Elsevier

Journal of Veterinary Behavior

Volume 16, November–December 2016, Pages 28-35
Journal of Veterinary Behavior

Research Paper
Separation anxiety in dogs: What progress has been made in our understanding of the most common behavioral problems in dogs?

https://doi.org/10.1016/j.jveb.2016.02.005Get rights and content

Abstract

Separation anxiety has long been recognized as an anxiety-related disorder in dogs that is only observed in the owner's real absence or perceived absence. Over the past four decades, this condition has been the most commonly discussed disorder in published studies, but etiology, treatment, and prevention remain elusive. A review of the literature indicates a lack of comparative studies. Little progress has been made to build a new understanding beyond the clinical symptomology. Future research employing more rigorous designs and systematically building upon a clearly defined research questions is needed to advance our knowledge of this most common behavioral problem in dogs.

Introduction

Companionship with a dog can provide great joy, and its experience is pleasantly fulfilling (Archer, 1997). According to the statistics, approximately 34%-36.5% of households own dogs in the USA, Canada, and Australia (American Veterinary Medical Association, 2012, Canadian Animal Health Institute, 2014, The Australian Companion Animal Council publication, 2010), following 24% in the UK (Pet Food Manufacture's Association, 2014), and 15.8% in Japan (Japan Pet Food Association, 2013). Any benefit of living with a dog can quickly diminish when pet dogs engage in undesirable behaviors that damage the human–animal bond. These dogs often end up relinquished and may or not find a new home. The behavior literature shows that the primary reason that an individual relinquishes a pet dog is usually related to some sort of behavioral problem (Salman et al., 2000, Shore, 2005).

Denenberg et al. (2005) found that the caseloads in three behavior specialty clinics in three different countries (the USA, Canada, and Australia) reported that aggression was the most common reason for canine referrals, followed by anxiety- and fear-related problems, such as separation anxiety or noise phobia. Bamberger's and Houpt (2006) retrospective 10-year study using the medical record of animal behavior clinic at Cornell University found a similar trend, where the most commonly reported problem was aggression, followed by separation anxiety. In the past four decades, separation anxiety has been the most commonly discussed disorder in published studies of experimental research and retrospective research in the fields of applied animal behavior and veterinary behavior. As a result, in the USA, there are two Food and Drug Administration–approved medications to treat this problem. Despite the number of published studies on this condition, there is still a persistent uncertainty regarding the terminology, etiology, treatment, and prevention, hence the need for this review.

This review, although comprehensive, is by no means exhaustive. The review will summarize the research published in peer-reviewed journals from the PubMed, Scopus, and Web of Science databases from 1970 to 2014. The following key words were used to retrieve articles regarding the current knowledge about separation-related behavioral problems in pet dogs: separation anxiety; separation-related behavior; separation-related problems; separation-related distress; and separation-related disorders. Dissertations, conference abstracts, or articles published in a language other than English were omitted.

In the terminology of canine behavior disorders, anxiety is defined as an emotion of apprehension to an anticipated danger or threat. The signs are recognized through physiological (e.g., autonomic arousal) and behavioral (e.g., hypervigilance) responses, but the stimulus of anxiety is oftentimes unidentified. If the stimulus is identifiable and present, the emotional response to the stimulus is defined as fear. When fear response is excessive and out of context, it is maladaptive and harmful to an individual that is defined as phobia (Landsberg et al., 2013). In this review, unless specifically defined in context, separation anxiety will be used interchangeably with the other terminologies that were commonly used in the articles reviewed here. Separation anxiety will be defined as an anxiety-related disorder in dogs whose signs are only observed in the owner's absence or perceived absence (Horwitz, 2000, Sherman and Mills, 2008).

In the 1970s, separation anxiety in pet dogs was described for the first time in the literature from a “scientific” point of view (Tuber et al., 1974, Voith, 1975, Borchelt and Voith, 1982, Voith and Borchelt, 1985). Until then, the general public perceived the behavior problems during the owner's absence as “a lack of obedience” or “the result of spoiling the dog.” When a “scientific view” about separation-related behavior problems was brought into the literature, supportive references were based on primarily on social species of birds (Hess, 1964), monkeys (Harlow and Zimmermann, 1959), or humans (Bowlby, 1977). When a dog study was cited, it was typically based on puppies' behavior in laboratory situations (Eliott and Scott, 1961). At that time, little information was available about separation-related behavior problems in privately owned adult dogs (Borchelt and Voith, 1982). In those early publications, the clinical signs noted were described as separation anxiety, which was manifested as one or a combination of signs (Table 1). A time line of the occurrence of clinical signs was emphasized as a differentiating feature between separation anxiety and other behavior problems. For separation anxiety, the clinical signs (Table 1) were to have been observed within minutes after being left alone or the signs of anxiety (e.g., panting, pacing, trembling, etc.) were to begin before the owner's departure. The other common characteristics of canine separation anxiety mentioned (Borchelt and Voith, 1982, Voith and Borchelt, 1985) were (1) the dog had constant contact with the owners when home, (2) the onset of the problem would often resume upon leaving the dog alone for a long vacation, (3) the anxiety can be triggered by a dramatic lifestyle change, such as divorce or the death of a family member, (4) the anxiety may develop in dogs with thunderstorm phobia or dogs left in a strange place (e.g., new apartment), and (5) separation anxiety occurs because dogs are highly social species and become attached to the people with whom they live. Voith and Borchelt (1985) stated that they found many dogs exhibit separation anxiety in a new home after being adopted from humane societies or found as strays, although consequence or a causal factor between the previous history and the development of separation anxiety in their paper was unknown.

These early reports consisted of descriptive studies with data collected from clinical cases, or conference abstracts from studies using small sample sizes with no control population. Both types of studies made it difficult to accumulate data that supported common risk factors or causal factors. For example, McCrave (1991) analyzed 55 canine cases from the Animal Behavior Clinic at the University of Pennsylvania and compared clinical data regarding the animal's age, breed, and sex and the owner's interaction with dogs with (n = 18) or without (n = 37) separation anxiety. The results showed no significant difference between two groups regarding the owners' interaction. No correlation was found between separation anxiety and other types of fear (such as fear of thunderstorms, unfamiliar adults, etc.). There was no significant difference in the dogs' sex and age between the two groups. Mixed breed dogs were overrepresented in the separation anxiety group, although this is likely a confounding variable, as more mixed breed dogs from a shelter and more dogs from a shelter have separation anxiety (McCrave, 1991, Serpell, 1995). Without using an adequate control group, a challenge in this type of study (Takeuchi et al., 2001), it is difficult to interpret the predisposition factors from this type of report.

Until the late 1990s, all the articles were clinical descriptions based on case observations, where no study-based data were obtained. Between 1999 and 2015, three clinical review articles and 28 research articles on separation anxiety in adult dogs were published in peer-reviewed journals (Table 2). Although this behavioral problem occurs only during the owner's absence (therefore, nobody watches the dog when it exhibits the signs), the majority of studies (18 of 28) were conducted by a subjective rating, such as a questionnaire completed by the owners, with or without an owner interview or clinician's assessment. Four of 18 studies were retrospective studies (Flannigan and Dodman, 2001, Takeuchi et al., 2001, McGreevy and Masters, 2008, Storengen et al., 2014) and reported the risk factors associated with separation anxiety. The eight studies of remaining 10 studies analyzed additional objective information using video/sound recordings of the affected dogs (Lund and Jørgensen, 1999, Blackwell et al., 2006, Parthasarathy and Crowell-Davis, 2006, Palestrini et al., 2010, Butler et al., 2011, Konok et al., 2011, Scaglia et al., 2013, Cannas et al., 2014). Half the studies (13 of 28) during this period focused on the treatment outcome after the diagnosis of separation anxiety, including randomized, double-blind, and placebo-controlled clinical trials with two Food and Drug Administration–approved medications (e.g., clomipramine and fluoxetine) (King et al., 2000, King et al., 2004, Sherman et al., 2007, Landsberg et al., 2008).

Collectively, several studies provided possible risk factors for separation anxiety, such as the animal's sex, breed, acquired resources, and family structure or owner's lifestyle; however, no consistent agreement has been found. The interpretation of the results is limited by the different sampling methods, definition of terminology, and definitions of the affected and unaffected groups. For example, in the review of three studies that tried to identify risk factors of separation anxiety using the information from questionnaires, two studies used clinical records of a behavior clinic retrospectively so that all dogs were evaluated in the clinic as well (Flannigan and Dodman, 2001, Storengen et al., 2014), whereas one study collected information directly from the general public by mail (McGreevy and Masters, 2008). Two studies showed that male dogs exhibited a higher frequency of separation anxiety (McGreevy and Masters, 2008, Storengen et al., 2014); yet in a third study, no sex-specific difference was observed (Flannigan and Dodman, 2001). Reports of higher breed-specific frequency were observed for golden retrievers, English springer spaniels, and English cocker spaniels from the study by Flannigan and Dodman (2001), whereas cocker spaniels, schnauzers, and Dachshunds were noted in the study by Storengen et al. (2014). There was no breed-specific association observed in the study by McGreevy and Master (2008). However, it is worth noting that these three studies were conducted in three different countries: the United States, Norway, and Australia. In addition, two retrospective studies were case-controlled studies, and their separation anxiety cases were diagnosed at the behavior clinic, whereas the control group of these studies included patients of the behavior clinics that were not diagnosed separation anxiety (Flannigan and Dodman, 2001, Storengen et al., 2014). Because of no epidemiologic data available in dogs, the study of Storengen et al. (2014) managed to use nationwide registration data from the Norwegian kennel club and data from large-sized small animal clinic in the region when they compare the results regarding sex and breed, whereas, the study of McGreevy and Masters (2008) was held only questionnaire based, and the owners scored their dog's separation-related distress in the questionnaire by five-point scale from 0; the sign is not present to 5; the sign is severe. The examples of these distress behaviors were shaking, excessive salivation, restlessness, and howling or chewing. Then, the responses were analyzed to identify the risk factors to predict an elevation of separation-related distress scores.

Although each study used relatively large samples [n = 200 in affected cases vs. n = 200 in control cases in Flannigan and Dodman (2001); n = 215 in affected cases vs. n = 3071 in Storengen et al. (2014) and n = 690 in McGreevy and Masters (2008)], it is quite possible that these geographic and data collection variables and definition of affected dogs should have contributed to the inconsistencies found in these reported results. When etiology and risk factors were pursued, using clinical population in behavior disorders is likely to have a pitfall of secondary outcome of the clinical signs and confounding variables as all dogs had some sort of behavior disorders. Although using the direct information from general public in McGreevy and Masters (2008) would fill up a bias of clinical population, however, subjective interpretation of regular owners has limited an accuracy of the definition so that if other information of these dogs in the survey was not combined to identify the target population (i.e., separation-related distress) and nonaffected population, the result should be carefully interpreted.

The clinical signs of separation anxiety are not specific and the etiologies appeared to be multifactorial, which led to several different terms being used by clinicians and researchers in the published literature. Terms such as separation anxiety, separation-related problems, separation reactions, separation-related distress, separation anxiety syndrome, or separation anxiety disorders were used, with no sufficient definition that could subsequently restrict comparisons with different studies. As is common in other behavioral problems with an unrevealed etiology, various authors have advocated symptom-based approaches to data collection to avoid inappropriate interpretations of animal's motivation (McGreevy and Masters, 2008, Palestrini et al., 2010). If only clinical outcomes as evaluated subjectively and the owner's satisfaction reports were included, we can conclude that we have made some positive progress (Cottam et al., 2008). Despite the complex etiologies that include dog–human interaction and context-related factors that have been described to influence attachment, it appears that owners more often complied with easy behavioral modification suggestions than more specialized advice (Takeuchi et al., 2000). Blackwell et al. (2006) also showed that a standard, universal behavioral, modification program was quite effective (75% of the 15 dogs improved during the first 6 weeks), and no significant difference was found in the treatment outcome compared to a tailored treatment plan.

Four decades have passed since the first publication on separation anxiety; however, further critical scientific evaluations are needed to address the etiologic factors of this common behavioral problem to advance our understanding. The following portion of the article will discuss the studies that were reviewed and sorted into content areas to select categories that represent the most inconsistent outcomes or debated areas. The categories selected to serve as a framework for this review are behavioral patterns with the time of separation, the presence of hyperattachment, and underlying fear/anxiety, including comorbidity of noise phobia.

Anxiety is considered to be adaptive response to enable avoidance of a perceived or anticipated threatening stimulus (Blackwell et al., 2013). If separation anxiety is one of adoptive responses to social isolation, one can assume that the dog will eventually be able to cope to the situation. Surprisingly, few studies used video analysis to capture the behavior of pet dogs in their home environment. Lund and Jorgensen's (1999) publication was the first article discussing the use of video recordings of 20 dogs with separation anxiety. In their findings, they proposed that the possible underlying motivation of this condition was frustration and fear. On the basis of the video analysis, 19 of 20 dogs in their study were whining that was considered to attention-soliciting behavior during the separation. Therefore, the authors thought that it indicated that fear was involved in the dog's emotional state being separated from the owner. It was also discussed that in some dogs, the level of frustration disinhibits displacement activities such as object play or precedes exploratory behavior that leads to destructive behavior like chewing and tearing apart. Frustration can also arouse the dog so that it lowers the threshold of response to external stimuli and can increase the level of activity and the vocalization. Prolonged frustration may lead to fear; therefore, the symptoms of autonomic responses (e.g., salivation, hyperventilation, and elimination) and destruction at door way were considered as an escape responses triggered by fear. The authors also confirmed the clinical information reported in previous articles (Borchelt and Voith, 1982, Voith and Borchelt, 1985), suggesting that the peak intensity of behavior signs such as vocalization and destruction occurred shortly after the owner's departure. All behavior signs decreased over time (4-hour period) unless the dog was rearoused by external stimuli. The article also showed a cyclical behavior pattern, where the separation-related behavior signs described previously, which were elicited by fear and frustration, occurred every 23-28 minutes.

Another study on dogs with separation anxiety that were recruited from a behavior clinic found that the clinical signs, such as vocalization and destruction, were observed immediately after the owner's absence (Palestrini et al., 2010). This study did not find a significant decrease in these behaviors after the passage of time, specifically 20-40 minutes after departure. However, barking and environment orientation tended to decrease over time, whereas panting had a tendency to increase over time.

Although these two studies only observed dogs with separation anxiety (Lund and Jørgensen, 1999, Palestrini et al., 2010), Rehn and Keeling (2011) video recorded dogs without separation anxiety at home and found no difference in their behavior. The recordings were made on three different occasions at 30 minutes, 2 hours, and 4 hours during the owner's absence. These studies suggested that the specific behavior pattern in dogs with separation anxiety included a peak intensity of distress immediately after the owner's departure, and some but not all behaviors tended to decrease over time (Lund and Jørgensen, 1999, Palestrini et al., 2010). However, the dogs without separation anxiety did not have a peak intensity of distress; therefore, the authors concluded that the dog behaviors during separation period did not change over time (Rehn and Keeling, 2011).

A study by Konok et al. (2011) supports this interpretation. In this study, dogs with and without separation anxiety were studied in a laboratory setting, although the definition of affected or not affected separation anxiety was based on the owners' self-report through questionnaires. It leaves the interpretation of this study limited as no other assessment such as video analysis of the dog's behavior at home or evaluation at behavior clinic were used. The researchers manipulated the duration of separation from the dog's owner—1, 3, and 5 minutes—to determine the effects on separation anxiety. During separation, they found that the activity level in the dogs without separation anxiety decreased over time (i.e., there was a difference between the duration of separation), but this did not occur in the dogs with separation anxiety. Although these results appeared to contradict the previous study, a laboratory setting (i.e., a strange place) was considered to activate separation anxiety more than a familiar place, such as the dog's home. The authors concluded that peak and decreased activity in the laboratory demonstrates the adaptability of dogs without separation anxiety, whereas persistent activity, regardless of the separation duration, in dogs with separation anxiety indicated their maladaptive behaviors in a strange place.

Scaglia et al. (2013) conducted a video behavioral analysis of dogs without separation anxiety and compared the results to another study of dogs with separation anxiety (Palestrini et al., 2010). Although they did not compare the behavioral changes for >90 minutes of absence, they did find that the unaffected dogs exhibited passive behaviors, such as sleeping or resting, significantly more often than the affected dogs. The authors concluded that the results suggested that most of the unaffected dogs that showed passive behavior during the owner's absence were understandable as their other behavior signs showed they were rather relaxed and less active in the familiar environment. Exploration behavior was reduced or absent when the dog did not perceive a new stimulus including the owner's absence. Rehn and Keeling (2011) found the similar result in their study with dogs that were not affected by separation anxiety.

These studies give insight into behavior patterns with regard to the time of separation. All video-based studies supported a specific pattern of behavioral signs associated with separation anxiety. The peak intensity of the anxiety-related behavioral signs occurred shortly after the owner's departure, and these same signs were not observed in dogs without separation anxiety. If the study was held at an unfamiliar place like a laboratory, the unaffected dogs experienced a peak of activity that quickly decreased, whereas affected dogs remained active. The conclusions from these studies suggested that separation anxiety reflected maladaptive behaviors in the affected dogs. These results validated the clinical observations of the behavior patterns commonly described in clinical reports.

Hyperattachment and overattachment are typical descriptions in the literature regarding dogs with separation anxiety or dogs exhibiting constant contact-seeking behaviors (e.g., needy or insecure dogs). There is no consistent terminology used to describe this type of behavior sign. Some clinical reviews mentioned hyperattachment as a causal factor of separation anxiety, when referring to social species (e.g., birds, monkeys) and dogs in a laboratory setting (Borchelt and Voith, 1982, Voith and Borchelt, 1985, McCrave, 1991). Simpson (2000), however, indicated that hyperattachment could exist in dogs with or without separation anxiety, and it is not a prerequisite or requirement for a diagnosis of separation anxiety.

Appleby and Pluijmakes (2003) proposed categorizing contact-seeking behaviors into three groups and suggested that an individual could move their behavioral pattern between groups in response to specific events. These three groups were group A: those that do not develop autonomy because of primary hyper-attachment; group B: those that transfer their dependence to one or more stimuli, usually social stimuli, through need or an increase in the stimuli's salience and/or availability; and group C: those that learn to depend on a range of stimuli without any narrow set of social or environmental stimuli becoming exceptionally salient. In their proposal, attachment or bonding was a high level of conditioned dependency that supported the dog's emotional stability (homeostasis). According to their theory, separation anxiety is derived from disrupted homeostasis; therefore, the dogs may try to reestablish it by creating proximity to their maintenance stimuli, which might be a salient human companion.

One clinical trial attempted to use hyperattachment as one of the inclusion criterion for diagnosing separation anxiety, defined as three specific behavior signs and was considered dogs must have all the three signs to suggest having a high degree of hyperattachment. These three signs were A: usually follow the owner about the house and tried to maintain physical contact with the owner (within a distance of 1 m); B: becomes distressed with increasing distance of separation from the owner; C: becomes distressed when the owner prepares to leave the home and greets the owner excessively on his/her return. Six dogs did not meet the first criteria (signs A). However, when the hyperattachment data were examined via two parameters, “dog follows the owner around the house” and “dog initiates interaction with the owner”, changes in the hyperattachment scores did not, by themselves, reflect an optimal improvement of separation anxiety. The authors of the study concluded that any relationship between hyperattachment and separation anxiety remained elusive (King et al., 2000). The conflicting findings on the presence of hyperattachment resulted in some studies including behaviors of “hyperattachment” (e.g., follow the owner from room to room, sleep close to the owner) as a definitive sign of separation anxiety (Gaultier et al., 2005, Storengen et al., 2014 and Karagiannis et al., 2015), whereas other researchers indicated that it was neither a definitive nor important sign for diagnosing the disorder (Overall, 2000, Sherman and Mills, 2008).

The retrospective studies of Flannigan and Dodman's (2001) and Storengen et al. (2014) showed that dogs with separation anxiety were more likely to follow their owners around the house compared to unaffected dogs. The former study also reported that the dogs with separation anxiety excitedly greeted their owners for over 2 minutes (Flannigan and Dodman, 2001). A survey study investigating the risk factors for separation anxiety suggested that the condition might not be purely an attachment-based problem, given that they found no significant difference in hyperattachment between the affected and unaffected dogs. However, subjective ratings are questionable in this type of study because the findings reflect the owner's perception and interpretation (McGreevy and Masters, 2008). Interestingly, although few in number, studies using video analysis also reported no significant differences in the type of proximity-seeking behaviors between dogs with and without separation anxiety (Parthasarathy and Crowell-Davis, 2006).

Konok et al. (2011) noted that the term hyperattachment did not exist in human developmental psychology, and therefore, it is not yet defined in terms of behavior. Their video analysis study also did not find that the dogs with separation anxiety exhibited more or more intense greeting behavior or showed preference for the owners' objects when they were left behind. Instead, Konok et al. (2011) found that the affected dogs could not easily be calmed down when reunited with their owner. Therefore, the authors concluded that these dogs may be insecure in their relationship with the owner and proposed to use a term from human developmental psychology, “ambivalent” relationship, for these dogs.

Although some survey studies reported excessive greeting by dogs with separation anxiety (Flannigan and Dodman, 2001), it is not clear whether the underlying motivation of the excessive greeting reflects hyperattachment or insecure attachment. Few video analysis studies have focused on greeting behaviors in dogs with separation anxiety. Rehn and Keeling (2011) reported that the duration of the owners' absence, but not the owners' behavior, affected greeting behaviors in dogs without separation anxiety. Future studies regarding greeting behavior in dogs with and without separation anxiety are warranted.

The presence of hyperattachment in dogs with separation anxiety remains questionable because there was no definitive evidence that it exists. Dogs with separation anxiety were not easily calmed down when reunited with the owner; however, excessive contact-seeking behaviors do not necessarily indicate hyperattachment or overattachment but may instead be a manifestation of other underlying motivations. This view is also supported by a recent study that focused on aged dogs and their attachment behavior (Mongillo et al., 2013). Their findings will be further discussed later in this article.

In an early clinical report, separation anxiety and thunderstorm or noise phobia was thought to be a comorbid condition (Voith and Borchelt, 1985). Because the clinical signs of separation anxiety and noise-related phobia overlap with each other, it is not surprising that owners may not be able to tell if their dog was affected by one or both conditions when completing the clinical evaluation questionnaire or participating in a behavior study. Several studies reported a similar positive association between separation anxiety and noise-related phobia (e.g., thunderstorms, fireworks) (Flannigan and Dodman, 2001, McGreevy and Masters, 2008, Cannas et al., 2014, Storengen et al., 2014) with a 30%-50% frequency of comorbidity (Flannigan and Dodman, 2001, Cannas et al., 2014, Storengen et al., 2014). However, a negative association between separation anxiety and noise-related phobia was also reported in other studies (McCrave, 1991, Blackwell et al., 2013, Herron et al., 2014). Further research addressing these inconsistent results and the frequency of comorbidity with other anxiety-related problems is needed because all the data relied on the owner's subjective report, resulting in gaps between the owner's interpretation and the dog's actual behavioral signs. So far, one study designed to address to fill up this gap. The study of Overall et al. (2001) was composed of two parts. First part involving 51 dogs was to determine the frequency of nonspecific clinical signs with separation anxiety alone, noise-related phobia alone, or a combination of these conditions. This part is served as a pilot study as the behavior questions were standardized in the next part of the study. In the second part of the study, the covariation of nonspecific signs of separation anxiety, thunderstorm phobia, and noise phobia was evaluated, and the comorbidity of these conditions was determined. All 141 dogs that had a noise reaction and/or separation anxiety presented to the behavior clinic in the course of a calendar year were evaluated using the same clinical and questionnaire evaluations based on the data from the 51 dog in the first part of the study. The result showed that 41 had separation anxiety alone, 1 had nonstorm phobia, 1 has storm-based noise phobia alone, 7 had separation anxiety and storm phobia, 21 had separation anxiety and noise phobia, 10 had noise and storm phobia, and 60 had all three conditions. That was suggested based on the statistical analysis that individual diagnoses (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 three conditions) were not independent. The authors discussed the further hypothesis of neurochemical and behavioral responses to these anxiety-provoking situations. Blackwell et al. (2013) noted the importance of a methodological approach using questionnaires and suggested that owners seemed to recognize their dog's behavioral responses to noises (e.g., bark, tremble/shake, hide, etc.) when an occurrence of specific behavioral responses was asked through a structured interview. However, the same owners did not count their dogs were fearful when the frequency of fearfulness was asked by a questionnaire as there was a gap in the population between two questions (i.e., behavioral signs to noises vs. the subjective interpretation of behaviors to noises [= dog's emotional state]). It suggested asking questions based on the owner's subjective interpretation might not reflect accurate frequency of the dog's behavior signs. Recently, Temesi et al. (2014) tried to develop a general questionnaire to facilitate the standardization of measurements of fear-related behavior in dogs. Their results said four traits were obtained (neuroticism, dog-related fear, human-related fear, and separation-related behavior) by principal component analysis, which provides measures for important aspects of fear-related behavior. Interestingly, their results showed that separation-related behavior was a separate trait from the other aspects of fear-related behavior. This may be why we have inconsistent findings regarding the comorbidity between separation anxiety and other types of fear/anxiety. Robust evidence underlying several types of fear/anxiety conditions in dogs with separation anxiety is not available.

Section snippets

Conclusions and future directions

The physiological and psychological benefits of pet ownership in modern society are well reported (Friedmann et al., 1980, Headey, 1999, Christian et al., 2013). However, as more and more households have full-time workers, dogs are more often being left alone. The impact can be seen by the increasing prevalence of separation anxiety, as evidenced by a U.S. telephone marketing survey reporting canine separation anxiety in 14%-17% of households (Sherman and Mills, 2008) and 20%-40% of caseloads

Acknowledgments

The author is grateful to the organizing committee (Karen Overall, Mark Vaudin, Adrian Burde, Paula Boyden, Clarissa Baldwin, Rosemary Smart, Aimee Llewellyn, and Andrew Higgins) on the meeting of Canine behaviour and cognition: evolution, genetics and applications for breeding and performance for their invitation to present on this topic and to Dr. Karen Overall for helpful comments on this article.

References (75)

  • J.N. King et al.

    Results of a follow-up investigation to a clinical trial testing the efficacy of clomipramine in the treatment of separation anxiety in dogs

    App. Anim. Behav. Sci.

    (2004)
  • V. Konok et al.

    The behavior of the domestic dog (Canis familiaris) during separation from and reunion with the owner: a questionnaire and an experimental study

    App. Anim. Behav. Sci.

    (2011)
  • G.M. Landsberg et al.

    Behavior problems in geriatric pets

    Vet. Clin. North Am. Small Anim. Pract.

    (2005)
  • G.M. Landsberg et al.

    Effectiveness of fluoxetine chewable tablets in the treatment of canine separation anxiety

    J. Vet. Behav.: Clin. Appl. Res.

    (2008)
  • J.D. Lund et al.

    Behaviour patterns and time course of activity in dogs with separation problems

    App. Anim. Behav. Sci.

    (1999)
  • E.A. McCrave

    Diagnostic criteria for separation anxiety in dog

    Vet. Clin. North Am. Small Anim. Pract.

    (1991)
  • P.D. McGreevy et al.

    Risk factors for separation-related distress and feed-related aggression in dogs: additional findings from a survey of Australian dog owners

    App. Anim. Behav. Sci.

    (2008)
  • M. Mendl et al.

    Dogs showing separation-related behaviour exhibit a “pessimistic” cognitive bias

    Curr. Biol.

    (2010)
  • P. Mongillo et al.

    Does the attachment system towards owners change in aged dogs?

    Physiol. Behav.

    (2013)
  • N. Ogata et al.

    Objective measurement of fear-associated learning in dogs

    J. Vet. Behav.: Clin. Appl. Res.

    (2006)
  • N. Ogata et al.

    Brain structural abnormalities in Doberman pinschers with canine compulsive disorder

    Prog. Neuropsychopharmacol. Biol. Psychiatry

    (2013)
  • K.L. Overall

    Natural animal models of human psychiatric conditions: assessment of mechanism and validity

    Prog. Neuropsychopharmacol. Biol. Psychiatry

    (2000)
  • K.L. Overall et al.

    Understanding the genetic basis of canine anxiety: phenotyping dogs for behavioral, neurochemical, and genetic assessment

    J. Vet. Behav.: Clin. Appl. Res.

    (2006)
  • K.L. Overall et al.

    Genetics and behavior: a guide for practitioners

    Vet. Clin. North Am.: Small. Anim. Pract.

    (2014)
  • C. Palestrini et al.

    Video analysis of dogs with separation-related behaviors

    App. Anim. Behav. Sci.

    (2010)
  • V. Parthasarathy et al.

    Relationship between attachment to owners and separation anxiety in pet dogs (Canis lupus familiaris)

    J. Vet. Behav.: Clin. Appl. Res.

    (2006)
  • T. Rehn et al.

    The effect of time left alone at home on dog welfare

    App. Anim. Behav. Sci.

    (2011)
  • E. Scaglia et al.

    Video analysis of adult dogs when left home alone

    J. Vet. Behav.: Clin. Appl. Res.

    (2013)
  • B.L. Sherman et al.

    Canine anxieties and phobias: an update on separation anxiety and noise aversions

    Vet. Clin. North Am.: Small. Anim. Pract.

    (2008)
  • L.M. Storengen et al.

    A descriptive study of 215 dogs diagnosed with separation anxiety

    App. Anim. Behav. Sci.

    (2014)
  • Y. Takeuchi et al.

    Differences in background and outcome of three behavior problems of dogs

    J. Am. Vet. Med. Assoc.

    (2001)
  • A. Temesi et al.

    Measuring fear in dogs by questionnaires: an exploratory study toward a standardized inventory

    App. Anim. Behav. Sci.

    (2014)
  • K. Tiira et al.

    Reliability and validity of a questionnaire survey in canine anxiety research

    App. Anim. Behav. Sci.

    (2014)
  • S. Vermeire et al.

    Neuro-imaging the serotonin 2A receptor as a valid biomarker for canine behavioural disorders

    Res. Vet. Sci.

    (2011)
  • S. Vermeire et al.

    Serotonin 2A receptor, serotonin transporter and dopamine transporter alterations in dogs with compulsive behaviour as a promising model for human obsessive-compulsive disorder

    Psychiatry. Res.

    (2012)
  • D.L. Walker et al.

    Role of the bed nucleus of the stria terminalis versus the amygdala in fear, stress, and anxiety

    Eur. J. Pharmacol.

    (2003)
  • U.S. Pet Ownership and Demographics Sourcebook

    (2012)
  • Cited by (67)

    • Separation-related Problems and Their Interaction with Physical Disease

      2024, Veterinary Clinics of North America - Small Animal Practice
    • Pampered pets or poor bastards? The welfare of dogs kept as companion animals

      2022, Applied Animal Behaviour Science
      Citation Excerpt :

      The most commonly reported clinical signs of separation-related problems are vocalization, elimination, and destructive behavior (Ogata, 2016; Amat et al., 2020;). However, dogs can also display other signs in relation to separation from their owner(s), such as behavioral indicators of stress or anxiety (panting, pacing, trembling, etc.), depression (lethargy, listlessness), or physiological signs (salivation, anorexia, vomiting, diarrhea, etc.) (Ogata, 2016; Amat et al., 2020; de Assis et al., 2020). These signs can more easily be overlooked by the owner, but are nonetheless indicators of an underlying negative affective state (Mendl et al., 2010; Karagiannis et al., 2015).

    • Pet dogs home alone: A video-based study

      2021, Applied Animal Behaviour Science
      Citation Excerpt :

      In a retrospective study of medical records of a veterinary clinic Bamberger and Houpt (2006) found a caseload of 14.4% for separation anxiety, which turned out to be the most common behaviour disorder after problems with aggressive behaviour. In a recent review on the topic Ogata (2016) emphasizes, separation anxiety being the most frequently discussed disorder in published studies of the past four decades, but still its aetiology and prevention remains elusive. The term ‘separation anxiety’ implies anxiety as the underlying motivation of the behavioural symptoms.

    View all citing articles on Scopus
    View full text