Introduction

Dog-owner relationships are typically researched as dyads of one dog-owner pair [1–4]. This tacitly assumes that each dog has only one adult owner and that all relevant information on that dog’s relationship with, and impact on, their human household can be derived from this person. However, it is estimated that approximately 1.6 million “family” homes in the United Kingdom (UK) contain adults, children, and at least one dog [5–9]. More than one person per household, including children, might therefore consider themselves the dog’s owner [10–13]. Dogs may be perceived to be integral family members in these homes [14,15], with children expected, and willing, to adopt varying levels of responsibility for their care depending on their age and what tasks their adult caregivers think are appropriate [12,16]. However, research is extremely limited on how co-habiting family members, including children, interact with and negotiate responsibility for the same dog. This paper seeks to start to fill that knowledge gap.

Dogs can be acquired by households for many reasons. Recently, mental health as a driver for acquisition has been highlighted; the COVID-19 pandemic led to a global surge in mental health problems amongst adults [17] and children [18], and a concomitant surge in ‘Pandemic Puppy’ acquisition [19,20], particularly amongst families [5]. Packer et al. [5] identified that 38% of UK adult owners acquired their ‘Pandemic Puppy’ to enhance their own or their family’s mental health. Mental health problems are estimated to cost the UK economy at least £117.9 billion annually [21], and dog ownership is therefore a very appealing potential solution. However, evidence to support the link between dog ownership and a mental health benefit during the COVID-19 pandemic is inconsistent for both children [22] and adults [11,23–32]. Furthermore, reviews of the evidence underpinning assertions about the benefits of pet ownership have identified only equivocal impacts on quality of life [33], happiness [34], loneliness [35], wellbeing [36], and in supporting people with formally diagnosed mental health conditions [37]. A variety of explanations for these apparently null findings have been proposed. These include that the studies did not account for other influences on wellbeing intrinsic to the owner [3,11] and that hedonic adaptation (the return of happiness to its previous baseline after a positive or negative event) may nullify initial benefits of pet acquisition over time [34]. Additionally, much of the existing research relies on quantitative measures which may not fully capture the complex nuances of pet-owner relationships [3,37], particularly regarding children [10,12,22,38,39].

Published studies have yet to explore the effects from ‘Pandemic Puppy’ acquisition versus existing dog ownership on child and adult mental health, or person-dog relationships, within the same household. Furthermore, outcomes from ‘Pandemic Puppy’ acquisitions explicitly aimed at providing companionship or improving mental health for children and their parents during the COVID-19 pandemic are not known. The aim of this study was to characterise the nature, benefits and challenges of dog-child interactions in households of UK adult caregivers and co-habiting children aged 8–17 years as reported by adults and children. We also sought to explore differences in mental health impact and human-dog bond during and after the COVID-19 pandemic.

Methods

Results

Demographic data

Four hundred and sixty-four responses to the survey were received (n = 376 from Respondent Group A and n = 88 responses from Respondent Group B). After data cleaning, n = 382 caregivers (82.3%; 2019 puppies group, n = 58; 'Pandemic Puppies' group, n = 324) remained who had provided a valid response about their households to Caregiver Questionnaire 1. Respondents from n = 299 (78.3%) of those 382 households (2019 puppies group, n = 42 households; 'Pandemic Puppies' group, n = 257 households) provided valid responses to Caregiver Questionnaire 2, describing n = 362 children (2019 puppies group, n = 51; 'Pandemic Puppies' group, n = 311). Child Questionnaire responses were provided by n = 216 (60%) of those children (2019 puppies group, n = 34; 'Pandemic Puppies' group, n = 182), representing n = 185 separate households (2019 puppies group, n = 28 households; 'Pandemic Puppies' group, n = 158 households). Fig 1 details the responses to each of the individual questionnaires comprising the survey and the manual cleaning process. Given the low response rate in the 2019 puppies group, any statistical comparisons were underpowered, and the following quantitative results therefore refer to both groups as a whole.

Download:

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

Fig 1. Schematic representation of the individual questionnaire responses and manual cleaning process prior to full analysis of the survey data.

Households did not have to complete all three linked questionnaires within the overall survey to be included in analyses.

https://doi.org/10.1371/journal.pone.0331179.g001

Child-dog interactions allowed by caregivers

All n = 301 caregivers who responded to the question about permitted child-dog interactions (Table 1) allowed their child(ren) to stroke or pat their dog, with the majority allowing their child(ren) to demonstrate other means of physical affection (hugging 95.0%, n = 286; kissing 89.0%, n = 268). Over one-third of parents (34.2%, n = 103,) allowed their child(ren) to interact with their dog whilst their dog was eating, with a minority (6.6%, n = 20) allowing their child(ren) to pull on their dog’s ears or tail. Seventeen of the 25 listed interactions (highlighted * in Table 1) have previously been reported to be associated with an increased risk of dog bites [38,49]; over 50.0% of children were permitted to perform eight of these.

Download:

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

Table 1. Permitted child-dog interactions collated from caregivers in response to the multiple-choice question “Which of the following is your child/children allowed to do with your dog?” (n = 301).

https://doi.org/10.1371/journal.pone.0331179.t001

Caregiver expectations versus realities of puppy/dog ownership

Just over one-third (37.3%, n = 139/373) of caregiver respondents to the questions regarding expectations versus realities of puppy training/management perceived at least one aspect of puppy training/management to have been worse than expected soon after acquisition. Of these, 30.2%, (n = 42/139) perceived two or more aspects to have been worse than expected. Dealing with their puppy nipping/biting their child during play was the aspect of puppy training/management perceived to be worse than expected by the most caregivers (Table 2).

Download:

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

Table 2. Caregiver expectations versus realities of puppy training/management soon after acquisition in response to the multiple-choice question “Thinking back to when your dog was a puppy, how did the following aspects of their care meet the expectations you had prior to acquiring them?”.

https://doi.org/10.1371/journal.pone.0331179.t002

First-time dog owners were significantly more likely to have perceived dealing with their puppy nipping/biting their children as worse than expected compared to caregivers with previous ownership experience (first-time owner: 24.4% versus previous dog owner: 12.4%; X² = 10.91, p = 0.004). First-time owners also found managing interactions between their child and a puppy worse than expected compared to those with previous ownership experience (first-time owner: 14.6% versus previous dog owner: 4.5%; X² = 11.44, p = 0.003) (Table 3).

Download:

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

Table 3. Impact of previous dog ownership experience on caregiver expectations versus realities of puppy training/management soon after acquisition. Significant differences at the 5% level identified by post-hoc comparisons are highlighted by emboldened text.

https://doi.org/10.1371/journal.pone.0331179.t003

Caregivers’ expectations versus realities of their children’s involvement in dog care

When considering their child(ren)’s change in involvement with aspects of their dog’s care at the time of survey completion compared to when their puppy was < 6 months of age, 49.2% of caregivers (n = 161/327) reported at least one aspect where their child(ren) were less involved than anticipated (Table 4). The aspects of dog care with the greatest increase over time in caregivers who felt their children were doing less than expected by the time of survey completion were walking (+11.9% increase) and feeding the dog (+9.6% increase). However, caregivers typically felt their children were currently more involved than expected, or as involved in as expected, in all five aspects of their dogs’ care. Dog walking was the aspect of care where the most caregivers felt involvement was greater than expected, whilst caregivers were most likely to score their children as currently less involved than expected in taking responsibility for the dog and playing with the dog.

Download:

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

Table 4. Expectations versus reality for child involvement in dog care by caregivers whilst their dog was < 6 months old (n = 335) and at the time of the survey in 2023 (n = 327).

https://doi.org/10.1371/journal.pone.0331179.t004

Caregivers’ prior dog ownership experience was significantly associated with their expectations versus realities of child involvement in dog care (Table 5). First-time dog owners were significantly more likely to have rated their child’s current involvement in dog care as less than expected for all areas except feeding.

Download:

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

Table 5. Impact of previous dog ownership experience on caregiver expectations versus realities of child involvement in dog care at the time of the survey. Significant differences at the 5% level identified by post-hoc comparisons are highlighted by emboldened text.

https://doi.org/10.1371/journal.pone.0331179.t005

Discussion

This mixed-methods study provides novel insights into the challenges and benefits for families from sharing their home with a puppy, including through the lens of the COVID-19 ‘Pandemic Puppy’ boom. Caregivers of approximately one quarter of children reported that their child had received a diagnosis of, or support for, a mental health problem – roughly in-line with NHS data which suggests 20.3% of children and young people in the UK had a mental health problem in 2023 [51]. Our qualitative data suggests that parents and children derived joy from dog ownership with dogs described as an indispensable mental health aide by a few parents, and a best friend by many children. However, challenges related to maintaining safe dog-child interactions, mismatched expectations, responsibility for the dog’s care, and the mental load of caring for a puppy plus children were evident.

Most children included in this study described comfort and happiness from living with a dog, results which concur with previous research [10,11,20,52,53]; the importance of dogs as attentive, non-judgemental partners and confidantes for children have been discussed elsewhere [39,54,55]. The necessity, and ability, of children to have close physical contact with their dog was also apparent, reflecting the findings of Baatz et al. [38]. That study identified that 7–12-year-old children’s affection and love for their dogs led to cuddles and hugs, though typically during play rather than for emotional comfort as suggested by the current study. Previously suggested motivations for children to cuddle dogs include: children’s perception that dogs “need” to be shown love through hugs, kisses and cuddles, reflecting their own perceived needs [12,56]; that pleasant affective touch with a dog might mechanistically mirroring close physical contact with humans by decreasing cortisol levels and increasing oxytocin levels, [57]; and that children learning how to regulate their emotions may seek social buffering of their stress responses through physical contact with pets [58]. Close physical contact with dogs is therefore likely to be a highly physiologically rewarding, and desirable, interaction for children. This was perhaps heightened further during the COVID-19 pandemic when other social interactions were limited [20]. Caregivers in the current study typically appeared highly supportive of children interacting closely and lovingly with their dogs, perhaps appreciating the dog performing some comforting aspects of childcare on their behalf.

However, dogs can find it stressful to be around children, particularly when children are highly emotional, noisy or seek prolonged close physical contact [38,39,59–61]. Dogs may also experience emotional contagion, reflecting (particularly negative) emotions of those around them [62–64]. Being bitten is therefore a major risk for a child of close physical contact with a stressed or aversive dog [39], and it was concerning that so few of the caregivers in the current study appeared to recognise this. Worryingly, parents interviewed by Baatz et al. [38], and caregivers involved in the current study, recalled struggling to stop children having close physical contact with a dog they did recognise to be displaying signs of aversion or distress, including growling. Given that parents [65] and children [60,66,67] may find it challenging to identify behavioural signs of anxiety in their dog, the incidence of interactions between children and dogs displaying aversive behaviours in the current study may have been higher than described. Further studies are needed to better understand the interactions between upset children and their family dogs to determine whether this interaction increases the bite risk, and if so to investigate mitigating strategies. Oxley et al. [68] have established that veterinary students can be taught to recognise dog aggression behaviours and change their own behaviour appropriately using a video simulation; similar initiatives may be beneficial for children.

First-time dog owners were revealed to have significantly more unrealistic expectations about the ease of puppy training (particularly their puppy’s nipping and biting), how easy it would be to manage relationships between their puppy and children, and their children’s likely involvement in puppy care. They also comprised 85% of those who had considered relinquishing their dog. Media imagery of the human benefits of dog ownership may lead to unrealistic expectations that puppies and children ‘grow up together’ in families without challenge or effort [69]. Our results suggest that first-time owners may commonly overestimate the ease of integrating a dog into their family home and may require additional support.

In common with previous research [16,20], many caregivers hoped their children would take on some responsibility for their dog’s care, but one in five reported that their child had taken less responsibility than expected, sometimes leaving those caregivers frustrated and stressed. Fifield and Forsyth [16] identified that children in New Zealand whose parents had bought them a dog to teach them about responsibility cared for that dog significantly less often than children with dogs bought for other reasons, though they did not establish why. Children involved in our study suggested that in some instances, expectations were too great. Previous research [12] identified that pet care responsibilities taken by children may be constrained by what parents think is safe or acceptable. Charmaraman et al. [10] revealed that in children aged 10–17 years, increased responsibility was significantly associated with a higher likelihood of identifying as a dog’s owner rather than them being a friend or sibling. It might therefore be important for parents to discuss ownership and how responsibility will work with their children prior to acquiring a new puppy, taking into account the age of those children when doing so [10,12,70]. Regular reappraisal may be needed; it was evident in our qualitative data that children’s desire, and perhaps ability, to take responsibility could change substantially over time. Longitudinal studies following the changes in responsibility for a dog’s care within a family over time, and its concurrent impact on their mental health, would be a vital addition to this research field.

Over twenty percent of caregivers had expected their child to play more with their dog. Our qualitative data revealed that multiple children described irritation or frustration at a dog seeking unwanted play with them. Caregivers, on the other hand, all described dog-child play interactions as positive and enjoyable for their children. This suggests these unwanted play interactions may not be observed by caregivers, and/or that children had not discussed these frustrations. Muldoon et al. [12] found that some children considered playing with a pet as itself being an aspect of taking responsibility, and their further work [56] identified that children may consider playing with a pet dog as important to that dog’s welfare. Additionally, playing with a dog has been described as a source of positive affect [57], and play was not included in recent lists of problem [71] or negative [72] canine behaviours. However, there is some existing evidence that play may not always be wanted by children; “getting mad with their dog” was logged by children in daily dog-relationship diaries [70]. Children of secondary school age reportedly play with their dogs less than primary school age children [73] so older children may experience more frustration if play-seeking does not diminish accordingly. Our results suggest that a dog seeking play could be a cause of negative affect, particularly in children, if that play is not reciprocally desired, and as such should be added to the list of owner-perceived problem behaviours. Predictors, and consequences, of this should be further explored.

Other owner-perceived canine problem behaviours were recorded, both qualitatively and quantitatively. Challenges posed by puppies jumping up and nipping were particularly prominent. Fortunately, most puppies outgrew these behaviours, but a few caregivers reported permanently damaged relationships between puppies and children because of these behaviours during puppyhood. In addition, several dogs were reactive towards other people or dogs, or pulled excessively on the lead whilst on walks, reducing the mental health benefits of those walks for those families. Unfortunately, unwanted behaviours are linked to relinquishment particularly when those behaviours were not part of owners’ expectations prior to acquiring a dog [74,75]; almost half of the caregivers considering relinquishment in the current study were motivated to do so by their dog’s problem behaviours.

Owners of 96.7% of a cohort of n = 985 Pandemic Puppies at the age of 21-months reported one or more persistent owner-reported problem behaviours in those dogs [71]. Causal links were drawn with the unusually isolated start in life of these puppies during the COVID-19 pandemic lockdowns and a concomitant absence of in-person training classes. Given the overlap in populations between that study and the current study, it is likely that some of those dogs were included here. However, our qualitative data shed further light on the negative effects these behaviours had on their families, reflecting the findings of Barcelos et al. [2] and Merritt et al. [76] that a dog displaying aggressive, anxious or fearful behaviours can be correlated with poorer well-being outcomes for their owners. The importance of dog training classes to reduce the probability of problem behaviours has been identified [71] and should be highlighted to new owners whenever possible. Ideally, they should also include children (either physically, if appropriate, or provide accessible resources for caregivers and children) to increase their awareness of canine body language and safe interactions.

This study offers further proof of concept that inclusion of children in research about the impact of family dogs is possible, and valuable. Extensive research has explored preferences within individual dogs for different people and for other dogs (e.g., [4,77–79]), but only one study [38] has previously sought to understand dog-human relationships beyond the single dog-owner dyad. Our qualitative data found that multiple different person-dog relationships co-exist in the same household, and those relationships change over time as dogs, and children, aged. The impact for adults and children of living with a dog was affected in some instances by other family members’ relationships with that dog and how different individuals adopted dog care responsibility. This research challenges the validity of the standard single dog-owner dyad model of human-dog interaction research. We suggest that future human-dog interaction research be broadened to include everyone living in the same household, and even non co-habiting family members who interact with the dog regularly [56].

In-keeping with a previous report [80], 95% of the dogs’ primary adult caregivers in this study were women, clearly positioning dog care as women’s work. All our female adult respondents were also caring for children. Many women spanned the menopausal age [81], a hormonal change which can be associated with increased feelings of stress and anxiety in addition to disruptive physiological changes. It was evident from the qualitative data that some found the addition of a dog to the household incredibly challenging. In light of growing evidence for a negative impact of the maternal mental load on women’s mental health [82], further work exploring dog ownership through the lens of the maternal mental load, and the menopausal period, is urgently needed, as this may be another risk factor for deteriorating dog-owner relationships, and potentially relinquishment.

Our study design had strengths and limitations. Its major strength was to include the experiences of both caregivers and children within the same household, as only a very small number of previous studies have done so [10,38]. Whilst data in the current study were collected cross-sectionally, respondents were invited to reflect on dog-owner relationships during the COVID-19 lockdowns and at the time of response. In doing so, we were able to capture descriptions of several apparently important drivers of change in dog-child relationships over time.

Our most significant limitation was the number of respondents in the 2019 group which meant we were unable to meaningfully analyse the quantitative mental health and owner-dog bond data collected. We were also unable to find child mental health measures which had been validated for retrospective use, further limiting the reliability of these data, hence their inclusion only in Appendix 2. Families were asked to remember events, relationships and emotions from almost three years prior to the study date, so those data are likely to be subject to recall bias. Additionally, the age limits for children we applied to the study applied at the time of data collection, not during the pandemic; the median age of the respondents who provided free-text data at the time of survey completion was 12 years, meaning these children were nine or ten years old at the start of the COVID-19 pandemic. The impact of the COVID-19 pandemic lockdowns on autobiographical memory, both for events and emotions, is the subject of ongoing research in children and adults [83–87]. This limited evidence to date suggests that most caregivers and children, particularly older ones, would be able to recall events and feelings at those times with some degree of accuracy due to unique nature of national lockdowns. This may have made acquisition of a dog during this period particularly memorable.

We hope to have mitigated some of the recall bias by starting the adult survey with specific questions about household events during the COVID-19 pandemic to provide some anchoring, such as whether individuals were furloughed and whether children were homeschooled. Further bolstering confidence in the validity of our qualitative data, all adult free-text responses contained clear references to the impact of the puppy or dog during the COVID-19 pandemic. Furthermore, almost all responses from children were written in the past tense and contained reference to the dog as a puppy, and/or to events consistent with the COVID-19 pandemic such as the impact of having a puppy when schools were closed, when walks were restricted, or their puppy helping with health anxiety, boredom or mental health problems during the pandemic. This suggests their relationship during the COVID-19 pandemic was being recalled. Even if it were the case that some children described their current relationship with their dog, there is such a dearth of qualitative data capturing child-dog relationships that these responses would still provide valuable insights. A more accurate method of data collection would have been to use contemporaneous daily diaries (e.g., [70]), but the unprecedented nature of the pandemic precluded this degree of academic preparedness.

Children’s qualitative responses were potentially constrained by their language development and writing ability. To mitigate these constraints, adult caregivers were invited to help children with their written responses where needed. We did not capture the total number of children of any age in the household, which would be a useful addition to future surveys. Families may have had additional children who were too old or young to participate but whose relationship with the dog may still have impacted the household, and our data hints that dog-child relationships may be different in single child households. Several parents did include description of dogs’ relationships with all their children, not just those of eligible ages, and this added further useful context. A few families mentioned additional dogs; determining whether the impact of all dogs in a household have an equal impact on families’ mental health would be useful future research.

Our study included elements of selection bias towards dogs that were still owned; the experiences of owners who acquired dogs between 2019 and 2021 then subsequently rehomed or relinquished them may have been substantially different but could not be captured by this study design. As such, more serious, negative impacts that precipitate relinquishment are likely to have been missed. Our sampling strategy incorporated owners of dogs who had previously been involved in our 'Pandemic Puppies' research, along with others recruited online. This self-selecting population could have biased those included towards caregivers of children who were more likely to have mental health problems, those with particularly good or bad memories of the COVID-19 lockdown experiences, and towards those interested in our research. However, our qualitative data suggests we had a wide mix of lockdown experiences, and that the prevalence of children with mental health problems in our population is in line with that of the UK population.

Conclusions

The COVID-19 pandemic precipitated many families to acquire puppies with the aim of improving their mental health. This study did not find puppy acquisition to be of consistent mental health benefit so this should not be a primary driver of puppy acquisition. Through inclusion of multiple household members in this research, we demonstrated that the consequences to children and caregivers from introducing a puppy to a family home are wide-ranging and may change with time. Individual family members build distinct and unique relationships with the same dog; however, those dog-human relationships, and their impacts on mental health within the family, appear inter-dependent. Rather than exploring single dog-owner dyads, future research on the relationships between dogs and people should include multiple household members to capture this complexity.

Many children benefit positively from their dog’s companionship and affection, though the close physical interactions inherent in many of these dog-child relationships may increase the risk of dog bites. Children may be unprepared for the restrictions and responsibilities associated with dog ownership and may find their dog’s play-seeking behaviours problematic. Caregivers may not anticipate specific challenges associated with having a young dog and children, such as aspects of the dog’s behaviour, setting boundaries for how the children and dog interact, and how responsibility for their care might be negotiated. Specific resources focusing on these potential problems should be developed for families seeking to acquire a new dog, particularly those who are entering dog ownership for the first time. Caring for a dog appears to be primarily women’s work, and the substantial impact of dog ownership on maternal mental health identified here requires further research attention.

Supporting information

Acknowledgments

The authors would like to thank all the respondents to this study, and the following who helped us to recruit participants: the British Veterinary Association (BVA), Royal Society for the Prevention of Cruelty to Animals (RSPCA), The People’s Dispensary for Sick Animals (PDSA), The Blue Cross, Battersea Dogs and Cats Home, Agria Pet Insurance, Marc Abraham and the All-Parliamentary Dog Advisory Welfare Group (APDAWG), Pets at Home, Vets4Pets, Viovet, the Animal Behaviour and Training Council (ABTC), the International Partnership for Dogs (IPFD), the Brachycephalic Working Group (BWG), Cariad Hound, Pets4Homes, The Doodle Trust, and many other organisations and individuals. Additional thanks to Dr Peter O’Kane and Hetta, Alby and Jenny Cory-Wright for their assistance in piloting the questionnaires.