Content uploaded by Karen Overall
Author content
All content in this area was uploaded by Karen Overall on Apr 18, 2014
Content may be subject to copyright.
S
eparation anxiety is one of the most common and
devastating behavioral conditions in pet dogs. It has
been estimated, for instance, that at least 14% of dogs
examined at typical veterinary practices in the United
States have signs of separation anxiety.
1
Unfortunately,
as is true for most behavioral conditions, signs associ-
ated with separation anxiety are nonspecific, making
an accurate diagnosis difficult. Dogs with separation
anxiety usually destroy objects in the house, destroy
sections of the house, or urinate, defecate, vocalize, or
salivate when they are left alone.
2-4
Dogs with profound
separation anxiety can be left alone for no more than
minutes before they begin to panic and exhibit the
behaviors associated with separation anxiety. In the
absence of treatment, affected dogs are often relin-
quished to a humane society or shelter, abandoned, or
euthanatized.
5,6
Elimination, destruction, and vocalization are the
most obvious and, hence, the most commonly reported
behaviors associated with separation anxiety.
2,3
It is
important, however, to realize that clients complain
about these behaviors most often because they are easy
to recognize and cause problems for the client them-
selves. Dogs that exhibit less obvious signs of distress
such as withdrawal, inactivity, salivation, whimpering
(or louder vocalizations if there are no nearby neigh-
bors), and pacing may not be identified as having sepa-
ration anxiety, even though they are equally affected,
and owners of such dogs seldom seek treatment for their
pets, because the dogs’ behaviors are not problems for
the clients. Including dogs with these less obvious signs,
the population of dogs with separation anxiety, although
unknown, is likely to be quite large, and all dogs with
separation anxiety can benefit from treatment.
Noise and thunderstorm phobias are among the
most commonly recognized disorders associated with
panic or phobic responses in dogs, although no data
exist on relative incidences of these problems. In our
experience, dogs with signs of separation anxiety fre-
quently also have signs of noise phobia or thunder-
storm phobia. This observation suggests 4 avenues of
investigation that merit exploration: whether the
development or expression of noise phobia, thunder-
storm phobia, and separation anxiety are codependent,
with the extent to which they covary suggesting differ-
ences in mechanisms for thresholds of anxiety-related
disorders; whether reactions to noises may predispose
dogs to other anxiety-related conditions; whether the
interactions among these disorders may have time pen-
etrance, such that the longer a dog has been affected
with 1 of these conditions, the more at risk it may be
either for a more complex form of that condition (eg, a
greater number or intensity of signs) or for other anx-
iety-related conditions; and whether the incidence and
comorbidity (ie, the cooccurrence of separate, but pos-
sibly related, conditions) of these conditions may be
underestimated, particularly in the absence of a ques-
tionnaire or evaluation tool that systematically
explores all responses to anxiety-related situations in
all dogs.
JAVMA, Vol 219, No. 4, August 15, 2001 Scientific Reports: Original Study 467
SMALL ANIMALS
Frequency of nonspecific clinical signs in dogs
with separation anxiety, thunderstorm phobia,
and noise phobia, alone or in combination
Karen L. Overall, VMD, PhD, DACVB; Arthur E. Dunham, PhD; Diane Frank, DVM, DACVB
Objective—To determine the frequency of nonspe-
cific clinical signs in dogs with separation anxiety,
thunderstorm phobia, noise phobia, or any combina-
tion of these conditions and determine whether these
conditions are associated in dogs.
Design—Case series.
Animals—141 dogs.
Procedure—Diagnoses were established using spe-
cific criteria. Owners of dogs completed a question-
naire on how frequently their dogs exhibited destruc-
tive behavior, urination, defecation, vocalization, and
salivation when the owners were absent and the
types and frequency of reactions to thunderstorms,
fireworks, and other noises.
Results—Associations of the 3 conditions and of var-
ious nonspecific clinical signs within and between
diagnoses were nonrandom. The probability that a
dog would have separation anxiety given that it had
noise phobia was high (0.88) and approximately the
same as the probability it would have separation anx-
iety given that it had thunderstorm phobia (0.86).
However, the probability that a dog would have noise
phobia given that it had separation anxiety (0.63) was
higher than the probability that it would have thun-
derstorm phobia given that it had separation anxiety
(0.52). The probability that a dog would have noise
phobia given that it had thunderstorm phobia (0.90)
was not equivalent to the converse (0.76).
Conclusions and Clinical Relevance—Results sug-
gested that dogs with any of these conditions should
be screened for the others. Interactions among these
conditions are important in the assessment and treat-
ment of dogs with > 1 of these conditions.
Responses to noise were different from those to
thunderstorms, possibly because of the unpre-
dictability and uncertainty of thunderstorms. (
J Am
Vet Med Assoc
2001;219:467–473)
From the Department of Clinical Studies, School of Veterinary
Medicine (Overall, Frank), and the Department of Biology
(Dunham), School of Arts & Sciences, University of Pennsylvania,
Philadelphia, PA 19104-6010.
Presented in part at the 1997 Annual Meeting of the American
Veterinary Society of Animal Behavior, Reno, Nev, and the 2000
Annual Meeting of the American Veterinary Society of Animal
Behavior, Salt Lake City, Utah.
01_02_0059.QXD 10/12/2005 9:46 AM Page 467
Specific criteria for the diagnosis of noise phobia,
thunderstorm phobia, and separation anxiety in dogs
have been published.
7
The use of such criteria allows
examination of the distribution of nonspecific signs of
anxiety-related disorders independent of diagnosis.
8-10
Nonspecific signs may be uniquely important for
assessing psycho-pathologically distinct groups,
11
and
an improved understanding of the intensity, duration,
and age of onset of certain signs may improve our
understanding of the variable behavioral phenotypes
associated with these conditions.
12
The use of explicit
diagnostic criteria is key to ensuring that observed het-
erogeneity is not an artifact of careless classification.
Use of explicit diagnostic criteria is also a powerful tool
for examining phenotypes and endophenotypes (ie,
components of complex phenotypes) and avoids the
underdiagnosis of comorbid conditions that is preva-
lent when somatic signs are used instead of indepen-
dent diagnostic criteria.
9,10
This type of approach is
essential for elucidating any underlying genetic sus-
ceptibility for behavioral or psychiatric conditions.
10
In
cases where there are strong genetic influences on dif-
ferent facets of temperamental expressions such as
fearfulness, the biological correlates of the behaviors
can be considered as related endophenotypes for anxi-
ety.
13,14
The purposes of the study reported here were to
evaluate the covariation (ie, the nonrandom associa-
tion) of nonspecific signs of separation anxiety, thun-
derstorm phobia, and noise phobia and determine the
comorbidity of these conditions in dogs.
Materials and Methods
For the purposes of this study, the necessary conditions
for a diagnosis of separation anxiety
7
were physical or behav-
ioral signs of distress exhibited by the dog only during an
actual absence of the owner (ie, when the owner is not at
home) or a virtual absence of the owner (ie, when the owner
is home, but the dog is denied access to the owner by a door
or some similar barrier). Sufficient conditions for a diagnosis
of separation anxiety were consistent intensive destruction,
elimination, vocalization, or salivation exhibited only during
an actual or virtual absence of the owner, with these behav-
iors most severe close to the time of separation and many
anxiety-related behaviors (eg, autonomic hyperactivity,
increased motor activity, and increased vigilance and scan-
ning) becoming apparent as the owner exhibits behaviors
associated with leaving.
Necessary and sufficient conditions for a diagnosis of
noise phobia
7
were sudden profound nongraded extreme
responses to noise that were manifest as intense active avoid-
ance, escape, or anxiety behaviors associated with activities
of the sympathetic branch of the autonomic nervous system.
Such behaviors could include catatonia or mania concomi-
tant with decreased sensitivity to pain or social stimuli.
Repeated exposure resulted in an invariant pattern of
response.
Necessary and sufficient conditions for a diagnosis of
thunderstorm phobia
7
were sudden profound nongraded
extreme responses to thunderstorms or any aspect of thun-
derstorms (eg, wind, noise, lightning, changes in barometric
pressure, rain, darkness, changes in ozone concentration)
that were manifest as intense active avoidance, escape, or
anxiety behaviors associated with activities of the sympathet-
ic branch of the autonomic nervous system. Such behaviors
could include catatonia or mania concomitant with
decreased sensitivity to pain or social stimuli. Repeated expo-
sure resulted in an invariant pattern of response.
The present study was conducted in 2 parts. In the first
part, medical records of 51 dogs that met the stated criteria
for separation anxiety, noise phobia, or thunderstorm phobia
and in which the diagnosis had been made during a 6-month
period in 1997 were reviewed. Information on presence and
frequency of attendant nonspecific signs was collected. The
Pearson product-moment correlation method was used to
identify significant pair-wise correlations for suites of behav-
iors and nonspecific signs.
Since 1998, owners of all dogs examined at the
Veterinary Hospital of the University of Pennsylvania
Behavior Clinic have been asked to complete a questionnaire
a
on behaviors exhibited by their dogs during actual and virtu-
al absences of the owners. The questionnaire asked owners to
indicate whether their dogs exhibited any of 5 behaviors
(destructive behavior, urination, defecation, vocalization,
and salivation) during actual or virtual absences of the own-
ers (yes, no, unknown) and, if their dogs did exhibit any of
these behaviors, to indicate how frequently they did so
(100% of the time, < 100% but > 60% of the time, 40 to 60%
of the time, > 0 but < 40% of the time). In addition, the ques-
tionnaire asked owners to indicate whether their dogs react-
ed to thunderstorms, fireworks, or other noises (yes, no,
unknown) and, if their dogs did react, how frequently they
did so (100% of the time, < 100% but > 60% of the time, 40
to 60% of the time, > 0 but < 40% of the time) and what
responses they exhibited (salivating, defecating, urinating,
destroying, escaping, hiding, trembling, and vocalizing).
In the second part of the present study, records for 141
dogs (81 males and 60 females) examined between January
1999 and January 2000 that were determined to have separa-
tion anxiety, noise phobia, or thunderstorm phobia and for
which responses to at least 1 of the items on the screening
questionnaire were positive were reviewed. Ambiguous data
or client responses that were based on only a singular occur-
rence of a particular behavior were excluded from the rele-
vant analyses.
Independence of diagnoses and nonspecific signs (ie,
client responses to the screening questionnaire) was evaluat-
ed by use of Cochran-Mantel-Haenszel (CMH) χ
2
tests and
by comparing results of these tests with results for a series of
1,000 CMH tests applied to randomized data sets generated
by Monte Carlo simulation assuming complete independence
of all diagnoses and client responses. For all analyses, values
of P < 0.05 were considered significant.
Results
In the first part of the study, 25 dogs had separa-
tion anxiety alone, 5 dogs had noise or thunderstorm
phobia alone, and 21 dogs had separation anxiety and
noise or thunderstorm phobia. The probability that
a dog had separation anxiety given that it had noise
or thunderstorm phobia [P(SA|N)] was 0.81
[(21/51)/(26/51)], whereas the probability that a dog
had noise or thunderstorm phobia given that it
had separation anxiety [P (N|SA)] was 0.46
[(21/51)/(46/51)]. Frequencies with which dogs exhib-
ited nonspecific anxiety-related signs (always [100% of
the time], mostly [< 100% but > 60% of the time], con-
sistently [40 to 60% of the time], seldom [< 20% of the
time], and never [0% of the time]) were not indepen-
dent of the nonspecific signs or of the specific condi-
tion, regardless of whether dogs had only 1 or multiple
conditions. Owners of dogs that always reacted in situ-
ations associated with noise phobia were more likely to
468 Scientific Reports: Original Study JAVMA, Vol 219, No. 4 , August 15, 2001
SMALL ANIMALS
01_02_0059.QXD 10/12/2005 9:46 AM Page 468
indicate that their dogs always destroyed, urinated,
vocalized, or salivated than were owners of dogs that
reacted less often in such situations. Dogs that had
noise phobia in addition to separation anxiety had
more severe signs (ie, higher frequency designations)
for all of the nonspecific signs than did dogs that had
separation anxiety alone (Table 1). Correlation analy-
sis indicated that the following pairs of signs occurred
together more frequently than would be expected:
defecation and urination (P < 0.001), vocalization and
destruction (P = 0.012), and salivation and vocaliza-
tion (P = 0.018). On the other hand, the reaction to
noise and vocalization occurred together less frequent-
ly than would be expected (P = 0.005).
Of the 141 dogs included in the second part of the
study, 41 (20 males and 21 females) had separation
anxiety alone, 1 (female) had thunderstorm phobia
alone, 1 (female) had noise phobia alone, 7 (4 males
and 3 females) had separation anxiety and thunder-
storm phobia, 21 (15 males and 6 females) had separa-
tion anxiety and noise phobia, 10 (4 males and 6
females) had thunderstorm phobia and noise phobia,
and 60 (38 males and 22 females) had all 3 conditions.
Males and females did not differ significantly in regard
to the relative frequencies of these conditions (Fisher
exact test; P = 0.22); therefore, data for males and
females were combined for all subsequent analyses.
For these dogs, the observed frequency of separa-
tion anxiety alone was significantly higher than
expected if these conditions had been completely inde-
pendent (likelihood ratio χ
2
test; P < 0.05; Fig 1). In
addition, the observed frequencies of noise phobia
alone, of separation anxiety in conjunction with thun-
derstorm phobia, and of separation anxiety in con-
junction with noise phobia were all significantly lower
than expected if these conditions had been completely
independent. On the other hand, the frequency of all 3
conditions was significantly higher than expected if
these conditions had been completely independent.
Therefore, the null hypothesis that individual diag-
JAVMA, Vol 219, No. 4, August 15, 2001 Scientific Reports: Original Study 469
SMALL ANIMALS
Table 1—Frequency of various nonspecific clinical signs in dogs with separation anxiety alone, noise
or thunderstorm phobia alone, or separation anxiety and noise or thunderstorm phobia
Percentage of dogs
Reaction
Frequency Urination Defecation Destruction Vocalization Salivation to noises
Dogs with separation anxiety alone (n = 25)
Always 4 4 8 58 21 4
Mostly 24 8 64 25 8 4
Consistently 16 24 8 8 8 0
Seldom 12 12 4 0 0 4
Never 44 52 16 8 63 88
Dogs with noise or thunderstorm phobia alone (n = 5)
Always 0 0 0 0 0 100
Mostly 0 0 0 0 0 0
Consistently 0 0 0 0 0 0
Seldom 20 20 0 0 0 0
Never 80 80 100 100 100 0
Dogs with both (n = 21)
Always 14 10 29 63 43 86
Mostly 10 5 38 0 5 5
Consistently 10 10 5 0 14 9
Seldom 18 24 19 16 0 0
Never 48 52 9 21 38 0
Always = 100% of the time. Mostly = More than 60% of the time. Consistently = 40 to 60% of the time. Seldom = Less than
20% of the time. Never = 0% of the time.
Figure 1—Observed and expected numbers of dogs with sepa-
ration anxiety alone (SA), thunderstorm phobia alone (TP), noise
phobia alone (NP), separation anxiety and thunderstorm phobia
(ST), separation anxiety and noise phobia (SN), thunderstorm
phobia and noise phobia (TN), or separation anxiety, thunder-
storm phobia, and noise phobia (ALL) in a study of 141 dogs
with behavioral disorders. The distribution of expected numbers
was obtained by analyzing the results of 1,000 Monte Carlo sim-
ulations, assuming complete independence among SA, TP, and
NP. *Observed frequency was significantly different (
P
< 0.05)
from the expected frequency.
01_02_0059.QXD 10/12/2005 9:46 AM Page 469
noses (separation anxiety alone, thunderstorm phobia
alone, noise phobia alone, separation anxiety and
thunderstorm phobia, separation anxiety and noise
phobia, thunderstorm phobia and noise phobia, all 3
conditions) were independent was rejected (likelihood
ratio χ
2
test; P < 0.001). The conditional probability
that a dog with thunderstorm phobia would also have
separation anxiety was 0.86, whereas the probability
that a dog with separation anxiety would also have
thunderstorm phobia was only 0.52. Similarly, the
probability that a dog with noise phobia would also
have separation anxiety was 0.88, whereas the proba-
bility that a dog with separation anxiety would also
have noise phobia was only 0.63. Finally, the probabil-
ity that a dog with noise phobia would also have thun-
derstorm phobia was 0.76, and the probability that a
dog with thunderstorm phobia would also have noise
phobia was 0.90.
Because only 1 dog had thunderstorm phobia
alone and only 1 dog had noise phobia alone, these
dogs were excluded from analyses of the association
between nonspecific signs and individual diagnoses.
For these analyses, CMH χ
2
tests were used to test
whether owner responses to specific questions on the
screening questionnaire were associated with the actu-
al diagnosis. Results of the analyses of association indi-
cated that frequency of vocalization during an actual
absence of the owner (P = 0.008; Fig 2), frequency of
destructive behavior during an actual absence of the
owner (P = 0.002; Fig 3), frequency of reactions to
thunderstorms (P = 0.001; Fig 4), and frequency of
reactions to fireworks (P = 0.001; Fig 5) were not inde-
470 Scientific Reports: Original Study JAVMA, Vol 219, No. 4 , August 15, 2001
SMALL ANIMALS
Figure 2—Frequency of vocalization during the owner's absence
in dogs with separation anxiety alone (SA), separation anxiety
and thunderstorm phobia (SATP), separation anxiety and noise
phobia (SANP), thunderstorm phobia and noise phobia (TPNP),
or separation anxiety, thunderstorm phobia, and noise phobia
(All). Owner response categories were defined as never (1),
> 0% but < 40% of the time (2), 40 to 60% of the time (3),
> 60% but < 100% of the time (4), and 100% of the time (5).
Frequency of the various owner response categories varied sig-
nificantly among diagnostic disorders (CMH χ
2
test with 4 df; Q
= 13.73; n = 114;
P
= 0.008).
Figure 3—Frequency of destructive behavior during the owner's
absence in dogs with various anxiety-related behavioral disor-
ders. Frequency of the various owner response categories var-
ied significantly among diagnostic disorders (CMH χ
2
test with 4
df; Q = 16.44; n = 136;
P
= 0.002).
See
Figure 2 for key.
Figure 4—Frequency of reactions during thunderstorms in dogs
with various anxiety-related behavioral disorders. Frequency of
the various owner response categories varied significantly
among diagnostic disorders (CMH χ
2
test with 4 df; Q = 91.63;
n = 131;
P
= 0.001).
See
Figure 2 for key.
01_02_0059.QXD 10/12/2005 9:46 AM Page 470
pendent of the actual diagnosis. In addition, P values
for whether frequency of urination during an actual
absence of the owner (P = 0.110; n = 132) and fre-
quency of salivation during an actual absence of the
owner (P = 0.085; n = 100) were independent of the
actual diagnosis were close to the cutoff for signifi-
cance. However, frequencies of vocalization, destruc-
tive behavior, urination, defecation, and salivation dur-
ing virtual absences of the owner and frequency of
defecation during an actual absence of the owner were
not associated with the actual diagnosis. One clear pat-
tern identified was that dogs with thunderstorm pho-
bia in conjunction with noise phobia never vocalized,
salivated, urinated, or engaged in destructive behavior
during an actual absence of the owner.
Discussion
The finding in the present study that separation
anxiety alone occurred significantly more often than
would be expected and that noise phobia alone
occurred significantly less often than would be expect-
ed under random conditions supported the concept
that although these conditions share some nonspecific
signs, they are separate entities. The finding that sepa-
ration anxiety in conjunction with thunderstorm pho-
bia and separation anxiety in conjunction with noise
phobia occurred significantly less often than would be
expected were these conditions independent but that
thunderstorm phobia in conjunction with noise phobia
and all 3 conditions together (separation anxiety, noise
phobia, and thunderstorm phobia) occurred signifi-
cantly more often than would be expected were these
conditions independent supports 2 important conclu-
sions. First, this suggests that noise phobia and thun-
derstorm phobia are different from each other and
affect the frequency and intensity of related behaviors
in dogs with other conditions differently. Second, this
suggests that the interaction of multiple pathologic
responses to noise likely reflects an altered dysfunc-
tional underlying neurochemical substrate or is the
result of one. In previous studies,
15-17
the extent to
which dynamic interactions among comorbid condi-
tions shaped the expression of behavioral phenotypes
was reflective of differential responses to psychotropic
medications. The newer psychotropic medications
such as tricyclic antidepressants, selective serotonin re-
uptake inhibitors, and monoamine oxidase inhibitors
use the same second messenger systems and transcrip-
tion pathways that are used to develop cellular memo-
ry or to learn something via mediation and induction
of long-term potentiation in the CA1 region of the hip-
pocampus.
18-20
In the present study, owner responses to questions
about the signs exhibited by their dogs and the fre-
quency of those signs were not independent of the
diagnosis (separation anxiety alone, thunderstorm
phobia alone, noise phobia alone, separation anxiety
and thunderstorm phobia, separation anxiety and
noise phobia, thunderstorm phobia and noise phobia,
and all 3 conditions together). Again, this finding sup-
ports the conclusion that the interaction of multiple
pathologic responses to situations associated with dis-
tress (noise phobia, separation anxiety, thunderstorm
phobia) may reflect an altered dysfunctional underly-
ing neurochemical substrate or be the result of one and
that the various phenotypes are a result of this dys-
function.
18-21
The roles played by arousal and reactivity cannot
be ignored if we are to understand dogs with anxiety-
related conditions such as separation anxiety, noise
phobia, and thunderstorm phobia. Some dogs respond
more quickly or react more intensely to a given stimu-
lus than do other dogs. At some level, this hyper-reac-
tivity is probably truly pathologic and represents yet
another phenotypic manifestation of some neurochem-
ical variation associated with anxiety. If so, the more
frequently the dog reacts to the anxiety-provoking
stimulus, the worse and more rapid the response. At
some point, any exposure can then result in a full-
blown nongraduated anxious reaction in which true
panic may be involved. Accordingly, anticipation and
early treatment is critical for these individuals, again
supporting the concept that behavioral phenotypes and
underlying neurochemical responses are linked in a
dynamic way. Early intervention can only be accom-
plished by understanding the spectrum of signs exhib-
ited in related conditions.
Approximately 20 to 30% of human patients with
major depressive disorder have panic attacks. The role
of panic has only recently been investigated in dogs.
b,c,d
Panic attacks share many of the attributes of noise and
thunderstorm phobia. In dogs with combinations of
separation anxiety, thunderstorm phobia, and noise
phobia, the signs of each appear worse and more
intense than in dogs with only 1 of these conditions.
22
If such dogs follow the pattern common for humans
with comorbid diseases, the comorbidity may result in
JAVMA, Vol 219, No. 4, August 15, 2001 Scientific Reports: Original Study 471
SMALL ANIMALS
Figure 5—Frequency of reactions to fireworks in dogs with vari-
ous anxiety-related behavioral disorders. Frequency of the vari-
ous owner response categories varied significantly among diag-
nostic disorders (CMH χ
2
test with 4 df; Q = 40.60; n = 94;
P
=
0.001).
See
Figure 2 for key.
01_02_0059.QXD 10/12/2005 9:46 AM Page 471
a longer persistence of clinical signs and a less favor-
able overall outcome.
23,24
In human medicine, child-
hood separation anxiety is seen as an important
antecedent in adult panic disorder and is a common
undercurrent in the dreams of human patients with
panic disorders.
25
This suggests that in humans, as in
pets, separation anxiety can occur apart from panic dis-
orders and other phobia-related conditions, but in
patients with 2 or more conditions, the interaction
among these conditions is an important factor in the
assessment and treatment. If the phobic reactions to
noise and thunderstorms are related to panic in dogs,
such interactions are important.
In the present study, the probability that a dog
with noise phobia would also have separation anxiety
was high (0.88) and approximately the same as the
probability that a dog with thunderstorm phobia
would also have separation anxiety (0.86). However,
the probability that a dog with separation anxiety
would also have noise phobia (0.63) was higher than
the probability that a dog with separation anxiety
would also have thunderstorm phobia (0.52). These
data, combined with the finding that the probability a
dog would have noise phobia given that it had thun-
derstorm phobia (0.90) was not equivalent to the con-
verse (0.76), support the hypothesis that neurochemi-
cal responses to noise are different from those to thun-
derstorms. The unpredictability and uncertainty asso-
ciated with thunderstorms may have a role in shaping
the neurochemical and behavioral responses to these
anxiety-provoking situations, suggesting areas for
future exploration of anxiety-related responses in dogs.
Results of this study suggest that the probability of
having noise phobia in conjunction with separation
anxiety was higher than would be expected if these 2
conditions were not associated. This strongly suggests
that adverse reactivity to noises in general may predis-
pose dogs to separation anxiety and that unless veteri-
narians carefully question clients about their pets’
behavior, both of these conditions could remain undi-
agnosed. The extent to which early fearful behaviors
contribute to the development of separation anxiety is
unknown. Overly fearful human youngsters are at risk
for later emotional distress, including anxiety and
depression, suggesting that such associations should
be investigated in veterinary behavioral medicine.
The specific neuroanatomy of anxiety-related
responses involves the locus ceruleus (LC), the princi-
pal noradrenergic nucleus in the brain and a region
rich in serotonergic nuclei. Dysregulation of the LC
appears to lead to panic and phobias in humans.
26
The
LC directly supplies the limbic system and sends effer-
ent neurons to Barrington’s plexus in the intestines,
accounting for many of the gastrointestinal signs in
affected dogs. Human patients with true panic and
phobic responses are more sensitive to pharmacologic
stimulation and suppression of the LC than are healthy
control patients.
26-28
These same neuroanatomic regions
are implicated in patients with panic disorders.
29
In
patients with panic, parahippocampal blood flow (a
marker of neuronal activity), blood volume, and oxy-
gen metabolism, determined by means of positive-
emission tomography, are asymmetric during nonpan-
ic conditions,
30
possibly because of an increase in nor-
epinephrine output by the LC that, in turn, stimulates
parahippocampal over-responsiveness. The effects of
altering the underlying neurochemical response of a
substrate associated with affect and learning has pro-
found implications for outcome and severity of signs in
human patients with comorbid panic disorders,
30-36
and
a common genetic predisposition for such neurochem-
ical responses has been implicated as a mechanism for
their cooccurrence.
37-38
Our data support the concept
that such phenomena are operating in dogs with sepa-
ration anxiety in conjunction with thunderstorm or
noise phobia. The extent to which reactions to noise
may predispose dogs to other anxiety-related condi-
tions, whether the interaction of nonspecific signs and
individual conditions may have time penetrance, and
whether the longer a dog has been affected with 1 con-
dition may increase its risk for a more complex form of
that condition or for other anxiety-related conditions
can best be addressed in prospective studies.
Finally, any time a dog exhibits destruction, elimi-
nation, vocalization, salivation, pacing, or withdrawal
only during virtual absences of its owner, regardless of
the age of the dog, the veterinarian should ascertain the
pattern of the behaviors and decide whether they meet
the criteria for a diagnosis of separation anxiety.
Questions about these behaviors should be a routine
part of the history-taking process during all examina-
tions, as separation anxiety is among the most common
problem in dogs and is often missed during its early
stages. Early intervention is important. Separation anx-
iety can be successfully treated but can also lead to
euthanasia or relinquishment of the dog if left untreat-
ed.
3,39,40
Because rescued and re-homed animals are par-
ticularly at risk for separation anxiety, compared with
other dogs, early intervention may stop the problem
from progressing and break the relinquishment cycle.
40,e
Furthermore, in our experience, dogs with a history of
separation anxiety that has resolved are at risk for more
severe signs if the condition should recur.
a
Copies of the questionnaire are available from the corresponding
author on request.
b
Overall KL. Assessment of potential canine models for panic: results
of the lactate sensitivity test comparing dogs with separation anxi-
ety and those with idiopathic fear (abstr), in Proceedings. Am Vet
Soc Anim Behav Annu Meet 1999;4.
c
Overall KL, Dunham AE, Acland G. Assessment of potential canine
models for panic: results of the lactate sensitivity test comparing
dogs with separation anxiety and those with idiopathic fear (abstr),
in Proceedings. 2nd World Meet Ethol 1999;8–17.
d
Overall KL, Dunham AE, Acland G. Responses of genetically fearful
dogs to the lactate test: assessment of the test as a provocative
index and application in mechanistic diagnoses (abstr). Mol
Psychiatry 1999;4:S125.
e
Overall KL. Demographic factors don’t distinguish dogs with sepa-
ration anxiety (oral presentation). Novartis roundtable on separa-
tion anxiety, 1997.
References
1. Novartis roundtable on separation anxiety, Birmingham,
England, April 1997. Greensboro, NC: Novartis AH, 1999.
2. Voith VL, Borchelt PL. Separation anxiety in dogs. Compend
Contin Educ Pract Vet 1985;7:42–53.
3. McCrave EA. Diagnostic criteria for separation anxiety in
the dog. Vet Clin North Am Small Anim Pract 1981;21:247–256.
472 Scientific Reports: Original Study JAVMA, Vol 219, No. 4 , August 15, 2001
SMALL ANIMALS
01_02_0059.QXD 10/12/2005 9:46 AM Page 472
4. King J, Simpson B, Overall KL, et al. Treatment of separa-
tion anxiety in dogs with clomipramine. Results from a prospective,
randomized, double-blinded, placebo-controlled clinical trial. J Appl
Anim Behav Sci 2000;67:255–275.
5. Houpt KA, Honig SU, Reisner IR. Breaking the human-com-
panion animal bond. J Am Vet Med Assoc 1996;208:1653–1659.
6. Miller DD, Staats SR, Partlo C, et al. Factors associated with
the decision to surrender a pet to an animal shelter. J Am Vet Med
Assoc 1996;209:738–742.
7. Overall KL. Clinical behavioral medicine for small animals. St
Louis: Mosby Year Book Inc, 1997;510–520.
8. Servant D, Parquet PJ. Le trouble panique. Presse Med
2000;29:905–910.
9. Goldenberg I, White K, Yonkers K, et al. The infrequency of
“pure culture” diagnoses among the anxiety disorders. J Clin
Psychiatry 1996;57:528–533.
10. Klein DF. Panic and phobic anxiety: phenotypes, endophe-
notypes, and genotypes. Am J Psychiatry 1998;155:1147–1149.
11. Loftus J, Delisi LE, Crow TJ. Factor structure and familiar-
ity of first-rank symptoms in sibling pairs with schizophrenia and
schizoaffective disorder. Br J Psychiatry 2000;17:15–19.
12. Mojtabai R, Rieder RO. Limitations of the symptom-orient-
ed approach to psychiatric research. Br J Psychiatry 1998;173:
198–202.
13. Goldsmith HH, Lemery KS. Linking temperamental fearful-
ness and anxiety symptoms: a behavior-genetic perspective. Biol
Psychiatry 2000;48:1199–1209.
14. Tarantino LM, Bucan M. Dissection of behavior and psychi-
atric disorders using the mouse as a model. Hum Mol Genet 2000;9:
953–965.
15. Gross C, Santarelli L, Brunner D, et al. Altered fear circuits
in 5-HT1A receptor KO mice. Biol Psychol 2000;48:1157–1163.
16. Nopoulos PC, Ceilley JW, Gailis EA, et al. An MRI study of
midbrain morphology in patients with schizophrenia: relationship to
psychosis, neuroleptics, and cerebellar neural circuitry. Biol
Psychiatry 2001;49:13–19.
17. Spitzer H, Desimone R, Moran J. Increased attention
enhances both behavioral and neuronal performance. Science
1988;240:338–340.
18. Daniel H, Levenes C, Crépel F. Cellular mechanisms of cere-
bellar LTD. Trends Neurosci 1998;21:401–407.
19. Salter MW. Src, N-methyl-D-aspartate (NMDA) receptors,
and synaptic plasticity. Biochem Pharmacol 1998;56:789–798.
20. Trotti D, Danboldt NC, Volterra A. Glutamate transporters
are oxidant-vulnerable: a molecular link between oxidative and exci-
totoxic neurodegeneration. Trends Pharmacol Sci 1998;19:328–334.
21. Blier P, de Montigny C. Current psychiatric uses of drugs
acting on the serotonin system. In: Baumgarten HG, Gothert M, eds.
Serotonergic neurons and 5-HT receptors in the CNS. Berlin: Springer,
1997;227–275.
22. Overall KL. Neurobiology and neurochemistry of fear and
aggression, in Proceedings. North Am Vet Conf 1997;11:33–39.
23. Fones CSL, Pollack MH, Susswein L, et al. History of child-
hood attention deficit hyperactivity disorder (ADHD) features
among adults with panic disorder. J Affect Disord 2000;58:99–106.
24. Lueger RJ, Lutz W, Howard KI. The predicted and observed
course of psychotherapy for anxiety and mood disorders. J Nerv Ment
Dis 2000;188:127–134.
25. Kalin NH. The neurobiology of fear. Sci Am 1993;268(5):
94–101.
26. Charney DS, Heninger GR. Abnormal regulation of nora-
drenergic function in panic disorders. Arch Gen Psychiatry
1984;43:1042–1058.
27. Ko GN, Elsworth JD, Roth RH, et al. Panic-induced eleva-
tion of plasma MHPG levels in phobic anxious patients. Arch Gen
Psychiatry 1983;40:425–430.
28. Pyke T, Greenberg H. Norepinephrine challenge in panic
patients. J Clin Psychol 1986;6:279–285.
29. Reiman EM, Fusselman MJ, Fox PT, et al. Neuroanatomical
correlates of anticipatory anxiety. Science 1989;243:1071–1074.
30. Katon W, Vitaliani PP, Russo J, et al. Panic disorder: spec-
trum of severity and somatization. J Nerv Ment Dis 1987;49:716–722.
31. Reich J, Noyes R, Yates W. Anxiety symptoms distinguish-
ing social phobia from panic and generalized anxiety disorders. J
Nerv Ment Dis 1988;176:510–513.
32. Breier A, Charney DS, Heninger GR. Major depression in
patients with agoraphobia and panic disorder. Arch Gen Psychiatry
1984;41:1129–1135.
33. Pollack MH, Otto MW, Sabatino S, et al. Relationship of
childhood anxiety to adult panic disorder: correlates and influence
on course. Am J Psychiatry 1996;153:376–381.
34. Klein DF. Panic disorder and agoraphobia: hypothesis hot-
house. J Clin Psychiatry 1996;31:193–199.
35. Charney DS, Grillon C, Bremner DJ. The neurobiological
basis of anxiety and fear: circuits, mechanisms, and neurochemical
interactions (part I). The Neuroscientist 1998;4:35–44.
36. Free NK, Wingert CN, Whitman RM. Separation anxiety in
panic disorder. Am J Psychiatry 1983;150:595–599.
37. Scherrer JF, True WR, Xian H, et al. Evidence for genetic
influences common and specific to symptoms of generalized anxiety
and panic. J Affect Disord 2000;57:25–35.
38. Smoller JW, Tsuang MT. Panic and phobic anxiety: defining
phenotypes for genetic studies. Am J Psychiatry 1998;155:
1152–1162.
39. Takeuchi Y, Houpt KA, Scarlett JM. Evaluation of treat-
ments for separation anxiety in dogs. J Am Vet Med Assoc 2000;
217:342–345.
40. Salman MD, Hutchison J, Ruch-Gallie R, et al. Behavior rea-
sons for relinquishment of dogs and cats to 12 shelters. J Appl Anim
Welfare Sci 2000;3:93–106.
JAVMA, Vol 219, No. 4, August 15, 2001 Scientific Reports: Original Study 473
SMALL ANIMALS
01_02_0059.QXD 10/12/2005 9:46 AM Page 473
