[Josh P]

Process of Science: Assume we have data indicating a strong positive correlation between acupuncture treatments and recovery of patients from, say, cocaine addiction. However, let's also assume that every hypothesis we have for a mechanism of action (i.e., how acupuncture could work to help cure addiction) can be shown to be false. The patients, however, all claim to know that the acupuncture is what cured them. Which of the following conclusions are supported by our data?


Must acupuncture be responsible for the patients - recovery?

[mgle3]

No. If you can only show a correlation, and every investigated mechanism of action is false, you wouldn't conclude that acupuncture is the cause of recovery.

Edit: Alties, this is a question of logic. If you only have correlation, it does not follow to conclude that acupuncture "must" be responsible.

Edit: "However, let's also assume that every hypothesis we have for a mechanism of action (i.e., how acupuncture could work to help cure addiction) can be shown to be false."

Did you miss this bit, Nosey?

Maybe people who seek help in overcoming cocaine addiction, whatever that help may be, are more likely to overcome it.

[hambert40]

Mechanism of action has historically not been used as a benchmark in medical studies, and a huge number of medical studies have been accepted without one --

About 80% of conventional drugs have unknown "mechanism of action".
Surgery is not supported by randomized studies that can prove a mechanism of action either. It's based on a hypothesis, trial and error. Then you have a statistical model of clinical experience.

So you will be left in a position that without any opposing evidence to the contrary that acupuncture was responsible for the patent's recovery and that the mechanism of action was "unknown" as it is in plenty of other studies.
You will have a positive correlation between the acupuncture treatments and the patients recovery, and it will invariably create a situation whereby other researchers will duplicate your work or not.

Then some armchair critic with no scientific credentials will come along and claim your study was "inconclusive".

[Ridiculous]

go to rehab.settle

you can know much more on this

[GAIL H]

>About 80% of conventional drugs have unknown "mechanism of action".
noseconenoesright. you can check this by your own eyes: descriptions of the most medications contain references that their mechanism is actually unknown! you can easily find the words like "presumably", "presumedly", "possibly" etc. and for each unknown "mechanism of action" there are hundreds of "scientific" theories, hypothesises, etc. (and note: for each "theory" and "hypo" there is at least one PhD thesis
So-called "scientific researches" in pharma is just almost RANDOM selection of agent/active substances and their proportions, and then a lot of tests during "high quality trials". Go and see the protocols of testings - you can see a huge numbers of "experiments" with varying parameters. This is just a permutation "science", there is no thinking at all. Am I wrong? Really? Ask SmartAzz, he knows.
This is "double/triple/... BLIND" method because they don't see what they are doing at all. Ha-ha-ha
Advice: turn on your own brains and think by yourself! Use common sense for analysis, don't let anybody to influence your opinion.
There is "Lies, Damned Lies, and Medical Science"