home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.med
- Path: sparky!uunet!cs.utexas.edu!sdd.hp.com!news.cs.indiana.edu!noose.ecn.purdue.edu!mentor.cc.purdue.edu!pop.stat.purdue.edu!hrubin
- From: hrubin@pop.stat.purdue.edu (Herman Rubin)
- Subject: Problems of studies Was: Re: Generic drug store brands.
- Message-ID: <C04tw5.AuL@mentor.cc.purdue.edu>
- Sender: news@mentor.cc.purdue.edu (USENET News)
- Organization: Purdue University Statistics Department
- References: <4zqr==b@dixie.com> <1992Dec31.093159.8158@spdcc.com> <C04L0w.3rB@monster.apd.saic.com>
- Date: Thu, 31 Dec 1992 16:31:17 GMT
- Lines: 25
-
- In article <C04L0w.3rB@monster.apd.saic.com> tomj@monster.apd.saic.com (Tom Johnson) writes:
- >dyer@spdcc.com (Steve Dyer) writes:
-
- >> What we do know is that Robaxin and placebo
- >>both seem to work equally well against back pain in controlled studies,
- >>making the measurement of any drug effect virtually impossible.
-
- >For what it's worth, this same question came up on yesterday's local
- >RadioDoc talk show. The RadioDoc confirms what Steve's been saying:
- >Oral muscle relaxants aren't distinguishable from placebos in
- >controlled studies.
-
- The typical controlled study is to have n_E subjects get the experimental
- treatment and n_P get the placebo, and observe how many in each group are
- changed. It is rare that subjects will get the experimental treatment
- sometimes and the placebo at other times. Now a drug can be effective
- on only a few, and this is unlikely even to be suspected on the first
- kind of study. However, if some benefit from the treatment but not the
- placebo, only the second kind of study will reveal it if the proportion
- is small and the placebo effect is present.
- --
- Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
- Phone: (317)494-6054
- hrubin@snap.stat.purdue.edu (Internet, bitnet)
- {purdue,pur-ee}!snap.stat!hrubin(UUCP)
-