In article <199605151305.GAA04597@dfw-ix4.ix.netcom.com> phb@ix.netcom.com (Dr. Paul H. Blackman ) writes:
>You wrote:
>>> The really interesting part is the last
>>>paragraph....in which he stated the limitations of his study. In a
>word,
>>>he acknowledged his study did not consider or analyze how many times
>the
>>>presence of a firearms in the house prevented crimes.
>>I think you are referring to a different study. We are discussing the
>>case-control study published in NEJM 1993;329:1084-91.
>Tim's right. The limitations are from the initial Kellermann & Reay
>piece (43-1), a non-CDC-funded study which presumably demonstrated
>Kellermann's bona fides to the CDC. Since then, Arthur has gotten a
>bit more casual about noting all the limitations. And his colleagues
>are even worse. An annotated bibliography produced by Harborview
>Medical Center, and available through its Web site
>(http://www.guninfo.org -- I think the bibliography may be
>http://www.guninfo.org/annotated.html -- at least that's what it says
>at the top), concludes a defense of the Kellermann et al.piece we are
>discussing:
>"We can see from that in all three study designs (randomized trial,
>cohort study, case-control study), it is possible to measure the
>protective value of firearms just by ascertaining the outcome of
>homicide...."
Woah there. A limitation on the earlier (43-1) study is not
necessarily a limitation on the later case-control study. The authors
of the bibliography are quite correct when they state that a
case-control study could measure a net protective effect of firearms.
Tim