The study from which these claims arise was also done by Dr. Arthur Kellermann and a bunch of his associates.  It was reported in volume 327 (8-13-92) of the New England Journal of Medicine (NEJM).

According to the researchers' abstract, they "studied all suicides that took place in the homes of victims in Shelby County, Tennessee, and King County, Washington, over a 32-month period."

"For each suicide victim (case subject), we obtained data from police or the medical examiner and interviewed a proxy.  Their answers were compared with those of control subjects from the same neighborhood, matched with the victim according to sex, race, and age range.  Crude and adjusted odds ratios were calculated with matched-pairs methods."

The results, again according to the abstract, were that:  "During the study period, 803 suicides occurred in the two counties, 565 of which (70 percent) took place in the home of the victim.  Fifty-eight percent (326) of these suicides were committed with a firearm.  After excluding 11 case subjects for various reasons, we were able to interview 80 percent (442) of the proxies for the case subjects."

Matching controls were identified for 99 percent of these subjects, producing 438 matched pairs.  Univariate analyses revealed that the case subjects were more likely than the controls to have lived alone, taken prescribed psychotropic medication, been arrested, abused drugs or alcohol, or not graduated from high school.  After we controlled for these characteristics through conditional logistic regression, the presence of one or more guns in the home was found to be associated with an increased risk of suicide (adjusted odds ratio, 4.8, 95 percent confidence interval, 2.7 to 8.5)."

The methods and deficiencies of this study were basically the same as for the later study about homicide and guns in the home.  Note that essentially zero of the people interviewed about victims of gun suicide could falsely deny that the victim or someone in the household possessed a gun.  Note, too, that the result was actually a ratio between 2.7 and 8.5.

SELECT FOR INFO ON THAT HOMICIDE STUDY if you wish at this point.

In the NEJM article, the researchers claimed that "limiting access to firearms could prevent many suicides," citing as support a Canadian study (Rich, '90) that had actually concluded that the reduction of suicides by use of guns which followed the increased regulation of the guns was totally offset by an increase of suicide by leaping from high places.  The researchers also cited six studies they claimed had studied variations in the rates of gun ownership and suicide across different places or over time.  Of these, four did not even try to measure gun ownership and one used a measure that was not valid.  None of their citations included 12 studies that really did measure the association between gun ownership rate and suicide rate, with 10 of them finding that there was no significant association.