Discussion in 'Latest ME/CFS Research' started by Dolphin, Sep 21, 2014.
I think (although am not sure) this won't show up in CFS systematic reviews as it didn't involve all CFS patients.
However, there were quite a lot of CFS patients in it: 40 out of 45 in the intervention group (88.9%) and 43 out of 45 in the control group (93.5%).
The authors do point out this limitation:
Here are the main results
Note that subjective fatigue is the fatigue subscale of the CIS-20 (8 questions). Fatigue severity is all 20 questions. So the fatigue subscale questions are counted in both - strip them away and the other 12 questions (concentration, motivation and activities) may not be significant.
Ta - some discussion of the protocol for this here: http://forums.phoenixrising.me/inde...t-seems-to-be-lacking-a-decent-control.20876/
On reading the protocol I was concerned that the control was worse than nothing.
Here's info on goal progress:
Mean of 4.16, although an improvement over the control group, is a long way from 10.
@Esther12 I was going to post something along the following lines. Then I noticed you already had - two years ago.
There was a big change in this score (from 41.56 to 120.67):
However, the change in the average daily step count was quite small: 6629 to 7077 (6.8% increase). This suggests people may have cut out other activities to do increased leisure-time activity (or perhaps they didn't do as much leisure-time activity as they claimed).
They improved on 4 measures and disimproved on 7. None of the changes look that large but still can be enough to change something from not statistically significant to statistically significant. But people with CFS tend to improve a small bit on average overtime and say they are a bit better again if they have had some sort of therapy, so not an ideal control group.
Discussion of the pedometer results:
Note that they groups were already at 6500-8500 before the trial.
Ref. 42 is the PACE Trial (the increase in the 6-minute walking test was small in the GET group, but there was no change in the CBT group).
More studies should probably highlight this issue. Although some/many of those who did not take part may have done so as they didn't feel they could exercise any more.
They don't mention the bigger trial of pragmatic rehabilitation, the FINE Trial, which flopped.
I only had a quick read of that (with a mild migraine), but thought I'd post my first thoughts.
So 'Leisure Time Physical Activity' is just self-report, while 'Physical Activity' was measured objectively with a pedometer and did not find a significant improvement.
Their control group certainly did much worse than PACE's SMC, but it's difficult to know if that shows my concerns about their control were justified - we need control for their control!
@Dolphin re Goal Progress: I was just thinking the same thing, especially considering the emphasis placed upon goals in this intervention in the protocol:
I don't really know what sort of scores are to be expected for an outcome measure like this though. Not sure how much time to invest in digging in to this.
I'm not too happy with their justification for what should be seen as a clinically significant improvement.
So according to their pre-determined primary outcome, this was a null result?
I've got some sympathy for the argument that the sample size was too small... but I think patients have suffered more than researchers because of exaggerated claims about the efficacy of these sorts of interventions. If the effect size is smaller than you expected then you need to run a larger trial, and ideally one with a better control group.
Or it could be that those patients most likely to benefit from this sort of intervention are those most likely to pursue it. It is possible that patients have some awareness of the nature of their problems and can make informed decisions about what therapies are likely to be most helpful.
Yeah, I noticed that too.
I'm just looking at the protocol (thanks to Esther12 for drawing my attention to its existence). There's no sign of quite a lot of these (including the CFS diagnosis as an outcome measure, which should assess whether they satisfy it or not at the end, otherwise it's not an outcome measure).
In no way is this a solid piece of evidence, but from popular science/medicine blog, I got some vague info on pedometer data that might be of interest to people:
2000-3000 steps: About half an hours walking.
Apparently "The average U.S. adult walks about 5,900 steps daily" - whoa - I suddenly feel like I was really active before falling ill.
These classifications could be totally made-up:
That info makes the improvement report by this 4 Steps physical activity programme look even worse, but also the starting levels of physical activity look quite high.
I didn't see data specifically for the YAMAX SW-200 pedometer, but I did find people saying it was reliable and accurate.
As I (and others) specifically predicted in 2012.
Right. Translation: CFS patients on the whole are active enough so that deconditioning does not explain their fatigue.
Some of the PhD this is a part of is now available here: https://openaccess.leidenuniv.nl/handle/1887/31422
The rest should be available next Jan, including this bit:
Maybe of interest?
"Subgroup analyses revealed that minimal contact interventions had additional beneficial effects upon fatigue (g= 0.96) and depression (g= 0.85)."
Yeah, I find that minimal contact with them perks me up a bit too.
New paper on this from twitter:
Loading Tweet... https://twitter.com/statuses/816329468830740480
Another null result when it comes to objective measures...
Also notable that 55% of participants were working before being involved in this study... (98% of participants were women).
You can also try a Google Site Search
Separate names with a comma.