Interesting study - but unfortunately badly written. At least the abstract anyway. Fails most of CONSORT for abstracts. E.g.: No info on participant N or characteristics. No data presented. Will read to see how reflects on overall quality of study.
-
-
Have a read David and let us know your thoughts. For me this type of study (very very difficult to do) is exactly what is needed to avoid the “well it’s not clear if it’s weight loss or carb restriction which is the effective part...”
3 replies 0 retweets 10 likes -
Replying to @Dr__Guess @dnunan79 and
I can't speak to the experimental design, but it appears that they've done dozens of anovas and t-tests without any control for multiple comparisons, which is not ideal
2 replies 0 retweets 10 likes -
Replying to @GidMK @Dr__Guess and
What I'm curious about is that LC group ate 2,5x more SFA yet looks like no change in LDL-C and apoB is not reported. This data is in the supplement although quite a remarkable result. Rises a lot of questions about the protocol.
2 replies 1 retweet 7 likes -
-
Replying to @Dr__Guess @POhukainen and
Makes sense when you look at the funders. Also looks like there's some sneaky outcome switching going on!pic.twitter.com/JoVM5Twfl8
3 replies 0 retweets 4 likes -
Replying to @GidMK @Dr__Guess and
Checked http://clinicaltrials.org entry. MetS not mentioned. Prim outcome = change in PFAs + lipoproteins. These reported 3rd/4th after MetS & secondary outcomes (e.g. DEXA). Unclear allocation concealment + all unblinded. n=16, P-values only and no 95% CIs anywhere in text.
1 reply 2 retweets 16 likes -
Use of "P < 0.05" and “trend” when >0.05 suggests wide CIs close to no stat difference. No mention by authors re: size of effect. Others can comment as to whether effect sizes = meaningful. For me = exploratory findings at best. Need replicating w >n, >robust, w <COI.
2 replies 1 retweet 14 likes -
Thanks for sharing your perspective. Personally I believe the findings despite the study's limitations (e.g., multiple comparisons) because they are consistent with previous research. Priors would have predicted similar outcomes IMO.
2 replies 0 retweets 5 likes -
This Tweet is unavailable.
That's actually a fantastic way to put it I'm stealing that 
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.