Data on the level of maximal scoring (on the Chalder Fatigue Scale) would be useful
I thought I'd point out that two of the authors (including the
corresponding author who was the Principal Investigator) should be well
aware of the problems of the ceiling effect of the Chalder Fatigue scale
for Chronic Fatigue Syndrome (CFS) patients as they devoted a whole paper
to analysing the use of the scale in CFS patients[1].
This study looked at the scores of 136 CFS patients, who were
attending an outpatient clinic, using the 14-item
Chalder Fatigue Scale with Likert scoring (i.e. 0, 1, 2, 3).
It found that "scores on ... six items "tiredness," "resting
more,""lacking energy,"
"feeling weak," "feeling sleepy or drowsy," and "starts things without
difficulty but gets weaker as goes on") were highly skewed with the
majority of patients reaching the maximum score." One item ("Starts
things without difficulty but gets weaker as goes on") is excluded from
the 11-item version of the Chalder Fatigue Scale but the
other 5 items remain.
They concluded: "Near-maximal scoring on six physical fatigue scale
items from the total of 14 items constituting the Chalder fatigue scale
supports the validity of scoring the physical fatigue scale on a two-point
scale (presence or absence) rather than the four-point Likert scoring.
Most of the variance in the total
fatigue scale is thus accounted for by the mental fatigue score."
To me, these results would have suggested using a different scale as,
in both scenarios (i.e. whether bimodal or Likert scoring is used),
patients can't be recorded as getting worse if they are already scoring
maximum at the start of the trial - an important issue that should be
investigated in rehabilitation trials in the CFS area given the high level
of adverse reactions that have been reported by patients from
rehabilitative interventions[2].
Another reason to use a different scale for the FINE trial in
particular is the fact that the trial, which was performed in patients'
homes, was designed to include severely affected patients who would have
difficulty attending an outpatient clinic[3] - the previous analysis found
high rates of maximal scoring in a group well enough to attend out-
patients so maximal scoring was certainly going to be an issue in a more
severely affected group.
Anyway, the authors concluded the bimodal scale was sufficient and
they then designed this trial using the bimodal scoring method[3]. I'm
sure many pharmacological and non-pharmacological studies could look
different if investigators decided to use a different scoring method or
scale at the end, if the results weren't as impressive as they'd hoped.
But that is not normally how medicine works. So, while it is interesting
that the researchers have shared this data, I think the data in the main
paper should be seen as the main data.
I have seen many studies where authors have given the percentage who
scored the maximum scores as the problem of floor and ceiling effects is a
well-recognised issue. In the spirit of full disclosure, it would be good
if the authors of this trial would now share that data for the bimodal
scoring (the scoring method used in the published paper) and also for the
Likert scoring system (0-3), now that they have presented other data using
that scoring method. Dr. Bart Stouten has already pointed out the problems
that exist with the bimodal scoring system for the group in this study but
we do not have the exact figures[4]:
"Reviewing the mean and standard deviation data in this study, I
calculated that between 62 (65%) and 78 (82%) patients in the Pragmatic
Rehabilitation (PR) group must have recorded the maximum score of 11 at
baseline (see
separate response). This indicates that a substantial proportion of
participants would not have been able to record an exacerbation of fatigue
following PR. Similarly, between 21 (26%) and 57 (70%) patients must have
recorded the maximum score at 70 weeks."
Given that the initial scores in the Pragmatic Rehabilitation (PR)
group on the Likert scoring scale were 29.39 (SD: 3.46), one can clearly
see that there was a large amount of maximal scoring on many or all of the
11-items in the Likert scale also.
References:
[1] Morriss RK, Wearden AJ, Mullis R. Exploring the validity of the
Chalder Fatigue scale in chronic fatigue syndrome. J Psychosom Res. 1998
Nov;45(5):411-7.
[3] Wearden AJ, Riste L, Dowrick C, Chew-Graham C, Bentall RP,
Morriss RK, Peters S, Dunn G, Richardson G, Lovell K, Powell P. Fatigue
Intervention by Nurses Evaluation--the FINE Trial. A randomised controlled
trial of nurse
led self-help treatment for patients in primary care with chronic fatigue
syndrome: study protocol. [ISRCTN74156610]. BMC Med. 2006 Apr 7;4:9. http://www.biomedcentral.com/1741-7015/4/9
Rapid Response:
Data on the level of maximal scoring (on the Chalder Fatigue Scale) would be useful
I thought I'd point out that two of the authors (including the
corresponding author who was the Principal Investigator) should be well
aware of the problems of the ceiling effect of the Chalder Fatigue scale
for Chronic Fatigue Syndrome (CFS) patients as they devoted a whole paper
to analysing the use of the scale in CFS patients[1].
This study looked at the scores of 136 CFS patients, who were
attending an outpatient clinic, using the 14-item
Chalder Fatigue Scale with Likert scoring (i.e. 0, 1, 2, 3).
It found that "scores on ... six items "tiredness," "resting
more,""lacking energy,"
"feeling weak," "feeling sleepy or drowsy," and "starts things without
difficulty but gets weaker as goes on") were highly skewed with the
majority of patients reaching the maximum score." One item ("Starts
things without difficulty but gets weaker as goes on") is excluded from
the 11-item version of the Chalder Fatigue Scale but the
other 5 items remain.
They concluded: "Near-maximal scoring on six physical fatigue scale
items from the total of 14 items constituting the Chalder fatigue scale
supports the validity of scoring the physical fatigue scale on a two-point
scale (presence or absence) rather than the four-point Likert scoring.
Most of the variance in the total
fatigue scale is thus accounted for by the mental fatigue score."
To me, these results would have suggested using a different scale as,
in both scenarios (i.e. whether bimodal or Likert scoring is used),
patients can't be recorded as getting worse if they are already scoring
maximum at the start of the trial - an important issue that should be
investigated in rehabilitation trials in the CFS area given the high level
of adverse reactions that have been reported by patients from
rehabilitative interventions[2].
Another reason to use a different scale for the FINE trial in
particular is the fact that the trial, which was performed in patients'
homes, was designed to include severely affected patients who would have
difficulty attending an outpatient clinic[3] - the previous analysis found
high rates of maximal scoring in a group well enough to attend out-
patients so maximal scoring was certainly going to be an issue in a more
severely affected group.
Anyway, the authors concluded the bimodal scale was sufficient and
they then designed this trial using the bimodal scoring method[3]. I'm
sure many pharmacological and non-pharmacological studies could look
different if investigators decided to use a different scoring method or
scale at the end, if the results weren't as impressive as they'd hoped.
But that is not normally how medicine works. So, while it is interesting
that the researchers have shared this data, I think the data in the main
paper should be seen as the main data.
I have seen many studies where authors have given the percentage who
scored the maximum scores as the problem of floor and ceiling effects is a
well-recognised issue. In the spirit of full disclosure, it would be good
if the authors of this trial would now share that data for the bimodal
scoring (the scoring method used in the published paper) and also for the
Likert scoring system (0-3), now that they have presented other data using
that scoring method. Dr. Bart Stouten has already pointed out the problems
that exist with the bimodal scoring system for the group in this study but
we do not have the exact figures[4]:
"Reviewing the mean and standard deviation data in this study, I
calculated that between 62 (65%) and 78 (82%) patients in the Pragmatic
Rehabilitation (PR) group must have recorded the maximum score of 11 at
baseline (see
separate response). This indicates that a substantial proportion of
participants would not have been able to record an exacerbation of fatigue
following PR. Similarly, between 21 (26%) and 57 (70%) patients must have
recorded the maximum score at 70 weeks."
Given that the initial scores in the Pragmatic Rehabilitation (PR)
group on the Likert scoring scale were 29.39 (SD: 3.46), one can clearly
see that there was a large amount of maximal scoring on many or all of the
11-items in the Likert scale also.
References:
[1] Morriss RK, Wearden AJ, Mullis R. Exploring the validity of the
Chalder Fatigue scale in chronic fatigue syndrome. J Psychosom Res. 1998
Nov;45(5):411-7.
[2] Kindlon T, Goudsmit EM. Graded exercise for chronic fatigue
syndrome: too soon to dismiss reports of adverse reactions. J Rehabil Med.
2010 Feb;42(2):184; author reply 184-6.
http://jrm.medicaljournals.se/article/full/10.2340/16501977-0493
[3] Wearden AJ, Riste L, Dowrick C, Chew-Graham C, Bentall RP,
Morriss RK, Peters S, Dunn G, Richardson G, Lovell K, Powell P. Fatigue
Intervention by Nurses Evaluation--the FINE Trial. A randomised controlled
trial of nurse
led self-help treatment for patients in primary care with chronic fatigue
syndrome: study protocol. [ISRCTN74156610]. BMC Med. 2006 Apr 7;4:9.
http://www.biomedcentral.com/1741-7015/4/9
[4] Stouten B. Fatigue scale.
http://www.bmj.com/cgi/eletters/340/apr22_3/c1777#235631
Competing interests:
None declared
Competing interests: No competing interests