3 ms·
> "The patients demonstrated improved brain function" So what? If you run trials with large enough sample size or allow sufficient researcher degrees of freedo
by nonbel 8y ago
> "The patients demonstrated improved brain function"
So what? If you run trials with large enough sample size or allow sufficient researcher degrees of freedom (multiple ways of measuring "brain function", reporting results whenever they look best, looking at subsets of the patients, etc), then you have 50% chance of the treatment group "demonstrating improvement".
For example, from the limited information available I see in this case they measured "brain function" in 5 different ways:
Change from baseline in cognitive: Alzheimer´s Disease Assessment Scale- Cognitive (ADAS-Cog) [ Time Frame: 78 weeks ]
Change from baseline in function: Alzheimer´s Disease Cooperative Study-Activities of Daily Living (ADCS-ADL) [ Time Frame: 78 weeks ]
Change from baseline in cognition: Mini-Mental State Examination (MMSE) [ Time Frame: 78 weeks ]
Change from baseline in global assessment: Clinical Dementia Rating-Sum of Boxes (CDR-SB) [ Time Frame: 78 weeks ]
Change from baseline in behavioral assessment: Neuropsychiatric Inventory (NPI) [ Time Frame: 78 weeks ]
https://clinicaltrials.gov/ct2/show/NCT02551809 https://clinicaltrials.gov/ct2/show/NCT02551809
Not that I agree with this type of analysis but it is going to be similar to what they did to determine "demonstrated improvement":
Lets say there was a 2.5% percent chance of "randomly" getting a result like theirs. Then (1-0.025)^5 ~ 88% of the time there would be "no demonstrated improvement" and ~12% of the time there would be.
How many clinical trials for Alzheimer's like this have been done? We'd expect at least ~1/8 of them to "demonstrate improvement" just based on how the studies are designed.
>"In any case, what is your expertise?"
The reason it is a scholarly practice to cite your sources is that the audience doesn't need to rely on heuristics like this. Don't take my word for it, just follow the links and assess the published evidence for yourself.