Correspondence

Guns in the Home

N Engl J Med 1994; 330:1157-1159April 21, 1994

Article

To the Editor:

The Journal has published several articles asserting that strict gun-control laws reduce the rates of suicide1,2 and homicide3-5. If such laws had a major effect on both suicide and homicide rates, one would expect that these rates would be positively correlated with each other. We tested this hypothesis by analyzing suicide and homicide rates for all 73 nations that listed these data in the 1991 Demographic Yearbook6. We used the computer program SPSS/PC+, version 4.0, to analyze the data. We found no correlation (R = 0.135). A scatter plot and quadratic and cubic regression showed no nonlinear correlation. This result is consistent with U.S. suicide and homicide rates. For example, the U.S. homicide rate reached its peak for this century in 1980, when the suicide rate was below the midpoint of its range. Firearm laws vary in strictness from one country to another. If strict laws were a determinant of both suicide and homicide rates, these rates would be positively correlated with each other. The lack of correlation does not rule out an effect on either suicide or homicide and no effect on the other, nor does it rule out minor effects on both. Nevertheless, claims that strict firearm laws result in large reductions in both suicide rates and homicide rates should be viewed with skepticism.

David C. Stolinsky, M.D.
Stefanie A. Stolinsky, M.S.
1200 N. State St., Los Angeles, CA 90033

6 References

1. 1

   Sloan JH, Rivara FP, Reay DT, Ferris JAJ, Kellermann AL. Firearm regulations and rates of suicide: a comparison of two metropolitan areas. N Engl J Med 1990;322:369-373
   Full Text | Web of Science | Medline

2. 2

   Boyd JH. The increasing rate of suicide by firearms. N Engl J Med 1983;308:872-874
   Full Text | Web of Science | Medline

3. 3

   Sloan JH, Kellermann AL, Reay DT, et al. Handgun regulations, crime, assaults, and homicide: a tale of two cities. N Engl J Med 1988;319:1256-1262
   Full Text | Web of Science | Medline

4. 4

   Mahler AJ, Fielding JE. Firearms and gun control: a public-health concern. N Engl J Med 1977;297:556-558
   Full Text | Web of Science | Medline

5. 5

   Loftin C, McDowall D, Wiersema B, Cottey TJ. Effects of restrictive licensing of handguns on homicide and suicide in the District of Columbia. N Engl J Med 1991;325:1615-1620
   Full Text | Web of Science | Medline

6. 6

   1991 Demographic yearbook. New York: United Nations, 1992.
   

Author/Editor Response

The authors reply:

To the Editor: Homicide and suicide are distinctly different forms of intentional injury. Each has multiple causes, and their descriptive epidemiologic characteristics differ markedly. Unadjusted comparisons (such as that presented by Stolinsky and Stolinsky) between international suicide and homicide rates are subject to bias from multiple confounding factors (e.g., social, economic, cultural, and behavioral factors), which can induce a spurious association or mask a real one.

The studies cited by Stolinsky and Stolinsky used retrospective cohort series or longitudinal time series to evaluate a possible association between the availability of firearms and the risk of violent death in defined populations. They sought to control for several sociographic and demographic factors. These studies offer a better test of the association than the correlational study conducted by Stolinsky and Stolinsky, which did not examine the availability of firearms directly.

One of us has undertaken two separate studies of a possible association between the availability of firearms and violent deaths in two Pacific Northwest regions in close proximity, which, though similar in many ways, have taken different approaches to the regulation (and therefore the availability) of handguns1,2. Taken together, these two studies suggested that populations living in areas with a greater availability of firearms had increased homicide rates, but the effect on suicide rates was smaller (except among people who were 15 to 24 years old).

In a cogent editorial accompanying the report on the first of these studies, Mercy and Houk made a compelling “call for science” and mapped future needs in the epidemiologic analysis of firearm-related intentional injuries3. They suggested that future studies shift from large ecologic analyses to those designed to assess the availability of firearms and an individual person's risk of violent death.

In two well-constructed case-control studies, Kellermann et al. did just that4,5. Examining three areas of the United States, they convincingly showed that gun ownership in the home is independently associated with an increased risk of homicide and suicide.

Injuries from firearms can and should be viewed not only as a social and criminologic problem but also as a public health issue. Further research is needed to evaluate any association between violent deaths and injuries and such factors as modification of firearms, education about firearm safety, and violence-resolution programs.

Information from future studies that are epidemiologically valid can be used to guide policies that can decrease the toll from firearm-related injuries and deaths.

John Henry Sloan, M.D., M.P.H.
St. Joseph Rehabilitation Hospital, Albuquerque, NM 87102

Thomas Koepsell, M.D., M.P.H.
University of Washington, Seattle, WA 98195

5 References

1. 1

   Sloan JH, Kellermann AL, Reay DT, et al. Handgun regulations, crime, assaults, and homicide: a tale of two cities. N Engl J Med 1988;319:1256-1262
   Full Text | Web of Science | Medline

2. 2

   Sloan JH, Rivara FP, Reay DT, Ferris JAJ, Kellermann AL. Firearm regulations and rates of suicide: a comparison of two metropolitan areas. N Engl J Med 1990;322:369-373
   Full Text | Web of Science | Medline

3. 3

   Mercy JA, Houk VN. Firearm injuries: a call for science. N Engl J Med 1988;319:1283-1285
   Full Text | Web of Science | Medline

4. 4

   Kellermann AL, Rivara FP, Somes G, et al. Suicide in the home in relation to gun ownership. N Engl J Med 1992;327:467-472
   Full Text | Web of Science | Medline

5. 5

   Kellermann AL, Rivara FP, Rushforth NB, et al. Gun ownership as a risk factor for homicide in the home. N Engl J Med 1993;329:1084-1091
   Full Text | Web of Science | Medline

Author/Editor Response

Stolinsky and Stolinsky's logic does not require a large correlation between total homicide and suicide rates. Current theories of weapon choice suggest that the level of firearm ownership will influence the proportion of violent injuries that are due to guns1. One might thus expect to find a relation between the fraction of homicides committed with guns and the fraction of suicides committed with guns. This relation has been documented in U.S. cities1.

There are reasons to be cautious in drawing inferences from the available international data. The samples are not representative, and recording practices vary. Nevertheless, there is evidence to suggest that the fractions of suicides and homicides committed with guns are also correlated across nations.

We located two sets of data on national homicide and suicide rates reported as means. The first set, reported by the World Health Organization, is for five-year periods between 1950 and 1969. We use the 1960-1964 period because it includes the largest number of countries (15)2,3. The second set of data, reported by Killias, includes homicide and suicide rates in 14 nations for various years between 1983 and 19864. The Pearson correlation between the fractions of homicides and suicides committed with guns was 0.69 in the first set of data and 0.49 in the second. Lester has reported similar findings with a third set of data5.

We suspect that cultural differences account for much of the variation in relative homicide and suicide rates between nations. Still, these results suggest that both types of mortality are sensitive to the availability of firearms.

Those interested in international studies of gun availability and violence should consult Killias's excellent work. Using survey data to measure national levels of firearm ownership, Killias found that access to guns is strongly correlated with both homicide and suicide rates4. The direct measurement of gun ownership is preferable to inferences based on the correlation between vital rates.

David McDowall, Ph.D.
Colin Loftin, Ph.D.
University of Maryland, College Park, MD 20742-8235

5 References

1. 1

   Cook PJ. The technology of personal violence. In: Tonry M, ed. Crime and justice: a review of research. Vol. 14. Chicago: University of Chicago Press, 1991:1-71.
   

2. 2

   World Health Organization. Suicides by means used, 1950-1969. World Health Statistics Report 1973;26:164-205
   

3. 3

   World Health Organization. Homicides, 1950-1970: five year totals of deaths by homicide specifying the victim's sex and the means used. World Health Statistics Report 1973;26:306-310
   

4. 4

   Killias M. International correlations between gun ownership and rates of homicide and suicide. Can Med Assoc J 1993;148:1721-1725
   Web of Science | Medline

5. 5

   Lester D. The availability of firearms and the use of firearms for suicide: a study of 20 countries. Acta Psychiatr Scand 1990;81:146-147
   CrossRef | Web of Science | Medline

Citing Articles (4)

Citing Articles

1. 1

   Stefanie A. Stolinsky, David C. Stolinsky. (2000) Letters to the Editor. The Journal of Trauma: Injury, Infection, and Critical Care 48:6, 1168-1169
   CrossRef

2. 2

   David C. Stolinsky, Stefanie A. Stolinsky. (1997) Letters to the Editor. The Journal of Trauma: Injury, Infection, and Critical Care 42:3, 573
   CrossRef

3. 3

   Petko T. Dontschev. (1995) Applied criminology. Current Opinion in Psychiatry 8:6, 376-379
   CrossRef

4. 4

   David C. Stolinsky, Stefanie A. Stolinsky. (1995) Letter to the Editor. The Journal of Trauma: Injury, Infection, and Critical Care 38:3, 464
   CrossRef