Overweight in Dogs: What the Evidence Shows and Which Lever Actually Works

1. Two Prevalence Figures That Differ by a Factor of Eight

1.1 What the Records Say

A study of 22,333 dogs under UK primary veterinary care found a one-year period prevalence of recorded overweight status of 7.1% (95% CI 6.7–7.4). Eight breeds showed increased odds compared with crossbred dogs after adjusting for confounders, with Pugs highest at an odds ratio of 3.12 (Pegram et al., 2021) — a breed in which excess weight compounds an airway that is already compromised (brachycephaly in dogs).

The operative word is recorded. The figure counts dogs whose clinical notes contain an entry about excess weight. It is a measure of what veterinarians wrote down, not of what the dogs weighed.

1.2 What Measurement Says

When dogs are actually scored rather than searched for in records, the figure changes completely. Among 696 dogs assessed at five UK veterinary practices, 38.9% were overweight and a further 20.4% obese — 59.3% above ideal in total. Only 35.3% had an ideal body shape, and 5.3% were underweight (Courcier et al., 2010).

Seven point one percent against fifty-nine point three. The two studies used different populations, different years and different methods of ascertainment. How much of the difference is down to recording and how much to differences between populations, periods and methods cannot be determined from these two studies. What is clear is only that recorded and directly measured prevalence lie far apart.

1.3 Why the Gap Matters

Three consequences follow for anyone working with dogs.

In some primary-care populations more than half of the dogs assessed directly were above ideal body condition, while only a much smaller proportion had excess weight recorded in their notes. A missing record does not show that nobody discussed the issue; it shows that documentation substantially underestimates measured prevalence.

For practice this means: a missing entry in the records says little about whether body condition is in fact ideal, or whether the subject has already been discussed.

Statistics drawn from clinical records can understate the problem where excess weight is not recorded consistently. Dog schools also meet dogs regularly whose body condition may not have been assessed systematically so far — which puts them in a position to raise the subject early.

2. How This Gets Studied, and Why the Mix Is Unusual

The evidence on canine weight is built from five designs of very unequal strength. The subject is unusual in having one genuine lifetime experiment and almost nothing else that is controlled, so it pays to know which claim rests on which.

2.1 The Lifetime Experiment

One study assigned the exposure, in littermate pairs, for the animals' whole lives. That design answers a causal question no observational study can, and nothing comparable has been done since. Its weakness is its narrowness: one breed, one kennel, and a restriction applied from puppyhood rather than an intervention in an adult dog.

2.2 Large Observational Cohorts

Clinical-record cohorts of tens of thousands of dogs can measure life span and risk across breeds, which no trial will ever afford. What they cannot do is separate cause from marker: a dog that is overweight at seven may already be a dog with reduced mobility, an undiagnosed condition, or an owner whose care differs in other ways. The association is solid; the mechanism is inferred.

2.3 Cross-Sectional Body-Condition Surveys

Physically scoring every dog in a clinic population gives the only honest prevalence figures and can identify associated factors. It cannot establish direction: a study finding less exercise in obese dogs cannot say whether the dogs move less because they are heavy or are heavy because they move less.

2.4 Uncontrolled Before-and-After Studies

Most of the weight-loss literature is of this kind: enrol overweight dogs, run a program, measure what changed. It is the only practical design for the question and it has two structural weaknesses. There is no group that did not lose weight, and the person rating the outcome is usually the owner who ran the program. Dropouts compound both: a study reporting on dogs that completed is reporting on a favorable selection.

2.5 Randomized Trials

There is very little, and what exists is small. A trial with thirteen dogs and no blinding is weak evidence in isolation — and it is still the only evidence that compares two interventions against each other rather than against nothing, which is a question no amount of observational data answers.

2.6 How to Weigh Them

Where several designs point the same way, the claim is usable. Life span is supported by a large cohort and by the lifetime experiment independently. Where only owner-reported before-and-after data exist — most of the quality-of-life and behavior findings — the direction is probably right and the size is not trustworthy (how behavior gets operationalized).

3. Nobody in the Room Can See It

3.1 Owner and Veterinarian Disagree Where It Counts

A study comparing owner and veterinary assessment covered 125 dogs belonging to 121 owners, with 117 complete questionnaires analyzed. For dogs the veterinarian scored as normal weight, owner and veterinarian agreed in about 80% of cases (49 of 61). For dogs scored as overweight — body condition 3.5 to 4.5 on a five-point scale — agreement fell to 53% (26 of 49), a significant difference (χ² = 7.96, P = 0.005) (White et al., 2011).

The direction of the disagreement is the finding. Among owners of overweight dogs who disagreed with the veterinarian, 39% considered their dog's weight acceptable or normal. One thought the dog needed to gain.

3.2 Why the Error Is Systematic

This is not inattention. It is a comparison problem.

Owners may calibrate against the dogs they see, and in populations where overweight is common the visual norm can shift upward. In that setting a dog at ideal body condition may look unusually lean by comparison. This is a plausible explanation for systematic underestimation rather than something the prevalence study itself tested.

Breed images reinforce it. Many breed photographs in circulation, including show-ring images, show dogs above ideal condition, so the mental template a new owner forms is already shifted.

And the change is gradual. Weight gain over three years is invisible to someone who sees the dog daily, which is why the comparison that works is a photograph from two years ago rather than a mirror.

3.3 What Follows for a Consultation

Owner self-assessment is not sufficient as screening. In the study cited, agreement between owner and veterinarian was markedly worse for overweight dogs than for normal-weight ones.

What works is the hands. Body condition scoring — ribs readily palpable without excess fat covering, a visible waist from above, and an abdominal tuck from the side — is a practical, repeatable assessment that both practitioner and owner can perform on the same dog and compare. It also requires a dog that tolerates being handled along the ribs and flank, which is training work in its own right (cooperative care) (why operational definitions matter).

Weight in kilograms is not a substitute. Without a breed-typical reference and a baseline for that individual, a number on a scale says nothing about body composition.

3.4 What Body Condition Scoring Does and Does Not Measure

Body condition scoring is the practical standard, and it should be used with its limits stated. It is a semi-quantitative judgment of subcutaneous fat and palpable bone cover, converted into a five- or nine-point scale. Two scales are in circulation, they are not interchangeable, and a score reported without its scale is uninterpretable.

It is a trained skill rather than an objective reading, and agreement between assessors is imperfect — which is one reason the owner-veterinarian disagreement above is not simply owner error. Heavy coat, muscle mass in working-built dogs, and breed conformation all make palpation harder in individual cases.

What makes it worth using anyway is that the alternatives are worse. Bodyweight alone carries no information about composition, visual impression is what fails in the data above, and the methods that measure fat directly belong in research settings. A repeated score by the same assessor on the same dog, with the scale named and the date recorded, is a usable series.

3.5 The Opposite Error

In the same survey that found 59.3% of dogs above ideal, 5.3% were underweight (Courcier et al., 2010). That figure is small and it is not zero, and over-restriction is a real failure mode in a household that has been told its dog is too heavy.

Weight loss that is too fast costs lean tissue as well as fat, which is the opposite of the aim in a dog whose joints need muscular support. The rates actually achieved in the studies cited here are modest: the osteoarthritis study averaged about 0.55% of body weight per week over sixteen weeks, and the international study's mean of 11.4% over three months works out at roughly 0.9% per week. Target rates are a veterinary prescription for the individual dog, and this article does not cite a guideline for them.

A dog losing faster than that without a deliberately restricted plan, or losing weight when nothing was changed, is a dog that needs examining rather than praising (why pain is missed in dogs).

4. What Excess Weight Actually Costs

4.1 Life Span, by Breed

The largest study on this question followed 50,787 neutered client-owned dogs of 12 breeds across approximately 900 North American veterinary hospitals, with recorded death dates for 14,316 of them (28.2%) (Salt et al., 2019).

Median life span was lower in overweight dogs in every one of the twelve breeds. The size of the difference varied considerably:

Yorkshire Terriers, 16.2 years at normal body condition against 13.7 overweight, a difference of 2.5 years. Dachshunds, 16.4 against 14.1, a difference of 2.3 years. Chihuahuas, 16.0 against 13.9. Beagles, 15.2 against 13.2. Pomeranians, 15.5 against 13.7. At the other end: German Shepherd Dogs, 12.5 against 12.1, a difference of 0.4 years. Labrador Retrievers, 13.3 against 12.7. Boxers, 12.4 against 11.8.

The hazard ratios follow the same pattern. Yorkshire Terrier 2.86 (99.79% CI 2.14–3.83), Dachshund 2.77, Chihuahua 2.42, Beagle 2.40, down to German Shepherd Dog 1.35 (1.05–1.73). For all twelve breeds the instantaneous risk of death was higher in overweight dogs throughout the age range studied, all at P < 0.001.

Two things are worth noting honestly. This is an observational study, so excess weight may partly be a marker for other things rather than purely a cause. The association was larger in several of the smaller breeds than in the larger breeds studied, which is the opposite of the usual assumption, but whether body size itself explains that pattern is unknown, and it sits alongside the unexplained inverse relationship between body size and longevity in dogs generally (cellular aging in dogs).

4.2 The One Experiment That Exists

Observational data on body condition cannot separate cause from correlation. One study can, because it assigned the exposure.

Forty-eight Labrador Retrievers were paired at six weeks of age by sex and weight within each of seven litters. During the first part of the study one dog in each pair was fed ad libitum and the other received 75% of its pair-mate’s intake. After 3.25 years the control dogs were also placed on a fixed ration to avoid excessive obesity, while the restricted dogs continued to receive 75% of the amount fed to their paired controls. Median life span was 13.0 years in the restricted group against 11.2 years in the control group, with delayed onset of late-life disease, osteoarthritis in particular (Kealy et al., 2002).

Roughly 1.8 additional years from feeding less, in a controlled lifetime design with littermate pairing. It remains the strongest evidence in the field, and it has not been repeated.

The limitations are real: one breed, one kennel environment, and a restriction imposed from puppyhood rather than a weight-loss intervention in an adult dog. It shows what lifelong leanness achieves, not what slimming a seven-year-old achieves.

4.3 The Comorbidity List Does Not Transfer From Humans

Here the literature contradicts what is usually said. A One Health review of obesity and its comorbidities across species examined the associations in people, cats and dogs, and found the canine picture to be markedly weaker. The obesity–diabetes association recognized in people and cats is not recognized in dogs, and far fewer data describe an association with canine neoplastic disease (Chandler et al., 2017).

That is an inconvenient finding for anyone who has used the human comorbidity list in a client conversation, and it should be stated rather than glossed. Canine diabetes mellitus is predominantly insulin-deficient rather than insulin-resistant, which is a different disease from the condition obesity drives in humans.

What this does not license is the conclusion that excess weight is harmless in dogs. The life-span data are strong, the orthopedic data give direct indications of a benefit from weight loss, and interventional before-and-after data support improvement in several quality-of-life domains after successful weight loss. The point is narrower and more useful: argue from the canine evidence, which is good enough on its own, rather than from a borrowed list that does not hold (hormones and behavior in dogs).

5. Pain and Movement: The Strongest Practical Argument

5.1 Weight Loss Can Reduce Lameness

Fourteen obese client-owned dogs with clinical and radiographic evidence of hip or elbow osteoarthritis completed a 16-week weight-loss study. Mean body weight reduction was 8.85% (95% CI 6.82–10.88), roughly 0.55% per week.

In this study, clinical lameness scores showed a statistically detectable improvement from about 6.10% weight loss onward. By the end, 82% of the dogs showed improvement on combined numerical rating scale scores, and kinetic gait analysis confirmed improvement in peak vertical force in the dogs lame on a forelimb from 8.85% reduction, with the difference between the worst-affected and the contralateral limb reducing by at least 80% for peak vertical force and vertical impulse at the final visit (Marshall et al., 2010).

5.2 Why That Number Is the Useful One

About six percent is a useful practical milestone rather than a biological threshold. For a 30 kg dog it is 1.8 kg; for a 10 kg dog it is 600 grams. In this small uncontrolled study improvement in clinical lameness scores became detectable at about that level, while the objective gait changes were evident closer to nine percent.

A smaller interim target can be more tangible for owners than a distant target weight alone, and it makes progress visible sooner.

The sample is fourteen dogs and the design is uncontrolled — there is no group that lost no weight — so the effect size should be held loosely. The objective gait measurement is what makes it more than an owner impression (chronic pain and osteoarthritis in dogs).

There is a behavioral corollary that belongs in the same conversation. Dogs in chronic pain have a lower threshold for defensive behavior, and pain in dogs is routinely missed because dogs do not display it the way people expect (why pain is missed in dogs). A weight problem and a behavior problem in the same dog are not always two problems.

6. Quality of Life

6.1 What Changed and What Did Not

Fifty obese dogs were enrolled in a weight-loss program with quality of life assessed by a validated owner-completed instrument scoring four domains; 30 completed the intervention.

After weight loss, vitality increased (P < 0.001), emotional disturbance decreased (P < 0.001) and pain decreased (P < 0.001). Anxiety did not change significantly (P = 0.09). Change in vitality score was positively associated with the percentage of weight lost and with body fat lost (German et al., 2012).

6.2 What That Design Can and Cannot Show

The association between the amount of weight lost and the change in vitality is interesting, and it does not rule out expectation effects: owners could see how much weight their dog had lost, so the more visible change could itself have driven the higher rating.

Against that, the raters were the owners, who knew their dogs were on a weight-loss program, and 20 of the 50 dogs did not complete. If the dogs that dropped out were doing worse, the completers are a favorable selection.

The anxiety result deserves attention precisely because it was negative. Three domains changed significantly and one did not, which argues against presenting the intervention as a global improvement in every aspect of quality of life — though on its own it does not separate true change from expectation (how quality of life is measured in dogs).

7. Where the Weight Comes From

7.1 Owner-Side Factors

The epidemiological study that measured body condition in 696 dogs also looked for associated factors, and found four: owner age, hours of weekly exercise, frequency of snacks and treats, and personal income. Awareness of the health risks of canine obesity was significantly lower among people in lower income brackets (Courcier et al., 2010).

Two of those four are modifiable by information and two are not. That is worth knowing before designing advice, because a leaflet addresses the treat frequency and does nothing about the rest.

The income finding deserves a cautious reading. The study shows an association between income and awareness of the health risks of obesity; it does not explain how that association arises.

One factor on that list deserves a caution. Hours of weekly exercise was associated with obesity in a cross-sectional design, which cannot say whether the dogs move less because they are heavy or are heavy because they move less — and the trial in section 8 suggests the causal arrow is weaker than the correlation implies (exercise and behavior).

7.2 Neutering

Two of the studies above touch this, and neither was designed to answer it. In the international weight-loss study, intact dogs lost more weight than neutered dogs and females more than males (Flanagan et al., 2017). The large life-span cohort studied neutered dogs exclusively, so it cannot compare.

What that supports is modest: in this weight-loss cohort neutered dogs lost less weight than intact dogs, and the study does not establish that neutering itself caused the difference. It is consistent with the known reduction in maintenance energy requirement after gonadectomy. What it does not support is a claim about how large the effect is, or that neutering causes obesity — a neutered dog fed at its new requirement does not gain weight, and the requirement is the part that changes.

The practical form of this is a timing point rather than a position on neutering. The weeks after the procedure are a sensible point at which to review the ration again (neutering and behavior).

7.3 Breed, and a Gene

Some of the variation is not about households at all. In Labrador Retrievers, a 14 base-pair deletion in the pro-opiomelanocortin gene (POMC) was found at an allele frequency of 12% and, among the other breeds tested, only in the closely related flat-coated retriever. The deletion was associated with body weight — a per-allele effect of 0.33 standard deviations — with adiposity, and with greater food motivation (Raffan et al., 2016).

So part of what owners experience as a dog that is always hungry has an identified molecular basis in at least one breed, and part of the breed difference in obesity risk is not a difference in owner behavior (behavioral genetics in dogs).

7.4 What the Gene Does Not Explain

A 12% allele frequency in one breed and its nearest relative explains a slice of the variation, not the 59.3% figure. Most overweight dogs do not carry it, and the breeds at highest recorded risk in the UK data — Pugs at the top — are not the breeds in which it was found.

The honest formulation is that food motivation has a heritable component, that it differs between individuals and between breeds, and that it changes how hard the same feeding plan is to run in a given household — not that it determines the outcome (breed versus behavior).

8. Diet and Exercise: What the Direct Comparison Found

8.1 Diet Against Exercise, Randomized

The advice an overweight dog's owner usually receives is to walk the dog more. One open-label randomized clinical trial tested that directly.

Thirteen overweight dogs, body condition score 6 to 9 on a nine-point scale, were randomized to either dietary caloric restriction or increased physical activity, with activity measured objectively by triaxial accelerometer. The activity group did become more active: vigorous activity increased significantly (P = 0.016). It was not enough. Median body weight decreased by 10% in the dietary restriction group against 2% in the physical activity group, and activity levels in the diet group did not change (Chapman et al., 2019).

In this small randomized trial median weight loss was 10% with dietary restriction against 2% with increased physical activity. The result favors dietary restriction for weight loss in this sample; the trial is far too small for the ten-to-two difference to be treated as a stable effect ratio.

8.2 Why This Is Not an Argument Against Exercise

Thirteen dogs is a small trial, and an open-label design cannot blind owners to which arm they are in. For weight reduction alone, the dietary intervention was markedly more effective in this sample than increased activity on its own.

Exercise nonetheless remains part of sensible weight management. Among other things it supports fitness, muscle mass, mobility and occupation, even though in this small trial it was weaker as a sole lever for weight reduction than the dietary intervention (exercise and behavior in dogs).

What it should not be is the only plan. An owner who adds a daily half hour of walking, changes nothing about intake, and sees two percent in three months concludes that their dog cannot lose weight.

8.3 What That Means for a Trainer

It relocates the conversation from the activity the trainer controls to the feeding the trainer does not. That is uncomfortable, and it is the honest position.

A trainer can still do three things that matter. Measure and show the body condition score, which nobody else in the dog's life is doing. Account for the food used in training, which is the one intake stream the trainer is directly responsible for. And decline to sell more exercise as a weight solution.

9. Does Weight Loss Actually Work in Practice?

9.1 The International Study

The largest multi-center weight-loss study to date covered 926 overweight dogs at 340 veterinary clinics in 27 countries over three months, using commercial weight-loss diets with an initial allowance of 250–335 kJ per kg target body weight to the power 0.75 per day (60–80 kcal on the same basis).

Of the 926 dogs, 896 (97%) lost weight, with mean weight loss of 11.4 ± 5.84%. Intact dogs lost more than neutered dogs and females more than males. Subjective scores for activity increased sequentially over the program (P < 0.001), as did quality of life (P < 0.001) (Flanagan et al., 2017).

The headline is that a structured program can produce substantial short-term weight loss: 97% of the enrolled dogs lost weight within three months. The study was industry-funded, several authors were employed by the manufacturer, and the diets were that manufacturer's commercial products — which makes independent replication worth asking for.

9.2 Food-Seeking Behavior Goes Down, Not Up

The finding in that study with the most practical value is counterintuitive. Scores for food-seeking behavior decreased sequentially through the program (P < 0.001).

The objection owners raise first is that a dog on reduced rations will beg constantly. These data say the opposite happened: begging scores fell as the program ran. One possible explanation is that a purpose-formulated weight-loss diet fed at a defined allowance produces greater satiety per calorie than simply feeding less of a maintenance diet, and that predictable meal structure matters too (diet and behavior in dogs). Fiber content and gastrointestinal satiety signaling are also discussed as possible mechanisms; the study itself could not test them (the gut–brain axis in dogs).

The caveat is the measurement: these were subjective scores recorded by owners who knew their dog was losing weight, in a study with no control arm. Such ratings can be influenced, at least in part, by expectation and observer effects. The finding is useful for countering an objection, not for asserting that restriction reduces hunger.

9.3 What the Study Cannot Tell Us

Three months is not the question that matters. The question is what the dog weighs in two years, and the study does not reach it.

Nor does a 97% success rate describe the general population. These were dogs whose owners enrolled in a structured program at a clinic, with a prescribed diet and scheduled weigh-ins. Dogs whose owners were told to feed less at a routine appointment are not in this sample, and neither are the dogs that dropped out before the first weigh-in.

The useful reading is conditional: in this selected group of participants a structured program went with substantial weight loss. Whether a comparable program is carried through in everyday practice, and the weight held, is a separate question.

10. The Behavior Side

10.1 Food Motivation Is Not a Character Flaw

The POMC finding has a consequence for how these conversations go. In at least one breed, part of what owners describe as a greedy dog is an identified genetic variant affecting appetite regulation. The dog is not being badly behaved and the owner is not being weak.

That reframing is worth more than it sounds. Weight conversations fail when they land as an accusation, and the owner who feels accused stops reporting accurately — which removes the only data the plan runs on.

It also sets a realistic expectation. A dog with high food motivation will not stop asking because the ration changed. The plan has to work with a dog that asks, rather than assume a dog that stops. How hard that is varies between individuals in ways that are not reducible to breed or to training history (temperament and coping styles).

10.2 Begging and What Maintains It

Intermittent reinforcement can make begging particularly persistent, because occasional successes are enough to keep further attempts going (reinforcement schedules).

Two implications follow. When reinforcement stops, begging may temporarily become more frequent or more intense, which is easily read as the diet making the dog hungrier (extinction and the return of learned behavior). And occasional individual successes can keep the behavior going even when most of the household stays consistent.

It is also worth separating a learned request from a physiological state before concluding which one is in front of you. Begging that persists immediately after feeding suggests that reinforcement history is contributing substantially, though it does not on its own rule out hunger or high food motivation (learned behavior versus emotional response). A dog that begins guarding food or chews after a ration reduction is telling you something different, and that one is worth taking seriously (resource guarding in dogs).

There is a separate question of what the dog is actually asking for. In some dogs attention, alongside food, can play a part in maintaining the behavior (attention-seeking behavior). Whether attention or food matters more can only be inferred from repeated changes to the consequences, not from a single observation.

10.3 Training With Food in an Overweight Dog

Reward-based training uses food, and food has calories. This is the point at which a dog school's practice intersects with the dog's body condition directly, and it is usually not counted.

The quantity is not negligible, and no study has measured it. Training rewards can contribute a meaningful fraction of daily energy intake, particularly in small dogs, and they are routinely left out when the ration is calculated.

The solution is bookkeeping rather than abstinence. Training food comes out of the daily ration rather than on top of it, the ration is weighed rather than scooped, and part of the dog's own measured food is used as the reward in low-difficulty work — reserving higher-value food for the training that genuinely needs it (how reward value works neurochemically).

Non-food reinforcement has a place here and should not be oversold. For some dogs play or access to sniffing works as well as food for some tasks; for others it does not, and substituting a weaker reinforcer to save calories buys worse learning.

11. The Older Dog, Where It All Arrives Together

11.1 Three Problems That Make Each Other Worse

In an old dog, excess weight rarely sits on its own. The combination that turns up is weight, joint pain and reduced activity, and each leg of it drives the other two: a painful dog moves less, a dog that moves less gains weight, and a heavier dog loads painful joints harder.

That loop is why the osteoarthritis weight-loss finding matters more in an old dog than in a young one. Intake is often the directly modifiable lever here, particularly where pain limits how much activity can be increased; at the same time pain management and adapted exercise belong in the overall plan (chronic pain and osteoarthritis).

11.2 Why Weight Loss Is Harder Here

Two reasons, pulling in opposite directions. Maintenance energy requirement falls with age and with reduced activity, so the ration that was correct at six is excessive at eleven without anything having changed. And the exercise lever, already the weaker one in a young dog, is weakest here — an arthritic dog cannot be walked into a calorie deficit.

The consequence is that adjusting energy intake often becomes especially important in an old dog, above all where pain or other conditions limit how much activity can be increased.

11.3 What Is Not a Weight Problem

Weight loss in an old dog without a deliberate restriction is a clinical finding, not a success. So is weight gain with no change in feeding.

And some of what looks like laziness is not. Reduced activity in an older dog can reflect pain, a developing cardiac or respiratory problem, or cognitive change, and the behavioral signs of cognitive dysfunction include altered activity levels (cognitive dysfunction syndrome). Attributing it to the weight, and the weight to the inactivity, closes a loop that should have started with an examination.

12. What a Dog School Can Actually Do

12.1 Measure, Do Not Estimate

Body condition score at intake, written down, with the owner's hands on the dog at the same time. It takes under a minute, it produces a shared baseline, and it converts a judgment into an observation.

Weight in kilograms at the same time, on the same scale, with the date. The number is uninformative on its own and highly informative as a series — which is the same logic that applies to any behavioral measure taken repeatedly rather than once (behavioral assessment in dogs).

A photograph from above and from the side. This is the comparison that works six months later, because the owner's memory of how the dog looked has moved with the dog.

12.2 Count the Treats

The one intake stream a trainer is responsible for is the one used in training, and counting it is both the honest thing and the cheapest intervention available. Treats weighed into a day's allowance, homework specified in grams rather than in handfuls, and the ration adjusted for what the training consumes.

This also removes a conflict of interest that is rarely named. A class that sells high-value treats and prescribes daily reward-based homework has a commercial interest in an intake stream it is not measuring.

12.3 Where the Boundary Is

Diet formulation, calorie prescription and the investigation of an endocrine cause are veterinary work, and a trainer who crosses into them is doing harm in two directions: the advice may be wrong, and it delays the examination.

Two findings make that boundary concrete. The one small trial that compared the two approaches directly favored dietary restriction over increased activity, which is the veterinarian's territory. And a dog that has gained weight without a change in feeding needs an examination rather than a plan.

What is squarely within a trainer's remit is the measurement, the training-food accounting, the behavior around food, and the conversation that gets an owner to the examination in the first place — which, given that six in ten dogs are above ideal and seven percent have it recorded, is frequently the limiting step.

13. Research Gaps

The prevalence discrepancy has not been studied as such. Recorded prevalence of 7.1% against measured prevalence of 59.3% in comparable populations is a measurement problem large enough to deserve its own investigation, and nobody has quantified how much of it is recording behavior.

There is no long-term follow-up. The largest weight-loss study ran for three months. What proportion of dogs that reach target weight hold it at two years, and what predicts regain, is unknown.

The diet-versus-exercise trial needs repeating at scale. Thirteen dogs, open label, one of the most practically consequential findings in the field. It should not rest on a sample that size.

Weight loss in an adult dog has not been shown to extend life. The lifetime restriction study used restriction from puppyhood; the life-span data for body condition are observational. Whether slimming a middle-aged dog recovers any of the lost years is untested and probably untestable at reasonable cost.

The canine comorbidity picture is underdescribed. The review that looked found the human associations largely not established in dogs. That is a gap in the evidence rather than evidence of absence, and it leaves the clinical argument resting on life span, orthopedics and quality of life.

Training-food intake has never been quantified. No study has measured what proportion of daily energy intake comes from training rewards in dogs attending classes, despite reward-based training being the recommended standard.

14. Conclusion

Excess weight in dogs is a common and clinically relevant problem that is recorded in clinical notes far less often than direct body-condition assessment would suggest. In one UK practice sample 59.3% of the dogs assessed directly were above ideal condition, while a separate study found 7.1% recorded overweight status in clinical records. Owners also underestimate excess weight comparatively often.

The consequences are documented without needing the borrowed human arguments. Median life span was lower in overweight dogs in all twelve breeds of a 50,787-dog study, by as much as 2.5 years in small breeds. The one lifetime experiment gave 13.0 years against 11.2 for a 25% restriction from puppyhood. In osteoarthritic dogs, clinical lameness scores improved from about 6% of body weight lost onward, with objective gait changes closer to nine percent. Owner-rated quality of life improved for vitality, emotional disturbance and pain, and did not change for anxiety.

The part that should change practice is the one direct comparison, small as it is. Dietary restriction produced 10% weight loss against 2% for increased activity, and the activity group genuinely moved more. Thirteen dogs cannot fix the size of that difference, but they are enough to say that the standard recommendation — walk the dog further — is the weaker of the two levers, and that recommending it alone sets an owner up to conclude their dog cannot lose weight.

For a dog school the practical position is narrow and defensible. Measure body condition and write it down, because almost nobody else does. Count the food used in training, because that is the intake stream the school is responsible for. Treat food-seeking as a learning problem with an intermittent schedule behind it rather than as evidence that the dog is starving. Send the diet question to the veterinarian, where the effective lever actually sits. And raise the subject at all, which, given the gap between 59.3% and 7.1%, is where the whole thing stalls.

Key Insights

Measured and recorded prevalence differ by a factor of eight. 59.3% above ideal when dogs are scored; 7.1% recorded in clinical notes.

Asking the owner does not work as screening. Among owners of overweight dogs who disagreed with the veterinarian, 39% called the weight normal.

Owners plausibly calibrate against an overweight population. A dog at ideal condition can look lean by that standard — an explanation, not a tested finding.

Overweight status went with shorter median life span in all twelve breeds. Up to 2.5 years in Yorkshire Terriers, 0.4 in German Shepherd Dogs.

The association was larger in several of the smaller breeds. Hazard ratios ran from 2.86 down to 1.35.

Lifelong 25% restriction gave 13.0 years against 11.2. Forty-eight littermate-paired Labradors, the only experiment of its kind.

Modest weight loss improved lameness. Detectable from about 6% — 1.8 kg on a 30 kg dog — in fourteen dogs, uncontrolled.

Dietary restriction outperformed added exercise in the one randomized comparison. 10% against 2% in thirteen dogs, and the exercise group really did move more.

Food-seeking scores fell during weight loss, they did not rise. Owner-reported, uncontrolled and industry-funded, but it answers the commonest objection.

The human comorbidity list does not transfer cleanly. The obesity–diabetes association recognized in people and cats is not established in dogs.

Appetite has a genetic component. A POMC deletion at 12% allele frequency in Labradors, associated with weight and food motivation.

Training food is an unmeasured intake stream. It comes out of the ration, not on top of it, and nobody has studied how much it is.

References

Chandler, M., Cunningham, S., Lund, E. M., Khanna, C., Naramore, R., Patel, A., & Day, M. J. (2017). Obesity and associated comorbidities in people and companion animals: a One Health perspective. Journal of Comparative Pathology, 156(4), 296–309. https://doi.org/10.1016/j.jcpa.2017.03.006

Chapman, M., Woods, G. R. T., Ladha, C., Westgarth, C., & German, A. J. (2019). An open-label randomised clinical trial to compare the efficacy of dietary caloric restriction and physical activity for weight loss in overweight pet dogs. The Veterinary Journal, 243, 65–73. https://doi.org/10.1016/j.tvjl.2018.11.013

Courcier, E. A., Thomson, R. M., Mellor, D. J., & Yam, P. S. (2010). An epidemiological study of environmental factors associated with canine obesity. Journal of Small Animal Practice, 51(7), 362–367. https://doi.org/10.1111/j.1748-5827.2010.00933.x

Flanagan, J., Bissot, T., Hours, M.-A., Moreno, B., Feugier, A., & German, A. J. (2017). Success of a weight loss plan for overweight dogs: the results of an international weight loss study. PLOS ONE, 12(9), e0184199. https://doi.org/10.1371/journal.pone.0184199

German, A. J., Holden, S. L., Wiseman-Orr, M. L., Reid, J., Nolan, A. M., Biourge, V., Morris, P. J., & Scott, E. M. (2012). Quality of life is reduced in obese dogs but improves after successful weight loss. The Veterinary Journal, 192(3), 428–434. https://doi.org/10.1016/j.tvjl.2011.09.015

Kealy, R. D., Lawler, D. F., Ballam, J. M., Mantz, S. L., Biery, D. N., Greeley, E. H., Lust, G., Segre, M., Smith, G. K., & Stowe, H. D. (2002). Effects of diet restriction on life span and age-related changes in dogs. Journal of the American Veterinary Medical Association, 220(9), 1315–1320. https://doi.org/10.2460/javma.2002.220.1315

Marshall, W. G., Hazewinkel, H. A. W., Mullen, D., De Meyer, G., Baert, K., & Carmichael, S. (2010). The effect of weight loss on lameness in obese dogs with osteoarthritis. Veterinary Research Communications, 34(3), 241–253. https://doi.org/10.1007/s11259-010-9348-7

Pegram, C., Raffan, E., White, E., Ashworth, A. H., Brodbelt, D. C., Church, D. B., & O’Neill, D. G. (2021). Frequency, breed predisposition and demographic risk factors for overweight status in dogs in the UK. Journal of Small Animal Practice, 62(7), 521–530. https://doi.org/10.1111/jsap.13325

Raffan, E., Dennis, R. J., O’Donovan, C. J., Becker, J. M., Scott, R. A., Smith, S. P., Withers, D. J., Wood, C. J., Conci, E., Clements, D. N., Summers, K. M., German, A. J., Mellersh, C. S., Arendt, M. L., Iyemere, V. P., Withers, E., Söder, J., Wernersson, S., Andersson, G., Lindblad-Toh, K., Yeo, G. S. H., & O’Rahilly, S. (2016). A deletion in the canine POMC gene is associated with weight and appetite in obesity-prone Labrador retriever dogs. Cell Metabolism, 23(5), 893–900. https://doi.org/10.1016/j.cmet.2016.04.012

Salt, C., Morris, P. J., Wilson, D., Lund, E. M., & German, A. J. (2019). Association between life span and body condition in neutered client-owned dogs. Journal of Veterinary Internal Medicine, 33(1), 89–99. https://doi.org/10.1111/jvim.15367

White, G. A., Hobson-West, P., Cobb, K., Craigon, J., Hammond, R., & Millar, K. M. (2011). Canine obesity: is there a difference between veterinarian and owner perception? Journal of Small Animal Practice, 52(12), 622–626. https://doi.org/10.1111/j.1748-5827.2011.01138.x