This study established that the COVID-19 restrictions study period was associated with differences in Labrador Retrievers’ lifestyle, routine care, insurance status, illness incidence and veterinary attendance for those dogs who were living in England and enrolled in Dogslife. To our knowledge, this is the first time data from a cohort study has been used to investigate associations between COVID-19 restrictions and canine lifestyles, health and veterinary care.
This study estimates that Dogslife owners reported that they exercised their dogs for an average 6.89 (95% CI: 1.65 – 12.13) minutes longer per week after the COVID-19 restrictions study period. This result initially seems in contrast with surveys of dog owners internationally, where dog owners reported their dogs had fewer walks (39), in Belgrade, where the number of minutes dog owners spent walking their dogs decreased (36) and in the UK, where dogs were walked less frequently and closer to home, but the number of minutes they were exercised for remained the same (15). However, the exercise reported in Dogslife was transformed from a categorical to a continuous variable prior to analysis and averaged across several different types of exercise. Research from Dogs Trust reported that dog owners played with or trained their dogs more frequently in lockdown, but the number of minutes and frequency of walks decreased and more time was spent walking on a lead, although the effect of breed was not accounted for (40). Therefore, it could be that Dogslife Labrador Retrievers were indeed walked less or similar amounts, but owners compensated with other forms of exercise due to spending more time with their dogs at home.
While to date there has been no peer reviewed research into dogs’ diet during lockdown, there have been several reports in the media of dog owners titbit-feeding more frequently during lockdown. The pet food brand Natural Instinct reported that 44% of the dog owners they surveyed gave their dogs more treats during lockdown (60), over half of owners surveyed by the charity Guide Dogs reported giving their dogs more treats (61) and 34% of dog owners surveyed by insurance group More Than reported giving their pets more treats (62). These surveys are in contrast with results from this study, which found decreased odds (0.36, 95% CI: 0.24 – 0.52) of owners reporting that they fed their dogs titbits SLV during the COVID-19 restrictions study period. An explanation is that a large proportion of Dogslife dogs were working, sporting or guide dogs: 9.48% prior to and 8.62% during the COVID-19 restrictions study period (Table 1). Owners would not have been able to take their dogs to sporting activities or training classes during lockdown, which may have reduced the number of titbits given as rewards. In support of this theory, one survey of 1833 Italian dog owners reported that 26% gave their dogs treats as a reward during training and sports activities (63) and a qualitative analysis of another survey in the UK revealed that some owners felt that treats should only be given as part of training (64). However, as Dogslife does not collect information on the reasons behind titbit-feeding or how regularly both pets and working, sporting or guide dogs are taken to training classes, it is impossible to infer whether a reduction in these activities was likely to cause such a large drop in the odds of titbit-feeding by owners.
The odds (1.20, 95% CI: 1.02 – 1.40) of owners reporting they had wormed their dog SLV increased during the COVID-19 restrictions study period. Despite the fact that many veterinary practices had limited services, they were still able to dispense wormers for dog owners to collect (49). Furthermore, one survey reported that only 69.3% of dog owners in the UK sought advice from their veterinarian about worming (65), so a large proportion of dog owners may have not been affected by changes in veterinary services in terms of their dogs’ worming routines. The higher rates of worming could be explained by improved compliance of owners due to having more time at home with their dogs and being less likely to forget (66). The results of this study are in contrast with reports of increased cases of Angiostrongylus vasorum lungworm after COVID-19 restrictions began in the UK, which was considered to be a result of a reduction in routine worming treatments (67).
A UK-based survey reported that people who bought puppies during the pandemic were more likely to be first-time pet owners and were more likely to buy from breeders who did not perform health checks on the puppies before sale (7). Therefore, a rise in inexperienced dog owners who bought from disreputable breeders may have exasperated the problem of poor parasite control measures. New puppy owners may have been reluctant to take them to a veterinarian in the pandemic and may not have received a parasite prevention plan (67). The demographics of Dogslife dog owners are considerably different: they buy Kennel Club registered Labrador Retrievers and a proportion of Dogslife dogs have another purpose in addition to being pets. Furthermore, 90.2% of dogs in the COVID-19 restrictions study period were over 1 year of age (Table 1), so most owners had at least some experience of dog ownership and would have probably already been in the routine of parasite prevention.
It is not surprising that there were reduced odds (0.70, 95% CI: 0.58 – 0.83) of Dogslife owners reporting that they had vaccinated their dog SLV during the COVID-19 restrictions study period. There was some confusion about the guidelines from the British Veterinary Association (BVA) about whether veterinary practices should administer vaccinations during the first 2020 lockdown period and while most ceased to administer them, some continued to prevent their emergency services being overwhelmed (68). The results reported here mirror the reduction in canine vaccination consultations reported by SAVSNET, who reported that the percentage change in vaccination consultation frequency in dogs during the first social distancing phase of lockdown (23rd March – 12th April 2020) compared to the same time period in 2019 approached approximately -95% and remained at around -40% by the 4th of July 2020 (69). Perhaps what is more pertinent is the implications of these results. The WHO have recently warned of the risk of measles outbreaks worldwide due to children missing their routine vaccines during the COVID-19 pandemic (70). In canine medicine, rises in cases of vaccine preventable diseases such of parvovirus and leptospirosis after the 2020 lockdown reported by SAVSNET are small yet concerning. Although these diseases currently seem under control, it is not clear whether there will be long-term consequences in immunity for puppies who missed their vaccinations within the optimal time frames, which are recommended to ensure that they are capable of making a primary immune response (71).
The odds (0.36, 95% CI: 0.24 – 0.52) of owners reporting that their dogs were insured were decreased during the COVID-19 restrictions study period, which could be linked to the financial status of owners. The BVA suggested that some owners may deal with the financial pressure of the pandemic in the short-term by cancelling their pet insurance (43). Furthermore, the price comparison site GoCompare reported a 26% increase in the price of dog insurance in May 2020 in comparison with May 2019, which may have made the decision to cancel or avoid renewing dogs’ insurance easier for some pet owners (72).
The illness incidence of Dogslife dogs did not change during the COVID-19 restrictions study period. However, as this study only included data between the 23rd of March and the 4th of July that were recorded up to a maximum of the 23rd of July in each year, it is possible that the long-term health of dogs was not yet affected. On the other hand, the odds of owners reporting that their dog experience a coughing episode during the COVID-19 restrictions study period was greatly reduced to a fifth of previous years (0.20, 95% CI: 0.04 – 0.92). A common cause of coughing in dogs is canine infectious respiratory disease, also known as “Kennel cough”, which transmits between dogs. has multiple viral and bacterial pathogens responsible for its causation and is endemic in the population (73,74). It is probable that the COVID-19 restrictions led to a reduction in dog socialisation and therefore the transmission of pathogens and that a subsequent reduction in coughing was reported by owners.
The odds of owners reporting that they took their dog to a veterinarian with an illness episode during the COVID-19 restrictions study period were reduced in comparison with previous years (0.58, 95% CI: 0.45 – 0.76). When specific illnesses were examined, the reduction in odds was still apparent in owners reporting that they took their dog to a veterinarian with scratching (0.46, 95% CI: 0.23 – 0.96), limping and lameness (0.56, 95% CI: 0.33 – 0.97), eye problems (0.12, 95% CI: 0.02 – 0.94) and accident or injury (not adjusted for age) (0.17, 95% CI: 0.03 – 0.90). The RCVS reported that in April 2020, 97% of practices it surveyed had limited their services to emergencies and ‘urgent cases’, although how these were defined by different practices were likely to differ and a further 29% had closed either a branch or their main practice (51). Owners were probably less likely to take their dog to the veterinarian when they perceived their dog had a milder problem, such as many cases of scratching or eye problems. They may have also been able to pick up their dogs’ repeat prescriptions for chronic or repetitive illnesses, such as many cases of limping or lameness. Accidents or injuries indicate more immediate problems which may require urgent veterinary treatment. However, this study did not differentiate between mild or more serious types of illness or identify chronic illnesses, so it is impossible to comment whether this was the case.
Alternatively, owners may have had a telephone consultation with their veterinarian about these complaints. In Dogslife, owners are asked “Did you take [dog name] to your vet for the [illness type]?” rather than if the owner received any veterinary advice or treatment. A limitation of this study is that the measurement of how veterinary care was accessed relied on Dogslife owners’ perception of what ‘taking their dog to the vet’ entails. SAVSNET reported an increase in phone consultations up to an average of about 3% of total consultations in the first and second phases of lockdown (23rd March – 10th May 2020) (54). This initially seems like a small percentage and if this estimation is accurate for Dogslife dogs it does not explain the large reduction in veterinary visits reported by owners. However, SAVSNET acknowledge these figures are likely to be an underestimation due to differences in the workflow and data recording of veterinary practices (54). In support of this, the RCVS survey reported that 100% of 451 practices that answered the question used remote consulting for existing clients, whilst 45% used it for new clients (51). Furthermore, HealthforAnimals reported that in a survey of 3258 pet owners in the US, UK, France and Brazil, 27% delayed or avoided contacting their veterinarians and the percentage of owners whose veterinary practices offered digital or remote services rose from 20% to 47% (75). Similarly, a survey of veterinarians in California reported that practices providing telehealth services rose from 12% prior to the pandemic to 38% between March the 15th and June the 15th 2020 (76). It is probable that the results in this study can be explained by a combination of Dogslife owners contacting their veterinarian less often due to concerns related to pandemic, experiencing an increase in telemedicine at their veterinary practices and having difficulties accessing veterinary care.
This study had several limitations which have not yet been discussed. We recommend that readers do not rely solely on the p-values reported to infer statistical significance and interpret our findings, but consider the confidence intervals and other results (49). Questionnaire data such as Dogslife is limited by social desirability bias (the tendency of survey and interview respondents to give answers they feel will be socially acceptable rather than those which reflect the truth) and ‘recall decay’ (the decrease in participants’ ability to accurately recall events as time to reporting increases). Previous Dogslife studies have attempted to account for recall decay (77), but such methods introduce subjective cut-offs and limit the data available, which was not deemed appropriate in the current study. An arbitrary cut-off was used to allow for some delay in illness reporting by owners and it is unlikely to have fully captured the true number of illnesses reported that were experienced in the study period. However, this was used for both the COVID-19 restrictions study period and the data in the same date range in previous years, so it is unlikely that there was a bias in reporting that affected the comparison of these time periods.
Furthermore, attrition (the loss of participants during the course of the study) may have affected the results of this study. It is typical for some Dogslife owners to not report for a long period between one questionnaire and the next one. This makes it difficult to determine the true attrition rates for Dogslife and all owners were considered as ‘available to report’, when this was unlikely. The illnesses included in this study were reported by owners and had not received veterinary diagnoses, so they probably had a wide variety of aetiologies. Dogslife is a study of Labrador Retrievers and the results reported here may not be generalisable to other dog breeds. Additionally, as has partly been discussed previously, the demography, dog owning experience and behaviour of Dogslife owners may differ from the general population of UK dog owners, especially during the COVID-19 restrictions study period with the increase of new dog owners. Finally, the cleaning, categorising and coding of the data included is subjective and relies on the researcher’s expertise and opinion.