Arthur Kellermann

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Arthur L. Kellermann (born 1955) is an American physician, epidemiologist, and professor. He became Dean of the F. Edward Hébert School of Medicine - "America's Medical School" - at the Uniformed Services University of the Health Sciences on September 7, 2013.[1] Prior to this position, he served as Director to the RAND Institute of Health (since January 1, 2011). He was the founding chairman of the department of Emergency Medicine at Emory University in Atlanta, Georgia, USA, and founding director of the Center Injury Control at Rollins School of Public Health, a collaborating center for injury and violence prevention of the World Health Organization. His writings include more than 200 scientific and lay publications on various aspects of emergency cardiac care, health services research, injury prevention and the role of emergency departments in the provision of health care to the poor.[2][3][4][5]

Kellermann co-chaired the Committee on the Consequences of Uninsurance of the Institute of Medicine of the United States National Academies, of which he is an elected member. Kellermann holds career achievement awards for excellence in science from the Society for Academic Emergency Medicine, and the Injury Control and Emergency Health Services Section of the American Public Health Association.[2] As a 2006-2007 Robert Wood Johnson Foundation Health Policy Fellow, he joined the Professional Staff of the United States House Committee on Oversight and Government Reform in Washington, D.C. In 2007 he was presented with the John G. Wiegenstein Leadership Award by the American College of Emergency Physicians, their highest award.[3][6]

Kellermann is well known for his research on the epidemiology of firearm related injuries and deaths. In a 1995 interview, Kellermann saw firearm and other injuries not as random, unavoidable acts but as preventable public health priorities: "I grew up around guns. My dad taught me how to shoot when I was eleven or twelve years old. Firearms are fascinating pieces of equipment. I enjoy the sport of shooting, although I rarely shoot anymore. However, as a clinician, as someone who is committed to emergency medicine, it is equally evident to me that firearm violence is wreaking havoc on public health."[7]

Several academic papers have been published severely questioning Kellerman's methodology, selective capture of data, and refusal to provide raw data from his gun-risk studies so as to substantiate his methods and result validity. While Kellerman has backed away from his previous statement that people are “43 times more likely” to be murdered in their own home if they own and keep a gun in their home, he still proposes that the risk is 2.7 times higher. The critiques included Henry E. Schaffer,[8] J. Neil Schuman, and criminologists Gary Kleck,[9] Don Kates, and others.[10]

In this polarized debate, Kellermann’s studies, which attempted to prove the risk of mortality associated with gun ownership, attracted criticism from gun-rights organizations and individuals.[11] The National Rifle Association of America contends that Dr Kellermann “severely understates defensive uses of guns,” and that his “conclusions provide anti-gunners propaganda.”[12]

Kellermann's published studies on gun ownership[edit]

Kellermann states that as an emergency room doctor, he noted that the number of gunowners injured by their own gun or that of a family member seemed to greatly outnumber the number of intruders shot by the gun of a homeowner, and therefore he determined to study whether or not this was in fact true. Kellerman's health-case approach, however, has been viewed as generally not having efficacy for assessing problems of sociology. Author Henry E. Schaefer M.D. published a rebuttal paper[10] to Kellerman’s 1993 publication that same year, in which Schaefer notes several flaws, starting with Kellerman's use of the case control method (CCM) of analysis. The generally-held view within academia is that CCM cannot demonstrate causation.


In his first publication on the subject, in 1986, Kellermann studied all gunshot related deaths in Seattle over six years, and found that

  • 54% of firearm-related deaths occurred in the home where the gun was kept
  • 70.5% of these (firearm-related deaths in the home where the gun was kept) involved handguns
  • 0.5% of these (firearm-related deaths in the home where the gun was kept) involved an intruder shot while attempting entry
  • 1.8% of these (firearm-related deaths in the home where the gun was kept) were judged by police as self-defense
  • there were 1.3 times as many accidental firearm-related deaths in the home where the gun was kept as self-protection shootings
  • there were 4.6 times as many criminal firearm-related homicides in the home where the gun was kept as self-protection shootings
  • there were 37 times as many suicides in the home where the gun was kept as self-protection shootings.

He concluded that "the advisability of keeping firearms in the home for protection must be questioned".[13] Critics of this study noted that it was restricted to firearm-related deaths, effectively excluding incidents in which gun owners used their firearm to injure and frighten away an intruder. But the study also excluded incidents in which individuals were non-fatally injured in a firearm accident, criminal assault or suicide attempt, as well as instances in which a homeowner used a gun to threaten or terrorize another member of the household, as sometimes occurs in the context of domestic violence. A subsequent Kellermann-led study identified both fatal and nonfatal injuries occurring in homes in 3 cities – Seattle WA, Memphis TN, and Galveston TX. It noted that for every time a gun in the home was used to kill an intruder, there were four accidental shootings, seven criminal assaults or homicides, and 11 attempted or completed suicides. (J of Trauma, August 1998. pp: 263-267). He then developed the now much criticized 43:1 ratio that states every time a gun is used in self-defense, it is 43 times more likely to be used in a homicide, suicide, or accidental shooting.

The ratio was numerically accurate but, according to gun-rights groups, misleading because it compared harmful life-taking uses of guns not to life-saving defensive uses (the benefit corresponding to the harms of lives taken with firearms), but rather only to the tiny subset of defensive uses that involve killing a criminal assailant, i.e. justifiable homicides. The NRA and other gun-rights groups argued the ratio that seemed to imply a sort of cost-benefit ratio for gun ownership was actually nothing of the kind because, allegedly, it did not take account of any benefits that corresponded to its costs.


In 1988, Kellermann published a study comparing robberies, burglaries, assaults, and homicides in Seattle, Washington and Vancouver, British Columbia, a city "similar to Seattle in many ways" that had "adopted a more restrictive approach to the regulation of handguns." The study found that

  • both cities had similar rates of burglary and robbery
  • in Seattle, the total rate of assaults with any weapon was modestly higher than that in Vancouver
  • rates of homicide by means other than guns were not substantially different in the two study communities
  • the rate of assaults involving firearms was seven times higher in Seattle than in Vancouver
  • the rate of being murdered by a handgun was 4.8 times higher in Seattle than in Vancouver.

The study concluded that restricting access to handguns may reduce the rate of homicide in a community by reducing the lethality of assaults.


In 1993, Kellermann responded to the criticism of his 1986 paper with a case-control study[14] of the rates of all homicides in the victim's home in Cleveland, Ohio, Memphis, Tennessee, and Seattle over five years, in homes where a gun was kept versus homes where a gun was not. This study found that

Kellermann's 1993 Table 4 Variables Included in the Final Conditional Logistic-Regression Model
Variable Adjusted Odds Ratio 95% CI
Any household member used illicit drugs 5.7 2.6-12.6
Home rented 4.4 2.3-8.2
Any household member hit or hurt in a fight in the home 4.4 2.2-8.8
Case subject or control lived alone 3.7 2.1-6.6
Gun or guns kept in the home 2.7 1.6-4.4
Any household member arrested 2.5 1.6-4.1
  • 23.9% of homicides occurred in the victim's home
  • 35.8% of the controls (homes where there was not a homicide) kept a firearm in their home
  • 45.4% of all victims of homicides in their home kept a firearm in their home
  • 57.7% - 64.7% of victims of firearm homicides lived in a home where one or more guns were kept (correction to original paper)
  • other protective measures, (reinforced doors, deadbolts, burglar alarms, and bars on the windows) were associated with small (about 0.8 times) reductions in risk of homicide in the home
  • after adjusting for other factors (such as a police-report history of violence in the home, a convicted felon in the home, drug or alcohol abuse in the home, race, etc.) there remained an independent 2.7 times increase in risk of homicide, specifically associated with a firearm in the home; this risk was not attributable to any particular "high risk" subgroup(s) identifiable by the above factors but was evident to some degree in all subgroups
  • this risk was essentially entirely attributable to being shot by a family member or intimate acquaintance with a handgun which was kept loaded and unlocked in the house
  • this risk was significantly less than the increased risk due to sociological factors (rental of a home instead of ownership, living alone) but close to that associated with the presence of a convicted felon in the home (see table at right).

These results were interpreted by Kellermann as confirming the 1986 finding that, on net, a firearm in the home represents a greater risk overall than the protection it may offer against intruders, either indirectly or by discouraging potential assaults. Kellermann noted that the study demonstrates the pervasiveness of domestic assault, as compared to better publicized crimes such as home invasion, but continued to stress the role of handguns in increasing the lethality of such assaults.

Critics of Kellermann's 1993 paper responded with a number of objections: (e.g. the guns/homicide association could simply reflect the fact that people already at risk of homicide are more likely to acquire guns for self-protection; the study population was urban and therefore higher risk in general, compared to suburban or rural areas), and (e.g. that members of rival gangs were tabulated as "family member or intimate acquaintance"; that the data was cherry-picked).

Kellermann also was criticized for not reporting what fraction of homicides in his sample were committed with guns kept in the victim's home.[15] If few homicides were committed with such guns, it would be unlikely that it was home gun ownership itself that caused an elevated risk of being murdered.

Critics also claimed that it was suspicious that Kellermann did not release his data immediately upon publication. SUNY-Buffalo's Lawrence Southwick, among others, publicly speculated "that Kellermann's full data set would actually vindicate defensive gun ownership." [1]

Centers for Disease Control and Prevention grants do not require the individual investigator to make data public until there are no more publications to be developed from them. After publishing additional analyses, Kellermann released the dataset to the Inter-university Consortium for Political and Social Research, the world’s largest archive for social science research [2]. ICPSR released the data for public access on May 30, 1997 [3]

Critics claimed, however, that this was only a truncated version of Kellermann’s full dataset, arguing that it omitted his crucial data on whether guns used in the firearm homicides he studied belonged to anyone in the victim’s household,[16] a critical issue in judging the plausibility of his conclusion that owning a gun caused an increased risk of being murdered. This information had clearly been gathered by Kellermann since it was used in another of his studies.[17] Once this information was taken into account, it was claimed that the effect of household gun ownership on the risk of homicide could not have been more than 6% of the effect that was estimated by Kellermann.[18]

Other work[edit]

Firearm injury prevention research accounts for a minority of Kellermann’s research output. Only 30 percent of his papers address the issue. He has published extensively in other areas of emergency medicine and public health, including studies of emergency cardiac care, use of diagnostic technologies in the emergency department, and a promising treatment for traumatic brain injury. He has also published research on the role of emergency departments in providing health care to the poor, the role of insurance [4] and the situation of the uninsured [5]. In recent years, he has written about domestic preparedness to respond to different forms of terrorism [6]. Kellermann was very instrumental in the planning and implementation of the American Heart Association's "Racing the Clock to Restart Atlanta's Hearts" initiative, one of the largest cardiopulmonary resuscitation and automated external defibrillator programs in the nation. He also played an important role in the Institute of Medicine’s three-volume report on the Future of Emergency Care in the United States, [7].


  1. ^
  2. ^ a b "Arthur Kellermann, M.D., M.P.H., F.A.C.E.P.", Robert Wood Johnson Foundation National Advisory Committee
  3. ^ a b "Emory Center for Emergency Control Faculty"
  4. ^ "Curriculum Vitae, Arthur L Kellermann, M.D., M.P.H., F.A.C.E.P.", University of Texas System News 2007
  5. ^ RAND Office of Media Relations release on November 17, 2010
  6. ^ "John G. Wiegenstein Leadership Award", American College of Emergency Physicians
  7. ^ "Accidents Don't Happen", John D. Thomas, Emory University Magazine, Summer 1995
  8. ^ Buckner, H. Taylor. "Comments on Kellerman – Guns in the Home". Retrieved 14 January 2013. 
  9. ^ Kleck, Ph. D., Gary. "Point Blank: Guns and Violence in America". Retrieved 14 January 2013. 
  10. ^ a b Schaefer, Henry D. "Serious Flaws in Kellerman et al (1993) NEJM". Retrieved 14 January 2013. 
  11. ^ "Trigger Points", Bob Thompson, The Washington Post Magazine, March 29, 1998
  12. ^ "22 Times Less Safe? Anti-Gun Lobby's Favorite Spin Re-Attacks Guns In The Home", National Rifle Association of America, Institute for Legislative Action, December 11, 2001
  13. ^ Kellermann AL; Reay DT (June 12, 1986), "Protection or peril? An analysis of firearm-related deaths in the home.", New England Journal of Medicine 314 (24): 1557–1560, doi:10.1056/NEJM198606123142406, PMID 3713749 .
  14. ^ Kellermann, Arthur L. and Rivara, Frederick P. and Rushforth, Norman B. and Banton, Joyce G. and Reay, Donald T. and Francisco, Jerry T. and Locci, Ana B. and Prodzinski, Janice and Hackman, Bela B. and Somes, Grant (1993), "Gun Ownership as a Risk Factor for Homicide in the Home", New England Journal of Medicine 329 (15): 1084–1091, doi:10.1056/NEJM199310073291506, PMID 8371731 
  15. ^ The Students of Dr. Mark Ferris’s Mathematical Statistics 460. (1994). Letter to the editor. New England Journal of Medicine, 330, 366
  16. ^ Kellermann, A. L., Hackman, B. B., Rivara, F. P., Rushforth, N. B., & the University of Tennessee (Producer). (1997). Home Safety Project, 1987-1992 (ICPSR version) [Computer file]. Ann Arbor, MI: Inter-university Consortium for Political and Social Research(Distributor)
  17. ^ Kellermann, A. L., Somes, G., Rivara, F. P., Lee, R. K., & Banton, J. G. (1998). Injuries and deaths due to firearms in the home. Journal of Trauma, Injury, Infection, and Critical Care,45, 263-267.
  18. ^ Kleck, G. Can Owning a Gun Really Triple the Owner’s Chances of Being Murdered? Homicide Studies 5, 64-71