CBD for pets: the dog and cat guide

Pet owners ask the same ten questions, in roughly the same order. This guide answers each one from the published dog and cat studies, the Merck Veterinary Manual, the ASPCA Animal Poison Control Center, the AVMA and the FDA. Every number carries a link. Where the evidence is thin, the text says so. Nothing here replaces a conversation with the veterinarian who knows your animal.

Is CBD safe for dogs? What the published studies measured

The first controlled trial came from Cornell in 2018. Dogs received 2 mg/kg of CBD oil or placebo every 12 hours for 4 weeks each in a crossover design, with a 2-week washout between arms. Blood work showed an elimination half-life of 4.2 hours and a statistically significant rise in alkaline phosphatase during the CBD arm, while owners reported no side effects (Gamble 2018). The study was small: 22 dogs recruited, 16 analyzed.

The longest trial ran 6 months. Twenty dogs took 4 mg/kg of a THC-free broad-spectrum distillate once daily, and twenty took placebo, with blood chemistry, hematology and urinalysis at weeks 2, 4, 10, 18 and 26. The authors reported a "transient elevation in alkaline phosphatase (ALP) in just over half of the dogs" receiving CBD, with no other liver markers moving and a strong correlation (r > 0.9) to the bone isoform of the enzyme (Bradley 2022). A separate 90-day study dosed 32 beagles at 5 mg/kg per day across three formats (broad-spectrum CBD alone, with CBG, and with CBDA) and concluded the cannabinoids were well tolerated; the same paper reports a post-market rate of 2.10 adverse events per 1 million administrations sold (Bookout 2024).

The escalation data are the most useful for owners worried about an accidental double serving. In a blinded study of 20 beagles, dogs tolerated a CBD-dominant oil up to the tenth escalating dose of roughly 62 mg/kg, and 94.9% of all recorded adverse events across the groups were mild. The more serious events, lethargy, low body temperature and loss of coordination, clustered in the two groups whose oils contained THC (Vaughn 2020). The signal across every study is the same: CBD at label-range doses produced mild, reversible findings in healthy adult dogs, and THC, not CBD, produced the serious ones.

None of this is FDA approval. The FDA has approved no cannabis product for any animal use and says it cannot ensure these products are safe or effective; as of its current Q&A, the agency had "not directly received any reports of adverse events associated with animals given cannabis products" (FDA). Several of the studies above were funded by product companies or trade groups, which the papers disclose. Read them as safety screens in healthy dogs, not as proof of benefit.

Can cats take CBD?

The 12-week study used a 50/50 CBD and CBDA fish-oil preparation in cats and a soft chew in dogs. Time to peak was 2 hours in cats and 1.4 hours in dogs. One of the eight cats showed a persistent rise in the liver enzyme ALT above the reference range for the whole trial, and the cats licked excessively in 35.4% of observation periods and shook their heads in 25.2%, which the authors recorded as adverse effects of oil administration (Deabold 2019). The lower serum concentrations are the key finding: a cat on the same mg/kg as a dog carries far less CBD in its blood.

The second study pushed doses much higher. Twenty healthy adult cats in five groups of four received up to 11 escalating oral doses of CBD oil, THC oil, a CBD/THC blend or placebo, with at least 3 days between doses. Every cat reached the top of its titration: 30.5 mg/kg of CBD, 41.5 mg/kg of THC, or 13.0:8.4 mg/kg of CBD:THC. All adverse events were mild and transient, blood counts and chemistry did not change meaningfully, and the authors noted that "Gastrointestinal AEs were more common with formulations containing MCT" while lethargy, low body temperature, wobbliness and a visible third eyelid were more common with the THC-containing oils (Kulpa 2021).

What this means in practice: a cat-specific product in a fish-oil base, given on food rather than squirted into the mouth, matches what the studies actually did. A 4 kg cat at the studied 2 mg/kg gets 8 mg per serving, which from a 33 mg/mL dog tincture is 0.24 mL, a volume most droppers cannot measure cleanly. The AVMA's updated report, published April 21, 2025, now covers safety and pharmacokinetic data for cats as well as dogs, which is the document a veterinarian will reach for (AVMA).

What is the difference between pet CBD and human CBD?

Xylitol is the clearest reason not to share a human gummy with a dog. The ASPCA Animal Poison Control Center says that in dogs the sweetener causes a sudden insulin release, dropping blood sugar within hours, and that liver enzymes often rise after ingestion; it names sugar-free peanut butter, candy, chewable and gummy vitamins and liquid prescription products among the sources. Its advice for a dog that seems normal is to call 888-426-4435, and for a dog showing signs to go to a veterinarian at once with the packaging (ASPCA). Most human CBD gummies are sweetened with sugar or other sugar alcohols, but the label has to be read every time.

Concentration is the second difference. A 1,500 mg per 30 mL human tincture delivers 50 mg per mL. A 10 kg dog at the studied 2 mg/kg needs 20 mg, which is 0.4 mL of that tincture and 0.6 mL of a 33 mg/mL dog tincture; the larger volume is easier to read on a 1 mL dropper. The dog studies used 2 mg/kg to 5 mg/kg per day (Gamble 2018; Bookout 2024), and a pet product is built so those numbers land on a dropper line.

Label accuracy is the third. A 2020 survey of 29 hemp products marketed for dogs found only 10 of 29 within 90% to 110% of their label claim, heavy metal contamination in 4 of 29 with lead the most common, and all 29 below the federal 0.3% THC line (Wakshlag 2020). The AVMA makes the same point, citing lab analyses showing products "labelled inaccurately with respect to both the identity and amount of active ingredient" (AVMA). A pet product with a current, batch-matched COA closes that gap; a human product bought for a dog usually arrives without one.

What are CBDA and CBGA?

The CBDA data come from six beagles dosed at 2 mg/kg of combined CBD/CBDA in three oral forms: a mixed-triglyceride oil, a lecithin and sesame oil base, and a chewable. Across the forms, "CBDA and THCA concentrations were 2- to 3-fold higher than CBD and THC" in serum, and the lecithin-sesame form held CBDA at roughly 192 to 196 ng/mL at weeks 1 and 2 against 44 to 75 ng/mL for the others, which the authors read as superior absorption or retention of CBDA (Wakshlag 2020). The half-lives were similar, about 4 hours for both CBD and CBDA.

The CBGA data are newer. Dogs given an equal CBG/CBGA hemp formulation at 2 mg/kg twice daily for 2 weeks, first fasted and then fed, held mean serum CBGA at 250 ng/mL against 5 ng/mL for CBG, with no difference between fed and fasted states, and the physical exams, blood counts and chemistry suggested no adverse events over the trial (Amstutz 2022). A 2026 follow-up gave eight beagles a whole-hemp extract supplying 1 mg/kg of CBDA plus 1 mg/kg of CBGA twice daily for 90 days, measured 24-hour exposures of 1,383 and 930 ng·h/mL, and reported that "All physical exam findings and blood work were unremarkable" (Koll 2026).

For a label, the practical question is which milligrams are being counted. A COA reports CBD, CBDA, CBG and CBGA on separate lines. Ask whether the number on the front of the bottle is CBD alone or the sum of the four, and compare it to the COA before working out a serving. The acid forms are not inert fillers; in dogs they are the compounds that show up most in blood.

Why is THC a problem for dogs?

The reason is anatomy. The Merck Veterinary Manual notes that dogs carry a higher concentration of CB1 receptors in the cerebellum and medulla oblongata, which is why loss of coordination and heart and breathing effects show up at doses a person would barely feel. In Pet Poison Helpline cases from January 2018 through February 2023, the most common signs were lethargy (30%), ataxia (21%) and vomiting (15%), with urinary dribbling, sound sensitivity, trembling, slow heart rate, dilated pupils and abnormal body temperature also reported. The manual describes "a high morbidity rate but an extremely low mortality rate," with the probable oral lethal dose above 3 g to 9 g of THC-dominant plant material per kilogram (Merck Veterinary Manual).

Edibles are the usual route. The ASPCA's veterinary resource explains that edibles smell and taste like baked goods, which is why dogs go for them, and that cats are more likely to chew the plant itself; it also warns that decontamination is often unnecessary and can do more harm than good once a dog is already dull and wobbly, because of the aspiration risk (ASPCA). The AVMA adds that chocolate, raisins or xylitol in the same edible worsen the prognosis and that "Several deaths have been reported due to cannabis toxicity," tied to complications such as aspiration (AVMA). Cornell's canine health center puts onset at 30 minutes to several hours and says signs may persist up to 72 hours, with most dogs recovering fully in 1 to 2 days (Cornell).

Two practical notes follow. First, over-the-counter human urine drug tests are unreliable in dogs; false negatives are very common, so a wobbly dog with a known exposure needs a veterinarian, not a test strip (Merck Veterinary Manual). Second, federal hemp is allowed up to 0.3% THC by dry weight (7 U.S.C. 1639o), so a full-spectrum product is not THC-free. A broad-spectrum pet product whose COA reports THC as not detected is the simplest way to keep THC out of the picture, and the suspected-exposure number is the same ASPCA line, 888-426-4435.

How do you give a dog a tincture?

Start with the math. A 1,000 mg tincture in a 30 mL bottle is 33.3 mg per mL, so a standard 1 mL dropper holds 33 mg at the top line and about 8 mg at the 0.25 mL mark. Draw to the line, check it at eye level, and dispense onto something the dog will eat on its own. Squirting into the mouth works for some dogs, but the cat data show how much an animal can waste by licking and shaking, and the same losses happen with a dog that fights the dropper.

Timing follows the pharmacokinetics. Dogs reached peak serum CBD at 1.4 hours and the elimination half-life was about 4.2 hours, which is why the trials dosed every 12 hours (Deabold 2019; Gamble 2018). For the acid cannabinoids, giving the dose fed or fasted made no measurable difference to serum CBG or CBGA (Amstutz 2022). Pick morning and evening meals and keep the schedule.

Keep a two-line log: date and serving, plus anything you notice. The studies flagged lethargy, loose stool and, in blood work, a transient rise in alkaline phosphatase (Bradley 2022); if a dog on CBD is due for routine blood work, tell the clinic so an ALP reading is interpreted with that in mind. Change one thing at a time, hold each change for 2 weeks, and bring the log to the veterinarian.

What does the AVMA say?

On evidence, the AVMA's position is that cannabis-derived products suggested for animals have not followed the FDA approval path, that the scientific evidence for their use in animals is limited, and that it supports well-controlled clinical research and pursuit of FDA approval by manufacturers. It notes that Epidiolex, one approved human CBD drug, plus three approved synthetic cannabinoid drugs, can be used extralabel by veterinarians under federal rules (AVMA). That is a narrow door and it runs through a veterinarian, not a retailer.

On regulation, the page tracks the January 2021 USDA hemp production rule, the FDA's run of warning letters, and the FDA's January 2023 statement that only a new regulatory pathway could provide oversight for CBD products. The AVMA records that the FDA does not intend to address CBD in animal food because it is "not apparent how CBD products could meet the safety standard" for substances in animal food (AVMA). The page also says laws on hemp-derived products in animals have been implemented in many states, which is why what a veterinarian may discuss or recommend differs by state.

The updated report itself, "Cannabis in Veterinary Medicine," keeps the earlier material on pharmacology, the endocannabinoid system and analytic testing, and adds safety and pharmacokinetic data for dogs and cats, clinical trials in dogs, cats and horses, and federal and state regulatory updates (AVMA). The social media toolkit and FAQ on the same hub are written for veterinarians answering client questions, and a client can ask for them by name.

What should a pet COA show?

Start with the potency table. The federal definition of hemp is a total THC concentration, including THCa, of not more than 0.3% on a dry-weight basis (7 U.S.C. 1639o), so the THC line matters even on a product sold as THC-free. Look for the lab's limit of quantitation next to any "ND" result: not detected at a 0.05% LOQ is a weaker statement than not detected at 0.01%. Then add up the cannabinoid lines and compare the sum to the front label, remembering that the acid forms are listed separately.

The 29-product survey shows why. Products were pulled from an internet search for hemp extracts marketed for dogs and tested at an ISO/IEC 17025 lab. Only 10 of 29 were within 90% to 110% of their label claim; 4 of 29 contained heavy metals, with lead the most prevalent at 3 of 29, and for 2 products the arsenic or lead level exceeded the limits Massachusetts sets for cannabis products. All 29 were below 0.3% THC, and the authors counted "only 18 of 29 being appropriately labeled" under FDA labeling rules, recommending that owners and veterinarians obtain a certificate of analysis before dosing (Wakshlag 2020).

The contaminant panels round it out: residual solvents for extracted products, heavy metals (arsenic, cadmium, lead, mercury), pesticides, and microbial counts. A COA dated years ago, or one for a different batch, says nothing about the bottle in hand. Each product page on this site links its current batch COA; the general reading guide is at /questions/what-is-a-coa.

How do you pick a strength by weight?

The table converts the studied range into dropper volumes for a 1,000 mg per 30 mL tincture. Divide pounds by 2.2 to get kilograms.

Body weight	2 mg/kg per day	5 mg/kg per day	Volume at 33 mg/mL
5 kg (11 lb)	10 mg	25 mg	0.3 mL to 0.75 mL
10 kg (22 lb)	20 mg	50 mg	0.6 mL to 1.5 mL
20 kg (44 lb)	40 mg	100 mg	1.2 mL to 3 mL
30 kg (66 lb)	60 mg	150 mg	1.8 mL to 4.5 mL

Two cautions on the numbers. The 2 mg/kg figure in the first trial was given twice daily, so the daily total was 4 mg/kg (Gamble 2018); the 6-month trial used 4 mg/kg once daily (Bradley 2022); and the 90-day trial used 5 mg/kg per day of total cannabinoids, not CBD alone (Bookout 2024). Most owners begin below the studied range, at roughly 0.5 mg/kg to 1 mg/kg, and move up in 2-week steps. That is a conservative reading of the data, not a recommendation from any study.

Strength also sets how long a bottle lasts. At 20 mg per day, a 1,000 mg bottle covers 50 days; at 60 mg per day, 16 days. For a large dog, a higher-concentration bottle keeps the volume on the dropper readable. For a cat, the 2 mg/kg used in the 12-week study puts a 4 kg animal at 8 mg per serving, which argues for a lower-concentration cat product rather than a dog tincture (Deabold 2019). The serving calculator on this page does the arithmetic; the veterinarian decides whether the plan fits the animal.

What do state rules say about pet products?

The FDA's Q&A states the federal position in three parts: it has approved no cannabis product for animals, it cautions pet owners against such products, and no hemp-derived ingredient is approved for animal food (FDA). Enforcement has been claims-driven. On May 26, 2022 the Center for Veterinary Medicine warned four companies selling CBD for food-producing animals, stating that "The FDA has not approved any human or animal products containing CBD" and pointing to the lack of data on residues in milk, eggs and meat (FDA CVM). Pet brands appear in the agency's letter index in 2015 and again on April 7, 2025 (FDA).

That is why this site uses structure and function language only, names no conditions, and asks reviewers to do the same. A product is marketed as supporting a routine, not as doing anything to a named problem.

Oregon is the clearest state example. From January 1, 2026, any hemp item sold to an Oregon consumer that is "intended for human or animal consumption or use" and is not purely topical must be registered with the OLCC and labeled to OLCC rules before sale, and the FAQ says the requirement applies "whether online or in person." Items with enough THC to count as adult-use are restricted to buyers 21 and older (OLCC). Other states gate hemp retail with licenses or registrations of their own. The pet tinctures in this catalog are broad-spectrum with THC reported as not detected, which keeps them in the least-restricted shipping lane; the state-by-state matrix, updated as rules change, lives at /learn/cbd-legality-by-state.