Disturbing news update- statistical manipulation

April 30, 2001

The piece below and the larger article in the link critique the AMA

Journal study allegedly showing that St. John's wort is ineffective

against depression (major, instead of the mild to moderate for which it

is indicated). The study data showed the opposite, despite Time Magazine

and similar misreporting. This is why people have so little faith in

medical testing. A shame, because good unbiased science is badly needed.

Considering how much better patients on high quality St. John's wort do

than on pharmaceutical SSRIs, especially with regard to side effects,

this cynical manipulation infuriates me.

Karen Vaughan

CreationsGarden

***************************************

>From Paul Bergner, with permission:

" So here's how the JAMA authors turned a successful trial of hypericum in

major depression into a failed one. Typical results for pharmaceutical

antidepressants are about 50% response to the drug (symptom scores

improved by 50% or more) and about 30% for placebo. In this trial, it was

about 52% for hypericum and 30% for placebo using the same criteria as

the metaanalysis that produced the typical results.. BUT, the authors in

the JAMA trial divided the responders into a " response " group and a

" remission " group. Of those who completed the trial, these

subgroups each showed a strong trend toward drug efficacy, but it did not

reach statistical significance (P=.07 for each subgroup). The combined

groups may have shown statistical significance, being a larger number of

patients, but the authors do not report on this one way or the other.

Reseachers used the HAM-D depression scale. Initially the subjects had a

score of at least 20, and a median of 22. On that scale, most patients

diagnosed with major depression according to DSM IV criteria score at

least 14. The researchers said that if patients reached a score of 7-12,

they were " responders " but had to score below 7 to be called

" remission. " There were 3 times as many patients in the remission group

as in the placebo, and this reached statistical significance (p=.02)

Results for the combined groups were typical of those for antidepressant

drugs taken as a group.

See the full article from Medical Herbalism (electronic preview article)

at

http://www.medherb.com/hypericum-jama.htm)

By all means link to it and publicize it. "

______________

GET INTERNET ACCESS FROM JUNO!

Juno offers FREE or PREMIUM Internet access for less!

Join Juno today! For your FREE software, visit:

http://dl.www.juno.com/get/tagj.

May 2, 2001

>

>The piece below and the larger article in the link critique the AMA

>Journal study allegedly showing that St. John's wort is ineffective

>against depression (major, instead of the mild to moderate for which it

>is indicated). The study data showed the opposite, despite Time Magazine

>and similar misreporting. This is why people have so little faith in

>medical testing. A shame, because good unbiased science is badly needed.

>

>Karen Vaughan

>CreationsGarden

a very interesting report on " good unbiased science " in the

pharmaceutical research is " Tales from the other drug wars " (guess which

drug wars).

Published by the Center for Health Services and Policy Research of the

U. of British Columbia. It can be downloaded (pdf format) from:

http://www.chspr.ubc.ca/drugwars2.pdf

rene

May 2, 2001

Karen,

This is frustrating news, so I'll immediately take some qi and blood movers.

Thanks for doing the homework.

Stephen

Karen S Vaughan [creationsgarden]

Monday, April 30, 2001 7:49 PM

; Paracelsus

Disturbing news update- statistical

manipulation