I no longer will trust any reports from NCBI, the supposed "health" agency.

[Archae]

Why?

Because they did a "study" of homeopathy and diarrhea.

http://www.ncbi.nlm.nih.gov/pubmed/8165068

[Wounded Bear]

1. So, giving water to treat dehydration works just as well as....

giving water to treat dehydration?

Is that about the extent of it?

[immoderate]

2. That study is 20 years old...

Is there a follow-up?

--imm

[Major Nikon]

4. Yes

http://www.ncbi.nlm.nih.gov/pubmed/17034278

[immoderate]

11. Ahah!

Homeopathy does not work in Honduras!

--imm

[Major Nikon]

3. They didn't do the study, they just published the results

...along with thousands of others. It was published in the American Academy of Pediatrics journal which is one of the most respected on the subject, which for that reason alone is why I suspect NCBI published it.

Although this particular study produced a statistically significant result in the treatment group, subsequent studies found the results not to be repeatable which basically proves it's woo. More data on woo is not a bad thing. More data on woo is a good thing.

[Archae]

5. So basically they simply made up the results?

[pnwmom]

6. No.

If you understand how science works, you know that many, if not most, initial studies cannot be repeated but that does NOT mean that anyone made up the results.

[Major Nikon]

9. That's one possibility

Another is that they simply didn't perform a large enough study to rule out the hypothesis. If you roll two dice enough times, the most common sum will be 7. If you only roll them 10 times, you might come up with something other than 7. That doesn't mean 7 is not the most likely result or that you made up the numbers.

[sharp_stick]

7. Yeah that makes a lot of sense

[MNBrewer]

8. Well

1. NCBI isn't a "health agency" and it doesn't "report" anything. It's a database for biotechnology and medical information, including indexing scientific journal articles.
2. NCBI didn't "DO" a study of homeopathy. They indexed the journal article that someone else wrote and had published.
3. The article is from the journal "Pediatrics" and was published in 1994.

[GeorgeGist]

12. TREAD WINNER!

[WillowTree]

10. So, because they published the results.......

......of a preliminary study 20 years ago that concluded that further investigation might be useful, you've decided that they're entirely unreliable? Really??

[JJChambers]

13. This OP... Epic fail. n/t