grabs

Steven Milloy

There is no doubt that gun-control advocates will exploit the tragic
shooting death of 6-year old Kayla Rolland by a 6-year old classmate and the
shooting of five persons in Pittsburgh restaurants. Adding fuel to this fire
may be a study published in the current Journal of the American Medical
Association.
A ban on the carrying of firearms in Cali and Bogota, Colombia, was
statistically associated with a reduction in homicides, according to a study
by University of Washington and Emory University researchers. But the study
is just more junk science-fueled, anti-gun propaganda.
The researchers, led by longstanding gun-control advocate Arthur
Kellerman, compared homicide rates on days when gun-carrying was banned with
rates on days when gun-carrying was legal. Enforcement of the gun bans
included establishment of police checkpoints and searching of individuals
during traffic stops and other routine law enforcement activity.
Based on 4,078 homicides during 1993 and 1994 in Cali and 9,106
homicides in Bogota during 1995 through August 1997, the researchers
reported the homicide rates were lower by 14 percent and 13 percent in Cali
and Bogota, respectively.
The researchers conclude: “Constitutional restrictions on police search
procedures would prevent transferring the methods used in Cali and Bogota to
any city in the United States. Our study suggests that police programs such
as those applied in Cali and Bogota can suppress interpersonal violence and
save lives. This kind of program may be suitable for regions of the world
where homicide rates are very high and programs of this type are
permissible.”
An accompanying editorial was less considerate of the Constitution. It
concluded: “Based on the quasi-experimental evidence on gun-carrying
enforcement, it seems implausible that further repeal of U.S. laws against
carrying concealed weapons would cause fewer gun injuries and homicides.”
But Mr. Kellerman, his co-authors and JAMA are going off only
half-cocked – if that.
First, even accepting the reported results at face value, it’s not
clear the Colombian experience with guns and homicide is even remotely
relevant to that of the United States. The Colombian homicide rate was 88
per 100,000 in 1993. The homicide rate in Cali was 124 per 100,000 in 1994.
The U.S. homicide rate for 1993 was 9.5 per 100,000. Since then, the
U.S. rate dipped to 6.3 percent, while the Colombian homicide rate rose to
about 15 times that in the United States. Much of this violence is related
to the undeclared civil war involving leftist guerrillas and right-wing
militias – both involved in drug trafficking.
But this isn’t the study’s only or even primary flaw.
The study data actually show that, in terms of crude homicide rate,
there were 50 percent and 37 percent more homicides in Cali and Bogota,
respectively, on days when the guns were banned. To produce the reported
results, the researchers applied to the crude rates a rarely used and
questionable statistical technique called “indirect standardization.”
It’s not necessary to discuss the merits of indirect standardization
because the underlying data are problematic. The authors note that 21
percent of the Cali homicides and 26 percent of the Bogota homicides were
not firearm related. Yet, these homicides were included in the analysis.
Given the relatively large percentage of irrelevant data and the
statistical marginality of the reported results, the study has little
scientific credibility. But I’m not surprised since Mr. Kellerman is
involved.
Mr. Kellerman is a pioneer in the study of gun violence as a health
issue. Reportedly, his interest was sparked by the death of Motown singer
Marvin Gaye, who had been shot by his father in a domestic argument. Mr.
Kellerman questions gun ownership on a moral and ethical level, asking
whether families are making their households and communities safer by owning
guns. His crusade has little use for good science and the scientific method.
Mr. Kellerman co-authored a 1986 New England Journal of Medicine study
reporting that of 398 persons killed in a home where a gun was kept, only
two were intruders shot while trying to enter. The study concluded that “the
advisability of keeping firearms in the home for protection must be
questioned.” But Mr. Kellerman and his co-author only considered the
relatively few cases involving death. They ignored the vast majority of
cases in which burglars or intruders were wounded or frightened away by the
use or display of a firearm.
Mr. Kellerman also has refused to share his study data with other
scientists, despite receiving federal grants through the Centers for Disease
Control and Prevention.
JAMA and Peter Cummings, a co-author of the JAMA study, may also have
less-than-scientific motives.
In September 1997, JAMA published a study co-authored by Mr. Cummings
reporting that in states with felony laws requiring safe gun storage,
accidental deaths of children by gunfire dropped by more than 40 percent.
Publication of the study coincided with the campaign for Initiative 676 in
Washington state, a provision that would, among other things, require
trigger locks on all handguns sold or transferred.
The then-JAMA editor George Lundberg denied the timing of the study had
anything to do with the Washington state initiative.
Sure. Mr. Lundberg eventually was fired in January 1999 by JAMA for
timing publication of an old survey of college students about oral sex to
coincide with President Clinton’s impeachment trial.
To the extent this study is used in the public policy debate about
firearms, it will be the gun-control advocates that are getting away with
murder.

Steven J. Milloy is a biostatistician, lawyer, publisher of
Junkscience.com and adjunct scholar at the Cato Institute.