I was just looking at this paper earlier. I am not sure I follow Enlander's argument properly. I guess the 8% figure comes from interpreting the regression coefficient of 0.81, since 1 = 10%? But
Table 4 is for
associations of baseline characteristics with physical function. The mean(SD) scores for this at baseline and at follow-up are found in
Table 2, 4.4/100 points is a 9% improvement over baseline for physical function, but greater improvements for fatigue: an average difference of 6.8 points in fatigue score is about 21% of the total scale of 33 points, or about 26% of the average baseline score i.e. 26.5 points (or about 44% if taking into account that 11/33 points in Likert scoring is the 'neutral' score).
I think this paragraph is also interesting, particularly when there was no control group for natural course:
"About 74% (620 of 834) of patients had a decreased Chalder Fatigue score at follow-up and 64% (534 of 834) had improved by >2 points (our definition of a clinically useful improvement). In contrast, only 50% (416 of 834) of patients had an increased SF-36 physical function score at follow-up and only 16% (131 of 834) had improved by >11* points. In total, 14% (120 of 834) had clinically useful improvements on both scales."
* The text states >22, but this must be a typo because earlier it mentions >11, which is consistent with how they derived this threshold in the first place i.e. 0.5 SD of baseline score.