Literature Review: Behavior Modification for Canine Separation Anxiety
Separation anxiety (SA) and separation-related behavior (SRB) are significant welfare concerns for companion dogs (Cannas et al., 2014; Lund & Jorgensen, 1999). SA has been shown to be associated with negative affective states (Scaglia et al., 2013), raise measurable cortisol (Shin & Shin, 2016), increase anxiety-related behaviors (Palestrini, 2010; Karagiannis et al., 2015) such as salivation, elimination in inappropriate areas, hyperventilation, appetite issues, diarrhea, vomiting, pacing, and repetitive behaviors (Takeuchi, 2000), and can even cause self-injurious behavior (Overall, 2000). Thus SA, while falling on a spectrum of severity, can cause poor welfare in all three welfare areas: health, behavior, and physiological/affective state of the dog.
Moreover, SA is not a rare phenomenon. In fact, it can be quite common, with some estimates showing that SRB complaints make up between 20 and 40% of cases seen at behavior practices (Borchelt & Voith, 1982; Flannigan & Dodman, 2001) and up to 30% in newly adopted dogs from shelters (Blackwell et al., 2016). Because of the potential seriousness of the impact to the dog’s welfare, and how the behavior can impact the life of the human caregiver (noise complaints; destroyed furniture, doorframes, or windows; inappropriate elimination; etc.), it is no surprise that SRB contributes to the relinquishment of dogs to shelters, with some estimates putting the percentage of dogs relinquished for SRB as high as 30% (Diesel et al., 2010; New et al., 1999).
My purpose in this article is to review the current literature on separation anxiety, focusing on the effectiveness of behavior modification techniques. It is my hope that understanding more about what the research says about the causes of SA and the best ways to address the problem—and about where there are gaps in the literature—can help behavior consultants develop more focused interventions and feel more confident in approaching what is often a particularly emotionally charged issue for clients.
What is separation anxiety?
Overall (2000) defines separation anxiety as “a condition that is manifest as extreme distress when the dog is separated from his owner(s). The presence of other animals in the household does not alleviate the distress. The condition is variable in its triggers and presentation.” King et al. (2000) defined it this way: “[SA] is characterized by signs of destructiveness, inappropriate elimination (defecation, urination) and vocalization when an affected dog is left alone or is separated from its attachment figure, usually its owner.” Flannigan and Dodman (2001) defined SA as “severe distress when an individual is distanced from other group members, but in canine behavioral terminology, this term is most often restricted to dogs that become upset when separated from their owner.” Appleby & Pluijmakers (2004) described it as “problematic behavior motivated by anxiety that occurs exclusively in the owner’s absence or virtual absence.”
In other words, it is generally accepted that SA is triggered by the separation from an attachment figure (owner) or when left alone, and the associated behaviors with separation anxiety are destructiveness, inappropriate elimination and vocalization, as well as physiological and affective distress.
While the definition for SA can seem simple, the reality of diagnosing SA is more complex. This is because the behaviors listed above may not be caused by anxiety (Turner, 1997). For example, inappropriate elimination could be caused by an untrained dog, a dog not receiving an appropriate amount of physical and mental stimulation could become destructive, or a dog with poor impulse control or basic training may vocalize excessively.
In addition, Lund or Jorgensen (1999) suggested that separation anxiety is sometimes caused by frustration and arousal and is not always related to fear and anxiety. For the purposes of this paper, it is assumed that we are dealing specifically with dogs that display the behaviors above due to an anxious experience related to an owner’s departure or separation. However, Lung & Jorgensen’s (1999) frustration proposal should be explored further, as should the work of Appleby and Pluijmakers (2004), who proposed three different motivations (hyperattachment to primary caregiver, change in social group or living environment, or recent phobic event or noxious experience). Having a better understanding of the different causes of separation anxiety may result in more effective diagnosis and treatment. In the research recommendation section below, I discuss one possible approach to test frustration versus fear-related vocalization.
Predictive factors and causes (etiology)
Sex and neuter status
Flannigan and Dodman (2001) retroactively looked at 400 cases from a behavior clinic and similar studies, and found that 60% of dogs presented for SRB were male (King et al., 2000; Podberscek et al., 1999; Storengen et al., 2014; Takeuchi, 2001). The reason for this uneven sex distribution is unknown, and it is important to note that some authors (Lund et al., 1996) found that the split between male and female dogs was closer to even. Neutering has not been found to contribute to SRB (Flannigan & Dodman, 2001; Takeuchi, 2001), but this correlation was found to be stronger in a recent study (Storengen et al., 2014).
There does seem to be evidence pointing to mixed-breeds being more likely than purebreds to display SRB (McCrave, 1991; Storengen, 2014; Takeuchi et al., 2001; Voith, 1994), but Voith (1993) found that purebred dogs were more likely than mixed-breeds to present symptoms of SA. However, it is difficult to know if mixed breeds are more likely to end up in a shelter than purebred dogs, and therefore are more likely to experience rehoming, which is suspected to be one of the potential causes of SA (Borchelt & Voith, 1982).
Recently, genetic factors have also been investigated. Zapata et al. (2016) discovered that known IGF1 and HMGA2 loci variants for small body size are associated with separation anxiety. While this is an extremely new area of research, it has the potential to provide a lot of input into the potential causes of SA.
Other behavioral issues
SA’s comorbidity with fear is often discussed, with some authors reporting that 30-50% of dogs with SA issues also have noise or thunderstorm phobias (Flannigan & Dodman, 2001; Storengen et al., 2014). However, other authors have found a negative association (Herron et al., 2014; McCrave, 1991). Ogata (2016) discussed these conflicting reports and indicated that more systematic evaluation of this comorbidity needs to be done as previous reports have largely relied on owner questionnaires, which have accuracy concerns.
Personality of the owner and dog
The personality of the owner and dog have also been investigated by looking at The Big Five personality traits in humans as well as scoring the personalities of the dogs (Konok et al., 2015). Owners who had a higher score on attachment avoidance were more likely to have a dog suffering from SRB. In addition, dogs scoring higher on the neuroticism scale were more prone to develop SRB. The assumption is that attachment-avoidant owners are less responsive to the needs of their dog, which can result in an insecure attachment. However, more research in this area is needed before drawing any firm conclusions.
Attachment style of the dog
Hyperattachment to the owner has been proposed by many authors as a critical component of separation anxiety. Flannigan and Dodman (2001), McCrave (1991) and Appleby and Pluijmakers (2004) all argued that this was a factor in at least some of the SA-related cases. However, Parthasarathy and Crowell-Davis (2006) devised an experiment to test this assumption in a lab as well as a home setting, and hyperattachment behaviors were no more likely in SA dogs when compared to a control group of non-SA dogs.
Amount of obedience training
There are mixed findings with regard to how obedience training is related to separation anxiety. Voith (1992) and Takeuchi et al. (2001) did not find an association between SA and obedience training. However, Takeuchi et al. did note that dogs with SA were less likely to lie down at the request of the owner and were generally trained less often. Flannigan and Dodman (2001) did find that fewer SA dogs had obedience training than a control group consisting of aggressive dogs, but the makeup of the control group may have been a confounding factor. Clark & Boyer (1993) found that owners who did obedience training with their dogs were significantly less likely to have a dog that displayed SRB.
Where the owners acquired the dog
The source of the dog may be related to SRB (at least in terms of resistance to treatment). Takeuchi (2000) found that SA dogs obtained from a shelter were less likely than dogs obtained from a breeder, pet store, or friend to respond to treatment. This agreed with the findings of McCrave (1991) and Serpel (1995). Flannigan and Dodman (2001) did not find as strong of a relationship as the previously mentioned studies and noted that only one study (Voith, 1994) directly investigated changing homes as a factor in SA, but did not find significant results.
Where the family lives
Household makeup has also been investigated. Takeuchi et al. (2001) found that dogs in apartments/urban environments were significantly more likely to display SA. However, it is unclear if it is of higher prevalence in these environments or simply reported more frequently since the behavior can impact neighbors and others in the community (vocalization). Flannigan and Dodman (2001) also found a significant association with households with only one person. However, this has yet to be found elsewhere.
Factors that are not associated with SA
Age of referral to behavior professional for treatment (Flanigan & Dodman, 2001; Voith 1994), spoiling activities (Flannigan & Dodman, 2001; McCrave, 1991), and the presence of another dog in the home (Flannigan & Dodman, 2001; McBride, 1995; Voith, 1994), have generally not been found to be associated with SA.
Our review agrees with the findings of Ogata (2016), that much work remains on the etiology of SA. Reports that are largely based on owner questionnaires are producing findings that conflict with each other. A more systematic approach, involving more video and direct observation, may be needed to improve knowledge in this area. It also may be possible that dogs of various motivations are being categorized as having SA due to similar SRB, when perhaps there may be better, more specific groupings.
Separation anxiety treatment: Behavioral medication
When a dog is diagnosed, or assumed to be suffering from SA, treatment falls into two general categories: pharmaceutical intervention and behavior modification. While the focus of this paper is on behavior modification techniques, it is important to examine the literature for effectiveness of medication prior to discussing behavior medication, as both are frequently used together.
Starting around the late 1990s, the use of an SSRI (selective serotonin reuptake inhibitors) like fluoxetine, or a tricyclic antidepressant such as clomipramine, has been investigated as a pharmaceutical intervention for SA. Most the research offers solid evidence for the effectiveness of these medications in alleviating SA symptoms when administered in conjunction with behavior modification.
Clomipramine has undergone numerous studies (Cannas et al., 2014; King et al., 2000, 2004; Poderberscek et al., 1999); almost all have shown clomipramine to be effective when used in conjunction with behavior modification. However, Poderberscek et al. (1999) found that clomipramine and behavior modification was no more effective than behavior modification and a placebo. However, more recent studies (Cannes et al., 2014, King et al., 2000, 2004) have demonstrated the effectiveness of this drug with behavior modification. Poderberscek et al. (1999) states that their study reaffirms the importance of behavior modification in treatment of SA.
Similarly, fluoxetine has been investigated and has been effective at significantly reducing SRB when used in conjunction with behavior modification (Ibanez & Anzole, 2009; Karagiannis et al., 2015; Simpson et al., 2007). Landsberg et al. (2008) investigated the effectiveness of using the drug alone (without behavior modification). While it was effective, the authors point out that it does not seem as effective as when it was combined with behavior modification, such as in Simpson et al.’s study (2007).
The behavior modification used in these studies was not standardized, but there were common themes, which will be discussed in the behavior modification section below. Schwartz (2003) hypothesizes that behavior modification teaches a dog how to respond so that eventually medication can be weaned off. However, no research supports this claim.
It is also important to note that medications such as clonidine (Ogata & Dodman, 2011) and trazodone (Gruen & Sherman, 2008) have both been effective when combined with either clomipramine or fluoxetine. However, how much more effective these medications are remains unclear.
Given the strong evidence supporting the use of pharmaceutical intervention in conjunction with behavior modification, dogs presenting with SA symptoms should be referred to a veterinary behaviorist or a veterinarian if there are time constraints on treatment (for example, if a dog will be relinquished to shelter due to eviction because of vocalization) or if there is a significant welfare concern.
Separation anxiety treatment: Behavioral modification
As mentioned above, behavior modification appears to be more effective when used in conjunction with certain behavior medication and has also been shown to be effective as a stand-alone treatment. However, after reviewing 14 research articles that tested treatment plans (Borchelt & Voith, 1982; Butler et al., 2011; Cannes et al., 2014; Cottam et al., 2008; Geurtsen et al., 2015; Ibanez & Anzole, 2009; Karagiannis, 2015; Kim et al., 2010; King et al., 2010, 2014; Poderscek et al., 1999; Shin & Shin, 2016; Simpson, 2007; Takeuchi, 2000), 18 different behavior modification recommendations were made. In order of prevalence, these were (total count in parenthesis):
- Eliminate all punishment (7)
- Low-key departures (6)
- Desensitization (5)
- Relaxation/stay training (4)
- Sitting for rewards such as petting, food, access to outdoors (4)
- Counterconditioning (3)
- Unpredictable departures (3)
- Independence training (3)
- Olfactory stimuli (3)
- Ignore dog on arrival (3)
- Auditory stimuli (2)
- Increase exercise (2)
- DAP (dog appeasing pheromone) collar (1)
- Compression vest (1)
- Interactive video game for the dog (1)
- Canine body language education (1)
- Discontinue reinforcement of fear behaviors (1)
- Routine schedule planning (1)
Takeuchi et al. (2000) noted that when owners were given more than five instructions, they saw significantly less effectiveness with their treatment program. For this reason, it is important to systematically test behavior modification recommendations so that the most effective instructions are the ones that are recommended. In addition, Takeuchi et al. (2000) and Ogata (2016) also mentioned that owners tend to display high compliance for easy, low-effort recommendations (avoidance of punishment, use of food toys, etc.) but are less compliant with more complex recommendations such as systematic desensitization to departures.
As just discussed, when looking at the total number of times a behavior modification recommendation was made in successful treatment programs, it is important to understand which recommendations are both effective and easy to implement. When examining the top ten recommended procedures, they can be divided into higher effort and lower effort.
- Relaxation/stay training
- Unpredictable departures
- Independence training
- Calm departures
- Sit for rewards
- Olfactory stimuli
- Ignoring dog on arrival
Thus, since the literature suggests that we keep recommendations to five or fewer instructions, behavior professionals should not add behavior modification exercises just because they require little effort. We really cannot determine from the literature which of these behavior modification pieces is most effective because they are usually recommended together. Future research should examine and systematically test these strategies so that the most efficient exercises can be recommended with confidence without overwhelming the dog owner.
This literature review identifies two broad research areas that need further investigation. First, because of the largely conflicting research on predictive factors and causes of separation anxiety, it is possible that our criteria for determining separation anxiety are too wide, resulting in too many dogs being grouped together when they have different underlying motivations (Ogata, 2016). As noted earlier, in addition to anxiety motivating the behavior, Lung & Jorgensen (1999) proposed that frustration may be the cause in some cases and Appleby and Pluijmakers (2004) proposed three different causes for SA: hyperattachment to the primary caregiver, change in social group or living environment, or recent phobic event or noxious experience. Add this to the dogs who have other motivations such as boredom, lack of mental or physical stimulation, and poor training or impulse control, and we are potentially grouping too many dogs together with different motivations that may require entirely different treatment.
Two possible ways to scientifically approach this is to either use a computer program that can help discriminate among canine vocalizations (Yin & McCowan, 2004) or use canine behavior experts to complete a free-choice profiling exercise to understand if there are different clusters of dogs under the SA classification (Walker et al., 2010). If so, being able to tease apart different motivations and classifications of SA dogs could help with diagnosis and potentially offer more efficient treatment options based on a more accurate categorization.
Second, even though we know that behavior modification is effective (see above), it is unclear which components of the behavior modification work best. Again, in looking at 14 studies that successfully tested treatments, 18 different recommendations were made. Because owners are significantly more likely to be successful with five or fewer instructions, it is important to know which of these interventions is having the desired effect, with more weight given to lower-effort interventions (Takeuchi et al., 2000). In particular, low-effort and frequently recommended instructions such as eliminating punishment, calm departures, sit for rewards, counterconditioning, olfactory stimuli and ignoring the dog on arrival should be examined before the frequently recommended higher-effort instructions of relaxation/stay training, desensitization, unpredictable departures, and independence training.
Separation anxiety is a serious welfare concern for companion dogs (Cannas et al., 2014; Lund & Jorgensen, 1999), due both to the impact on the dog’s welfare and the high occurrence of people seeking help at behavior practices (Borchelt & Voith, 1982; Flannigan & Dodman, 2001). While there are a fair number of research articles on separation anxiety, their findings are often contradictory (Ogata, 2016). In terms of predictive factors and causes, the lack of consensus may suggest that too many dogs with different motivations (anxiety, frustration, boredom, etc.) are being lumped together. Future research should examine how dogs categorized as having SA may differ in their motivation. Two possible ways of examining this are to either use a computer program that has been used to discriminate among dog vocalization in the past (Yin & McCowan, 2004) or by using free-choice profiling (Walker et al., 2010). Taking a simple anxiety versus frustration approach may be an appropriate starting point. Ideally, if it were possible to discriminate among different types of barks when a dog is alone, that could be indicative of different motivations, which may then require different treatments.
What we do know, however, is that behavioral medication such as fluoxetine or clomipramine combined with behavior modification can be highly effective at reducing and eliminating SA. However, what is unclear from the research is what components of a behavior modification plan are the most effective. It would be beneficial for the research community to start systematically testing the most frequently recommended low-effort interventions to determine which are effective. Only when we know that those do not work as expected should we examine the higher-effort recommendations (unless the higher effort solution is suspected of providing a high success rate).
Finally, it is important for behavior professionals to be aware of the high effectiveness of the combination of behavior medication and behavior modification when treating separation anxiety. When cases present a serious time constraint (owner may be evicted due to vocalization) or if there is a serious welfare concern (severe distress or self-injurious behavior), those dogs should be referred to a veterinary behaviorist or a veterinarian who works with behavior issues. Only when there is not a time constraint and when the impact to the dog’s welfare is not a major concern, should behavior modification alone be considered. When recommending behavior modification strategies, it is important to remember that giving five or fewer instructions and using low-effort exercises will likely increase compliance and desired results.
Blackwell, E. J., Casey, R. A., & Bradshaw, J. W. (2016). Efficacy of written behavioral advice for separation-related behavior problems in dogs newly adopted from a rehoming center. Journal of Veterinary Behavior: Clinical Applications and Research, 12,
Butler, R., Sargisson, R. J., & Elliffe, D. (2011). The efficacy of systematic desensitization for treating the separation-related problem behaviour of domestic dogs. Applied Animal Behaviour Science, 129(2-4), pp. 136-145.
Cannas, S., Frank, D., Minero, M., Godbout, M., & Palestrini, C. (2010). Puppy behavior when left home alone: Changes during the first few months after adoption. Journal of Veterinary Behavior: Clinical Applications and Research, 5(2), pp.94-100.
Cannas, S., Frank, D., Minero, M., Aspesi, A., Benedetti, R., & Palestrini, C. (2014). Video analysis of dogs suffering from anxiety when left home alone and treated with clomipramine. Journal of Veterinary Behavior: Clinical Applications and Research, 9(2), pp.50-57.
Clark, G. I., & Boyer, W. N. (1993). The effects of dog obedience training and behavioural counselling upon the human-canine relationship. Applied Animal Behaviour Science, 37(2), pp.147-159.
Cottam, N., Dodman, N. H., Moon-Fanelli, A. A., & Patronek, G. J. (2008). Comparison of remote versus in-person behavioral consultation for treatment of canine separation anxiety. Journal of Applied Animal Welfare Science, 11(1), pp.28-41.
Diesel, G., Brodbelt, D., & Pfeiffer, D. U. (2010). Characteristics of relinquished dogs and their owners at 14 rehoming centers in the United Kingdom. Journal of Applied Animal Welfare Science, 13(1), pp.15-30.
Geurtsen, A., Lamers, M. H., & Schaaf, M. J. (2015, September). Interactive Digital Gameplay Can Lower Stress Hormone Levels in Home Alone Dogs—A Case for Animal Welfare Informatics. In International Conference on Entertainment Computing, pp. 238-251. Spr
Gruen, M. E., & Sherman, B. L. (2008). Use of trazodone as an adjunctive agent in the treatment of canine anxiety disorders: 56 cases (1995–2007). Journal of the American Veterinary Medical Association, 233(12), pp.1902-1907.
Herron, M. E., Lord, L. K., & Husseini, S. E. (2014). Effects of preadoption counseling on the prevention of separation anxiety in newly adopted shelter dogs. Journal of Veterinary Behavior: Clinical Applications and Research, 9(1), pp.13-21.
Ibáñez, M., & Anzola, B. (2009). Use of fluoxetine, diazepam, and behavior modification as therapy for treatment of anxiety-related disorders in dogs. Journal of Veterinary Behavior: Clinical Applications and Research, 4(6), pp.223-229.
Karagiannis, C. I., Burman, O. H., & Mills, D. S. (2015). Dogs with separation-related problems show a “less pessimistic” cognitive bias during treatment with fluoxetine (ReconcileTM) and a behaviour modification plan. BMC Veterinary Research, 11:80.
King, C., Buffington, L., Smith, T. J., & Grandin, T. (2014). The effect of a pressure wrap (ThunderShirt®) on heart rate and behavior in canines diagnosed with anxiety disorder. Journal of Veterinary Behavior: Clinical Applications and Research, 9(5), pp
King, J. N., et. al, (2004). Results of a follow-up investigation to a clinical trial testing the efficacy of clomipramine in the treatment of separation anxiety in dogs. Applied Animal Behaviour Science, 89(3-4), pp.233-242.
King, J. N., et. al, (2000). Treatment of separation anxiety in dogs with clomipramine: Results from a prospective, randomized, double-blind, placebo-controlled, parallel-group, multicenter clinical trial. Applied Animal Behaviour Science, 67(4), pp.255-2
Konok, V., Kosztolányi, A., Rainer, W., Mutschler, B., Halsband, U., & Miklósi, Á. (2015). Influence of owners’ attachment style and personality on their dogs’ (Canis familiaris) separation-related disorder. PLoS ONE, 10(2): e0118375
Landsberg, G. M., Melese, P., Sherman, B. L., Neilson, J. C., Zimmerman, A., & Clarke, T. P. (2008). Effectiveness of fluoxetine chewable tablets in the treatment of canine separation anxiety. Journal of Veterinary Behavior: Clinical Applications and Rese
Lund, J. D., Agger, J. F., & Vestergaard, K. S. (1996). Reported behaviour problems in pet dogs in Denmark: age distribution and influence of breed and gender. Preventive Veterinary Medicine, 28(1), pp.33-48.
Ogata, N. (2016). Separation anxiety in dogs: What progress has been made in our understanding of the most common behavioral problems in dogs? Journal of Veterinary Behavior: Clinical Applications and Research. Pages
Note: This paper was not available at the time of publication.
Ogata, N., & Dodman, N. H. (2011). The use of clonidine in the treatment of fear-based behavior problems in dogs: An open trial. Journal of Veterinary Behavior: Clinical Applications and Research, 6(2), pp.130-137.
Parthasarathy, V., & Crowell-Davis, S. L. (2006). Relationship between attachment to owners and separation anxiety in pet dogs (Canis lupus familiaris). Journal of Veterinary Behavior: Clinical Applications and Research, 1(3), pp.109-120.
Podberscek, A. L., Hsu, Y., & Serpell, J. A. (1999). Evaluation of clomipramine as an adjunct to behavioural therapy in the treatment of separation-related problems in dogs. The Veterinary Record, 145(13), pp.365-369.
Shin, Y., & Shin, N. (2016). Evaluation of effects of olfactory and auditory stimulation on separation anxiety by salivary cortisol measurement in dogs. Journal of Veterinary Science, 17(2), pp.153-158.
Voith, V. L., E. McCrave, and N. Lung, (1994). Profiles of dogs with separation anxiety and treatment approaches.
Voith, V. L., Goodloe, L., Chapman, B., & Marder, A. (1993). Comparison of dogs presented for behavior problems by source of dog. Paper presented at annual meeting of the American Veterinary Medical Association, Seattle, WA.
Voith, V. L., Wright, J. C., & Danneman, P. J. (1992). Is there a relationship between canine behavior problems and spoiling activities, anthropomorphism, and obedience training? Applied Animal Behaviour Science, 34(3), pp.263-272.
Brian Burton is the Co-Founder of Instinct Dog Behavior & Training LLC in New York, New York. He has worked with hundreds of owners and dogs with aggression, fear, and anxiety issues. His primary role with Instinct is Behavior Consulting. Brian is currently working toward his MA in Animal Behavior & Conservation at Hunter College. Brian grew up fearful of dogs but overcame his fear and found a lifelong passion through shelter and rescue group volunteer work. He owns a 1-year old rescue rat terrier mix, Joey, currently in training. Brian’s previous rat terrier mix, Sammy, CD, CGC, RAE2 finished as the top-ranking mixed breed in the 2013 and 2014 AKC Rally National Championships.