home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!news.larc.nasa.gov!grissom.larc.nasa.gov!kludge
- From: kludge@grissom.larc.nasa.gov (Scott Dorsey)
- Newsgroups: misc.health.alternative
- Subject: Re: homeopathy
- Date: 3 Jan 1993 22:31:11 GMT
- Organization: NASA Langley Research Center and Reptile Farm
- Lines: 51
- Message-ID: <1i7pffINNd7l@rave.larc.nasa.gov>
- References: <1992Dec29.183006.1832@netcom.com> <1hssu9INNgpt@rave.larc.nasa.gov> <1993Jan3.201728.4425@netcom.com>
- NNTP-Posting-Host: grissom.larc.nasa.gov
-
- In article <1993Jan3.201728.4425@netcom.com> kaminski@netcom.com (Peter Kaminski) writes:
- >In article <1hssu9INNgpt@rave.larc.nasa.gov> kludge@grissom.larc.nasa.gov
- >(Scott Dorsey) writes:
- >
- >>That's true, and without accurate double-blind testing, you don't really
- >>have much empirical evidence.
- >
- >No, you *do* have empirical evidence (by definition), you just don't have
- >bias from possible placebo effects removed. And without a control group,
- >you wouldn't have bias from other factors removed, either.
-
- You have evidence that is possibly invalid, and is therefore no evidence
- at all. As you have pointed out, a control group is essential (and something
- I just generally assumed as part of the double-blind methodology). The
- difficulty is that placebo effects are so amazingly powerful.
-
- >>This is a real problem with human beings, since we are such complex
- >>organisms and there are so many possible influences.
- >
- >It's a problem in clinical tests, wherein the goal is to control each
- >variable except the one being tested.
-
- Yes, but if you don't control all variables except the one being tested,
- you can never be sure of your results. It's just not possible to control
- all variables in the case of human beings, which is what makes testing on
- humans so difficult. I find it difficult enough to accept a lot of
- carefully performed tests because of the number of confounding variables,
- let alone accept hearsay.
-
- >But in homeopathic prescribing, it's not really a problem. The homeopath
- >just notes the value of a whole bunch of variables and finds the remedy
- >that's been empirically noted to be effective for that particular symptom
- >picture.
-
- Yes, but what is effective for one person might not be effective for
- another. Just because it's been noted in one case doesn't mean that it's
- any good at all. Even if it's been noted in a whole lot of cases doesn't
- mean that it will do anything. It's possible that some medication might
- have significant interaction with some environmental variable and work
- very well for people in one area, but be useless on a group of people that
- live somewhere else. Without accurate testing, you'll never know.
-
- >>After looking at how much testing the FDA does, how extensive
- >>it is, and how lousy a job they do, it's obvious that testing anything on
- >>humans with any amount of rigor is not an easy task.
- >
- >Yup. Reductionism tends to create lots of little details to take care of.
-
- But without a reductionist approach, you'll never be able to clean those
- little details up. And who knows? Some of them may be crucial.
- --scott
-