al. Prevalence and incidence of chronic fatigue syndrome in Wichita, Kansas. Arch Intern Med. 2003;163:1530–1536. doi: 10.1001/archinte.163.13.1530. [DOI] [PubMed] [Google Scholar]