A Systematic Review of Pet Attachment and Health Outcomes in Older Adults

Abstract

Research suggests that older adults might obtain health benefits from pet ownership; however, results are mixed. Pet attachment is suggested as both a mechanism for the relationship and a reason for differences in the association of pet ownership with health outcomes. This systematic review examines evidence for the relationship between pet attachment and health outcomes among older adults. The Open Science Foundation-registered review began with 20,795 candidate articles. We limited our review to the 58 articles that consisted of original research, published in peer-reviewed journals between 1965 and June 2025, written in English, included older adults (age ≥ 50 years) or were limited to only older adults, and examined the relationship between pet attachment and health outcomes. The articles included analyses of psychological (n = 53), social (n = 27), or physical (n = 2) health outcomes. Pet attachment was assessed with 19 tools; most frequently the Lexington Attachment to Pets Scale (n = 21) and the Pet Attachment Questionnaire (n = 8). The studies were not consistently of high quality according to OCEBM criteria. Except for grief, which was consistently related to pet attachment, the findings do not support a clear relationship between pet attachment and health outcomes in older adults. Findings suggest that the relationship between pet attachment and health outcomes may be more pronounced in younger than in older adults.

1. Background

The number of older adults in the worldwide population is increasing rapidly and, despite losses as a result of the COVID-19 pandemic, is expected to double by 2050 [1,2]. Numerous academics have investigated the connection between human physical, mental, and social health and owning a companion animal. The literature has identified a variety of relationships, including good, neutral, and negative ones, leaving general conclusions unclear. There is an accumulation of evidence suggesting that pet ownership may impact mental and physical health. Pet ownership (PO) has been linked to a slower decline in both physical [3] and mental [4] functioning. Furthermore, five years ago, a systematic review [5] on PO and pet interactions among older adults examined 145 published research articles and reported a relationship between PO and physical health and exercise, including cardiovascular health, indicating that pet owners may be healthier than their non-pet owning counterparts. This paper also reported a relationship between PO and depression; pet owners are less likely to be depressed than non-owners. The findings of the studies with respect to the relationship between PO and their owners’ health are mixed. In several studies, PO is related to some outcomes but not others, and in most studies, the quality of the evidence is not strong.

The quality of the relationship between pet owners and their pets may determine whether POs benefit from their pets. The American Veterinary Medical Association [6] defined the human-companion animal bond as a dynamic, mutually beneficial relationship shaped by activities that are vital to both parties’ health and wellbeing. According to Ainsworth [7], attachment figures have four characteristics. Many researchers argue that pets fulfill those four roles because they are comforting, enjoyable, missed when absent, and sought in times of distress [8]. These characteristics create the ties between the caregivers and their attachment figures [9].

Attachment appears to be an important component of the relationship between people and their pets and has been associated with a number of positive outcomes for older adults (e.g., [10,11]). It is possible that attachment is a mechanism driving health-related outcomes associated with PO in older adults.

Researchers have also investigated a variety of variables that affect an older adult’s ability to form attachments to an animal and possibly benefit from that bond. These include life experiences involving animals, prior attitudes and attachments to pets, health, living conditions, time spent with a pet, and socioeconomic status [12]. Moreover, people who had a strong attachment to pets early in life might have a stronger attachment to pets later in life than those who have never had a profound connection with an animal. Alternatively, some older people who have never lived with a pet before may not see a reason to add one to their home at this stage in life, and they may consider having a pet as a burden [13].

The way attachment to pets is related to human outcomes is varied and complex. Some studies show intriguing results indicative of some interesting but potentially complicated positive outcomes. For example, one study [14] showed that participants with higher levels of pet attachment demonstrated better executive functioning and that women (but not men) who interacted with their dogs had higher levels of oxytocin. On the other hand, results from previous research indicate that elders who cannot adequately provide for their dog have higher levels of depression than those who can [15], suggesting that attachment to pets for infirm older adults who do not have help in maintaining their pet in their home may have negative effects on mental health. Further, like the loss of a family member, the death of a companion animal causes suffering; yet attempts to embrace this suffering cognitively broaden one’s viewpoint. While the experience of separation sorrow from losing a pet increases with the strength of attachment to the pet, post-traumatic growth was enhanced by the application of a positive cognitive emotion control technique [16]. In other words, experiencing and coping with pet loss may offer an opportunity for growth and resilience.

Hill [17] suggested that the degree of the human–animal bond and the perception of human social support are correlated in a curvilinear fashion. Thus, bonds that are incredibly strong or weak may be connected to a lack of ability to develop resilience and overcome obstacles.

Given the conflicting outcomes associated with the relationship between pet attachment and the health of older adults, we concluded that a systematic review was warranted to gain a deeper understanding of the relationship between pet attachment and a variety of health outcomes, with the goal of providing a framework and suggestions for future investigations. The purpose of this manuscript is to review, evaluate, and discuss the known research on attachment to pets and its relationship to health outcomes among older adults.

2. Method of Systematic Review

This systematic review was registered with The Open Science Foundation at https://osf.io/juaw8 (accessed on 28 July 2025). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [18] were used to describe and document the search strategy (Figure 1). The inclusion criteria for article selection were as follows: (1) peer-reviewed journal article based on original research (data-based); (2) published in English; (3) published after 1965; and (4) participants must include people over the age of 50 years. The search included all articles available prior to June 2025. Search terms used and databases searched are included in

Table 1. Search terms used to identify articles for potential inclusion in the systematic review from each database searched.

Pet OR companion animal OR dog OR cat OR canine OR feline OR horse OR bird OR fish OR rabbit OR guinea pig OR alpaca OR equine OR hamster OR reptile OR hedgehog OR cricket OR insect OR reptile OR snake OR amphibian OR mammal

AND

Attachment OR bond OR emotional bond OR attitude OR Lexington Attachment to Pets OR MDORS OR CDORS OR Pet Attachment Scale

AND

Older adult OR elderly OR geriatric OR aging OR senior citizen OR senior OR older people OR old people OR aged

NOT

Pet scan OR CT scan OR Computerized Tomography OR CAT scan OR veterinary care OR positron emission tomography

and

Table 2. Databases searched for original research articles fitting the inclusion criteria in the systematic review.

Combined Database Search	Additional Separate Searches
Academic Search Complete	Anthrozoös
Alt Health Watch	Google Scholar
CINAHL Plus with Full Text	PubMed and Ovid MEDLINE
Google Scholar	Social Sciences Abstracts and Social Sciences Citation Index
Environment Complete
ERIC
Health Source: Waltham.com
Nursing/Academic Edition
LGBT Life with full text
PsychARTICLES APA
Psychology and Behavioral Sciences Collection,
PsycINFOAPA,
SocINDEX with Full Text

.

The PRISMA flowchart (Figure 1) documents the process of identification of articles relevant to the topic of the relationship between pet attachment and health outcomes of older adults. After the searches were completed, a list of all articles that were identified was created by combining the searches. The articles were screened by title first, then those that had not been eliminated were screened by abstract, and finally the remaining articles were screened by reviewing the entire articles. Two of the authors participated in blind screening at each stage. Discrepancies were identified and adjudicated at each stage.

Once articles were finalized, the quality of each article was assessed with a modified version of the Oxford Center on Evidence-Based Medicine (OCEBM) evidence level system [19]. The entire list of articles, including the authors, references, design type, OCEBM rating, sample size, outcomes assessed, tools used to assess pet attachment, and tools used to assess each health outcome, are included in a table about each outcome in which they are evaluated.

Review of the articles allowed us to separate them based on outcomes. We examined the health outcomes investigated in each article based on the three realms within the biopsychosocial model [20]: psychological, social, and biological categories. Psychological outcomes were defined as perceptions of self, social as perceptions of others, and physical as characteristics that are measured physically. One characteristic, stress biomarkers, is included as both a psychological and a physical outcome. It is important to note that the outcomes were not mutually exclusive in that any given paper could have included outcome measures that fit into more than one realm. The outcomes analyzed with respect to attachment were identified for each article, and the number of articles addressing each outcome was tabulated. Topics (e.g., depression) within each realm (e.g., psychological) that had five or more papers addressing a given outcome were discussed together. Topics that were discussed in fewer than five papers were presented in a general discussion.

Tables were created for each outcome as it was analyzed. Each table contains information about the articles that included the topic, including the design, group studied, area, age group, OCEBM categorization, and number of pet owners, as well as the tools used to assess pet attachment, tools used to assess the outcome, and the relationship between pet attachment and the outcome provided in the Results Section of the article. The area was defined as the places where the participants were located based on the information published in the article. If one or more locations were not specified, the location of the first author was included. Articles were identified as addressing older adults only or adults. Older adults only were defined as having all participants aged 50 years or older. Adults were defined as articles that include individuals aged 18 or above and have an upper age limit of 50 years or above. One article [21] includes individuals as young as 12 years old. In that case, the article is classified as “adult”, and the expanded definition is noted at the bottom of the relevant tables. A few articles contain analyses that were conducted separately for different age groups [22] or cover only very narrow age groups [23,24]. In those cases, the applicable age ranges are noted in the relevant tables.

3. Results

The initial search identified 20,795 articles (see Figure 1). An additional 55 articles were identified through external sources. Duplicates (n = 36) were removed prior to screening. The titles and abstracts were screened, and articles were removed that were not in English (n = 327), not related to older adults, published before 1965, not in academic journals, not full-text (n = 107,280), and not topically relevant by title (n = 9421). Additional articles were excluded if the abstract identified that they were not related to older adults and pet ownership (n = 132). Full texts of 209 articles were requested, and 208 of these articles were retrieved and reviewed. One article could not be retrieved; it was not available electronically and could not be obtained through interlibrary loan. Articles were excluded if they were not original research or not published in a peer-reviewed journal (n = 73), were not related to pet attachment (n = 15), or did not include pet attachment measures (n = 61) unless they were qualitative studies, which were retained. A total of 58 articles were identified that met the criteria of examining the relationship between pet attachment and health outcomes in older adults (aged 50 years or older) and are included in this review. A complete list of articles and the designs, ages of the participants (range, mean, and standard deviation), participant sub-populations, research quality rating, country (countries) of residence of the study participants, pet owner sample size, tool(s) used to assess pet attachment, and abbreviations for the tools used to assess the outcomes in each article are included in Supplemental Materials Sheet A: Articles included. The names and abbreviations for the tools used to assess psychological, social, and physical health outcome are included in Appendix A.

Pet attachment was measured using a variety of scales (

Table 3. Tools used to measure pet attachment and the number of articles in which each tool was used.

Abbreviation	Tool	Frequency
CAAS	Companion Animal Attachment Scale [43]	3
CABS	Companion Animal Bonding Scale [44]	1
CCA	Comfort from Companion Animals Scale [38]	3
CDORS	Cat/Dog Owner Relationship Scale [45]	1
HRS	Health and Retirement Study [46]	1
I-1	Single Item (1–10) [14,36,37]	3
INT	Interview	3
LAPS	Lexington Attachment to Pets Scale [39]	21
MDORS	Monash Dog Owner Relationship Scale [47]	1
MSPSS	Multidimensional Scale of Perceived Social Support—modified for pets [48]	1
OPR	Owner-Pet Relationship Questionnaire [49]	3
PAI	Pet Attitude Inventory [50]	1
PALS	Pet Attachment and Life Impact Scale [51]	3
PAQ	Pet Attachment Questionnaire [40]	8
PAS-A	Pet Attachment Scale (Albert) [42]	4
PAS-H	Pet Attachment Survey (Holcomb) [41]	4
PAS-M	Pet Attachment Scale (Meehan) [52]	2
PAS-T	Pet Attitude Scale (Templer) [53]	1
PRS	Pet Relationship Scale [54]	2

). Most articles used one scale to measure pet attachment; however, eight articles [25,26,27,28,29,30,31,32] used more than one scale, three articles were qualitative and used interviews [33,34,35], and three articles [14,36,37] used a one-item rating (1 to 10, with 10 representing the most attachment or strongest bond possible). Notably, several different scales were referred to as the Pet Attachment Scale or Pet Attachment Survey and abbreviated as PAS; thus, the individual scales are identified by adding the first initial of the first author’s last name. Some scales were also used with different names in different articles. For example, the Comfort from Companion Animals Scale [38] was used with different names but the same source. The most commonly used tool was the Lexington Attachment to Pets Scale (LAPS, 21 articles) [39], followed by the Pet Attachment Questionnaire (PAQ, 8 articles) [40] and Holcomb’s (PAS-H [41]) and Albert’s (PAS-A [42]) Pet Attachment Scales, four articles each. Additional tools were used less frequently (see Table 3). Seventeen of the scales were scored in such a way that higher pet attachment was indicated with a higher score. Two scales, the PAS-H and the PAQ, both of which scored different aspects of attachment style, were scored in such a way that a lower score indicated better attachment. The PAS-H assesses relationship maintenance and intimacy, and the PAQ assesses insecure, avoidant, and anxious attachment styles.

Articles addressed a large variety of health outcomes, and many of them examined several outcomes. A total of 53 articles addressed the relationship between pet attachment and psychological, 27 to social, and 27 to physical health outcomes (see

Table 4. Summary of articles containing an analysis of the relationship between pet attachment and each psychological, social, and physical outcome.

Outcomes	n	Social or Physical Outcome	n
Psychological Health Outcomes	53	Social Health Outcomes	27
Depression	26	Loneliness	12
Emotional/psychological health/wellbeing/QoL	20	Social support	7
		Social wellbeing/function/QoL	7
Anxiety	11	Social isolation	2
Distress/stress	11	Daily lives-social behaviors	1
Grief	7	Family membership	1
Positive emotions/affect	6	Positive relations with others	1
Cognition	5	Satisfaction with social support	1
Life satisfaction †	5	Seclusion	1
Anger ††	3	Social interaction	1
Executive function †††	3	Social networks	1
Purpose/meaning in life †	3
Happiness †	2	Physical Health Outcomes	27
Memory †††	2	Physical wellbeing, health, QoL	13
Negative affect ††	2	Physical activity	7
Autonomy †	1	Activities of daily living (ADLs) ***	3
Bad mood ††	1	Healthy lifestyle **	3
Caregiver burden ††	1	Physical performance ***	3
Coping with uncertainty †	1	Cardio-respiratory fitness ***	2
Emotional balance †	1	Dog walking **	2
Emotional exhaustion ††	1	Gait speed ***	2
Emotional instability ††	1	Illness Behavior **	2
Emotional ties †	1	Inflammation biomarker *	2
Environmental mastery †	1	Participation in activities **	2
Good mood †	1	Vitality/energy *	2
Guilt ††	1	Alcohol consumption **	1
Immoderation ††	1	Basal metabolic index (BMI)	1
Overwhelmed ††	1	Cannabidiol (CBD) consumption **	1
Personal growth †	1	Pain *	1
Psychopathology ††	1	Perceived activeness ** Physical limitations-disease related ***	1 1 1
Regular schedule †	1
Resilience †	1	Physical role disability ***	1
Self-acceptance †	1	Smoking **	1
Self-compassion †	1	Stress biomarkers *	1
Self-consciousness ††	1	Surgery (health care) decision **	1
Somatoform symptoms ††	1
Vulnerability ††	1

† = included in additional positive psychological outcomes. †† = included in additional negative psychological outcomes. ††† = included in cognition. * = included in health- or disease-related indicator, ** = included in health or health-related behavior, *** = included in physical function. QoL = quality of life.

). The analyses presented included 323 evaluations of the relationship between pet attachment and a psychological (109), social (160), or physical (54) outcome (see Supplementary Sheet B: Topic Count).

Only 35 articles (41%) included participants limited to older adults. The average age of participants in the articles limited to older adults from which averages could be obtained ranged from 63 to 75 years, and from the articles that included older adults, ranged from 24 to 60.5 years. Only older adult participants were included in a minority of the articles addressing outcomes in the psychological (38%; 20/53) and social realms (44%; 12/27), and in most of the articles (56%; 15/27) addressing outcomes in the physical realm.

Articles examined 1 to 12 different health outcomes. Most of the articles (49/58) examined the relationship between pet attachment and multiple outcomes (

Table 5. Number of topics addressed in relation to pet attachment per article (n = 58).

Number of Topics	Number of Articles	% of Articles
1	9	15.5%
2	14	29.3%
3	10	19.0%
4	8	13.8%
5	5	8.6%
6	1	1.7%
7	3	5.2%
8	2	3.4%
9	5	8.6%
12	1	1.7%

). Fourteen articles examined two outcomes, ten examined three outcomes, and thirty-four examined four or more outcomes.

Thirty-eight percent of the articles (22/58) examined health outcomes within only one realm, while most examined outcomes within multiple realms. Eighteen articles included only psychological health outcomes, one included only social health outcomes, and three included only physical health outcomes. Thirteen articles examined the relationship between pet attachment and health outcomes in all three realms: 10 in psychological and physical realms, 12 in psychological and social, and one in social and physical outcomes.

The topics are organized in the order of psychological, social, and biological or physical health outcomes. Outcomes that are addressed in four or fewer articles are evaluated in one or more groups within each realm of outcomes.

3.1. Topic 1: Psychological Outcomes

Articles that examined the relationship between pet attachment and psychological variables explored one to eight different psychological outcomes (

Table 6. Number of psychological health topics addressed in relation to pet attachment per article (n = 53).

Number of Topics	Number of Articles	%
1	23	43.4%
2	8	18.9%
3	9	15.19%
4	7	13.2%
5	1	1.9%
6	1	1.9%
7	2	3.8%
8	2	3.8%

). Most of the articles (53/58) included at least one psychological health outcome. A majority of the articles (30 of 58) included more than one psychological outcome. One psychological outcome was included in 23 articles, two were in 8 articles, three were in 9 articles, and four or more were in 7 articles.

The most frequent psychological health outcomes (Table 4) from the 53 articles that were examined in relation to pet attachment were depression (26 articles), emotional or psychological wellbeing or quality of life or mental health (18 articles), anxiety (11 articles), psychological stress or psychological distress (11 articles), and grief (7 articles). Positive emotions or affects were addressed in six articles, and cognition was addressed in five articles. Four or fewer articles addressed 29 additional outcomes, including life satisfaction (four articles), anger, executive function, and purpose or meaning in life (three articles each), and happiness, memory, and negative affect (two articles each). Another 20 psychological health outcome topics were addressed in one article each.

3.1.1. Depression

Pet attachment was examined in relation to depression in 26 articles (Table 7). A previous review of articles examining depression, pet ownership, and attachment to pets or attachment style [55] included 40 articles. Of those articles, a total of 20 were excluded from this review because they did not include older adults (n = 3), did not examine the relationship between pet attachment and depression (n = 8), examined human attachment styles rather than pet attachment (n = 2), were not peer reviewed, included unpublished dissertations (n = 6), or were not possible to retrieve (n = 1). Five additional articles were identified that examined the relationship between pet attachment and depression in samples that included older adults [37,56,57,58,59]. All 26 studies used quantitative research methods. Thirteen of the 26 studies were OCEBM category 5, seven were 3b, five were 4, and one was 2c (lower OCEBM category numbers indicate higher quality). As reported in the previous review, the most common tool used to assess pet attachment was the LAPS (14 articles); the next most common approach was a single item specific to the study (three articles) or the CAAS (three articles), a precursor of the LAPS. Two articles used the CCA, two used the CAAS, and four articles used other tools. Ten of the articles included older adults only, while the remaining 16 articles included, but were not limited to, older adults.

Likewise, a variety of tools were used to assess depression. The two most frequently used were the Center for Epidemiological Studies Depression Scale (CES-D; eight articles) and the Geriatric Depression Scale (GDS; five articles). Three different versions of the Patient Health Questionnaire (PHQ), two questions (one article), eight questions (one article), and nine questions (two articles) were used in a total of four articles. The Beck Depression Inventory (BDI) and the Depression, Anxiety, and Stress Scale (DASS) were used in two articles each, and five other scales were used once each.

While 23 of the studies used cross-sectional time frames, three included longitudinal components, either as surveys conducted at multiple times or using combinations of retrospective and prospective data to evaluate changes over time [36,60,61]. Three studies employed representative samples [22,62,63], while the others used convenience samples. All three longitudinal studies focused on changes in depression during the COVID-19 pandemic in the adult population, including older adults [36,60,61]. There were no commonalities in the surveys used to assess pet attachment or depression. In a large survey (n = 5061) that included only a single item to rate pet attachment, dog and cat owners reported higher attachment during COVID-19 than before it. Respondents reported that pets helped them cope with depression about COVID-19, and pet attachment was related to lower depression during the first three months of COVID-19 restrictions [36]. An article that focused on changes in mental health symptoms from before to during the pandemic (n = 1942) found that the relationship between pet attachment and changes in symptoms depended on the initial symptom level. For adults with moderate to high symptoms prior to the pandemic, pet attachment was related to higher odds of reductions in symptom severity. In contrast, among those with more severe mental health symptoms before the pandemic, pet attachment was related to maintaining severe symptoms over the same time period [60]. In an article that focused on how adults coped during the pandemic, pet attachment buffered the relationships between depression and the use of CBD and alcohol, but not the relationship between depression and cigarette smoking. Pet attachment also alleviated the indirect effects of job insecurity on depression that are due to increased stress [61].

Table 7. Articles that included the relationship between pet attachment and depression (n = 26).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Depression Measures
[64] Antonacopoulos, 2010	C	LA	CAN	A	5	66	LAPS	=, +S, no M social support	CES-D
[65] Branson, 2019	C	CO	USA	OA	5	41	LAPS	=	GDS
[14] Branson, 2016	C	H	USA	OA	3b	48	I-1	=	GDS
[66] Budge, 1998	C	PO	NZ	A	5	176	PAS-H	=	MHI
[67] Ellis 2024	C	PO	UK	A	5	1359	LAPS	+	PHQ-9
[37] Fernandes, 2024	C	PO	POR	OA	4	130	I-1	=	GDS
[68] Friedmann, 2020	C	PO	USA	OA	5	90	LAPS	=	CES-D
[69] Fritz, 1996	C	CG	USA	A	3b	124	LAPS	=	GDS-S
[62] Garrity, 1989	C, R	PO	USA	OA	5	208	CAAS	−	CES-D
[70] Ingram, 2019	C	CA, DO	USA	A	5	122	LAPS	=	CES-D
[36] Kogan, 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	−	1 item
[56] Koodaly, 2025	C	PO	USA	OA	4	178	LAPS	=, ME	CES-D
[71] Krause-Parello, 2012	C	W	USA	OA	5	159	PAS-A	−, ME	PGWB
[57] Kulari, 2025	C	E	POR	A	4	279	LAPS	+	DASS
[60] McDonald, 2021	L	PO	USA	A	4	1942	LAPS	+	BSI
[72] Miller, 1990	C	W	USA	OA	3b	53	PRS	=	GDS
[15] Miltiades, 2011	C	RU, DO	USA	OA	5	117	LAPS	+	CES-D
[27] Reevy, 2020	C	CO	USA	A	5	1239	LAPS, PAQ	+	IPIP
[63] Shoesmith, 2023	C, R	MI	UK	A	3b	81	CCA	=	PHQ-2
[22] Stallones, 1990	C, R	PO	USA	A †	5	152	CAAS	=	CES-D
[28] Teo, 2019	C	PO	AUT	A	3b	322	PAQ, OPR	−	DASS
[58] Testoni, 2017	C	BPO	ITA	A	4	159	LAPS	=	BDI
[61] Wan, 2023	L	E	USA	A	5	187	CCA	=, M stress to	PHQ-8
[73] Watson, 1993	C	EW	USA	A	3b	42	CAAS	=	BDI
[74] Wells, 2022	C	UKO	UK	A	5	146	LAPS	+	PHQ-9
[59] Zebrowska, 2024	C	PO	USA	OA	3b	140	LAPS subset	=	CES-D

† Ages 45–64. Design: C = cross-sectional; L = longitudinal; R = representative sample. Group: BPO = bereaved pet owners; CA = cancer patients; CG = caregivers of Alzheimer’s patients; CO = cat owners; DO = dog owners; E = employees; EW = employed women; H = homebound; LA = lives alone; M = men; MI = individuals with severe mental Illness; PO = community-residing pet owners or recent pet owners; RU = rural; UKO = UK residents; W = women. Area: AUT = Austria; CAN = Canada; ITA = Italy; NZ = New Zealand; POR = Portugal; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS = Companion Animal Attachment Scale [43]; CCA = Comfort from Companion Animals Scale [38]; I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAQ = Pet Attachment Questionnaire [40]; PAS-A = Pet Attachment Scale-Albert [42]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-M = Pet Attachment Scale-Meehan [52]; PAS-T = Pet Attitude Scale-Templer [53]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; M = moderator; ME = mediated. Depression Measures: BDI = Beck Depression Inventory [75]; BSI = Brief Symptom Inventory [76]; CES-D = Center for Epidemiological Studies Depression Scale [77]; DASS = Depression, Anxiety, and Stress Scales-21 [78,79]; GDS = Geriatric Depression Scale [80]; GDS-S = Short form of Geriatric Depression Scale [80,81]; IPIP = International Personality Item Pool [82]; MHI = Mental Health Inventory [83]; PGWB = Psychological General Wellbeing Schedule Depressed Mood Subscale [84]; PHQ-2 = Patient Health Questionnaire-2 [85]; PHQ-8 = Patient Health Questionnaire [86]; PHQ-9 = Patient Health Questionnaire-9 [87].

Table 7. Articles that included the relationship between pet attachment and depression (n = 26).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Depression Measures
[64] Antonacopoulos, 2010	C	LA	CAN	A	5	66	LAPS	=, +S, no M social support	CES-D
[65] Branson, 2019	C	CO	USA	OA	5	41	LAPS	=	GDS
[14] Branson, 2016	C	H	USA	OA	3b	48	I-1	=	GDS
[66] Budge, 1998	C	PO	NZ	A	5	176	PAS-H	=	MHI
[67] Ellis 2024	C	PO	UK	A	5	1359	LAPS	+	PHQ-9
[37] Fernandes, 2024	C	PO	POR	OA	4	130	I-1	=	GDS
[68] Friedmann, 2020	C	PO	USA	OA	5	90	LAPS	=	CES-D
[69] Fritz, 1996	C	CG	USA	A	3b	124	LAPS	=	GDS-S
[62] Garrity, 1989	C, R	PO	USA	OA	5	208	CAAS	−	CES-D
[70] Ingram, 2019	C	CA, DO	USA	A	5	122	LAPS	=	CES-D
[36] Kogan, 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	−	1 item
[56] Koodaly, 2025	C	PO	USA	OA	4	178	LAPS	=, ME	CES-D
[71] Krause-Parello, 2012	C	W	USA	OA	5	159	PAS-A	−, ME	PGWB
[57] Kulari, 2025	C	E	POR	A	4	279	LAPS	+	DASS
[60] McDonald, 2021	L	PO	USA	A	4	1942	LAPS	+	BSI
[72] Miller, 1990	C	W	USA	OA	3b	53	PRS	=	GDS
[15] Miltiades, 2011	C	RU, DO	USA	OA	5	117	LAPS	+	CES-D
[27] Reevy, 2020	C	CO	USA	A	5	1239	LAPS, PAQ	+	IPIP
[63] Shoesmith, 2023	C, R	MI	UK	A	3b	81	CCA	=	PHQ-2
[22] Stallones, 1990	C, R	PO	USA	A †	5	152	CAAS	=	CES-D
[28] Teo, 2019	C	PO	AUT	A	3b	322	PAQ, OPR	−	DASS
[58] Testoni, 2017	C	BPO	ITA	A	4	159	LAPS	=	BDI
[61] Wan, 2023	L	E	USA	A	5	187	CCA	=, M stress to	PHQ-8
[73] Watson, 1993	C	EW	USA	A	3b	42	CAAS	=	BDI
[74] Wells, 2022	C	UKO	UK	A	5	146	LAPS	+	PHQ-9
[59] Zebrowska, 2024	C	PO	USA	OA	3b	140	LAPS subset	=	CES-D

† Ages 45–64. Design: C = cross-sectional; L = longitudinal; R = representative sample. Group: BPO = bereaved pet owners; CA = cancer patients; CG = caregivers of Alzheimer’s patients; CO = cat owners; DO = dog owners; E = employees; EW = employed women; H = homebound; LA = lives alone; M = men; MI = individuals with severe mental Illness; PO = community-residing pet owners or recent pet owners; RU = rural; UKO = UK residents; W = women. Area: AUT = Austria; CAN = Canada; ITA = Italy; NZ = New Zealand; POR = Portugal; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS = Companion Animal Attachment Scale [43]; CCA = Comfort from Companion Animals Scale [38]; I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAQ = Pet Attachment Questionnaire [40]; PAS-A = Pet Attachment Scale-Albert [42]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-M = Pet Attachment Scale-Meehan [52]; PAS-T = Pet Attitude Scale-Templer [53]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; M = moderator; ME = mediated. Depression Measures: BDI = Beck Depression Inventory [75]; BSI = Brief Symptom Inventory [76]; CES-D = Center for Epidemiological Studies Depression Scale [77]; DASS = Depression, Anxiety, and Stress Scales-21 [78,79]; GDS = Geriatric Depression Scale [80]; GDS-S = Short form of Geriatric Depression Scale [80,81]; IPIP = International Personality Item Pool [82]; MHI = Mental Health Inventory [83]; PGWB = Psychological General Wellbeing Schedule Depressed Mood Subscale [84]; PHQ-2 = Patient Health Questionnaire-2 [85]; PHQ-8 = Patient Health Questionnaire [86]; PHQ-9 = Patient Health Questionnaire-9 [87].

The previous review indicated that there was either a positive or no relationship between pet attachment and depression, but that the relationship might differ in older adults. The current review includes 20 of the articles that were included in that review and addressed pet attachment and depression. Four of those articles were cross-sectional surveys of older adults only [14,15,65,72]. Two articles assessed pet attachment with the LAPS, one with a single item and one with the PRS; three assessed depression with the GDS, and one with the CES-D. Three of the articles found no relationship between pet attachment and depression, while in one, pet attachment was positively related to depression. The three articles that did not reveal a relationship included cat owners who attended senior centers in suburban and urban areas (n = 41, mean age 72.6 years) [65], homebound older adults living in a sparsely populated area who received meals delivered through a public service agency (n = 48, mean age 74.4 years) [14], and community-residing older adults (n = 53, mean age 73.0 years) in the USA [72]. The study that revealed a positive relationship was among rural adult women (n = 117, mean age 68.4 years), also in the USA [15]. Among those articles included in both reviews, 14 included older adults, but most included younger individuals. Eight of these articles reported no relationship between pet attachment and depression [22,61,63,64,66,69,70,73], four reported a negative relationship between pet attachment and depression [28,36,62,71], and two [60,67] reported a positive relationship between pet attachment and depression. None of the four reports of a negative relationship assessed pet attachment with the LAPS. All of the articles that reported a positive relationship between pet attachment and depression, in the articles that included, but were not limited to, older adults, used the LAPS to assess pet attachment. Examining the six articles that were limited to older adults, one article found a positive relationship [15], two found a negative relationship [62,71], and three found no relationship between pet attachment and depression [14,65,72]. All three of the articles that found no relationship between pet attachment and depression used the GDS to assess depression. One article focusing on older women found that pet attachment mediated the relationship between loneliness and depression; pet attachment helped cope with the loneliness [71]. The association between attachment and depressive symptoms can be related to other factors. In the article about recently diagnosed cancer patients (n = 122) who owned dogs, the association between attachment and depressive symptoms depended in part on treatment status [70]. Among those receiving treatment, there was a direct relationship between attachment and depression. Among those who had completed cancer treatment, there was an inverse relationship between depression and attachment.

The six articles that were not included in the prior review but were added in the current review included six cross-sectional studies, four with only older adults [37,56,59,68] and two that included, but were not limited to, older adults [57,58]. Five of these articles used the LAPS, and one used a single item to assess pet attachment. None of the articles that focused exclusively on older adults found a positive relationship between pet attachment and depression. In these articles, depression was assessed with the CES-D or GDS. One article [56] also examined whether pet attachment moderated the relationship between depression and physical health outcomes among 178 community-residing older adults who owned pets. They found that pet attachment did not moderate the relationship between depression and any physical health outcomes, including functional assessments and physical wellbeing. In the two articles that included younger and older adults, pet attachment was assessed with the LAPS, and depression was assessed with the DASS [57] or the BDI [58]. The studies were conducted in Portugal and Italy, respectively. Among Portuguese health care workers who owned pets (n = 279), higher pet attachment was associated with more symptoms of depression. In contrast, among 159 Italians who had experienced the death of a pet, pet attachment was not related to depression [58].

Summarizing the results of the articles evaluating the association of pet attachment with depression among older adults only, 9% of the articles report that higher attachment is associated with higher depression, 72% report no relationship, and 18% report that higher attachment is related to lower depression. In contrast, among articles that include older adults but are mainly younger participants, 31% report a positive relationship, 56% report no relationship, and 12% report a negative relationship (

Table 8. Summary of the direction of the relationship between pet attachment and depression in analyses in articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	5, 31%	1, 9%	6
=	9, 56%	8, 72%	17
−	2, 12%	2, 18%	4
Total	16	11	27

+ = positive relationship, = = no relationship, − = negative relationship.

). While the distributions of the results in the two groups of articles do not differ significantly [Chi-square (2 df) = 1.86, p = 0.39], the lower proportion of articles with a positive relationship between pet attachment and depression in older adults bears additional investigation. These findings do not support the conjecture that older adults procure or retain pets to treat their depression [88,89].

3.1.2. Psychological/Emotional Wellbeing/Health or Quality of Life

Twenty articles fell within this outcome. Pet attachment was examined in relation to only psychological or emotional wellbeing, an individual’s perception of their own mental or psychological health, in 17 articles that used scales that specifically assessed psychological wellbeing and one qualitative article (see

Table 9. Articles that relate pet attachment to emotional or psychological health, or wellbeing, or quality of life (n = 20).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Psychological Wellbeing or Mental Health Measures
[90] Allen, C. J., 2022	C	DO	AU	A	5	639	MDORS	+, ME	SPANE
† [66] Budge, R. C. 1998	C	PO	NZ	A	5	176	PAS-H	+ wellbeing = mental health	MHI
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA	N/A	11	INT	−	INT
[91] Demeter, N., 2024	C	STR	ISR	OA	5	25	PAQ	=	WHOQOL-BREF
[88] Friedmann, E.,	L	PO	USA	OA	2c	214	LAPS	=	SF-12 MCS
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	=	SF-12 MCS
[25] Hardie, S., 2023	C	PO	AU, USA, UK	A	5	238	CDORS, MSPSS	= M social support to	PWS
[70] Ingram, K.M., 2019	C	CA, DO	USA	A	5	122	LAPS	=	FACA
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	=	ISAI, SPHQ
[56] Koodaly, L., 2025	C	PO	USA	OA	4	178	LAPS	=, M Physical wellbeing to	SF-12 MCS
* [60] McDonald, S. E., 2021	C	PO	USA	A	4	1942	LAPS	M COVID stress to	BSI
* [26] Northrope, K., 2024	C	DO	USA and other countries	A	2c	607	LAPS, PAQ	−, + Avoidant attachment + Anxious attachment ME, M, anxious attachment to people to	BSI
[92] Raina, P., 1999	L	PO	CAN	OA	2c	286	LAPS	=, = changes	OHS
[24] Schwarzmueller-Erber, G., 2020	C	HR, DO	AU	A ††	5	124	PALS	+ HR/= DO	FAHW
[63] Shoesmith, E., 2023	C, R	MI	UK	A	3b	81	CCA	=	ONS
[29] Tan, J. S. Q., 2021	C	PO	SGP	A	3b	431	PAS-H, PAQ	+	RAND
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	+ + Secure pet attachment	WHOQOL-BREF
[93] Watt, D., 2007	C	PO	AU	OA	5	32	PAS-M	+/=	PWS
[94] White N, 2017	C	ADO	UK	A	5	73	PAQ	+ Anxious attachment	QOLS
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	=	SF-36

† Both psychological wellbeing and mental health outcomes. †† = ages 45 to 82 years. * Mental health outcome only. Design: C = cross-sectional; L = longitudinal; Q = qualitative; R = representative sample. Group: ADO = assistance dog owners; CA = cancer patients; DO = dog owners; HR = recreational horseback riders; LPO = Latino pet owners; MI = individuals with severe mental illness; PO = community-residing pet owners or recent pet owners; STR = stroke survivors. Area: AU = Australia; CAN = Canada; ISR = Israel; NZ = New Zealand; SGP = Singapore; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CCA = Comfort from Companion Animals Scale [38]; CDORS = Cat/Dog Owner Relationship Scale [45]; I-1 = Single Item (1–10) [14,36,37]; INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; MDORS = Monash Dog Owner Relationship Scale [47]; MSPSS = Multidimensional Scale of Perceived Social Support—modified for pets [48]; OPR = Owner–Pet Relationship Questionnaire [49]; PAI = Pet Attitude Inventory [50]; PALS = Pet Attachment and Life Impact Scale [51]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-M = Pet Attachment Scale-Meehan [52]; PRS = Pet Relationship Scale [54]. Highlighted findings use scales with lower attachment indicated by a higher score and reported in that manner in the article. The directions of associations are retained in this table, but discussed in the context of more attachments having a higher score in the summary tables. + = positive; = = no relationship; − = negative; M = moderator; ME = mediated. Wellbeing and Mental Health Measures: BSI = Brief Symptom Inventory [95]; FACA = Functional Assessment of Cancer Therapy-General Scale Version 4 [96,97]; FAHW (German questionnaire for habitual wellbeing) [98]; INT = Interview; ISAI = Iowa Self Assessment Inventory [99]; MHI = Mental Health Inventory [83]; OHS = Ontario Health Survey [100]; ONS = Wellbeing—Office for National Statistics [UK] Health & Lifestyle survey [101]; PWS = Psychological Wellbeing Scale [102]; QOLS = Quality of Life Scale [103]; RAND = RAND 36-item Health Survey [104]; SPHQ = Self-Perceived Health Questionnaire [Johnson, 2002 #43]; SF-12 MCS = Medical Outcome Study Health Survey Short Form-12 Mental Component Score [105,106]; SF-36 = Medical Outcome Study Health Survey Short Form-36 [107]; SPANE = Scale of Positive and Negative Experience [108]; WHOQOL-BREF = World Health Organization Quality of Life assessment, Brief Version [109].

). Additionally, two articles [26,60] assessed mental health by assessing an individual’s perceptions of their symptoms, and one article [66] included assessment of both psychological wellbeing and mental health. Sixteen of the 17 articles [24,25,28,29,32,49,56,63,66,68,70,88,90,91,92,93,94] that included psychological or emotional wellbeing used quantitative and one [33] used qualitative research methods. Both articles that focused only on mental health [26,60] were quantitative. The OCEBM ratings of quality were category 2c for three articles [26,88,92], 3b for four articles [28,29,63,68], 4 for two articles [56,60], and 5 for ten articles [24,25,32,49,66,70,90,91,93,94].

The quantitative articles that evaluated the relationship between pet attachment and psychological/emotional wellbeing used 10 different scales to assess pet attachment. Most of the articles used one scale, but four [25,28,29,32] used multiple scales. The most frequently used scale was the LAPS (7 articles), followed by the PAQ (3 articles), and the MDORS/CDORS, OPR, and PAS-H (2 articles each). Other scales that were used in one article included the CCA, MSPSS, PALS, PAS-M, or the combination of the PRS and PAI.

Assessment of psychological or emotional wellbeing also used many different tools; no tool was used in more than three articles (see Table 9). Five of the quantitative articles used three inter-related scales: the RAND-36 (1 article), the Medical Outcomes Study Short Form 36 (SF-36; 1 article), and the Short Form-12 (3 articles). RAND-36 is an adaptation of the SF-36 intended for public use, and the SF-12 is a shortened form of the SF-36. Tools used twice include the PWS and the WHO-QOL BREF. Eight other tools were used once. Both quantitative articles focused on mental health and used the Brief Symptom Inventory (BSI) to assess mental health.

Nine of the quantitative articles [32,49,56,68,88,91,92,93] and the qualitative article [33] included only older adult participants; the remaining seven quantitative articles focused on wellbeing [25,28,29,63,70,90,94], and all three articles that included analyses focused on mental health included a mix of older and younger participants. Among the nine quantitative articles that included only older adults, two [88,92] were longitudinal, and the seven others [24,32,49,56,68,91,93] were cross-sectional; all of these articles included convenience samples. All of the articles about mental health were cross-sectional, although one [60] asked about previous mental health symptoms and then evaluated changes in mental health based on these retrospective data. Among the cross-sectional articles, there was no relationship between pet attachment and psychological/emotional quality of life in healthy community-residing older adults in the USA [32,56,68] (N’s = 178, 24, 90) or Australia (n = 179) [49]. Similarly, Israeli older adult stroke survivors (n = 25) did not demonstrate a relationship between pet attachment and psychological wellbeing [91]. In a sample of 32 Australian suburban older adults living in their own homes or a retirement community, pet attachment was related to the environmental mastery component of psychological quality of life, assessed with the PWS, but not to autonomy, personal growth, positive relations with others, purpose in life, or self-acceptance [93]. The qualitative article also reported no relationship between pet attachment and psychological wellbeing among Australian older adults [33]. In the longitudinal article, there was no relationship between Canadians’ pet attachment and older adults’ (n = 286) psychological wellbeing or their changes in psychological wellbeing over a year [92].

The nine [24,25,28,29,63,66,70,90,94] articles that included a mix of older and younger participants all reported on cross-sectional quantitative analyses, and all but one [63] included convenience samples. Four of the articles that included, but were not limited to, older adults reported positive relationships between pet attachment and psychological wellbeing [28,29,66,90] and one reported a negative relationship [94]. Those articles included samples of dog owners (n = 639) [90] and pet owners (n = 322) [28] in Australia, pet owners (176) in New Zealand [66], pet owners (n = 431) in SGP [29], assistance dog owners (n = 73) in the UK [94], as well as dog owners and horse owners who rode them for recreation in Austria (n = 124) [24]. Austrian adult owners did not demonstrate a relationship between pet attachment and psychological wellbeing, while recreational horseback riders demonstrated a positive relationship during these activities [24]. The other three articles, including a representative sample (n = 81) of UK individuals with severe mental illness [63], a convenience sample of pet owners from Australia, the UK, and the USA (n = 238), as well as a few other countries [25], and American cancer patients who owned dogs (n = 122) [70], did not demonstrate relationships between pet attachment and psychological wellbeing.

In addition to examining direct relationships, articles suggested that pet attachment may play an important role in mediating or moderating the effects of other variables on psychological wellbeing. In an article focusing on Australian adult dog owners (n = 639), the strength of the human–animal bond, as assessed with the MDOR S, mediated the associations of family loneliness and romantic loneliness with poorer psychological wellbeing [90]. However, it did not moderate the relationship between structural pet support and psychological wellbeing in adult pet owners (n = 238) from AU, the UK, the USA, and other countries [25]. In an article that focused on American older adult pet owners residing in the community (n = 178), pet attachment moderated the relationship between physical wellbeing and psychological wellbeing [56]. The higher the attachment, the less strong the relationship, so that at high attachment, poor physical wellbeing was not associated with poor psychological wellbeing, while at low attachment, the two variables were strongly associated with each other.

The relationship between pet attachment and mental health appears complex. Pet attachment was not related to mental health in a cross-sectional study from New Zealand [66], although pet attachment was related to psychological wellbeing, and dog attachment was related to higher severity of mental health symptoms in 607 individuals from several countries who completed the online survey [26]. Avoidant and anxious dog attachment is also associated with worse mental health. The relationship between attachment and the severity of mental health problems was mediated by closeness, dependence, and anxious adult attachment to other people. Gender moderated the relationship between pet attachment and mental health; there was a significant relationship in women but not in men. The second article, focusing on mental health, evaluated the role of pet attachment in moderating changes in mental health symptoms in adults (n = 1942) from before to after the COVID-19 pandemic [60]. Again, the relationship was complex. The less severe the mental health symptoms prior to the pandemic, the greater the relationship between pet attachment and reduction in symptoms. For individuals with severe mental health symptoms prior to the pandemic, the higher the attachment, the greater the odds of retaining more severe symptoms.

Summarizing the findings with respect to the relationship between pet attachment and psychological wellbeing and related outcomes (

Table 10. Summary of the number of analyses reporting the direction of the relationship between pet attachment and emotional or psychological wellbeing or health or quality of life in the quantitative articles including older adults (Adults) and those limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	7	1	8
=	5	8	13
−	4	1	5
Total	16	10	26

Note: Some articles contain multiple outcomes. Relationship: + = positive relationship, = = no relationship, − = negative relationship of pet attachment and the health outcome.

), in 50% of the articles, pet attachment was not related to these outcomes in older adults. In articles limited to older adults, 80% demonstrated no relationship, 10% demonstrated a negative relationship, and 10% demonstrated a positive relationship between pet attachment and psychological wellbeing and related outcomes. In articles including older adults, the relationships between pet attachment and psychological wellbeing were distributed across three different valences: 44% demonstrated a positive relationship, 31% demonstrated no relationship, and 25% demonstrated a negative relationship. It is clear that there is no consistent relationship between pet attachment and psychological wellbeing, mental health, or quality of life in either group. The quality of the evidence was equivalent for articles that included but were not limited to older adults, 56% of which had OCEBM ratings of 5, and those limited to older adults, 56% of which had ratings of 5. The evidence suggests a more positive relationship in younger adults and no relationship in older adults, especially in those with health challenges such as needing an assistance dog or having been diagnosed with cancer.

3.1.3. Anxiety

Pet attachment was examined in relation to anxiety in 11 articles [27,28,33,36,56,59,63,66,67,68,73] (

Table 11. Articles that included the relationship between pet attachment and anxiety (n = 11).

Article	Design Type	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Anxiety Measures
[66] Budge, R. C. 1998	C	PO	NZ	A	5	176	PAS-H	=	MHI
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA	N/A	11	INT	=	INT
[67] Ellis, A., 2024	C	PO	UK	A	5	1359	LAPS	+, ME	GAD-7
[68] Friedmann, E. 2020	C	PO	USA	OA	3b	90	LAPS	=	PSS
[36] Kogan, L.R. 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	−	QU
[56] Koodaly, L. 2025	C	PO	USA	OA	4	178	LAPS	=, M psychological wellbeing to	PSS
[27] Reevy, G. 2020	C	CO	USA	A	3b	1239	LAPS, PAQ	+	IPIP
[63] Shoesmith, E., 2023	C, R	MI	UK	A	3b	81	CCA	−	GAD-2
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	=	DASS
[73] Watson, N. L. 1993	C	EW	USA	A	3b	42	CAAS	=	STAI
[59] Zebrowska M, 2024	C	PO	USA	OA	3b	140	LAPS subset	−, S−	GAD-7

Design: C = cross-sectional; L = longitudinal; Q = qualitative; R = representative sample. Group: CO = cat owners; EW = employed women; MI = individuals with severe mental Illness; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CAN = Canada; NZ = New Zealand; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS = Companion Animal Attachment Scale [43]; CCA = Comfort from Companion Animals Scale [38]; I-1 = Single Item (1–10) [14,36,37]; INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative. Anxiety Measures: DASS = Depression, Anxiety, and Stress Scales-21 [78,79]; GAD-2 = Generalized Anxiety Disorder Scale-2 [110]; GAD-7 = Generalized Anxiety Disorder Scale—7 [111]; I-1 = Single Item (1–10) [14,36,37]; INT = Interview; IPIP = International Personality Item Pool [82]; MHI = Mental Health Inventory [83]; QU = Questions developed for this study; PSS = Perceived Stress Scale [112] (subset); STAI = State-Trait Anxiety Inventory [113].

). Ten of the 11 articles used quantitative research methods. The OCEBM ratings of quality were category 2c for one article [36], 3b for six articles [27,28,59,63,73], 4 for one article [56], and 5 for two articles [66,67]. Three of the quantitative articles [56,59,68] and the qualitative article [33] included only older adult participants; the remaining seven articles had a mix of older and younger participants, with younger participants predominating.

The ten quantitative articles used seven different scales (Table 3) to assess pet attachment. Most of the articles used one method, but two articles [27,28] used two different scales. The most used pet attachment scale was the LAPS, which was used in five of the articles. The PAQ was used twice. One article used a single item, and four other scales, the PAS-H, CCA, CASS, and OPR, were used once each.

Likewise, anxiety was also assessed with seven different scales in the 10 quantitative articles. The most commonly used instrument was the Generalized Anxiety Disorder (GAD) scale; the GAD-7 was used in two articles, and the GAD-2 was used in one. Other anxiety measurement tools found in these articles include a subset of items from the Perceived Stress Scale (PSS) used in two articles and the DASS-21, International Personality Item Pool (IPIP), Mental Health Inventory (MHI), State-Trait Anxiety Inventory (STAI), and a single newly developed item each used in one article. The qualitative article featured interviews of older adults who discussed feelings of anxiety surrounding pet ownership and attachment.

The longitudinal survey of pet owners assessed the impact of COVID-19 on pet attachment and its relationship to changes in anxiety over the first few months of the pandemic [36]. The other eight quantitative articles that assessed the relationship between pet attachment and anxiety were cross-sectional. One of these surveys [63] used data from a representative sample; the other surveys used convenience samples.

The four articles that were limited to older adults were based on studies conducted in the USA. The three quantitative articles all used the LAPS to assess pet attachment. Pet attachment was not related to anxiety in two articles about cross-sectional samples of generally affluent community-dwelling older adults (N’s = 90, 178, respectively) with anxiety assessed with the PSS [56,68] or in a sample of pet owners (n = 140) with a heavy skew towards survivors of child abuse and with anxiety assessed by GAD-7 [59].

Pet attachment was also negatively related to anxiety in a subset of dog but not cat owners [59]. In one sample of health community-residing older adults (n = 178), pet attachment moderated the relationship between anxiety and physical wellbeing. In pet owners with high anxiety, the greater the pet attachment, the greater their physical wellbeing; in less anxious older adults, pet attachment was not related to physical wellbeing [56]. In the qualitative article about older adult pet owners, participants indicated that increased anxiety levels in older adults with pets (n = 11) over travel and concern for abandoning their beloved animal were associated with greater attachment [33].

Of the seven survey articles that included but did not limit their sample to older adults, three found no association between pet attachment and anxiety [28,66,73]. These articles were conducted in NZ, AU, and the USA (respectively). Two of them focused on pet owners (N’s = 176, 322) [28,66], and the third focused on employed women (n = 42) [73]. There was no commonality in the tools used to assess pet attachment or anxiety among the three articles.

Higher pet attachment was related to lower anxiety in two cross-sectional articles that included older adults [36,63]. One of these articles included a large sample (n = 5061) from a number of English-speaking countries but used study-developed single-item measures to assess both pet attachment and anxiety [36]. The other article also used a very short measure, the GAD-2, to assess anxiety and surveyed individuals (n = 81) with severe mental illness [63].

Two articles that included, but were not limited to, older adults found a positive relationship between pet attachment and anxiety [27,67]. Both articles included an assessment of pet attachment with the LAPS and were based on online surveys. Greater pet attachment in a UK sample (n = 1359) was correlated with higher anxiety levels, which opposed their initial hypothesis [67]. Further, in this sample, loneliness mediated the relationship between pet attachment and anxiety. In a survey conducted in the U.S. (n = 1239), pet attachment was positively related to anxiety [27]. In this sample, younger age was also related to higher anxiety.

Overall, these articles lack consistency in evidence regarding the relationship between pet attachment and anxiety levels (see

Table 12. Summary of the direction of the relationship between pet attachment and anxiety in analyses in quantitative articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	3	0	3
=	3	3	6
−	0	1	1
Total	6	4	10

+ = positive relationship, = = no relationship, − = negative relationship.

) in pet owners, which prompts the need for further research into this topic. Among older adults, 75% (3/4) of the analyses did not demonstrate a relationship between pet attachment and anxiety, and 25% (1/4) demonstrated a negative relationship. In contrast, among analyses of adults, including older adults, 50% (3/6) demonstrated no relationship between pet attachment and anxiety, and 50% (3/6) demonstrated a positive relationship. Based on the evidence, there is no conclusive support for an association between pet attachment and anxiety in older adults. The pattern of results suggests that pet attachment could be associated with anxiety about pets but not general anxiety in older adults.

3.1.4. Distress or Stress

Pet attachment was examined in relation to psychological distress or stress in 11 articles [14,21,26,28,30,59,61,66,74,114,115] (

Table 13. Articles that relate pet attachment to psychological distress or stress (n = 11).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Distress or Stress Measures
[14] Branson, S., 2016	C, B	H, RU	USA	OA	3b	48	I-1	=	CRP
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	=	MHI
[21] Joseph, N., 2019	CC	PO	IND	A †	3b	122	LAPS	=	PSS
[114] le Roux, M. C., 2020	C	PO	ZA	A	4	3329	CCA	=	PSS
[26] Northrope, K., 2024	C	DO	USA and other countries	A	2c	607	LAPS, PAQ	+, ME	BSI
[30] Peacock J, 2012	C	PO	AU	A	5	150	PAS-H, OPR	+	BSI-18
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	−, ME	DASS
[61] Wan, M., 2022	L	E	USA	A	5	187	CCA	=, M job insecurity to	SS
[74] Wells, D. L., 2022	C	UKO	UK	A	3b	146	LAPS	=	PSS- 4
[115] Wu, C. S. T., 2018	C	PO	HK	A	5	288	CABS	−, ME	PSS-C
[59] Zebrowska M, S 2024	C	PO	USA	A	3b	140	LAPS subset	=	KPDS

† ages 12 + years. Design: B = biomarkers; C = cross-sectional; CC = case–control; L = longitudinal. Group: DO = dog owners; E = employees; H = homebound; PO = community-residing pet owners or recent pet owners; RU = rural; UKO = UK residents. Area: AU = Australia; HK = Hong Kong; IND = India; NZ = New Zealand; UK = United Kingdom; USA = United States of America; ZA = South Africa. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CABS = Companion Animal Bonding Scale [44]; CCA = Comfort from Companion Animals Scale [38]; I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; ME = mediated. Stress/Psychological Distress Measures: BSI = Brief Symptom Inventory [95]; BSI-18; 18 Item BSI [76]; CRP = Salivary C-Reactive Protein [116]; DASS = Depression, Anxiety, and Stress Scales-21 [78,79]; KPDS = Kessler Psychological Distress Scale [117]; MHI = Mental Health Inventory [83]; PSS = Perceived Stress Scale [112]; PSS-4 = Perceived Stress Scale 4 item version14; PSS-C = Perceived Stress Scale; Chinese Version [118]; SS = Stress Scale from BSI [119].

), all of which used quantitative research methods. The OCEBM ratings of quality were 2c for one article [26], 3b for five articles [14,21,28,59,74], 4 for one article [114], and 5 for four articles [30,61,66,115].

The 11 quantitative articles that evaluated the relationship between pet attachment and stress or psychological distress used seven different scales to assess pet attachment. Most of the articles used one scale, but three [26,28,30] used multiple scales. The most frequently used scale was the LAPS (four articles), followed by the CCA, PAQ, PAS-H, and OPR (two articles each), and the CABS or a single item (one article each).

Stress or distress was assessed with nine different tools in the 11 articles. One article used a biomarker [14]. Three different versions of the Perceived Stress Scale were used in a total of four articles [21,74,114,115]; the only other tool used in more than one article was the Brief Symptom Inventory, which was used in two articles [26,30].

Ten articles presented cross-sectional designs, while one presented a longitudinal two-wave design [61]. One article [14] included only older adult participants and used a biomarker and a survey to assess stress; the remaining 10 articles had a mix of older and younger participants, with younger participants predominating, and used survey tools to assess stress or distress. One article included only housebound older adults (n = 48) and assessed pet attachment with a single item, describing no association of pet attachment with stress. Among the ten articles that included older adults among their participants, two articles demonstrated positive relationships between pet attachment and distress, two articles demonstrated negative relationships between pet attachment and distress, and six articles demonstrated no relationship between pet attachment and psychological distress. The two articles that demonstrated positive relationships included 150 pet owners from Australia [30] and 607 dog owners from the USA and other countries [26]. In the former article [30], based on a mail survey, pet attachment was an independent predictor of psychological distress after controlling for demographic factors, but it did not moderate the relationship between loneliness and psychological distress. In the latter article [26], higher dog attachment and insecure dog attachment were related to mental distress, and the relationship between attachment and distress was mediated by anxious attachment styles toward other people. The article also demonstrated that gender moderated the relationship between higher dog attachment and higher distress; it was significant for women but not for men.

The two articles that demonstrated negative relationships between pet attachment and psychological wellbeing included 322 Australian adult pet owners [28] and 288 Chinese adult pet owners [115]; both sets of analyses controlled for other variables, including demographic characteristics of the participants. In the Australian article, higher secure pet attachment specifically was related to lower distress relationships between a strong attachment to dogs and poorer mental health. The relationship between stronger attachment and mental distress was moderated by gender; it was significant for women but not for men. In the Chinese article, pet attachment mediated the relationship between time spent caring for a pet and stress; the more time spent caring, the greater the attachment and the lower the stress.

The six cross-sectional surveys that found no association between pet attachment and stress or distress included 176 New Zealand pet owners [66], 122 Indian pet owners [21], 3329 South African pet owners [114], 146 UK resident pet owners [74], and 140 American pet owners [59]. The one article about a longitudinal study of employees (n = 147) during the COVID-19 pandemic demonstrated no direct relationship between pet attachment and stress. However, pet attachment moderated the relationship between stress and the use of negative coping strategies, such as marijuana and alcohol use [61].

Overall, the evidence that there is a direct relationship between pet attachment and stress/distress is not conclusive. In articles focused on adult participants, two analyses demonstrated positive relationships, six analyses demonstrated no relationship, and two analyses demonstrated negative relationships between pet attachment and distress. Only one article that focused only on older adults reported an analysis of the relationship between pet attachment and distress; it found no relationship between the two variables. No conclusion about the relationship between pet attachment and stress/distress is possible based on the limited information available.

3.1.5. Grief over Pet Loss

Pet attachment was examined in relation to grief over pet loss in seven articles [31,34,58,120,121,122,123], one of which [34] used qualitative methodology, and one of which used both qualitative and quantitative methodology [122] (

Table 14. Articles that relate pet attachment to grief over pet loss (n = 7).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Grief Measures
[120] Barnard-Nguyen, S., 2016	C, R	BPO	USA	A	2C	409	LAPS	+	PBQ
[121] Brown, C. A., 2023	C	BPO	CAN	OA	5	98	PALS	+	PBQ, WDG
[34] Kwong MJ, 2011.	Q	ADO	CAN	A	Q	25	INT	+
[123] Lykins, A. 2025	C	BPO	AU	A	4	496	PAQ	+	PBQ, CBI
[31] Planchon, L. A., 2002	C	BPO	USA	A	5	454	PAS-H, PAS-T	+	NGS
[58] Testoni, I.,2017	C	BPO	ITA	A	4	159	LAPS	+	PBQ
[122] Wilson, D.M., 2021	C, Q	W, BPO, LA	CAN	OA	5	12	PALS	+	PBQ

Design: C = cross-sectional; Q = qualitative; R = representative sample. Group: ADO = assistance dog owners; BPO = bereaved pet owners; LA = lives alone; W = women. Area: AU = Australia; CAN = Canada; ITA = Italy; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; PALS = Pet Attachment and Life Impact Scale [51]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-T = Pet Attitude Scale-Templer [53]. Relationship of pet attachment to health outcome: + = positive. Grief Measures: CBI = Continuing Bonds Inventory (unpublished); NGS = Number of Grief Symptoms [124]; PBQ = Pet Bereavement Questionnaire [125]; WDG = Witnessing of Disenfranchised Grief [126].

). The OCEBM ratings of quality for the articles involving quantitative components were category 2c for one article [120], 4 for two articles [58,123], and 5 for two articles [31,122].

Six articles included five different tools to assess pet attachment. Two articles used the LAPS, and two articles used the PALS. One article used the PAQ, and one article [124] used the PAS-H and the PAS-T.

One tool, the Pet Bereavement Questionnaire, was used to assess grief in five [58,120,121,122,123] of the six quantitative articles, including both of those that focused exclusively on older adults. One article [34] used the Number of Grief Symptoms to assess grief. Two articles [121,123] also included secondary scales to assess specific aspects of grief.

Two articles included only older adults [121,122]. The qualitative article and the four remaining quantitative articles had a mix of older and younger participants, with younger participants predominating.

The participants described in all of the articles were individuals who had experienced pet loss due to death, except for the qualitative article [34], which included individuals who had lost an assistance dog due to its retirement or death. One article included a representative sample of individuals (n = 409) who had euthanized a pet within the past year [120]; all other quantitative articles included convenience samples. In the two articles focused on Canadian older adults whose pets had died naturally or via euthanasia, women who lived alone (n = 12) [122] and whose pets died during the past year and a larger group of similar older adults (n = 98) [121] were very attached to their pets and experienced considerable grief.

The five articles that included adults of all ages included similar conclusions. In the qualitative article about individuals who lost the use of an assistance dog, attachment was included as an important theme around dog loss. Most participants also expressed their intense grief over the loss of their service dogs [34]. Among veterinary clients from the USA who had recently experienced dog or cat loss (n = 63) and college students who had experienced pet loss (n = 391), higher pet attachment was associated with greater grief [31]. There was a direct relationship between pet attachment and grief among Italian adults (n = 159) who had experienced the death of a pet [58]. Similarly, pet attachment was directly related to grief in Americans (n = 409) who had euthanized a pet within the last year [120]. In a large sample (n = 496) of Australians who had experienced the death of a pet within the last three years, grief was correlated with both attachment anxiety and attachment avoidance, independent of each other and of time since loss, which were both significant predictors of grief [123].

Although the quality of the articles was not generally high, the findings were consistent. Based on the evidence from both articles about older adults and those about all adults, pet attachment is related to grief when the pet dies or is euthanized. Similar effects also likely occur when a service dog retires from service. It is possible that those with the highest grief following the loss of the pet tend to report higher attachment to their pet as a way of explaining the depth of their feelings. To fully understand the connection between attachment and pet loss, one needs to measure attachment to the pet prior to their passing and then assess grief.

3.1.6. Positive Affect or Emotions

Pet attachment was examined in relation to positive affect or emotions in six articles [36,66,67,70,74,90], all of which are quantitative methodologies (

Table 15. Articles that relate pet attachment to positive affect (n = 6).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Positive Affect Measures
[90] Allen, C. J., 2022.	C	DO	AU	A	5	639	MDORS	=, ME	SPANE
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	+	MHI
[67] Ellis, A., 2024	C	PO	UK	A	5	1359	LAPS	+, ME	PANAS
[70] Ingram, KM, 2019	C	CA, DO	USA	A	5	122	LAPS	=	CES-DS
[36] Kogan, L. R., 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	+	QU
[74] Wells, D. L., 2022	C	UKO	UK	A	3b	146	LAPS	−	SPANE

Design: C = cross-sectional; L = longitudinal. Group: CA = cancer patients; DO = dog owners; PO = community-residing pet owners or recent pet owners; UKO = UK residents. Area: AU = Australia; CAN = Canada; New Zealand; UK = United Kingdom; USA = United States of America. Age: A = adult. OCEBM = OCEBM rating. Pet attachment scales: I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; MDORS = Monash Dog Owner Relationship Scale [47]; PAS-H = Pet Attachment Survey-Holcomb [41]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative. Positive Affect Measures: CES-DS = Centers for Epidemiologic Studies—Depression Scale Positive Affect Subscale [127]; MHI = Mental Health Inventory [83]; PANAS = Positive and Negative Affect Schedule [128]; QU = questions developed for this study; SPANE = Scale of Positive and Negative Experience [108].

). The OCEBM ratings of quality for the articles involving quantitative components were category 2c for one article [36], 3b for one article [58,74,123], and 5 for four articles [66,67,70,90].

Pet attachment was assessed with the LAPS in half of the articles [67,70,74] and three different tools in the other three studies. Positive affect was assessed with the Scale of Positive and Negative Experience (SPANE) in two articles [74,90] and different tools in the other four articles. All six articles included, but were not limited to, older adult participants and included convenience samples.

One article was longitudinal [36], examining how pet attachment impacted the response to the first few months of the COVID-19 epidemic. Although this article examined positive affect, it did not include summary scores. In this article, pet attachment was related to the five aspects of positive affect involved with coping with the stress and social isolation of the pandemic: higher self-compassion, ability to maintain a regular schedule, sense of purpose and meaning, and coping with uncertainty.

The remaining five articles were cross-sectional. Two cross-sectional articles reported no association between pet attachment and positive affect; they included Australian dog owners (n = 639) [90] and American cancer patients (n = 122) [70]. In the former article, bond strength mediated the relationship between family loneliness and positive affect; the strength of the bond reduced the negative effects of family loneliness and romantic loneliness on positive affect. Two cross-sectional articles reported a positive association between pet attachment and positive affect; they included 176 pet owners in NZ [66] and 1359 pet owners in the UK [67]. In the latter article, loneliness mediated the relationship between pet attachment and positive affect. In the path analysis model, pet attachment and loneliness both had a significant direct negative effect on positive affect, and attachment had a direct positive effect on positive affect. Attachment also had an indirect negative effect on positive affect through loneliness. In one article about UK resident pet owners (n = 146), higher pet attachment was related to lower positive affect.

The articles that assess the relationship of pet attachment to positive affect fail to show a consistent pattern. Three of six analyses of adults showed a positive relationship, two showed no relationship, and one showed a negative relationship. Further, there were no articles about older adults to compare with the articles that included them. It is not possible to ascertain the relationship or whether the pattern of older adults may be different from that of younger individuals.

3.1.7. Cognitive Health Outcomes

A group of five articles [14,32,65,68,129] includes examination of the relationship between pet attachment and cognition and/or specific aspects of cognition (Table 16), such as executive function (three articles) [65,68,129] and memory (two articles) [68,129]. All five articles include only older adults and are convenience samples of populations in the USA. One article includes only cat owners, one includes only Latino pet owners, and one includes only homebound older adults. One article is longitudinal [129]; the others are cross-sectional. The articles are OCEBM categories 2c [129], 3b [14,68], and 5 [32,65].

Overall, there was no relationship between cognitive measures and pet attachment in the community-dwelling participants. In one article, among 48 homebound older adults, pet attachment was related to executive function [65]. In the only longitudinal article, pet attachment (n = 214) was related to slower deterioration in executive function but faster deterioration in long-term memory over an average of 7.8 years after controlling for age and number of comorbidities [129]. There was no relationship between pet attachment and changes in short-term memory.

Six analyses examined the relationship between pet attachment and cognition in older adults; in five of them, no relationship was demonstrated. No articles examined the relationship between pet attachment and cognition in the broader adult population. Taken together, these findings suggest that pet attachment is not related to cognitive functioning in generally healthy older adults living in the community. They also suggest that there may be differences in the relationship between pet attachment and cognition among more impaired older adults and those who are generally healthy. Further, pet attachment may be related to changes in cognitive function with aging.

3.1.8. Other Positive Psychological Outcomes

Thirteen articles related to pet attachment to positive psychological health outcomes that are assessed in four or fewer articles (Table 4). Fourteen different positive psychological health outcomes (autonomy, bad mood, coping with uncertainty, emotional ties, emotional balance, environmental mastery, happiness, good mood, life satisfaction, maintaining a regular schedule, personal growth, self-acceptance, self-compassion, and sense of purpose or meaning) were assessed in 13 articles (Table 17). The 13 articles include 1 with an OCEBM rating of 2c [36], 3 with a rating of 3b [17,68,69], 4 with a rating of 4 [23,56,114,130], 4 with a rating of 5 [25,32,66,93], and one qualitative article [35]. Five of the articles include older adults only, seven of the articles include but are not limited to older adults, and one article includes older adult women and adult men [69].

Pet attachment was assessed with ten different tools. Five articles used the LAPS. One article used two different tools, the PRS and the PAI. All other quantitative articles used different tools to assess pet attachment. The qualitative article used an interview to assess pet attachment.

Table 16. Articles that relate pet attachment to cognition and related outcomes (n = 5).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Cognition and Related Measures
[65] Branson, S. M., 2019	B, C	CO	USA	OA	5	41	LAPS	= cognitive function	MoCA
[14] Branson, S., 2016	B, C	H, RU	USA	OA	3b	48	I-1	+ executive function	CLOX 1
[129] Friedmann, E., 2025	L	PO	USA	OA	2c	214	LAPS	−↓ executive function +↓ memory (LTR) =↓ memory (STR)	Trails CVLT CVLT
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	= memory = cognition = executive function	CVLT WAIS Trails
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	= cognition	ISAI

Design: B = biomarkers; C = cross-sectional; L = longitudinal. Group: CO = cat owners; H = homebound; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; RU = rural. Area: USA = United States of America. Age: OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; PAI = Pet Attitude Inventory [50]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; ME = mediated. Relationship of pet attachment to cognitive outcomes: + = positive; = = no relationship; −↓ = slower deterioration; =↓ = no relationship to deterioration; +↓ = faster deterioration. Cognition and Related Outcome Measures: CLOX 1 = Clock Drawing Executive Test [131]; CVLT = California Verbal Learning Test [132]; ISAI = Iowa Self-Assessment Inventory [99]; MoCA = Montreal Cognitive Assessment [133]; Trails: Trail-Making Test (Trails A; B; B-A) [134]; WAIS = Weschler Adult Intelligence Scale—Revised [135].

Table 16. Articles that relate pet attachment to cognition and related outcomes (n = 5).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Cognition and Related Measures
[65] Branson, S. M., 2019	B, C	CO	USA	OA	5	41	LAPS	= cognitive function	MoCA
[14] Branson, S., 2016	B, C	H, RU	USA	OA	3b	48	I-1	+ executive function	CLOX 1
[129] Friedmann, E., 2025	L	PO	USA	OA	2c	214	LAPS	−↓ executive function +↓ memory (LTR) =↓ memory (STR)	Trails CVLT CVLT
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	= memory = cognition = executive function	CVLT WAIS Trails
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	= cognition	ISAI

Design: B = biomarkers; C = cross-sectional; L = longitudinal. Group: CO = cat owners; H = homebound; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; RU = rural. Area: USA = United States of America. Age: OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; PAI = Pet Attitude Inventory [50]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; ME = mediated. Relationship of pet attachment to cognitive outcomes: + = positive; = = no relationship; −↓ = slower deterioration; =↓ = no relationship to deterioration; +↓ = faster deterioration. Cognition and Related Outcome Measures: CLOX 1 = Clock Drawing Executive Test [131]; CVLT = California Verbal Learning Test [132]; ISAI = Iowa Self-Assessment Inventory [99]; MoCA = Montreal Cognitive Assessment [133]; Trails: Trail-Making Test (Trails A; B; B-A) [134]; WAIS = Weschler Adult Intelligence Scale—Revised [135].

Table 17. Articles that relate pet attachment to additional positive psychological outcomes (n = 13).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Other Positive Psychological Measures
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	= emotional ties	MHI
[130] Curl, A. L., 2021	C, R	PO	USA	OA	4	188	HRS	+ life satisfaction, ME through dog walking and social support to life satisfaction	LS item
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	= happiness	H item
[69] Fritz, C. L., 1996	C	CG	USA	OA, W A, M	3b	38 27	LAPS	= Life satisfaction = Life satisfaction	LSI-Z
[25] Hardie, S., 2023	C	PO	AU, USA, UK	A	5	238	CDORS, MSPSS	M perceived pet support to life satisfaction (moderator pet emotional closeness)	SLS
[17] Hill, L., 2020	C	PO	AU	A	3b	392	LAPS	No M social support to resilience	CD-RISC
[32] Johnson, R.A., 2002	C	LPO	USA	A	5	24	PRS, PAI subset	= emotional balance	ISAI
[36] Kogan, L. R., 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	+ self-compassion, + ability to maintain a regular schedule, +sense of purpose and meaning, + coping with uncertainty.	QU
[56] Koodaly, L., 2025	C	PO	USA	OA	4	178	LAPS	= Happiness	H item
[114] le Roux, M. C., 2020	C	PO	ZA	A	4	3329	CCA	= Life satisfaction	SLS
[35] Mueller, R. 2024	Q	DO, M	USA	OA		12	INT	+ Meaning and purpose	N/A
[23] Turner, D.C. 2003	C	CO	SUI	A ††	4	516	LAPS	+ M living with partners to good mood	EWL
[93] Watt, D. 2007	C	PO	AU	OA	5	32	PAS-M	− Environmental mastery = autonomy, = personal growth, = purpose in life, = self-acceptance	PWS

†† aged 45 to 53 years. Design: C = cross-sectional; L = longitudinal; Q = qualitative; R = representative sample. Group: CG = caregivers of Alzheimer’s patients; CO = cat owners; DO = dog owners; LPO = Latino pet owners; M = men; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CAN = Canada; NZ = New Zealand; SUI = Switzerland; UK = United Kingdom; USA = United States of America; ZA = South Africa. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CCA = Comfort from Companion Animals Scale [38];CDORS = Cat/Dog Owner Relationship Scale [45]; HRS = Health and Retirement Study [46]; I-1 = Single Item (1–10) [14,36,37]; INT = interview; LAPS = Lexington Attachment to Pets Scale [39]; MSPSS = Multidimensional Scale of Perceived Social Support—modified for pets [48]; PAI = Pet Attitude Inventory [50]; PALS = Pet Attachment and Life Impact Scale [51]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-M = Pet Attachment Scale-Meehan [52]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; M = moderator. Additional Positive Psychological Outcome Measures: CD-RISC = The Connor–Davidson Resilience Scale [136]; EWL = List of Adjectives [137]; H item = happiness (single item 0–10) [68]; ISAI = Iowa Self-Assessment Inventory [99]; LS item = Life Satisfaction (Single Item 0–4) [46]; LSI-Z = Life Satisfaction Index-Z [138]; MHI = Mental Health Inventory [83]; PWS = Psychological Wellbeing Scale [102]; SLS = Satisfaction with Life Scale [139]; QU = questions developed for this study; Trails: Trail-Making Test (Trails A, B, B-A) [134].

Table 17. Articles that relate pet attachment to additional positive psychological outcomes (n = 13).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Other Positive Psychological Measures
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	= emotional ties	MHI
[130] Curl, A. L., 2021	C, R	PO	USA	OA	4	188	HRS	+ life satisfaction, ME through dog walking and social support to life satisfaction	LS item
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	= happiness	H item
[69] Fritz, C. L., 1996	C	CG	USA	OA, W A, M	3b	38 27	LAPS	= Life satisfaction = Life satisfaction	LSI-Z
[25] Hardie, S., 2023	C	PO	AU, USA, UK	A	5	238	CDORS, MSPSS	M perceived pet support to life satisfaction (moderator pet emotional closeness)	SLS
[17] Hill, L., 2020	C	PO	AU	A	3b	392	LAPS	No M social support to resilience	CD-RISC
[32] Johnson, R.A., 2002	C	LPO	USA	A	5	24	PRS, PAI subset	= emotional balance	ISAI
[36] Kogan, L. R., 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	+ self-compassion, + ability to maintain a regular schedule, +sense of purpose and meaning, + coping with uncertainty.	QU
[56] Koodaly, L., 2025	C	PO	USA	OA	4	178	LAPS	= Happiness	H item
[114] le Roux, M. C., 2020	C	PO	ZA	A	4	3329	CCA	= Life satisfaction	SLS
[35] Mueller, R. 2024	Q	DO, M	USA	OA		12	INT	+ Meaning and purpose	N/A
[23] Turner, D.C. 2003	C	CO	SUI	A ††	4	516	LAPS	+ M living with partners to good mood	EWL
[93] Watt, D. 2007	C	PO	AU	OA	5	32	PAS-M	− Environmental mastery = autonomy, = personal growth, = purpose in life, = self-acceptance	PWS

†† aged 45 to 53 years. Design: C = cross-sectional; L = longitudinal; Q = qualitative; R = representative sample. Group: CG = caregivers of Alzheimer’s patients; CO = cat owners; DO = dog owners; LPO = Latino pet owners; M = men; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CAN = Canada; NZ = New Zealand; SUI = Switzerland; UK = United Kingdom; USA = United States of America; ZA = South Africa. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CCA = Comfort from Companion Animals Scale [38];CDORS = Cat/Dog Owner Relationship Scale [45]; HRS = Health and Retirement Study [46]; I-1 = Single Item (1–10) [14,36,37]; INT = interview; LAPS = Lexington Attachment to Pets Scale [39]; MSPSS = Multidimensional Scale of Perceived Social Support—modified for pets [48]; PAI = Pet Attitude Inventory [50]; PALS = Pet Attachment and Life Impact Scale [51]; PAS-H = Pet Attachment Survey-Holcomb [41]; PAS-M = Pet Attachment Scale-Meehan [52]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; M = moderator. Additional Positive Psychological Outcome Measures: CD-RISC = The Connor–Davidson Resilience Scale [136]; EWL = List of Adjectives [137]; H item = happiness (single item 0–10) [68]; ISAI = Iowa Self-Assessment Inventory [99]; LS item = Life Satisfaction (Single Item 0–4) [46]; LSI-Z = Life Satisfaction Index-Z [138]; MHI = Mental Health Inventory [83]; PWS = Psychological Wellbeing Scale [102]; SLS = Satisfaction with Life Scale [139]; QU = questions developed for this study; Trails: Trail-Making Test (Trails A, B, B-A) [134].

The articles included 11 measures of positive health outcomes. Two articles employed the SLS, and two articles employed a single happiness item. The eight other articles contained unique measures of positive health outcomes.

Five of the articles [35,56,68,93,130] focused on older adults, seven [17,23,25,32,36,66,114] focused on adults, including older adults, and one [69] focused on older adult women and adult men. Among the articles about older adults, two articles addressed the relationship between pet attachment and life satisfaction. In a representative sample of older adult pet owners, pet attachment was positively related to life satisfaction, and pet attachment was a mediator through dog walking and social support to contribute to life satisfaction [130]. In contrast, among 38 older adult female pet owners who were caregivers for individuals with Alzheimer’s disease, there was no relationship [69]. In two articles focused on related samples of healthy community-residing older adult pet owners [56,68], pet ownership was not related to happiness. In the qualitative article focused on 12 older USA adult men who owned dogs [35], pet attachment was associated with meaning and purpose in life. The final article focused on older adults [93], surveyed 32 Australian pet owners, and found no relationship between pet attachment and purpose in life, autonomy, self-acceptance, or personal growth. Pet attachment also was related to lower environmental mastery in this sample.

Articles addressing direct effects of pet attachment to positive psychological health outcomes in adults, including but not limited to older adults, did not demonstrate relationships. Pet attachment was not related to life satisfaction [114] in South African pet owners (n = 3329), USA men who were caregivers for Alzheimer’s patients (n = 27) [69], and an online survey of pet owners from Australia, the UK, and the USA (n = 238) [25]. The articles focused on adults, including older adults, also did not demonstrate significant associations of pet attachment with emotional balance [32] in USA Latino pet owners (n = 24) or emotional ties [66] in New Zealand pet owners (n = 176).

Several articles examined pet attachment as a moderator or mediator in the relationships between other characteristics and positive social outcomes. Among older adults (n = 188), pet attachment influenced life satisfaction through the mediators of the amount of dog walking and frequency of social contacts [130]. Two additional evaluations include moderation in the relationship between pet attachment and positive psychological outcomes in adults. The strength of the bond did not significantly moderate the relationship between social support and resilience among adult Australian pet owners (n = 392) [17]. The authors suggested that this was due to a quadratic relationship between pet attachment and social support. The relationship between cat attachment and good mood depended on the living situation. Cat attachment was related to good mood among 44–53-year-old Swiss adults (n = 516) living with partners but not among those without partners [23].

The articles focused on older adults only present no consistent pattern of the relationship between pet attachment and other positive psychological outcomes (

Table 18. Summary of findings from analyses in articles examining the relationship between pet attachment and additional positive psychological outcomes.

Relationship	N	+	=	−	Other
Life satisfaction	4	[130]	[69]; [114]		[130]
Purpose/meaning in life	3	[35]; [36]	[93]
Happiness	2		[56]; [68]
Autonomy	1		[93]
Coping with uncertainty	1	[36]
Emotional balance	1		[32]
Emotional ties	1		[66]
Environmental mastery	1			[93]
Good mood	1	[23]			[23]
Personal growth	1	[93]
Regular schedule	1	[36]
Self-acceptance	1		[93]
Self-compassion	1	[36]

+ = positive relationship, = = no relationship, − = negative relationship. Bold indicates older adults only in the article.

). Within these articles, three outcomes have a positive relationship with pet attachment, two outcomes have no relationship, and one outcome has a negative relationship. Among the articles that focus on adults, including older adults, most of the articles demonstrate a positive relationship between pet attachment and additional positive health outcomes. Eight of the outcomes showed only positive relationships, three outcomes included only no relationships (purpose/meaning in life, autonomy, coping with uncertainty, good mood, personal growth, regular schedule, self-acceptance, and self-compassion), one outcome (life satisfaction) had both positive and no relationships, and one outcome (emotional mastery) had a negative relationship with pet attachment.

Based on the evidence provided in the articles about additional positive health outcomes as summarized in Table 18, there is no negative relationship between pet attachment and these positive health outcomes in older adults. There are only a few articles with only older adults to distinguish whether patterns of relationships are different in older adults compared to adults in general.

3.1.9. Additional Negative Psychological Outcomes

Twelve articles related to pet attachment to negative psychological health outcomes are assessed in four or fewer articles (Table 4). Thirteen different negative psychological health outcomes were assessed in these articles (

Table 19. Articles that relate pet attachment to negative psychological outcomes (n = 12).

Article	Design	Group	Area	Age	OCEBM	Sample Size	Attachment Scale	Relationship	Other Negative Psychological Measures
[90] Allen, C. J., 2022	C	DO	AU	A	5	639	MDOR S subscale	= negative affect	SPANE
[120] Barnard-Nguyen, S., 2016	C, R	BPO	USA	A	2C	409	LAPS	+ anger + guilt	PBQ
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	= emotional instability	MHI
[67] Ellis, A., 2024	C	PO	UK	A	5	1359	LAPS	= Negative affect	PANAS
[69] Fritz, C. L., 1996	C	CG	USA	OA, W A, M	3b	38 27	LAPS	= Caregiver burden	ACBI
[36] Kogan, L. R., 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	= Overwhelmed	QU
[30] Peacock J, 2012	C	PO	AU	A	5	150	PAS-H, OPR	+ Somatoform symptoms	BSI
[27] Reevy, G. M., 2020	C	CO	USA	A	3b	1239	LAPS, PAQ	+ Anger + Vulnerability = Immoderation = Self-consciousness = Anger/avoidant attachment = Vulnerability/avoidant attachment = Immoderation/avoidant attachment = Self-consciousness/avoidant attachment = Anger/anxious attachment = Vulnerability/anxious attachment = Immoderation/anxious attachment = Self-consciousness/anxious attachment	IPIP
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	− Psychopathology − Secure attachment with psychopathology	WHOQOL-BREF, BSI
[23] Turner, D.C. 2003	C	CO	SUI	A ††	4	516	LAPS	= Bad mood No M living with partners to bad mood	EWL
[61] Wan, 2022	L	E	USA	A	5	187	CCA	= Emotional exhaustion, M job support to strain reactions, stress to emotional exhaustion, ME	EES
[73] Watson, N. L. 1993	C	EW	USA	A	3b	42	CAAS	= Anger	STAEI

†† aged 45 to 53 years. Design: C = cross-sectional; L = longitudinal; R = representative sample. Group: BPO = bereaved pet owners; CG = caregivers of Alzheimer’s patients; CO = cat owners; DO = dog owners; E = employees; EW = employed women; M = men; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CAN = Canada; NZ = New Zealand; SUI = Switzerland; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS= Companion-Animal Attachment Scale [43]; CCA = Comfort from Companion Animal Scale [38]; I-1 = Single Item (1–10) [14,36,37]; LAPS = Lexington Attachment to Pets Scale [39]; MDORS = Monash Dog Owner Relationship Scale [47]; OPR = Owner–Pet Relationship Questionnaire [49]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative; ME = mediated. Additional Negative Psychological Outcome Measures: ACBI = Alzheimer’s Caregiver Burden Interview [140]; EES = Emotional Exhaustion scale [141]; EWL = List of Adjectives [137]; IPIP = International Personality Item Profile [82]; MHI = Mental Health Inventory [83]; PANAS = Positive and Negative Affect Schedule [128]; PBQ = 16-item Pet Bereavement Questionnaire [125](PBQ); SPANE = Scale of Positive and Negative Experience [108]; STAEI = State-Trait Anger Expression Inventory [142]; QU = questions developed for this study.

). The 12 articles include 2 with an OCEBM rating of 2c [36,120], 4 with a rating of 3b [27,28,69,73], 1 with a rating of 4 [23], and 5 with a rating of 5 [30,61,66,67,90]. None of the articles were limited to older adults; one set of analyses includes older adult women and adult men [69].

Pet attachment was assessed with the LAPS in five of the 12 articles that addressed the relationship of pet attachment to negative psychological outcomes. One article included the PAQ in addition to the LAPS. One article included the OPR and the PAQ-H, and another included the OPR and the PAQ. The PAQ-H was also used alone in one article. Four different scales were used to assess pet attachment in the four other articles. Negative psychological outcomes were assessed with different scales in each article.

The majority of the articles describe no relationship between pet attachment and negative psychological health outcomes (Table 19 and

Table 20. Summary of findings from analyses in articles examining the relationship between pet attachment and negative psychological health outcomes.

Psychological Health Outcome	N	+	=	−
Anger	3	[27]; [120]	[27]; [73]; [27]
Negative affect	2		[67,90]
Bad mood	1		[23]
Caregiver burden	1		[69]
Emotional exhaustion	1		[61]
Emotional instability	1		[66]
Guilt	1	[120]
Immoderation	1		[27]; [27]; [27]
Overwhelmed	1		[36]
Psychopathology	1			[28]; [28]
Self-consciousness	1		[27]; [27]; [27]
Somatoform symptoms	1	[30]
Vulnerability	1	[27]	[27]; [27]

Numbers in the table indicate the article in the reference list. Each number in [brackets] represents one analysis. + = positive relationship, = = no relationship, − = negative relationship. Bold = Older adults only. Note: Articles can examine multiple outcomes within each category, multiple listings of an article indicates examination of multiple outcome variables within each type of negative health outcome.

). Three topics were addressed in two or more articles. Three articles addressed the relationship between anger and pet attachment [27,73,120]. Anger was related to overall cat attachment in an online sample (n = 1239) of USA cat owners [27] but not in a USA sample of women employed by the American Medical Association (n = 42) who keep pets [73]. Anger also was not related to avoidant or anxious attachment in the USA cat owners [27]. Anger and guilt were related to pet attachment in pet euthanasia cases (n = 409) [120].

Two articles showed no relationship between pet attachment and negative affect. One of these articles included 1359 pet owners from the UK [67], and the other included 639 dog owners from Australia [90].

One article included an analysis of the moderating and mediating roles of pet attachment in the relationship between job insecurity during the COVID pandemic and emotional exhaustion through stress [61]. Pet attachment support moderated the relationship of stress with emotional strain reactions in adult USA pet owners (n = 187). In this article, pet attachment support also buffered the conditional indirect effects of job insecurity through stress to emotional strain reactions.

In one article, pet attachment was related to a negative psychological outcome in adults. In a mail survey of 150 Australian adult pet owners [30], there was a positive association between pet attachment and somatoform symptoms, physical symptoms not completely explained by a medical condition that cause significant distress or impairment in function. These symptoms are particularly distressing due to the extremity of individuals’ thoughts and focus on them.

One article [28] addressed the relationship between pet attachment and the broader construct of psychopathology. In 322 Australian adult pet owners, pet attachment was not related to psychopathology, and secure pet attachment also was not related to this outcome.

Only one of the analyses addressing less frequently assessed negative psychological health outcomes focused exclusively on older adults, and it revealed no relationship between pet attachment and caregiver burden. Among the 24 analyses of the relationship between pet attachment and other negative psychological outcomes in adults, 5 showed that higher pet attachment was related to greater negative psychological outcomes, 17 showed that pet attachment was not related to negative psychological outcomes, and 2 showed that higher pet attachment was related to lower negative psychological outcomes. Based on the evidence provided in the articles about these negative health outcomes (Table 20), there is no consistent relationship between pet attachment and negative psychological health outcomes in adults.

3.2. Topic 2: The Relationship Between Pet Attachment and Social Health Outcomes

Twenty-seven articles included examination of the relationship between pet attachment and social health outcomes. Articles that examined the relationship between pet attachment and social variables explored one to four social health outcomes (

Table 21. Number of social health topics addressed in relation to pet attachment per article (n = 27).

Number of Topics	Number of Articles	%
1	21	77.8%
2	4	14.8%
3	2	7.4%

). Most of the articles (21/27) that explored the relationship between social variables and pet attachment examined only one social outcome. Four articles included two social health outcomes. Two articles included three social health outcomes: a qualitative study included social support, social isolation, and loneliness [33], and a quantitative study included social isolation, loneliness, and satisfaction with social support [37].

The most frequent of the social health outcomes from the 27 articles (Table 4) that related pet attachment to aspects of social health was loneliness (12 articles). Social support/companionship was examined in seven articles, and social wellbeing, function, or quality of life was examined in relation to pet attachment in seven articles. Social isolation was related to pet attachment in two articles each, and seven additional social health outcomes were addressed in one article each.

3.2.1. Loneliness

Twelve articles examined the relationship between pet attachment and their owners’ loneliness (Table 4). One of the methods was qualitative [33], and the remainder [10,11,32,37,57,64,65,67,71,74,90] were quantitative (

Table 22. Articles that relate pet attachment to loneliness (n = 12).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Loneliness
[90] Allen, C. J., 2022	C	DO	AU	A	5	639	MDORS subscale	+ family loneliness + romantic loneliness. = social loneliness	SELSA-S
[64] Antonacopoulos, N. M. D., 2010	C	LA	CAN	A	5	66	LAPS	=, M social support to	UCLA-3
[65] Branson, S. M., 2019	B, C	CO	USA	OA	5	41	LAPS	=	UCLA-R
[33] Chur-Hansen, A.,2009	Q	PO	AU	OA		11	INT	+	INT
[67] Ellis, A., 2024	C	PO	UK	A	5	1359	LAPS	=	UCLA-R
[37] Fernandes, J., 2024	C	PO	POR	OA	4	130	I-1	=	I-L
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	=	ISAI
[10] Krause-Parello, C. A. 2008	C	W, LA	USA	OA	5	159	PAS-A	+	UCLA-R
[11] Krause-Parello, C. A., 2013	C	W	USA	OA	5	191	PAS-A	+	UCLA-R
[71] Krause-Parello, C.A. 2012	C	W	USA	OA	5	159	PAS-A	+	UCLA-R
[57] Kulari, G., 2025	C	E	POR	A	4	279	LAPS	+, ME	UCLA-3
[74] Wells, D. L., 2022	C	UKO	UK	A	3b	146	LAPS	+	UCLA-3

Design: B = biomarkers; C = cross-sectional; Q = qualitative. Group: CO = cat owners; DO = dog owners; E = employees; LA = lives alone; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; UKO = UK residents; W = women. Area: AU = Australia; CAN = Canada; POR = Portugal; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: I-1 = Single Item (1–10) [14,36,37]; INT = interview; LAPS = Lexington Attachment to Pets Scale [39]; MDORS = Monash Dog Owner Relationship Scale [47]; PAI = Pet Attitude Inventory [50]; PAS-A = Pet Attachment Scale-Albert [42]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; M = moderator; ME = mediated. Loneliness Measures: I-L = Single loneliness item [37]; I-1 = Single Item (1–10) [14,36,37]; INT = interview; ISAI = Iowa Self-Assessment Inventory [99]; SELSA-S = Short version of the Social and Emotional Loneliness Scale for Adults [143]; UCLA-3 = UCLA Loneliness Scale V3 [144]; UCLA-LS = UCLA Loneliness Scale [145].

). One article was OCEBM category 3b, two were 4, and eight were 5. All were cross-sectional and used convenience samples.

Pet attachment was assessed with six tools: the LAPS (5 articles), the PAS-A (3 articles), the MDOR S (1 article), a single item (1 article), and the combination of the PRS and PAI (1 article).

Most of the quantitative articles (8/11) used a form of the UCLA loneliness scale to assess loneliness. The UCLA Revised Loneliness Scale was used in five articles, and the UCLA-3 Loneliness Scale was used in three. A single item was used in one article, the short version of the Social and Emotional Loneliness Scale for Adults was used in one article, and the ISAI was used in one article.

Six of the quantitative articles [10,11,32,37,65,71] and the qualitative article [33] focused on older adults only; the remaining five included, but were not limited to, older adults. Pet attachment was related to loneliness in two articles [11,71] focused on older women in the USA (n = 159, 191) in the community, and one [10] focused on similar older women (n = 151) who lived alone. Three additional articles found no relationship between pet attachment and loneliness in USA cat owners (n = 41) [65], USA Latino pet owners [32], and Portuguese pet owners (n = 130) [37].

When considering the five articles that included older adults within a broader sample of adults, one article assessed the relationship between pet attachment and three different aspects of loneliness [90]. In that sample of Australian dog owners (n = 639), dog attachment was positively associated with family loneliness and romantic loneliness, and not with social loneliness. An adult sample of 146 UK residents [74] and an adult sample of 279 Portuguese health care employees [57] who kept pets also demonstrated a positive relationship between pet attachment and loneliness. In contrast, in a large UK sample (n =1359), pet attachment was not related to loneliness. Among adults who lived alone in CAN (n = 66), pet attachment was also not related to loneliness [64]. However, the relationship between pet attachment and loneliness depended on human social support; among pet owners with low levels of human social support, high attachment to pets was related to higher loneliness. In the article focusing on Portuguese adult pet owners employed in health care (n = 279), pet attachment completely mediated the relationship between loneliness and depression [57].

Based on the 12 articles that addressed the relationship between pet attachment and loneliness, pet attachment is not negatively related to lower loneliness (

Table 23. Summary of the direction of the relationship between pet attachment and loneliness in analyses from the quantitative articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	4	3	7
=	3	3	6
−	0	0	0
Total	7	6	13

+ = positive relationship, = = no relationship, − = negative relationship. Note: some articles contained multiple outcomes thus the total analyses are greater than the number of articles.

) The distribution of the relationships is similar for articles that include older adults and articles that are limited to older adults. Higher pet attachment is related to higher loneliness in 57% (4/7) of the analyses among adults and 50% (3/6) of the analyses among older adults. Higher pet attachment is not related to lower loneliness in any of the analyses. However, pet attachment may reduce the negative health consequences of loneliness. The data suggest that older adults who are lonely, especially those who live alone, may become more attached to their pets. Longitudinal examinations are required to confirm this temporal relationship.

3.2.2. Social Support or Companionship

Seven articles addressed the relationship between pet attachment and social support. (Table 4). The articles included six [11,17,21,32,49,57] that employed quantitative and one [33] that employed qualitative methods (

Table 24. Articles that relate pet attachment to social support (n = 7).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Social Support
[33] Chur-Hansen, A.,2009	Q	PO	AU	OA		11	INT	=	INT
[17] Hill, L., 2020	C	PO	AU	A	3b	392	LAPS	=, ∩	MSPSS
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	=	ISAI
[21] Joseph, N., C., 2019	C	PO	IND	A †	3b	122	LAPS	+	MOSSS
[11] Krause-Parello, C. A., 2013	C	W	USA	OA	5	191	PAS-A	+	UCLA-R
[57] Kulari, G., 2025	C	E	POR	A	4	279	LAPS	+	UCLA-3
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	=	MDSS

† Ages 12 + years. Design: C = cross-sectional; Q = qualitative. Group: E = employees; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; W = women. Area: AU = Australia; IND = India; POR = Portugal; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: INT = interview; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner-Pet Relationship Questionnaire [49]; PAI = Pet Attitude Inventory [50]; PAS-A = Pet Attachment Scale-Albert [42]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive; = = no relationship; ∩ = inverted U relationship. Social Support Measures: INT = interview; ISAI = Iowa Self-Assessment Inventory [99]; MDSS = Multi-Dimensional Support Scale [146]; MOSSS = Medical Outcomes Study Social Support Survey [147]; MSPS = Multi-Dimensional Scale of Perceived Social Support [148]; UCLA-3 = UCLA Loneliness Scale V3 [144]; UCLA-R = UCLA Loneliness Scale-Revised [145].

).

All of the quantitative articles are cross-sectional surveys. Two of them are OCEBM category 3b, one is category 4, and three are category 5. Pet attachment was assessed with the LAPS in three of the articles and the PAS-A, OPR, and PRS + PAI in one article each. Two articles included an assessment of social support with different versions of the UCLA Loneliness Scale, and every other assessment of social support was by a different tool.

Four of the articles, including the qualitative one, were conducted in older adults, and three were conducted in all adults, including older adults. In the qualitative article [33], the older adult pet owners (n = 11) were highly or extremely attached to their pets, and all participants considered pet companionship to be very important. Among the other articles focused only on older adults, one of 191 community USA older adults who owned pets [11] showed a positive relationship between pet attachment and social support, while two articles focused on 24 USA older Latino pet owners [32] and 179 Australian older adult pet owners [49] found that there was no association of pet attachment with social support.

Results reported in articles that included, but were not limited to, older adults also reported inconsistent results. Two articles, including 191 pet owners employed in health care in Portugal [57] and 122 pet owners in India [21], reported a positive relationship between pet attachment and social support. In the former article, pet attachment was specifically related to support from family, friends, and significant others. The latter article specified that the relationship was with the emotional component of social support. In contrast, an article focused on 392 Australian pet owners [17] found no relationship between pet attachment and social support. The authors suggested an inverted U-shaped relationship between pet attachment and social support, where very high and very low attachment were associated with low human social support.

The series of seven articles does not provide systematic evidence for a relationship between pet attachment and social support or companionship (

Table 25. Summary of the direction of the relationship between pet attachment and social support in the analyses in articles that include older adults and are limited to older adults.

Relationship	Adults	Older Adults	Combined
+	2	1	3
=	1	3	4
−	0	0	0
Total	3	4	7

+ = positive relationship, = = no relationship, − = negative relationship.

). Sixty-seven percent (2/3) of the analyses of the relationship between pet attachment and social support demonstrated a positive relationship among adults, and 25% (1/4) of the analyses demonstrated this relationship among older adults. No analyses demonstrated negative relationships between pet attachment and social support or companionship in either age group. The data clearly suggests that there is no negative relationship between the two variables. The suggestion of an inverted U-shape distribution for the relationship between pet attachment and health outcomes [17,33] merits evaluation. Such a relationship might explain the lack of correlations in some cases, as well as the presence of positive or negative relationships in other situations. Given the results presented above, this idea merits evaluation with regard to this and the other health outcomes.

3.2.3. Social Wellbeing, Social Function, or Social Quality of Life

Seven articles address social wellbeing, social function, or social quality of life (Table 4 and

Table 26. Articles that relate pet attachment to social wellbeing, social function, or social quality of life (n = 7).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Social Wellbeing
[91] Demeter, N., 2024	C	STR	ISR	OA	5	25	PAQ	=	WHOQOL-BREF
[70] Ingram, KM 2019	C	CA, DO	USA	A	5	122	LAPS	=	FACA-27
[24] Schwarzmueller-Erber, G., 2020	C	HR, DO	AU	A †	5	124	PALS	+ during activity with an animal = after activity with an animal	FAHW
[29] Tan, J. S. Q., 2021	C	PO	SGP	A	3b	431	PAS-H, PAQ	=	RAND
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	+ − insecure attachment	WHOQOL-BREF
[94] White N, 2017	C	ADO	UK	A	5	73	PAQ	+ anxious attachment = insecure attachment = avoidant attachment	QoLS
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	=	SF-36

† Age range 45 to 82 years. Design: C = cross-sectional. Group: ADO = assistance dog owners; CA = cancer patients; DO = dog owners; HR = recreational horseback riders; PO = community-residing pet owners or recent pet owners; STR = stroke survivors. Area: AU = Australia; ISR = Israel; SGP = Singapore; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PALS = Pet Attachment and Life Impact Scale [51]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]. Highlighted findings use scales with lower attachment indicated by a higher score and reported in that manner in the article. The directions of associations are retained in this table but discussed in the context of more attachment having a higher score in the summary tables. Relationship of pet attachment to health outcome: + = positive; = = no relationship; − = negative. Social Wellbeing/Functioning/Quality of Life Measures: FACA-27 = 27-item Functional Assessment of Cancer Therapy-General scale Version 4 [96]; FAHW = German Questionnaire for Habitual Wellbeing [98]; QoLS = Quality of Life Scale [103] RAND = RAND 36-item Health Survey [104]; SF-36 = Medical Outcome Study Short Form 36 Health Survey [107]; WHOQOL BREF = World Health Organization [109].

). These related concepts will be termed social wellbeing for simplicity in this analysis. Two of the articles were OCEBM category 3b [28,29] and five were category 5 [24,49,70,91,94]. All of the articles were cross-sectional. The populations from which the participants in these articles resided varied considerably. The seven articles originated in six countries: two were from Australia, and the others were from Austria, Israel, Singapore, the United Kingdom, and the USA.

The articles used five different scales to assess pet attachment, and three articles used a combination of two scales. The articles also used six different scales to assess social wellbeing. Six of the articles included separate assessments of social aspects of wellbeing or quality of life. One article [94] used a composite assessment that combined psychological, social, and physical aspects of quality of life into one score.

Most of the articles that examined the relationship between pet attachment and social wellbeing were conducted in adults. Five articles [28,29,70,94] focus on the relationship of pet attachment to social wellbeing on adults. Among 25 older adult stroke survivors from Israel [91] and 179 older adult pet owners from Australia [49], pet attachment was not related to social wellbeing. Among the 124 Austrian 45–82-year-old recreational horseback riders and dog owners, the relationship differed depending on the timing of the assessment [24]. Attachment to the respective animals was related to social wellbeing while they were participating in the activities with their animals, but not after they participated. There was no relationship between pet attachment and social wellbeing in USA dog owners (n = 122) who had recently been diagnosed with cancer [70]. Similarly, there was no relationship between pet attachment and social wellbeing in a sample of 431 adult pet owners who resided in Singapore [29].

Two studies of adults [28,94] examined the relationship of overall pet attachment and the relationship of some specific aspects between pet attachment and social wellbeing. Overall pet attachment was negatively related to social wellbeing among 322 Australian pet owners, but insecure pet attachment was not [28]. Among 73 adult assistance dog owners [94], pet attachment was associated with social wellbeing but insecure and avoidant pet attachment were not.

Two of the articles [49,91] focused on the relationship between pet attachment and social wellbeing in older adults. There is no evidence that overall pet attachment is related to social wellbeing in community-residing older adults (

Table 27. Summary of the direction of the analyses of the relationship between pet attachment and social wellbeing in the articles including older adults and limited to older adults.

Relationship	Adults	Older Adults	Combined
+	3	0	3
=	5	2	7
−	1	0	1
Total	9	2	11

+ = positive relationship, = = no relationship, − = negative relationship. Note: Some articles contained multiple outcomes.

).

In the majority of analyses of articles addressing adults (56%, 5/9) and older adults (100%, 2/2), there was no relationship between pet attachment and social wellbeing. The results together suggest that the relationship between pet attachment and social wellbeing depends on the situation and the type of attachment.

3.2.4. Relationship Between Pet Attachment and Other Social Outcomes

The eight social outcome topics listed in Table 4 that were addressed in fewer than five articles (dogs as family members, human social network size, influence of pets on owners’ daily lives, positive relations with others, satisfaction with social support, seclusion, social interactions, and social isolation) were addressed in eight articles (

Table 28. Articles that relate pet attachment to other social outcomes (n = 8).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Other Social Outcomes
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA	Q	11	INT	− Social interactions	INT
[37] Fernandes, J., 2024	C	PO	POR	OA	4	130	I-1	= satisfaction with social support = social isolation	ESSS SIS
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	+ influence of pets on owners’ daily lives	PIDL
[36] Kogan, L. R., 2021	L	PO	USA, CAN, UK, AU, Other	A	2c	5061	I-1	− Social isolation	QU
[35] Mueller, R. 2024	Q	M, DO	USA	OA	Q	12	INT	+ dogs as family	INT
[22] Stallones, L., 1990	C, R	PO	USA	A †	5	152	CAAS	− human social network size	SNI
[23] Turner, D.C., 2003	C	CO	SUI	A ††	4	516	LAPS	= seclusion	EWL
[93] Watt, D., 2007	C	PO	AU	OA	5	32	PAS-M	= positive relations with others	PWS

† Ages 45–64 years; †† ages 45 to 53 years. Design: C = cross-sectional; L = longitudinal; Q = qualitative; R = representative sample. Group: CO = cat owners; DO = dog owners; M = men; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CAN = Canada; SUI = Switzerland; POR = Portugal; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS = Companion Animal Attachment Scale [43]; I-1 = Single Item (1–10) [14,36,37]; INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; PAS-M = Pet Attachment Scale-Meehan [52]. Relationship of pet attachment to health outcome: + = positive, = = no relationship, − = negative. Other Social Outcome Measures: BSNI = Social Network Index [149] (modified); ESSS Satisfaction with Social Support Scale, Portuguese version [150,151]; FACA-27 = 27-item Functional Assessment of Cancer Therapy-General scale Version 4 [96]; FAHW = German Questionnaire for Habitual Wellbeing [98]; PIDL = Pet Influence on Owner’s Daily Lives [68]; PWS = Psychological Wellbeing Scale [102]; QU = Questions developed for this study; SIS = social isolation scale [152,153].

). One of the outcomes, social isolation, was addressed in two articles each. The remaining seven topics were each addressed in one article.

Two of the articles were qualitative [33,35], and the six remaining articles [22,23,36,37,68,93] were quantitative. One of the articles was OCEBM classification 2c, one was 3b, two were 4, and two were 5. One of the quantitative articles was longitudinal [36] and the others were cross-sectional [22,23,37,68,93].

Both qualitative articles and two quantitative articles focused on older adults only.

In the qualitative articles, high attachment was associated with older men who owned dogs (n = 12), considering their dogs as family members [35]. High pet attachment was mentioned as taking priority over social interactions with people [33], and with the older adults preferring to remain at home with their pets rather than visiting relatives [33]. These findings are consistent with the cross-sectional survey that found that the more attached older adults (n = 90) are to their pets, the more they played a role in decisions about activities in their daily lives [68]. The other quantitative articles found no relationship between pet attachment and satisfaction with social support or to social isolation among Portuguese older adult (n = 130) pet owners [37]. Pet attachment also was not associated with positive relations with others among (n = 32) Australian older adult pet owners [93].

The remaining three articles included, but were not limited to, older adults. The only longitudinal article focused on changes that adults (n = 5061) experienced during the early stages of the COVID-19 pandemic. Pet attachment was related to reduced feelings of social isolation from before to after the first few months of the lockdown [36]. In a cross-sectional survey of Swiss cat owners (n = 516), cat attachment was not related to feelings of seclusion [23], and in a representative survey of USA adults, pet attachment was not related to social network size [22].

These articles taken together (

Table 29. Summary of the direction of the relationship between pet attachment and other social outcomes in analyses in the articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults
Negative Social Outcomes
+
=	Seclusion [23] ††	Social Isolation [37]
−	Social isolation [36]
Positive Social Outcomes
+		Dogs as family [35] Influence of pets on owners’ lives [68]
=		Satisfaction with social support [37] Positive relations with others [93]
−	Human social network size [22] †	Social interactions [33]

† Ages 45–64 years, †† ages 45 to 53 years. + = positive relationship, = = no relationship, − = negative relationship. Note: Some articles contained multiple outcomes.

) suggest a complex relationship between pet attachment and social outcomes in adults. When the valences of the negative outcomes are converted to positive and the directions of the relationships are reversed, among all adults, in one analysis (33%; 1/3), pet attachment is positively related to positive social outcomes; in one analysis (33%; 1/3), pet attachment is not related to a positive social outcome; and in one analysis (33%; 1/3), pet attachment is negatively related to a positive social outcome. In contrast, among older adults, two analyses (33%; 2/6) reveal that pet attachment is related positively to positive social outcomes, three analyses (50%; 3/6) reveal no relationship, and one analysis (17%; 1/6) reveals a negative relationship between pet attachment and a positive social outcome. This suggests that there may be different patterns in the relationship between pet attachment and social outcomes in older adults than in younger ones. 3.

3.3. Topic 3: The Relationship Between Pet Attachment and Biological or Physical Health

The 27 articles that examined the relationship between pet attachment and physical health outcome variables explored one to four different outcomes (

Table 30. Number of physical health topics addressed in relation to pet attachment per article (n = 27).

Number of Topics	Number of Articles	% of Articles
1	13	48.1%
2	6	22.2%
3	3	11.1%
4	5	14.8%

). A majority of the articles (14/27) included more than one topic. One biological or physical health variable was included in 13 articles, two were included in 6 articles, three were included in 3 articles, and four were included in 5 articles.

The most frequent of the physical health outcomes from the 27 articles related to pet attachment to physical aspects of health were as follows: physical wellbeing, health, or quality of life (13 articles) and physical activity (7 articles). Many other outcomes were present in three or fewer articles (Table 4). The other articles were grouped into three thematic categories for examination: disease or health indicators, physical function, and health or health-related behaviors.

3.3.1. Physical Wellbeing, Health, or Quality of Life (N = 13)

Thirteen articles focused on the relationship between pet attachment and physical wellbeing (

Table 31. Articles that relate to pet attachment to physical wellbeing, health, or quality of life (n = 13).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Physical Wellbeing
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA	N/A	11	INT	=	INT
[91] Demeter, N., 2024	C	STR	ISR	OA	5	25	PAQ	=	WHOQOL-BREF
[129] Friedmann, E., 2025	L	PO	USA	OA	2c	214	LAPS	=, =↓ CO −↓	SF-12 PCS
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	=	SF-12 PCS
[70] Ingram, KM 2019	C	CAN	USA	A	5	122	LAPS	=	FACA-27
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	=	ISAI
[56] Koodaly, L., 2025	C	PO	USA	OA	4	178	LAPS	=, M Psychological wellbeing to	SF-12 PCS
[10] Krause-Parello, C. A., 2008	C	W, LA	USA	OA	5	159	PAS-A	−, ME loneliness to physical health	PGWB
[24] Schwarzmueller-Erber, G., 2020	C	HR, DO	AU	A †	5	124	PALS	=	FAHW
[29] Tan, J. S. Q., 2021	C	PO	SGP	A	3b	431	PAS-H, PAQ	+	RAND
[28] Teo, J.T., 2019	C	PO	AU	A	3b	322	PAQ, OPR	+	WHOQOL-BREF, BSI
[94] White N, Mills D, Hall S., 2017	C	ADO	UK	A	5	73	PAQ	+ Anxious attachment = Insecure attachment = Avoidant attachment	QoLS
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	=	SF-36

† Ages 45 to 64. Design: C = cross-sectional; L = longitudinal; Q = qualitative. Group: ADO = assistance dog owners; CA = cancer patients; DO = dog owners; HR = recreational horseback riders; LA = lives alone; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; STR = stroke survivors; W = women. Area: AU = Australia; ISR = Israel; SGP = Singapore; UK = United Kingdom; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAI = Pet Attitude Inventory [50]; PALS = Pet Attachment and Life Impact Scale [51]; PAQ = Pet Attachment Questionnaire [40]; PAS-A = Pet Attachment Scale-Albert [42]; PAS-H = Pet Attachment Survey-Holcomb [41]; PRS = Pet Relationship Scale [54]. Highlighted findings use scales with lower attachment indicated by a higher score and reported in that manner in the article. The directions of associations are retained in this table but discussed in the context of more attachment having a higher score in the summary tables. Relationship of pet attachment to health outcome: + = positive, = = no relationship, − = negative, M = moderator, ME = mediated, −↓ = slower deterioration, =↓ = not related to deterioration. Physical Wellbeing/Functioning/Quality of Life Measures: BSI = Brief Symptom Inventory [95]; FACA-27 = 27-item Functional Assessment of Cancer Therapy-General scale Version 4 [96]; FAHW = German Questionnaire for Habitual Wellbeing [98]; PGWB = Psychological General Wellbeing Index [84]; QoLS = Quality of Life Scale [103]; RAND = RAND 36-item Health Survey [104]; SF-12 PCS = SF-12 Physical Component Score [105,106]; SF-36 = Medical Outcome Study Short Form 36 Health Survey [107]; WHOQOL BREF = World Health Organization [109].

), individuals’ perceptions of their physical health. All except one [33] of the articles [10,24,28,29,32,49,56,68,70,91,94,129] used quantitative research methods. Seven of the articles were OEBCM rating group 5, three were group 3b, one was group 2c, and one was group 4.

The most common tools used to assess pet attachment were the LAPS, used in four articles, and the PAQ, used in four articles, alone in two articles and in combination with the PAQ-H and the OPR in one article each. The OPR was also used alone in two articles; the PALS, PAS-A, and a combination of the PAS and the PAI were used in one article each.

Eight of the articles included older adults only. The remaining four articles included, but were not limited to, older adults; one of them [24] included only adults aged 45 to 64 years. All articles included used convenience samples.

A variety of tools were used to assess physical wellbeing. Six of the quantitative articles used three inter-related scales: the RAND-36 (1 article), the Medical Outcomes Study Short Form 36 (SF-36; 1 article), and the Short Form-12 (SF-12; 4 articles). RAND-36 is an adaptation of the SF-36 intended for public use, and the SF-12 is a shortened form of the SF-36. The WHO QOL-BREF was used in two articles, and the FACA-27, FAHW, PGWB, and QoLS were each used in one article.

Thirteen articles used a cross-sectional time frame to evaluate the relationship between pet attachment and physical wellbeing; additionally, one [129] evaluated the relationship between pet attachment and changes in physical wellbeing.

Nine articles focused exclusively on older adults. The qualitative article [33] found no direct relationship between pet attachment and physical wellbeing. The authors suggested an inverted U relationship. In the seven other articles focusing exclusively on older adults, only two showed a direct relationship between pet attachment and physical wellbeing. One of the articles demonstrated a systematic relationship between pet attachment and physical wellbeing among 159 community-dwelling older adult women; higher pet attachment was related to poorer physical wellbeing [10]. Pet attachment was associated with poorer general physical health among older adult USA female pet owners who live alone (n = 159) [10]. In that article, pet attachment also mediated the relationship between loneliness and poor general health, alleviating the negative impact of loneliness on general health. Among Israeli (n = 25) older adult stroke survivors [91], USA (n = 90) healthy community-residing older adults (n = 90) [129], and Australian (n = 179) older adult pet owners [49], pet attachment was not related to physical wellbeing. Among community-residing older adult pet owners (n = 214), greater pet attachment was related to slower deterioration in physical wellbeing, with a mean of 7.5 years of aging [68], although it was not related to initial physical wellbeing. In a smaller group of these pet owners (n = 178), pet attachment moderated the relationship between mental wellbeing and physical wellbeing [56] despite the lack of a relationship between pet attachment and physical wellbeing. The higher the attachment, the lower the relationship between mental and physical wellbeing. For older adults with low mental wellbeing, the higher the attachment, the greater the physical wellbeing. For individuals with higher mental wellbeing, pet attachment was not related to physical wellbeing.

Among the articles that included, but were not limited to, older adults, pet attachment, and physical wellbeing were not related to each other in Austrian 45- to 64-year-old (N = 124) dog owners or individuals who participated in recreational horseback riding [24]. Dog attachment also was not related to physical wellbeing in USA patients (n = 122) who were recently diagnosed with cancer and owned dogs. Insecure pet attachment and avoidant pet attachment also were not related to physical wellbeing in 73 adult UK assistance dog owners [94].

Among adult pet owners, pet attachment was positively associated with physical wellbeing in two articles [28,29] and one aspect of pet attachment was related to physical wellbeing in another [94]. In 431 adult pet owners in SGP [29] and 322 adult pet owners in Australia [28], pet attachment was associated with better physical wellbeing independent of age, gender, and core conditions. In the latter article, specific aspects of wellbeing, such as physical health, were examined in UK assistance dog owners (n = 73). Anxious dog attachment was related to physical wellbeing [94], but insecure and avoidant attachment were not.

To summarize (

Table 32. Summary of the direction of the direct relationship between pet attachment and physical wellbeing or quality of life in analyses in the articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	2	0	2
=	5	5	10
−	1	1	2
Total	8	6	14

+ = positive relationship, = = no relationship, − = negative relationship. Note: Some articles contained multiple outcomes.

), pet attachment was not related to physical wellbeing in 71% (10/14) of the analyses of this relationship. Pet attachment was related to greater physical wellbeing in 14% (2/14) of the analyses and to lower physical wellbeing in 14% (2/14) of the analyses. Pet attachment was not directly related to physical wellbeing in any of the analyses limited to older adult participants. This suggests that pet attachment is not related to physical wellbeing among older adults. The articles that include, but are not limited to, older adults were not consistent in the pattern of the relationship between pet attachment and physical wellbeing. This suggests that pet attachment may be related to physical wellbeing in younger individuals, at least under some circumstances. While there was not a direct relationship in older adults, pet attachment moderated the relationship between mental wellbeing and physical wellbeing [56], and pet attachment was related to changes in physical wellbeing with aging [129] in older adults. These findings suggest a more complex role of pet attachment as a contributor to physical wellbeing.

3.3.2. Physical Activity

Seven articles evaluated the relationship between pet attachment and pet owner physical activity levels (

Table 33. Articles that relate pet attachment to physical activity (n = 7).

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measures of Physical Activity
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA	Q	11	INT	=	INT
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	=	CE
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	− Times exercise per week	ISAI
[154] Lu, J., 2023	C, R	PO	CHN	OA	4	879	LAPS	+, M Loneliness to	HPLP-CE
[29] Tan, J. S. Q., 2021	C	PO	SGP	A	3b	431	PAS-H, PAQ	+	PA Survey
[155] Taniguchi, Y., 2024	C	DO	JAP	A	4	1041	PAS-A reduced	+	IPAQ-SF
[23] Turner, D.C. 2003	C	CO	SUI	A †	4	516	LAPS	=	EWL

† ages 45 to 53 years. Design: C = cross-sectional; Q = qualitative; R = representative sample. Group: CO = cat owners; DO = dog owners; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners. Area: AU = Australia; CHN = China; JAP = Japan; SGP = Singapore; SUI = Switzerland; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; PAI = Pet Attitude Inventory [50]; PAQ = Pet Attachment Questionnaire [40]; PAS-A = Pet Attachment Scale-Albert [42]; PAS-H = Pet Attachment Survey-Holcomb [41]; PRS = Pet Relationship Scale [54]. Relationship of pet attachment to health outcome: + = positive, = = no relationship, − = negative, M = moderator. Physical Activity Measures: CE = caloric expenditure [156]; EWL = List of Adjectives [137]; HPLP-CE = Health-Promoting Lifestyle Profile for Chinese Elderly Scale [157]; IPAQ-SF = International Physical Activity Questionnaire Short Form [158]; ISAI = Iowa Self-Assessment Inventory [99]; PA Survey = Physical Activity Survey (SGP Ministry of Health) [159].

). One article was qualitative and employed interviews with pet owners [33]. The other six articles were cross-sectional surveys of pet owners [23,29,32,68,154,155]. Five of the quantitative articles used convenience samples, while one [155] used a representative sample of Chinese pet owners.

The OCEBM ratings were category 3b for two of the articles [29,68], category 4 for three of the articles [23,154,155], and category 5 for one article [32].

Pet attachment was assessed with six different tools in the five quantitative articles. The LAPS was used in three articles [23,68,154], and the PAS-A was used in one article [156]. Two articles used two scales; the PRS and PAI were used together in one article [32], and the PAS-H and PAQ were used together in another article [29].

Physical activity was assessed differently in each article. All estimated activity was based on participant estimates of activity. One article [68] calculated total calories expended based on days per week and how many minutes per day participants complete particular activities. Other articles included the number of times participants exercised per week [32], how frequently participants participated in physical activity of different intensities [155], and questions about exercise included in lifestyle [29,154] or mood adjective [23] questionnaires. The qualitative article included interview questions that touched on the impact of the subject’s pet on their physical activity level [33].

Four of the articles exploring the relationship between pet attachment and physical activity included only older adults [32,33,68,154]. Three cross-sectional surveys that were limited to older adults had different findings with regard to the relationship between pet attachment and pet owners’ physical activity. Among USA community-residing generally healthy older adult pet owners (n = 90), pet attachment was not associated with calories consumed per day [68]. In a group of USA-residing Latino pet owners (n = 24), pet attachment was associated with exercising fewer times per week [32]. Pet attachment moderated the negative effect of loneliness on health-promoting lifestyle behaviors, including physical activity, in a representative sample (n = 879) of Chinese pet owners [154]. In the qualitative article that included interviews with older women (n = 11), the participants were very attached to their pets, and some reported that their pets, including dogs and cats, encouraged them to be more physically active [33].

Three articles focused on the relationship between pet attachment and physical activity in samples that included, but were not limited to, older adults [23,29,155]. Dog attachment was positively correlated with physical activity among Japanese dog owners (n = 1041), as well as with dog walking frequency [156]. Pet attachment was also associated with more frequent moderate- to high-intensity exercise among pet owners residing in SGP (n = 431) during the COVID-19 social isolation [29]. Cat attachment was not related to activity [23] in a survey of 516 Swiss cat owners aged 45 to 53 years.

Overall, there is conflicting evidence on the relationship between pet attachment and physical activity of older adult pet owners (

Table 34. Summary of the direction of the direct relationship between pet attachment and physical activity in analyses in the articles including older adults (Adults) and limited to older adults (Older Adults).

Relationship	Adults	Older Adults	Combined
+	2	0	2
=	1	3	4
−	0	1	1
Total	3	4	7

+ = positive relationship, = = no relationship, − = negative relationship. Note: Some articles contained multiple outcomes.

). The articles suggest that the relationship between pet attachment and physical activity might be less consistent in older adults than in younger populations. The results suggest that there may be an effect for younger, but not older, adult pet owners. There may be differences between the relationships for dog and cat owners, but these require further exploration. The apparent differences in the relationship between pet ownership and physical activity may also be due to cultural differences between the areas where the articles were conducted.

3.3.3. Other Physical Health Outcomes

Twenty additional physical health outcomes were evaluated in relation to pet attachment in fewer than five articles (Table 4). These topics, contained in 19 articles, are divided into the following three broad categories for further review: (1) health- or disease-related indicators; (2) physical function; and (3) health and health-related behaviors. More than half (14/19) of the 19 articles that address the relationship between pet attachment and other physical health outcomes address one of these topics. Four different topics were addressed in one article, three different topics were addressed in two articles, and two topics were addressed in two articles.

Physical Health- or Disease-Related Indicators (Six Articles)

Six articles addressed the relationship between pet attachment and physical health and disease-related indicators (

Table 35. Articles that relate pet attachment to other physical health outcomes according to topic groups.

Article	Design	Group	Area	Age	OCEBM	Pet Owners	Attachment Scale	Relationship	Measure of Physical Health
Part 1: Health- or disease-related indicators (n = 6)
[65] Branson, S. M., 2019	C, B	CO	USA	OA	5	41	LAPS	= SB =CRP	SB, CRP
[14] Branson, S., 2016	C, B	H, RU	USA	OA	3b	48	I-1	= CRP	CRP
[66] Budge, R. C., 1998	C	PO	NZ	A	5	176	PAS-H	+ # physical health symptoms	ILL
[160] Stephens, M.B., 2012	C	DO	USA	A	5	75	LAPS	= BMI	calculation
[29] Tan, J. S. Q., 2021	C	PO	SGP	A	3b	431	PAS-H, PAQ	+ energy	RAND
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	= Vitality = Pain	SF-36
Part 2: Physical function (n = 6)
[129] Friedmann, E., 2025	L	PO	USA	OA	2c	214	LAPS	−↓ physical performance, −↓ usual-gait speed, −↓ rapid-gait speed, =↓ cardio-respiratory fitness	HABCPPB UGS RGS 400 M
[68] Friedmann, E., 2020	C	PO	USA	OA	3b	90	LAPS	= rapid gait speed	RGS
[32] Johnson, R.A., 2002	C	LPO	USA	OA	5	24	PRS, PAI subset	= ADLs	ISAI
[72] Miller, M., 1990	C	W	USA	OA	3b	53	PRS	= physical limitations	OARS
[92] Raina, P., 1999	L	PO	CAN	OA	2c	286	LAPS	=↓ ADLs	ADLS
[49] Winefield, H. R., 2008	C	PO	AU	OA	5	179	OPR	= Physical role disability	SF-36
Part 3: Health and health-related behaviors (n = 8)
[33] Chur-Hansen, A., 2009	Q	PO	AU	OA		11	INT	= Dog walking = Life activities	INT
[130] Curl, A. L. 2021	C, R	DO	USA	OA	4	188	HRS	+ Dog walking	HRS
[91] Demeter, N., 2024	C	STR	ISR	OA	5	25	PAQ	= Real-life activities/anxious attachment − real-life activities/avoidant attachment	ACS
[62] Garrity, T. F.,1989	C, R	PO	USA	OA	3b	339	CAAS	= illness behavior	IB
[154] Lu, J., 2023	C, R	PO	CHN	OA	4	879	LAPS	+, M loneliness to health-promoting lifestyle	HPLP-CE
[30] Peacock J, 2012	C	PO	AU	A	5	150	PAS-H, OPR	− willingness to undergo surgery	QU
[22] Stallones, L., 1990	C, R	PO	USA	A †	5	152	CAAS	+ illness behavior	IB
[61] Wan, M., 2022	L	E	USA	A	5	187	CCA	M stress and CBD, M stress to alcohol No M stress to cigarettes	SUS

† Aged 45 to 64 years. Design: B = biomarkers; C = cross-sectional; L = longitudinal; R = representative sample. Group: CO = cat owners; DO = dog owners; E = employees; H = homebound; LPO = Latino pet owners; PO = community-residing pet owners or recent pet owners; RU = rural; STR = stroke survivors; W = women. Area: AU = Australia; CAN = Canada; CHN = China; ISR = Israel; NZ = New Zealand; SGP = Singapore; USA = United States of America. Age: A = adult; OA = older adult. OCEBM = OCEBM rating. Pet attachment scales: CAAS = Companion Animal Attachment Scale [43]; CCA = Comfort from Companion Animal Scale [38]; HRS = Health and Retirement Study [46]; I-1 = Single Item (1–10) [14,36,37]; INT = Interview; LAPS = Lexington Attachment to Pets Scale [39]; OPR = Owner–Pet Relationship Questionnaire [49]; PAI = Pet Attitude Inventory [50]; PAQ = Pet Attachment Questionnaire [40]; PAS-H = Pet Attachment Survey-Holcomb [41]; PRS = Pet Relationship Scale [54]. Highlighted findings use scales with lower attachment indicated by a higher score and reported in that manner in the article. The directions of associations are retained in this table but discussed in the context of more attachment having a higher score in the summary tables. Relationship of pet attachment to other physical health outcomes: + = positive, = = no relationship, − = negative, M = moderator, −↓ = slower deterioration, =↓ = no relationship to deterioration. Other Physical Health Measures: # physical health symptoms = number of physical health symptoms; 400 M = 400 m walk time [161,162]; ACS = Activity Card Sort, Recovering Version [163]; ADLS = Activities of Daily Living (ADL) scale adapted from [107]; BSI = Brief Symptom Inventory [95]; CE = Caloric expenditure [155]; CRP = Salivary C Reactive Protein [116]; EWL = List of Adjectives [137]; HPLP-CE = Health-Promoting Lifestyle Profile for Chinese Elderly Scale [157]; HABCPPB = Health ABC Physical Performance Battery [164]; HRS = Health and Retirement Study Module 9 [46]; IB = illness behavior [22]; ILL = Inventory of Limbic Languidness [165]; IPAQ-SF = International Physical Activity Questionnaire Short Form [158]; ISAI = Iowa Self-assessment Inventory [99]; OARS = OARS Multidimensional Functional Assessment Activities of Daily Living, Symptom Severity, Activity Limitations [166]; PA Survey = Physical Activity Survey, SGP Ministry of Health [159]; QU = Study specific questions; RGS = Rapid gait speed [167]; SB = stress biomarkers (salivary cortisol, interleukin-1beta) [116]; SF-36 = Medical Outcome Study Health Survey Short Form-36 [107]; SUS = Substance Use Scale [168]; UGS = Usual gait speed [167]. WHOQOL-BREF = World Health Organization Quality of Life assessment, brief version [109].

). All of the articles were quantitative and cross-sectional [14,29,32,49,65,66,160]. Two of them involved salivary samples [14,65]. All of them were cross-sectional and used convenience samples. The OCEBM ratings were 5 for four articles and 3b for two articles.

Pet attachment was assessed with the LAPS in two articles [65,160] and differently in each of the other articles. One article used each of the following: OPR [49], a single item [14] and the PAS-H [66]. The final article [29] used the PAS-H in combination with the PAQ.

Six outcomes were assessed in this subset of articles with four separate methods. Salivary biomarkers were used to assess the inflammatory biomarker and C-reactive protein (CRP) in two articles [14,65] and to assess stress biomarkers of cortisol and Interluken 1β in one of these articles [65]. BMI was calculated based on the veterinary clients’ estimates of their own heights and weights [160]. Two related validated tools, the SF-36 and its publicly accessible version, the RAND, were used for the assessment of vitality/energy [29,49] and pain [49]. The final article used the Inventory of Limbic Languidness to assess the number of physical health symptoms [66].

Among the six articles that focused on the relationship between pet attachment and aspects of physical health or disease symptoms or indicators, three included older adults only. Two of these articles included the evaluation of stress or inflammatory biomarkers using salivary samples. There was no association of C-Reactive Protein (CRP), an inflammatory biomarker, with pet attachment among homebound older adults (n = 48) in the USA [14]. Stress biomarkers and CRP also were not associated with cat attachment among older adult cat owners (n = 41) in the USA [65]. In the remaining article that focused on Australian older adult pet owners (n = 179), pet attachment was not related to vitality or to pain [49].

Three articles that include, but are not limited to, older adults included examination of the relationship between pet attachment and aspects of physical health or disease symptoms or indicators. In a postal survey of New Zealand adult pet owners (n = 176), pet attachment was related to more physical health symptoms [66]. Higher pet attachment was also related to having higher energy in 75 pet-owning residents of Singapore [29]. Pet attachment was not related to BMI among 75 USA dog owners who were clients of a veterinary practice [160].

Physical Function (Six Articles)

Six articles involved the relationship between pet attachment and physical function (Table 35). All of the articles were quantitative; two were longitudinal [92,129], and the others were cross-sectional [32,49,68,72]. All of the articles included convenience samples. The OCEBM rating categories were 5 for two articles, 3b for two articles, and 2c for two articles.

Half of the articles used the LAPS (3 articles) to assess pet attachment. Two articles used the PRS, one alone [72] and one with the PAI [32], and one article used the OPR [49].

Three groups of outcomes were assessed in the articles about the relationship between pet attachment and physical function. Three articles addressed activities of daily living (ADLs) as a physical function outcome [32,72,92], although one [72] addressed it from the perspective of limitations on physical activity. Two articles used a combination of physical functional measures to assess physical functions [68,129]. Both articles included measures of gait speed over a 6 m course. Other measures included a physical performance battery and a 400 m walk to measure cardiorespiratory performance. The final outcome of physical role disability was assessed with the SF-36 [49].

All six articles addressing the relationship between pet attachment and physical function included only older adult pet owners. The three articles related to activities of daily living (ADLs) were conducted on convenience samples of older adults. In the two cross-sectional studies, pet attachment was not related to ADLS in the USA sample of 24 Latino older adult pet owners [32] and not related to the physical limitations of 53 female older adult pet owners [72] after controlling for emotional wellbeing and depression. In a one-year longitudinal survey of pet attachment and changes in ADLs [92], Canadian older adult pet owners’ (n = 286) attachment to their pets was not related to changes in the ADL.

The two articles that used actual functional measures to assess physical function found that older adult pet owners’ (n = 90) physical function was not related to pet attachment in cross-sectional analysis [68]. However, in longitudinal analysis of a larger sample of older adult pet owners (n = 214) from the same article, pet attachment moderated the deterioration of physical function measures, including rapid and usual gait speed and physical performance, but not cardiorespiratory fitness over an average of 7.5 years.

The article that examined the association of pet ownership with physical role disability revealed no relationship among 179 Australian older adult pet owners [49].

Health and Health-Related Behaviors (Eight Articles)

Eight articles addressed the relationship between pet attachment and health and health-related behaviors (Table 35). One of the articles was qualitative [33]. The remaining seven articles were quantitative [22,30,61,62,91,130,153]. One article [61] was longitudinal; all other articles used a cross-sectional design. The OCEBM ratings for the quantitative articles were category 5 for four articles, category 4 for two articles, and category 3b for one article. Four of the articles used representative samples: two of the USA adult population [22,62], one of the USA’s older adults [130], and the other part of the Chinese population [153]. The other articles used convenience samples. Pet attachment was assessed with different tools in each quantitative article. One article [30] used two scales.

The relationships between pet attachment and eight different health or health-related behaviors were investigated in the eight articles. Three topics were investigated in one article, two topics were investigated in one article, and the remaining articles addressed one topic each.

The behavioral outcomes were addressed with unique tools (Table 34), except for illness behavior, including a combination of health care visits, medications, and hospitalizations, which was addressed with the same tool in two articles [22,62].

Five of the articles, including the qualitative one, were limited to older adults; the other three included, but were not limited to, older adults.

Two articles limited to older adults included discussion of the relationship between pet attachment and participation in life activities, and two included investigations of dog walking. Participation in life activities, evaluated by assessing current participation in 88 real-life activities as a percent of participation in these activities prior to their strokes, was not related to anxious pet attachment among 25 Israeli older adult dog owners who had survived a stroke for at least six months [91]. However, avoidant dog attachment was related to less participation in life activities in this group; the lower the avoidant attachment, the more complete the return to pre-stroke activity levels. Dog walking was associated with greater pet attachment in a representative sample of 188 USA older adult dog owners [130]. The higher the attachment, the more time owners spent walking their dogs. The quantitative article [33] found no direct relationship between Australian older adults’ pet attachment and either life activities or dog walking. Based on the high levels of attachment in the participants (n = 11), the authors suggested an inverted U relationship where high and low attachment would be associated with less activity and lower dog walking, and moderate attachment would be associated with more of both.

Two additional articles used representative samples to examine the relationship between pet attachment and health or health-related behavior in the USA [62] older adult pet owners (n = 339) and Chinese [153] older adult pet owners (n = 879). In the USA sample, older adults’ pet attachment was related to illness behavior, which was operationalized as a composite of the number of medical visits, the number of hospitalizations, the number of prescription medications, and the number of activities compromised due to health in the last six months or year. The higher the attachment, the more illness behaviors [62]. In a telephone survey of pet owners aged 45 to 64 years old obtained from a representative population of the USA (n = 152), pet attachment was related to medical visits and medication use; the higher the attachment, the more of these illness-related behaviors [22]. In the Chinese article, older adults’ attachment to their pets was related to health-promoting lifestyles and moderated the negative impact of loneliness on their participation in a health-promoting lifestyle, including nutrition, physical activity, health responsibility, interpersonal support, self-actualization, and stress management. This effect was independent of relationships with other people.

Three articles addressed the relationships between pet attachment and health-related behavior outcomes in two convenience samples and one representative sample of adults that included but were not limited to older adults. In an article designed similarly to that by Garrity, reported above, illness behavior was examined in relation to pet attachment in a representative sample of 152 pet-owning USA adults. In this article, pet attachment was positively related to illness behavior [22].

Three substance-related behaviors were assessed in relation to pet attachment in one article that included 187 adult pet owners in the USA who were employed and working from home during the COVID-19 pandemic [61]. Pet attachment moderated the positive relationship between stress and CBD and alcohol consumption but not the positive relationship between stress and cigarette consumption; higher stress was associated with more CBD and alcohol consumption, but the higher the pet attachment, the less positive those relationships were.

The final article in this group evaluated the relationship between pet attachment and adults’ decisions about whether to have surgery if they would have to leave their pets [30]. Among 150 Australian adult pet owners, pet attachment was negatively related to willingness to have surgery; the greater the attachment, the less willing the pet owners were to have the surgery.

The articles that examine the relationship between pet attachment and other physical health outcomes are summarized in

Table 36. Summary of findings from analyses in articles examining the relationship between pet attachment and other physical outcomes.

Relationship to Pet Attachment	n	+	=	−	Other
1. Health- or disease-related indicators
Inflammation biomarker	2		[14]; [65]
Vitality/energy	2	[29]	[49]
Basal metabolic index (BMI)	1		[160]
Pain	1		[49]
Physical health symptoms	1	[66]
Stress biomarkers	1		[65]
2. Physical Function
Activities of daily living	2		[32]		=↓ [92]
Cardio-respiratory fitness	2				=↓ [129]
Gait speed—rapid	2		[68]		−↓ [129]
Physical performance	2				−↓ [129]
Gait speed—usual					−↓ [129]
Physical limitations -disease-related	1		[72]
Physical role disability	1		[49]
3. Health- and Health-related Behavior
Dog walking	2	[130]	[33]
Illness behavior	2	[22]	[62]
Participation in activities	3	[91]	[33]; [91]
Alcohol consumption	1				M stress to [61]
Cannabidiol (CBD) consumption	1				M stress to [61]
Healthy lifestyle	1	[154]			M Loneliness to [154]
Smoking	1				No M stress to [61]
Surgery (health care) decision	1			[30]

Numbers in the table indicate the article in the reference list. Each number in [brackets] represents one analysis. + = positive relationship, = = no relationship, − = negative relationship, =↓ = no difference in deterioration, −↓ = slower deterioration. Bold indicates older adults only in the article.

. The preponderance of evidence suggests that there is no direct relationship between pet attachment and these outcomes. When the categories of other outcomes are examined separately, the preponderance of evidence does not support a relationship between pet attachment and physical function outcomes. Among the several analyses that examined the relationship between pet attachment and health or disease indicator outcomes, three were in articles focused on adults and four were in articles focused on older adults. Seventy-five percent of the analyses focused on older adults demonstrated no relationship, while 33% of the articles focused on adults demonstrated positive, nil, and negative relationships between pet attachment and positive health indicators. Among the analyses that addressed the relationship between pet attachment and other physical function outcomes, it was related to worse physical function in one of the analyses (among adults) and not related to physical function in three of the analyses among older adults and one among adults. Only two analyses evaluated the relationship between pet attachment and health behaviors in adults. One demonstrated no relationship, and the other demonstrated a relationship between greater pet attachment and more unhealthy health behavior. However, there is some evidence that pet attachment moderates changes in physical function with aging.

4. Discussion

When the outcomes reported in the articles, as summarized in the results, were considered together (

Table 37. Summary of direct relationships between pet attachment and health outcomes in articles addressing adults, including older adults, and those addressing older adults only *.

	Adults			Older Adults
Outcome Relationship	+	=	−	+	=	−
Negative Psychological Outcomes
Depression	5	9	2	1	8	2
Anxiety	3	3	0	0	3	1
Distress/Stress	2	6	2	0	1	0
Grief	5	0	0	2	0	0
Other negative psychological	5	17	2	0	1	0
Negative Social Outcomes
Loneliness	4	3	0	3	2	0
Other negative social	0	1	1	0	1	0
Negative Physical Outcomes
Health/disease indicators (negative)	1	1	0	0	3	0
Physical function (negative)	1	0	0	0	1	0
Health behaviors (negative)	0	1	0	0	0	0
Positive Psychological Outcomes
Psychological/emotional wellbeing/health/QoL	7	5	4	1	7	1
Positive affect	3	2	1	0	0	0
Cognition	0	0	0	1	5	0
Other positive psychological	5	3	0	2	8	1
Positive Social Outcomes
Social support	2	1	0	1	3	0
Social wellbeing	3	5	1	0	2	0
Other positive social	0	0	1	2	2	1
Positive Physical Outcomes
Physical wellbeing	2	5	1	0	5	1
Physical activity	2	1	0	1	2	1
Health/disease indicators (positive)	1	0	0	1	0	0
Physical function (positive)	0	0	0	0	2	0
Health behaviors (positive)	0	0	1	3	3	0
Summaries of Information
Psychological Outcomes Negative	20	28	4	8	11	4
Social Outcomes Negative	4	4	1	4	4	0
Physical Outcomes Negative	2	1	1	0	5	0
All Negative Health Outcomes Summary	26	33	6	12	20	4
Psychological Outcome Summary Positive	15	11	6	5	12	1
Social Outcome Summary Positive	5	4	1	5	7	0
Physical Outcome Summary Positive	5	6	0	5	21	2
All Positive Health Outcomes Summary	25	21	7	15	40	3
All Outcomes Converted to Positive	31	54	33	19	60	15

* A valence was not assigned to BMI. + = positive relationship, = = no relationship, − = negative relationship.

), pet attachment was not directly related to most health outcomes. Among the 212 analyses of the relationship between pet attachment and health outcomes reported in the 58 articles, 54% (114) showed no relationship between the outcome and pet attachment. Twenty-four percent (50) of the analyses showed a relationship indicating that pet attachment was associated with better health outcomes, and 23% (48) of the analyses showed a relationship indicating that pet attachment was associated with worse health outcomes.

4.1. Valence of Health Outcomes and the Relationship Between Pet Attachment and Outcomes

The relationship of health outcomes with pet attachment was not different overall depending on the valence of the health outcome assessed [Chi square (92 df) = 0.143, p = 0.93]. Higher pet attachment was associated with better health outcomes in 10% (10/101) of the negative health outcomes and 36% (40/110) of the positive health outcomes. Pet attachment was not associated with health outcomes in 53% (53/101) of the negative health outcomes and 55% (61/110) of the positive health outcomes. Pet attachment was associated with worse health outcomes in 38% (38/101) of the negative health outcomes and 9% (10/106) of the positive health outcomes.

The relationship between pet attachment and health outcomes was examined according to the age groups of the study participants and the valences of the health outcomes within the age groups.

4.1.1. Differences in the Relationship of Pet Attachment with Negative Health Outcomes Between Adults and Older Adults

Examining all negative health outcomes, there was not a significant difference in the distribution of positive, null, and negative relationships of pet attachment with the health outcomes between adults and older adults [Chi square (2df) = 0.458, p = 0.80]. This finding suggests that negative health outcomes are related to pet attachment similarly across the age spectrum of pet owners,

4.1.2. Differences in Relationships of Pet Attachment with Positive Health Outcomes Between Adults and Older Adults

Examining all positive health outcomes, there was a significant difference in the distribution of positive, null, and negative relationships of pet attachment with positive health outcomes between adults and older adults [chi squared (2 df) = 9.813, p ≤ 0.007]. Among the 53 analyses involving adults, 47% (25) of outcomes investigated demonstrated a positive relationship, and 40% (21) demonstrated no relationship between pet attachment and the positive health outcome. In contrast, among the 58 analyses involving only older adults, 26% (15) of outcomes investigated demonstrated a positive relationship, and 70% (40) demonstrated no relationship between pet attachment and a positive health outcome. This analysis seems to indicate that adults may reap more health benefits related to higher attachment to their pets than do older adults.

There were no significant differences in the relationship between pet attachment and negative health outcomes between adults and older adults [Chi sq (2 df) = 1.468, p = 0.79].

As indicated in the results, only one outcome, grief, was uniformly associated with pet attachment. This positive relationship was uniform across all articles reporting on it, in both articles focused exclusively on older adults and those that included older adults. This indicates that uniformly, adults and older adults who are highly attached to their pets experience greater levels of grief than those who are not as attached to their pets.

4.2. Research Quality

The findings of the relationship between pet attachment and health outcomes are not based on the highest quality research. Forty-five percent of the articles (25/55) are of OCEBM category 5, 20% (11/55) are category 4, 25% (14/55) are category 3b, and 11% (5/55) are category 2c. The higher the OCEBM rating, the lower the quality of research. This suggests that higher quality research is required in order to draw stronger conclusions.

4.3. Shape of the Relationship Between Pet Attachment and Health Outcomes

Two sets of authors, including Bennett, Beckwith, Chur-Hanson, Hill, and Winefield [17,33], suggested an inverted U relationship between pet attachment and health outcomes. In this case, at low attachment, there would be a positive relationship between pet attachment and health outcomes; at moderate attachment, there would be no relationship between pet attachment and health outcomes; and at high attachment, there would be a negative relationship between pet attachment and health outcomes. We decided to investigate whether this possibility was supported by the data from the articles included in this review. Nineteen different scales were used to assess pet attachment (Table 3). Only the LAPS was used sufficiently frequently, 21 times, to provide a variety of attachment scores in different articles. Data on the distribution of LAPS scores were provided in 16 articles. Mean LAPS scores were adjusted for each article into the original 0–69 range, and standard errors were calculated. Mean LAPS scores ranged from 41.4 [15] to 64 [68]. The distribution of the scores according to age groups is illustrated in Figure 2. Visual examination revealed that the articles that focused on older adults had either high or low attachment, while those that focused on adults were intermediate. However, the three articles [56,68,129] that had the highest attachment scores were from related samples (ns = 214, 90, 178; ages: Ms = 63.0, 71.8, 64.8, SDs = 7.8, 9.8, 8.7 years) of affluent, largely white, community-residing older-adult USA pet owners. The three articles with the lowest attachment scores were from community-residing, older adult, USA, cat owners (n = 41, age: M = 72.6, SD = 7.3 years) [65]; rural, white, older adult, USA, dog owners (n = 111, age: M = 68.4, SD = 7.8 years) [15]; and a random sample of community-residing, older adult, Canadian, pet owners (n = 286, age: M = 73.0, SD = 6.3 years) [92].

To avoid the difficulty of defining LAPS scores as high, medium, or low [17,33], we divided the scores into intervals with a width of five points and examined the relationship of pet attachment with health outcomes in each interval (

Table 38. Summary of the relationship of pet attachment as assessed with the LAPS to health outcomes categorized by the mean LAPS score in the article.

LAPS Score Range Variable	41–45	46–50	51–55	56–60	61–65
Depression	+ [15]	= [65], + [74]	+ [27]	= [70], + [57], + [60]	= [56],= [68]
Psychological Wellbeing		=↓ [92]	− [26]	= [70]	= [56], = [68], = [129]
Anxiety			+ [123]		= [56], = [68]
Stress/distress		= [65], [32], = [74]	+ [26]
Grief			+ [120]
Positive emotions/affect		− [74]		= [70]
Cognition					= [68]
Anger			+ [120], + [27]
Executive function		+ [65]			= [68], −↓ [129]
Happiness					= [56], = [68]
Memory					= [68], =↓ [129], +↓ [129]
Guilt			+ [120]
Immoderation			= [27]
Self-consciousness			= [27]
Vulnerability			+ [27]
Loneliness		= [65], + [74]		+ [57]
Social support		+ [21]	= [17]	+ [57]
Social wellbeing/function/QoL				= [70]
Daily lives–social behavior					+ [68]
Physical wellbeing, health, QoL				= [70]	= [68], =↓ [129]
Activities of daily living (ADLs)		=↓ [92]
Cardio-respiratory fitness					=↓ [129]
Gait speed					= [68], −↓ [129]
Physical performance					−↓ [129]
Basal metabolic index (BMI)				= [161]
Summary	n	n	n	n	n
+	1	4	8	4	2
=		6	3	6	17
−		1	1	0	3

Numbers in the table indicate the article in the reference list. Each number in [brackets] represents one analysis. Relationship between pet attachment and health outcomes: + = positive, = = no relationship, − = negative, −↓ = slower deterioration, =↓ = no difference in deterioration, +↓ = faster deterioration. Bold indicates an article with only older adult participants.

). This crude evaluation does not support the inverted U hypothesis based on the high number of assessments of outcomes with no relationship between pet attachment and the outcome in the highest interval (LAPS score means of 61–65), the preponderance of outcomes with no relationship between pet attachment in the interval of 46 to 50, and the large number of outcomes with positive relationships to attachment in the LAPS score 51–55 interval. This is an example where more research is needed to more precisely evaluate the inverted U-shaped function. It is also possible that the different attachment scales used in this examination are not assessing the same aspects of attachment. These scales appear to assess a global construct thought of as attachment to pets, but we know from the human attachment literature that people have different attachment styles [7]. Many researchers have drawn parallels between pet–owner and infant–parent attachment styles, and at least one study demonstrated that the proportion of these different attachment styles exists in dogs and their owners [169]. A minority of the articles included in this review assess pet attachment styles. Examples include articles by Northrope, Ellis, Joseph, Kwong, Reevy, Tan, Teo, and White [21,26,27,28,29,34,67,94]. It is possible that the blunt measures of attachment being used in most of the articles do not flesh out enough information to provide the precision needed to assess the presence of a U-shaped function. For example, we do not know how secure and insecure attachments result in different attachment scores as measured by these global measures. It is entirely possible, perhaps likely, that secure attachment to a pet is positively related to positive health outcomes, while insecure attachments are inversely related to the same outcomes, but there are too few studies examining attachment orientation to conduct a separate review or to draw consistent conclusions. The lack of information about attachment styles may contribute to the variability in the relationships reported in the published manuscripts and partially explain the inconsistent results. This suggests the need for a new measure of attachment to be developed that can assess both global attachment and attachment style.

There was no evidence of ceiling effects interfering with finding relationships. See the distribution of scores for LAPS in Figure 2. It is important to note that the maximum LAPS score is 69 [39]. Some articles scored the LAPS differently (original 0–3 for each item vs. 1–4 or 1–5 for each item or item average). LAPS scores were converted to the original scale for comparison.

4.4. Longitudinal Analyses: Pet Attachment and Changes in Health Outcomes over Time

Two articles [92,129] used longitudinal designs to address the relationship between pet attachment and changes in health outcomes. An additional article [60] examined the relationship between pet attachment and changes in health outcomes based on cross-sectional data collection. The preponderance of evidence (7/9 outcomes) was that pet attachment was not related to changes in health outcomes among older adults. Pet attachment was associated with slight decreases in executive function and with faster decreases in long-term memory. In one article, there was no relationship between pet attachment and changes in mental health in older adults.

4.5. Pet Attachment as a Moderator of Relationships

Nine articles [17,23,25,26,56,61,64,65,154] evaluated pet attachment as a moderator of relationships that predicted health outcomes (

Table 39. Summary of analyses of pet attachment as a moderator in predictions of psychological, social, and physical health outcomes.

Article	Age	Y/N	Predictor	Outcome
Psychological Health Outcomes
[64] Antonacopoulos	A	N	Social support	Depression
[56] Koodaly	OA	N	Gait speed	Depression
[56] Koodaly	OA	N	Physical performance	Depression
[56] Koodaly	OA	N	Physical wellbeing	Depression
[61] Wan	A	Y	Stress	Depression
[25] Hardie	A	N	Social support	Psychological wellbeing
[56] Koodaly	OA	Y	Physical wellbeing	Psychological wellbeing
[56] Koodaly	OA	N	Gait speed	Psychological wellbeing
[56] Koodaly	OA	N	Physical performance	Psychological wellbeing
[60] McDonald	A	Y	COVID stress	Psychological wellbeing
[26] Northrope	A	Y	Anxious attachment to people	Psychological wellbeing
[56] Koodaly	OA	N	Gait speed	Anxiety
[56] Koodaly	OA	N	Physical performance	Anxiety
[56] Koodaly	OA	Y	Physical wellbeing	Anxiety
[61] Wan	A	Y	Job insecurity	Stress
[56] Koodaly	OA	N	Gait speed	Happiness
[56] Koodaly	OA	N	Physical performance	Happiness
[56] Koodaly	OA	N	Physical wellbeing	Happiness
[17] Hill	A	N	Social support	Resilience
[23] Turner	A	Y	Partnered living situation	Good mood
[25] Hardie	A	Y	Pet support (moderator pet emotional closeness)	Life satisfaction
[23] Turner	A	N	Partnered living situation	Bad mood
[61] Wan	A	Y	Stress	Emotional exhaustion
Social Health Outcomes
[64] Antonacopoulos	A	Y	Social support	Loneliness
Physical Health Outcomes
[154] Lu	OA	Y	Loneliness	Health-promoting lifestyle
[61] Wan	A	N	Job insecurity	Behavior—cigarette use
[61] Wan	A	Y	Job insecurity	Behavior—CBD use
[61] Wan	A	Y	Job insecurity	Behavior—alcohol use

Age: A = adults; OA = older adults only; Y/N: Y = yes, pet attachment moderated the relationship; N = no, pet attachment did not moderate the relationship.

), excluding those examined in longitudinal analyses.

In 28 instances, pet attachment was examined as a moderator of relationships of predictors with psychological (n = 23), social (n = 1), or physical (n = 4) health outcomes. These results reveal that in 13 of 28 instances where authors indicated that they examined moderation, pet attachment moderated important relationships, mostly affecting relationships predicting psychological outcomes. Two articles were responsible for 12 psychological [56] and one physical [155] health moderation analysis for older adults. Moderation occurred in three of them: two psychological and one physical. Seven out of 11 analyses of pet attachment as a moderator of predictors of psychological outcomes, the only analysis of pet attachment as a moderator of predictors of social outcomes, and two of three analyses of pet attachment as a moderator of physical (behavior) outcomes demonstrated significant moderation in adults. This is a promising area for future research.

Five articles [17,23,25,56,61,64] indicated that non-significant moderation was found in at least one analysis, and all except one [17] of them also showed significant moderation. It is possible that many other moderation effects were examined but not reported due to a lack of significance. Evaluation of pet attachment as a moderator of relationships that predict health outcomes is a promising area for future research, and we emphasize the need for transparency in reporting analyses that resulted in non-significant findings.

4.6. Limitations

While this review has pulled together a large amount of information, there are limitations in the analyses presented herein. In some cases, we have provided crude analyses as a way of giving the reader a thoughtful evaluation of the information available to us, but to be more precise in our conclusions, we would need to base them on higher-quality studies using very precise instruments. This review is only as strong as the research that has been conducted, and there are inherent flaws in pet ownership research. For example, pet ownership itself is measured in a wide variety of ways, some of which are more refined. A simple question of whether there is a pet in the home. Some studies seek additional information, such as primary responsibility for and time spent with the pet. Finally, most of the research is based on individuals from developed and/or Western countries. It is not clear how the responses and results would differ in other cultures and/or countries.

5. Conclusions

This systematic review examined 58 peer-reviewed studies on the topic of the relationship between pet attachment and health outcomes in older adults. With the exception of grief, the findings do not support a clear relationship between pet attachment and health outcomes for older adults. They suggest that the relationship between pet attachment and health outcomes, especially positive psychological outcomes, is not the same for older adults as for younger individuals. Further research is necessary to more precisely examine the relationship of pet attachment to health outcomes in older adults.