Abstract
While the symbiotic nature of human-dog relationships and perceived benefits to human health have attracted much scientific interest, the mechanisms through which human-dog interactions may confer health benefits to humans are still poorly understood. The aim of this systematic literature review was to synthesize evidence of physiological changes associated with human-dog interactions with relevance to human health. Electronic databases were systematically searched (PubMed, MEDLINE with full text, Scopus, PsycINFO, CINAHL Plus with Full Text, and Web of Science Core Collection) for relevant studies. Of the 13,072 studies identified, 129 met the inclusion criteria, with approximately half being randomized trials (Level 2) based on the Oxford Centre for Evidence Based Medicine level system. Measures employed to study human physiological changes associated with human-dog interactions most commonly involved cardiac parameters and hormones, with negligible research of brain activity. The main positive findings were increases in heart rate variability and oxytocin, and decreases in cortisol with human-dog interactions. These physiological indicators are consistent with activation of the parasympathetic nervous system (PNS) and oxytocinergic system (OTS), and down-regulation of hypothalamic-pituitary-adrenal (HPA) axis activity. These results provide evidence of specific pathways through which human-dog contact may confer health benefits, likely through relaxation, bonding, and stress reduction. However, these findings should be interpreted contextually due to limitations and methodological differences. Previous research using other biological variables was limited in quantity and quality, thus impeding firm conclusions on other possible mechanisms. Further research is needed in some psychophysiological domains, particularly electroencephalography, to better understand central nervous system (CNS) effects. The findings of this review have implications for human-dog interactions to positively affect several stress-sensitive physiological pathways and thus confer health benefits. This supports their incorporation in various clinical, non-clinical, and research settings to develop evidence-based interventions and practices for cost-effective and efficacious ways to improve human health.
Keywords
Systematic review; Psychophysiology; Human-animal interaction; Animal-assisted intervention; Dogs
Introduction
The enduring relationship between humans and dogs is unique. With a shared evolutionary history, the symbiotic relationships between humans and dogs have existed for thousands of years (Thalmann et al., 2013). Dogs have adapted to socializing with humans, showing sensitivity to humans' emotional states (Albuquerque et al., 2016), social gestures (Hare et al., 2002), and even forming complex attachment relationships similar to those between infants and caregivers (Payne et al., 2015). In modern times, dogs have been employed in various contexts due to their responsiveness to human direction, such farm work, security, and assisting humans with disabilities (Payne et al., 2015). Pet ownership has also become increasingly common across various countries and cultures, with approximately 30 % of households having one or more pet dogs in Europe, Australia, China, and Japan (Islam and Towell, 2013; Murray et al., 2015). There has also been a deliberate inclusion of dogs into human health and wellbeing interventions, known as Animal-assisted interventions (AAI), an umbrella term encompassing both Animal-Assisted Therapy (AAT) and Animal-Assisted Activity (AAA). AAAs have been defined as informal human-animal interactions and interventions conducted by human-animal teams that are goal oriented for motivational, educational, and recreational purposes, while AATs are goal oriented, structured, and individualized therapeutic interventions often delivered or directed by licensed healthcare professionals as part of a treatment process (Griffin et al., 2011; Jegatheesan et al., 2014; Kruger and Serpell, 2010).
Systematic literature reviews (SLRs) investigating the empirical evidence for the popular opinion that dogs are beneficial to human health have documented benefits to mental health, physical health, and quality of life (Hughes et al., 2020; Yakimicki et al., 2019). However, existing research does not provide conclusive evidence regarding the explanatory pathways through which such benefits are conferred. Although previous SLRs on AAI and pet ownership have investigated psychological and psychosocial benefits to humans (Jones et al., 2019; Kamioka et al., 2014; O'Haire et al., 2015), no systematic review exists for physiological variables. While Beetz et al. (2012b) reviewed the effects of human-animal interactions on psychosocial and physiological variables, this was a narrative review of studies which were not chosen systematically. Furthermore, a meta-analytic study by Ein et al. (2018) that investigated the therapeutic effect of human-animal interactions has limitations in generalizability across different types of human-animal interactions due to the focus on interactions with certified therapy animals that are unfamiliar to participants while excluding personal pets. Further synthesis of evidence is crucial to understand the potential benefits of human-dog interactions on the physical and psychological health of humans and the underlying mechanisms through which potential health benefits may be conferred, in seeking cost-effective and efficacious ways of improving health (Teo and Thomas, 2019; Wilkes, 2009).
Several mechanisms and theoretical approaches have been proposed to account for the direct and indirect benefits to psychological and physical health of humans from human-dog interactions. Potential indirect benefits to human health include the development of attachment relationships with pets (Kurdek, 2008), an increase in perceived social support (McConnell et al., 2011), and increased physical activity (Curl et al., 2017; Cutt et al., 2007). In contrast, to elucidate the direct effects that human-dog interactions may have on human health, some studies have measured objective physiological biomarkers of stress and wellbeing.
There are a number of reliable physiological approaches to measurement of stress and wellbeing, including non-invasive physiological indices of neuroendocrine and immune functioning (Clark et al., 2020). Cortisol concentrations are commonly used as indices of stress and wellbeing in humans (Hellhammer et al., 2009). In response to acute stressors, physiological changes are driven by the autonomic nervous system (ANS) and hypothalamic-pituitary-adrenal (HPA) axis to adapt to environmental stressors, cope with challenges, and maintain homeostasis. This activation of the HPA axis due to stress triggers an increase in the production of the stress hormone cortisol (Marcus et al., 2012). Concurrently, the dynamic relationship between the sympathetic and parasympathetic branches of the ANS modulates involuntary processes such as blood pressure (BP) and heart rate (HR) (Waxenbaum et al., 2021). Chronic stress is detrimental to human health due to a cascade of physiological changes to the neuroendocrine, immune, and cardiovascular system. This can result in long-term elevation of cortisol levels and negatively affect immune function and recovery by down-regulating inflammatory responses (Schneiderman et al., 2005; Steptoe et al., 2007). Furthermore, cortisol has been linked with high BP, a risk factor of cardiovascular disease (Iob and Steptoe, 2019). Therefore, cortisol may be a useful measure of health-related changes associated with human-dog interactions.
Additionally, oxytocin, the neuropeptide produced by the hypothalamus and released by the posterior pituitary, has been linked with social interaction, bonding, and wellbeing (Beetz et al., 2012a; Olff et al., 2013). Oxytocin has been found to be increased with warm touch (Holt-Lunstad et al., 2008). It has also been identified as key in the regulation of complex social behaviors such as attachment and social support-seeking behaviors (Cochran et al., 2013; Thomas and Larkin, 2018). Furthermore, via the regulation of neuroendocrine stress responses, oxytocin has been found to be positively related to quality of life (Tang et al., 2019). On the contrary, oxytocin has been associated with unpleasant experiences such anxiety and distress (Glenk et al., 2020; Taylor et al., 2006). These findings reveal the complex yet significant role of oxytocin in facilitating complex social behaviors and in modulating the outcome of the stress response (Winter and Jurek, 2018). It is therefore of interest to determine how human-dog interactions are related to oxytocin levels.
Salivary alpha-amylase (AA) has also been identified as a biomarker of psychosocial stress as it is validated, reliable, non-invasive, and unaffected by salivary flow rate, with increased levels reflecting an increased activation of the sympathetic nervous system (Nater and Rohleder, 2009; Rohleder et al., 2006). Additionally, stress-related immune impairment has been studied by analysing salivary Immunoglobulin A (IgA) which serves as an index of humoral immune response and regulation of ANS activity (Krause-Parello, 2012). Salivary IgA has been found to decrease in response to acute stressors which increases susceptibility to illnesses, and can be increased by interventions that reduce stress (Proctor and Carpenter, 2002).
Cardiac measures also provide information on the fluctuations in stress levels which are relevant to health. There is a possible relaxing effect of interaction with animals as seen in lowered cardiovascular activity, reflected in lowered levels of HR and BP (Campo and Uchino, 2013; Vormbrock and Grossberg, 1988). This lowered cardiovascular activity has been postulated to be a result of tactile interaction with animals (Vormbrock and Grossberg, 1988), the presence of perceived noncritical social support (Cohen and Wills, 1985), or classical conditioning of a relaxation response (Vormbrock and Grossberg, 1988). Heart rate variability (HRV), the variation in the time between single heart beats, is a non-invasive measure of cardiac ANS function and is often used to indicate interactions between the heart and brain (Shaffer et al., 2014). HRV metrics include time-domain and frequency-domain measurements which reflect the complex dynamic relationship between the parasympathetic and sympathetic nervous system (Shaffer and Ginsberg, 2017). Analyses of these metrics reveal activity of the ANS, with higher HRV reflecting activation of the parasympathetic nervous system (PNS) and an increased state of relaxation (Kim et al., 2018). Conversely, lower HRV has been linked to health conditions such as major depressive disorder and coronary heart disease (Liao et al., 2002; Stapelberg et al., 2012).
Research on the health and therapeutic implications of the human-animal interactions has grown exponentially in the last 20 years (Rodriguez et al., 2021). Numerous benefits to human health such as reduced stress, lowered HR and BP, and increased socio-emotional functioning have been documented (Arhant-Sudhir et al., 2011; Wells, 2019). However, findings have been inconsistent. Some studies report no differences in such measures of health and stress, while others have even found negative associations such as an increased risk of mental and physical health conditions amongst pet owners (Herzog, 2011; Parslow and Jorm, 2003). Such inconsistency in findings has been attributed to the large variability in methodological design and outcome measures (Rodriguez et al., 2021). Furthermore, many studies only collect self-report measures and lack objective physiological data to support findings (Clark et al., 2020). Given these limitations in the literature, it is important to conduct a focused review of those studies including physiological measures. Identification of the psychophysiological mechanisms that underlie potential health benefits of human-dog interaction may provide a framework which guides burgeoning interventions in this field (Virués-Ortega and Buela-Casal, 2006). This knowledge is needed to develop evidence-based practice in clinical and non-clinical settings to optimize the human-dog relationship for improved human health (O'Donnell and Hetrick, 2016; Odendaal, 2000).
The current systematic literature review aims to examine the evidence for the benefits to humans of human-dog interactions, identify plausible psychophysiological mechanisms involved, and provide directions for future research. It will address the following research question: What is the psychophysiological evidence for mechanisms through which health benefits may potentially be conferred to humans through human-dog interactions? Studies reviewed will include all types of human-dog interactions irrespective of the nature of the interaction, such as temporary contact, regular contact, or ownership to provide a comprehensive view of the psychophysiological changes involved. This review will focus on the physiological pathways involved in human-dog interactions as previous reviews have largely considered psychological and psychosocial benefits to humans (Jones et al., 2019; Kamioka et al., 2014; O'Haire et al., 2015).
Section snippets
Data sources and search strategy
A systematic search of the literature using six electronic databases - namely PubMed, MEDLINE with full text, Scopus, PsycINFO, CINAHL Plus with Full Text, and Web of Science Core Collection - was performed in June 2021. A protocol was created a priori based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and registered with PROSPERO (ID: CRD42021249898) (Page et al., 2021). No ethical approval was required. The search terms used are listed in Table...
Results of literature search
The literature search identified 13,072 publications, of which 4907 were removed as duplicates. Of the 8165 originally identified publications, 174 passed the initial stage of title and abstract screening. A detailed full-text review of these 174 articles was conducted and a further 45 articles were excluded for not meeting the inclusion criteria. The final sample included 129 articles which were published between 1983 and 2021. A summary of these studies is listed in Table 4. The Preferred...
Discussion
This SLR is the first to systematically review the changes in health-related indices in human-dog interactions as measured by physiological variables. Overall, the reviewed 129 studies showed evidence for significant cardiac and neuroendocrine changes in humans following human-dog interactions, particularly increased heart rate variability, decreased cortisol, and increased oxytocin concentrations in more than half of the studies that measured these variables. These findings provide evidence...