A New Decade of ME Research: The 11th Invest in ME International ME Conference 2016
Mark Berry presents the first in a series of articles on the 11th Invest in ME International ME Conference in London ...
Discuss the article on the Forums.

Effects of Bedtime Very Low Dose Cyclobenzaprine on Sleep in FM patients (2011 double-blind study)

Discussion in 'Sleep' started by CFS_for_19_years, Feb 23, 2016.

  1. CFS_for_19_years

    CFS_for_19_years Hoarder of biscuits

    Messages:
    2,016
    Likes:
    5,110
    USA
    http://www.jrheum.org/content/38/12/2653.long
    (2011)
    Effects of Bedtime Very Low Dose Cyclobenzaprine on Symptoms and Sleep Physiology in Patients with Fibromyalgia Syndrome: A Double-blind Randomized Placebo-controlled Study





    I've been taking flexeril since it was first introduced in the 1980's. I take it now 10mg at a time as needed for muscle spasms, but I will ask my GP if I could try taking it as a sleep aid every night at a lower dose, and still take the larger dose as needed for muscle spasms.
     
    Last edited: Feb 23, 2016
    Little Bluestem, Woolie and Dolphin like this.
  2. Woolie

    Woolie Senior Member

    Messages:
    1,930
    Likes:
    14,558
    What a refreshingly clear, well-designed and carefully reported study! Thanks for posting, @CFS_for_19_years!
     
  3. *GG*

    *GG* Senior Member

    Messages:
    6,167
    Likes:
    4,772
    Concord, NH
    Isn't the low number of participants not so great?

    GG
     
  4. Woolie

    Woolie Senior Member

    Messages:
    1,930
    Likes:
    14,558
    Yea, it would be nice to see this replicated in a larger group. But then they'd probably have to drop some of their sleep measures, would be difficult to do with huge numbers of people.

    It was also not a registered trial, and so not stated what the primary outcome measures were. Having pre-defined primary outcomes measures helps to limit the number of chances you have to find a significant effect somewhere. Because the authors measured 7 outcome variables, they had 7 chances to get something significant, which is probably too high. (same problem for the sleep outcomes; there were a lot of measures examined, and not all were significant).

    Still, to look at it the other way, using a small group underpowers your study, so you're less likely to find significant effects. If this study were replicated with a larger group, it might be really impressive (look at how many people the PACE authors had to include to get anywhere near a statistically reliable effect - that's because the effects of GET and CBT are so minute, you need a lot of power to find them). The main reason for needing a larger group is to check your patients' responses are truly representative of the group you're studying, and not some quirk of the particualr peoe you chose for the various treatment conditions.
     
    Valentijn likes this.
  5. CFS_for_19_years

    CFS_for_19_years Hoarder of biscuits

    Messages:
    2,016
    Likes:
    5,110
    USA
    The size of a study can be small and yet have statistically significant results (p <0.05) if the differences between the control group and the treated group are large enough. If you look at the 7 outcome measures, all of the differences between the control group and the treated group are significant (p<0.05). Also, they compared the baseline measurements vs. the 8-week measurements for the treated group and results were statistically significant.

     
  6. Woolie

    Woolie Senior Member

    Messages:
    1,930
    Likes:
    14,558
    I don't think that's true of the between-group comparisons, though, @CFS_for_19_years (most of the p values you report in the quote above are comparisons of baseline and after-treatment, not group comparisons*). And the group comparisons are what really count (otherwise, why bother to have a placebo control at all?)

    (*Edit: You're not supposed to compare a significant p value for one group with a non-significant p value for another and conclude there is a treatment effect. p values can't be compared in this way. You need to directly compare the treatment and control groups in the same analysis. Not a criticism of you though - many smart, graduate-level Psychology students still have trouble with this issue)
     
    Last edited: Feb 29, 2016
  7. CFS_for_19_years

    CFS_for_19_years Hoarder of biscuits

    Messages:
    2,016
    Likes:
    5,110
    USA
    You're right. Some of the outcome measures failed to show differences between and treated group and the non-treated. Thanks for noticing that. The reported measures were very tedious to read and understand, but you've got a keen eye.;)
     
    Valentijn and Woolie like this.
  8. Woolie

    Woolie Senior Member

    Messages:
    1,930
    Likes:
    14,558
    Not really. More a reflection of what a dull nerd I am. :rolleyes: If you look at enough of these kinds of reports, the key stuff just kind of starts leaping out at you.
     
    CFS_for_19_years likes this.

See more popular forum discussions.

Share This Page