Breeds of dogs involved in fatal human attacks

/Home/Bites/Information about Dog and Breeds

836 Vet Med Today: Special Report JAVMA, Vol 217, No. 6, September 15, 2000
Special Report
From 1979 through 1996, dog attacks resulted in
more than 300 human dog bite-related fatalities
(DBRF) in the United States.1-3 Most victims were children.
Studies indicate that pit bull-type dogs were
involved in approximately a third of human DBRF
reported during the 12-year period from 1981 through
1992, and Rottweilers were responsible for about half
of human DBRF reported during the 4 years from 1993
through 1996. These data have caused some individuals
to infer that certain breeds of dogs are more likely
to bite than others and should, therefore, be banned or
regulated more stringently.4,5 The purposes of the study
reported here were to summarize breeds associated
with reported human DBRF during a 20-year period
and assess policy implications.
Procedure
We collected data from The Humane Society of the
United States (HSUS) and media accounts related to
dog bite attacks and fatalities, using methods from previous
studies.1-3 The HSUS maintains a registry of human
DBRF, including date of death, age and sex of decedent,
city and state of attack, number and breeds of dogs
involved, and circumstances relating to the attack. To
supplement HSUS reports, as in the past, a database6 was
searched for accounts of human DBRF that occurred in
1997 and 1998. Our search strategy involved scanning
the text of newspapers and periodicals for certain words
and word combinations likely to represent human DBRF
followed by a review of articles containing those terms.
Data obtained from HSUS and news accounts were
merged to maximize detection of human DBRF and
avoid duplicate reports. One new human DBRF from
1996 was identified in the 1997 and 1998 reports and
was added to the existing data for 1996.
A human DBRF was defined as a human death
caused by trauma from a dog bite. In addition to
excluding 9 human deaths, as described in previous
reports (eg, dying of rabies from a dog bite, strangling
on a leash or scarf pulled by a dog, dying from fire ant
From the Division of Unintentional Injury Prevention, National Center for Injury Prevention and Control, US Department of Health and
Human Services, US Public Health Service, Centers for Disease Control and Prevention, 4770 Buford Hwy NE (MS K-63), Atlanta, GA 30341
(Sacks, Gilchrist); The Humane Society of the United States, 2100 L Street, NW, Washington, DC 20037 (Sinclair, Lockwood); and the
Division of Education and Research, American Veterinary Medical Association, 1931 N Meacham Rd, Ste 100, Schaumburg, IL 60173
(Golab). Dr. Sacks’ present address is the National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control
and Prevention, 4770 Buford Hwy NE (MS K-45), Atlanta, GA 30341. Dr. Sinclair’s present address is Shelter Veterinary Services, 9320 Jarrett
Ct, Montgomery Village, MD 20886.
Use of trade names and commercial sources is for identification purposes only and does not imply endorsement by the authors or their affiliated
agencies.
The authors thank Dr. Suzanne Binder for technical assistance.
Embargoed for Release Until 8 AM, September 15, 2000
Breeds of dogs involved in fatal human attacks
in the United States between 1979 and 1998
Jeffrey J. Sacks, MD, MPH; Leslie Sinclair, DVM; Julie Gilchrist, MD;
Gail C. Golab, PhD, DVM; Randall Lockwood, PhD
Objective—To summarize breeds of dogs involved in
fatal human attacks during a 20-year period and to
assess policy implications.
Animals—Dogs for which breed was reported involved
in attacks on humans between 1979 and 1998 that
resulted in human dog bite-related fatalities (DBRF).
Procedure—Data for human DBRF identified previously
for the period of 1979 through 1996 were combined
with human DBRF newly identified for 1997
and 1998. Human DBRF were identified by searching
news accounts and by use of The Humane Society of
the United States’ registry databank.
Results—During 1997 and 1998, at least 27 people
died of dog bite attacks (18 in 1997 and 9 in 1998). At
least 25 breeds of dogs have been involved in 238
human DBRF during the past 20 years. Pit bull-type
dogs and Rottweilers were involved in more than half of
these deaths. Of 227 reports with relevant data, 55
(24%) human deaths involved unrestrained dogs off
their owners’ property, 133 (58%) involved unrestrained
dogs on their owners’ property, 38 (17%) involved
restrained dogs on their owners’ property, and 1 (< 1%)
involved a restrained dog off its owner’s property.
Conclusions—Although fatal attacks on humans
appear to be a breed-specific problem (pit bull-type
dogs and Rottweilers), other breeds may bite and
cause fatalities at higher rates. Because of difficulties
inherent in determining a dog’s breed with certainty,
enforcement of breed-specific ordinances raises constitutional
and practical issues. Fatal attacks represent
a small proportion of dog bite injuries to humans and,
therefore, should not be the primary factor driving
public policy concerning dangerous dogs. Many practical
alternatives to breed-specific ordinances exist and
hold promise for prevention of dog bites. (J Am Vet
Med Assoc 2000;217:836–840)
JAVMA, Vol 217, No. 6, September 15, 2000 Vet Med Today: Special Report 837
bites after being pushed on a mound by a dog, or dying
from a motor vehicle or bicycle crash while being
chased by a dog), for 1997 and 1998, we excluded 3
additional deaths: death resulting from infection secondary
to a dog bite, death attributable to trauma from
being knocked over but not bitten, and death resulting
from myocardial infarction, which was caused by an
individual being chased but not bitten. For the 20-year
study, we excluded 4 human deaths from attacks by
guard or police dogs “at work” and approximately 90
deaths when breed information for the attacking dog
was unavailable; thus, this study included approximately
72% of cases of human DBRF and is not
exhaustive.
We tallied data in 2 ways to provide alternatives
for breed data interpretation. First, we used a human
death-based approach in which we counted whether a
particular breed was involved in a death. When multiple
dogs of the same breed were involved in the same
fatal episode, that breed was counted only once (eg, if
10 Akitas attacked and killed a person, that breed was
counted once rather than 10 times). When crossbred
dogs were involved in a fatality, each suspected breed
in the dog’s lineage was counted once for that episode.
Second, we tallied data by dog. When multiple dogs of
the same breed were involved in a single incident, each
dog was counted individually. We allocated crossbred
dogs into separate breeds and counted them similarly
(eg, if 3 Great Dane-Rottweiler crossbreeds attacked a
person, Great Dane was counted 3 times under crossbred,
and Rottweiler was counted 3 times under crossbred).
Data are presented separately for dogs identified
as pure- and crossbred. Lastly, dogs were classified as to
whether they were on or off the owners’ property and
restrained (eg, chained or leashed) or unrestrained at
the time of the attack.
Results
Fatalities during 1997 and 1998—During 1997
and 1998, at least 27 people died as the result of dog
bite attacks (18 people in 1997 and 9 in 1998). Of 27
human DBRF, 19 (70%) were children (1 was &#8804; 30 days
old, 3 were between 7 and 11 months old, 9 were
between 1 and 4 years old, and 6 were between 5 and
11 years old), and 8 were adults (ages 17, 44, 64, 70,
73, 75, 75, and 87). Approximately half (n = 15 [56%])
of the human DBRF were male.
Five (19%) deaths involved unrestrained dogs off
the owners’ property, 18 (67%) involved unrestrained
dogs on the owners’ property, 3 (11%) involved
restrained dogs on the owners’ property, and 1 (4%)
involved a restrained dog off the owner’s property.
Eighteen (67%) deaths involved 1 dog, 5 (19%)
involved 2 dogs, and 4 (15%) involved 3 dogs. Sixty
percent of attacks by unrestrained dogs off the owners’
property involved more than 1 dog.
Fatal attacks were reported from 17 states
(California [4 deaths]; Georgia and North Carolina [3
each]; Kansas, Texas, and Wisconsin [2 each]; and
Alaska, Arkansas, Colorado, Florida, Kentucky,
Massachusetts, Michigan, Missouri, New York, South
Dakota, and Tennessee [1 each]).
Some breed information was reported for all 27
attacks. As in recent years, Rottweilers were the most
commonly reported breed involved in fatal attacks, followed
by pit bull-type dogs (Table 1). Together, these
2 breeds were involved in approximately 60% of
human deaths.
Twenty-year data—Some breed information was
available for 238 human DBRF. More than 25 breeds of
dogs were involved in DBRF during the past 20 years
(Table 2). Of 227 human DBRF for which data were
1979– 1981– 1983– 1985– 1987– 1989– 1991– 1993– 1995– 1997–
Breed 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Total
Purebred
Pit bull-type 2 5 10 9 11* 8 6 5 4* 6 66
Rottweiler 0 0 1 1 3 1 3 10 10 10 39
German Shepherd Dog 2 1 4* 1 1 4* 2 0 2 0 17
Husky-type 2 1 2 2 0 2 2 1 2 1 15
Malamute 2 0 3 1 0 2 3 1 0 0 12
Doberman Pinscher 0 1 0 2 2 2 1 0 0 1 9
Chow Chow 0 1 0 0 0 2 3 0 2 0 8
Great Dane 3 1 0 0 0 0 0 1 1 1 7
Saint Bernard 1 2 1 0 0 0 0 0 0 3 7
Crossbred
Wolf-dog hybrid 0 1 1 2 1 4 1 2 2 0 14
Mixed-breed 0 3 1 2 1 2 0 1 1 1 12
German Shepherd Dog 0 2 0 2 2 2† 0 1 2 0 10†
Pit bull-type 0 1 0 3 2† 3 1 1 0 0 10†
Husky-type 0 1 1 2 1 1 0 0 0 0 6
Rottweiler 0 0 0 0 1† 1 0 1 1 2 5†
Alaskan Malamute 0 0 0 0 0 2 1 0 0 0 3
Chow Chow 0 0 0 0 0 1 0 1 1 0 3
Doberman Pinscher 0 0 0 0 1 0 0 0 0 1 0
Saint Bernard 0 0 0 0 0 0 1 0 0 0 1
Great Dane 0 0 0 0 0 0 0 0 0 1† 0†
No. deaths for which 10 20 26* 24 22 34* 24 25 26* 27 238
breed was known
*Numbers differ from previous reports because police/guard dogs "at work" were excluded, and 1 new DBRF was identified as occurring in 1996. †A purebred dog
and a crossbred dog of this breed were involved in a single fatality; therefore, that breed is counted only once in the total column.
Table 1—Breeds of dogs involved in human dog bite-related fatalities (DBRF) in the United States, by 2-year period, between 1979 and
1998. Death-based approach of counting most frequent purebreds and crossbreds involved in 7 or more human DBRF
838 Vet Med Today: Special Report JAVMA, Vol 217, No. 6, September 15, 2000
available, 55 (24%) deaths involved unrestrained dogs
off the owners’ property, 133 (58%) involved unrestrained
dogs on the owners’ property, 38 (17%)
involved restrained dogs on the owners’ property, and
1 (< 1%) involved a restrained dog off the owner’s
property.
Four hundred three dogs were responsible for
these attacks. There were almost twice as many dogs
involved in off-owner-property attacks, compared with
attacks occurring on the owners’ properties. In 160
human deaths, only 1 dog was involved; in 49 deaths,
2 dogs were involved; and in 15 deaths, 3 dogs were
involved. Four and 7 dogs were involved in 3 deaths
each; 5, 6, and 10 dogs were involved in 2 deaths each;
and 11 and 14 dogs were responsible for 1 death each.
Discussion
Ideally, breed-specific bite rates would be calculated
to compare breeds and quantify the relative dangerousness
of each breed. For example, 10 fatal attacks by
Breed X relative to a population of 10,000 X’s (1/1,000)
implies a greater risk than 100 attacks by Breed Y relative
to a population of 1,000,000 Y’s (0.1/1,000).
Without consideration of the population sizes, Breed Y
would be perceived to be the more dangerous breed on
the basis of the number of fatalities.
Considering only bites that resulted in fatalities,
because they are more easily ascertained than nonfatal
bites, the numerator of a dog breed-specific human
DBRF rate requires a complete accounting of human
DBRF as well as an accurate determination of the
breeds involved. Numerator data may be biased for 4
reasons. First, the human DBRF reported here are likely
underestimated; prior work suggests the approach
we used identifies only 74% of actual cases.1,2 Second,
to the extent that attacks by 1 breed are more newsworthy
than those by other breeds, our methods may
have resulted in differential ascertainment of fatalities
by breed. Third, because identification of a dog’s breed
may be subjective (even experts may disagree on the
breed of a particular dog), DBRF may be differentially
ascribed to breeds with a reputation for aggression.
Fourth, it is not clear how to count attacks by crossbred
dogs. Ignoring these data underestimates breed
involvement (29% of attacking dogs were crossbred
dogs), whereas including them permits a single dog to
be counted more than once. Therefore, we have elected
to present data separately for purebred and crossbred
dogs to demonstrate at least 2 alternative counting
methods. Relative rankings do not differ greatly
whether one focuses only on purebred dogs or includes
crossbred dogs. The crossbreed issue is also problematic
when estimating denominators (ie, breed-specific
population sizes).
The denominator of a dog breed-specific human
DBRF rate requires reliable breed-specific population
data. Unfortunately, such data are not currently available.
Considering American Kennel Club registration
data7 for Rottweilers in parallel with fatality data for
that breed indicates that as the breed has soared in pop-
Death-based approach Dog-based approach
Breed Purebred Crossbred Total Purebred Crossbred Total
Pit bull-type 66 11* 76* 98 20 118
Rottweiler 39 6* 44* 60 7 67
German Shepherd Dog 17 11* 27* 24 17 41
Husky-type (includes at least 2 Siberian) 15 6 21 15 6 21
Malamute 12 3 15 13 3 16
Wolf-dog hybrid 0 14 14 0 15 15
Mixed-breed (NOS) 0 12 12 0 47 47
Chow Chow 8 3 11 8 13 21
Doberman 9 1 10 12 1 13
Saint Bernard 7 1 8 7 1 8
Great Dane 7 1* 7* 11 2 13
Labrador Retriever 1 4 5 1 7 8
Akita 4 0 4 4 0 4
Sled-type (NOS) 3 0 3 12 0 12
Bulldog 2 1 3 2 1 3
Mastiff 2 1 3 4 1 5
Boxer 2 1 3 4 1 5
Collie 0 3 3 0 6 6
Bullmastiff 1 1 2 1 1 2
Hound-type (NOS) 1 1 2 1 1 2
Retriever-type (NOS) 1 0 1 1 0 1
Chesapeake Bay Retriever 1 0 1 1 0 1
West Highland Terrier (NOS) 1 0 1 1 0 1
Terrier-type (NOS) 1 0 1 1 0 1
Japanese Hunting Dog (NOS) 1 0 1 1 0 1
Newfoundland 1 0 1 1 0 1
Coonhound 1 0 1 1 0 1
Sheepdog (NOS) 1 0 1 1 0 1
Australian Shepherd 0 1 1 0 3 3
Rhodesian Ridgeback 1 0 1 1 0 1
Cocker Spaniel 1 0 1 1 0 1
*A purebred dog and a crossbred dog of this breed were involved in a single fatality; therefore, that breed is counted only
once in the total column.
NOS
Not otherwise specified.
Table 2—Breeds of dogs involved in human dog bite-related fatalities between 1979 and 1998, using
death-based and dog-based approaches
JAVMA, Vol 217, No. 6, September 15, 2000 Vet Med Today: Special Report 839
ularity, so have Rottweiler-related deaths (24,195 registrations
from 1979 through 1982 and 0 deaths; 272,273
registrations from 1983 through 1990 and 6 deaths; and
692,799 registrations from 1991 through 1998 and 33
deaths). However, official registration or licensing data
are likely to be biased, as owners of certain dog breeds
may be less likely than those owning other breeds to
register or license their dogs4 and, thus, should not be
used to calculate these rates. Finally, it is imperative to
keep in mind that even if breed-specific bite rates could
be accurately calculated, they do not factor in ownerrelated
issues. For example, less responsible owners or
owners who want to foster aggression in their dogs may
be drawn differentially to certain breeds.
Despite these limitations and concerns, the data
indicate that Rottweilers and pit bull-type dogs
accounted for 67% of human DBRF in the United States
between 1997 and 1998. It is extremely unlikely that
they accounted for anywhere near 60% of dogs in the
United States during that same period and, thus, there
appears to be a breed-specific problem with fatalities.
Although the fatality data are concerning, one must
broaden the context to consider both fatal and nonfatal
bites when deciding on a course of action. Nonfatal dog
bites continue to be a public health problem in the
United States. Although this and prior reports1-3 document
more than 330 DBRF during a 20-year period,
these tragedies represent only the most severe manifestation
of the problem. In 1986, nonfatal dog bites resulted
in an estimated 585,000 injuries that required medical
attention or restricted activity.8 By 1994, an estimated
4.7 million people (1.8% of the US population) sustained
a dog bite; of these, approximately 800,000 (0.3%
of the US population) sought medical care for the bite
(332,000 in emergency departments), and 6,000 were
hospitalized.9-11 This 36% increase in medically attended
bites from 1986 to 1994 draws attention to the need for
an effective response, including dog bite prevention programs.
Because (1) fatal bites constitute less than
0.00001% of all dog bites annually, (2) fatal bites have
remained relatively constant over time, whereas nonfatal
bites have been increasing, and (3) fatal bites are rare at
the usual political level where bite regulations are promulgated
and enforced, we believe that fatal bites should
not be the primary factor driving public policy regarding
dog bite prevention.
Several interacting factors affect a dog’s propensity
to bite, including heredity, sex, early experience,
socialization and training, health (medical and behavioral),
reproductive status, quality of ownership and
supervision, and victim behavior. For example, a study
in Denver of medically-attended dog bites in 1991 suggested
that male dogs are 6.2 times more likely to bite
than female dogs, sexually intact dogs are 2.6 times
more likely to bite than neutered dogs, and chained
dogs are 2.8 times more likely to bite than unchained
dogs.12 Communities have tried to address the dog bite
problem by focusing on different factors related to biting
behavior.
To decrease the risk of dog bites, several communities
have enacted breed-specific restrictions or bans. In
general, these have focused on pit bull-type dogs and
Rottweilers. However, breeds responsible for human
DBRF have varied over time. Pinckney and Kennedy13
studied human DBRF from May 1975 through April
1980 and listed the following breeds as responsible for
the indicated number of deaths: German Shepherd Dog
(n = 16); Husky-type dog (9); Saint Bernard (8); Bull
Terrier (6); Great Dane (6); Malamute (5); Golden
Retriever (3); Boxer (2); Dachshund (2); Doberman
Pinscher (2); Collie (2); Rottweiler (1); Basenji (1);
Chow Chow (1); Labrador Retriever (1); Yorkshire
Terrier (1); and mixed and unknown breed (15). As
ascertained from our data, between 1979 and 1980,
Great Danes caused the most reported human DBRF;
between 1997 and 1998, Rottweilers and pit bull-type
dogs were responsible for about 60% of human DBRF.
Indeed, since 1975, dogs belonging to more than 30
breeds have been responsible for fatal attacks on people,
including Dachshunds, a Yorkshire Terrier, and a
Labrador Retriever.
In addition to issues surrounding which breeds to
regulate, breed-specific ordinances raise several practical
issues. For optimal enforcement, there would need
to be an objective method of determining the breed of
a particular dog. Pedigree analysis (a potentially timeconsuming
and complicated effort) combined with
DNA testing (also time-consuming and expensive) is
the closest to an objective standard for conclusively
identifying a dog’s breed. Owners of mixed-breed or
unregistered (ie, by a kennel club) dogs have no way of
knowing whether their dog is one of the types identified
and whether they are required to comply with
breed-specific ordinances. Thus, law enforcement personnel
have few means for positively determining a
dog’s breed and deciding whether owners are in compliance
or violation of laws.
Some municipalities have attempted to address
this classification issue of unregistered and mixedbreed
dogs by including within their ordinances a
description of the breed at which the ordinance is
directed. Unfortunately, such descriptions are usually
vague, rely on subjective visual observation, and result
in many more dogs than those of the specified breed
being subject to the restrictions of the ordinance.
When a specific breed of dog has been selected for
stringent control, 2 constitutional questions concerning
dog owners’ fourteenth amendment rights have been
raised: first, because all types of dogs may inflict injury
to people and property, ordinances addressing only 1
breed of dog are argued to be underinclusive and, therefore,
violate owners’ equal protection rights; and second,
because identification of a dog’s breed with the certainty
necessary to impose sanctions on the dog’s owner is prohibitively
difficult, such ordinances have been argued as
unconstitutionally vague, and, therefore, violate due
process. Despite such concerns, a number of breed-specific
ordinances have been upheld by the courts.14-16
Another concern is that a ban on a specific breed
might cause people who want a dangerous dog to simply
turn to another breed for the same qualities they
sought in the original dog (eg, large size, aggression
easily fostered). Breed-specific legislation does not
address the fact that a dog of any breed can become
dangerous when bred or trained to be aggressive. From
a scientific point of view, we are unaware of any formal
840 Vet Med Today: Special Report JAVMA, Vol 217, No. 6, September 15, 2000
evaluation of the effectiveness of breed-specific legislation
in preventing fatal or nonfatal dog bites.
An alternative to breed-specific legislation is to regulate
individual dogs and owners on the basis of their
behavior. Although, it is not systematically reported, our
reading of the fatal bite reports indicates that problem
behaviors (of dogs and owners) have preceded attacks in
a great many cases and should be sufficient evidence for
preemptive action. Approaches to decreasing dangerous
dog and owner behaviors are numerous. The potential
importance of strong animal control programs is illustrated
by our data; from 1979 through 1998, 24% of
human DBRF were caused by owned dogs (typically
more than 1) that were roaming off the owners’ property.
Some deaths might have been averted through more
stringent animal control laws and enforcement (eg, leash
laws, fencing requirements). Although the bite prevention
effectiveness of such animal control ordinances and
programs has not been systematically evaluated, freeroaming
dogs and dogs with menacing behavior are
problems that need to be addressed even if they do not
bite (eg, causing bicycle or car crashes).
Generic non–breed-specific, dangerous dog laws
can be enacted that place primary responsibility for a
dog’s behavior on the owner, regardless of the dog’s
breed.17 In particular, targeting chronically irresponsible
dog owners may be effective.18 If dog owners are
required to assume legal liability for the behavior and
actions of their pets, they may be encouraged to seek
professional help in training and socializing their pets.
Other options include enforcing leash laws and laws
against dog fighting. We noticed in the fatal cases, that
less than one half of 1% of DBRF were caused by
leashed animals off the owners’ property. Subdivisions
and municipalities that outlaw fences or limit fences to
heights insufficient for controlling large dogs may be
increasing the probability of children interacting with
unsupervised dogs. Scientific evaluations of the effects
of such regulations are important.
Education of dog owners can address several issues:
(1) understanding breed profiles19,20 may assist owners in
selecting the appropriate dog for their lifestyle and training
abilities, (2) convincing owners to seriously consider
the sex and reproductive status of their dogs is important
because male and sexually intact dogs are more likely
to bite than are female and neutered dogs,12 and (3)
teaching owners about the importance of socialization
and training may decrease their likelihood of owning a
dog that will eventually bite.
Veterinarians play a key role in educating pet owners,
but because many dogs that bite may not be seen
by a veterinarian prior to the bite incident, programs
that encourage responsible ownership must also be
presented through other venues. Public education
strategies should include school-based and adult educational
programs addressing bite prevention and basic
canine behavior, care, and management. Programs
should strive to ensure that dogs receive proper socialization,
exercise, and attention; that they are given adequate
food, water, shelter, and veterinary care; that
they are neutered if they are not maintained for legitimate
and responsible breeding purposes; and that they
are trained humanely and confined safely. However,
like breed-specific legislation, all these approaches
appear formally unevaluated for effectiveness.
Targeting and evaluation of prevention efforts
requires improved surveillance for fatal and nonfatal
dog bites. Dog bites should be reported as required by
local or state ordinances, and reports of such incidents
should include information about the circumstances of
the bite, ownership, breed, sex, reproductive status of
the dog, history of prior aggression, and the nature of
restraint prior to the bite incident. Collection of data
on the entire dog population (eg, breed, age, sex)
would help resolve comparative risk issues and may be
accomplished by combining paperwork on mandatory
rabies immunizations with registration of breed and
sex. Only with numerator and denominator data and
with formal evaluations of the impacts of strategies
tried by various communities will we be able to make
science-based recommendations for decreasing the
number of dog bites. In the interim, adequate funding
for animal control agencies, enforcement of existing
animal control laws, and educational and policy strategies
to reduce inappropriate dog and owner behaviors
will likely result in benefits to communities and may
well decrease the number of dog bites that occur.
References
1. Sacks JJ, Sattin RW, Bonzo SE. Dog bite-related fatalities in
the United States, 1979–1988. JAMA 1989;262:1489–1492.
2. Sacks JJ, Lockwood R, Hornreich J, et al. Fatal dog attacks,
1989–1994. Pediatrics 1996;97:891–895.
3. Centers for Disease Control. Dog bite related fatalities—
United States, 1995–1996. Morbid Mortal Weekly Rep 1997;46:463–467.
4. Lockwood R. Humane concerns about dangerous dog laws.
University of Dayton Law Rev 1988;13:267–277.
5. Lockwood R, Rindy K. Are “pit bulls” different? An analysis
of the pit bull terrier controversy. Anthrozoos 1987;1:2–8.
6. NEXIS-LEXIS [online database available at http://www.lexisnexis.
com/lncc]. Dayton, Ohio: Lexis-Nexis Group; 1999.
7. American Kennel Club. Dog registration statistics, Jan 1,
1990–Dec 31, 1998. New York: American Kennel Club.
8. Sosin DM, Sacks JJ, Sattin RW. Causes of nonfatal injuries in
the United States, 1986. Accid Anal Prev 1992;24:685–687.
9. Sacks JJ, Kresnow M, Houston B. Dog bites: how big a problem?
Injury Prev 1996;2:52–54.
10. Weiss HB, Friedman D, Coben JH. Incidence of dog bite
injuries treated in emergency departments. JAMA 1998;279:51–53.
11. Quinlan KP, Sacks JJ. Hospitalizations for dog bite injuries.
JAMA 1999:281:232–233.
12. Gershman KA, Sacks JJ, Wright JC. Which dogs bite? A
case-control study of risk factors. Pediatrics 1994;93:913–917.
13. Pinckney LE, Kennedy LA. Traumatic deaths from dog
attacks in the United States. Pediatrics. 1982;69l:193–196.
14. Pollock S. Banning pit bulls in the District of Columbia.
Memorandum to the Washington Humane Society from the Arnold
and Porter Law Firm. Washington, DC, 1999.
15. Burt MR. Canine legislation: can dogs get a fair shake in
court? J Am Vet Med Assoc 1997;210:1139–1142.
16. Wapner M, Wilson JF. Are laws prohibiting ownership of pit bulltype
dogs legally enforceable? J Am Vet Med Assoc 2000;216:1552–1554.
17. Companion animals section and division of higher education
programs. Guidelines for regulating dangerous or vicious dogs.
Washington, DC: Humane Society of the United States, 1987.
18. Lockwood R. Dangerous dogs revisited. The Humane Society
News 1992;37:20–22.
19. Hart BL, Miller MF. Behavioral profiles of dog breeds. J Am
Vet Med Assoc 1985;186:1175–1180.
20. Hart BL, Hart LA. Selecting pet dogs on the basis of cluster
analysis of breed behavior profiles and gender. J Am Vet Med Assoc
1985;186:1181–1185.