Please help evaluate the Chalder Fatigue Scale

[Jenny]

Do you have problems with tiredness? 2 3 2
Do you need to rest more? 3 3 2
Do you feel sleepy or drowsy? 2 3 1
Do you lack energy?2 3 2
Do you have trouble starting things?2 3 1
Do your muscles have less strength?2 3 1
Do you feel weak?2 3 2
Do you have difficulty concentrating1 2 1
Do you find it more difficult to find the correct word?1 2 1
Do you make slips of the tongue when speaking?1 1 1
How is your memory?1 11

Hi Graham

My scores are above. I'm 'moderate' - been ill 29 years with some periods of near remission.

I've only just seen this - I wonder if you'd get more responses if you gave the thread a different title. The last bit of titles are often obscured - so I didn't realise you wanted us to complete it.

Having said that I'm not sure what to suggest (!), but perhaps something like 'Your scores on Fatigue scale needed'?

Thanks very much for doing this - sounds good!

Jenny

 

[Graham]

Thanks Jenny - you are number 38 on the list. I'll see if I can work out how to change the title - a very good suggestion - thanks!

 

[Enid]

Chalder is psychiatric nursing - do her findings relate to medicine.

 

[Sparrow]

If information about how long I've been ill, etc. would help, I would be happy to provide that too. Whatever helps.

I crashed in 1997 and several times through my University career as well. I just didn't know it was ME then, and the crashes weren't as severe as my more recent problems.

 

[Graham]

Thanks Sparrow. I'm wondering about whether how long people have been ill could be relevant to what I am doing. If I do think I'll need it, I have a list of all the contributors, and I'll go back to them and explain why. That also means I will have to separate sudden onset (certain date) with gradual onset (uncertain date).

I'm still sorting out the ideas that the results are already giving me - it's a slow process getting the brain cells to respond!

 

[Marco]

Thanks Sparrow. I'm wondering about whether how long people have been ill could be relevant to what I am doing. If I do think I'll need it, I have a list of all the contributors, and I'll go back to them and explain why. That also means I will have to separate sudden onset (certain date) with gradual onset (uncertain date).

I'm still sorting out the ideas that the results are already giving me - it's a slow process getting the brain cells to respond!

Hi Graham.

I really appreciate what you are trying to do with this.

Just a quick question.

I assume that there is particular relevance in the good and bad spells needing to last a week or more (to tie in with how the Chalder scale was scored in PACE perhaps?).

Reason being I'm one of the longer term patients (25 years plus) and I'd say my pattern of illness has more or less settled down to a constant state with fluctuations for better or worse lasting no more than a few days (bad) or a few hours (good).

I'm not sure my scores would be very helpful if only relatively longer term fluctuations are relevant?

 

[Graham]

Hi Marco

That's fine - a number of people have sent me scores that have been pretty steady - some haven't changed at all. I focused on a week or more because I didn't want to include when someone is pretty stable but then has an couple of bad or good days for some odd reason. It is entirely arbitrary on my part.

I want to see if there are hidden ceilings in the Chalder scale (so far, it doesn't look like it, but that will take a more careful analysis), and to get some sort of measure of how much the average ME patient varies in the course of a year. Both of those have not been taken into consideration, and may be significant, particularly with such small average changes in the PACE trial.

39 responses now.

Thanks

 

[Jenny]

Hi Graham

Will you be looking at the reliability (internal consistency) and validity (eg correlation with self-assessed severity) of the scale?

Jenny

 

[Graham]

Hi Jenny

Already there seems to be no consistency between mild, moderate and severe and the scores (although I only have 1 mild score, and Goudsmit gives me another 3), so I doubt if any calculation will be needed - the graphic shows it clearly.

How could I check for reliability?

 

[Marco]

Hi Marco

That's fine - a number of people have sent me scores that have been pretty steady - some haven't changed at all. I focused on a week or more because I didn't want to include when someone is pretty stable but then has an couple of bad or good days for some odd reason. It is entirely arbitrary on my part.

I want to see if there are hidden ceilings in the Chalder scale (so far, it doesn't look like it, but that will take a more careful analysis), and to get some sort of measure of how much the average ME patient varies in the course of a year. Both of those have not been taken into consideration, and may be significant, particularly with such small average changes in the PACE trial.

39 responses now.

Thanks

Thanks Graham.

I'll have some figures for you asap.

As I do have bad spells lasting a number of days and shorter 'better' spells I believe I could legitimately describe those as a bad week or good week respectively compared to the average.

 

[Jenny]

Hi Jenny

Already there seems to be no consistency between mild, moderate and severe and the scores (although I only have 1 mild score, and Goudsmit gives me another 3), so I doubt if any calculation will be needed - the graphic shows it clearly.

How could I check for reliability?

Hi Graham - see below:

SPSS FAQ
What does Cronbach's alpha mean?

Cronbach's alpha is a measure of internal consistency, that is, how closely related a set of items are as a group. A "high" value of alpha is often used (along with substantive arguments and possibly other statistical measures) as evidence that the items measure an underlying (or latent) construct. However, a high alpha does not imply that the measure is unidimensional. If, in addition to measuring internal consistency, you wish to provide evidence that the scale in question is unidimensional, additional analyses can be performed. Exploratory factor analysis is one method of checking dimensionality. Technically speaking, Cronbach's alpha is not a statistical test - it is a coefficient of reliability (or consistency).
Cronbach's alpha can be written as a function of the number of test items and the average inter-correlation among the items. Below, for conceptual purposes, we show the formula for the standardized Cronbach's alpha:

GRAHAM - I'VE TRIED TO PASTE THE FORMULA HERE BUT CAN'T FOR SOME REASON - IT'S GIVEN HERE:

http://www.ats.ucla.edu/stat/spss/faq/alpha.html


One can see from this formula that if you increase the number of items, you increase Cronbach's alpha. Additionally, if the average inter-item correlation is low, alpha will be low. As the average inter-item correlation increases, Cronbach's alpha increases as well (holding the number of items constant).

 

[Graham]

Thanks for that Jenny. I haven't come across that before.

Sadly, I am a mathematician, so I only believe in using things when I understand how they work and what the underlying concepts and assumptions are (such as standard deviation really should only be used for Normal-like distributions). It looks like I have got a challenge on my hands here, as the explanation only shows you how to calculate it, and I can't work out what underlies it all. More research and brain matter needed!

This will probably go on the back burner for a while though, as the amount of work is piling up. A pleasure to be enjoyed later!

 

[Marco]

Scores are for Now; Good and Bad compared to pre illness :

Do you have problems with tiredness? 2 2 3
Do you need to rest more? 2 2 3
Do you feel sleepy or drowsy? 2 2 3
Do you lack energy? 2 2 3
Do you have trouble starting things? 2 2 3
Do your muscles have less strength? 2 2 3
Do you feel weak? 2 2 3
Do you have difficulty concentrating 3 3 3
Do you find it more difficult to find the correct word? 2 2 3
Do you make slips of the tongue when speaking? 2 2 3
How is your memory? 2 2 3

A pretty boring set of figures!

Just a couple of observations and I may not be representative.

'Fatigue' is the wrong word. Its a lack of energy; feeling inflamed and then being incapable of doing things whether physical or mental.

There are definitely ceiling effects. If I was to be completely honest, compared to pre illness, I would have to score each item as 'much more than normal' but this didn't allow any latitude to record the exacerbation of symptoms during a crash.

During a crash I am basically incapable of doing anything apart from crawling into bed and riding it out.

As stated previously I have very brief (a few hours) spells where I feel completely normal i.e. I would score 1 on every item. As this doesn't last and is often followed by PEM then averaged over a week I would have to score this as the same as my normal.

Please feel free to disregard these scores if you don't feel they are representative.

I would class my present condition as moderate - unable to work. During a crash moderate/severe - bedbound.

 

[Jenny]

Thanks for that Jenny. I haven't come across that before.

Sadly, I am a mathematician, so I only believe in using things when I understand how they work and what the underlying concepts and assumptions are (such as standard deviation really should only be used for Normal-like distributions). It looks like I have got a challenge on my hands here, as the explanation only shows you how to calculate it, and I can't work out what underlies it all. More research and brain matter needed!

This will probably go on the back burner for a while though, as the amount of work is piling up. A pleasure to be enjoyed later!

Hi Graham

I've sent you a pm - Cronbach's alpha is just a measure of how far items in a scale hang together. If it's low (less than around 0.7) this suggests the scale isn't clearly measuring one particular construct (here, self-reported fatigue). It seems unlikely on the face of things that it is only measuring one construct - it seems to be attempting to measure a range of things, all quite badly. Any respected psychological journal would want data on a scale's internal consistency before accepting it for publication.

Jenny

 

[Graham]

If I was to be completely honest, compared to pre illness, I would have to score each item as 'much more than normal' but this didn't allow any latitude to record the exacerbation of symptoms during a crash.

Thanks Marco - you are 39 on the list. Actually your scores are not at all unusual.

May I quote your sentence above please in our report (anonymously)? I have another similar one from another person: I'm now struggling to put into words how I interpret the limitations of the Chalder scale, and it would help.

 

[Marco]

Thanks Marco - you are 39 on the list. Actually your scores are not at all unusual.

May I quote your sentence above please in our report (anonymously)? I have another similar one from another person: I'm now struggling to put into words how I interpret the limitations of the Chalder scale, and it would help.

Of course you can Graham.

Thinking about this scale (not very deeply I admit) I would expect that the internal reliability would depend very much on the cohort being studied. Whilst we PWME experience physical and cognitive energy limitations other diseases might show more of one and less of the other or exclusively either physical or cognitive 'fatigue'.

Regarding the comparison with how you were pre-illness, the scale may have some utility as a screening device. It is relatively straightforward to compare whether you are the same or worse post onset compared to pre onset as any deviation from having no problems might suggest the need for further investigation.

The problem comes in trying to use the scale for measuring changes during chronic illness. Regardless of good spells or bad spells, short of a miraculous full recovery, you are always going to score your capabilities as worse than pre-illness (and again if honest much worse and therefore no means of tracking deterioration).

What would make more sense would be to ask patients to indicate if their condition has changed in the last 'X' period for better or for worse and if so, then to ask how much better or worse they feel on each item now compared with the earlier period.

Of course all this subjective comparison of one time period to another is subject to recall bias (rose tinted memories of how you used to be) and report bias (rose tinted reporting of how you are now following cognitive re-education).

 

[Graham]

Thanks Marco

I totally agree. My son went down with ME when he was 8: he is now 30. He hasn't got the foggiest idea what it is like to feel well.

By the way, you are number 40 - I miscounted!

 

[slayadragon]

Graham, where did you get this scale?

When I looked at the Chalder 1993 paper, it looked like she took out the one about "Do you have problems finding the right word" (which you have here) and left in "Do you have problems thinking clearly" (which you left out).

"Based on the above mentioned analysis, three items (Do you start things without difficulty but get weak as you go on ?; Do you find it more difficult to find the correct word?; Have you lost interest in the things you used to do? were eliminated from the scale, leaving 11 items. Cronbachs alpha for this revised version was 0.8903. As the principal components analysis revealed two possible solutions Cronbachs alpha was also calculated for the two sets of items, (physical fatigue items 1-8; mental fatigue items 9-14). Results were 0.845 and 0.82 1 respectively."

http://wwwcache1.kcl.ac.uk/content/1/c6/01/47/68/PDF-109.pdf

Most of the papers (e.g. PACE study) looking at this scale refer to this 1993 paper, and not to anywhere else, so I was guessing that was what they used.

Here's what I was thinking (from that paper) that the scale would include.

Thanks much for your help.

Best, Lisa

*


Do you have problems with tiredness?
Do you need to rest more?
Do you feel sleepy or drowsy?
Do you have problems starting things?
Are you lacking in energy?
Do you have less strength in your muscles?
Do you feel weak?
Do you have difficulty concentrating?
Do you have problems thinking clearly?
Do you make slips of the tongue when speaking?
How is your memory?

 

[slayadragon]

Usual:

Do you have problems with tiredness? 2
Do you need to rest more? 2
Do you feel sleepy or drowsy? 1
Do you lack energy? 2
Do you have trouble starting things? 1
Do your muscles have less strength? 1
Do you feel weak? 2
Do you have difficulty concentrating 1
Do you find it more difficult to find the correct word? 2
Do you make slips of the tongue when speaking? 2
How is your memory? 1

*

"Bad Patch" (when doing a heavy detox last winter)

Do you have problems with tiredness? 3
Do you need to rest more? 3
Do you feel sleepy or drowsy? 3
Do you lack energy? 3
Do you have trouble starting things? 3
Do your muscles have less strength? 3
Do you feel weak? 3
Do you have difficulty concentrating 3
Do you find it more difficult to find the correct word? 3
Do you make slips of the tongue when speaking? 3
How is your memory? 3

 

[Graham]

Thanks for the answers Lisa, and the comments. The Chandler scale is not clear cut at all. I got the questions from the actual protocol paper for the PACE trial, although I actually got the order slightly wrong, and the phrasing is marginally different (e.g. do you have less strength in your muscles vs do your muscles have less strength). That is because originally I used someone else's summary of what the questions were. I also fiddled the last one about memory a bit, because strictly speaking it works the reverse to the others, if you just use the "more than usual" description.

The only thing is that you haven't given me scores for a good spell, and I am very interested in the variations in the scores over the course of a year. I'd be very grateful if you could add that one on, then you can be number 45 on the list.