Reliability of physiological, psychological, and cognitive variables in chronic fatigue syndrome

[Dolphin]

Res Sports Med. 2005 Jul-Sep;13(3):231-41.

Reliability of physiological, psychological, and cognitive variables in chronic fatigue syndrome.

Wallman KE, Morton AR, Goodman C, Grove R.

Source
School of Human Movement and Exercise Science, The University of Western Australia, Crawely, Western Australia, Australia.
kwallman@cyllene.uwa.edu.au

Abstract*

The purpose of this study was to assess the reliability of specific physiological, psychological, and cognitive variables in 31 chronic fatigue syndrome (CFS) subjects and 31 matched control subjects.

All variables were assessed weekly over a 4-week period and reliability was determined using an intraclass correlation coefficient (ICC).

Results ranged from moderately to highly reliable for all variables assessed, except for mental and physical fatigue, which were of questionable reliability in both groups (ICC = 0.61 and 0.65, respectively, for the CFS group; 0.62 and 0.52 for the control group).

A Pearson product-moment correlation analysis that compared exercise performance with all psychological variables assessed, demonstrated a significant relationship between exercise performance and depression (r = .41, P = .02) in week 3 only, suggesting minimal association between objective performance and psychological responses.

These correlation results support a central, as opposed to a peripheral, basis to the sensation of fatigue in CFS.

PMID: 16392538 [PubMed - indexed for MEDLINE]

*I gave each sentence its own paragraph

 

[Dolphin]

This study asked people to do exercise tests every seven days, so that four were completed.

They tried to control for activity the day before the test with:

Subjects were required to follow the same daily routine 24 hours prior to each test.

They compared the results of the exercise test (W kg-1 i.e. Watts per kilogram) to four variables (Physical Fatigue, Mental Fatigue, Depression and Anxiety) at four time time-points.

There was only one statistically significant correlation out of the 16 and they say this could even be due to chance as there was 16 comparisons.

It is somewhat interesting that Physical Fatigue (Chalder Fatigue Questionnaire) and Mental Fatigue (Chalder Fatigue Questionnaire) didn't correlate with any of the exercise test results.
[Aside: although all r-values (correlations) for these two were in the expected direction i.e. negative i.e. lower fatigue scores were associated with better exercise results].

 

[Dolphin]

They reported an published finding from an earlier study:

A further variable assessed, but not reported in this adjunct study, were scores for accuracy recorded on a cognitive test (Stroop colour word test), which were significantly lower in the CFS group (P = 0.002).

Wallman KE, Morton AR, Goodman C, Grove R (2004) Physiological responses during a submaximal cycle test in chronic fatigue syndrome. Medicine & Science in Sport & Exercise 36: 1682-1688.

 

[Dolphin]

Both mental and physical fatigue showed poor reliability in both the CFS and control groups while the other variables were much more reliable.

The authors discuss why this might be important in the introduction:

To date, few studies have accounted for symptom variation when assessing the efficacy of interventions in CFS or when comparing variables between CFS and control subjects. Yet, failure to account for possible variation in physiological, psychological, and cognitive variables when conducting trials in CFS can lead to the reporting of results that do not accurately reflect the nature of CFS. This conjecture is supported by various researchers who note that the reporting of single-session measures may yield inconclusive results as levels of fatigue and effort can vary over extended periods of time (Fuentes et al. 2001; Gantz and Holmes 1990; Kane, Gantz, and Di Pino 1997). In addition, Jason et al. (1999) note that single-session assessment procedures often can obscure the fluctuating nature of CFS, and that this could prevent health care professionals from understanding the true complexities of the disorder. Consequently, Jason et al. (1999) propose the use of multiple-measures made over the course of different time points in order to provide more accurate estimates of the true ability of CFS sufferers.

Addressing the fluctuating nature of symptoms in CFS becomes even more important if a study does not have the subject numbers designated by an appropriate power analysis, as is the case for many published physiological studies. Take, for example, an insufficiently powered intervention trial that performs baseline testing when the majority of subjects are feeling extremely unwell. If the majority of these subjects feel comparatively better on the day that postintervention assessment is made, then one cannot be certain whether the final results refiect the efficacy of the intervention or the fiuctuating nature of the variables assessed.