ORIGINAL RESEARCH published: 14 April 2021 doi: 10.3389/fpsyt.2021.622060 The Value of Companion Dogs as a Source of Social Support for Their Owners: Findings From a Pre-pandemic Representative Sample and a Convenience Sample Obtained During the COVID-19 Lockdown in Spain Jonathan Bowen 1,2 , Antonio Bulbena 1,3 and Jaume Fatjó 1,3* 1 Affinity Foundation Chair for Animals and Health, Department of Psychiatry and Forensic Medicine, School of Medicine, Autonomous University of Barcelona, Barcelona, Spain, 2 Queen Mother Hospital for Small Animals, Royal Veterinary College, North Mymms, United Kingdom, 3 Institut Hospital del Mar d’Investigacions Mèdiques - Institut de Neuropsiquiatria i Edited by: Addiccions, Mar Health Park, Barcelona, Spain Virginia Torres-Lista, Catholic University Santa Maria La Antigua, Panama Dogs are a source of companionship and comfort for their owners, but the degree to Reviewed by: which this might translate into real emotional and social support has not been quantified. Hamdi Chtourou, Emotional and social support are essential to help people to get through personal University of Sfax, Tunisia crises such as bereavement. In this study we characterize the social support owners Marine Grandgeorge, University of Rennes 1, France obtain from their dogs, provide evidence of how widespread this social support is *Correspondence: amongst dog-owners, and show how social support from dogs can increase during Jaume Fatjó a crisis (using the COVID-19 pandemic as an example). We collected data from a jaume.fatjo@uab.es representative population-based sample of Spanish dog-owners and found that most Specialty section: respondents said that their dogs helped them to get through tough times. They got This article was submitted to comfort from physical contact with their dogs, shared activities with them and treated Public Mental Health, a section of the journal them as confidants in a similar way to friends and family. These are all key aspects Frontiers in Psychiatry of social support, and dogs offer the advantage of being more available than human Received: 27 October 2020 sources of support. It would be expected that the support that dogs provide would Accepted: 11 March 2021 be increased during a time of personal crisis and when we looked at data collected Published: 14 April 2021 from a convenience sample of Spanish dog-owners during the COVID-19 confinement Citation: Bowen J, Bulbena A and Fatjó J that is what we found; during the confinement owners engaged in more shared (2021) The Value of Companion Dogs activities with their dogs, hugged them more often and turned to them more as a as a Source of Social Support for Their Owners: Findings From a source of companionship and comfort (p < 0.0001 in all cases). However, although Pre-pandemic Representative Sample owners did confide more in their dogs (p < 0.0001), the effect was not as great as and a Convenience Sample Obtained for other aspects of social support. We suspect that this is because people were During the COVID-19 Lockdown in Spain. Front. Psychiatry 12:622060. able to use telecommunications such as video conferencing to maintain their human doi: 10.3389/fpsyt.2021.622060 confidant relationships. Our findings indicate that dogs can substitute for humans Frontiers in Psychiatry | www.frontiersin.org 1 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs as sources of some kinds of social support when conventional sources are unavailable. Our conclusion is that where a dog is present in a household, it should be regarded as an important resource for social support. This should be considered when designing clinical interventions and when public health decisions are being made. Keywords: social support, crisis, dog, bereavement, COVID-19 INTRODUCTION human-dog relationships and interactions between people, particularly between the mothers and their children (14). A recent systematic review suggests that the COVID-19 Previous research has identified pets as one of the potential pandemic is having significant psychological effects on the sources of social support that could contribute to mental general populations of many countries including the USA, resilience (6, 15, 16). However, to the best of our knowledge China, Italy and Spain (1). In particular, the pandemic has no previous research has explored how companion dogs might had a substantial effect on both the prevalence of grief and its fit into the multi-faceted framework of social support, in the management. The process of grieving has been negatively affected general population. by reduced social interactions due to restrictions and lockdowns, In this study we sought to confirm our hypotheses that financial insecurity, fear of contagion, and the limitations on the the social support people get from their companion dogs is a holding of funerals and burials that have resulted from pandemic widespread phenomenon, that the types of support that people control measures (2). The pandemic is also having detrimental get from their dogs is comparable in character to that which is effects on positive lifestyle factors, such as physical activity and obtained from people, and that social support from dogs would healthy dietary habits (3). increase during a time of crisis. There is a very robust relationship between access to social support and positive indicators of physical and mental health METHODS (4), and social support has been one of the main protective factors against stress and anxiety during the pandemic (1). The Two samples of data were collected in Spain. The main focus stress-buffering hypothesis proposes that social support acts as of the study was a representative sample population that was a buffer to reduce the negative impact of stressful situations obtained in early 2019, a year before the COVID-19 pandemic. on health and well-being, as well as to compensate individual This data was used to establish the types and levels of support trait vulnerabilities, such as neuroticism and introversion (5, that dog owners would be expected to obtain from their pets. 6). However, during the pandemic there has been a profound A second, convenience sample, was collected during the initial disruption of social support networks and access to regular face- period of the COVID-19 pandemic lockdown in Spain. This to-face medical services, including counseling and psychological data was used to provide evidence for the effect that a personal support to the bereaved, which has exacerbated existing crisis might have on the degree to which people seek such social vulnerabilities (7). This is a situation in which pet dogs could support from pets. The data from the representative sample become a supplementary source of support, not only in the face population was collected anonymously, so it was not possible of highly stressful events, but also as a buffer to help people to to re-contact the respondents during the pandemic. For both cope with the background of every day low-level stressors. populations, a questionnaire based on the Monash Dog Owner Social support is an overarching construct that includes Relationship Scale (17) (MDORS) was used to evaluate the social several somewhat independent components. Two of the main support obtained by the owner from the pet dog. Data was features of any given source of social support are availability analyzed at the level of the individual item scores for the scales and closeness (4). Availability relates to how easily a source that were used, as the subscale scores for these scales were of social support can be accessed, and closeness includes originally designed to measure the human-animal bond, and interdependence, shared activities and emotional support (8). include items that are not relevant to social support. Another valuable role in a relationship is that of “confidant;” someone with whom to share personal thoughts. Self-disclosure Spanish Representative Sample Population is the main characteristic of a confidant relationship (9). Physical Five hundred and one responses from dog owners were collected contact is also important to social support, and has positive in Spain during March-April 2019 by Ipsos MORI for the effects on socio-emotional, physical and psychological well-being Affinity Foundation Chair for Animals and Health and the (10). Although not part of social support, for many people Affinity Foundation. Data were collected from an Ipsos Panel there is a psychological benefit to having opportunities to care representative sample of the population of Spanish dog owners. for others (11). The sample population included dog owners who were adult According to recent estimates, 24% of European households Spanish residents (above 18 years of age). The sampling method own at least one dog (12) and most people who live with dogs applied national representative quotas related to the socio- consider them to be like family members (13). Also, several demographic features of the owner and general characteristics studies have found neurophysiological similarities between of the dog. Data was obtained following a standard CAWI Frontiers in Psychiatry | www.frontiersin.org 2 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs (Computer Assisted Web Interviewing) procedure. Respondents One reason for changing the scoring system was that, as we will completed demographic questions about themselves and their see, scores for many items of MDORS were already maximal history of dog ownership, as well as the MDORS (17). for most respondents in the representative sample. A scale from A previously prepared, standardized, back-translated Spanish “much more” to “much less” than before the COVID confinement language version of MDORS questionnaire was used (18). In the not only restored the range for each item, but also made it clear conventional scoring of MDORS, the items within the perceived to respondents that they were being asked to compare their costs subscale are reverse scored. This means that a high score for current situation (during the lockdown confinement) with prior the perceived costs subscale equates to a low-level of perceived to the lockdown. costs. Although this is confusing, it means that high scores for all three subscales indicate a better human-animal bond, and a single Statistics value for quality of relationship can be calculated by combining Data was tabulated, mean scores and other descriptive statistics the scores from the subscales. This was not relevant to our study, were calculated in Microsoft Excel (version 16. 2020). After so to reduce confusion all items were scored in the same way finding that data was not normally distributed (D’Agostino- (none were reverse scored). For example, “Strongly agree” was Pearson test), Spearman rank correlation was used to investigate scored 5 for all items, regardless of which subscale they originated correlations between variables in the representative sample. A from. This makes it easier to interpret the results, within the Bonferroni correction was applied to these correlations, so that context of the present study, particularly in combination with the the threshold for significance was 0.0026 (0.05 divided by 19). data from the COVID lockdown convenience sample. For the convenience sample COVID-19 population, a single- sample Wilcoxon Signed Rank Test was used to determine whether responses for each modified MDORS item were Spanish COVID Convenience Sample significantly different from the mid-point response “The same as Population before the confinement.” A Bonferroni correction was applied to Two weeks after the start of the COVID lockdown in Spain, those contrasts, so that the threshold for significance was 0.002 an online questionnaire about the impact of the lockdown (0.05 divided by 25). Statistical tests were performed using Prism on pets and household was distributed through social media version 8 (Graphpad Software, 2020). of the Affinity Foundation Chair for Animals and Health, the Affinity Foundation, AVEPA/VetBonds (Asociación de RESULTS Veterinarios Españoles Especialistas en Pequeños Animales), the Grupo de Especialidad de Etología Clínica de Avepa (GrETCA), Spanish Representative Sample Population the Fundación para el Asesoramiento y Acción en Defensa The mean age of the 501 respondents was 41.7 years of age, de los Animales (FAADA), Single Track Ltd, and veterinary with 47.9% being male and 52.1% being female. The mean clinics associated with AVEPA. The lockdown included the number of dogs per household was 1.3 (SD 0.6). A table of the following measures; social distancing, the closure of schools and individual item scores is presented in Table 1, with the items universities, banning of mass gatherings and public events, and sorted according to their location in the three MDORS sub-scales the suspension of all non-essential economic activities (19). Data (interaction, emotional closeness and perceived costs). collection ended after 2 weeks. The survey was hosted on the Ninety-four point six percent and 88.4% of respondents SurveyGizmo R platform (which is now known as Alchemer R ). played games with their dogs or had their dog with them while Seven hundred and ninety-four responses were collected from relaxing at least once a day or once every few days, respectively, dog owners (no data from cat owners was included). and 69.6% of respondents told their dogs things that didn’t tell Respondents completed a modified version of the Cat/Dog- anyone else once a day or at least once a week. Eighty-six-point Owner Relationship scale (C/DORS), that was developed from eight percent of respondents hugged their dogs at least once a day the MDORS by the authors, for the measurement of the human- or once every few days, and 80% agreed or strongly agreed with animal bond between owners and their cats or dogs (20). The the statement “my dog helps me get through tough times.” differences between MDORS and C/DORS mostly relate to items To understand the relationship between perceived costs from the owner-pet interaction sub-scale, because, for example, and the interaction and emotional closeness items, a series of cat owners don’t usually take their cats with them to visit friends correlations was calculated. An additional measure, “How hard and family. Three items were excluded from this part of the is it to look after your dog?,” was chosen as a factor, because in study, either because they were in direct contravention of the the convenience sample collected during the COVID lockdown, lockdown regulations (“How often do you take your dog to visit this item of perceived cost was the only one for which a large people?” and “How often do you take your dog in the car?”), percentage of the population reported an increase. The results or because they had the potential to cause substantial distress to are summarized in Table 2. In the table, significant correlations people during a time of crisis (“How traumatic do you think it are highlighted in bold. will be for you when your dog dies?”). Scoring of the items was The score for “How hard is it to look after your dog?” was also different from the standard implementation of C/DORS or negatively correlated with more of the interaction and emotional MDORS: All items were scored on a 5-point scale from “Much closeness items than the overall perceived costs score was. Within more than before the confinement,” which was scored +2, to the interaction sub-scale, physical contact (hugging and kissing) “Much less than before the confinement,” which was scored −2. and shared activities (play and relaxation time together), were Frontiers in Psychiatry | www.frontiersin.org 3 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs TABLE 1 | Individual item scores for MDORS items (representative sample). Scores (%) Mean score SD 5 4 3 2 1 % scoring 4 or 5 Meaning of score 4 or 5 Interaction How often do you play games with 4.7 0.7 79.8 14.8 3.0 1.4 1.0 94.6 At least once a day or once your dog? every few days How often do you have your dog with 4.5 1.0 77.4 11.0 4.8 1.8 5.0 88.4 At least once a day or once you while relaxing, ie watching TV? every few days How often do you hug your dog? 4.5 1.0 71.0 15.8 5.8 3.6 3.8 86.8 At least once a day or once every few days How often do you kiss your dog? 3.7 1.7 56.9 12.2 3.4 3.0 24.5 69.1 At least once a day or once every few days How often do you give your dog food 3.5 1.3 27.7 29.1 21.0 11.0 11.2 56.8 At least once a day or once treats? every few days How often do you take your dog to 3.5 1.5 37.9 18.8 18.1 6.0 19.2 56.7 Once a week or once a fortnight visit people? How often do you groom your dog? 3.4 1.1 15.6 39.1 21.5 16.6 7.2 54.7 At least once a day or once every few days How often do you take your dog in 2.9 1.2 9.0 26.9 24.6 27.5 12.0 35.9 At least once a day or once the car? every few days How often do you buy your dog 3.0 1.0 7.2 22.0 37.5 27.3 6.0 29.2 Once a week or once a fortnight presents? Emotional If everyone else left me my dog would 4.6 0.7 69.8 20.8 8.6 0.8 0.0 90.6 Strongly agree or agree closeness still be there for me. My dog provides me with constant 4.6 0.7 67.2 23.0 8.4 1.0 0.4 90.2 Strongly agree or agree companionship. How traumatic do you think it will be 4.4 0.8 58.7 28.7 10.4 1.2 1.0 87.4 Very traumatic or traumatic for you when your dog dies? My dog is there whenever I need to 4.5 0.8 60.9 26.1 11.0 1.4 0.6 87.0 Strongly agree or agree be comforted. I wish my dog and I never had to be 4.4 0.8 58.5 26.7 11.2 3.0 0.6 85.2 Strongly agree or agree apart. My dog helps me get through tough 4.3 0.9 50.7 29.3 16.2 2.8 1.0 80.0 Strongly agree or agree times. My dog gives me a reason to get up 4.2 0.9 44.3 33.5 16.8 4.2 1.2 77.8 Strongly agree or agree in the morning. My dog is constantly attentive to me. 4.0 1.0 33.7 39.9 17.8 7.0 1.6 73.6 Strongly agree or agree How often do you tell your dog things 3.7 1.6 50.3 19.3 6.8 1.6 22.0 69.6 Once a day or once a week you don’t tell anyone else? I would like to have my dog near me 3.9 1.0 35.5 32.3 24.4 6.8 1.0 67.8 Strongly agree or agree all the time. Perceived My dog costs too much money. 2.8 1.0 4.0 21.9 36.7 24.8 12.6 25.9 Strongly agree or agree costs There are major aspects of owning a 2.4 1.2 4.8 15.1 20.8 34.7 24.6 19.9 Strongly agree or agree dog I don’t like. It is annoying that I sometimes have to 2.3 1.1 3.6 13.8 24.6 29.7 28.3 17.4 Strongly agree or agree change my plans because of my dog. How often do you feel that looking 1.8 1.2 4.4 9.8 13.6 9.6 62.6 14.2 Once a day or once a week after your dog is a chore? My dog makes too much mess. 2.3 1.1 4.4 9.8 20.8 37.9 27.1 14.2 Strongly agree or agree How often does your dog stop you 2.0 1.2 3.0 9.4 19.5 16.0 52.1 12.4 Once a day or once a week doing things you want to? It bothers me that my dog stops me 2.1 1.1 4.0 8.0 22.2 29.7 36.1 12.0 Strongly agree or agree doing things I enjoyed doing before I owned it. How often do you feel that having a 1.6 1.1 3.6 5.4 8.8 12.0 70.2 9.0 Once a day or once a week dog is more trouble than it is worth? How hard is it to look after your dog? 2.2 0.9 1.0 4.8 28.7 40.5 25.0 5.8 Very difficult or difficult Within each sub-scale group, items are sorted in descending order of the percentage of respondents scoring 4 or 5. Frontiers in Psychiatry | www.frontiersin.org 4 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs TABLE 2 | Correlations between individual MDORS items and the scores for perceived costs and “How hard is it to look after your dog” (representative sample). Perceived costs score “How hard is it to look after your dog?” Mean SD Spearman r P (two-tailed) 95% confidence Spearman r P (two-tailed) 95% confidence score interval interval Interaction How often do you play games 4.7 0.7 −0.29 <0.0001 −0.37 to −0.2 −0.27 <0.0001 −0.35 to −0.18 with your dog? How often do you have your dog 4.5 1.0 −0.32 <0.0001 −0.4 to −0.23 −0.26 <0.0001 −0.34 to −0.17 with you while relaxing, ie watching TV? How often do you hug your dog? 4.5 1.0 −0.24 <0.0001 −0.32 to −0.15 −0.23 <0.0001 −0.31 to −0.14 How often do you kiss your dog? 3.7 1.7 −0.18 <0.0001 −0.27 to −0.1 −0.22 <0.0001 −0.31 to −0.13 How often do you give your dog 3.5 1.3 −0.10 0.0201 −0.19 to −0.01 −0.15 0.0005 −0.24 to −0.06 food treats? How often do you take your dog 3.5 1.5 −0.12 0.0057 −0.21 to −0.03 −0.16 0.0003 −0.25 to −0.07 to visit people? How often do you groom your 3.4 1.1 −0.06 0.1651 −0.15 to 0.03 −0.18 <0.0001 −0.27 to −0.09 dog? How often do you take your dog 2.9 1.2 0.01 0.8161 −0.08 to 0.1 −0.12 0.0079 −0.21 to −0.03 in the car? How often do you buy your dog 3.0 1.0 0.00 0.9925 −0.09 to 0.09 −0.06 0.1499 −0.15 to 0.03 presents? Emotional If everyone else left me my dog 4.6 0.7 −0.28 <0.0001 −0.36 to −0.19 −0.23 <0.0001 −0.31 to −0.14 closeness would still be there for me. My dog provides me with 4.6 0.7 −0.33 <0.0001 −0.41 to −0.25 −0.28 <0.0001 −0.36 to −0.19 constant companionship. How traumatic do you think it will 4.4 0.8 −0.30 <0.0001 −0.38 to −0.22 −0.26 <0.0001 −0.34 to −0.17 be for you when your dog dies? My dog is there whenever I need 4.5 0.8 −0.28 <0.0001 −0.37 to −0.2 −0.29 <0.0001 −0.37 to −0.21 to be comforted. I wish my dog and I never had to 4.4 0.8 −0.37 <0.0001 −0.44 to −0.28 −0.33 <0.0001 −0.41 to −0.25 be apart. My dog helps me get through 4.3 0.9 −0.29 <0.0001 −0.37 to −0.20 −0.29 <0.0001 −0.37 to −0.21 tough times. My dog gives me a reason to get 4.2 0.9 −0.26 <0.0001 −0.35 to −0.18 −0.25 <0.0001 −0.34 to −0.17 up in the morning. My dog is constantly attentive to 4.0 1.0 −0.11 0.0153 −0.20 to −0.02 −0.18 <0.0001 −0.26 to −0.09 me. How often do you tell your dog 3.7 1.6 −0.08 0.0606 −0.17 to 0.01 −0.14 0.002 −0.23 to −0.05 things you don’t tell anyone else? I would like to have my dog near 3.9 1.0 −0.29 <0.0001 −0.37 to −0.21 −0.27 <0.0001 −0.35 to −0.19 me all the time. For convenience of reference, the order of items is as in Table 1. The threshold for significance after Bonferroni correction is p < 0.0026. Significant correlations are highlighted in bold. both negatively correlated with both perceived costs (Spearman of dogs per household was 1.6 (SD 1.02). The mean duration r −0.24, −0.18, −0.29, and −0.32, respectively, p < 0.0001 in all of confinement reported by respondents was 3.2 weeks (SD cases) and “how hard is it to look after your dog?” (Spearman 1.19). Table 3 shows the mean score and standard deviation for r −0.23, −0.22, −0.27, and −0.26, respectively, p < 0.0001 in each C/DORS item, as well as the results for the single-sample all cases). Within the emotional closeness sub-scale, only “my Wilcoxon test. Apart from “How often do you buy your pet dog is constantly attentive to me,” and “how often do you tell presents?,” which was insignificant (p = 0.12), all results were your dog things you don’t tell anyone else?,” were not correlated significantly different from the mid-point response “The same with both perceived costs and “how hard is it to look after your as before the confinement” at the level of p < 0.0001. The dog?.” Although all of these correlations were significant, they percentage of respondents giving each response is also presented were modest in size. in Table 3, along with the overall percentages that answered more or much more than before, the same as before, or less or Spanish COVID Convenience Sample Data much less than before. The mean age of the 794 respondents was 40.7 years of age, with For the interaction and emotional closeness items, 12.7– 90.2% being female and 9.8% being male. The mean number 62.2% of respondents indicated a change of more or much more Frontiers in Psychiatry | www.frontiersin.org 5 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs TABLE 3 | Individual item scores for C/DORS items and the results of the single-sample Wilcoxon signed rank test (convenience sample, COVID). % % much % same % much Wilcoxon signed rank test more or as less or Mean SD 2 1 0 −1 −2 more before less Sum of ranks (W) p Interaction How often do you play games with 0.66 0.72 8.9 53.0 33.2 4.4 0.5 61.9 33.2 4.9 121,267 <0.0001 your dog? How often do you have your dog with 0.55 0.73 13.6 28.2 57.8 0.4 0.0 41.8 57.8 0.4 55,596 <0.0001 you while relaxing, i.e., watching TV? How often do you hug your dog? 0.57 0.72 11.1 37.4 49.6 1.5 0.4 48.5 49.6 1.9 74,350 <0.0001 How often do you kiss your dog? 0.35 0.81 10.3 23.6 59.2 5.0 1.9 33.9 59.2 6.9 35,250 <0.0001 How often do you give your dog food 0.30 0.64 2.4 31.3 61.0 4.2 1.1 33.7 61.0 5.3 33,529 <0.0001 treats? How often do you groom your dog? 0.21 0.59 2.4 21.8 70.5 4.5 0.8 24.2 70.5 5.3 17,271 <0.0001 How often do you buy your dog 0.03 0.58 2.1 10.6 76.7 9.5 1.1 12.7 76.7 10.6 2,100 0.12 presents? Emotional If everyone else left me, my dog 0.34 0.66 10.1 14.2 75.1 0.6 0.0 24.3 75.1 0.6 19,106 <0.0001 closeness would still be there for me. My dog provides me with constant 0.67 0.72 14.9 37.3 47.6 0.2 0.0 52.2 47.6 0.2 86,138 <0.0001 companionship. My dog is there whenever I need to 0.31 0.63 8.2 15.5 75.9 0.3 0.1 23.7 75.9 0.4 17,767 <0.0001 be comforted. I wish my dog and I never had to be 0.41 0.75 13.4 15.7 69.5 0.9 0.5 29.1 69.5 1.4 26,972 <0.0001 apart. My dog helps me get through tough 0.58 0.73 13.5 31.6 54.3 0.5 0.1 45.1 54.3 0.6 64,423 <0.0001 times. My dog gives me a reason to get up 0.20 0.56 4.8 12.7 80.4 1.6 0.5 17.5 80.4 2.1 9,667 <0.0001 in the morning. My dog is constantly attentive to me. 0.77 0.74 16.5 45.7 36.3 1.4 0.1 62.2 36.3 1.5 123,265 <0.0001 How often do you tell your dog things 0.26 0.54 4.6 16.8 78.2 0.3 0.1 21.4 78.2 0.4 14,470 <0.0001 you do not tell anyone else? I would like to have my dog near me 0.33 0.65 8.0 18.8 71.7 1.4 0.1 26.8 71.7 1.5 23,268 <0.0001 all the time. Perceived My dog costs too much money. −0.05 0.38 0.3 1.5 93.8 1.8 2.6 1.8 93.8 4.4 −749 <0.0001 costs There are major aspects of owning a −0.27 0.71 0.2 3.1 77.5 7.7 11.5 3.3 77.5 19.2 −13,403 <0.0001 dog I do not like. It is annoying that sometimes I have to −0.23 0.63 0.3 1.4 82.7 6.8 8.8 1.7 82.7 15.6 −8,321 <0.0001 change my plans because of my dog. How often do you feel that looking −0.21 0.62 0.0 2.1 83.8 5.4 8.7 2.1 83.8 14.1 −7,348 <0.0001 after your dog is a chore? I feel that my dog makes too much −0.08 0.62 0.9 6.4 82.9 3.5 6.3 7.3 82.9 9.8 −3,724 <0.0001 mess. How often does your dog stop you −0.19 0.59 0.4 2.4 82.0 8.7 6.5 2.8 82.0 15.2 −7,909 <0.0001 doing things you want to? It bothers me that my dog stops me −0.18 0.59 0.5 1.1 86.2 4.8 7.4 1.6 86.2 12.2 −5,041 <0.0001 doing things I enjoyed doing before I owned it. How often do you feel that having a −0.17 0.64 0.9 3.4 81.6 6.4 7.7 4.3 81.6 14.1 −7,023 <0.0001 dog is more trouble than it is worth? How hard is it to look after your dog 0.25 0.80 5.3 29.0 54.4 8.5 2.8 34.3 54.4 11.3 30,829 <0.0001 For convenience of reference, the order of items is as in Table 1. The threshold for significance after Bonferroni correction is p < 0.002. during the COVID-19 lockdown, with present-buying for the my plans because of my dog,” and “How often does your dog stop dog showing the lowest percentage of change and “My dog is you doing things you want to.” “How hard it is it to look after constantly attentive to me” the highest. Perceived costs items your dog?” was the only perceived costs item that increased in a tended to stay the same or be reduced, with the greatest reported large percentage of respondents (34.3%). reductions being for “there are major aspects of owning a dog Looking at items that are associated with support, 61.9% of that I don’t like,” “It is annoying that sometimes I have to change respondents played games with their dogs more or much more, Frontiers in Psychiatry | www.frontiersin.org 6 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs 41.8% had their dogs with them while relaxing more or much convenience sample, the data is exposed to various demographic more, and 48.5% hugged their dog more or much more. Fifty- and recruitment biases. two point two percent of respondents indicated that their dog provided them with constant companionship more or much Representative Sample Population more, and 45.1% said that the dog helped them through tough Although numerous studies have collected information about the times more or much more. human-animal bond using convenience samples, the results have not been generalizable because of biases in the sex, age and other demographic features of the studied populations (15, 16, 22). To DISCUSSION our knowledge, our study is the first to present findings from a representative sample, and as a result we are able to make The context of this paper was the global COVID-19 pandemic reasonable generalizations about how valuable dogs are in the and Spanish national confinement lockdown that occurred social support networks of their owners. from 14th March (2020) onward. Many families were MDORS includes a range of items grouped into sub-scales of isolated at home, people were unable to work, and their owner-dog interaction, emotional closeness and perceived costs. households experienced financial, emotional, health and lifestyle Unlike other measures of the human-animal bond, MDORS impacts. As the pandemic developed, many families also has been tested for both reliability and validity (23). It also experienced grief. has widespread use (23), including in studies of dog ownership The Monash Dog Owner Relationship Scale (MDORS) that we satisfaction (24) the performance of seizure detection dogs used with the representative sample population includes items (25), the effects of service dogs on the psychosocial health and that measure aspects of social support, as this is one of the well-being of individuals (26), and the effects of dog-owner benefits proposed as the basis for a functioning relationship in relationship on perceived stress and happiness (27). Although the social-exchange theory (17). In our study we abstracted those studies of associations between MDORS and biological measures social support items to evaluate them outside the subscales that in dogs and people are limited, there is some evidence of an MDORS uses to quantify the human-animal bond. Due to their association between MDORS parameters and plasma oxytocin anonymity, it was not possible for us to re-contact the members levels in people and dogs (28). of the representative sample panel during the COVID pandemic. Some of the MDORS items are more directly relevant to the Had we collected data from a convenience sample population and social support that the owner gets from the dog. These include compared that with our representative data, differences would direct statements about support, such as “My dog helps me get have been confounded by demographic biases. through tough times.” However, MDORS also includes items that So, when we collected data during the COVID-19 lockdown do not necessarily relate to support, such as “how often do you in Spain we changed the scoring of the scale to a relative measure take your dog in the car.” For this study we focussed on those for each item (a 5-point scale from “much more” to “much less” items which relate most closely to social support and caregiving. than before the lockdown). Also, we wanted to look at cat and Table 4 presents those MDORS items which are most relevant dog owners, so we used the C/DORS scale that is an adaptation to social support, grouped according the previously mentioned of MDORS for use with both species. C/DORS contains several characteristics, with a summary of the percentage of respondents additonal items on pet-owner interaction that are more specific who gave the most positive answers. to cats. However, for the present study the items included are the Two items in MDORS stand out as are direct statements same for the two populations, aside from those items which were of the degree of support that people get from their dogs; “My excluded in the Spanish lockdown study because they involved dog helps me get through tough times” and “My dog gives me activities that were potentially in breach of the law (such as a reason to get up in the morning.” Both of these statements traveling to other people’s homes). gained a high level of agreement, with 80% of respondents The findings from the lockdown study have already been agreeing or strongly agreeing with the former (50.7% strongly published (21), but in that paper we gave only a broad overview agreed with the statement). Lack of motivation is a common of the effect of the lockdown on the lifestyle, quality of life and characteristic of people who are in an anhedonic or depressive behavior of people and their pets. We found that emotional state, so the fact that 77.8% of respondents agreed or strongly and lifestyle impacts on the household were those most strongly agreed with the statement “My dog gives me a reason to get up associated with a perceived negative effect on personal quality of in the morning” is a powerful indicator of the importance of the life during the confinement, but we did not look in detail at the human-dog relationship. data from C/DORS and the support people got from their pets. However, agreement with direct statements of this kind may The discussion will prioritize the findings from the not reflect the actual support people get from their dogs. If it representative sample population obtained prior to the pandemic is real, the social support people get from their dogs should in detail and separately from the findings from the COVID-19 conform to what we know about the characteristics of social lockdown, because it offers a more generalizable insight into the support between people. social support people get from their companion dogs. Findings from the sample obtained during the COVID-19 lockdown Availability should be considered only as an illustration of how social Availability is one of the main predictors of perceived social support might be realized during a crisis, since, as it is from a support (4). It is worth noting that in Spain, 26% of families Frontiers in Psychiatry | www.frontiersin.org 7 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs TABLE 4 | Key items from MDORS that relate to social support, with the Almost three out of four dog owners in this study declared percentage of respondents who gave the positive responses for each item that their dogs are constantly attentive to them. This proactive (representative sample population). attitude perceived by dog owners is consistent with previous % scoring Meaning of score 4 research on human-dog interactions, which have found that 4 or 5 or 5 dogs are able to detect and interpret primary human emotional responses, including expressions of negative affective states Stated support (30, 31). For example, dogs proactively approach people more My dog helps me get through tough 80.0 Strongly agree or agree often if they are crying than showing other emotionally neutral times. vocalizations (32). This ability can be partially explained by My dog gives me a reason to get up 77.8 Strongly agree or agree in the morning. selection pressure favoring the survival of dogs that exhibited human-oriented behaviors during the process of domestication Availability (33). Recent evidence suggests that dogs are able not only to If everyone else left me my dog would 90.6 Strongly agree or agree still be there for me. express emotional contagion but also to show prosocial helping My dog provides me with constant 90.2 Strongly agree or agree behaviors, regardless of whether they are truly intentional or companionship. only perceived as such by their caretakers (31). In our study, the My dog is there whenever I need to 87.0 Strongly agree or agree majority of owners perceived their dogs to be readily available be comforted. and actively motivated to provide them with emotional support. My dog is constantly attentive to me. 73.6 Strongly agree or agree Shared activity Shared Activities How often do you play games with 94.6 At least once a day or Playing games with the dog and relaxing with the dog are two your dog? once every few days important shared activities. More than 90% of participants in the How often do you have your dog with 88.4 At least once a day or present study said that they play games with their dogs at least you while relaxing, i.e., watching TV? once every few days once a day or once every few days. Just under 90% of respondents Confidant stated that they had their dog with them while they were relaxing How often do you tell your dog things 69.6 Once a day or once a at least once a day or once every few days. you don’t tell anyone else? week Interactive play is considered one of the main benefits of Physical contact interacting with dogs, together with going on walks (34, 35), and How often do you hug your dog? 86.8 At least once a day or interactive play with dogs seems to be particularly beneficial in once every few days promoting prosocial behavior in children (35). The frequency How often do you kiss your dog? 69.1 At least once a day or once every few days of other shared activities included in the MDORS, such as Opportunity for care giving going with the owner to visit people and on car trips are more How often do you give your dog food 56.8 At least once a day or influenced by factors outside the relationship, including lifestyle treats? once every few days preferences, the dog’s dislike of travel or its inability to be left How often do you groom your dog? 54.7 At least once a day or alone at times. In our opinion play and time relaxing are the best once every few days overall indicators of the level of shared activities between people How often do you buy your dog 29.2 Once a week or once a and their dogs within MDORS. presents? fortnight Self-Disclosure and the Confidant Self-disclosure is the main characteristic of a confidant relationship (9). In this study, the majority of dog owners reported telling their dogs things they didn’t tell anyone else live with at least one dog (12), making dogs a readily availability at least once a day or once a week (69.6%). This fits with source of social support for many families. Four items from the findings from a study comparing self-disclosure to partners and MDORS refer to the availability of support, and all achieved a companion dogs, which found that for dog owners the dog played high level of agreement from respondents in the representative a similar confidant role as the person’s partner, with people sample, with 69.8% strongly agreeing with the statement “If showing greater willingness to talk to their dog about depression, everyone else left me, my dog would still be there for me,” jealousy, anxiety, calmness, apathy, and fear-related emotions, 67.2% strongly agreeing with “My dog provides me with constant compared with a human confidant (36). companionship” and 60.9% strongly agreeing with “My dog is Although it is easy to be dismissive of this confidant there whenever I need to be comforted.” Of these, the statement relationship between owner and dog because the dog is incapable about the dog still being there for the person even after everyone of understanding what is being disclosed, evidence from the else left could be seen as an indication of the emotional support use of text-based chatbots and artificial intelligence (AI) based a grieving person might get from their dog. This finding is in counseling systems suggest that people benefit from disclosure agreement with previous research that indicated that living with even when they know that the counselor is inanimate. The a dog seems to help people to get through the early stages of first evidence that computerized chatbots could be convincing bereavement, particularly if the person actively seeks support conversational partners that were able to elicit disclosure came from the dog (29). from the ELIZA studies at MIT (37); one of the response Frontiers in Psychiatry | www.frontiersin.org 8 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs scripts for this system was a convincing simulation of Rogerian Opportunities for Care Giving psychotherapy (38). More effective chatbots are being developed The majority of participants in our study reported that they to use AI to analyze user inputs and patterns of communication engaged in activities related to caring for the dog on a daily or in order to respond to human emotions (39), which is nearly daily basis. Helping others, including family members, something that dogs do naturally. The fact that so many friends and neighbors, can be as beneficial to health and well- owners regularly engage in personal disclosure to their dogs being as receiving support. Caregiving has been associated with should be considered an important indicator of the nature of psychological benefits for the caregiver, particularly when caring the relationship. behavior is not perceived as a burden (11). We also explored the association between perceived costs on aspect of interaction and emotional closeness. Most of the key indicators of social support in MDORS were negatively correlated Physical Contact with perceived costs or difficulty looking after the dog, as shown Touch and physical contact are important for socio-emotional, in Table 5. physical and psychological well-being (10, 40, 41). In a laboratory This suggests that increased perceived costs impair the owner’s study of the effects of physical contact on scores for loneliness, perception of, and ability to use, the dog as a source of support. touch contact was found to reduce the perception of loneliness, Interestingly, a time-consuming activity like grooming the dog especially among single people (42). In a study observing the naturally occurring touch contact between married partners whilst they were discussing personal stressors, disclosers who TABLE 5 | Correlations between key MDORS items associated with social received more touch contact perceived that they were better support and perceived costs and stated difficulty with looking after the dog. able to overcome their stressors (43). They also reported greater decreases in self-reported stress, greater increases in self-esteem, Correlation (Spearman r) and had a more positive view of their partners. Touch may Perceived “How difficult is even have effects on immune response that are of relevance costs it to look after to the current pandemic. In a study of the effects of stress- score your dog” buffering social support and hugging on rate of infection after exposure to the common cold virus, Cohen et al. found that Stated support perceived support protected against the rise in infection risk My dog helps me get through tough −0.29 −0.29 associated with increasing frequency of social conflict, and that times. 32% of this effect was associated with hugging (44). In a study My dog gives me a reason to get up −0.26 −0.25 in the morning. of psychiatric assistance dogs (PAD) used to help people with Availability mental health disorders, patients were found to make use of If everyone else left me my dog would −0.28 −0.23 tactile and body contact with dogs to help reduce their anxiety still be there for me. (45). A psychiatric assistance dog is a specific type of service dog My dog provides me with constant −0.33 −0.28 that has been trained to assist its owner, and in that study PAD companionship. owners showed reduce rate of hospitalization and medication My dog is there whenever I need to −0.28 −0.29 use, and an increase in ability to attend appointments. be comforted. In the present study, 86.8% of respondents said that they My dog is constantly attentive to me. −0.18 hugged their dog at least once every day or every few days. Shared activity Likewise, 69.1% of respondents kissed their dog at least once How often do you play games with −0.29 −0.27 every day or every few days. For people living alone, these may your dog? be valuable sources of physical contact. There is also evidence How often do you have your dog with −0.32 −0.26 from a study of plasma levels of oxytocin in dogs and their you while relaxing, ie watching TV? owners during an interaction experiment, that the level of this Confidant hormone was associated with increased frequency of kissing the How often do you tell your dog things −0.14 dog (23). During the pandemic, social distancing, restrictions you don’t tell anyone else? on social interactions between people, and individual concerns Physical contact about disease transmission have meant that many people have How often do you hug your dog? −0.24 −0.23 been isolated from sources of physical contact. For many people, How often do you kiss your dog? −0.18 −0.22 this included reducing physical contact with resident friends Opportunity for care giving and family members who worked outside the household in How often do you give your dog food −0.15 occupations associated with increased disease risk (such as treats? healthcare). However, the social restrictions and isolation that How often do you groom your dog? −0.18 the whole population has experienced during the pandemic is How often do you buy your dog presents? just a taste of what many disadvantaged, elderly people, and people with disabilities or mental illness experience as part of Only correlations that were significant after Bonferroni correction are presented (non- their everyday lives. significant correlations are left blank). Frontiers in Psychiatry | www.frontiersin.org 9 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs showed a weak but significant negative correlation with perceived one of the key indicators, with “my dog is constantly attentive to cost, suggesting that it is not generally perceived as a burden by me,” “How often do you play games with your dog,” “How often most dog owners who participated in our study. do you hug your dog,” and “My dog provides me with constant companionship” being increased the most. However, “How often do you tell your dog things that you don’t tell anyone else” was CONVENIENCE SAMPLE POPULATION increased for only 21.4% of respondents, and this may reflect the (COVID-19 PANDEMIC) fact that the use of video-conferencing technology has enabled many people to maintain a degree of communication with their The COVID-19 pandemic provided an opportunity to gather human confidants. supporting evidence for the findings from the representative Perceived costs items tended to stay the same or be reduced, sample population. This was a time of isolation, stress and which is unsurprising given that the data was collected during the bereavement that affected entire nations. We would expect lockdown; a time when having a dog would be expected to have people to turn to their dogs for support, and for the relevant less of an impact on owners’ plans and activities. Although we key measures of MDORS to be increased. This is what we would expect that the lockdown would have imposed a significant found (see Table 6). The average time of confinement was financial burden on households, the level of agreement with the 3.2 weeks, which may be regarded as quite short. However, statement “My dog costs too much money” hardly changed. previous studies indicate that periods of quarantine and home confinement as short as 10 days have been associated with negative psychological consequences (46). LIMITATIONS OF THIS STUDY AND Increases were seen in each of the key areas of social support, FUTURE WORK but were greatest for shared activity, physical contact and availability. Interestingly, these would be the areas of human Using a representative sample population we were able to make social interaction that would seem to be the most affected by a general characterization of the social support people get from the pandemic confinement, which implies that the social support dogs, and find out how widespread it is. However, we were unable provided by the dog dynamically adapted to fill gaps in social to follow-up the same population during the pandemic, so we interaction and support created by the lockdown. had to adapt a version of the same scale to provide a relative Forty-five-point one percent of respondents said that their measure for each item, and we collected data from a convenience dog helped them through tough times more or much more than sample. These were compromises that we felt were worthwhile, before the confinement. There were increased ratings for every given the opportunity to collect data during such an unusual TABLE 6 | Percentage of respondents in the COVID convenience sample who rated key MDORS items as much more, more or the same as before the lockdown. % Much more % Same % More or same Mean % reporting or more as before as before an increase Stated support My dog helps me get through tough times. 45.1 54.3 99.4 31.3 My dog gives me a reason to get up in the morning. 17.5 80.4 97.9 Availability If everyone else left me my dog would still be there for me. 24.3 75.1 99.4 40.6 My dog provides me with constant companionship. 52.2 47.6 99.8 My dog is there whenever I need to be comforted. 23.7 75.9 99.6 My dog is constantly attentive to me. 62.2 36.3 98.5 Shared activity How often do you play games with your dog? 61.9 33.2 95.1 51.9 How often do you have your dog with you while relaxing, i.e., watching TV? 41.8 57.8 99.6 Confidant How often do you tell your dog things you don’t tell anyone else? 21.4 78.2 99.6 21.4 Physical contact How often do you hug your dog? 48.5 49.6 98.1 41.2 How often do you kiss your dog? 33.9 59.2 93.1 Opportunity for care giving How often do you give your dog food treats? 33.7 61.0 94.7 23.5 How often do you groom your dog? 24.2 70.5 94.7 How often do you buy your dog presents? 12.7 76.7 89.4 Mean percentage reporting an increase for each groups of items. Frontiers in Psychiatry | www.frontiersin.org 10 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs event, but they limit the value of the lockdown data. Future The authors of that paper went on to test this hypothesis studies should address these limitations. During the lockdown using primary healthcare providers (PHPs; family physicians, we also did not explore the relationship between individual nurses, and social workers) who were trained to include circumstances and the social support obtained. For example, how communication with patients about their pets as part of their the type and level of social support from the dog related to the service. As hypothesized, this was found to be a good way to quality of a person’s wider social support network, the stresses the open up communication that had positive effects on practice person experienced, their physical and mental health, and how and relationships with patients (48). Patients responded quickly technological solutions (such as video calling and social media) and openly to questions about their pets, and this enabled had mitigated the non-physical aspects of social isolation. These PHPs to learn more about aspects of lifestyle such as physical are areas that require further study. activity, as well as about patients’ family members, social capital and housing. PHPs were also able to use discussion about pets to leverage improvements in social capital, physical PRACTICAL IMPLICATIONS exercise, controlling unhealthy behaviors and the therapeutic benefits of pets. Our findings indicate that dog owners treat their dogs as a Talking with patients about their pets as a way to improve source of social support that offers a unique combination of communication, establish rapport, gather information and characteristics; emotional responsiveness, physical contact, being leverage support and behavioral change could be of increased non-judgmental and unquestioning in the face of personal importance in situations in which the patient has limited disclosure, and apparently unaffected by the underlying causes mobility or access to social capital, as we have seen during the of the person’s distress. In addition, what is qualitatively lacking pandemic confinement. in each of these characteristics, when compared with the support that might be provided by a person, may be made up for by the immediacy of availability. CONCLUSION Our data from the COVID-19 pandemic show that dogs can The findings from the representative sample population indicate be a valuable source of social support during a time of crisis, and that the majority of dog owners feel that their dogs help them in particular a source of physical contact to people during a time through tough times, and that the support they get conforms of social isolation. to known characteristics of social support such as availability, This suggests that the social support dogs provide should be shared activity, physical contact and acting as a confidant. In factored into public health decisions about how to manage crises, the convenience sample pandemic population, shared activity, such as a pandemic, and interventions to help people who are physical contact and availability items were, in decreasing order, in distress. This could include interventions to take advantage of those which were most increased. The owner-dog relationship the presence of a dog in the household as part of psychological therefore seems to be highly adaptable to compensate for changes or counseling interventions or finding ways to alleviate perceived in other sources of social support. If we want to take make the costs and difficulties of ownership that we have shown can impair most of these benefits, we have two options; firstly to create perceived support from the dog in a number of important areas. interventions that take advantage of the presence of a dog in For example, if we want dog owners to gain the maximum the household, and secondly to identify and minimize perceived benefit from having a dog during a period of personal crisis, costs for the owner so that the dog can be most effective as a interventions could be targeted to make owning the dog easier. social support. On an individual basis, this could include helping with the costs of food and veterinary care or helping a person who is elderly or has a disability to exercise their dog. During a national DATA AVAILABILITY STATEMENT emergency, such as the COVID-19 lockdown, this could include making public announcements that the availability of dog food The raw data supporting the conclusions of this article will be and healthcare will not be affected, and that dogs will continue made available by the authors, without undue reservation. to be able to go outside (as we found in our previous study that these were specific concerns dog owners had). Hodgson et al. identified 4 ways in which pets could benefit ETHICS STATEMENT human health; as builders of social capital, as agents of harm The studies involving human participants were reviewed reduction, as motivators for healthy behavior change, and as and approved by Social Sciences Research Ethical Review potential participants in treatment plans (47). They suggested Board (SSRERB) at the Royal Veterinary College. The that pets could motivate regular exercise, encourage activities patients/participants provided their written informed consent to of daily living, catalyze social interactions and a sense of participate in this study. community, encourage harm reduction (e.g., quitting smoking), and augmenting plans for the treatment of chronic disease. The simplest way to activate these benefits during healthcare AUTHOR CONTRIBUTIONS interventions would be to ask people about their pets, as this is a non-challenging way to open a dialogue and discover details JB and JF designed the study, performed the analysis and drafted about a person’s social support network, and lifestyle. the paper. AB assisted with the design of the study, interpretation Frontiers in Psychiatry | www.frontiersin.org 11 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs of the results and final drafting of the paper. All authors ACKNOWLEDGMENTS contributed to the article and approved the submitted version. The authors would like to thank Elena García, Patricia Darder, FUNDING Juan Argüelles for their assistance in organizing and collecting data for the convenience sample used in this study. We would This study was funded as part of the research activities of the also like to thank Germán Sánchez for helping us to collect the Affinity Foundation Chair for Animals and Health. data for the representative sample. REFERENCES 19. BOE (Official Gazettes). Real Decreto 463/2020, Vol. 67. p. 25390–400. Available online at: https://www.boe.es/eli/es/rd/2020/03/14/463 (accessed 1. Xiong J, Lipsitz O, Nasri F, Lui LMW, Gill H, Phan L, et al. Impact of COVID- March 14, 2020). 19 pandemic on mental health in the general population: a T systematic 20. Howell TJ, Bowen J, Fatjó J, Calvo P, Holloway A, Bennett PC. Development review. J Affect Disord. (2020) 277:55–64. doi: 10.1016/j.jad.2020.08.001 of the cat-owner relationship scale (CORS). Behav Process. (2017) 141:305– 2. Wallace CL, Wladkowski SP, Allison Gibson A, White P. Grief during the 15. doi: 10.1016/j.beproc.2017.02.024 COVID-19 pandemic: considerations for palliative care providers. J Pain 21. Bowen J, Garcia E, Darder P, Argüelles J, Fatjó J. The effects of the Spanish Symptom Manage. (2020) 60:e70–6. doi: 10.1016/j.jpainsymman.2020.04.012 COVID-19 lockdown on people, their pets, and the human-animal bond. J 3. Ammar A, Brach M, Trabelsi K, Chtourou H, Boukhris O, Masmoudi L, Vet Behav. (2020) 40:75–91. doi: 10.1016/j.jveb.2020.05.013 et al. On Behalf of the ECLB-COVID19 Consortium. Effects of COVID- 22. Brooks HL, Rushton K, Lovell K, Bee P, Walker L, Grant L, et al. The power 19. Home confinement on eating behaviour and physical activity: results of support from companion animals for people living with mental health of the ECLB-COVID19 international online survey. Nutrients. (2020) problems: a systematic review and narrative synthesis of the evidence. BMC 12:1583. doi: 10.1159/000512852 Psychiatry. (2018) 18:31. doi: 10.1186/s12888-018-1613-2 4. Rush MM. Perceived social support: dimensions of social interaction among 23. Payne E, Bennett PC, McGreevy PD. Current perspectives on attachment sober female participants in alcoholics anonymous. J Am Psychiatr Nurses and bonding in the dog–human dyad. Psychol Res Behav Manag. (2015) Assoc. (2002) 8:114–19. doi: 10.1067/mpn.2002.126673 8:71–9. doi: 10.2147/PRBM.S74972 5. Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol 24. Herwijnen IRV, van der Borg JAM, Naguib M, Beerda B. Bull. (1985) 98:310–57. doi: 10.1037/0033-2909.98.2.310 Dog ownership satisfaction determinants in the owner-dog 6. Li F, Luo S, Mu W, Li Y, Zheng X, Xu B, et al. Effects of sources of social support relationship and the dog’s behaviour. PLoS ONE. (2018) and resilience on the mental health of different age groups during the COVID- 13:e0204592. doi: 10.1371/journal.pone.0204592 19 pandemic. BMC Psychiatry. (2021) 21:16. doi: 10.1186/s12888-020-03012-1 25. Martos Martinez-Caja A, De Herdt V, Boon P, Brandl U, Cock H, Parra J, 7. Ammar A, Chtourou H, Boukhris O, Trabelsi K, Masmoudi L, Brach et al. Seizure-alerting behavior in dogs owned by people experiencing seizures. M, et al. On Behalf of The Eclb-Covid Consortium. COVID-19 home Epilepsy Behav. (2019) 94:104–11. doi: 10.1016/j.yebeh.2019.02.001 confinement negatively impacts social participation and life satisfaction: 26. Rodriguez KE, Bibbo J, O’Haire ME. The effects of service a worldwide multicenter study. Int J Environ Res Public Health. (2020) dogs on psychosocial health and wellbeing for individuals with 17:6237. doi: 10.3390/ijerph17176237 physical disabilities or chronic conditions. Disabil Rehabil. (2020) 8. Berscheid E, Snyder M, Omoto AM. The relationship closeness inventory: 42:1350–8. doi: 10.1080/09638288.2018.1524520 assessing the closeness of interpersonal relationships. J Pers Soc Psychol. (1989) 27. Gonzalez-Ramirez MT, del Quezada-Berumen L, Vanegas-Farfano M, 57:792–807. doi: 10.1037/0022-3514.57.5.792 Landero-Hernandez R. The effects of dog-owner relationship on perceived 9. Isaacs J, Soglian F, Hoffman E. Confidant relations in Italy. Eur J Psychol. stress and happiness. Hum Anim Interact Bull. (2018) 6:44–57. (2015) 11:50–62. doi: 10.5964/ejop.v11i1.741 28. Handlin L, Nilsson A, Ejdebäck M, Hydbring-Sandberg E Uvnäs-Moberg 10. Field T. Touch for socioemotional and physical well-being: a review. Dev Rev. K. Associations between the psychological characteristics of the human–dog (2010) 30:367–83. doi: 10.1016/j.dr.2011.01.001 relationship and oxytocin and cortisol levels. Anthrozoös. (2012) 25:215– 11. Schulz R, Sherwood PR. Physical and mental health effects 28. doi: 10.2752/175303712X13316289505468 of family caregiving. Am J Nurs. (2008) 108 (9 Suppl.):23– 29. McNicholas J, Collis G. Animals as social supports: insights for understanding 7. doi: 10.1097/01.NAJ.0000336406.45248.4c animal-assisted therapy. In: Fine AH, editor. Handbook on Animal-Assisted 12. FEDIAF (European Pet Food Industry Federation). Facts and Figures 2019 Therapy: Theoretical Foundations and Guidelines for Practice. San Diego, CA: Report. (2019). Available online at: https://fediaf.org/images/FEDIAF_facts_ Academic Press (2006). p. 49–71. and_figs_2019_cor-35-48.pdf (accessed October 27, 2020). 30. Huber A, Barber AA, Faragó T, Müller CA, Huber L. Investigating emotional 13. Julius H, Beetz A, Kotrschal K, Turner D, Uvnäs-Moberg K. Attachment to contagion in dogs (Canis familiaris) to emotional sounds of humans and Pets: An Integrative View of Human-Animal Relationships With Implications conspecifics. Anim Cogn. (2017) 20:703–15. doi: 10.1007/s10071-017-1092-8 for Therapeutic Practice. Cambridge: Hogrefe Publishing (2013). 31. Sanford EM, Burt ER, Meyers-Manor JE. Timmy’s in the well: 14. Nagasawa M, Mitsui S, En S, Ohtani N, Ohta M, Sakuma Y, et al. Social empathy and prosocial helping in dogs. Learn Behav. (2018) evolution. Oxytocin-gaze positive loop and the coevolution of human-dog 46:374–86. doi: 10.3758/s13420-018-0332-3 bonds. Science. (2015) 348:333–36. doi: 10.1126/science.1261022 32. Custance D, Mayer J. Empathetic-like responses in domestic dogs (Canis 15. Kurdek LA. Pet dogs as attachment figures for adult owners. J Fam Psychol. familiaris) to distress in humans: an exploratory study. Anim Cogn. (2012) (2009) 23:439–46. doi: 10.1037/a0014979 15:851–59. doi: 10.1007/s10071-012-0510-1 16. Gee NR, Mueller MK. Systematic review of research on pet ownership 33. Hare B, Brown M, Williamson C, Tomasello M. The domestication of and animal interactions among older adults. Anthrozoös. (2019) 32:183– social cognition in dogs. Science. (2002) 298:1634–36. doi: 10.1126/science. 207. doi: 10.1080/08927936.2019.1569903 1072702 17. Dwyer F, Bennett PC, Coleman GJ. Development of the monash 34. Westgarth C, Christley RM, Christian HE. How might we increase physical dog owner relationship scale (MDORS). Anthrozoos. (2006) activity through dog walking?: a comprehensive review of dog walking 19:243–56. doi: 10.2752/089279306785415592 correlates. Int J Behav Nutr Phys. (2014) 11:83. doi: 10.1186/1479-5868-11-83 18. Calvo P, Bowen J, Bulbena A, Tobeña A, Fatjó J. Highly educated men establish 35. Wenden EJ, Lester L, Zubrick SR, Ng M, Christian HE. The relationship strong emotional links with their dogs: a study with monash dog owner between dog ownership, dog play, family dog walking, and pre-schooler relationship scale (MDORS) in committed Spanish dog owners. PLoS ONE. social-emotional development: findings from the PLAYCE observational (2016) 11:e0168748. doi: 10.1371/journal.pone.0168748 study. Pediatr Res. (2020) 1–7. doi: 10.1038/s41390-020-1007-2 Frontiers in Psychiatry | www.frontiersin.org 12 April 2021 | Volume 12 | Article 622060 Bowen et al. Social Support From Companion Dogs 36. Evans-Wilday AS, Hall SS, Hogue TE, Mills DS. Self-disclosure with dogs: upper respiratory infection and illness. Psychol Sci. (2014) 26:135– dog owners’ and non- dog owners’ willingness to disclose emotional topics. 47. doi: 10.1177/0956797614559284 Anthrozoos. (2018) 31:353–66. doi: 10.1080/08927936.2018.1455467 45. Lloyd J, Johnston L, Lewis J. Psychiatric assistance dog use for 37. Weizenbaum J. ELIZA-A computer program for the study of natural language people living with mental health disorder. Front Vet Sci. (2019) communication between man and machine. Comput Linguist Assoc Comput 6:166. doi: 10.3389/fvets.2019.00166 Linguist. (1966) 9:36–45. doi: 10.1145/365153.365168 46. Hawryluck L, Gold W, Robinson S, Pogorski S, Galea S, Styra R. SARS control 38. Bassett C. The computational therapeutic: exploring Weizenbaum’s and psychological effects of quarantine, Toronto, Canada. Emerg Infect Dis. ELIZA as a history of the present. AI Soc. (2019) 34:803– (2004) 10:1206–12. doi: 10.3201/eid1007.030703 12. doi: 10.1007/s00146-018-0825-9 47. Hodgson K, Barton L, Darling M, Antao V, Kim FA, Monavvari A. Pets’ impact 39. Bendig E, Erb B, Schulze-Thuesing L, Baumeister H. The next generation: on your patients’ health: leveraging benefits and mitigating risk. J Am Board chatbots in clinical psychology and psychotherapy to foster mental health – Fam Med. (2015) 28:526–34. doi: 10.3122/jabfm.2015.04.140254 a scoping review. Verhaltenstherapie. (2019) 1–13. doi: 10.1159/000501812 48. Hodgson K, Darling M, Freeman D, Monavvari A. Asking about pets 40. Gallace A, Spence C. The science of interpersonal touch: an overview. Neurosci enhances patient communication and care: a pilot study. Inquiry. (2017) Biobehav Rev. (2010) 34:246–59. doi: 10.1016/j.neubiorev.2008.10.004 54:46958017734030. doi: 10.1177/0046958017734030 41. Jakubiak BK, Feeney BC. Affectionate touch to promote relational, psychological, and physical well-being in adulthood: a theoretical model Conflict of Interest: The authors declare that the research was conducted in the and review of the research. Per Soc Psychol Rev. (2017) 21:228– absence of any commercial or financial relationships that could be construed as a 52. doi: 10.1177/1088868316650307 potential conflict of interest. 42. Tejada AH, Dunbar RIM, Montero M. Physical contact and loneliness: being touched reduces perceptions of loneliness. Adapt Human Behav Physiol. Copyright © 2021 Bowen, Bulbena and Fatjó. This is an open-access article (2020) 6:292–306. doi: 10.1007/s40750-020-00138-0 distributed under the terms of the Creative Commons Attribution License (CC BY). 43. Jakubiak BK, Feeney BC. Interpersonal touch as a resource to facilitate positive The use, distribution or reproduction in other forums is permitted, provided the personal and relational outcomes during stress discussions. J Soc Pers Relat. original author(s) and the copyright owner(s) are credited and that the original (2019) 36:2918–36. doi: 10.1177/0265407518804666 publication in this journal is cited, in accordance with accepted academic practice. 44. Cohen S, Janicki-Deverts D, Turner RB, Doyle WJ. Does hugging No use, distribution or reproduction is permitted which does not comply with these provide stress-buffering social support? A study of susceptibility to terms. Frontiers in Psychiatry | www.frontiersin.org 13 April 2021 | Volume 12 | Article 622060