In fact, as people with ME/CFS go through good and bad patches, the changes differ greatly between individuals. For some people, the number of items that cause problems (the bimodal score) stays the same – just the severity changes. For others there is no change at all. In a few the changes are really very large: a gap of up to 10 points on the bimodal score and 18 on the Likert score.

Here is an attempt to illustrate this for the first 60 in our survey: it is easy to see that the variations in scores are widely different. Each line represents a patient; the colour represents whether that patient is mild, moderate or severe, and the line goes from their score in a bad patch to that in a good patch. Dots show no change, and vertical lines show no change in the bimodal score.


Longer lines clearly show greater changes. The further across a line goes indicates how many factors change from being a problem to not being a problem.

Out of the 60 results here, 45 of them have a bimodal score of 11 on a bad patch; 20 of them stay there on a good patch.

Looking at the general pattern of the lines, two things seem to be clear. One is that, for approximately half of these individuals, changes in Likert scores are matched with changes in Bimodal scores (these are the diagonal lines), so either scoring system for them registers their improvements. But for many of those at the worse end of the scale, with bimodal scores of 11, if they improve at all, they stay at a Bimodal score of 11 or 10. On average, lines go up 4 Likert points for every Bimodal point across.

The conclusion to draw from this is that there is a deep well at the bottom of the scale that fails to reflect the severity of the illness, and it is difficult for patients at this level of illness to register changes. Adding to that is the fact that many lines end at 11/33, suggesting that the scale is does not go low enough. In fact most lines stop at the bottom of a column. This adds doubt as to whether the scale is able to measure deterioration fully, which in turn could relate to the debate on whether the use of this scale in the PACE trial is sensitive enough to record harm.

Several people commented on this when returning the questionnaire. Here are two examples:

I think grading the questions is hard too as, compared with when I was well I am very bad indeed as on my best days I still can’t sit up in a wheelchair for longer than a couple of mins, but you automatically try to think of grades and put some as less bad simply because you are even worse in bad patches!

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.


Here is a table showing the changes in the results of the whole survey, from a bad patch to a good patch within a year. Click on each of the text boxes below to get an explanation of how it works, and to get the values entered.

 If one person scores 9
 Here are all of the
Notice that 35 (28%) of

summary          more          details          further details          survey



pdf version  




item1a contact/comments > Home < Return to Summary survey further details more > 6. We need better