ArticlePDF Available

Which dogs bite? Case–control study of risk factors

  • Colorado Dept. Public Health


Dog bites cause an estimated 585,000 injuries resulting in the need for medical attention yearly and children are the most frequent victims. This study sought to determine dog-specific factors independently associated with a dog biting a nonhousehold member. A matched case-control design comprising 178 pairs of dogs was used. Cases were selected from dogs reported to Denver Animal Control in 1991 for a first-bite episode of a nonhousehold member in which the victim received medical treatment. Controls were neighborhood-matched dogs with no history of biting a nonhousehold member, selected by modified random-digit dialing based on the first five digits of the case dog owner's phone number. Case and control dog owners were interviewed by telephone. Children aged 12 years and younger were the victims in 51% of cases. Compared with controls, biting dogs were more likely to be German Shepherd (adjusted odds ratio (ORa) = 16.4, 95% confidence interval (CI) 3.8 to 71.4) or Chow Chow (ORa = 4.0, 95% CI 1.2 to 13.7) predominant breeds, male (ORa = 6.2, 95% CI 2.5 to 15.1), unneutered (ORa = 2.6, 95% CI 1.1 to 6.3), residing in a house with > or = 1 children (ORa = 3.5, 95% CI 1.6 to 7.5), and chained while in the yard (ORa = 2.8, 95% CI 1.0 to 8.1). Pediatricians should advise parents that failure to neuter a dog and selection of male dogs and certain breeds such as German Shepherd and Chow Chow may increase the risk of their dog biting a nonhousehold member, who often may be a child. The potential preventability of this frequent public health problem deserves further attention.
1994;93;913-917 Pediatrics
Kenneth A. Gershman, Jeffrey J. Sacks and John C. Wright
Which Dogs Bite? A Case-Control Study of Risk Factors
the World Wide Web at:
The online version of this article, along with updated information and services, is located on
Online ISSN: 1098-4275.
Copyright © 1994 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005.
American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007.
has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the
PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
PEDIATRICS Vol. 93 No. 6 June 1994 913
Which Dogs Bite? A Case-Control Study of Risk Factors
Kenneth A. Gershman, MD, MPH*; Jeffrey J. Sacks, MD, MPH; and John C. Wright, PhD
ABSTRACT. Objective. Dog bites cause an estimated
585 000 injuries resulting in the need formedical attention
yearly and children are the most frequent victims. This
study sought to determine dog-specific factors indepen-
dently associated with a dog biting a nonhousehold
Methods. A matched case-control design comprising
178 pairs of dogs was used. Cases were selected from dogs
reported to Denver Animal Control in 1991 for a first-bite
episode of a nonhousehold member in which the victim
received medical treatment. Controls were neighborhood-
matched dogs with no history of biting a nonhousehold
member, selected by modified random-digit dialing
based on the first five digits of the case dog owner’s phone
number. Case and control dog owners were interviewed
by telephone.
Results. Children aged 12 years and younger were the
victims in 51% of cases. Compared with controls, biting
dogs were more likely to be German Shepherd (adjusted
odds ratio (ORa) 16.4, 95% confidence interval (CI) 3.8
to 71.4) or Qiow Chow (OR = 4.0, 95% CI 1.2 to 13.7)
predominant breeds, male (ORa 6.2, 95% CI 2.5 to 15.1),
unneutered (ORa Z6, 95% CI Li to 6.3), residing in a
house with 1 children (ORa 35, 95% CI 1.6 to 7.5), and
chained while in the yard (ORa 2.8, 95% CI 1.0 to 8.1).
Conclusions. Pediatridans should advise parents that
failure to neuter a dog and selection of male dogs and
certain breeds such as German Shepherd and Chow Chow
may increase the risk of their dog biting a nonhousehold
member, who often may be a child. The potential pre-
ventability of this frequent public health problem de-
serves further attention. Pediatrics 199493:913-.917; dog
bite, epidemiology, risk factor.
ABBREVIATIONS. DMAS, Denver Municipal Animal Shelter; Cm,
confidence interval; OR, odds ratio.
Dog bites are an underrecognized public health
problem.1’ Every year in the United States, dog bites
cause about 20 deaths3 and an estimated 585 000 in-
juries resulting in need for medical attention or re-
stricted activity.4 Children are the most frequent
victims.2’5-9 A survey of 3238 Pennsylvania school
children determined that by, 12th grade, 46% of stu-
dents had been bitten by a dog and 17% had received
medical attention for dog bites.1#{176}Among children,
From the *,ffion of Field Epidemiology, Epidemiology Program Office,
Centers for Disease Control and Prevention, Atlanta, GA; INational Center
for Injury Prevention and Control, Centers for Disease Control and Preven-
tion, Atlanta, GA; and §Department of Psychology, Mercer University, 1400
Coleman Aye, Macon, GA 31207.
Received for publication Aug 26, 1993; accepted Nov 10, 1993.
Reprint requests to (K.A.G.) at present address: Division of Disease Control
and Environmental Epidemiology, Colorado Department of Health, 4300
Cherry Creek Dr South, Denver, CO 80222-1530.
PEDIATRICS (ISSN 0031 4005). Copyright © 1994 by the American Aced-
emy of Pediatrics.
dog bites frequently involve the face,2 resulting in se-
vere lacerations.1’ Dog bites may cause infection,’214
cause disability,’5 and incur substantial costs.15
Dog bites may be characterized according to the
dog, the victim, the dog-victim interaction, and the
environment. Dog-specific factors associated with
biting include breed,’6’9’’9 gender,6’9”7’182#{176} age,6’17
and size.2”7’#{176}These previous studies, however, have
been potentially flawed by the choice of the compari-
son group or by the lack of a comparison group. For
example, several studies used licensed or registered
dogs as the denominator for bite-rate calculations or
as a comparison group.5’6’16’18 Licensed or registered
dogs are unlikely to be representative of the entire
dog population. Additionally, none of these studies
have used multivariate analysis to assess the inde-
pendent contribution of bite-related factors while
controlling for the potentially confounding effects of
other factors.
We conducted a matched case-control study to de-
termine dog-specific factors independently associ-
ated with biting a person who was not a member of
the dog’s household. The identification of such fac-
tors, especially modifiable ones, could help reduce the
number of dog bite injuries.
Study Population
We identified biting dogs (cases) from all 1991 reports to the
Denver Municipal Animal Shelter (DMAS), the animal control
agency for Denver County. Eligible cases were dogs reported to
DMAS in 1991 for biting a nonhousehold member and whose
victim received medical treatment as indicated on the bite report.
We excluded dogs if they had bitten a nonhousehold member
before the reported bite in 1991 because owners, in response, may
have changed dog-rearing practices, discipline, and training, and
because dogs that repeatedly bite are likely to be removed from
the household. We also excluded dogs if more than one dog was
involved in the bite episode, the dog had been owned for <6
months before the reported bite, the owner was not a Denver
County resident, or the owner’s telephone number was not listed
on the bite report.
To identify control dogs (nonbiting dogs) from the same geo-
graphic area as case dogs, we used the first five digits of the phone
number of the owner of the case dog and randomized the last two
digits. We then called households until an eligible control dog was
found. We excluded dogs from being controls if they had bitten a
nonhousehold member or been acquired by the owner after July
1991 (to ensure at least 6 months ownership). For households with
multiple dogs, we randomly selected one for participation in the
We ascertained information about case and control dogs
through structured telephone interviews of the owners, conducted
by trained interviewers from the Telephone Survey Unit of the
Colorado Department of Health. Because of the need to determine
eligibility and explain the study to respondents, interviewers were
not blinded to case or control status. interviewers were aware of
the general purpose of the study but not of any specific study
questions. We collected information regarding the dog’s charac-
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
teristics (breed, sex, age, weight, neuter status), house and outdoor
environment, discipline and training, behavior, and owner’s dog-
rearing practices. We defined predominant breed as whatever
breed the owner considered the dog. If the owner specified only
one breed, we asked if the dog was purebred. If the owner men-
tioned more than one breed, we asked which breed they consid-
ered predominant. We abstracted DMAS bite reports for the age
and sex of the victim; the location, severity and circumstances of
the bite; and license and rabies vaccine status. All data were
double-keypunched to ensure accurate data entry.
Statistical Analysis
We performed umvariate analysis of the results with the use of
SAS statistical software for personal computers.21 We used the
McNemar’s test to compare categorical variables and the Wilcoxon
rank sum test to compare continuous variables. We used EGRET
statistical software for personal computers to perform multiva-
nate conditional logistic regression analysis. The initial (full)
model included meaningful variables significant at the P .05
level in univariate analysis, as well as several variables of a priori
interest that approached significance. We tested the addition to the
full model of individual effect modifiers (interaction terms) that
might be epidemiologically meaningful with a likelihood ratio
test. We used a stepwise, backward, variable-selection procedure
based on the likelihOod ratio test to determine the order and extent
of variable deletion. In addition, we restricted the final model to
matched pairs in which the bite victim was 12 years of age, the
median age of bite victims in this study.
Of the 991 dog bites reported to DMAS in 1991, we
identified 357 potentially eligible cases from bite re-
ports (representing approximately 94% of all poten-
tially eligible cases in 1991; due to filing problems at
DMAS, the other reports were unavailable). Of these,
114 (31.9%) owners were unlocatable by phone (non-
working phone number, owner had moved, no an-
swer, or owner not there on repeated attempts), 33
(9.2%) were ineligible (dog had previously bitten a
nonhousehold member, dog owned for <6 months,
owner said no bite had occurred, or owner was not a
Denver resident), 10 (2.8%) owners refused to be in-
terviewed, and for 22 (6.2%) no control was found.
This left 178 cases (50% of those identified as poten-
tially eligible; 18% of all reported dog bites to DMAS
in 1991) that we included in this report with their
matched controls.
The median age of the bite victims of case dogs was
12 years (range, 1 to 83 years); 64.7% of bite victims
were males. The anatomic locations of bites were as
follows: 62 (34.8%) upper extremities; 51 (28.7%)
lower extremities; 41 (23.0%) face, head, or neck; 15
(8.4%) trunk; and 9 (5.1%) some combination of ex-
tremities and trunk. Of the 83 bite victims 12 years
of age, 33 (40%) were bitten on the face, head, or neck.
Although not standardized, bite severity was indi-
cated on report forms for 135 (75.8%) incidents; 103
(76.3%) of these were minor bites and 32 (23.7%) were
recorded as severe. Bite report forms indicated where
the bite episode occurred for 101 (56.7%) of the inci-
dents. Of these, 51 (50.5%) took place on the sidewalk,
street, alley, or playground (no further characteriza-
tion of these locations in relation to the owner’s house
was made); 30 (29.7%) in the owner’s yard; 14(13.9%)
in the owner’s house; and 4 (4.0%) in the victim’s yard.
Data on whether bites were provoked was not sys-
tematically recorded on bite report forms.
Dogs predominantly of Chihuahua, Golden Re-
triever, Labrador Retriever, Poodle, Scottish Terrier,
and Shetland Sheepdog breeds were more common
among nonbiting than among the biting dogs (Table
1). None of the cases and only one control dog was a
Pit Bull Terrier (new ownership of Pit Bull Terriers has
been prohibited in Denver County since 1989). Dogs
predominantly of German Shepherd, Chow Chow,
Collie, and Akita breeds were substantially more fre-
quent among biting than nonbiting dogs. The total
numbers of dogs predominantly of Collie (n = 9) and
Akita (n = 5) breeds were small compared with the
total numbers of German Shepherd (n = 47) and
Chow Chow (n = 40) predominant breed dogs; there-
fore, subsequent breed analyses focus on German
Shepherds and Chow Chows.
Several dog characteristics were associated with
biting (Table 2). Biting dogs were significantly more
likely than nonbiters to be Chow Chow or German
Shepherd predominant breed, male, not neutered,
>50 pounds, and <5 years of age.
Several environmental factors were also associated
with biting (Table 2). Biting dogs were significantly
more likely to reside in homes with one or more chit-
dren 10 years of age and to be chained while in the
yard. Of the 83 dogs chained while in the yard (cases
plus controls), 44 (53%) had growled or snapped at
visitors to the house. This behavior was also reported,
however, of 116 (44%) of 263 dogs not chained while
in the yard (P = .20, test).
Among measures of discipline and training (Table
2), biting dogs were significantly less likely than non-
biting dogs to have been disciplined by a takedown
or stringup maneuver (methods sometimes used to
discipline dogs with aggression problems); however,
only a few dogs were disciplined by these methods.
Only five dogs (four cases and one control) had re-
ceived guard or attack training. No measures of ag-
gressive behaviors or obedience were significantly as-
sociated with biting (Table 2).
TABLE 1. Predominant Breed9 Distribution of 178 Biting
and 178 Nonbiting Dogs, Denver, 1991
Predominant Breed No. (%) P Values
Dogs Dogs
Akita 5 (2.8)
0 (0.0) .06
Chihuahua 2 (1.1) 6 (3.4) NS
Chow Chow 31 (17.4) 9 (5.1) <.001
Cocker Spaniel 8 (4.5) 10 (5.6) NS
Collie 8 (4.5) 1 (0.6) .04
Doberman Pinscher 6 (3.4)
5 (2.8)
German Shepherd 34 (19.1)
13 (7.3) <.01
Golden Retriever
2 (1.1) 13 (7.3) .01
Labrador Retriever
9 (5.1) 14 (7.9)
Poodle (standard) 4 (22) 14 (7.9) .03
Scottish Terrier 3 (1.7) 7 (3.9) NS
Shetland Sheepdog
2 (1.1) 6 (3.4) NS
18 (10.1) 18 (10.1)
All other breeds
46 (25.8) 62 (34.8)
Total 178 (100.0) 178 (100.0)
* Owners were asked what breed they consider their dog if more
than one breed was specified, they were asked which breed they
considered to be predominant.
t Only breeds represented by frequencies 5 in either the biting or
nonbiting group are listed.
:1:Unmatched analysis conducted with Yates corrected Chi-square
test unless otherwise noted.
§ Fisher’s exact test (two-tailed).
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
TABLE 2. Characteristics of 178 Biting and 178 Nonbiting Dogs, Denver, 1991
No/To tal No. (%)*
Odds Ratiot
(95% CI)
Biting Dogs
Nonbiting Dogs
Predominant breedt
Chow Chow 28/128 (22) 9/156 (6) 5.5(2.1-14.2)
German Shepherd 31/131 (24) 12/159 (8) 3.4 (1.6-7.6)
Male sex 136/178 (76) 91/178 (51) 3.0 (1.9-4.8)
Age <5 years
106/177 (60) 84/177 (47) 1.7 (1.1-2.7)
Weight >50 lbs 94/169 (56) 68/174 (39) 1.9 (1.2-3.0)
Not neutered
100/176 (57) 52/177 (29) 3.5 (2.2-5.7)
Not purebred
97/171 (57) 81/172 (47) 1.5 (0.9-2.2)
Got as stray 15/177 (8) 8/176 (5) 2.3 (0.9-5.9)
Got from pet store 9/177 (5) 8/176 (5) 1.1 (0.4-3.1)
1 child in houses
102/178 (57) 53/178 (30) 2.7 (1.8-4.2)
1 other dogs in house
70/178 (39) 55/178 (31)
1.4 (0.9-2.1)
>8 h/d in yard
88/173 (51) 60/162 (37) 1.6 (1.0-2.6)
Chained while in yard 55/174 (32) 28/171 (16) 2.4 (1.4-4.0)
Ever went to obedience school
21 /175 (12) 34/174 (20) 0.6 (0.3-1.0)
Ever trained at home 45/178 (25) 32/177 (18) 1.6 (0.9-2.6)
Ever guard/attack-trained
Ever disciplined by takedown/stringupli
4/174 (2)
5/173 (3)
1 /177 (1)
14/177 (8)
0.3 (0.1-0.9)
Obedience score 70/173 (40) 50/169 (30) 1.5 (1.0-2.3)
Ever nipped household member 47/178 (26) 46/177 (26) 1.0 (0.6-1.6)
Ever bit household member 19/177 (11) 18/178 (10)
1.1 (0.6-2.2)
Ever growled/snapped at visitors 90/178 (51) 74/178 (42) 1.4 (0.9-2.1)
Barks excessively at passers by
14/178 (8) 11/178 (6) 1.3 (0.6-2.8)
Owner behavior
Not licensed in past year 86/172 (50) 40/170 (24) 3.3 (2.0-5.3)
No rabies vaccine in past year
33/176 (19) 16/173 (9) 2.5 (1.3-5.0)
Registered with AKC/UKC**
34/170 (20) 31 /173 (18)
1.2 (0.7-2.0)
Female dogs
1 litter
19/40 (47) 8/86 (9) 7.0(1.2-42.3)
* Totals may vary for different variables because of missing data or for predominant breed because of mutually exclusive categories.
t Odds ratios are from matched univariate analysis. CI denotes confidence interval.
: Owners were asked what breed they considered their dog; if more than one breed was specified, they were asked which breed they
considered to be predominant. The “unexposed” or reference group of 100 biting and 147 nonbiting dogs is composed of all dogs for which
the owner did not mention Chow Chow or German Shepherd as one of the breeds; Akitas and Coffies are included. Chow Chow-German
Shepherd and German Shepherd-Chow Chow mixes are excluded from the analysis.
§ Children 10 years of age.
IIA “takedown” is defined as pinning a dog to the floor/ground on its back while holding it by the scruff of the neck. A “stringup” is
defined as lifting a dog off the ground by its chain.
I Obedience score is the sum of one point each for a dog regularly on command: sitting, staying, coming to owner, lying down, and
walking on its leash without pulling maximum score = 5 points.
** American Kennel Club/United Kennel Club.
Among cases, the owners’ report of license and vac-
one status compared with information abstracted
from the DMAS bite reports showed substantial dis-
agreement. Current licensure was confirmed by the
bite report for only 44% of case dogs which the owner
reported as licensed in the past year; for rabies vac-
cination this figure was 66%.
Nine factors remained in the multivariate condi-
tional logistic regression model (Table 3). Biting dogs
were significantly more likely than control dogs to be
German Shepherd or Chow Chow predominant
breeds, to be male, to reside in a house with one or
more children, and not to be neutered. Biting dogs
were also more likely to be chained while in the yard;
this association reached borderline significance.
When we restricted this model to those cases in which
the bite victim was a child 12 years of age, elevated
odds ratios of similar magnitude were obtained. The
variables for the Chow Chow predominant breed and
for those not neutered, however, were no longer statis-
tically significant, because wider 95% confidence inter-
vals resulted from the smaller sample size (Table 3).
This study of dog bites, we believe, is the first to use
a multivariate approach to determine dog-specific
factors independently associated with biting. Our
study has several potential limitations. We were able
to reach only half of potentially eligible biting dog
owners. Our selection of cases from reported bites to
nonhousehold members in which victims sought
medical attention is not representative of all bites. We
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
TABLE 3. Multivari
ing, Denver, 1991
ate Models of Risk Factors for Dogs Bit-
Variable All Ages Victims
s12 Years Old
AOR* 95% CIt
AOR* 95% CIt
Predominant breeds
German Shepherd
16.4 (3.8-71.4)
22.1 (2.4-207.4)
Chow Chow 4.0 (1.2-13.7) 3.7 (0.8-18.4)
6.2 (2.5-15.1) 5.3 (1.4-19.8)
1 child in house 3.5 (1.6-7.5) 6.9 (1.8-26.1)
Not neutered 2.6 (1.1-6.3)
2.3 (0.7-7.3)
Chained while in yard
2.8 (1.0-8.1) 5.4 (0.7-39.4)
No obedience school 1.9 (0.7-4.9) 1.4 (0.2-8.2)
Purebred 1.7 (0.7-4.0) 1.8 (0.5-6.9)
Weight >50 lbs 1.5 (0.7-3.1) 1.3 (0.4-4.2)
* AOR, adjusted odds ratio; adjusted for all other variables in the
t CI, confidence interval.
:1The “unexposed” or reference group of 100 biting and 147 non-
biting dogs, all dogs for which the owner did not mention Chow
Chow or German Shepherd as one of the breeds; Akitas and
Collies are included. Chow Chow-German Shepherd and German
Shepherd-Chow Chow mixes are excluded from the analysis.
restricted our study to bites of nonhousehold mem-
bers, because bites involving the owner or owner’s
family may involve different scenarios, risk factors,
and likelihood of reporting. The majority of re-
ported bites appear to occur to nonhousehold
Although we did not verify the validity of reported
bite events, we used the victims’ seeking medical at-
tention as a surrogate measure of events likely to be
real bites. To the extent that some nonbites may have
been misclassified as bites, this would have biased
odds ratios of true risk factors toward the null. If bite
victims of certain breeds such as Chow Chow or Ger-
man Shepherd are more likely than those of other
breeds to report bites or to seek medical attention,
then the associations we found between biting and
these breeds could be partly spurious. In contrast, it
is extremely unlikely that bite victims knew their at-
tacking dog’s sex, neuter status, or whether children
reside in the same house and based their decision to
report the bite and seek medical attention on this in-
formation. Thus, these latter associations appear real.
We did not verify predominant breed as stated by
the owner; however, we ascertained breed similarly
forboth cases and controls. Because of small numbers,
we were unable to assess in multivariate analysis
whether breeds other than Chow Chow and German
Shepherd (eg, Akita, Coffie, and Pit Bull Terrier) were
more likely to bite. Additionally, we did not assess the
role of the victim’s behavior in dog bite events.
Our findings are in agreement with previous stud-
ies which have indicated that male dogs9”8’#{176}and
German Shepherds”4”6”8 are overrepresented
among biting dogs. Our finding that Chow Chow
is also a high-risk breed for biting has not been
previously reported.
Canine behavioral literature has, like our study,
suggested that intact males are more aggressive than
neutered ma1es. Unlike our findings, however, the
literature suggests that unneutered female dogs
may be less likely to bite than neutered female
dogs.2 We were unable to further assess the role
of having one or more litters as an independent risk
factor for female dogs biting.
The increased risk of biting for dogs residing in
houses with one or more children has not been pre-
viously reported. This association might be explained
partly by dogs having greater opportunity to express
protective (of the home, yard, or owner), possessive
(approached while in possession of food, toys, or ob-
jects), or fear-induced (approached, reached for, or
threatened) aggression26 in the context of young play-
mates visiting with household children.
Our finding that being chained in the yard may be
a risk factor for biting is in agreement with prior stud-
ies which have demonstrated that chained dogs ac-
count for a substantial proportion of seriousz’ and fa-
talbites.3 A dog may be chained as the result of having
exhibited aggressive behavior which itself may be a
risk factor for biting, rather than chaining somehow
causing a dog to bite. One measure of aggressive be-
havior may be growling or snapping at visitors to the
house. Our results, however, showed no significant
difference in this behavior for dogs chained while in
the yard and those not chained, suggesting that chain-
ing was not likely to have been the result of aggressive
An estimated 36.5% of American households
owned a dog in 1991 for a total dog population of 53.5
miffion. Given the large numbers of canines and the
magnitude of the dog bite problem, more attention
needs to be devoted to the prevention of dog bites.
Prevention strategies have been proposed which fo-
cus on victims, dogs, and owners including: educa-
tional programs on canine behavior especially di-
reeled at children,29 laws for regulating dangerous or
vicious dogs,30’ and educational programs regarding
responsible dog ownership. The effectiveness of
these strategies has not been assessed. Improved sur-
vefflance for dog bites is needed if we are to under-
stand better how to reduce the incidence of dog bites
and evaluate prevention efforts.
Our study suggests that owners, through their se-
lection and treatment of a pet, may be able to reduce
the likelihood of owning a dog that will eventually
bite. Further study is needed to confirm our findings,
especially in other geographic areas where different
breed propensities for biting may exist. In the mean-
time, given the numbers of dog bites and the high
proportion of victims who are children, we believe
that the potential preventabffity of this public health
problem deserves further attention by pediatricians
and parents. Pediatricians currently offer anticipatory
injury prevention guidance to parents.32 We urge pe-
diatricians to also advise parents that failure to neuter
a dog and selection of male dogs and certain breeds,
such as German Shepherd and Chow Chow, may in-
crease the chances of their dog biting a nonhousehold
member, who often may be a child.
We thank Dr Eugene Pei and Paula Uoyd of the Denver Mu-
nidpal Animal Shelter for providing access to bite reports, Dr
Richard Hoffman of the Colorado Department of Health for gen-
eral support throughout the study, and Marcie-jo Kresnow and
Barbara Houston of the National Center for Injury Prevention and
Control, Centers for Disease Control for data analysis (M.K.) and
data processing (B.H.) assistance.
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
1. Lauer EA, White WC, Lauer BA. Dog bites: a neglected problem in
accident prevention. AJDC. 1982;136:202-204
2. Harris D, Imperato PJ, Oken B. Dog bites-an unrecognized epidemic.
Bull NY Acad Med. 197450:981-1000
3. Sacks II. Sattin RW, Bonzo SE. Dog bite-related fatalities from 1979
through 1988. JAMA. 1989262:1489-1492
4. Sosin DM, Sacks JJ, Sattin RW. Causes of nonfatal injuries in the United
States, 1986. Acrid Anal Prey. 199224:685-687
5. Berzon DR. The animal bite epidemic in Baltimore, Maryland: review
and update. Am I Public Health. 197868:593-595
6. Parrish HM, Clack FB, Brobst D, Mock if. Epidemiology of dog bites.
Public Health Rep. 1959;74:891-903
7. Beck AM, Loring H, Lockwood R The ecology of dog bite injury in St.
Louis, MISSOUrI. Public Health Rep. 197590:262-267
8. Moore RM, Zehmer RB, Moulthrop JI, Parker RL Surveillance of animal-
bite cases in the United States, 1971-1972. Arch Environ Health. 197732.
9. Morton C. Dog bites in Norfolk, VA. Health Seru Rep. 197388:59-65
10. Beck AM,Jones BA. Unreported dog bites in children. Public Health Rep.
11. Karlson TA. The incidence of facial injuries from dog bites. JAMA.
12. Kizer KW. Epidemiologic and clinical aspects of animal bite injuries.
JACEP. 1979;8:134-141
13. Galloway RE. Mammalian bites. J Emerg Med. 1987;6:325-331
14. Aghababian RV, Conte JE. Mammalian bite wounds. Ann Emerg Med.
15. Berzon DR. DeHoff JB. Medical costs and other aspects of dog bites in
Baltimore. Public Health Rep. 197489:377-381
16. Szpakowski NM, Bonnett BN, Martin SW. An epidemiological investi-
gation into the reported incidents of dog biting in the city of Guelph.
Can Vet J. 198%30937-942
17. Daniels TJ. A study of dog bites on the Navajo reservation. Public Health
Rep. 1986;101:50-59
18. Hanna ii, Selby LA. Characteristics of the human and pet populations
in animal bite incidents recorded at two Air Force bases. Public Health
Rep. 198196:580-584
19. Avner JR, Baker MD. Dog bites in urban children. Pdiatrics. 199188:
20. Wright JC. Reported dog bites: are owned and stray dogs different?
Anthrozoos. 1990;4:113-119
21. SASPC Release 6.03. Cary, NC: SAS Institute, 1987
22. EGRET Statistical Package. Seattle, WA: Statistics and Epidemiology
Research Corporation; 1990
23. Borchelt PL Aggressive behavior of dogs kept as companion animals:
classification and influence of sex, reproductive status and breed. AppI
Animal Ethol. 1983;10:45-61
24. Wright JC, Nesselrote MS. Classification of behavior problems in dogs:
distributions of age, breed, sex and reproductive status. App! Anim
Behav Sci. 1987;19:169-178
25. O’Farrell V, Peachey E. Behavioural effects of ovariohysterectomy on
bitches. I Small Anim Pract. 199031:595-598
26. Borchelt PL, Voith VL. Classification of animal behavior problems. Vet
Clin North Am Small Anim Pract. 1982;12:571-585
27. Wright JC. Severe attacks by dogs: characteristics of the dogs, the vic-
tims, and the attack settings. Public Health Rep. 1985;100:55.-61
28. Rowan AN. Companion animal demographics and unwanted animals
in the United States. Anthrozoos. 19925:222-225
29. Wright JC. Canine aggression toward people: bite scenarios
and prevention. Vet Clin North Am Small Anim Pract. 199121:299-314
30. Guidelines for Regulating Dangerous or Vicious Dogs. Washington, DC:
Companion Animals Section and Division of Higher Education Pro-
grams, Humane Society of the United States; 1987
31. LOCkWOOd R, Rindy K. Are “Pit Bulls” different? An analysis of the Pit
Bull Terrier controversy. Anthrovxs. 1987;1:2-8
32. TIPP: The Injury Prevention Program. Elk Grove Village, IL: American
Academy of Pediatrics, 1989
England must take the credit-or blame-for the reinvention of boxing. The
sport was a popular part of the Roman games but had vanished by the 5th century.
It returned some 1200 years later, when bare-fist prizefights began to be held in and
around London. With help from the Marquess of Queensberry, boxing spread
around the world, making money for a considerable number of boxing promoters
and a smaller number of boxers.
It is appropriate, then, that the British Medical Association should be actively
involved in examining the sport. In its latest report, The Boxing Debate, which was
issued last week, it repeats its call for a ban on boxing and asks for an independent
inquiry into its safety.
The briefest reading of the report should persuade even boxing’s proponents of
the need for an inquiry. In its appendix the report prints abstracts of recent research
on what happens to boxers after they have been battered in the ring.
For professional boxers, several studies make unpleasant reading. One using
computerised tomography found 87 per cent of boxers, in a sample of 18, showed
evidence of brain damage. Another records that 15 out of 19 young boxers register
as impaired on a battery of neuropsychological tests.
Particularly disturbing are three studies which show that changes found in the
brains of ex-boxers are immunochemically similar to those seen in Alzheimer’s
disease. That raises the possibility that even boxers who retire from the ring healthy
may pay the price in middle age with early onset of Alzheimer’s disease.
Time to throw in the towel? New Scientist. June 19, 1993:3.
Noted by J.F.L., MD
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
1994;93;913-917 Pediatrics
Kenneth A. Gershman, Jeffrey J. Sacks and John C. Wright
Which Dogs Bite? A Case-Control Study of Risk Factors
& Services
Updated Information
including high-resolution figures, can be found at:
This article has been cited by 6 HighWire-hosted articles:
Permissions & Licensing
its entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or in
Information about ordering reprints can be found online:
by guest on November 12, 2007 www.pediatrics.orgDownloaded from
... This may be because children are less predictable than adults and often misinterpret dog body language, especially warnings of imminent aggression (Meints et al., 2014). Gershman et al. (1994) also suggested this increased risk might be explained partly by dogs living with children having greater opportunities to express protective, possessive, or fear-induced aggression in the context of young play-mates visiting household children. Contrastingly, this study also found that dogs owned by families with two children were less likely to be reported as showing aggression than those living with only one child. ...
... One possible reason for this is that children with conspecific playmates, i.e. siblings, might spend less time engaging with the dog than children without siblings. A better understanding of how children interact with dogs and how this might influence aggressive behaviour will be beneficial for public safety education campaigns as children have been found to be most at risk of dog bites (Gershman et al., 1994;Rajshekar et al., 2017) and are also at higher risk than adults of dying from rabies (Tang et al., 2005). ...
This study aimed to identify the risk factors for aggressive behaviour in domestic dogs (Canis familiaris) in Mainland China. This information has never been estimated before for owned dogs in China, therefore, there has been a lack of information to help guide veterinarians and dog behaviourists when giving advice to owners of dogs considered to have such behaviour problems. In order to establish this information, questionnaires were completed in electronic and paper format by dog owners: 2575 completed questionnaires were received. The majority of owners (2215, 86 %) reported that their dogs exhibited at least one behaviour they considered a problem: the main behaviour problem categories reported by owners were barking (871, 33.8 %), lunging and biting other dogs or people (586, 22.7 %), urinating inappropriately (566, 22 %), being destructive (537, 21.9 %), and eating non-food items (354, 13.7 %). When asked specifically about aggressive (any of growling, baring teeth, snarling, snapping or biting) and biting behaviours, 1578 owners (61 %) reported their dogs regularly being aggressive to other dogs or people, and 883 (34 %) reported that their dog had bitten other dogs or people. By using binomial logistic regression analyses, significant risk and protective factors were found for aggressive (X²(24, N = 2575) = 112.613, Nagelkerke R² = .058, p < 0.001) and biting (X²(24, N = 2575) = 101.087, Nagelkerke R² = .053, p < 0.001) behaviour. Variable categories with the largest odds ratios (>2.5) for being reported for aggressive behaviour include: terriers; and little time spent with the owner. Variable categories with the largest odds ratios (>2.00) for being reported for biting include: living with children older than 10 years of age; and being acquired from ‘other types’ of sources (compared to e.g. being found as a stray, which was protective). These results are very similar to those found in other geographical areas, for example in the USA, Sweden and Taiwan. The investigation of risk factors for aggressive behaviour will help veterinarians and dog behaviourists provide more appropriate advice to owners regarding their dogs’ aggressive behaviour, which will in turn improve dog welfare and public safety in China.
... In a recent review, Urfer and Kaeberlein [36][37][38][39][40][41][42][43][44] reported that there is consistent evidence that desexing is associated with an increased risk of obesity in dogs of both genders and that sex steroids induce a decrease in caloric intake, at least in female dogs. Edney and Smith [44,45] suggested that after orchiectomy, food intake remained unchanged while energy consumption decreased, whereas Hopkins et al. [14] and Hart [24] assumed that an orchiectomy reduction in activity is related to a decline in roaming behavior. Heidenberger and Unsheim [15] also observed a post-orchiectomy decrease in activity, but they suggested that this was due to an increase in body weight, not vice versa. ...
... A series of studies published primarily between the 1970s and 1990s suggested that surgical castration of male dogs improved their behavior regarding roaming, mounting, and urine marking [14,[23][24][25][26][27][28]31,45,46], possibly due to the reduction in testosterone [29,30]. These results confirm our findings on mounting behavior that, in the experimental group, showed a significant decrease across time: nine months after gonadectomy, it was significantly lower in gonadectomized dogs than in controls, more markedly in males. ...
Full-text available
Due to the lack of unequivocal scientific evidence, gonadectomy’s effects on dogs’ behavior are still debated. Since veterinarians differ in their opinion, there may be considerable diversity in the advice received by owners. This study aimed to evaluate the effects of gonadectomy on dog behavior across time. Ninety-six dog owners (48 control dogs and 48 experimental dogs) were interviewed twice (T0 and T1, nine months later) to obtain information about their dog’s behavior. No change was found in the eating behavior or weight of dogs of both groups. Compared to T0, at T1, experimental dogs were reported to show less mounting behavior, pull on the leash, and roaming behaviors. Marking behavior did not vary across time for both groups of dogs. A tendency to reduce owner-directed aggression was observed at T1 for experimental male dogs, while no change was observed for male controls. The literature reports conflicting information about the effect of gonadectomy on behavior, suggesting that further studies about this topic should be undertaken.
... Very few hospitalized cases reported or identified breed of the dog inflicting the injury [31,[94][95][96][97][98]. Cognitive biases regarding a particular breed of dog causing serious injury may result in those recipients being more likely to attend for treatment [99]. ...
Full-text available
Objectives: To assess the scientific literature pertaining the risk factors for injuries among victims of animal bite injuries. Data and sources: A systematic review of scientific literature published until May 2020 was carried out in the following databases: PubMed, Cochrane Library, Google Scholar and Journals@ovid. Study selection: A total of 924 records were found, of which 29 articles fulfilled the inclusion criteria and were analyzed. There was a male preponderance in most of the studies with male/female ratio ranging from 0.75:1 to 2.1:1. The age range varies from 0 to 19 years with the mean age varying from 3.6 to 8 years. Pitbulls, Rottweiler's, German shepherds, Bull terriers, Labradors and Dobermans were breeds with higher risk of attack. The animals were familiar to the victim (own, friends, neighbors) in 27–98% instances. Most cases of animal bite injuries were recorded during Summer and Spring months. Head and neck followed by extremities was found to be most inflicted area. Conclusions: The sociodemographic characteristics of victim as well as the biting animal affect the circumstances leading to biting episode. However, the results should be interpreted with caution due to the high heterogeneity among studies and moderate quality evidence.
... Additionally, they urge pediatricians to advise parents that failure to neuter a dog and selecting a male dog of specific breed may increase the chances of a bite in the household. 52 Resolving the public health concern related to dog bites must also be undertaken, and there is currently no literature that has done this. It is important to gain insight on the policies and legislation that are in place in the cities and surrounding communities we live in. ...
Full-text available
Introduction Dog bites are one of the leading causes of non-fatal emergency room visits in children. These injuries not only cause physical harm but can lead to long-term psychological stress. This study evaluated the current literature related to pediatric dog bite injuries to identify research gaps which should be prioritized to improve a major public health concern. Methods We performed a keyword search of PubMed, Scopus, and OVID Medline databases (January 1980– March 2020) for all published studies focused on dog bite injuries in the pediatric population (≤18 years of age) using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Results Out of 1859 abstracts screened, 43 studies involving 86 880 patients were included. Twenty-nine studies were retrospective chart reviews characterizing the epidemiology of dog bites and their associated treatment outcomes; six were prospective cohort studies; two were cross-sectional studies; and six were experimental studies. Synthesized results demonstrate that children <9 years of age suffer the greatest burden of injuries, with children <6 years of age at higher risk of more severe injuries involving the head, neck, and face. Conclusion Studies analyzing the prevention or psychosocial consequences of dog bites injuries are needed.
... Considering the housing conditions of dogs expressing aggressive biting behaviors, living exclusively outdoors and being chained seemed to be the main environmental conditions. This supports the reports of Gershman, Sacks, and Wright (1994), who found that dogs chained in the yard is a major risk factor for biting in the USA. Our results also showed that most dogs involved in the biting incidents were known to have a history of aggressive behavior towards people and other dogs. ...
Using a web-based questionnaire with 29 close- and open-ended questions about 400 self-reported dog-biting incidents in Slovenia, this research investigated the contexts of dog bites, focusing on characteristics of the dogs and the descriptions of the situations to identify the main risk factors for the occurrence of dog bites. Even though it has been suggested that most dog bites occur during initially non-aggressive interactions with the dog (e.g., during petting, playing), in our study we discerned a wide variety of contexts, including those in which the person did not intend to interact with the dog. Most victims reported unprovoked bites during fast movements near the dog, while coming into close proximity, and during incidents without a reason. These incidents more likely occurred in public than private places and were associated with purebred dogs with a history of aggression.
... Fatal dog attacks have increased in the USA [10]. Gershman et al. [11] identified the tethering of dogs as one of the risk factors for increasing behavioral problems. However, the latter may also be due to the fact that during the time of Gershman's study, both the number of dogs and the number of people owning dogs increased. ...
Full-text available
Long-term tethering of dogs, or their keeping under unsuitable conditions can result in issues related to changes in their behavior as they may not satisfy their basic needs of life. These needs are discussed in this paper, along with cases when dogs unnecessarily have to endure cruelty and pain. The unavoidable tethering of a dog must not cause trauma and must be arranged in a way that it guarantees physical comfort. Failure to meet the basic needs of an animal may result in manifestation of fear and subsequent aggressiveness. Owners of animals are responsible for their life and health, and their obligations include eliminating the possibility of them hurting themselves or other beings. The relevant adopted legislative provisions should provide protection to animals and be enforceable, which currently appears rather difficult. Controlling and observation of the legislative provisions related to the tethering of dogs raises some difficulties for animal protection inspectors. It is necessary to focus on the specificities of keeping conditions of various dog breeds and on their individual features. Based on research and the relevant Slovak legislative provisions, this paper discusses various views on the practice of tethering dogs from the point of view of public safety and the ethical consequences of permanent dog tethering. Data on dog tethering in Slovakia were evaluated based on a survey and Slovak legal rules governing this issue were analyzed along with various views of public safety and the ethical consequences of permanent dog tethering.
... These results are concerning given the growing trend of dog ownership in Queensland. Our findings suggest that paediatric dog bite cases admitted to ED are close to 15-fold more likely to suffer bites to the head/face/neck which is consistent with previous literature indicating that paediatric patients are more likely to suffer bites to the this region of the body compared to adults [10][11][12][13]15,22,[36][37][38]. Our results also indicate that paediatric patients are also more likely to sustain dog bite injuries to the lower body although at lower rates that head/face/neck. ...
Full-text available
Dog bites are a recognized public health issue due to their impact on human and animal health/welfare. This study aimed to investigate demographic and geographic disparities in the epidemiology of dog bites presentations reported to the emergency departments of the four main public hospitals in the Metro South region of Brisbane, Queensland, Australia.Dog bite patient hospitalization data geolocated to the street address were collected from clinical records management systems from the four main public hospitals in the Metro South Hospital Health Service region of Queensland for a 5-year period (ie. 01/07/2013 to 30/06/2017). We investigated the epidemiology of three clinical outcomes including probability of paediatric cases (paediatric vs. adult), probability of dog bites to the head (head injury vs. other injury), and probability of re-presentation to the ED following their initial dog bite (yes or no) by way of univariable then multivariable Bernoulli logistic regression models including patient postcode as a random effect. Residual semivariograms were created to identify spatial trends in the medical geography of dog bites and binomial geostatistical models were created to predict the probability of the outcomes of interest in Brisbane Metro south and surrounding suburbs.Our results demonstrate that compared to adult dog bite cases, paediatric dog bite cases were significantly associated with bites to the head or face or neck (OR 14.65, P
This chapter contextualizes the dog-human relationship in the dog's origin as a scavenger on the fringes of human settlements over 15,000 years ago. It then reviews the evidence for unique evolved cognitive structures in dogs that could explain their success in a human-dominated world. Failing to find evidence of unique human-like social-cognitive capacities I then review uncontroversial facts of dogs' basic behavioral biology, including reproductive and foraging behavior and, particularly, affiliative and attachment-related behaviors. This leads to consideration of dogs' social behavior, both conspecific and toward other species, especially humans. I draw attention to a seldom-noted apparent contradiction between dogs' stronger affectional bonds toward humans than toward members of their own species. Dogs' social groups also show steeper social hierarchies accompanied by more behaviors indicating formal dominance than do other canid species including wolves. I resolve this contradiction by proposing that dogs' intense sensitivity to social hierarchy contributes to their willingness to accept human leadership. People commonly control resources that dogs need and also unknowingly express behaviors which dogs perceive as formal signs of dominance. This may be what Darwin was referring to when he endorsed the idea that a dog looks on his master as on a god. Whatever the merits of this idea, if it serves to redirect behavioral research on dogs in human society more toward the social interactions of these species in their diverse forms of symbiosis it will have served a useful function.
Technical Report
Advice of the French Food Safety Agency on the risk of dog bites and the relevance of breed specific laws made by a subgroup of the Animal Health and Welfare Committee. An evaluation of risk process : identification of the hazard, evaluation of risk i.e emission X expostion and consequences. Advice given on demand of Department of Agriculture related to Laws of 1999, 2007 and 2008 concerning dangerous dogs. Relevance of categorization of dog breeds is discussed as well as the methods of behavioural evaluation.
Full-text available
A detailed analysis of all the reported dog bites that occurred over a 2-year period in St. Louis, Mo., provided new insight not only into the severity of the problem, but also the environmental context for injury. Dog bite is a major medical problem that affects at least 1 of every 222 people and specifically 1 of every 83 children, 5 to 9 years old. Nearly 20 percent of all the children bitten were injured on the head or face, a source of concren and expense for all concerned. Nearly 10 percent of all bites were classified as serious. In only 25 percent of all injuries did the victim's behabior involve the dog at the time ofe victim interacting with the dog's owner. The victim was on the dog owner's property in about u9 percent of the incidents, and in about 48 percent of the cases the bite took pla-e near the owner's property. Bite incidents go up whenever the weather is conductive to street activity. More than 85 percent of all the biting dogs had owners. These results indicate that society's views of dog bite injury, which tend to minimize the problem and find fault with the victim, must be re-evaluated. It is time to place less emphasis on the victim and even the animal and review thae public health implications of dog ownership
Full-text available
Beginning in 1965, the number of reported dog bites began to increase; between 1965 and 1972, the annual number increased by 37%, from 27,699 to 37,896. In order to explore some of the factors associated with dog bites, a sample of 1,869 dog bites reported to the New York City Health Department between 1965 and 1970 was selected and analyzed for each year of the period under study and in aggregate. A variety of host, agent and environmental factors such as the age and sex of the victim, the anatomic site of the bite, time of day, season, geographic area where the incident occurred and size of the dog were analyzed. A tabulation of the data is presented with conclusions about the risk factors associated with dog bites. The findings also suggest the basis for designing a program to prevent such injuries which, for a variety of reasons, are becoming an increasingly common hazard of city life.
A descriptive retrospective study was undertaken to evaluate 250 records of dog bites collected by the Guelph Health Unit during 1986 and 1987. The average reported dog bit rate was 160/100,000 people per year. The lower extremities were bitten in 31.3% of cases, hands 26.5%, face 19.7%, torso 10.4%, and arms received 8.4% of bites. Owners of biting dogs were located 97.6% of the time, and 70% of these dogs were vaccinated for rabies. Incidents occurred 60.1% of the time within the dog's home territory, 14.4% on the street, 13% on a neighboring property, 6.3% in parks, 2.4% around schools, 1.4% at stores, and 2.4% others. Forty-one percent of attacks were unprovoked, but most attacks could be explained by the classifications of canine aggression. Although several breeds had high bite rates, only mixed and German Shepherd breeds had a population attributable fraction in excess of two percent.
During the 1986 National Health Interview Survey (NHIS), data on injuries resulting in a doctor visit or restricted activity for at least a half day were collected and assigned E-codes. Based on 603 injuries, the estimated number of nonfatal injuries for civilian, noninstitutionalized U.S. residents in 1986 was 60,212,000. The most frequent cause of injury was a fall (11,547,000), followed by motor vehicle traffic crashes (4,361,000) and adverse effects of drugs and biologics (3,363,000). While cause-specific detail was limited by small numbers of injuries in the sample, the NHIS can provide a valuable snapshot of the causes of nonfatal injuries.
By combining data from the National Center for Health Statistics and computerized searching of news stories, we identified 157 dog bite-related fatalities that occurred in the United States from 1979 through 1988. Of the 157 deaths, 70% occurred among children who were less than 10 years of age. The death rate for neonates was almost 370 times that of adults who were 30 to 49 years of age. Pit bull breeds were involved in 42 (41.6%) of 101 deaths where dog breed was reported, almost three times more than German shepherds, the next most commonly reported breed. The proportion of deaths attributable to pit bulls increased from 20% in 1979 and 1980 to 62% in 1987 and 1988. Pit bull attacks were almost twice as likely to be caused by strays as attacks by other breeds. Extrapolated estimates suggest 183 to 204 dog bite-related fatalities from 1979 through 1988. To prevent such deaths, we recommend stronger animal control laws, public education regarding dog bites, and more responsible dog ownership. Parents and physicians should be aware that infants left alone with a dog may be at risk of death.
Reservation-wide dog-bite statistics indicate a bite rate on the Navajo Reservation that is comparable to that of a large city. Detailed analysis of 772 bite reports was made to determine the characteristics of biters and their victims. This included an assessment of the behavioral antecedents leading up to the bite incident; 98.4 percent of all cases for which a possible cause could be ascertained were provoked in some way. Both dog control and public education measures need to be taken to reduce the frequency of dog bites.
In 1981, more than 3,200 Pennsylvania children, ages 4 to 18 years, were surveyed about their dog bite histories and attitudes toward animals. Dog bites were much more common than previously reported: 45 percent of children had been bitten during their lifetimes, and 15.5 percent had been bitten in 1980, more than 36 times the rate reported to health authorities. In 1980, the highest bite rate occurred among children 7-12 years old (20 percent). Children were bitten more frequently by the dogs owned by their neighbors, followed by their own dogs, than by strays or by dogs whose owners were not known. Boys were bitten twice as frequently as girls by neighbors' dogs and strays; the bite rates from family dogs were identical in boys and girls. Despite the high bite rates, being bitten was not significantly associated, in most groups of children studied, with a dislike of dogs. These positive attitudes toward dogs may lead to inadequate precautions against bites and to biases in the reporting of bites to health authorities.
The public health hazard of dog bites has grown to epidemic preparations. It has also created a significant financial burden in terms of medical costs and other losses for those injured. Of 214 responses (45%) to 480 questionnaires, 145 bite victims reported medical costs averaging $38.50; the highest medical bills were $756, $721, and $350. Sixty three persons (30%) were covered by medical assistance or insurance plans and had no information on their medical costs. For 45 persons reporting medical costs and at least one other type of expense, the average total cost was $49. The average hospital charge reported by 19 of 20 hospitals surveyed was $32. One in five patients was bitten by the family dog, two in five by a neighborhood dog, and two in five by dogs belonging to strangers or persons unknown. Fewer than 10% of those bitten by dogs not belonging to them were reimbursed by the dog owner for the expense incurred. Two hunderd and nine patients visited a medical facility; 80 reported disability days; the longest period was more than 3 mth. Children reported up to 14 days absence from school, and adults lost up to 14 days from work. The hospitals reported that more than half of the bites were on the leg, more than one in four on the hands and arms, and one in eight on the face, neck, and head.
• Dog bites are a common but neglected pediatric problem. To clarify the epidemiology of dog bites and to learn if parents would welcome counseling aimed at preventing bites, 455 families (960 children) in a Denver pediatric practice were surveyed. One hundred ninety-four children (20.2%) had been bitten at least once, with the majority of bites occurring before the child was aged 5 years. Forty-three percent of the bites prompted a visit to a physician and 16.5% received sutures. German shepherds were responsible for 17% of the incidents, more than expected relative to their popularity as pets. The dogs usually were owned by a neighbor (40.2%) or the victim's family (31%). Approximately half of the bites were believed to be unprovoked. Seventy-seven percent of the parents believed that dog bite prevention warranted discussion with their physician. Dog bites are an important pediatric problem, and parents should be counseled accordingly during well-child visits. (Am J Dis Child 1982;136:202-204)