Animals · Explainer

Separation anxiety in dogs and what behaviour research shows

Separation anxiety in dogs refers to distress that appears in the context of separation, but behaviour research shows that the broader category of separation-related behaviour is not one uniform disorder. Dogs may vocalise, pace, destroy objects, eliminate indoors, pant, salivate or become unusually still, and the same outward sign can arise from different emotional or medical causes.

This article compares observational studies, owner surveys, a controlled behaviour trial, a diagnostic framework and veterinary welfare guidance. It focuses on what can be inferred from timing, video, prevalence and risk-factor data, and where a veterinarian or veterinary behaviourist is needed to distinguish separation distress from boredom, confinement problems, noise sensitivity or illness.

Dog lying beside a window with toys nearby in a quiet living room
A dog rests beside a window with toys on the rug Editorial illustration

Separation-related behaviour, often shortened to SRB, is an umbrella description for behaviour that appears when a dog is left without access to human company. The label is useful because it describes the context without assuming that every dog is experiencing the same emotion. Sargisson's 2014 review used the more familiar term separation anxiety, but also noted that the research base included several forms of separation-related problems rather than one tightly defined disorder.

A large diagnostic-framework study made that problem explicit. De Assis et al. (2020) analysed a broad questionnaire in 2,757 dogs with at least one separation-related sign, then used complete data from 762 dogs to identify seven behavioural components and four broader clusters. The authors argued that outwardly similar problems can plausibly involve different combinations of social panic, frustration, elimination, reactivity, noise sensitivity and boredom, so the phrase separation anxiety should not be treated as a biological diagnosis by itself.

That distinction matters for welfare as well as terminology. Destruction or barking can be inconvenient for a household, but their meaning depends on timing, accompanying body signs and whether the same behaviour occurs when people are present. The same context-first principle is useful across animal behaviour research, because visible behaviour is evidence to interpret rather than a diagnosis on its own.

Behaviour patterns when a dog is left alone

Video studies show why the first minutes after departure are especially informative. Palestrini et al. (2010) filmed 23 dogs with separation-related problems for 20 to 60 minutes in their homes; most showed at least one sign within 10 minutes, with mean latency of 3.25 minutes for vocalising and 7.13 minutes for destruction. Across the recorded period, vocalising occupied 22.95% of observed time on average, while panting, passivity and environmental orientation were also common.

The visible signs are diverse, and some are easy to miss after the owner returns. Sherman and Mills (2008), alongside ASPCA and RSPCA welfare guidance, describe patterns including vocalisation, destructive behaviour around exits, indoor elimination, pacing, salivation and other signs of distress during separation. Pre-departure agitation and unusually intense greeting can add context, but neither is specific enough to establish the cause by itself.

  • Vocalisation such as barking, whining or howling that is linked to absence rather than ordinary environmental triggers.
  • Destructive or escape-directed behaviour, particularly around doors, windows or confinement points.
  • Autonomic or motor signs such as panting, salivation, trembling, pacing or repetitive movement.
  • Elimination indoors, reduced eating or marked passivity that appears during separation.
  • Changes before departure or immediately after reunion that support, but do not prove, a separation-linked pattern.

Sherman and Mills (2008) emphasise that diagnosis requires both a behavioural history and consideration of medical and behavioural differential diagnoses. A young dog may chew because of exploratory behaviour, barking may follow outside noises, toileting can reflect incomplete house training or illness, and frantic escape can be driven by confinement itself. Separation-related distress becomes more plausible when the pattern is reliably tied to departure or loss of access to a particular person and is supported by what happens on video.

Prevalence and risk factors depend on how studies define the problem

Estimates of separation-related behaviour vary because studies count different signs, recruit different populations and use different reporting methods. The same uncertainty affects risk-factor findings: an association observed in one sample may not hold under another case definition (Salonen et al., 2020; Harvey et al., 2022; Dale et al., 2024).

Prevalence depends on the case definition

There is no single prevalence figure that can be applied to all pet dogs. Salonen et al. (2020) surveyed 13,715 Finnish dogs across 264 breeds and classified 5% as having a high level of separation-related behaviour under a relatively strict questionnaire rule. Harvey et al. (2022), using a UK cohort followed through changing pandemic routines, reported 22.1% with SRB at the February 2020 baseline and 17.2% at follow-up; among 1,187 dogs with no SRB signs in February that were left alone in October, 117 (9.9%) showed new signs.

A 2024 Generation Pup analysis illustrates how a broader case definition can produce a much higher estimate. Dale et al. (2024) classified 68 of 145 six-month-old puppies, or 46.9%, as showing at least one defined SRB only when left alone; pacing was the most frequent individual sign at 14.5%. These figures are not contradictory in a simple sense, because age, recruitment, observation, threshold and the number of behaviours counted were different.

Study

Population and design

Reported separation-related finding

Interpretation limit

Salonen et al. 2020

13,715 Finnish pet dogs; owner questionnaire

5% met the study's high SRB definition

Strict threshold; self-selected owner survey

Harvey et al. 2022

1,807 UK dogs followed across 2020 routine changes

22.1% at baseline, 17.2% at follow-up; 9.9% (117/1,187 initially SRB-free dogs left alone in October) showed new signs

Pandemic-era cohort and owner report

Dale et al. 2024

145 puppies at six months; longitudinal owner data

46.9% met a broad SRB case definition

Small exploratory puppy sample; broad definition

Risk factors are associations, not causes

Older studies identified several possible associations, but they are not consistent enough to support a simple profile of the dog that will develop the problem. Flannigan and Dodman (2001) reported links with some household and acquisition variables in clinical cases, and Sargisson's 2014 review concluded that findings for sex, shelter origin and early experience were mixed. Later work has reinforced that uncertainty: differences in case definition and sampling can reverse an apparent association.

Comorbidity is clearer than many individual risk factors. In Salonen et al. (2020), dogs in the separation-related group were 4.1 times more often classified as hyperactive or impulsive, 3.4 times more often inattentive and 2.8 times more likely to be fearful than dogs outside that group. Those are relative associations within one questionnaire dataset, not proof that one trait causes another or that a single underlying disorder explains all of them.

The Generation Pup study also found several early-life and owner-interaction variables associated with SRB at six months, including large odds ratios for some training, sleep and return-response categories. Dale et al. (2024) explicitly described the work as exploratory and correlational, with a final analytical sample of 145 puppies and wide confidence intervals for several estimates. Those results are useful for generating hypotheses, but intervention studies are needed before they can establish preventive rules.

Clinical assessment separates separation distress from similar problems

A clinical assessment starts with the pattern rather than one dramatic symptom. The timing of vocalising, destruction, elimination or pacing in relation to departure is compared with the dog's behaviour when people are present, and a veterinarian can review health, age, sensory change, pain, urinary or gastrointestinal problems and other medical explanations. A sudden change in an older dog, for example, has a different evidential context from a long-standing pattern that repeatedly begins after departure, because physical illness, sensory decline or altered cognition can change both activity and elimination. Medical differentials belong within pet health assessment, not as an afterthought to a behaviour label.

Video is especially valuable because owners cannot directly observe a dog while they are away. Palestrini et al. (2010) showed that filming can capture the timing and sequence of separation-related behaviours, while de Assis et al. (2020) emphasised that separation-related problems are heterogeneous and require careful differential assessment. The RSPCA similarly notes that less obvious signs may be missed when assessment relies only on property damage or neighbour reports.

The differential assessment also considers boredom, juvenile exploration, noise sensitivity, barrier frustration and learned patterns that occur outside separation. De Assis et al. (2020) found behavioural clusters consistent with more than one underlying process, which is why a veterinarian or veterinary behaviourist may ask about triggers, routines, confinement, prior experiences and the dog's behaviour across different settings. A diagnosis should not be inferred from a single accident, damaged door or episode of barking; the interpretation instead depends on contrasts across contexts.

  • A close and repeatable link between departure and the onset of signs supports a separation-related interpretation more strongly than an isolated event.
  • Behaviour that also occurs while people are present can point towards additional triggers, such as noise sensitivity, frustration, incomplete house training or medical problems.
  • Video can show whether distress begins immediately, builds gradually or follows an outside event, helping separate owner absence from coincidental environmental stimulation.
  • Sudden onset, especially when accompanied by changes in mobility, elimination, appetite, sleep or orientation, increases the importance of a veterinary medical assessment.

What behaviour management research can and cannot establish

Behavioural management research examines whether separation-related signs can improve, while welfare guidance explains why those signs matter to the dog. Evidence of benefit does not remove the need to assess the individual causes of distress or recognise the limitations of available treatment studies (Blackwell et al., 2006; Sargisson, 2014).

Evidence for behaviour modification

Controlled evidence for behavioural treatment exists, but the literature is smaller and older than the prevalence literature. Blackwell, Casey and Bradshaw (2006) evaluated a generic behaviour-modification programme over 12 weeks; 56% of owners in the treated groups reported significant improvement and another 25% reported slight improvement, with owner reports broadly supported by video. Most untreated dogs continued to show a similar degree of separation-related behaviour over the same period.

The result supports the principle that behaviour can change, but it does not prove that one protocol fits every dog. Sargisson (2014) reviewed studies of systematic desensitisation and counterconditioning and concluded that behavioural approaches were central to management, while also highlighting inconsistent research on predisposing factors and the practical difficulty of owner adherence. Modern diagnostic work adds another caution: if different dogs reach the same outward signs through different emotional processes, management may need to be individualised.

For that reason, this evidence is better read as support for structured, monitored behaviour work than as a do-it-yourself sequence of steps. RSPCA guidance recommends veterinary assessment when separation-related distress is suspected, and Sherman and Mills (2008) place diagnosis within a wider medical and behavioural assessment. The controlled trial did not establish that the same management approach works equally across distinct behavioural subtypes.

Why reducing distress is a welfare goal

A dog that repeatedly panics, paces, vocalises or tries to escape during absence is experiencing more than a household nuisance. RSPCA guidance treats separation-related behaviour as a welfare concern and recommends veterinary assessment when distress is suspected, while ASPCA guidance warns that escape attempts can cause injury. The welfare objective is therefore to understand and reduce the negative state, not merely suppress the visible behaviour.

Stress research in other animal contexts also shows why an outward stress response should not be equated with a single diagnosis or long-term outcome. The same distinction appears in work on stress during animal translocation, where acute responses and chronic welfare consequences are separated conceptually. In dogs, the practical implication is similar: behaviour, context, duration and recovery all matter when judging severity.

What is established and what remains uncertain

Several conclusions are reasonably secure. Separation-related behaviour is defined by context, often begins soon after departure, and can include both obvious and subtle signs; video adds information that owner recall alone can miss. Survey studies also show that separation-related signs co-occur with other behavioural traits and can change when household routines change, as Harvey et al. (2022) observed during 2020. Across these studies, the most reproducible information comes from the relationship between behaviour and absence rather than from any single symptom, breed characteristic or household feature, which is why definitions and observation methods must be stated when prevalence estimates are compared.

What is not established is a single cause, a universal prevalence figure or a reliable checklist of individual risk factors that predicts which dog will develop the problem. Breed, sex, neuter status, source, early experience and owner behaviour have all produced associations in some datasets, but the direction and strength vary across studies. The 2024 puppy findings are promising for hypothesis generation, yet their correlational design cannot show that changing one factor will prevent separation-related behaviour.

Separation-related behaviour is a context, not a single diagnosis. The same visible sign can reflect anxiety, fear, frustration, boredom or another problem, so evidence from timing, video and differential assessment matters more than the label alone.

The strongest practical message from the research is therefore diagnostic modesty. Behavioural improvement is possible, and controlled evidence supports structured behaviour modification, but the individual dog's emotional state and medical context need to be understood before management is selected. Welfare guidance from the RSPCA treats suspected distress as a reason for veterinary assessment rather than for a generic routine.

Frequently asked questions

Can a dog have separation anxiety without destroying things?

Yes. Destruction is only one possible separation-related sign. Some dogs mainly vocalise, pace, pant, salivate, stop eating or become unusually still when left alone. Because subtle signs can disappear before the owner returns, video during an ordinary absence can reveal distress that would otherwise leave little physical evidence.

How soon after an owner leaves can separation-related signs begin?

They can begin within minutes in affected dogs, although timing varies. In a 2010 video study of 23 clinical cases, most dogs showed signs within 10 minutes of departure, with vocalising appearing earlier on average than destruction. Timing is informative, but it is not diagnostic without context and differential assessment.

Can puppies show separation-related behaviour?

Yes. The 2024 Generation Pup study identified separation-related behaviours in six-month-old puppies using owner questionnaires. Its broad case definition classified 46.9% of the 145 analysed puppies as showing at least one defined sign only when left alone. The authors cautioned that the exploratory associations do not prove specific preventive causes.

Can separation-related behaviour start after a change in routine?

A routine change can be associated with new signs, but it is not the only explanation. In a UK cohort followed during 2020, dogs whose alone-time patterns changed substantially were at increased risk of developing new SRB signs. A new behaviour still needs assessment for medical, environmental and other behavioural causes.

Does getting another dog reliably solve separation anxiety?

Research does not support another dog as a universal solution. Some separation-related problems involve loss of access to a particular person, while others may reflect frustration, noise sensitivity or boredom. Whether another animal changes the response depends on the individual dog and household rather than on a general rule.

Sources

  1. Sargisson RJ. Canine separation anxiety: strategies for treatment and management. Veterinary Medicine: Research and Reports. 2014;5:143–151. DOI 10.2147/VMRR.S60424.
  2. Salonen M, Sulkama S, Mikkola S, Puurunen J, Hakanen E, Tiira K, Araujo C, Lohi H. Prevalence, comorbidity, and breed differences in canine anxiety in 13,700 Finnish pet dogs. Scientific Reports. 2020;10:2962. DOI 10.1038/s41598-020-59837-z.
  3. de Assis LS, Matos R, Pike TW, Burman OHP, Mills DS. Developing Diagnostic Frameworks in Veterinary Behavioral Medicine: Disambiguating Separation Related Problems in Dogs. Frontiers in Veterinary Science. 2020;6:499. DOI 10.3389/fvets.2019.00499.
  4. Harvey ND, Christley RM, Giragosian K, Mead R, Murray JK, Samet L, Upjohn MM, Casey RA. Impact of Changes in Time Left Alone on Separation-Related Behaviour in UK Pet Dogs. Animals. 2022;12(4):482. DOI 10.3390/ani12040482.
  5. Dale FC, Burn CC, Murray J, Casey R. Canine separation-related behaviour at six months of age: Dog, owner and early-life risk factors identified using the Generation Pup longitudinal study. Animal Welfare. 2024;33:e60. DOI 10.1017/awf.2024.56.
  6. Palestrini C, Minero M, Cannas S, Rossi E, Frank D. Video analysis of dogs with separation-related behaviors. Applied Animal Behaviour Science. 2010;124(1–2):61–67. DOI 10.1016/j.applanim.2010.01.014.
  7. Blackwell E, Casey RA, Bradshaw JWS. Controlled trial of behavioural therapy for separation-related disorders in dogs. Veterinary Record. 2006;158(16):551–554. DOI 10.1136/vr.158.16.551.
  8. Flannigan G, Dodman NH. Risk factors and behaviors associated with separation anxiety in dogs. Journal of the American Veterinary Medical Association. 2001;219(4):460–466. DOI 10.2460/javma.2001.219.460.
  9. ASPCA. Separation Anxiety. Accessed 7 November 2026. https://www.aspca.org/pet-care/dog-care/common-dog-behavior-issues/separation-anxiety
  10. Sherman BL, Mills DS. Canine anxieties and phobias: an update on separation anxiety and noise aversions. Veterinary Clinics of North America: Small Animal Practice. 2008;38(5):1081–1106. DOI 10.1016/j.cvsm.2008.04.012.
  11. RSPCA. Recognising separation-related behaviour and anxiety in dogs. Accessed 7 November 2026. https://www.rspca.org.uk/adviceandwelfare/pets/dogs/behaviour/separationrelatedbehaviour
Gethin Morrow

Gethin Morrow — author

Gethin Morrow is an Animal Biology graduate with a master’s-level specialization in Veterinary Epidemiology. His earlier work involved preparing clear, accessible materials on disease prevention for sheep, pig and poultry keepers. At Cabdirect, Gethin covers l...

Related

About this section