Cooperative Care in Dogs: What the Evidence Supports

1. The Problem Cooperative Care Is Meant to Solve

1.1 What Happens at the Clinic

Veterinary visits are one of the few situations in which dogs are routinely restrained, handled by strangers and subjected to procedures they cannot consent to or escape. The measured consequences are substantial. In a study of 135 dogs at a German veterinary practice, up to 78.5 percent were classified as fearful on the examination table, fewer than half entered the clinic calmly, and 13.3 percent had to be dragged or carried in (Döring, Roscher, Scheipl, Küchenhoff & Erhard, 2009).

Those figures describe ordinary appointments, not emergencies. They are the baseline against which any handling approach has to be judged (how fear and anxiety are measured in dogs).

1.2 What Cooperative Care Claims to Be

Cooperative care teaches an animal to participate voluntarily in husbandry and medical procedures rather than being restrained through them. In the applied behavior analysis literature the approach is characterized by the learner's option to leave the situation, with start and stop behaviors described as specific techniques for arranging that option (Novack, Schnell-Peskin, Feuerbacher & Fernandez, 2023). In practice: the behavior is broken into small reinforced steps, the animal is taught a position it can hold and leave, and a signal is established by which it can indicate readiness or ask for a pause.

What defines the approach is voluntary participation: the animal can influence whether the procedure starts and continues, and can leave. An explicitly trained start and stop signal is one way of operationalizing that, not the only one. Direct canine evidence now exists and is small and mixed rather than absent; what has not been isolated is the specific contribution of an explicit, reliably answered opt-in and opt-out contingency compared with otherwise identical training without it.

1.3 How to Read the Evidence Here

Three kinds of source run through this article. A small number of controlled canine studies test specific interventions — a training program, owner presence, a handling technique. A larger body of descriptive work documents how fearful dogs are in clinics and which factors go with that fear. And the training procedures themselves come from applied behavior analysis and animal husbandry training, where they are standard practice.

2. What the Clinic Does to Dogs

2.1 Fear Is the Norm, Not the Exception

The Döring figures are not an outlier: a review concluded that most dogs and cats are fearful during veterinary visits and that some show aggression as a result (Riemer et al., 2021). This matters clinically as well as ethically — a frightened animal is harder to examine, subtle signs of pain are easier to miss, and the risk to staff rises.

2.2 The Visit Starts Before the Examination

Fear responses begin in the car park and waiting room. In the same sample, entry behavior already separated dogs into those who walked in calmly and those who had to be pulled or carried (Döring et al., 2009), and what follows is layered onto a dog already under load (what is known about accumulating load).

2.3 The Environment Itself Contributes

Some of the load may come from the building rather than the procedure, although the direct evidence is thinner than the recommendation. In a controlled comparison, high background noise during a routine physical examination was associated with a higher respiratory rate, while the other measured behavioral and physiological responses did not differ significantly (Stellato et al., 2019b). Noise, flooring, smells and the proximity of other animals are modifiable without training the dog at all, which is an argument from welfare and plausibility rather than from a large effect.

2.4 What Goes With Fear and Aggression at the Clinic

A survey of dog owners examined which factors were associated with veterinary-related fear and aggression. For fear, dogs were rated as more fearful if they were neutered at or before one year of age, had their first nail trim later in life, showed severe non-social or stranger-directed fear, were stressed or aggressive during body handling, had shown a negative change in behavior after an aversive clinic experience, or had owners who reported being nervous in the clinic themselves (Stellato, Flint, Dewey, Widowski & Niel, 2021).

For aggression at the clinic, the associated factors included fear of the veterinary clinic, fear of being toweled, stress or aggression during body handling — and the owner's use of positive punishment in routine training. These are cross-sectional associations from owner report, so they cannot establish direction. They do describe the population a cooperative care program is aimed at, and the punishment finding is consistent with what the wider literature reports about aversive methods (what punishment costs).

3. The Evidence for Training Interventions

3.1 Owner-Led Handling Practice

The obvious intervention is to have owners practice examination-style handling at home and visit the clinic without a procedure. That recommendation has been tested. Dogs with pre-existing veterinary fear were allocated to a four-week program in which owners performed exam-style handling and visited the clinic weekly, or to a control group with no instructions, with mock veterinary appointments before and after (Stellato, Jajou, Dewey, Widowski & Niel, 2019a).

Twenty-seven owners assigned to the training completed the study, but 12 of them — 44 percent — failed to meet the training requirements, so the efficacy analysis compared the 15 compliant training dogs with 22 controls. That is essentially a per-protocol analysis of the households that did the work, which makes the compliance figure a finding in its own right. During the examination, trained dogs showed less reduced posture than control dogs, but more lip licking during entry and examination. Owners reported visible improvement, and general fear scores were lower at the second examination for trained dogs. The authors concluded that few behavioral indicators changed in the predicted direction, and that the training should still be recommended until better evidence exists.

3.2 What That Study Does and Does Not Show

It does not show that desensitization and counterconditioning fail. Fifteen analyzed dogs, four weeks of practice and a mock appointment as the outcome measure — the design could easily miss a real effect. It also mixed two interventions, handling practice and clinic visits, without separating them.

What it does show is that the standard recommendation, in owners' hands, did not produce a clean improvement in observable fear, and that compliance is a serious practical constraint (how graduated protocols are built and where they fail).

3.3 A Cooperative Care Program With a Cooperation Signal

The approach has also been tested as a package. In a controlled trial with blinded examiners, 47 dogs were allocated semi-randomly — balanced for age, sex, neuter status, veterinary fear and training experience, with owner preferences taken into account — to a training group or a control group, and 40 completed the study, 22 and 18 respectively. Dogs and owners attended a standardized veterinary visit twice, about 140 days apart, and in between the training group completed an average of ten group sessions of cooperative care training plus practice at home, including a front-paw target as a cooperation signal (Wess et al., 2022).

The results do not support the enthusiastic version of the method. Heart rate and heart rate variability showed that the examination was more stressful than the waiting room, but there were no significant group differences across visits. Tympanic membrane temperature showed a group-by-visit interaction, and compliance — completing all steps of the examination — was actually lower in the training group at the second visit. The authors describe the transfer of trained skills to an examination performed by a blinded team as poor. Lower compliance should not be read automatically as worse welfare: the authors also considered whether the training produced more active resistance in dogs that had previously coped by holding still.

A small pilot study with a long interval between visits cannot settle the question, and the result is not evidence that the training harms dogs. It is, however, direct evidence that a full cooperative care program, cooperation signal included, did not automatically produce calmer, more compliant patients in a blinded examination.

3.4 A Handling and Collaboration Package

A second controlled study looked at the clinic side. Twenty-eight dogs were examined four times over eight weeks; all received identical care at the first visit and were then randomized, with 14 dogs receiving procedures designed to reduce stress and to enlist their collaboration during the examination and 14 receiving routine care. Heart rate, serum cortisol, neutrophil-lymphocyte ratio and creatine kinase were measured at each visit, and a composite stress index was calculated from them (Squair, Proudfoot, Montelpare & Overall, 2023).

The physiological results were mixed rather than null. Heart rate, neutrophil-lymphocyte ratio and creatine kinase did not differ significantly between groups, but the intervention group showed a greater overall decrease in serum cortisol from the first to the fourth visit, and its composite stress index improved significantly over that period while the control group's did not (Squair et al., 2023).

The behavioral data from the same randomized study point the same way. Clinical stress scores did not differ at the baseline visit or when dogs entered the hospital or the examination room. At the fourth visit, however, intervention scores decreased or stayed low while the dogs were weighed and at the beginning and end of the physical examination, and control scores were significantly higher during those periods (Squair et al., 2024).

This design cannot isolate the consent element, because the intervention combined handling, environmental and collaborative changes, and the homework differed between groups. What it does show is that a package of this kind produced measurable differences against standard care in a randomized comparison.

3.5 Handling and Protocol Changes on the Clinic Side

The other half is what the practice does. The review of mitigation measures lists non-threatening body language, generous use of food or toys, low-stress handling, brief pauses, and adjusting the procedure to the animal's body language (Riemer et al., 2021), and is explicit that where controlled studies are missing it draws on guidelines and expert texts.

That is an honest description of the state of the field. Much of what is recommended in low-stress handling is plausible and widely endorsed, and packages of it have now been tested (Squair et al., 2023, 2024).

Individual handling components are incompletely studied but no longer untested. In a randomized block design with 97 shelter dogs, five conditions were compared during a two-minute examination: passive restraint, muzzle-hold restraint, basket muzzle, soft muzzle and full-body restraint. Fear scores were lower under passive restraint than under each of the other four conditions, dogs in the muzzle-hold condition made more escape attempts, other negative behaviors differed between treatments, and physiological measures did not differ (Cisneros, Carroll, Moody & Stellato, 2025). It used shelter dogs in a short examination and does not validate a complete handling protocol, but it is direct evidence that technique changes how dogs respond. An observational teaching-laboratory study found the same pattern, with passive handling associated with lower fear scores and more restrictive techniques with higher fear responses; with 13 repeatedly examined dogs and no experimental assignment, it supports an association rather than a causal effect (Nakonechny, Pulver, Artemiou & Stellato, 2026).

3.6 Owner Presence

One clinic-side variable has been examined directly: whether the owner stays, and what the owner does. Owner presence was associated with indicators of fear during routine veterinary examinations (Stellato, Dewey, Widowski & Niel, 2020). In a crossover study in which owners either petted and talked to their dog during the examination or were present without interacting, the examination produced acute stress responses in both conditions, while owner interaction was associated with fewer attempts to jump off the table and with attenuated heart rate and maximal ocular surface temperature (Csoltova, Martineau, Boissy & Gilbert, 2017).

A third study complicates the picture. In a crossover with 25 dog-owner dyads at two standardized consultations, the presence or absence of the owner had no significant effect on stress-related behavior or on how easily the dog could be handled during the examination phase; with the owner present, dogs entered the consultation room more willingly and appeared more relaxed during the exploration phase (Girault, Priymenko, Helsly, Duranton & Gaunet, 2022).

Controlled results are therefore not uniform across measures or stages of the visit. Some outcomes improve when the owner stays, active interaction attenuates several stress-related measures, one study found no effect during the examination itself, and none of them removes the stress response.

3.7 What Happens After a Bad Visit

One finding from the risk-factor survey deserves separate attention: dogs whose behavior changed for the worse after an aversive clinic experience were among those rated as more fearful at the clinic (Stellato et al., 2021). Owner report cannot establish that the experience caused the change, and dogs that react badly may also be the ones who get handled more firmly.

4. What Cooperative Care Actually Consists Of

4.1 The Building Blocks

Stripped of its terminology, cooperative care combines elements that are well understood in learning terms. The procedure is broken into small approximations, each reinforced. A stationing behavior gives the dog a defined position — a chin rest, a mat, a platform, a paw on a target — that it can hold and leave. A start or consent signal is built, in which the dog's own behavior begins the procedure. And a pause or stop signal is respected when the dog moves out of position.

4.2 Why the Consent Signal Is the Interesting Part

Stationing and shaping are ordinary training; what distinguishes the approach is that the dog's behavior influences whether the procedure starts and continues, converting an uncontrollable event into a partly controllable one. Predictable conditions are generally associated with better welfare than unpredictable ones in captive animals, and predictability and control are closely linked in husbandry settings (Bassett & Buchanan-Smith, 2007); that is the theoretical case for the approach (what the research on uncontrollable experience shows).

In dogs specifically, that case has not been tested against a comparison condition. No canine study has compared a procedure with a functioning consent signal against the same procedure without one, holding everything else constant.

4.3 The Signal Only Works If Someone Answers It

Ignoring a trained stop signal breaks the contingency the dog has learned and may produce frustration or make the signal useless. The cooperative care trial provides a real example of the contingency failing in transfer: owners were told they could pause or stop the examination at any time, yet very few did so when their dogs left the target. Loss of control and frustration were proposed by the authors as possible explanations for the poor transfer, but those emotional states were not measured, and no study has compared an ignored stop signal with having no such signal at all (Wess et al., 2022).

The practical consequence is that the training target is the human at least as much as the dog. Teaching the stationing behavior is one part of the task; maintaining the stop contingency consistently across owners and clinic staff is plausibly the harder implementation problem.

4.4 What the Dog Can and Cannot Consent To

"Consent" here describes a trained contingency, not an ethical or legal concept. A dog that holds a chin rest for a nail trim has learned that staying produces reinforcement and that leaving stops the procedure. It has not understood the purpose, evaluated alternatives or agreed to treatment.

That distinction matters. Emergencies, painful procedures and legally required treatments happen whether or not the dog cooperates, and an approach that implies otherwise sets up household and animal for a worse experience when choice is not available (why a behavior is not evidence of an inner state).

4.5 Choice, Control and What They Are Worth

The welfare argument for cooperative care rests on two variables: predictability, meaning the animal can anticipate what happens, and controllability, meaning its behavior influences what happens. In research on other species both are associated with lower stress responses to the same physical event (Bassett & Buchanan-Smith, 2007), and the absence of control is the defining feature of the procedures used to produce learned helplessness.

Applied to husbandry, that argument is reasonable enough to act on and weaker than it sounds when stated precisely. What the dog controls is usually the timing and duration of a procedure that will happen anyway: a dog that can pause a nail trim has more control than one that cannot, but not the option of declining it.

4.6 When the Procedure Cannot Wait

The honest limit of the approach is the situation it was not built for. A laceration, a foreign body, a seizure — these do not pause for a chin rest, and they are the moments in which a dog with no experience of firm handling may be handled firmly for the first time.

5. What Is Borrowed From Other Species

5.1 Zoo and Laboratory Practice

Voluntary participation in husbandry — presenting a limb for blood draw, entering a crate, holding still for an injection — has a long history in zoo and laboratory animal management. In zoos, positive reinforcement training is used to make complex veterinary procedures possible without sedation or restraint; in one study formally trained animals responded more quickly to keeper cues than partially trained or untrained ones, which the authors suggested might reflect lower fear of humans, although fear itself was not measured (Ward & Melfi, 2013). A scoping review of 20 peer-reviewed zoo studies found shaping, often with clicker training, to be the most common method, no use of coercion, and more than half of the trained behaviors intended to ease handling, with others serving sample collection and medical treatment (Boundey, Nikitins & Fernandez, 2025). In laboratory primate management, training has the animals cooperate in aspects of their own care, using target training, stationing and systematic desensitization and counterconditioning (Westlund, 2015). Modern canine cooperative care protocols closely parallel that practice, often using the same terminology, though a documented transfer from one field to the other is another matter.

That is a strength: these procedures were refined on animals that cannot be restrained safely. It is also a limitation: that literature concerns other species and procedures.

5.2 The Mechanism Claim

The usual explanation is that predictability and control reduce stress responses, drawing on experimental work in other species. Applied to a dog on an examination table, the account is plausible and unmeasured. Whether a dog that has learned a chin rest experiences a veterinary examination differently, or simply performs a trained behavior while feeling much as it did before, has not been established. No single established stand-alone measure can settle it: qualitative behavioral assessment and several physiological and behavioral parameters are promising candidates, but their interpretation remains context-dependent (Csoltova & Mehinagic, 2020; Flint et al., 2024).

5.3 Measuring Whether It Worked

The available measures are the usual ones: behavior, heart rate and heart rate variability, cortisol, and whether the procedure can be completed. They disagree often enough that a single measure is a weak basis for a claim; in a controlled study of 60 dogs several standard parameters differentiated arousal only within negative-valence scenarios (Flint et al., 2024).

5.4 Where the Approach Came From in Dog Training

In companion dog training, cooperative care protocols are widely taught through courses, books and practitioner networks. That practical literature is far larger than the peer-reviewed outcome literature: the protocols are detailed and written for ordinary households, and they circulate without the controls that would establish whether they work better than careful conventional handling.

Because the approach is tied to a training philosophy, disagreements about it are conducted as arguments about values rather than evidence, which makes it harder to say the accurate thing — the rationale is good, the canine outcome data are limited, and both are true at once.

6. Medical and Pharmacological Context

6.1 Pain Changes What Is Possible

A dog that resists handling may be defending a painful area rather than showing a training deficit. In a review of 100 behavior cases, a conservative estimate put a painful component in roughly a third (Mills et al., 2020). Training that repeatedly asks a dog to tolerate contact with a painful region may be ineffective and inappropriate until the pain has been assessed and treated (how chronic pain changes behavior).

6.2 Medication Around Visits

The drug with the most direct evidence here is dexmedetomidine oromucosal gel. In a randomized, double-blind, placebo-controlled crossover trial with 40 dogs with a history of fear at the clinic, it reduced stress and fear vocalization, avoidance and other stress-related activity during the examination (Hauser et al., 2020). In a multicenter placebo-controlled field study, 76 dogs were enrolled and randomized and 74 were treated; the investigator could carry out the intended examination and one short procedure with an excellent rating for 40.7 and 33.3 percent of the two dose groups, against 4.3 percent on placebo, with more relaxed posture on entering the room. No dog was classified as clinically sedated, although 7.4 percent of the dogs in the lower-dose and 12.5 percent in the higher-dose dexmedetomidine group were rated as drowsy (Korpivaara, Huhtinen, Aspegrén & Overall, 2021).

Two other drugs have been tested for the same situation. Trazodone at 9–12 mg/kg given 90 minutes before transport reduced signs of stress in 20 dogs (Kim et al., 2022), while a comparable gabapentin trial in 22 dogs produced mixed results, with most outcomes not differing from placebo (Stollar et al., 2022).

The decision belongs to a veterinarian. Such medication may reduce situational distress enough to make handling safer or more feasible; whether it improves cooperative care learning or later training progress has not been tested (what the canine drug trials show and what they do not).

6.3 When Training Is the Wrong Answer

A dog that panics on entry, cannot take food in the building, or has already bitten during handling should not have a newly trained chin rest as the sole plan for an imminent procedure. Training can still start in a safe context at home, alongside a veterinary behavioral consultation, a medical work-up, and a plan for procedures that cannot wait, decided in advance rather than improvised.

6.4 Sedation Is Not a Failure

Where distress or defensive behavior makes a necessary procedure unsafe or highly aversive, pre-visit medication or planned sedation may be the better option; the choice and protocol are veterinary decisions. The argument for considering it early is that it avoids the struggle and the rehearsal of panic.

Framing sedation as the outcome of a training failure gets this backwards, and it pushes households into forcing highly aversive procedures, which may worsen later responses and complicate subsequent training. The decision belongs to the veterinarian; the useful contribution from the training side is to say early that the dog will not be ready in time.

7. Building the Behavior

7.1 Start Where the Dog Already Is

Start with an inventory: which body areas can be touched, for how long, in which room, by whom. Plans commonly start above the dog's threshold, and the resulting struggle teaches what the approach is meant to prevent.

7.2 The Station

A chin rest, a mat, a platform or a nose target gives the procedure a defined start and end. Its value is clarity: the dog has one thing to do, and leaving it is an unambiguous signal the handler can answer. Building it to fluency in a neutral context comes before any handling.

7.3 Adding the Procedure in Small Steps

Handling is added in increments that keep the dog in position — touching the paw before holding it, the clipper running nearby before it touches a nail. The criterion is not whether the dog tolerated the step but whether it stayed and returned willingly on the next repetition (how reinforcement is scheduled in practice).

7.4 Reading the Dog While You Work

The behaviors to watch are small: head turning, lip licking, blinking, freezing, a lowered body. They are not a code with fixed meanings, but a change in frequency during a session is useful feedback (what dogs' signals do and do not show).

7.5 Generalizing to the Clinic

A behavior trained on the kitchen floor is not a behavior at the clinic. Transfer needs the same graduated treatment: surface, room, handler, smells and audience all change, and each change is a step. Practice visits without a procedure are one half of the tested intervention (Stellato et al., 2019a).

7.6 Four Procedures and Where They Break

Nail trimming is a frequent cooperative care project and often has a history of bad experiences behind it. In practice the paw hold is often the harder part rather than the clipper: dogs restrained for trimming often object to the grip before the tool appears. Splitting the paw hold from the cutting is a common practical solution rather than a tested one.

Injections and blood draws depend on someone else's timing. The dog can be taught to hold a position and accept a limb being held and wiped, but the needle is placed by someone under time pressure. Rehearsing everything except the needle, and agreeing who says stop, is the realistic version. Generalizing any of this to unfamiliar staff who work quickly is its own training step.

Grooming is a plausible application because sessions are long and involve repeated handling. An observational study across grooming phases found heart rate variability declining from arrival through restraint, nail trimming, bathing and drying, lowest during drying, while behavioral suppression increased; suppression was inversely related to heart rate variability, so stillness during grooming may reflect coping rather than comfort (Hart & Hart, 2026). The study tested no cooperative care intervention, and in practice time constraints in commercial grooming may limit how consistently such protocols are applied. Where a groomer will not work that way, the training will not transfer to that setting.

8. Practical Recommendations

8.1 Separate the Three Levers

Environment, handling and training are separate interventions with different costs. Noise, flooring, waiting time and whether the owner stays can be changed today; handling style is a practice decision; training depends on the household doing homework. Some environmental and handling changes can be made immediately, whereas training requires repeated practice by the household (Stellato et al., 2019b, 2020; Cisneros et al., 2025).

8.2 Plan for the Emergency Now

Every plan should decide in advance whether and under what circumstances pre-visit medication, sedation or a muzzle trained calmly beforehand may be needed if the dog requires a procedure before the training is ready. That avoids the kind of experience the risk-factor data associate with later fear (Stellato et al., 2021).

8.3 Keep Sessions Short and Records Simple

A short log — what was practiced, how long the dog stayed, what it looked like afterwards — is what makes progress visible. Given the compliance figures in the one study available, the realistic question is not what the ideal protocol looks like but what a household will actually do (Stellato et al., 2019a).

8.4 Respect the Signal or Drop It

If the household or the practice will not stop when the dog leaves the station, introducing the consent element has little point: the contingency the dog is supposed to learn does not exist. Whether that is worse than never training a signal has not been tested, but it is certainly not what the method is meant to deliver.

8.5 Work With the Practice, Not Around It

A household can prepare a dog and still lose the visit. Asking the practice in advance what is possible — a longer slot, the first or last appointment of the day, waiting in the car, an examination on the floor, the owner staying, a pause when the dog leaves the station — turns the plan into something that can run. Not every such adjustment helps every dog: in a randomized crossover study with 37 dogs, examination on the floor showed no advantage over the table in body language, heart rate or respiratory rate, so the location is better matched to the individual dog than applied as a rule (Mercier, Wilson, Bazin, Fiset & Seksel, 2025).

It works the other way too: practices that want calmer patients need households that arrive prepared, with food where medically appropriate and a behavior the dog can perform. Cooperative care is a shared protocol or it ends at the clinic door.

9. Summary at a Glance

Fear at the clinic is the norm — Up to 78.5 percent of dogs were classified as fearful on the examination table, fewer than half entered the clinic calmly, and 13.3 percent had to be dragged or carried in (Döring et al., 2009).

The majority finding holds across the literature — A review concludes that most dogs and cats are fearful during veterinary visits, with aggression in some (Riemer et al., 2021).

Owner-led training: mixed results and poor compliance — Trained dogs showed less reduced posture but more lip licking, owner-rated fear improved, and 44 percent of the training-group owners did not meet the requirements (Stellato et al., 2019a).

A full cooperative care program transferred poorly — With blinded examiners and 40 dogs, heart rate and heart rate variability showed no group differences, and compliance was lower in the training group at the second visit (Wess et al., 2022).

A handling and collaboration package did better — In 28 randomized dogs, the intervention group showed a greater cortisol decrease and an improved composite stress index, and lower clinical stress scores at the fourth visit while weighed and during the examination (Squair et al., 2023, 2024).

Owner presence and interaction can matter, but not uniformly — Owner interaction attenuated heart rate and eye temperature and reduced attempts to jump off the table (Csoltova et al., 2017), presence was associated with indicators of fear (Stellato et al., 2020), and a crossover study found no effect of presence during the examination itself (Girault et al., 2022).

The building may contribute — High background noise during a routine examination was associated with a higher respiratory rate, with no significant differences in the other measures (Stellato et al., 2019b).

Punishment at home goes with aggression at the clinic — Owner-reported use of positive punishment in routine training was among the factors associated with veterinary-related aggression (Stellato et al., 2021).

Medication can reduce distress during visits — Dexmedetomidine gel reduced several stress-related behaviors in a 40-dog crossover trial and improved completion of examinations and minor procedures in a placebo-controlled field study (Hauser et al., 2020; Korpivaara et al., 2021); trazodone reduced stress signs in 20 dogs (Kim et al., 2022), while gabapentin produced mixed results in 22 (Stollar et al., 2022).

Pain belongs on the differential — Roughly a third of 100 reviewed behavior cases had a painful component (Mills et al., 2020).

10. Research Gaps and Critical Appraisal

The contribution of the opt-in contingency is untested. Cooperative care has been studied as a package (Wess et al., 2022; Squair et al., 2023, 2024), but no canine study isolates the additional effect of an explicit, reliably answered start and stop contingency from the other components of the training.

The available trials are small and their results mixed. Fifteen analyzed training dogs against 22 controls with 44 percent non-compliance (Stellato et al., 2019a); 22 against 18 dogs with no group differences in heart rate or heart rate variability, lower compliance in the training group at the second visit and poor transfer to a blinded examination (Wess et al., 2022); 14 against 14 dogs in a package intervention (Squair et al., 2023).

Interventions are bundled. Handling practice at home and clinic visits without procedures were tested together, as were most low-stress handling packages. Which component does the work is unknown.

Outcome measures do not map cleanly onto emotional state. Behavioral, physiological and owner-reported measures moved differently within the same studies. In a controlled comparison, several commonly used parameters differentiated arousal only within negative-valence conditions, while others varied across valence as well, and qualitative behavioral assessment was the best single candidate for positive valence (Flint et al., 2024). No single measure currently provides a stand-alone readout of whether a cooperating dog is experiencing positive welfare (Csoltova & Mehinagic, 2020).

Descriptive data dominate the field. Prevalence of fear, risk factors and owner perceptions are well described; controlled intervention studies are few, and several rely on cross-sectional owner report (Stellato et al., 2021).

Durability is unstudied. The longest interval in the available work is a few months (Wess et al., 2022). Whether dogs trained this way remain easier to handle over years, and whether any effect survives a painful or emergency procedure, has not been examined.

The groomer and the home are barely studied. Most controlled canine work concerns the veterinary clinic. One case study compared nail clipping with and without restraint and found more stress signals in the restrained group; because the groups reflected pre-existing routines rather than random assignment, choice, restraint and training history were confounded, so it cannot isolate the effect of an opt-in contingency (Sydänheimo, Freeman & Hunt, 2023). Nail trimming, grooming, ear and eye treatment and medication at home are important handling contexts outside the clinic, and they have almost no controlled literature.

Owner adherence and canine compliance are different problems, and both look difficult. Forty-four percent of owners did not meet the training requirements in the four-week program (Stellato et al., 2019a), while the dog-side measure — completing all steps of the examination — was lower in the trained group at the second visit of the cooperative care trial (Wess et al., 2022). Neither has been studied as a target in its own right.

11. Conclusion

Cooperative care is a well-reasoned response to a well-documented problem, and it is no longer untested. The problem is not in doubt: up to 78.5 percent of dogs were fearful on the examination table in one clinic sample, fewer than half entered calmly (Döring et al., 2009), and a review of the field finds that most dogs and cats are fearful during veterinary visits (Riemer et al., 2021). The response — breaking procedures into small reinforced steps, giving the dog a station it can hold and leave, and stopping when it leaves — is sound learning practice, paralleling husbandry training in zoo and laboratory animal management, and consistent with what is known about predictability and control in other species. The direct canine evidence is small and mixed. The controlled test of the standard owner-led program found less reduced posture but more lip licking, improved owner ratings, and 44 percent of owners not meeting the training requirements (Stellato et al., 2019a). A blinded pilot trial of a full cooperative care program including a cooperation signal found no group differences in heart rate or heart rate variability, lower compliance in the training group at the second visit, and poor transfer of trained skills to the examination (Wess et al., 2022). A randomized 28-dog study of adaptive, collaborative handling found improvements in selected physiological and behavioral stress measures across repeated visits (Squair et al., 2023, 2024). What no study has isolated is the additional effect of an explicit, reliably answered start and stop contingency, as distinct from the rest of the training package. Meanwhile, several clinic-side variables have been tested directly and are unglamorous: whether the owner stays (Stellato et al., 2020; Girault et al., 2022), background noise (Stellato et al., 2019b), handling technique (Cisneros et al., 2025) and medication around the visit (Hauser et al., 2020; Korpivaara et al., 2021; Kim et al., 2022). Pain is a separate clinical consideration: painful conditions are common among referred behavior cases and can change responses to handling (Mills et al., 2020), although treating pain has not been tested as a cooperative care intervention. For practice, that argues for using cooperative care for what it reliably provides — a structured, reinforcement-based way to work below the dog's threshold, with a clear signal for stopping — while being honest that the evidence for it in dogs is thin, that a trained stop signal only functions as intended when handlers answer it — no study has established whether an ignored signal is worse than none — and that emergencies will still require handling the dog has not agreed to.

Key Insights (Takeaways)

• The problem is well documented; the solution has been tested a few times with mixed results. Up to 78.5 percent of dogs were fearful on the examination table in one sample (Döring et al., 2009), while a blinded trial of a full cooperative care program found poor transfer to the examination and lower compliance in the trained group (Wess et al., 2022).

• What defines cooperative care is voluntary participation; an explicit start and stop signal is one way of arranging it. Packages have been studied (Wess et al., 2022; Squair et al., 2023, 2024), but the added contribution of that signal has not been isolated.

• A signal only works if it is answered. In the controlled cooperative care trial most owners did not react when the dog interrupted the cooperation signal (Wess et al., 2022); whether an ignored signal is worse than none has not been tested.

• "Consent" describes a trained contingency, not agreement. Emergencies and painful procedures happen regardless, which is why every plan should settle in advance whether and when sedation, muzzle training or pre-visit medication may be needed.

• Environment and handling can be changed without homework. Owner interaction during the examination attenuated heart rate and eye temperature (Csoltova et al., 2017), and high background noise was associated with a higher respiratory rate (Stellato et al., 2019b) — neither requires homework from the household.

• Pain and medication belong in the plan. Dexmedetomidine oromucosal gel has the strongest direct evidence for veterinary visits (Hauser et al., 2020; Korpivaara et al., 2021). About a third of reviewed behavior cases had a painful component (Mills et al., 2020), and pre-visit trazodone reduced stress signs in a placebo-controlled trial (Kim et al., 2022).

• Adherence is the bottleneck nobody plans for. Forty-four percent of owners did not meet the training requirements (Stellato et al., 2019a), and in the cooperative care trial the dogs' compliance during the examination was lower in the trained group at the second visit (Wess et al., 2022). A protocol a household can sustain beats an ideal one it cannot.

References

Bassett, L., & Buchanan-Smith, H. M. (2007). Effects of predictability on the welfare of captive animals. Applied Animal Behaviour Science, 102(3–4), 223–245. https://doi.org/10.1016/j.applanim.2006.05.029

Boundey, A., Nikitins, F., & Fernandez, E. J. (2025). An examination of husbandry training in zoos: A scoping review. Applied Animal Behaviour Science, 292, 106805. https://doi.org/10.1016/j.applanim.2025.106805

Cisneros, A., Carroll, A. D., Moody, C. M., & Stellato, A. C. (2025). Handle with care: Dogs show negative responses to restrictive handling restraints and tools during routine examinations. Applied Animal Behaviour Science, 286, 106601. https://doi.org/10.1016/j.applanim.2025.106601

Csoltova, E., & Mehinagic, E. (2020). Where do we stand in the domestic dog (Canis familiaris) positive-emotion assessment: A state-of-the-art review and future directions. Frontiers in Psychology, 11, 2131. https://doi.org/10.3389/fpsyg.2020.02131

Csoltova, E., Martineau, M., Boissy, A., & Gilbert, C. (2017). Behavioral and physiological reactions in dogs to a veterinary examination: Owner-dog interactions improve canine well-being. Physiology & Behavior, 177, 270–281. https://doi.org/10.1016/j.physbeh.2017.05.013

Döring, D., Roscher, A., Scheipl, F., Küchenhoff, H., & Erhard, M. H. (2009). Fear-related behaviour of dogs in veterinary practice. The Veterinary Journal, 182(1), 38–43. https://doi.org/10.1016/j.tvjl.2008.05.006

Flint, H. E., Weller, J. E., Parry-Howells, N., Ellerby, Z. W., McKay, S. L., & King, T. (2024). Evaluation of indicators of acute emotional states in dogs. Scientific Reports, 14(1), 6406. https://doi.org/10.1038/s41598-024-56859-9

Girault, C., Priymenko, N., Helsly, M., Duranton, C., & Gaunet, F. (2022). Dog behaviours in veterinary consultations: Part 1. Effect of the owner's presence or absence. The Veterinary Journal, 280, 105788. https://doi.org/10.1016/j.tvjl.2022.105788

Hart, D. S., & Hart, S. J. (2026). Behavioural suppression and autonomic stress responses in dogs during routine grooming procedures. Applied Animal Behaviour Science, Article 107146. https://doi.org/10.1016/j.applanim.2026.107146

Hauser, H., Campbell, S., Korpivaara, M., Stefanovski, D., Quinlan, M., & Siracusa, C. (2020). In-hospital administration of dexmedetomidine oromucosal gel for stress reduction in dogs during veterinary visits: A randomized, double-blinded, placebo-controlled study. Journal of Veterinary Behavior, 39, 77–85. https://doi.org/10.1016/j.jveb.2020.05.002

Kim, S. A., Borchardt, M. R., Lee, K., Stelow, E. A., & Bain, M. J. (2022). Effects of trazodone on behavioral and physiological signs of stress in dogs during veterinary visits: A randomized double-blind placebo-controlled crossover clinical trial. Journal of the American Veterinary Medical Association, 260(8), 876–883. https://doi.org/10.2460/javma.20.10.0547

Korpivaara, M., Huhtinen, M., Aspegrén, J., & Overall, K. (2021). Dexmedetomidine oromucosal gel reduces fear and anxiety in dogs during veterinary visits: A randomised, double-blind, placebo-controlled clinical pilot study. Veterinary Record, 189(12), e832. https://doi.org/10.1002/vetr.832

Mercier, P., Wilson, C., Bazin, I., Fiset, S., & Seksel, K. (2025). Is it more stressful for dogs to be examined on a table versus the floor – A randomized crossover study. Journal of Veterinary Behavior, 82, 20–27. https://doi.org/10.1016/j.jveb.2025.09.006

Mills, D. S., Demontigny-Bédard, I., Gruen, M., Klinck, M. P., McPeake, K. J., Barcelos, A. M., Hewison, L., Van Haevermaet, H., Denenberg, S., Hauser, H., Koch, C., Ballantyne, K., Wilson, C., Mathkari, C. V., Pounder, J., Garcia, E., Darder, P., Fatjó, J., & Levine, E. (2020). Pain and problem behavior in cats and dogs. Animals, 10(2), 318. https://doi.org/10.3390/ani10020318

Nakonechny, L., Pulver, V., Artemiou, E., & Stellato, A. C. (2026). Veterinary student wellbeing and its relationship with dog handling and fear responses during a clinical examination laboratory. Frontiers in Veterinary Science, 13, 1910028. https://doi.org/10.3389/fvets.2026.1910028

Novack, L. I., Schnell-Peskin, L., Feuerbacher, E., & Fernandez, E. J. (2023). The science and social validity of companion animal welfare: Functionally defined parameters in a multidisciplinary field. Animals, 13(11), 1850. https://doi.org/10.3390/ani13111850

Riemer, S., Heritier, C., Windschnurer, I., Pratsch, L., Arhant, C., & Affenzeller, N. (2021). A review on mitigating fear and aggression in dogs and cats in a veterinary setting. Animals, 11(1), 158. https://doi.org/10.3390/ani11010158

Squair, C., Proudfoot, K., Montelpare, W., & Overall, K. L. (2023). Effects of changing veterinary handling techniques on canine behaviour and physiology part 1: Physiological measurements. Animals, 13(7), 1253. https://doi.org/10.3390/ani13071253

Squair, C., Proudfoot, K., Montelpare, W., Doucette, T., & Overall, K. L. (2024). Effects of changing veterinary handling techniques on canine behaviour and physiology part 2: Behavioural measurements. Animal Welfare, 33, e43. https://doi.org/10.1017/awf.2024.37

Stellato, A., Dewey, C., Widowski, T., & Niel, L. (2020). Evaluation of associations between owner presence and indicators of fear in dogs during routine veterinary examinations. Journal of the American Veterinary Medical Association, 257(10), 1031–1040. https://doi.org/10.2460/javma.2020.257.10.1031

Stellato, A., Flint, H. E., Dewey, C. E., Widowski, T. M., & Niel, L. (2021). Risk-factors associated with veterinary-related fear and aggression in owned domestic dogs. Applied Animal Behaviour Science, 241, 105374. https://doi.org/10.1016/j.applanim.2021.105374

Stellato, A., Hoffman, H., Gowland, S., Dewey, C. E., Widowski, T. M., & Niel, L. (2019b). Effect of high levels of background noise on dog responses to a routine physical examination in a veterinary setting. Applied Animal Behaviour Science, 214, 64–71. https://doi.org/10.1016/j.applanim.2019.03.009

Stellato, A., Jajou, S., Dewey, C. E., Widowski, T. M., & Niel, L. (2019a). Effect of a standardized four-week desensitization and counter-conditioning training program on pre-existing veterinary fear in companion dogs. Animals, 9(10), 767. https://doi.org/10.3390/ani9100767

Stollar, O. O., Moore, G. E., Mukhopadhyay, A., Gwin, W., & Ogata, N. (2022). Effects of a single dose of orally administered gabapentin in dogs during a veterinary visit: A double-blinded, placebo-controlled study. Journal of the American Veterinary Medical Association, 260(9), 1031–1040. https://doi.org/10.2460/javma.21.03.0167

Sydänheimo, A., Freeman, M., & Hunt, K. (2023). Cooperative care does not scare – Use of cooperative care training in routine husbandry in dogs. Animal Behaviour and Welfare Cases, 2023, abwcases20230010. https://doi.org/10.1079/abwcases.2023.0010

Ward, S. J., & Melfi, V. (2013). The implications of husbandry training on zoo animal response rates. Applied Animal Behaviour Science, 147(1–2), 179–185. https://doi.org/10.1016/j.applanim.2013.05.008

Wess, L., Böhm, A., Schützinger, M., Riemer, S., Yee, J. R., Affenzeller, N., & Arhant, C. (2022). Effect of cooperative care training on physiological parameters and compliance in dogs undergoing a veterinary examination – A pilot study. Applied Animal Behaviour Science, 250, 105615. https://doi.org/10.1016/j.applanim.2022.105615

Westlund, K. (2015). Training laboratory primates – benefits and techniques. Primate Biology, 2, 119–132. https://doi.org/10.5194/pb-2-119-2015