3 ms·

Unfortunately you have a fundamental misunderstanding of health experimental design. Double-blind means that neither the participants nor the experimenters know

by mabcat 11y ago

Unfortunately you have a fundamental misunderstanding of health experimental design. Double-blind means that neither the participants nor the experimenters know which participant is allocated to which group. It doesn't say anything about how many groups there are or what is done to each group. I haven't heard of Dr Bredesen, but they could certainly run a double-blind study to validate their treatment regime. Treatments that haven't been validated in this way have a high risk of being ineffective but appearing effective due to experimenter error, placebo effect, demand characteristics, etc. Requiring convincing evidence in the form of a double-blind trial isn't purism, it's sensible caution based on experience. edit: you're also on shaky ground statistically. There's a rule of thumb that if interactions between several variables are present in a treatment effect, 70% of that effect will show up on the individual variables (as main effects). It's very unlikely that there exists a treatment where all the effect is on a 35-variable interaction and none of the effect is visible when looking at each variable singly.

schoen 11y ago
> It's very unlikely that there exists a treatment where all the effect is on a 35-variable interaction and none of the effect is visible when looking at each variable singly. Maybe something about genetic modifications where 35 different point modifications to the genome have to happen in order to allow synthesis of a working protein? If you measured the effect of each point modification, you might measure it as zero. I'm analogizing from trying to do statistical tests of whether a password is correct. At least in systems that aren't vulnerable to timing attacks, you can't just test part of the password and see if you get a little bit more access to the system as a result. :-)
mhkool 11y ago
You need to read my comment again. I never stated nor think what you wrongly assume. Your reasoning about the work of Dr Bredesen shows that you a re ignoring the facts. reversal of Alzheimer in 9 of 10 patients is a very significant results since nobody has ever reversed Alzheimer. The fact that the sample was small is only a reason to do a followup study with a larger number of patients, which is exactly what he is doing. The desire to reduce protocols to only a single variable is based on the wrong assumption that treatments can be successful with only one variable changed and excludes all treatments where more variables play a role.

by mabcat 11y ago

schoen 11y ago

> It's very unlikely that there exists a treatment where all the effect is on a 35-variable interaction and none of the effect is visible when looking at each variable singly. Maybe something about genetic modifications where 35 different point modifications to the genome have to happen in order to allow synthesis of a working protein? If you measured the effect of each point modification, you might measure it as zero. I'm analogizing from trying to do statistical tests of whether a password is correct. At least in systems that aren't vulnerable to timing attacks, you can't just test part of the password and see if you get a little bit more access to the system as a result. :-)

mhkool 11y ago

You need to read my comment again. I never stated nor think what you wrongly assume. Your reasoning about the work of Dr Bredesen shows that you a re ignoring the facts. reversal of Alzheimer in 9 of 10 patients is a very significant results since nobody has ever reversed Alzheimer. The fact that the sample was small is only a reason to do a followup study with a larger number of patients, which is exactly what he is doing. The desire to reduce protocols to only a single variable is based on the wrong assumption that treatments can be successful with only one variable changed and excludes all treatments where more variables play a role.