Introduction

In recent years, the use of medical cannabis has received renewed interest in both human and veterinary medicine. After the discovery of the endocannabinoid system in the late 1980s [1], researchers have been exploring the system as a potential treatment target [2]. Cannabis is the popular name for the plant Cannabis sativa and products derived from the plant. Cannabis sativa contains over 100 cannabinoids with cannabidiol (CBD) and Δ9-tetrahydrocannabinol (THC) being the two main cannabinoids [3]. In human medicine, a substantial amount of clinical research concerning medical cannabis has been conducted and is currently ongoing. A search in the WHO International Clinical Trials Registry Platform in September 2022, showed 265 active clinical trials registered [4]. Several pre-clinical and pharmacokinetic studies on veterinary cannabinoids have been conducted [5–18] but only a limited number of clinical trials in dogs have been published. The majority of studies have focused on investigating the effects of cannabinoids on osteoarthritis-related pain, but there have also been reports of studies on idiopathic epilepsy, behavioural modification, and canine atopic dermatitis [16, 19–31].

There is an increasing availability of commercial cannabis-based extracts, nutraceuticals, supplements, feed additives, and treats aimed at companion animals. Unlike medical products, these types of cannabis-based products are not subject to regulation, including standardized evaluation of cannabinoid quality and quantity, screening for contaminants, and microbial testing. In a market review by Bonn-Miller et al. (2017), the content and labelling of 84 commercially available CBD extracts on the US market were analyzed. Of these, only 32% correctly labeled the actual CBD content, with 42% of products being under-labelled, and 26% being over-labelled. In addition, non-labelled cannabinoids, including THC, were found in up to 20% of the products [32]. Similar results have been reported in commercially available CBD extracts in the European Union, with some products containing potentially hazardous contaminants and most lacking information about THC content [33]. A study by Wakshlag et al. (2020) analyzed the content of THC, CBD, other selected cannabinoids, terpenes, and heavy metals in 29 full-spectrum commercially available veterinary products. They found that only 10 of the 29 products fell within 10% of the label claims regarding cannabinoid content, and 4 of the 29 products contained heavy metals [34]. When dog owners self-prescribe these unregulated CBD products, the lack of consistency in labelling content poses a significant challenge to accurately determining dosage and assessing effects. In the worst-case scenario, this inconsistency may pose a health risk to the dog, particularly concerning THC and heavy metal intoxication. In Denmark recreational cannabis is prohibited, and medical cannabis has been illegal up until 2018 when a national pilot programme was passed by the Danish government, allowing human physicians to prescribe medical cannabis for certain indications. The purpose of the programme was to enable the prescription of medical cannabis to patients experiencing a lack of effect from conventional therapy. Veterinary prescription is not included in the programme [35]. In Denmark, there are no approved veterinary medical cannabis products, and no veterinary cannabinoid supplements or feed additives are registered by the Danish Veterinary and Food Administration, Ministry of Food, Agriculture and Fisheries. As a result, it is illegal to sell or prescribe cannabinoids for the treatment of veterinary patients. There are many anecdotal reports on the use and effect of cannabinoid products for various health conditions in companion animals. Three surveys have investigated the use of veterinary cannabinoids among companion animal owners in the US and Canada. The results showed that the most frequent reasons for use were managing seizures and pain, and there was a generally positive perception of its effectiveness by the owners [36–38].

This study aimed to investigate the use of unauthorized cannabinoids in Danish dogs, including indications for use as well as the owner-perceived effect. An underlying incentive was furthermore to provide information on indications with high owner-perceived effects, which future academic research could build upon. Given the lack of legal veterinary cannabinoid products, an anonymous survey was deemed the most appropriate method to obtain this information.

Material and methods

Results

Out of 2,091 respondents who initiated the survey, 2,002 respondents completed the survey for a completion rate of 95.7%. Only completed surveys were included in the analyses.

Demographics

The crude demographics showed respondents were concentrated around the two largest cities in Denmark (capital region, n = 1,044, 52% and central Jutland, n = 343, 17%). The weight of the respondents’ dogs was distributed as 24% (n = 486) being under 10 kg, 24% (n = 470) between 10 and 20 kg, 25% (n = 501) between 20 and 30 kg, and 27% (n = 545) over 30 kg (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of the respondents.

https://doi.org/10.1371/journal.pone.0296698.t001

The indication for the use of cannabinoids

Pain management was the most common reason for using cannabinoids, with 36% (n = 376) of respondents reporting this as a reason. The second most common indication was allergy (11%, n = 122), followed by disease prevention/well-being (9%, n = 100) and seizure control (4%, n = 40). Of the inappropriate free text responses nine respondents recorded “epilepsy” in the free text and not in the predefined category “seizure control”. Similarly, eight respondents recorded “well-being” as free text, instead of indicating “disease prevention or well-being purposes” in the predefined category. Two additional major categories of indication were identified from the 28% (n = 299) of respondents who typed in other indications as free text. These were “behavioural issues” (n = 154), and “dermatological disease” (n = 34). Overall, the behavioural issue category was the second-largest indication group and included conditions such as anxiety, stress, fireworks-, and thunder phobia. The remaining indications were miscellaneous and included: neurological disease including dementia, intervertebral disc disease, and meningitis (n = 21); respiratory disease including pneumonia, cough, and excess sputum production (n = 12); cancer including treatment of splenic tumor, of an undiagnosed tumor, and cancer prevention (n = 3), palliative purposes (n = 5); reproductive issues including stimulation of neonatal weight gain, stabilizing hormonal status in a pregnant bitch, and mastitis (n = 2); immunological disease including autoimmune disease (n = 2); as an anti-inflammatory agent (n = 2); renal disease (n = 2); pancreatitis (n = 1); heart disease (n = 1); prostatic disease (n = 1); as bacteriostatic agent (n = 1); immune-stimulation (n = 1), and prevention of medical side effects from other treatments (n = 1). The corrected distribution of indications for cannabinoid use is reflected in Table 2, which considers the adjustments made to the categorical sums.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. The distribution of the indications for which owners administered cannabinoids to their dogs.

https://doi.org/10.1371/journal.pone.0296698.t002

Owner-perceived efficacy of cannabinoid treatment

Out of the 752 respondents who had used cannabinoids for their dogs, 48% (n = 363) reported a “very convincing and good effect”, 29% (n = 217) indicated “convincing with some effect”, whilst 18% (n = 137) were uncertain of the effect (“possible, but not convincing effect”), and only 5% (n = 35) reported observing “no effect”.

Difference in owner-perceived efficacy between indication groups.

When analyzing the indication-specific owner-perceived effect of the cannabinoid treatment, only respondents reporting one treatment indication were included, as it was not possible to distinguish the effect between various indications. Of the 752 respondents confirming use of cannabinoids, 69% (n = 516) used cannabinoids for only one indication and the majority reported some degree of effect. For the three largest indication groups, pain management (n = 249), behavioural issue (n = 103), and allergy (n = 47), the perceived overall effect was high (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. The owner-perceived effect of cannabinoid treatment in different indication groups.

https://doi.org/10.1371/journal.pone.0296698.t003

The respondent group using cannabinoids for pain management reported the highest percentage of perceived effect. Seventy-seven percent (n = 193) indicated either “very convincing and good effect” or “convincing with some effect”. Only five percent (n = 13) reported “no effect”, and 17% (n = 43) were unsure of the effect (“possible, but not convincing effect”). For behavioural issues, the majority (67%, n = 69) observed an effect. Nine percent (n = 9) reported “no effect”, and 24% (n = 25) reported a “possible, but not convincing effect”. When used for treatment of allergy, 57% (n = 27) reported a positive effect with an equal distribution of “very convincing and good effect” (30%, n = 14) and “convincing with some effect” (28%, n = 13). When the categories for effect in the allergy group were analyzed individually, the largest number of respondents were in the category “possible, but not convincing effect” with 34% (n = 16) indicating a lack of obvious effect. Nine percent (n = 4) reported “no effect”. The remaining indication groups had low total respondent numbers (n = 1–25). The indication that had the highest percentage of owners not perceiving any effect (12%, n = 4) was when used for seizure control (n = 34). As differences in the owner-perceived effect between the indication groups could reflect a true difference in response to treatment of different disease states a Kruskal-Wallis test was performed and provided weak evidence of a difference χ ²(9, N = 516) = 19.808, p < 0.019, E² = 0.40. A multiple comparison post hoc test with a Dunn’s test did however not show any significant differences in perceived effect between the indication groups (p >0.05, adjusted using the Bonferroni correction).

Discussion

The study explored the use and perceived efficacy of cannabinoid treatment in dogs in Denmark. The results indicate that cannabinoids are used for a variety of reasons, and most are purchased online. The majority of respondents perceived that the treatment improved their dogs’ medical condition or well-being. Despite the lack of licensed cannabinoid-containing products for pets, and even though it is illegal to sell or prescribe cannabinoids for animals in Denmark, this survey of over 2,000 dog owners found that 38% of the respondents had used a cannabinoid product. Three prior surveys in the US and Canada have explored the use and owner-perceived efficacy of cannabinoids among dog owners with sample sizes of 106, 632, and 1,068 [36–38]. The proportion of dog owners using cannabinoids was higher in the previous US and Canadian studies (79.8% and 78.3%) compared to the current Danish study [36, 38]. The difference may be due to study selection bias or the difference in the availability of cannabinoid products in North America and Denmark. In Denmark, the recreational use of cannabis is illegal, and human medical cannabis prescription is in its infancy, which may affect owner’s awareness and willingness to use cannabinoids in their dogs. This assumption is supported by a Slovenian study which found a slight association between owners’ personal experience with cannabinoid and their use in pets [39].

The most often stated indications for cannabinoid treatment were pain management, behavioural issues, and allergy. Similar results were reported in the North American surveys, where pain management, anxiety, and inflammatory disease were the most common indications [36, 38]. Besides these indications, the respondents reported a very large range of indications for which they used cannabinoids.

The majority of dog owners perceived a positive effect of the cannabinoid treatment for various indications. This could be due to a placebo effect or selection bias, or it could reflect a real therapeutic effect or improved quality of life for the treated dogs. However, there are no veterinary clinical trials to support or refute these findings for indications other than osteoarthritis-related pain, noise-induced anxiety, aggression, voluntary activity, canine atopic dermatitis, and seizures [16, 19–24, 26–28, 30, 31]. The indication with the highest owner-perceived efficacy was pain management where 77% experienced a positive effect with the majority reporting an obvious and good effect. The most common cannabinoid used in this group was CBD. Support for beneficial effects of CBD in the management of osteoarthritis-related pain in dogs has been reported in three double-blinded, randomized clinical trials [16, 20, 24]. Gamble et al. (2018) performed a randomized, placebo-controlled, cross-over, and double-blinded study with administration of 2 mg/kg CBD every 12 hours in 16 dogs. At two and four weeks into treatment, the Canine Brief Pain Inventory (CBPI) and a Hudson activity score showed a decrease in pain and an increase in activity compared to baseline (p < 0.01). The subjective veterinary clinical assessment in the study also showed a decrease in pain scores from baseline (p < 0.02) [16]. Similar findings were reported by Verrico et al. (2020) who investigated the effect of two CBD doses and liposomal CBD in 20 dogs with osteoarthritis-related pain. They found a decrease in pain estimated by the Helsinki Chronic Pain Index (HCPI) (p <0.01) in dogs treated with either 1.2 mg/kg/day CBD or 20 mg/day liposomal CBD [20]. Only one double-blinded, cross-over, and placebo-controlled study has used objective evaluation methods in addition to the subjective Clinical Metrology Instruments (CMI) in the investigation of the effect of 2.5 mg/kg CBD every 12 hours in osteoarthritis-related pain. Even though they found similar results in the subjective measurements (Liverpool Osteoarthritis in dogs and CBPI) as in the two previous studies, the objective measurements from accelerometry and objective gait analysis with pressure-sensitive walkway did not identify a significant difference between CBD and placebo groups after 6 weeks of treatment [24]. The disparity between the results from owner-driven subjective assessment tools (CMIs) and the objective assessment methods could indicate that the observations are reflections of CBD’s effect on other factors influencing the quality of life (better quality of sleep, reduced anxiety, better ability to cope with pain) than a true analgesic effect. Recent systematic reviews and meta-analyses’ have been conducted in the human field investigating the efficacy of cannabinoid treatment in different pain phenotypes. When evaluating the cannabinoid efficacy in chronic non-cancer pain patients, moderate quality of evidence for a small beneficial effect was reported [40], whilst when used in adult cancer-related pain patients the conclusion was that the effect of cannabinoids in addition to opioid treatment did not reduce pain in these patient phenotypes [41]. In our study, it was not possible to recognize which pain phenotypes (inflammatory pain, neuropathic pain, cancer-induced pain, chronic pain, acute pain) the owners were treating, but if dogs experience the same pain phenotypes as is described in humans, there might be certain subgroups of dogs in which cannabinoids are more effective than others. In contrast to the veterinary studies the human studies are based on cannabinoid products containing both CBD and THC, which could affect the outcome and efficacy of pain management. The second largest indication category was behavioural issues which included but was not restricted to treatment of anxiety, stress, firework- and thunder phobia. The owner-perceived effect of CBD treatment was high (67%). These findings are not supported by current studies investigating behavioural modification in dogs after CBD treatment. One study investigated the effect of 1.25 mg/kg CBD daily on aggression and stereotypic behaviour and did not recognize a difference between the treatment group and placebo group (p = 0.078) [23]. A second study aimed to investigate the effect of 0.7 mg/kg CBD every 12 hours on firework phobia and did not register any anxiolytic effect (plasma cortisol levels, pulse, anxiety-related behaviour) after the CBD treatment [22]. Interestingly the category with the highest percentage of owners not observing an effect was in the seizure group where 12% reported no effect. McGrath et al. (2019) published the first clinical trial using 2.5 mg/kg CBD every 12 hours as adjunctive therapy in epileptic dogs and reported a reduction in seizure frequency in the treatment group compared to the placebo group (p <0.01). Supporting these findings is a study by Garcia et al. (2022) who reported a 50% reduction in epileptic activity in six out of 14 dogs when treated with 2 mg/kg CBD compared to no reduction in epileptic activity in the placebo group (p = 0.02) [31]. The relatively high percentage of perceived lack of effect compared to other indications in our study could reflect several causes. The seizure group had few individuals compared to the pain management and behavioural issue group as well as seizure activity can be a very obvious clinical sign that is difficult to neglect or overlook by the owner. Clinical signs in the other categories can be relatively difficult to assess for owners and may even be overlooked and could thereby have been perceived as treatment effects by the owner.

One surprising finding was that 20 respondents indicated that they had purchased cannabinoid products through their veterinarian. The veterinary market in Denmark is highly regulated and it is illegal for a veterinarian to prescribe cannabinoids. The indications for use of the veterinarian-prescribed cannabinoids were pain management, behavioural issues, cancer, appetite stimulation, kidney disease and allergy. Even though many in vitro and in vivo studies have shown promising tendencies for treatment of all of the above-mentioned indications [42, 43], it is only the use in osteoarthritis-related pain management, canine atopic dermatitis and behavioural issue that is evidence-based and supported by canine clinical trials [16, 20, 22, 23, 28, 30].

Our study suggests that CBD drops/oil are the most popular type and formulation of cannabinoids and was used by 93% of the respondents confirming use of cannabinoids. It is not possible from the survey to identify the precise cannabinoid content or concentration in the used products. None of the respondents indicated using registered human medical cannabinoid products, which are the only products with controlled and documented cannabinoid content legally available in Denmark. A concern with the use of unauthorized cannabinoid products is the lack of content specification and batch analysis. The reported products in this study have the potential of not containing any cannabinoids. It has been documented that a discouraging number of veterinary products labelled as containing specific types and amounts of cannabinoids, had divergent content [34]. Furthermore, some respondents indicated using pure hemp seed oil which does not contain cannabinoids except if contaminated [44]. The uncertainty of content in the cannabinoid products is challenging for both dosing recommendations and effect evaluation, especially because the appropriate therapeutic dose could vary for different medical conditions. For seizure control in dogs, there is an indication that plasma concentration levels correlate with the effect [21], just as it has been shown that low-dose CBD (0.5 mg/kg/day) did not have the same positive effect on pain scores as higher CBD (1.2 mg/kg/day) concentrations [20]. Besides the challenges in dose recommendations and effect evaluation when using products with either unknown or varying cannabinoid content, it is especially critical regarding the THC concentration. Dogs are more sensitive to the psychoactive properties of THC compared to humans, possibly due to larger quantities of CB1 receptors in the brain, which can lead to high morbidity rates when exposed to THC [45]. In our study, very few reported intentionally using products primarily containing THC (4%).

Our study found that respondents living in Greater Copenhagen (Capital area) were less likely to supplement their dogs with cannabinoids. The reason for this is not clear, but it could be speculated to be linked to the higher educational level of the general population in this area, making these owners more adherent to evidence-based therapies, and less motivated for “alternative treatments”. This statement is however contradicted in the Slovenian study which did not find owner educational status as a predictor of treating pet animals with cannabinoids [39].

The authors would like to address some of the limitations of this study. First, it should be emphasized that this survey of owner-perceived effects should not be seen as a validation of the efficacy of any of the mentioned cannabinoid formulations or types. The study was designed as a short simple questionnaire for distribution via social media to secure as many respondents as possible. The potential bias with this type of self-selected convenience sampling is that dog owners with very strong opinions for or against cannabinoid use could be more motivated to participate (selection bias) and caution should be taken in concluding that this is a true representation of the general population. Secondly, the recognition of treatment effect is based on owner observations with a lack of standardized efficacy evaluation tools and without placebo or control groups (detection bias), resulting in a risk of reporting high numbers of both placebo and nocebo effects. Furthermore, the survey does not consider the owner’s capability to evaluate medical conditions and responses. Lastly, the treatments reported are not standardized or even documented regarding cannabinoid profile, content, or dosage.

Conclusion and perspectives

Danish dog owners administer cannabinoids for a broad range of different medical and behavioural conditions, despite no legal products being available on the Danish market. The majority of the owners perceive a very good or good effect of the cannabinoid treatment, especially when used for pain management. Most owners purchase the cannabinoid product online and most often use CBD drops/oil.

While caution must be exercised in interpreting the results, this study supports the need for more evidence-based knowledge within the companion animal field of cannabinoids, as some dog owners self-prescribe cannabinoids for a wide variety of medical conditions in their dogs. Pet owners do request information on cannabinoid products and currently, it is challenging for veterinarians to provide evidence-based information and dosing recommendations. Pet owners are to a high degree left to seek information and recommendations from commercial websites, family, friends, or anecdotal sources, which may not be factual or impartial.

As the majority of studies have focused on osteoarthritis-related pain, it is possible that cannabinoids may have a therapeutic rationale for additional veterinary indications or health-related conditions, not yet explored. Based on the findings of this study the authors suggest conductance of larger double-blinded, randomized and controlled clinical and dose-escalating trials, especially in areas of different pain phenotypes, behavioural modulation, and allergy. Future studies should be conducted using quality-controlled products with defined and documented cannabinoid content, profile, and concentrations and importantly should include objective assessment methods.

Supporting information

Acknowledgments

The authors would like to acknowledge and express their sincere thanks to Assistant Professor Anna Mueller, DVM, Ph.D. for input and management of the recruitment advert on the University Hospital for Companion Animals’ Facebook page enabling the high respondent numbers, to the Danish Kennel Klub and their magazine “Hunden”, as well as our veterinary colleagues who distributed the online survey and not least to all the Danish dog owners who showed interest and contributed to the study. We would also like to thank the Section of Biostatistics, Statistical Advisory Service, University of Copenhagen for their assistance with the statistical analysis.