Content uploaded by Nicholas Dodman
Author content
All content in this area was uploaded by Nicholas Dodman on Oct 17, 2014
Content may be subject to copyright.
JAVMA, Vol 238, No. 6, March 15, 2011 Scientific Reports 731
SMALL ANIMALS
Interdog household aggression (sometimes described
as sibling rivalry, whether or not the dogs involved are
genetically related) is a distressing problem for dog own-
ers and a common reason that owners seek behavioral
advice or veterinary medical treatment. It is not uncom-
mon for 1 or both dogs to require treatment for injuries
or for owners to need medical attention for injuries re-
ceived while interceding in a dog fight. Interdog house-
hold aggression tends to result in more severe injuries
than do fights between dogs from separate households.1
The authors of several publications have examined
the issue of interdog household aggression.1–8 The most
thorough description of the problem with treatment and
outcome assessment was published in 1996 by Sherman
et al.1 The authors reported that typically 1 dog insti-
gates fights and that often this is the younger of the 2
Interdog household aggression:
38 cases (2006–2007)
Kathryn M. Wrubel, phd; Alice A. Moon-Fanelli, phd;
Louise S. Maranda, dvm, phd; Nicholas H. Dodman, bvms, dacvb
Objective—To analyze factors associated with interdog household aggression and deter-
mine treatment outcomes.
Design—Retrospective case series and survey.
Animals—38 pairs of dogs with interdog household aggression. Each pair of dogs was
considered 1 case.
Procedures—Records of dogs with interdog household aggression that were examined
during initial or follow-up consultations at a veterinary teaching hospital from December 5,
2006, to December 5, 2007, were analyzed for clinical features. Data regarding outcome,
owner compliance, and efficacy of recommended treatments obtained by use of a follow-
up survey were evaluated.
Results—Most cases (30/38 [79%]) of interdog household aggression involved same-sex
pairs; 26 of 38 (68%) cases involved 1 female or a pair of females. Instigators and recipients
of aggression were clearly identified in 27 of 38 (71%) cases; most instigators were the
younger of the pair (20/27 [74%]) or were newer additions to the household (19/27 [70%]).
Fight-eliciting triggers included owner attention, food, excitement, and found items. Some
dogs had risk factors for behavior problems such as a history of living in multiple households
(21/51 [41%]), adoption after 12 weeks of age (20/51 [39%]), or being acquired from a shel-
ter (17/51 [33%]). Effective treatment recommendations included implementing a so-called
nothing-in-life-is-free program, giving 1 dog priority access to resources, and administering
psychotropic medication. Frequency and severity of fighting were significantly reduced af-
ter consultation. Owners reported a 69% overall improvement following treatment.
Conclusions and Clinical Relevance—Most treatment strategies were considered effec-
tive. Consistency and predictability of social interactions are essential in resolving interdog
household aggression. (J Am Vet Med Assoc 2011;238:731–740)
dogs or is a newer addition to the household. The study
also found that fights most often occur between dogs of
the same sex (usually pairs of females). Fight-eliciting
triggers include physical resources, owner proximity or
attention, and access to confined spaces; another com-
mon trigger for fights is excitement, which may occur
during greetings, play sessions, walks, or automobile
rides.1 In addition to actual fights, typical behaviors
observed in cases of interdog household aggression in-
clude mounting, blocking, standing over the other dog,
posturing, staring, and vocalization.
It has been postulated that interdog household ag-
gression occurs when dogs attempt to establish or rees-
tablish a hierarchy2,5 or when a dominant-subordinate
relationship is contested.1 Fighting in interdog house-
hold aggression cases often begins when circumstances
are changing (eg, an older dog becomes sick or dies, a
new dog is added to the household, or a younger dog
becomes socially mature). Hierarchies are dynamic and
are established by many factors including competition
over resources that are valuable to 1 or both dogs, mem-
From the Departments of Clinical Sciences (Wrubel, Moon-Fanelli,
Dodman) and Environmental and Population Health (Maranda), Cum-
mings School of Veterinary Medicine, Tufts University, North Grafton,
MA 01581. Dr. Wrubel’s present address is InTown Veterinary Group
Inc, Animal Behavior Services, Massachusetts Veterinary Referral Hos-
pital, 20 Cabot Rd, Woburn, MA 01801. Dr. Moon-Fanelli’s present
address is Animal Behavior Consultations LLC, Brooklyn Veterinary
Hospital, 150 Hartford Rd Rt 6, Brooklyn, CT 06234. Dr. Maranda’s pres-
ent address is Division of Clinical Research, University of Massachusetts
Medical School, 55 Lake Ave N, Worcester, MA 01655.
Address correspondence to Dr. Dodman (nicholas.dodman@tufts.edu).
Abbreviations
AKC American Kennel Club
CI Confidence interval
VAS Visual analogue scale
732 Scientific Reports JAVMA, Vol 238, No. 6, March 15, 2011
SMALL ANIMALS
ories of past encounters over resources, motivational
differences, and responsiveness to threat or deference
behaviors.9
The outcome for interdog household aggression is
reported to be worse if the instigating dog is the young-
er of the 2 dogs, if a human has been bitten, or if the
owner cannot predict when the aggression will occur.1
If an older dog in the home cannot fend off the chal-
lenge of a younger dog or a dog does not recognize def-
erence and becomes a chronic instigator, severe injuries
may result. In such circumstances, dogs may need to be
kept separated for safety purposes.
Alliance aggression is a type of interdog household
aggression in which dogs fight in the presence of their
owners and are frequently reported to coexist peace-
fully when left unsupervised in the household.10 Dogs
with alliance aggression typically fight over owner at-
tention, and many skirmishes occur around doorways
or hallways when the dogs simultaneously try to greet
their owners. These fights may be intense because the
owner is associated with valuable resources. It has been
postulated that owners may contribute to interdog
household aggression and leave unresolved conflict be-
tween the dogs if they treat their dogs equally, show
sympathy to a dog that instigates fights, or punish a dog
that instigates fights.1–5 Domestic canines are sensitive
to inequity in offering rewards,11 and this could lead to
subsequent conflict. Interdog household aggression can
be a response to inconsistent interactions in the house-
hold, which may create anxiety and tension between
the dogs. Dogs that have lived in multiple households
or shelters may have received mixed messages from
previous owners or cohabitating dogs. Also, owners in
the same home may behave inconsistently and respond
differently in the same situation.
Interdog household aggression may also result
from a lack of communication or miscommunication
between the dogs involved.8 If a dog does not display
or recognize communication signals properly, its inap-
propriate actions may lead to an aggressive encounter.
A dog may be at risk to act inappropriately if it was not
well socialized with other dogs, was orphaned or hand-
raised, or was born without littermates.
After examination to rule out physical disorders
such as endocrine abnormalities or pain, treatments
for interdog household aggression include avoidance of
fight-eliciting triggers, the use of a head collar or trail-
ing leash for physical control, and the use of an aversive
citronella spray or a physical barrier such as a board to
separate fighting dogs. Use of a muzzle is not typically
advised as it may increase anxiety or give 1 dog a physi-
cal advantage.
A common behavioral approach to resolving inter-
dog household aggression includes selection of 1 dog
to consistently receive the first access to resources and
the best resources (eg, food, treats, toys, owner atten-
tion, having a leash put on, and access to doorways
and furniture). Determination of which of the 2 dogs
to support (ie, which dog will be provided priority ac-
cess to resources) is typically based on age, sex, health,
and tenacity2; this may be difficult in a clinical setting
where time spent with the dogs is limited and owners
may not describe behavioral interactions accurately.
One method is to support the younger, healthier, larger,
or more confident dog, and the aggressor may possess
these qualities.3 Alternatively, a senior support program
may be suggested in which the eldest dog or the dog
that was acquired first receives support.5 It has been
recommended that clinicians wait up to 6 weeks before
reversing the rank order in cases where this approach
appears ineffective.5
A so-called nothing-in-life-is-free program may be
incorporated into the treatment plan to create better
communication and consistent interactions between
the dogs and owner. This entails making 1 or both dogs
work for their resources (including petting and atten-
tion) by obeying commands. Owners are instructed to
ignore attention-seeking behavior and to initiate and
terminate interactions with 1 or both dogs.
Other behavioral approaches to reduce interdog
household aggression include systematic desensitiza-
tion and counterconditioning methods.1 These tech-
niques involve gradually, incrementally, and repeatedly
exposing the dogs to each other and teaching them that
pleasant experiences occur when the other dog is near.
In extreme cases, separation and gradual reintroduction
of the dogs may be advised.
If a dog has extreme emotional attachment to its
owner and this results in fights, independence training
may provide a solution. This method involves encour-
aging, rewarding, and training the dog to do things in-
dependently (away from the owner) and can be particu-
larly helpful in cases of alliance aggression.
Medications that increase circulating concentrations
of serotonin in the brain can reduce reactivity, anxiety,
and responsiveness to aggression or fight-eliciting trig-
gers. The serotonergic system is involved in impulsive
aggression.12 Studies have shown that serotonin agonists
decrease aggression and that lower concentrations of se-
rotonin metabolites are found in the CSF of aggressive
dogs than in that of nonaggressive dogs.13 Fluoxetine,
a selective serotonin reuptake inhibitor, is a commonly
administered medication that is useful for treatment of
aggression problems and is reported to reduce aggression
in dogs after approximately 3 weeks of treatment.14
Research suggests that a low-protein, all-natural diet
can reduce some forms of aggression in dogs; high cir-
culating concentrations of protein are thought to block
serotonin precursors from crossing the blood-brain bar-
rier, and certain artificial preservatives are thought to
have negative effects on behavior.15 Exercise also helps
to relax an anxious dog and to increase concentrations of
serotonin in the brain, which helps to reduce reactivity.16
Few scientific studies in the literature have fo-
cused on interdog household aggression. The purpose
of the study reported here was to examine interdog
household aggression in detail from signalment to
outcome and to provide in-depth information on the
history of the dogs involved and other comorbid diag-
noses that could contribute to the problem. We also
sought to evaluate various treatment recommenda-
tions by use of a follow-up survey to assess outcomes,
owner compliance, and efficacy of the recommended
treatments, and to describe an effective treatment
strategy for improving interdog household aggression
problems on the basis of these findings.
JAVMA, Vol 238, No. 6, March 15, 2011 Scientific Reports 733
SMALL ANIMALS
Materials and Methods
Case selection—Hard-copy and computer data-
base files of the Animal Behavior Clinic at Tufts Univer-
sity Cummings School of Veterinary Medicine, Grafton,
Mass, were searched for records of dogs with interdog
household aggression that were examined during initial
or follow-up visits from December 5, 2006, to Decem-
ber 5, 2007. Each pair of dogs involved in interdog ag-
gression was considered 1 case. Although some dogs
(9/38 [24%]) had individual records, in the majority of
cases (29/38 [76%]), 1 record contained information
for both dogs. One case was excluded from the study
because the owner had been provided a written remote
consultation prior to the dogs’ initial examination at
the clinic.
Behavior consultations—Both dogs in each case
were examined during a 90-minute consultation, which
included review of a detailed history form completed
by the owner prior to consultation, discussion of be-
havioral issues and history with the owners, and ob-
servation of the dogs. A suitable treatment plan was
developed for each pair of dogs. Owners were offered
6 months of unlimited follow-up by telephone or
e-mail as part of the consultation process, and recheck
appointments were scheduled in 6-month increments.
Each consultation was performed by a veterinary be-
haviorist (NHD; 19/38 [50%] cases), a certified applied
animal behaviorist (AM-F; 16/38 [42%]), or both clini-
cians (3/38 [8%]). All dogs in the study were healthy,
and medical records were obtained from the referring
veterinarian. If medication was recommended, the vet-
erinary behaviorist examined the dog and prescribed
the drug. Dogs that received medication were examined
every 6 months, and blood samples were obtained at
each of these visits for biochemical analysis as a mea-
sure of liver function (circulating activities of ALP, ALT,
AST, and GGT; concentrations of albumin, total pro-
tein, globulin, bilirubin [total, direct, and indirect], tri-
glycerides, glucose, chloride, potassium, sodium, and
total CO2; and calculation of the albumin-globulin ratio
and anion gap).
Study design and records review—The 2-part
study included a detailed analysis of consultation re-
cords and of the results of postconsultation outcome
surveys. Data collected from case records included sig-
nalment and history information provided in a ques-
tionnaire completed by clients, discharge orders that
detailed the clinician’s observations, and recommenda-
tions approved by the primary clinician in charge of the
case. Outcome surveys were sent via mail to owners on
April 24, 2008, with a follow-up sent to those who did
not respond by July 23, 2008. Information collected
from surveys included self-reported owner compliance
and case outcomes.
History questionnaire—Descriptive data recorded
in history formsa by dog owners included age, weight,
breed, sex, and neuter status of each dog as well as in-
formation regarding duration of ownership and sources
from which dogs were obtained. A chart in the history
form allowed owners to indicate applicable risk factors
for behavior problems for their dogs, which included
a history of living in multiple households, adoption of
the dog when it was > 12 weeks old (ie, after the sensi-
tive period for socialization17), acquisition from a shel-
ter or pet shop, and being orphaned or born with no
littermates.
Owners provided information about the dogs by re-
sponding to open-ended questions. This included the
number of dogs in the household and the overall social
relationship between the dogs. Owners were asked to de-
scribe the fighting in general and to provide details regard-
ing the most recent fight. If multiple fights were described,
all were included in the analysis. Information reported
regarding incidents of interdog household aggression in-
cluded age of onset; whether 1 dog was known to con-
sistently instigate fights; number, frequency, duration, and
intensity of fights; whether fights resulted in injuries to
dogs or owners; and the types of injuries acquired, if ap-
plicable. Owners also described fight-eliciting triggers and
the methods they had used to stop fights. Owners were
asked whether the fighting had increased or decreased in
intensity, frequency, or duration.
A diagnostic form18 used to rate aggressive behavior
in 28 situations that typically elicit owner-directed ag-
gression from dogs was evaluated as part of each record.
Owners checked the most severe category of response
(not tried, no aggressive response, growling, lip-lifting,
snapping, and biting) for each situation. Situations de-
scribed on the form were designed to assess resource
guarding, guarding of personal space (ie, the area im-
mediately around the dog), and responses to human
behaviors potentially perceived as threatening by the
dog. If multiple responses were indicated for a particu-
lar situation, the 1 category that indicated the strongest
aggression response was used for analysis. For purposes
of analysis, owner-directed aggression was considered
clinically relevant in dogs that had growling, lip-lifting,
snapping, or biting responses indicated for > 5 of the 28
listed situations.
Owner assessments indicated on a chart19 used to
obtain diagnostic information for separation anxiety
were also analyzed. Owners rated each dog’s response
as none, mild, moderate, or severe for each of 13 situ-
ations used to determine the dog’s responses to owner
preparations to leave the house, what the dog’s actions
were when the owners were not at home, and how the
dog behaved when owners returned. Dogs for which 3
or more responses on the separation anxiety chart were
indicated as mild, moderate, or severe were considered
to have clinically relevant separation anxiety for pur-
poses of analysis. Owners were also asked whether their
dog appeared to be excessively frightened by various
(listed) items and situations, which included noises,
vehicles, people carrying things, car rides, appliances,
stairs, and slippery floors; they were also allowed to de-
scribe other items or situations.
The history form included open-ended questions
that asked for descriptions of each dog’s medical his-
tory and details regarding lifestyle and training history.
Questions were asked concerning the dog’s social in-
teractions with humans and other animals (ie, wheth-
er the dog had aggressive or fearful behaviors toward
members of the household, unfamiliar humans or ani-
mals, or veterinary staff).
734 Scientific Reports JAVMA, Vol 238, No. 6, March 15, 2011
SMALL ANIMALS
Follow-up surveys—Follow-up surveysa that con-
tained 28 questions were sent via mail to the owners
of all dogs identified in the first part of the study to
gain information on owner compliance and treatment
outcome. Surveys were returned by owners 8 to 62
months after the initial consultation. Owners were
asked how often fights had occurred between the dogs
before and after consultation and what the triggers were
for fights. Severity of fights was assessed by use of a
scoring system modified from a 6-point bite assessment
scale20 (Appendix). Owners scored the dogs’ bites on
a scale of 1 to 6 (where 1 = least severe and 6 = most
severe; a bite of this severity may have occurred once
or multiple times), and the highest number recorded
was subsequently analyzed. Owners were also asked
open-ended questions regarding which treatment rec-
ommendations they had implemented and which they
found to be most helpful. If medication was prescribed,
the owner was asked to indicate the type of medica-
tion and the duration of time that it had been given. A
15-cm VAS was used to subjectively assess the amount
of overall change in interdog household aggression
from 0 (left end of the scale, which indicated that fight-
ing was much worse) to 15 cm (right end of the scale,
which indicated substantial improvement). A neutral
response (ie, mark at 7.5 cm) represented no change.
Owners were asked to place a mark anywhere on the
line that indicated the amount of overall improvement
or regression in the interdog household aggression
problem, after considering the number, intensity, and
duration of fights. Locations of marks on the lines were
measured in centimeters to determine the perceived
amount of change. A monetary incentive, which con-
sisted of a $50 discount for a recheck appointment, was
offered for all owners who returned the follow-up sur-
vey. Investigators were aware of the sources of surveys
at the time of evaluation.
Statistical analysis—Because it was assumed that
the data were not normally distributed, nonparametric
statistical tests were used to analyze the data, includ-
ing medians with 95% CIs, Wilcoxon signed rank tests
for 2 nonindependent groups, and McNemar χ2 analy-
sis for matched pairs. For each record, 81 pieces of de-
scriptive data were entered for analysis.
Results
The records search revealed 38 cases of interdog
household aggression for which dogs were evaluated
during initial or recheck appointments from December
5, 2006, to December 5, 2007. All cases identified were
analyzed for the study except for 1 case, which was ex-
cluded because the owner had been provided a remote
consultation via e-mail from the Tufts Behavior Clinic
before the initial clinical evaluation was performed at
the hospital. Dogs that were examined during recheck
appointments in the time frame of the study had first
been evaluated between May 8, 2003, and July 12, 2006.
All cases seen prior to 2007 were recheck appointments
except for 1 case seen in late December 2006.
At the time of data collection, 25 of 38 (66%) cas-
es involved a single consultation, 9 (24%) had a sin-
gle consultation and 1 follow-up appointment, and 4
(11%) had 2 to 4 follow-up appointments. Of the cases
that had follow-up appointments, 3 pairs of dogs were
first evaluated between 2003 and 2005, and 7 were first
evaluated in 2006; the remaining 3 cases had an ini-
tial consultation in 2007 with the follow-up appoint-
ment during the same year. In these cases, the origi-
nal detailed history form and data from the clinician’s
discharge orders at the time of the initial appointment
were entered for analysis, as well as recommendations
made to clients at follow-up appointments.
Both dogs in each case were evaluated at the clinic
and had information in the record, but not all informa-
tion was available in the records for every dog. In cases
where information was more complete for 1 dog, this
was typically for the instigator of the fights.
Ancillary diagnoses—There were ancillary behav-
ioral diagnoses made by clinicians for 1 or both dogs
at the time of the behavioral consultation in 16 of 38
(42%) cases. These included owner-directed aggression,
generalized anxiety, hyperactivity, poor self-control, ex-
cessive barking, fear aggression toward humans, sepa-
ration anxiety, fear aggression toward unfamiliar dogs,
redirected aggression, urine marking, predatory aggres-
sion, and fear of humans. Because medical histories on
the questionnaires were provided by the owners and not
all were complete, other medical history was not ana-
lyzed further.
Breed, weight, sex, and neuter status—On the
basis of owner descriptions, dogs in the study were
grouped according to AKC breed group definitions.
Most (56/76 [74%]) dogs in the study were purebred.
All AKC breed groups were represented in the study as
follows: sporting (n = 14), hound (12), working (9),
herding (8), terrier (6), toy (5), and nonsporting (2).
Three dogs were of breeds not recognized by the AKC,
and 12 were of mixed breeds. Breeds of 5 dogs in the
study were unknown. Pairs of dogs that fought with
each other were of the same breed in 38 of 76 (50%)
cases and were related to each other in 6 of 38 (16%)
cases.
Dogs ranged in weight from 3.2 to 75.0 kg (7 to 165
lb; median, 23 kg [51 lb]; n = 47 dogs). Pairs of dogs that
fought with each other were females in 18 of 38 (47%)
cases, males in 12 of 38 (32%) cases, and both sexes in
8 of 38 (21%) cases. Overall, female dogs were involved
in 26 of 38 (68%) interdog household aggression cases.
Neuter status was available for 70 of 76 (92%) dogs, and
of these, 67 (96%) were neutered (1 female and 2 males
were sexually intact). Both sexually intact males in the
study were in the same household, and the sexually in-
tact female lived with a spayed female.
Age of onset—Median age of dogs at the time of
the initial consultation was 51 months (95% CI, 39 to
60 months; n = 75 dogs). Median age of dogs at onset
of interdog household aggression was 36 months (95%
CI, 15 to 42 months; n = 33 dogs).
Number of dogs in household—The median num-
ber of dogs in each household was 2 (range, 2 to 5; n
= 36 cases). The dogs involved in interdog household
aggression were the only dogs in the home in 21 of 38
JAVMA, Vol 238, No. 6, March 15, 2011 Scientific Reports 735
SMALL ANIMALS
(55%) cases. In a small number of cases, other dogs in
the household had been in fights with one of the pair
involved in the interdog household aggression, but this
had only happened rarely.
Instigators and recipients of aggression—Instiga-
tors and recipients of aggression were clearly identi-
fied in 27 of 38 (71%) cases. Instigators were a newer
addition to the household than were recipients in 19
of these 27 (70%) cases, and instigators were younger
than the recipients of aggression in 20 (74%) of these
cases. At the time of initial consultation, the median age
of instigators was 48 months (95% CI, 29 to 60 months;
n = 26), and median age of recipients of aggression was
72 months (95% CI, 41 to 84 months; 26). Complete
records were available for 26 of 27 instigators of aggres-
sion and for 7 of 27 recipients; because of this dispar-
ity, statistical analysis of the differences between these
groups could not be performed.
Training and social relationships between dogs—
Some form of training (obedience training, agility class-
es, or flyball) had been provided for 30 of 48 (63%)
dogs. Information was obtained from 28 of 38 owners
regarding the overall relationship between dogs with
interdog household aggression. Fifteen owners (54%)
reported that their dogs were not compatible in gen-
eral or appeared to ignore each other most of the time
or that 1 dog seemed afraid of the other dog. Eleven
owners (39%) reported that their dogs were compatible
most of the time, and 2 owners (7%) reported that their
dogs were sometimes compatible.
Fight-eliciting triggers—The most commonly re-
ported triggers for fights between dogs with interdog
household aggression were owner attention (16/35
[46%]), food (16/35 [46%]), excitement (11/35 [31%]),
and found items or toys (9/35 [26%]). Other fight-elic-
iting triggers that were identified included changes in
the home environment; 1 dog becoming weak or in-
jured; loud or sudden noises; passing through doorways
or sharing walkways; access to dog beds, furniture, or
crates; confinement in tight spaces; and the presence of
crowds or visitors.
Duration, number, and frequency of fights—The
owner-estimated duration of fights ranged from 3 sec-
onds to 30 minutes; however, further analysis was not
performed because many owners reported that fights
would have continued if they did not intervene. At the
time of the initial assessment, the median total number
of fights reported was 4 (95% CI, 3 to 5; n = 10). Fur-
ther analysis of this variable was not performed because
quantifiable information in the records was deemed
insufficient.
Owners reported frequency of fights in 17 cases. Of
these owners, 12 indicated that their dogs fought sever-
al times each week, and 5 reported that fights occurred
≤ 2 times monthly (range, 1/d to 1/y). Of 18 owners that
reported the duration of intervals between fights, 10 in-
dicated intervals of ≤ 1 week, 5 described intervals of 2
weeks to 1 month, and 3 reported that fights occurred
1 to 1.5 years apart. When asked if there had been a
change in the frequency or intensity of fights prior to
the initial consultation, 20 of 28 (72%) owners reported
that fights had become more frequent, more intense, or
both (without a specified time frame); in 4 of 28 (14%)
cases, no change in frequency was reported, and fights
had become less frequent in 4 of 28 (14%) cases be-
cause owners had avoided triggers or kept their dogs
separated for periods of time.
Severity of fights—The severity of fights described
in preconsultation history forms ranged from no bites
(eg, only growling and lunging) to those in which at
least 1 of the dogs received substantial injuries and
might have been killed. Injuries to dogs described
by owners included scratches, lacerations, lost teeth,
puncture wounds, and severe ocular trauma; some in-
juries required surgical repair and some infections were
reported. Owners indicated that veterinary medical
treatment was needed for 1 or both dogs after a fight
in 15 of 30 (50%) cases, and in 6 (40%) of these cases,
1 or both dogs had required veterinary treatment after
> 1 fight. Owners reported that they required medical
treatment themselves at least once after attempting to
intercede in fights in 3 of 30 (10%) cases.
Methods used to manage and prevent fights—
Information was available regarding preconsultation
management and prevention of dogfights for 35 of 38
cases. Methods owners reported they had used to man-
age fights prior to the initial consultation included non-
intervention, physically separating the dogs (ie, pulling
them away from each other during fights), keeping 1
dog leashed, throwing blankets to separate the dogs,
giving obedience commands, muzzling 1 or both dogs,
and aversive methods (eg, citronella spray, water spray,
or shock collars) or responses (eg, striking, shouting,
or physically forcing a dog down on its side or back).
Nineteen of 35 (54%) owners reported physically sepa-
rating the dogs from each other during fights, 18 (51%)
indicated that aversive methods or responses were used,
and 8 (23%) used obedience commands to attempt in-
tervention. Fifteen of 36 (42%) owners that responded
to the question of whether fights could be interrupted
without physically intervening replied that they could
not; 18 (50%) reported it was very difficult to interrupt
fights or that they had to physically pull the dogs away
from each other. Owners could usually interrupt fights
without physically intervening in 3 of 36 (8%) cases.
Methods used to prevent fights between the dogs
prior to consultation included avoidance of triggers, prais-
ing the dogs for nonaggressive behavior when together,
giving consistent preference to 1 dog in terms of priority
access to resources, isolation for short periods of time in
response to undesired or aggressive behavior, crating 1 or
both dogs, and keeping the dogs separated in the house-
hold. Nine of 35 (26%) owners kept the dogs separated in
their households as a method of fight prevention.
Risk factors for behavior problems—Of 51 dogs in
the study for which these data were available, 21 (41%)
were reported to have lived in multiple households, 20
(39%) were ≥ 12 weeks of age when adopted, 17 (33%)
were acquired from pet shelters, 8 (16%) were acquired
from pet shops, and 3 (6%) were orphaned or had no
littermates. These data were not statistically analyzed be-
cause of the lack of a control population for comparison.
736 Scientific Reports JAVMA, Vol 238, No. 6, March 15, 2011
SMALL ANIMALS
Comorbid behavior problems related to social in-
teractions—Dogs in the study were assessed for prob-
lems with social interactions involving aggressive or
fearful behaviors on the basis of owners’ reports. Of 48
dogs in the study for which these data were reported,
19 (40%) were reported to act aggressively toward un-
familiar dogs, 13 (27%) had some type of aggressive
behavior toward human members of the household,
13 (27%) had aggressive behavior toward unfamiliar
humans, and 3 (6%) had aggressive behavior toward
humans at the referring veterinarian’s office. Fearful be-
haviors were reported for 17 of 48 (35%) dogs when
unfamiliar humans were encountered, for 15 of 48
(31%) when the veterinarian’s office was visited, and
for 11 of 48 (23%) when unfamiliar dogs were encoun-
tered. No dogs were reported to act fearful of human
members in the household.
Owner-directed aggression, separation anxiety,
and other behavioral diagnoses—Owners completed
diagnostic charts for owner-directed aggression and
separation anxiety for 44 dogs in the study; of these, 9
(20%) had a clinical diagnosis of owner-directed aggres-
sion, and 22 (50%) were considered to have clinically
relevant separation anxiety. Twenty-three of 76 (30%)
dogs in the study were determined to have phobias or
severe anxiety. Phobias were triggered by noises in 20
of these 23 (87%) dogs; 1 dog had a phobia regarding
insects, and 2 dogs each had a diagnosis of generalized
anxiety.
Treatment recommendations—Recommendations
most commonly made to owners of dogs involved in
interdog household aggression cases included giving
1 dog priority access to resources (36/38 [95%] cases)
and implementation of a nothing-in-life-is-free program
for 1 or both dogs (32/38 [84%]). Use of a head collar,
trailing leash, or both to improve physical control of 1
or both dogs (30/38 [79%]); psychotropic medication
for 1 or both dogs (30/38 [79%]); obedience training
to improve the dogs’ responses to commands (27/38
[71%]); and avoidance of triggers for fights (23/38
[61%]) were also recommended to most owners.
Instructions for giving 1 dog priority access to re-
sources included consistently providing this dog with
the first and best resources. Owners were told to attend
to this dog after a fight and give this dog praise in front
of the other dog. The dog chosen to be supported (ie,
to receive first and best access to resources) was deter-
mined by 1 of the 2 investigators (NHD or AM-F) after
consideration of many factors including each dog’s tem-
perament, level of independence, and presence of any
comorbid fear- or anxiety-related conditions. In most
cases for which this treatment was recommended, cli-
nicians advised use of a senior support program (32/36
[89%] cases) in which the older dog of the pair (30/32
[94%] cases) or the dog that had lived in the household
for the greatest amount of time (2/32 [6%]) was chosen
to receive support. The younger or newer dog in the
household was recommended to receive this support at
a later time in 4 of 32 (13%) cases in which the senior
support program was initially implemented. Neither
dog was chosen to receive priority access to resources
in 2 of 38 (5%) cases; the dogs were siblings in 1 of
these cases, and dogs in the remaining case were of the
same age and had been obtained at the same time.
Owners were instructed to implement a noncon-
frontational nothing-in-life-is-free program in which 1
or both dogs were required to work for resources, atten-
tion-seeking behavior was to be ignored, and owners
were to initiate and terminate interactions with the dogs
in 32 of 38 (84%) cases. Medication was prescribed by
the veterinary behaviorist (NHD) for 1 or both dogs in
30 of 38 (79%) cases in which underlying anxiety was
thought to contribute to reactivity or aggression. The
most frequently prescribed medication was fluoxetine.
Owners were advised on how to avoid and eliminate
fight-eliciting triggers. Owners were also instructed on
methods to safely intervene or break up fights. A head
collar was recommended and fitted for the fight instiga-
tor or for both dogs in 30 of 38 (79%) cases.
Nonpunitive, positively reinforced training meth-
ods were recommended to owners in 27 of 38 (71%)
cases. Other recommendations included feeding a pre-
servative-free, low-protein diet; increasing the amount
of exercise the dogs received; implementing indepen-
dence training; and, in severe cases, separating and re-
introducing the dogs.
Follow-up survey responses and analysis—Twenty-
five of 38 (66%) owners completed and returned fol-
low-up surveys regarding the outcome of treatment for
interdog household aggression. Surveys were returned
a median of 15 months (range, 8 to 62 months) after
the initial consultation.
Owners that returned outcome surveys reported
that presence of a human member of the household was
a fight-eliciting trigger in 14 of 25 (56%) cases. Both
dogs passing through doorways simultaneously were a
trigger in 9 of 25 (36%) cases.
Six of 20 (30%) owners reported that they did not
leave their dogs together unsupervised. Of those that
did leave the dogs alone together and could state with
certainty whether fights occurred in this situation, 13 of
14 reported that their dogs did not fight when left alone
together. Owners were asked whether there had been a
change after the consultation in which dog instigated
the majority of fights, and 15 of 16 owners indicated
that there was no change.
Data obtained from the survey regarding the fre-
quency of fights before and after consultation were
transformed into number of fights that occurred each
month. Dogs for which the survey was returned (n = 19
cases) had a median of 4 fights/month (95% CI, 0 to 10)
prior to consultation and < 1 fight/month (median, 0.4
fights/mo [95% CI, 0 to 1]) after consultation. Results
of a Wilcoxon matched pairs signed rank test indicated
that the frequency of fights declined significantly (P =
0.001) after consultation.
Twelve of 25 (48%) survey respondents reported
that prior to consultation, 1 or both of their dogs re-
quired veterinary treatment as a result of injuries in-
curred from fights. After consultation, 4 of 22 (18%)
owners reported that 1 or both of their dogs required
veterinary treatment as a result of injuries incurred
from fights. However, results of a McNemar χ2 analysis
for matched pairs did not meet the criteria for signifi-
cance (P = 0.05).
JAVMA, Vol 238, No. 6, March 15, 2011 Scientific Reports 737
SMALL ANIMALS
The median score for severity of fights prior to con-
sultation (determined by use of a modified 6-point bite
assessment scale) was 3 (95% CI, 2 to 3; n = 25). The
median severity score after consultation was 2 (95% CI,
1 to 2; n = 25). Results of a Wilcoxon matched pairs
signed rank test indicated that fights were significantly
(P < 0.001) less severe after consultation.
Overall change in aggressive behaviors—The
VAS scores indicative of the overall change in the in-
terdog household aggression problem were measured
in centimeters from the center (7.5 cm; indicative of
no change), and median values were determined. None
of the 24 respondents indicated that the problem had
worsened after consultation. The 15-cm mark was ac-
cepted as 100% improvement; a median overall im-
provement of 69% was indicated by owners (95% CI,
46% to 93%). Owners reported some amount of im-
provement in 23 of 24 (96%) cases, and the minimum
improvement reported was 15% better than no change.
When owners were asked to estimate the interval of
time between the initial consultation and a noticeable
improvement in the aggression problem, the median re-
sponse was 5.2 weeks (95% CI, 2 to 13 weeks; n = 12
responses). Owners of dogs from 2 of 3 cases that were
first assessed between 2003 and 2005 returned surveys
for this part of the study; 1 owner indicated a 70% over-
all improvement, and the other indicated a 21% overall
improvement. Owners reported a 72% overall improve-
ment in cases in which 2 male dogs were fighting (n
= 7), a 75% overall improvement in cases in which a
male and female dog were fighting (5), and a 57% over-
all improvement in cases in which 2 female dogs were
fighting (11).
Owners who implemented a nothing-in-life-is-free
program for their dogs (n = 15) indicated an 89% im-
provement from no change on the VAS, whereas own-
ers who were not advised to implement the program or
chose not to do so (8) indicated a 28% improvement.
Owners who implemented a senior support program
(18) indicated a 67% improvement from no change on
the VAS.
In cases in which medication was prescribed for
both dogs (n = 8), owners indicated a 57% improvement
from no change on the VAS, and in cases where medica-
tion was prescribed for only 1 dog (12), an 88% improve-
ment was indicated; in cases in which no medication was
prescribed (3), a 91% improvement was indicated.
Negative outcomes—Two of 25 (8%) survey re-
spondents reported that their dogs were living sepa-
rately in their households all of the time prior to the
initial consultation (median duration, 4.3 months), and
3 of 24 (13%) owners reported that their dogs were liv-
ing separately after consultation (median duration, 4.2
years). One of the dogs had been placed in a different
household by the owner because of the interdog house-
hold aggression problem in 1 of 24 (4%) cases. None
of the dogs in the study were euthanized because of
interdog household aggression; however, 1 dog was eu-
thanized for another behavior problem.
Self-reported owner compliance—Clients had been
offered 6 months of unlimited follow-up via telephone
or e-mail as part of the consultation process; 20 of 25
(80%) survey respondents indicated that they had used
1 or both of these options. Owners were asked to report
the treatment recommendations that they implemented
for their dogs; the treatments most commonly imple-
mented were psychotropic medication (21/24 [88%]),
giving 1 dog priority access to resources (19/24 [79%]),
a nothing-in-life-is-free program (15/24 [63%]), use of a
head collar for 1 or both dogs (12/24 [50%]), increased
exercise (12/24 [50%]), and keeping the dogs separated
and gradually reintroducing them (11/24 [46%]).
Helpful recommendations—Owners were asked
which clinician recommendations they found most
helpful. They reported that the most useful recommen-
dations were medication (15/23 [65%]), implementing
a nothing-in-life-is-free program (11/23 [48%]), giving
1 dog priority access to resources (11/23 [48%]), use
of a head collar for 1 or both dogs (5/23 [22%]), and
keeping the dogs separated and gradually reintroducing
them (6/23 [26%]).
Medication—Survey respondents indicated that
psychotropic medication had been prescribed for 1 or
both dogs in 22 of 25 (88%) cases. Prescribed medica-
tions included fluoxetine, paroxetine, amitriptyline, ser-
traline, and buspirone. One dog received medication in
14 of 22 (64%) cases, and both dogs received medica-
tion in 8 of 22 (36%) cases. Fluoxetine was prescribed
for 25 of 30 (83%) dogs that received medication. In 2
cases, the medication was changed because of poor re-
sults or adverse effects associated with the first drug ad-
ministered. At the time of survey administration, 20 of
22 (91%) owners were still administering psychotropic
medication to their dogs, and the median duration of this
treatment was 17 months (95% CI, 10 to 24 months).
Discussion
A study of free-ranging dogs found that agonistic
interactions among group members were scarce and
communication was subtle; aggression was typically
in the form of threats by means of growling and snarl-
ing, and chasing and severe fighting behaviors were not
observed.21 A household environment induces physical
restrictions and limitations on dogs that can contribute
to the development of interdog household aggression
(ie, a dog cannot leave). Limited resources in a house-
hold can cause them to have a higher value, which may
result in tension between cohabiting dogs.8
Owner attention and intervention can be integral
in interdog household aggression problems, and 13
of 14 owners that responded in follow-up surveys in
the present study reported that their dogs did not fight
when left alone together, indicating that most of these
cases likely involved alliance aggression. Dogs are also
sensitive to inequity of reward, and if an owner pro-
vides differential treatment, dogs will recognize this.11
More resources are available when an owner is present;
thus, the simple presence of the owner may not be the
sole trigger that elicits the fight, although the owner’s
attention is a resource worth competing over.
The results of the study reported here concur with
those of previous studies1,22,23 that indicated interdog
738 Scientific Reports JAVMA, Vol 238, No. 6, March 15, 2011
SMALL ANIMALS
household aggression is more common between dogs
of the same sex and that female-female pairs are most
often affected. In the present study, responses from
owners that returned the follow-up survey indicated
a 72% overall improvement in cases in which 2 male
dogs were fighting (n = 7), a 75% overall improvement
in cases in which a male and female dog were fighting
(5), and a 57% overall improvement in cases in which 2
female dogs were fighting (11). Sherman et al1 reported
that fights among female dogs tended to be the most
intense, but outcome results in that study indicated
that female-female pairs did not have worse behavioral
outcomes than those of other sex combinations. It has
been suggested that an owner can reduce the risk of
having an interdog household aggression problem by
means of adding a male dog to the household.1 Results
of the present study revealed that the median age of
onset of interdog household aggression is 36 months,
which is approximately the time that dogs reach social
maturity and develop the confidence to use aggression
to control a situation in which they are uncomfortable.
According to the findings of Sherman et al1 and of
the present study, approximately half of all cases of in-
terdog household aggression involve dogs of the same
breed. Emotional responses, physiologic and behav-
ioral arousal, and the tendency toward aggression vary
among dog breeds.17,24,25 Dogs of the same breed may be
similarly motivated for particular resources or perceive
value in the same resources, including time and atten-
tion from the owner, which is limited in the confines of
a household. This may also be true for dogs of different
breeds in the same household. These factors could lead
to competition, stress, anxiety, and conflict.
Dogs involved in interdog household aggression
cases in the study reported here were found to have
several risk factors for behavior problems. Of 51 dogs
for which the information was reported, 21 (41%) had
lived in multiple households, 20 (39%) were adopted
after the sensitive period for socialization17 (ie, after
12 weeks of age), 17 (33%) were acquired from a shel-
ter, 8 (16%) were acquired from pet shops, and 3 (6%)
were orphaned or had no littermates. These dogs could
have developed anxiety because of previous suboptimal
life experiences. Factors such as these could result in
a dog’s not learning proper signals for communication
with other dogs that could otherwise diffuse potentially
agonistic situations. There is, however, evidence to sug-
gest that animals raised in isolation are able to demon-
strate normal social behavior when introduced to con-
specifics several months later.26 Dogs that have lived in
multiple households could have anxiety about what to
expect in their surroundings. Inconsistent interactions
or lack of leadership in the current or a previous home
could also worsen an already fragile situation. Unfortu-
nately, we were unable to analyze differences between
dogs with interdog household aggression and a control
population, so it is unknown whether dogs in the pres-
ent study had a higher percentage of these risks factors
than did unaffected dogs.
Ancillary diagnoses had been determined for dogs
in 16 of 38 (42%) interdog household aggression cas-
es. Many dogs in the present study (22/44 [50%]) had
some degree of separation anxiety. These dogs may
have formed a very strong attachment to their owners,
which could be linked to the apparent prevalence of
alliance aggression among dogs reported here. The un-
expectedly high number of dogs with separation anxi-
ety in our study may also be a reflection of generalized
anxiety underlying the competitive aspects of interdog
household aggression.
In agreement with the findings of another study,1
9 of 44 (20%) dogs in the present study with interdog
household aggression also had owner-directed aggres-
sion, indicating that a proportion of these dogs also
have conflict with their owners and are confident us-
ing aggression to try to control situations in which they
are uncomfortable. Of dogs in the present study, 19 of
48 (40%) had aggressive behavior toward unfamiliar
dogs, and 11 of 48 (23%) had fearful behaviors toward
unfamiliar dogs. These dogs may not have been prop-
erly socialized with other dogs or may have had previ-
ous negative learning experiences with other dogs, so
some fights between household dogs could have been
attributable to tension and difficulty in establishment
of trust with a cohabiting dog. Other behavior prob-
lems involving fear and anxiety were also commonly
reported. Of dogs in the study reported here, 13 of 48
(27%) had some aggressive behavior toward unfamil-
iar humans, and 17 of 48 (35%) were reported to have
some fearful behavior around strangers. Twenty-three
of 76 (30%) study dogs had phobias or generalized
anxiety. One limitation of this retrospective study was
that we were unable to perform statistical analysis for
features of dogs categorized as instigators of aggression
versus those categorized as recipients, which would
have provided some interesting insight into this issue.
Future studies should investigate differences between
instigators and recipients in regard to risk factors, social
interactions, aggression, or anxiety in other contexts.
The most commonly recommended treatments in
the present study (ie, giving 1 dog priority access to
resources, a nothing-in-life-is-free program, and psy-
chotropic medications) were also rated most effective
by survey respondents. The frequency and severity of
fights decreased significantly after consultation; the
number of cases in which veterinary medical treatment
was needed as a result of injuries incurred during fights
decreased by 30% after consultation. Owners reported
some amount of improvement in the interdog house-
hold aggression problem in 96% (23/24) of cases, with
a minimum improvement of 15% and a median 69%
improvement over no change after consultation.
Sherman et al1 reported some degree of improve-
ment in 39 of 66 (59%) interdog household aggression
cases; in that study, 27 of 66 (41%) owners reported
that there was no change or that the situation had
worsened. The most commonly recommended treat-
ments in that study were systematic desensitization and
counterconditioning methods, and owner compliance
with these methods is often poor,7 which may explain
the difference in outcomes between that study and the
study reported here.
Survey respondents in cases in which a nothing-
in-life-is-free program was implemented reported an
89% improvement in interdog household aggression,
whereas those who were not recommended to use the
JAVMA, Vol 238, No. 6, March 15, 2011 Scientific Reports 739
SMALL ANIMALS
program or chose not to do so reported only a 28% im-
provement. The nothing-in-life-is-free program used
in the present study did not involve making the dog
assume progressively more submissive postures to
earn resources or exposure techniques, which some
programs with the same name entail27; instead, it in-
volved making owners interact with their dog in a cue-
response-reward format when valuable resources were
present.24 This program provides structure and consis-
tency in the household through enhanced communica-
tion and predictable social interactions around resourc-
es, which substantially reduces conflict and anxiety in
the authors’ clinical experience. Owners help dogs to
learn appropriate behavior in this program by offering
resources as rewards for desired behavior and by not
rewarding inappropriate behavior. This program also
improves the predictability of the cohabiting dogs’ be-
haviors, which reduces conflict.
In the present study, a 67% improvement was report-
ed for cases in which a senior support program (wherein
the oldest dog in the house or the dog that was in the
home first received priority access to resources) was im-
plemented. The fact that the fight instigator was often the
younger of the 2 dogs or a newer addition to the house-
hold could be the reason for the success of this program
because that dog may be comparatively insecure and thus
instigate conflict. In a study21 of free-ranging dogs, adult
males and females were not observed to have submissive
behaviors toward juvenile dogs. Giving 1 dog priority ac-
cess to resources reduces conflicts that result in aggressive
behavior because it allows the dogs to successfully coexist
in a household without the stress of living in an unpredict-
able environment.
It is important to be aware that dominance is a rela-
tive measure, rather than an absolute property of indi-
viduals, and dominance hierarchies among dogs are not
absolute and strict.28–30 Pack structure is thought to be
dynamic and marked by continuous change. It has been
reported31 that dogs living in households may maintain
several competitive hierarchies determined separately
for each resource. It has also been suggested32 that dogs
have no real hierarchy but have preferences for differ-
ent resources in the environment; however, the stability
of the social structure is also a highly valued resource,
and dogs may relinquish a resource to maintain social
harmony.
Use of a nothing-in-life-is-free program and giv-
ing 1 dog priority access to resources may also initiate
counterconditioning protocols for each dog. Dogs learn
the order in which each will receive valuable resources,
and the dog that receives a resource second must wait
until the first dog has an opportunity to earn the re-
source. Also, both dogs may be asked to engage in a
behavior such as sitting or lying down when a valuable
resource is present, which could diffuse tension dur-
ing situations that might have elicited fights in the past
because the dogs must behave in a controlled manner
and thus have more relaxed behavior around resources.
In the majority of cases of interdog household ag-
gression in the present study, 1 or both dogs received
psychotropic medication. The cases in which neither
dog was prescribed medication were rated as having the
most success, followed by cases in which 1 dog was pre-
scribed medication; cases in which both dogs were pre-
scribed medication were rated as having the least suc-
cess. In cases in which medication was prescribed for 1
or both dogs, the fights were severe or there was reason
to believe that underlying anxiety was severe enough
to warrant pharmacological intervention. This type of
medication would help to make these dogs less reactive
and, ideally, less aggressive.
It has been reported24 that most dogs that have a
diagnosis of owner-directed aggression are not so-called
dominant dogs with confident personalities. In agree-
ment with the findings of the present study, those au-
thors reported that aggressive dogs often act uncertain,
fearful, or submissive in some situations but become ag-
gressive when conflict arises. They suggested this may
be attributable to inconsistent interactions that lead to a
dog’s inability to predict what will happen next. In dogs
of the present study, giving 1 dog priority access to re-
sources and enforcing a nothing-in-life-is-free program
reduced conflict, likely because these methods provide
consistency and predictability; the dog learns how to
successfully behave in the environment in order to re-
ceive resources. Our study revealed that psychotropic
medication, which may reduce a dog’s reactivity to anx-
iety-eliciting triggers, was also an effective adjunctive
treatment to reduce interdog household altercations at-
tributed to lack of environmental or social predictabil-
ity. In a typical household, dogs rely on their owners to
provide everything they need, and owners themselves
can become a resource to fight over. The findings of the
present study indicate that interdog household aggres-
sion can be successfully treated and that consistency
and predictability of social interactions are essential in
resolving the issue.
a. Copies of forms are available on request from the corresponding
author.
References
1. Sherman CK, Reisner IR, Taliaferro LA, et al. Characteristics,
treatment, and outcome of 99 cases of aggression between dogs.
Appl Anim Behav Sci 1996;47:91–108.
2. McKeown D, Luescher AU. Canine competitive aggression: a
clinical case of “sibling rivalry.” Can Vet J 1988;29:395–396.
3. Overall KL. Animal behavior case of the month. Aggression between
2 pugs in a household. J Am Vet Med Assoc 1995;207:305–307.
4. Beaver BV. Clinical classification of canine aggression. Appl
Anim Ethol 1983;10:35–43.
5. Dodman NH, Arrington D. Animal behavior case of the month.
J Am Vet Med Assoc 2000;217:1468–1472.
6. van Kerkhove W. A fresh look at the wolf-pack theory of com-
panion-animal dog social behavior. J Appl Anim Wel Sci 2004;7:
279–285.
7. King T. Comment on van Kerkhove’s commentary. J Appl Anim
Wel Sci 2004;7:293–294.
8. Mertens P. The concept of dominance and the treatment of ag-
gression in multidog homes: a comment on van Kerkhove’s
commentary. J Appl Anim Wel Sci 2004;7:287–291.
9. Lindsay SR. Handbook of applied dog behavior and training. Vol-
ume 2: etiology and assessment of behavior problems. Ames, Iowa:
Iowa State University Press, 2001.
10. Luescher A. Behavioral disorders. In: Ettinger SJ, Feldman EC,
eds. Textbook of veterinary internal medicine. 6th ed. St Louis:
WB Saunders Co, 2004;183–189.
11. Range F, Horn L, Virany Z, et al. The absence of reward induces ineq-
uity aversion in dogs. Proc Natl Acad Sci U S A 2009;106:340–345.
740 Scientific Reports JAVMA, Vol 238, No. 6, March 15, 2011
SMALL ANIMALS
12. Peremans K, Audenaert K, Coopman F, et al. Estimates of re-
gional cerebral blood flow and 5-HT2A receptor density in im-
pulsive, aggressive dogs with 99mTc-ECD and 123I-5-I-R91150.
Eur J Nucl Med Mol Imaging 2003;30:1538–1546.
13. Reisner IR, Mann JJ, Stanley M, et al. Comparison of cerebrospi-
nal fluid monoamine metabolite levels in dominant-aggressive
and non-aggressive dogs. Brain Res 1996;714:57–64.
14. Dodman NH, Donnelly R, Shuster L, et al. Use of fluoxetine
to treat dominance aggression in dogs. J Am Vet Med Assoc
1996;209:1585–1587.
15. Dodman NH, Reisner I, Shuster L, et al. Effect of dietary protein con-
tent on behavior in dogs. J Am Vet Med Assoc 1996;208:376–379.
16. Jacobs BL. Serotonin, motor activity and depression-related dis-
orders. Am Sci 1994;82:456–463.
17. Scott JP, Fuller JL. Genetics and the social behavior of the dog.
Chicago: The University of Chicago Press, 1965;84–224.
18. Dodman NH, Smith A, Holmes D. Comparison of the efficacy of
remote consultations and personal consultations for the treat-
ment of dogs which are aggressive towards their owners. Vet Rec
2005;156:168–170.
19. Cottam N, Dodman NH, Moon-Fanelli AA, et al. Comparison of
remote versus in-person behavioral consultation for treatment of
canine separation anxiety. J Appl Anim Welf Sci 2008;11:28–41.
20. Association of Pet Dog Trainers website. Dr. Ian Dunbar’s dog bite
scale. An assessment of the severity of biting problems based on
an objective evaluation of wound pathology. Available at: www.
apdt.com/veterinary/assets/pdf/Ian Dunbar Dog Bite Scale.pdf.
Accessed Dec 29, 2010.
21. Pal SK, Ghosh B, Roy S. Agonistic behaviour of free-ranging
dogs (Canis familiaris) in relation to season, sex and age. Appl
Anim Behav Sci 1998;59:331–348.
22. Borchelt PL. Aggressive behavior of dogs kept as companion
animals: classification and influence of sex, reproductive status
and breed. Appl Anim Ethol 1983;10:45–61.
23. Bamberger M, Houpt KA. Signalment factors, comorbidity, and
trends in behavior diagnoses in dogs: 1,644 cases (1991–2001).
J Am Vet Med Assoc 2006;229:1591–1598.
24. Luescher AU, Reisner IR. Canine aggression toward familiar
people: a new look at an old problem. Vet Clin North Am Small
Anim Pract 2008;38:1107–1130.
25. Duffy DL, Hsu Y, Serpell JA. Breed differences in canine aggres-
sion. Appl Animal Behav Sci 2008;114:441–460.
26. MacDonald KB, Ginsburg BE. Induction of normal prepubertal
behavior in wolves with restricted rearing. Behav Neural Biol
1981;33:133–162.
27. Voith VL, Borchelt PL. Diagnosis and treatment of domi-
nance aggression in dogs. Vet Clin North Am Small Anim Pract
1982;12:655–663.
28. Borchelt PL, Voith VL. Dominance aggression in dogs. In: Voith
VL, Borchelt PL, eds. Readings in companion animal behavior.
Trenton, NJ: Veterinary Learning Systems, 1996;230–239.
29. Drews C. The concept and definition of dominance in animal
behaviour. Behaviour 1993;125:283–313.
30. Bradshaw JWS, Blackwell EJ, Casey RA. Dominance in domestic
dogs—useful construct or bad habit? J Vet Behav 2009;4:135–
144.
31. Wickens SM, Bradshaw JWS. Dominance relationships in
groups of domestic dogs (Canis familiaris). Appl Anim Behav Sci
1993;35:291–298.
32. Semyonova A. The social organization of the domestic dog: blow-
ing up the dominance myth. The Hague: The Carriage House
Foundation, 2003.
Severity level Threat or bite characteristics
1 Growling, lunging, or snarling behavior occurred without teeth touching skin (ie, mostly intimidation behavior).
2 Teeth touched skin, but no puncture wounds were inflicted. Marks or minor scratches from paws and nails (minor surface
abrasions) may have been incurred.
3 Punctures were half the length of a canine tooth and resulted in 1 to 4 holes from a single bite. No tears or slashes were
incurred, and the recipient was not shaken side to side.
4 One to 4 holes from a single bite, with 1 or more holes deeper than half the length of a canine tooth. Contact and punctures
were incurred from more than the canine teeth. Tears, slash wounds, or both resulted. One dog clamped its teeth down and the
other dog was shaken or slashed.
5 Multiple bites at severity level 4 or greater incurred in a concerted, repeated attack.
6 Any bite that resulted in death of a dog.
Information was modified from a 6-point bite assessment scale.20 The scoring system was used by owners of dogs with interdog household
aggression to score the severity of fights as part of a follow-up survey to determine owner compliance and efficacy of recommended treatments.
The highest number recorded from a single bite or multiple biting episodes was analyzed.
Appendix
Modified bite assessment scale.
