Impact of pet ownership and animal-assisted therapy on suicidal ideation and suicide deaths: A scoping review

Abstract

Background

There is growing research on therapeutic benefits of animals. However, their impact on suicide-related outcomes such as suicidal ideation (SI), suicide attempts (SA), and suicide deaths remains unclear. This scoping review consolidates existing literature on the role of pet ownership and animal assisted therapy (AAT) on suicide-related outcomes.

Methods

A PRISMA-compliant systematic search of MEDLINE, EMBASE, and PsycINFO was conducted up to March 10, 2025. Articles reporting the impact of pet ownership or AAT on SI, SA, or suicide deaths were included. Screening and data extraction were performed in duplicate, with risk of bias assessed using appropriate tools for each study design except for case reports and interviews. Study findings were descriptively summarized.

Results

25 studies were included: one RCT, five pre- and post-studies, two cohort studies, one case-control study, three case reports/series, and 13 surveys/interviews. For pet ownership, a cohort study (n=709) found a positive association between time with dog and SI (r=0.17, p<0.001). Two other observational studies yielded non-significant results against suicide deaths. For AAT, one pre- and post-study of veterans (n=71) reported significant SI reductions (t(54)=4.87, p<0.001) while another pre- and post-study of adolescents (n=30) also found significant SI reductions (χ² McNemar=6.75; p<0.05). Four other experimental studies, including one RCT, yielded non-significant results. Qualitative findings suggested pet ownership and AAT may have a perceived protective effect against SI. No articles reported on SA. Most articles had a high or moderate risk of bias.

Conclusion

While there is a reported perceived protective effect of pet ownership and AAT against SI, the evidence for pet ownership and AAT remains inconclusive in relations to SI, SA, and suicide deaths. Future research should focus on high-quality with stringent methodology, large-scale, experimental studies with standardized outcome measures to clarify the role of animals in suicide prevention.

Introduction

The role of animals has garnered increasing attention within the mental health field due to their potential therapeutic benefits. Studies have highlighted the positive effects of animal interactions on various psychological outcomes, including stress reduction, improved mood, and enhanced social interactions.^1–3 These benefits are especially significant for people with mental disorders, who often face complex challenges, including an increased risk of suicide.⁴ Consequently, investigating the roles of pet ownership and animal-assisted therapy (AAT) in addressing suicide-related outcomes–such as suicidal ideation (SI), suicide attempts (SA), and suicide deaths–is crucial, given their potential in suicide prevention.^5–7

Suicide is a significant global public health crisis, ranking among the leading causes of death worldwide. In 2021, 746,000 people died by suicide, equating to approximately one death every 40 seconds.⁸ Despite advancements in prevention efforts and interventions, suicide continues to be the primary cause of death among individuals aged 10 to 34, with a 35% increase in suicide rates from 1999 to 2018, and persisting at high levels across various mental disorders.^9–12 These statistics highlight the critical need for innovative, accessible, and effective interventions to supplement existing suicide prevention efforts.

Social isolation, emotional distress, and a lack of coping mechanisms are well-established risk factors for suicide.⁴ Traditional treatment modalities–including pharmacological interventions such as conventional antidepressants, ketamine/esketamine, and psychotherapeutic approaches–have demonstrated efficacy in reducing suicidal symptoms.¹³ However, many individuals continue to experience persistent SI or remain untreated due to barriers such as stigma, limited mental health resources, or personal reluctance to seek care.⁴ Thus, alternative interventions that enhance emotional support and social connectedness, such as pet ownership and AAT, warrant further exploration.

Pet ownership, in the context of this review, refers to the long-term care and companionship of domesticated animals within a home setting. Pets can offer emotional support, reduce loneliness, and encourage routine and responsibility, which may serve as protective factors against suicide risk.⁵ AAT, on the other hand, is a structured therapeutic intervention involving interactions between individuals and trained animals, facilitated by a healthcare professional or therapist. AAT has been studied in various clinical populations for its potential to reduce stress, enhance emotional well-being, and improve social engagement.⁶ Given these mechanisms, research on the impact of pet ownership and AAT on suicide-related outcomes is essential to determine whether these approaches can play a mitigating role.

Despite growing interest in the psychological benefits of human-animal interactions, the extent to which pet ownership and AAT can be integrated into suicide prevention strategies remains an emerging area of study. This scoping review aims to explore and consolidate the existing literature on pet ownership and AAT in the context of suicidal thoughts and behaviors in humans, including key findings, knowledge gaps, and methodological approaches. Ultimately, this review aims to provide insights that can inform future research directions and clinical practice related to the use of animals in suicide prevention efforts.

Methods

This scoping review was conducted adhered to the the Joanna Briggs Institute (JBI) methodology for scoping reviews.¹⁴ We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension Statement for Reporting of Scoping Reviews (PRISMA-ScR) (eTable 1).¹⁵ The protocol was uploaded to Open Science Framework a priori at https://osf.io/ewqx2/.

Search strategy

MEDLINE, EMBASE, and PsycINFO were searched from inception to March 10, 2025 using a search strategy developed in consultation with a health sciences librarian (RS). Terms relating to pets, animal-assisted therapy, and suicide (eTable 2) were used. A further manual search was conducted on Google Scholar using the above search terms and within the references of eligible literature to identify additional relevant articles.

Inclusion and exclusion criteria

Peer-reviewed literature was included if it met the following criteria: (1) included human participants and (2) reported the outcomes that pet ownership or AAT had on SI, SA, or suicide deaths, regardless of other concurrent treatment. SI refers to thoughts of engaging in suicide, ranging from passive contemplation of death to active planning with intent. SA is defined as a non-fatal, self-directed act carried out with the intent to die, while suicide death refers to a fatal self-inflicted act. We considered AATs to be any health intervention, meant to improve physical, social, emotional, or cognitive functioning, with animals as an integral part of the treatment. We considered pet ownership to be the act of living with or caring for a domesticated animal without any specific therapeutic training. Pet ownership status may be self-reported as “pet ownership” or “pet companionship.”

We did not set any limits based on study design, setting, geographic location, publication date, or language. We excluded review articles, study protocols, and conference abstracts for which a full version of the document that provided additional information was not available.

Study screening and data extraction

Articles identified through the search were imported into Covidence software for abstract and full-text screening.¹⁶ Three reviewers (CZ, NF, HC) screened titles and full texts independently and in duplicate, with disagreements resolved by consensus mediated by a third reviewer.

Two reviewers (CZ, HC) independently extracted relevant data from the included literature onto a Microsoft Excel extraction form designed a priori. All extractions were done independently and in duplicate with discrepancies resolved by consensus. Our primary outcome was the impact of pet ownership or AAT on SI, SA, and suicide deaths in humans. We extracted measures of the primary outcome as reported by the authors. We collected comprehensive demographic data and information regarding intervention, study design, mental/physical conditions, and control/comparator groups when applicable.

Risk of bias

The risk of bias in the included literature was assessed using Cochrane's Risk of Bias 2 (RoB 2) tool for randomized controlled trials,¹⁷ the Risk Of Bias In Non-randomised Studies of Interventions (ROBINS-I) tool for non-randomized controlled trials,¹⁸ the National Institute of Health (NIH) Study Quality Assessment Tool for cohort studies, cross-sectional studies, case-control studies, pre- and post-studies, and case series,¹⁹ and the Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for qualitative surveys.²⁰ Two independent reviewers (CZ, PY) assessed the risk of bias independently and in duplicate with discrepancies resolved by consensus. The risk of bias in case reports and interviews was not evaluated.

Results

The initial searches retrieved 248 records. 193 unique records were retrieved after removing 55 duplicates. We screened these records by title and abstract, excluding 140. We then assessed the full text of 53 potentially relevant articles, excluding 28 (eFile 1). This scoping review included 25 articles published between 2009 and 2024 (Fig. 1).

Fig. 1.

PRISMA article selection flow diagram.

Descriptive summary

The 25 included articles are composed of eight interviews,^21–28 five surveys,^5,29–32 five pre- and post-studies,^6,33–36 two cohort studies,^37,38 two case series,^39,40 one case report,⁴¹ one randomized controlled trial (RCT),⁴² and one case-control study.⁴³ Their countries of origin include the United States (n=11), the United Kingdom (n=7), Australia (n=5), Canada (n=1), and Spain (n=1). The age range of participants was 10-84 years. Ten articles included individuals with post-traumatic stress disorder (PTSD), eight involved participants with mood or anxiety disorders, two examined psychotic disorders, and one focused on autism spectrum disorder. No article included only healthy participants (i.e., without a mental disorder), although eight articles did not report participants’ mental disorder status. Most articles did not report sample sizes for each disorder or the diagnostic criteria used. Nine articles focused on veterans, while one each addressed homeless individuals, older adults, young adults, and adolescents, respectively. 14 articles examined pet ownership, while 11 examined purpose-trained animals (e.g., service or therapy animals). Dogs were the most reported animal (n=26), followed by cats (n=12); horses (n=2), birds (n=2), and “other pets” (n=1) were also reported. No study specified the breeds of each animal species. The descriptive summary of each included article is presented in tabular format in Tables 1 and 2. Further detail is presented in eTables 3 and 4 in the supplementary materials.

Table 1.

Study characteristics, results, and risk of bias of the included articles on pet ownership and suicidal ideation and deaths (n=14). Statistically significant results are marked with * and underlined.

Author(s), year(location)	Study design	Population and health conditions	Methods	Intervention and comparator (if applicable)	Results and adverse events	Risk of bias score (green=low, yellow=moderate, red=high) and tool
Barcelos et al., 2023(UK)	Prospective cohort study	709 adult dog owners.Age: 18-84 years.Sex: 91.3% female.Health conditions: 25.1% reported experiencing a mental health condition, and 27.5% had a physical health condition.	Methodology: Conducted online with participants completing weekly questionnaires over 4 weeks.Assessment: Suicidal ideation was measured using a single question about the occurrence of suicidal thoughts in the last seven days.Analysis: A random intercept cross-lagged panel model (RI-CLPM) was used to test correlations and potential causal relationships.	Dog-related factors: 17 factors, including physical health of the dog, running or jogging with the dog, walking the dog, failure to do something for the dog, dog greeting at the door, friendly conversations due to the dog, aggressive behavior, fearful behavior, and training the dog.	Results:* - Proportion of time spent with dog was correlated with suicidal ideation (r=0.17, p≤0.001). - No significant causal relationships were found between the factors and suicidal ideation. Adverse events: None reported.	Moderate risk, NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies
Batty and Bell, 2018(UK)	Prospective cohort study	67,441 pet owners with data on covariates and mortality. 47 suicide deaths. No age or sex data was reported.Health conditions: No reported psychiatric or physical diagnosis.	Methodology: Used data from the Health Surveys for England collected in 1995–1997, 2001, 2002, and 2004. Up to 17.1 years of follow-up (median=10.4 years).Assessment: Census questionnaire. Participants were asked if they kept household pets inside their homes and what pets they owned.Analysis: The study used hazard ratios (HRs) to evaluate the risk of suicide related to pet ownership. Covariates were adjusted for age, sex, socioeconomic status, mental health problems, smoking status, somatic illness, and marital status.	Animal companionship (Pet ownership): Specifically, dog, cat, other pets, any pet, or no pet. The study did not involve an active intervention but rather observed natural differences in pet ownership and their associations with suicide risk.	Results: - Overall Pet Ownership Suicide Risk: HR=1.0, 95% CI (0.5 to 1.7) after multivariable adjustment, indicating no significant association between any pet ownership and suicide risk. - Dog Ownership Suicide Risk: HR=1.3, 95% CI (0.7 to 2.6) after multivariable adjustment; not statistically significant. - Cat Ownership Suicide Risk: HR=0.3, 95% CI (0.1 to 1.1) after multivariable adjustment; not statistically significant, though it suggests a potential protective trend. - Other Pets Suicide Risk: HR=1.5, 95% CI (0.6 to 3.9) after multivariable adjustment; not statistically significant. Adverse events: None reported.	Low risk, NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies
Helsing and Monk, 1985(USA)	Case-control study	48 individuals who committed suicide.Age and sex: 15-24 years: 4 (all male); 25-44 years: 15 (10 males, 5 females); 45-64 years: 22 (14 males, 8 females); 65+ years: 7 (5 males, 2 females).Sex: 33 males and 15 females.Controls: Set #1: 96 living individuals matched to the suicide cases by race, sex, and age. Set #2: 96 individuals who died of other causes in the same year as the suicide cases, matched by race and sex.Health conditions: No reported psychiatric or physical diagnosis.	Methodology: Used data from a health census conducted in Washington County in 1975. Pet owners were tracked over 8 years (1975 to 1983) to see who committed suicide as per the county's death records.Assessment: Households were asked about pet ownership, specifically dogs and cats.Analysis: Matched case-control analysis using Schlesselman's procedures for outcomes with two controls per case. The study calculated odds ratios (ORs) and 95% confidence intervals (CIs) to examine the relationship between pet ownership and suicide.	N/A	Results: - Overall Pet Ownership: 56.2% of suicide cases and 56.2% of matched controls owned pets. OR=1.00, 95% CI (0.50 to 2.01); not statistically significant. - Dog Ownership: 41.7% of suicide cases owned dogs, compared to 42.7% of controls. OR=0.95, 95% CI (0.20 to 5.93); not statistically significant. - Cat Ownership: 29.2% of suicide cases owned cats, compared to 25.0% of controls. OR=1.22, 95% CI (0.37 to 4.00); not statistically significant. Subgroup analysis: - Men: OR=1.06, 95% CI (0.46 to 2.46). - Women: OR=0.88, 95% CI (0.06-13.7). - Non-farm Residences: OR=1.19, 95% CI (0.24-5.93). - Suicides (1975-1978): OR=0.73, 95% CI (0.08-6.42). - Suicides (1978-1983): OR=1.21, 95% CI (0.24-6.05). - Second Control Set (Matched Deaths): OR=0.91, 95% CI (0.49-1.69); no significant difference compared to individuals who died of other causes. Combined analysis (all controls): - OR=0.98, 95% CI (0.66-1.45); not statistically significant.	Low risk, NIH Quality Assessment Tool for Case-control Studies
Hutton, 2016(Australia)	Case series	5 pet owners Age: 30-58 years (range).Health conditions: HIV for 7-30 years. No other reported psychiatric or physical diagnosis.	Methodology: Conducted semi-structured interviews.Assessment: Participants were asked semi-structured questions to gather narratives on the interplay between their attachment to a companion animal, projection, and feelings of death anxiety.Analysis: Responses were analyzed through a psychoanalytic lens.	N/A	Results: - Projection of Death Anxiety: Individuals with HIV might project their death anxiety onto their companion animals, using the animals as a means to cope with the fear of their mortality. - Buffer Against Suicidal Ideation: The presence of companion animals provided participants with a sense of responsibility and a tangible reason to maintain their health (e.g., taking medications). This connection to the animals helped alleviate suicidal thoughts and provided a protective factor against self-harm. - Anticipatory Grief: Participants expressed anxiety about the eventual death of their animals, indicating that they might use this projected anxiety as a way to avoid confronting their mortality. The bond with the animals provided comfort but also intensified their death-related fears. - Attachment and Merging: Some participants experienced a blurring of boundaries between themselves and their animals, describing the animals as "part of their soul." This intense attachment to the animal offered emotional support and a sense of being needed.	Low risk, NIH Quality Assessment Tool for Case Series
Cooke, 2013(USA)	Case report	1 middle-aged womanHealth conditions: A possible history of mild depression. No other reported psychiatric or physical diagnosis. Her pet dog was found dead beside her in her car due to carbon monoxide poisoning.	N/A	N/A	Results: - The bond with a pet might involve intense attachment, possibly substituting for human relationships, and the killing of a pet during suicide could serve multiple psychological functions: 1) Projection of Altruism: The individual may believe they are sparing the pet from suffering in their absence. 2) Attachment and Loss: The pet's presence might represent the individual's past life, and its death could symbolize a change in the person's self-image. 3) Protective Role: It is also suggested that pets can sometimes serve as a protective factor against suicide by providing a sense of responsibility and care.	N/A
Smith et al., 2022(UK)	Cross-sectional survey	1087 veterinary professionals. 86.6% were pet owners. Most common pets: Dogs (73%), cats (54.5%), horses (18.2%).Age: 39.3% aged 31-40 years, 34.5% aged 18-30 years, 16.1% aged 41-50 years, and 10.1% aged 51 years and over.Gender: 86.9% female, 12.0% male, 1.1% identifying differently.Health conditions: No psychiatric or physical diagnosis was specified.	Methodology: Conducted online surveys over 4 weeks in June and July 2021.Assessment: Suicidal ideation was assessed with a question about thoughts of taking one's life in the past 12 months.Analysis: Multiple linear regression models assessed associations between pet ownership, physical activity, and mental health outcomes. F-tests were used to calculate p-values, and variables were adjusted for age, gender, employment status, relationship status, and others.	N/A	Results: - Pet Ownership: Pet owners reported higher levels of depression than non-pet owners (PHQ-9 mean score: 8.8 vs. 7.4; F[1, 1064]=7.19, p=0.007). Authors report no significant difference in anxiety or suicidal ideation but did not provide a F statistic or p value. - Dog Ownership: Associated with lower anxiety (GAD-7) scores (F[1, 918]=5.56, p=0.019) and lower suicidal ideation (χ²(1)=5.04, p=0.025). - Horse Ownership: Linked to better mental health indicators, including lower anxiety (F[1, 918]=5.60, p=0.018), depression (F[1, 920]=8.34, p=0.004), and suicidal ideation (χ²(1)=4.32, p=0.038). - Cat Ownership: Associated with higher levels of depression (F[1, 920=4.37, p=0.037) and suicidal ideation (χ²(1)=10.52, p=0.001).	High risk, Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for Qualitative Surveys
Douglas et al., 2021(USA)	Cross-sectional survey	187 undergraduate student pet owners.Age: 18-25 years (range), 18.89 (mean, SD±1.24).Gender: 71 men, 115 women, and 1 transgender.Health conditions: No psychiatric or physical diagnosis was specified.	Methodology: Conducted online via Qualtrics for class credit, utilizing the psychology department's participant pool.Assessment: Suicide risk was measured with the Suicidal Behaviors Questionnaire-Revised (SBQ-R), a 4-item scale. Reliability: Cronbach's α = 0.79. Used a cutoff score of 7 for nonclinical samples; 18.7% of the sample were at elevated suicide risk.	N/A	Results: - Indirect Effect: Higher overall pet attachment was indirectly associated with lower perceived burdensomeness (PB), which was in turn associated with lower suicide risk (Point estimate=-0.45, 95% CI -0.91 to -0.04). - Thwarted Belongingness (TB): No significant indirect association between overall pet attachment and suicide risk through TB (Point estimate=-0.23, 95% CI -0.65 to 0.07). - Follow-up Analysis: Direct negative association between overall attachment and suicide risk (F[1 , 152]=5.00, β=-0.70, 95% CI -1.32 to -0.08, p=0.03). - Attachment Anxiety: Positively associated with suicide risk through both TB and PB. Indirect association with suicide risk via PB (Point estimate=0.04, 95% CI 0.02 to 0.06) and TB (Point estimate=0.03, 95% CI 0.01 to 0.05). - Attachment Avoidance: Positively associated with increased TB, which was in turn linked to higher suicide risk (Point estimate=0.03, 95% CI 0.01 to 0.06). No significant association was found for PB (Point estimate=0.02, 95% CI -0.01 to 0.05).	Moderate risk, Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for Qualitative Surveys
Love, 2021(USA)	Cross-sectional survey with thematic analysis	71 people with a history of suicidal behaviors. No age or sex data was reported.Health conditions: No psychiatric or physical diagnosis was specified.	Methodology: Conducted online using Reddit, an anonymous social media site, to recruit participants.Assessment: A brief, open-ended survey question: "What role did your pets have in your life at the time you were experiencing suicidal thoughts or behaviors?"Analysis: Thematic analysis was conducted on participants' responses. The responses were coded, and a codebook was developed to identify themes.	N/A	Results: - Pets as a Protective Influence (70.4% of responses): 1) Comfort: Pets provided companionship, emotional support, and a calming effect. 2) Distraction: Pets provided a diversion or interference that helped participants focus on something other than their suicidal thoughts. 3) Reason to live: Pets gave participants a sense of purpose, hope, or an obligation to continue living to care for their animals. - Pets Played No Role (14.1% of responses): Pets did not influence their suicidality, either because the pets were not present or because they perceived them as "just pets." - Pets as a Risk Factor (5.6% of responses): Pets caused additional stress, particularly due to health or behavioral issues, which exacerbated their suicidal feelings.	Moderate risk, Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for Qualitative Surveys
Hawkins et al., 2021(UK)	Cross-sectional survey with thematic analysis	119 adult pet owners. No age or sex data was reported.Health conditions: 41 participants had a diagnosed psychiatric disorder (depression [26.9%], anxiety [49.6%], PTSD [5%], OCD [0.8%], and schizophrenia [0.8%]), while 70 were experiencing mental health difficulties but were not formally diagnosed. No other psychiatric or physical diagnosis was specified.	Methodology: Conducted online via a 30-minute survey. Recruitment was done using social media platforms such as Facebook and Twitter.Assessment: Open-ended survey questions regarding participants' experiences with their pets and mental health, exploring both the benefits and challenges of pet ownership.Analysis: Thematic analysis was used and facilitated through NVivo software. This involved reading participants' quotes, identifying recurring themes, generating initial codes, and searching for overarching key themes.	N/A	Results: - Pets provide a protective influence by reducing suicidal thoughts. The sense of responsibility for their pets and the worry about leaving them behind were significant factors in preventing participants from attempting suicide. For some, pets gave life meaning and a "reason to live," thereby helping to mitigate suicidality.	Moderate risk, Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for Qualitative Surveys
Barcelos et al., 2021(UK)	Interview	36 dog owners.Age: 18-74 years (range).Gender: 18 males, 18 females.Health conditions: All participants had autism spectrum disorder. The majority (61.1%) were moderately to severely depressed or anxious. Mental health status was screened using the PHQ-9 and the GAD-7. No other psychiatric or physical diagnosis was specified.	Methodology: Conducted using remote interviews via an online video conferencing tool. Participants were recruited through national/regional autism organizations, social media, and personal contacts.Assessment: Semi-structured interviews adapted from a previous framework, focusing on dog-related activities' impact on participants’ hedonic and eudaimonic well-being.	N/A	Results: - Suicide Preventive Role: 16.7% of participants (6/36) reported that their dog prevented them from taking their own lives. The primary reasons cited included the dog's affection and the need to care for the animal. - Pathways to Suicide Prevention: 1) Demonstrations of Affection/Love: Dogs showing affection toward their owners (e.g., greeting them, physical touch) provided emotional support. 2) Care for the Dog: Caring for the dog (e.g., feeding, walking) gave the participants a sense of responsibility and a reason to continue living. 3) The Dog's Presence: Simply having the dog present was a protective factor. 4) Living Alone with the Dog: The idea that their dog would be left alone and suffer if they were gone was a significant deterrent to suicide for some participants.	N/A
Scanlon et al., 2021(UK)	Interview	20 homeless or vulnerably housed dog owners.Age: 23-65 years (range).Gender: 18 males, 2 females.Health conditions: No psychiatric or physical diagnosis was specified.	Methods: Conducted face-to-face.Assessment: Semi-structured interviews with open and closed questions to gather data on participants' living circumstances, experiences of accessing services, and their relationship with their dog(s).	N/A	Results: - Mutual Rescue: A theme of "mutual rescue" emerged, where participants felt that they had rescued their dogs from negative situations, and in turn, the dogs provided them with emotional support and motivation to take care of themselves. - Anticipatory Grief: Participants expressed deep concern about the possibility of losing their dogs, with some indicating that their pets were their primary reason for living. - Dogs served as a protective factor against suicidal thoughts. For example, one participant mentioned that losing their dog would likely lead them to follow a negative path, suggesting that the dog's presence had a life-saving impact.	N/A
Young et al., 2020(Australia)	Interview	12 older adult pet (dogs, cats, birds, reptiles) owners.Age: 60-83 years (range).Sex: 5 males, 7 females.Health conditions: No psychiatric or physical diagnosis was specified.	Methodology: Conducted face-to-face, mostly in the participants' homes with their pets present. Some interviews with non-metropolitan participants were conducted over the phone.Assessment: Semi-structured interviews containing open-ended questions about pet ownership, the nature of the human-animal relationship, and how pets affected their health.	N/A	Results: - 12 participants indicated some connection between their pets and suicide protection. - Function: Pets played a functional role in keeping participants active and engaged, providing direct (active care needs) or indirect (companionship) support. - Presence: The pets' mere presence offered comfort and reduced feelings of loneliness. - Known-ness: Participants felt "known" by their pets, experiencing a sense of mutual understanding and connection. - Reciprocity: The sense of a reciprocal relationship, where participants provided care and, in return, received unconditional love and companionship from their pets, which created a strong protective bond.	N/A
Wisdom et al., 2009(USA)	Interview	117 pet owners.Age: 48.8 years (mean).Sex: 52.2% women.Health conditions: All participants had serious mental illness (i.e., schizophrenia, schizoaffective disorder, bipolar disorder, or affective psychosis). No other psychiatric or physical diagnosis was specified.	Methods: Conducted face-to-face among members of a Health Maintenance Organization.Assessment: Data collection included in-depth interviews exploring mental health history and personal experiences, as well as paper-and-pencil questionnaires that gathered information about pet ownership, pet importance, recent mental health symptoms, hospitalizations, social support, and other demographic factors.	N/A	Results: - Some participants reported that pets played a crucial role in preventing suicidal ideation or behavior. For instance, one participant mentioned that the thought of their responsibility toward their pets was a motivating factor for continuing to live, stating, "I can't leave because the rabbits need me.”	N/A
Hawkins et al., 2024	Interview	16 pet owners. 6 had a dog, 6 had a cat, 4 had both a dog and cat.Age: 18-26 (range), 22 years (mean).Gender: 87.5% female.Health conditions: 12 participants had formal diagnoses of anxiety, depression, or both. Unknown number of participants had borderline personality disorder, OCD, ADHD, psychosis, autism spectrum disorder.	Methods: Conducted 1:1 online interviews or via telephone.Assessment and analysis: Semi-structured interviews to collect mental health status, professional help-seeking behaviour, pet type and ownership duration, motivation for acquiring pet. Thematic analysis was used to generate key themes from the data.	N/A	Results: - Participants reported that pets improved depression by acting as a reason to get out of bed and engage in daily life, fostering a sense of purpose. In cases of severe mental health struggles and suicidality, pets were viewed as protective factors, preventing self-harm and suicidal thoughts by offering emotional support, routine, and a strong bond. Some participants noted that since acquiring their pet, they had not experienced suicidal thoughts again.	N/A

Table 2.

Study characteristics, results, and risk of bias of the included articles on AAT and suicidal ideation and deaths (n=11). Statistically significant results are marked with * and underlined.

Author(s), year(location)	Study design	Population and health conditions	Methods	Intervention and comparator (if applicable)	Results and adverse events	Risk of bias score (green=low, yellow=moderate, red=high) and tool
Richersen et al., 2023(USA)	Randomized, two-arm, parallel-design, multicenter clinical trial	181 veterans were paired with a dog (97 with a service dog, 84 with an emotional support dog).Age: 50.6 years (mean, SD±13.6), 22-79 years (range).Gender: 80% male.Health conditions: All participants had PTSD. Most participants had comorbid conditions, including major depressive disorder (37%), panic disorder (13%), generalized anxiety disorder (7%), alcohol use disorder (4%), and substance use disorder (1%). No other psychiatric or physical diagnosis was specified.	Methods: Conducted across 3 medical centers. Veterans were randomly assigned to receive either a service dog or an emotional support dog. 18-month follow-up period.Assessment: Columbia–Suicide Severity Rating Scale (C-SSRS).Analysis: Linear mixed repeated-measures models and generalized linear mixed models were used to analyze changes over time in the primary and secondary outcomes.	Service Dog Group (intervention): Trained service dogs were paired with veterans to assist in mitigating PTSD symptoms. Emotional Support Dog Group (comparator): Emotional support dogs provided comfort but were not trained in specific PTSD-related tasks.	Results: - Both emotional support dog and service dog groups showed a reduction in suicidal behavior and ideation at follow-up, but was not statistically significant. At 18 months, the suicidal behavior and ideation rate was 28% for the emotional support dog group versus 14% for the service dog group (adjusted effect estimate=2.98, 95% CI=-1.88 to 2.09, p=0.032). Adverse events: None reported.	High risk, Cochrane's Risk of Bias 2 tool for Randomized Controlled Trials
Mikaelsen et al., 2023(USA)	Pre- and post-study	Veterans experiencing moral injury were recruited. No other participant, age, sex, or health conditions data was reported.	Methods: Conducted pre- and post assessments following 8 weeks of participation in an Equine-assisted learning (EAL) program combined with chaplain care. Sessions took place in a ranch setting where veterans interact with horses.Assessment: The pre- and post assessments focus on self-reported changes in emotional well-being, ability to connect with others, optimism, general mood, and present-moment awareness.	EAL program combined with chaplain care lasting 8 weeks, with 90-minute weekly sessions.	Results: - Suicide-Specific Findings: Some veterans reported a reduction in suicidal thoughts following participation in the EAL program, attributing their experiences with horses to this improvement. No statistical analysis was conducted. Adverse events: None reported.	High risk, NIH Study Quality Assessment Tool for Before and After Studies
Sherman et al., 2023(Australia)	Pre- and post-study	16 veterans.Age: 50.88 years (mean, SD±12.88), 34-74 years (range).Sex: 14 males (87.5%), 2 females (12.5%).Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Conducted pre- and post assessments following 3, 6, and 12 months post-matching with an assistance dog.Assessment: 4 questions related to suicidality from the National Survey of Mental Health and Wellbeing at each follow-up. Semi-structured interviews were conducted 3 months post-matching.	Assistance Dog. The matching process was personalized, considering factors such as veterans' physical abilities, verbal interaction capabilities, medical information, and the personality match between the dog and the veteran.	Results: - 34% reduction in the proportion of participants reporting suicidality from baseline to 3 months. Not statistically significant (p=0.061). Futher reductions in the proportion of participants reporting suicidality from baseline to 6 and 12 months, but % was not reported. Not statistically significant (6 months p=0.164, 12 months p=0.409). - The overall change in suicidality across time pints was not statistically significant, with an Exp(β) of 1.286 (95% CI=0.48 to 3.45, p=0.618). - Qualitative: 3 veterans explicitly mentioned the role of their dogs in reducing suicidality, providing them with a sense of purpose and responsibility. Adverse events: None reported.	Low risk, NIH Study Quality Assessment Tool for Before and After Studies
Muela et al., 2021(Spain)	Pre- and post-study	30 adolescents in residential child care under protection by authorities.Age: 15.50 years (mean, SD±1.60), 14-17 years (range).Sex: 11 males, 19 females.Suicide risk factors: 53.3% had previously made suicide attempts. 70% had suicidal ideation within the last six months. 60% had devised a suicide plan. 80% engaged in non-suicidal self-harm. 76.7% had a psychiatric diagnosis and were receiving psychopharmacological treatment.Health conditions: No psychiatric or physical diagnosis was specified.	Methods: Conducted pre- and post-assessments 2 weeks after the intervention.Assessment: SENTIA (Suicidal Behavior in Adolescents Scale) measured suicidal ideation, communication, planning, non-suicidal self-harm, and help-seeking behavior.	OverCome-AAI Program: Six 90-minute sessions held weekly over six weeks, involving emotional regulation techniques and interaction with two trained Labrador Retriever dogs. The sessions incorporated progressive muscle relaxation, mindfulness, and a suicide safety plan personalized for each participant.	Results:* - Suicidal Ideation: Significant reduction (χ² McNemar=6.75; p≤0.05). - Suicide Plans: Significant reduction (χ² McNemar=12.07; p≤0.05). - Non-Suicidal Self-Harm: Significant reduction (χ² McNemar=12.07; p≤0.05). - No participants made a suicide attempt during the program. - There was a notable increase in help-seeking behavior post-intervention (χ² McNemar=8.10; p<0.05), with 85% of those who had suicidal ideation seeking help, compared to 38.1% before the intervention. - Girls continued to show higher rates of suicidal ideation (35.3%) and non-suicidal self-harm (29.4%) compared to boys (11.1% and 22.2%, respectively) at the post-test. Adverse events: None reported.	Moderate risk, NIH Study Quality Assessment Tool for Before and After Studies
Scotland-Coogan et al., 2020(USA)	Pre- and post-study	71 veterans initially enrolled, with 55 completing the program (77% completion rate).Age: 34 participants (62%) were aged 18-49 years, 21 (38%) were aged 50-90 years.Sex: 44 males, 11 females.Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Conducted pre- and post assessments at end of intervention (14 weeks).Assessment: Trauma Symptom Inventory-2 (TSI-2) to assess PTSD-related symptoms and suicidality.	Service Dog: Veterans were taught how to train their own service dogs (either provided by the program or existing pets) over 14 weekly sessions (60 minutes). In addition to dog training, veterans had access to supportive counseling services. A social worker was available during all sessions to provide emotional support or formal therapy if requested.	Results:* - Suicidal Ideation: Significant reduction from pretest (mean=5.29, SD±4.77) to posttest (mean=2.53, SD±3.16), with t(54)=4.87, p≤0.001. - Suicidal Behavior: Significant reduction from pretest (mean=2.38, SD±3.44) to posttest (mean=0.78, SD±1.67), with t(54)=3.87, p≤0.001. Adverse Events: None reported.	Moderate risk, NIH Study Quality Assessment Tool for Before and After Studies
Whitworth et al., 2019(USA)	Pre- and post-study	32 veterans (15 in the service dog training group, 15 in the control group).Age: 50.9 years (mean, SD±16).Gender: 87% male (n=26), 13% female.Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Conducted pre- and post assessments at end of intervention (14 weeks).Assessment: Trauma Symptom Inventory-2 (TSI-2) to assess PTSD-related symptoms and suicidality.	Service Dog: Veterans participated in 60-minute weekly sessions for 14 weeks, where they were trained to work with a service dog to improve their daily functioning. Some veterans used their own companion dogs, while others were provided with dogs from the program.	Results: - Suicidality: There were reductions in suicidality, but the changes were not statistically significant (t=1.36, p>0.05). - Suicidal Ideation: There were reductions but the changes were not statistically significant (t=1.78, p>0.05). - Suicidal Behavior: There were reductions but the changes were not statistically significant (t=0.9, p>0.05). Adverse events: None reported.	Low risk, NIH Study Quality Assessment Tool for Before and After Studies
Altschuler, 2018(USA)	Case series	5 cases are mentioned, including veterans with PTSD who have undergone AAT.Age and sex: Not explicitly detailed for all cases; predominantly male veterans.Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Compilation of case reports in various media outlets, describing veterans undergoing AAT, including service dogs (n=2), equine therapy (n=2), and working with parrots (n=1).Assessment: Not explicitly mentioned; suicidality is discussed in relation to anecdotal reports from veterans.	N/A	Results: - Service Dogs: One veteran credited his dog with preventing a suicide attempt by interrupting him during the attempt. - No descriptions relating to suicide for Equine Therapy or Parrot Therapy.	High risk, NIH Study Quality Assessment Tool for Case Series
Lloyd et al., 2019(Australia)	Cross-sectional survey	199 participants out of 600 eligible clients of “mindDog” responded to the survey.Age: 47 years (median), 10-75 years (range).Gender: 77% female.Health conditions: depression (84%), anxiety (social 61%, generalized 60%), PTSD (62%), panic attacks (57%). Other diagnoses without % included Obsessive-compulsive disorder, autism spectrum disorder, and eating disorders. No other psychiatric or physical diagnosis was specified.	Methods: Online survey were administered to participants who are registered with the charity “mindDog,” an Australian organization that helps people with mental health conditions acquire and train psychiatric assistance dogs (PADs).Assessment: The study used an online questionnaire with multiple-choice, forced-choice, binary (yes/no), and free-text responses.Analysis: Chi-square tests for independence were used for statistical analysis.	N/A	Results: - 46% of participants reported a decrease in the use of psychiatric or healthcare services, attributed to a reduction in suicide attempts, hospitalizations, and medication use. 30% of participants reported an increase in service use due to their improved ability to attend appointments, with the PAD providing confidence to venture outdoors. 24% reported no change. - Reduced suicide attempts were reported as a significant reason for decreased healthcare utilization. - No significant associations were found between participants' mental health diagnoses and the tasks performed by their PADs, nor between diagnosis and changes in health service utilization.	Moderate risk, Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices for Qualitative Surveys
Yarborough et al., 2018(USA)	Interview	41 veterans with service dogs, plus 8 caregivers and 6 dog trainers.Age: 44.7 years (mean, SD±11.9), 24-66 years (range).Sex: 31.7% female.Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Conducted in person and over the phone. Observational data were collected during training sessions.Assessment and analysis: Semi-structured interviews were used to assess the veterans' experiences. Thematic analysis was employed to analyze the interview data.	N/A	Results: - Service dogs were reported to reduce suicidal impulses by giving veterans a sense of purpose and requiring them to take care of the dog. (e.g., one veteran noted that the service dog helped prevent a suicide attempt, providing companionship and a reason to get out of bed.	N/A
McLaughlin et al., 2019(Australia)	Interview	7 veterans with service dogs.No age data was reported.Sex: 6 males, 1 female.Health conditions: All participants had PTSD. No other psychiatric or physical diagnosis was specified.	Methods: Conducted in person with participants recruited from a veteran support organization that trains service dogs.Assessment and analysis: Two semi-structured focus groups to collect qualitative data. Thematic analysis was used to generate key themes from the data.	N/A	Results: - Several participants reported that their service dogs directly contributed to preventing suicidal ideation or actions. The dogs provided distraction, a sense of responsibility, and emotional support during crises.	N/A
Pavelich et al., 2024(Canada)	Interview	5 military veterans. 4 completed the study.Age: 43 (average), 36-51 (range).Sex: 5 males.Health conditions: All participants had PTSD and self-identified substance use concerns.	Methods: Conducted in person with participants recruited from a service dog training program.Assessment and analysis: Semi-structured interviews conducted over 18 months (baseline, months 1, 3, 6, 9, 12, and 18) to collect qualitative data. Thematic analysis was used to generate key themes from the data.	N/A	Results: - Service dogs were reported to reduce suicidal thoughts by providing unconditional, non-judgmental support seen as distinct from human relationships. - The dogs reportedly provided PTSD symptoms relief, reduced depression and hopelessness, and decreased substance use. - Being a service dog client increased social interactions and reintegration to society via dog-walking responsibilities, as well as connection to a supportive community within the service dog training program.	N/A

Pet ownership

11 articles assessed the impact of pet ownership on SI, including one cohort studies,³⁸ one case series,³⁹ and nine surveys or interviews.^{5,21,23,25,26,28,29,31,32} Three articles assessed the impact of pet ownership on suicide deaths, including one cohort study,³⁷ one case-control study,⁴³ and one case report.⁴¹ No articles assessed SA.

Observational studies

One observational study examined SI. Barcelos and colleagues³⁸ conducted a prospective cohort study of 709 dog owners and found a positive correlation between time spent with dog and SI (r=0.17, p<0.001), with owners spending more time with their dogs if they reported thinking about suicide. They identified that failure to fulfill dog-related responsibilities, the presence of aggressive or fearful dog behavior, and the owner's lack of control over the dog, were associated with increased feelings of loneliness, depression, and anxiety.

In terms of suicide deaths, Batty and Bell,³⁷ in their prospective cohort study that included 47 suicide deaths out of 67,441 pet owners, found no significant association between pet ownership and suicide risk after adjusting for various covariates (HR=1.0, 95% CI 0.5 to 1.7). These covariates included education, presence of mental health problems, presence of somatic illness, cigarette smoking status, and marital status. The type of pet owned was also considered, with dog (HR=1.3, 95% CI 0.7 to 2.6), cat (HR=0.3, 95% CI 0.1 to 1.1), and other pets (HR=1.5, 95% CI 0.6 to 3.9) having no significant association with suicide risk. Similarly, Helsing and Monk⁴³ also did not find significant associations between pet ownership and suicide (OR=1.00, 95% CI 0.50 to 2.01) in their multi-year census-based case-control study of 48 people who died by suicide. The type of pet owned was also not significantly associated with suicide, namely dogs (OR=0.95, 95% CI 0.20 to 5.93) and cats (OR=1.22, 95% CI 0.37 to 4.00). Further subgroup analyses also found no significant associations with suicide, including among men (OR=1.06, 95% CI 0.46 to 2.46), women (OR=0.88, 95% CI 0.06 to 13.7), or those living in non-farm residences (OR=1.19, 95% CI 0.24 to 5.93).

Case series and reports

One case series examined SI. Hutton³⁹ presented a case series of five pet owners living with HIV who self-reported strong attachments to their pets that acted as buffers to SI due to concerns about their pets’ potential distress upon their death.

With regard to suicide deaths, Cooke⁴¹ presented a case report of a middle-aged woman with a possible history of mild depression who died by “extended suicide” (homicide-suicide in which the individual who ends their own life also kills one or more others) with her pet dog, who had previously served as a self-reported protective factor against suicide, suggesting that attachment to pets can, in some cases, serve as a complicating factor in suicide risk assessment.

Survey and interviews

Smith and colleagues²⁹ found in a survey of 1087 veterinarians that pet owners, in general, did not significantly differ in SI compared to non-pet owners (34.0% vs. 32.2%, no F statistic or p-value reported). However, owning certain types of pets was associated with varying mental health outcomes and SI. Dog owners, specifically, had lower anxiety scores on the GAD-7 (F[1, 918]=5.56, p=0.019) and SI (χ²(1)=5.04, p=0.025) compared to non-pet owners, while cat owners reported more depression (F[1, 920]=4.37, p=0.037) and SI (χ²(1)=10.52, p=0.001). Thematic analyses of 119 and 71 adult pet owner responses by Hawkins et al.³² and Love,³¹ respectively, identified various psychological benefits of pet ownership that contributed to the participants' reduced SI, including increased hedonic tone, comfort, distraction, motivation, behavioral activation, social connections, and reduced anxiety symptoms. However, some participants also reported increased negative feelings and emotional strain associated with pet ownership, which exacerbated their SI. Echoing these findings, qualitative interviews of adult pet owners by Barcelos et al. (n=36),²¹ Scanlon et al. (n=20),²³ Young et al. (n=12),²⁵ Wisdom et al. (n=117),²⁶ and Hawkins et al. (n=16)²⁸ emphasized the protective role of pets in suicide prevention, particularly in reducing SI via the provision of companionship, empathy, and a sense of purpose or belonging. Furthermore, Douglas and colleagues⁵ found in a survey of 187 undergraduate student pet owners that overall attachment to pets was negatively associated with perceived burdensomeness (PB), which was in turn associated with lower suicide risk (point estimate=-0.45, 95% CI -0.91 to -0.04). However, overall attachment to pets was not associated with thwarted belongingness (TB)–another factor associated with suicide risk (point estimate=-0.23, 95% CI -0.65 to 0.07). They also found that attachment anxiety was indirectly associated with suicide risk via PB (point estimate=0.04, 95% CI 0.02 to 0.06) and TB (point estimate=0.03, 95% CI 0.01 to 0.05), while attachment avoidance was indirectly associated wih suicide risk via TB (point estimate=0.03, 95% CI 0.01 to 0.06) but not PB (point estimate=0.02, 95% CI -0.01 to 0.05).

Animal-assisted therapy

11 articles assessed the impact of AAT on SI, including one RCT,⁴² five pre- and post-studies,^6,33–36 one case series,⁴⁰ and four surveys or interviews.^22,24,27,30 No articles assessed SA or deaths.

Experimental studies

In a pre- and post-study, Sherman and colleagues³⁵ demonstrated that 16 veterans with PTSD paired with assistance dogs (participating in 14 weekly sessions, 60 minutes in length) experienced a 34% reduction in the proportion of participants reporting SI from baseline to 3 months, however this reduction was not statistically significant (p=0.061). Qualitatively, three out of 16 veteran participants attributed their dogs with suicide prevention by providing comfort and purpose. However, their study did reveal a 50% reduction in the number of participants meeting PTSD diagnostic criteria over the 12-month period (Exp(β)=0.207, 95% CI=0.07 to 0.66). Similarly, Scotland-Coogan and colleagues³⁴ noted significant decreases in PTSD-related symptoms (t(53)=5.38, p<0.001) and SI (t(54)=4.87, p<0.001) among 71 veterans in a service dog training program, although older participants (those over age 50; n=21) did not show significant improvements in SI or behavior (p=0.059), potentially due to a floor effect, since initial scores on these scales were low in this group. In contrast, Whitworth and colleagues³³ found no significant impact on SI (t=1.78, p>0.05) and behaviors (t=0.9, p>0.05) among 15 veterans in a similar service dog program compared to a waitlist control group. In a pre- and post-study of veterans with self-identified moral injury (defined by the authors as an injury to a person's trust in self, others, community, or God, which leads to symptomatic features such as guilt, shame, spiritual/existential confict, depression, and a hypervigilance to reoccurring betrayal), Mikaelsen and colleagues³⁶ reported that chaplain care and equine-assisted interventions improved self-reported stress management, resilience, community engagement, and suicidal thoughts at 8 week follow-up. However, the sample size, formal psychiatric diagnoses, and other descriptive statistics were not reported.

Beyond pre- and post-studies, Richerson and colleagues⁴² conducted an RCT that reported no significant differences in suicidal behaviors or ideation between veterans with PTSD after 18 months of being paired with a service dog (intervention; n=97) or emotional support dog (control; n=84) (adjusted effect estimate=2.98, 95% CI=-1.88 to 2.09, p=0.032), where the service dog group showed a reduced rate of SI (14% vs. 28%). Among the AAT experimental studies, only one pre- and post-study by Muella and colleagues⁶ examined an intervention aimed at reducing suicidal behaviors in non-veterans, specifically in adolescents. They found that 30 adolescents with high suicide risk who underwent six weekly sessions of group-based dog-assisted therapy experienced significant reductions in SI (χ² McNemar=6.75; p<0.05), suicide plans (χ² McNemar=12.07; p<0.05), and self-harm (χ² McNemar=12.07; p<0.05), with 85% of participants requesting help during a crisis post-intervention compared to 38.1% before the intervention (χ² McNemar=8.10; p<0.05). Importantly, some participants reported feeling more confident in seeking help for suicidal thoughts and engaging in non-suicidal self-harm.

Surveys, interviews, and case series

Among 199 surveyed psychiatric assistance dog users living with a variety of psychiatric disorders (e.g., depression, anxiety, PTSD, panic attacks, OCD, autism spectrum disorder, and eating disorders), reductions in the use of healthcare services were reported by 46% of participants, attributed mainly to decreased SAs, hospitalizations, and medication requirements.³⁰ Articles focusing on veterans with PTSD consistently highlighted positive outcomes over time, with participants reporting improvements in sleep, increased community participation, and decreased reliance on medications for mental and physical health, alongside a sense of responsibility towards their service dogs, which served as deterrents to suicide and distractions from SI.^22,24,27,40 None of the described descriptive and case series studies provided quantitative evidence or statistical analysis relating to SI beyond sample size.

Risk of bias

Risk of bias was assessed for all articles except interviews and case reports. Overall, 5/17 articles were determined to have a low risk of bias, 7/17 a moderate risk, and 4/17 a high risk (Tables 1 and 2). Among the experimental studies, 2/5 pre- and post-studies were assessed as low risk, 2/5 as moderate risk, and 1/5 as high risk. The single RCT was rated as high risk. Among the observational studies, 1/2 cohort studies were determined to have a low and moderate risk of bias each, while the single case control study was also rated as low risk. Among the descriptive studies, 4/5 survey studies were found to have a moderate risk, with 1/5 rated as high risk. Finally, within the case series, 1/2 studies was determined to have a low risk, and the other high risk.

Discussion

This scoping review identified 25 articles exploring the impact of pet ownership or AAT on suicide-related outcomes, specifically, SI and suicide deaths. Pet ownership was evaluated in 14 studies, with a range of pet types, primarily dogs and cats, while AAT was investigated in 11 studies, predominantly among veterans with PTSD. Suicide deaths was assessed by three articles (two observational studies and one case report) in the context of pet ownership only.

The evidence regarding pet ownership's impact on suicide-related outcomes is heterogeneous. Quantitative findings were inconsistent. All observational studies included in the review failed to establish statistically significant associations between pet ownership and suicide risk with the exception of one cohort study, which unexpectedly found that the time spent with dog by the owner was significantly correlated with SI.³⁸ This finding may reflect the increased reliance on dogs as emotional support among individuals with poor mental health, coupled with heightened awareness of negative aspects of the dog's behavior or health, potentially reinforcing distress in a bidirectional relationship. This hypothesis aligns with qualitative findings that highlighted the emotional strain of caregiving responsibilities and distress related to pet health or behavioral issues. However, an alternative explanation could involve reverse causality: individuals with greater SI might tend to isolate themselves socially and focus more on caring for their dog, as a form of emotional coping. In this case, the increased time spent with the dog could reflect an intensified bond due to the individual's emotional state, rather than being a protective factor. These explanations suggest that the relationship between pet ownership and SI might be more complex than initially thought. Importantly, this study did not control for confounding variables such as sociodemographic factors, physical health conditions, and psychiatric conditions, contributing to its increased risk of bias. Additionally, the two observational studies that examined pet ownership and suicide deaths did not find a statistically significant relationship.^37,43 Descriptive studies, however, often highlighted the owner-reported protective role of pets, including companionship to mitigate loneliness, a sense of responsibility or purpose, engagement in pet-care activities as distractions or forms of behavioral activation, and facilitation of social connections; these factors have the potential to deter SI. Dogs, in particular, emerged as notably beneficial, often associated with reduced anxiety and improved psychological well-being. However, this finding reflects a bias in the literature, as most studies predominantly focused on dog ownership. Only one descriptive study compared the influence of different species, such as cats and horses.²⁹ Collectively, these mixed findings suggest that pet ownership alone may not universally reduce SI or prevent suicide.

The evidence regarding AAT's impact on suicide-related outcomes is also heterogeneous. Descriptive studies highlighted similar perceived benefits against SI as did descriptive studies of pet ownership. The mechanisms underlying these benefits are thought to include enhanced emotional regulation, reduced PTSD symptoms, increased help-seeking behaviors, and the establishment of purposeful routine–factors that mirror the benefits reported by pet owners. Experimental studies largely yielded non-significant results, with only one pre- and post-study reporting a significant reduction in SI among veterans with PTSD³¹ and another pre- and post-study finding significant reductions in SI, plans, and self-harm, alongside improved crisis-help-seeking behaviors among adolescents with high suicide risk.⁶ The significant results highlight AAT's potential as an adjunct to traditional suicide prevention strategies, especially in populations experiencing significant emotional or social isolation. However, given the considerable number of studies with non-significant findings, conclusions cannot be drawn regarding AAT's efficacy in suicide prevention. Furthermore, challenges were reported regarding the use of AAT, including difficulties related to animal training, perceived increase in public attention, and financial burdens.^21,23 Importantly, these challenges were not linked to increased SI.

A recurrent theme among the included articles is that a person's attachment style plays a pivotal role in shaping the effectiveness of both pet ownership and AAT in reducing SI. Individuals with secure attachment styles may derive significant psychological benefits from human-animal interactions, viewing pets as a "safe haven" or "secure base" that provides unconditional love, emotional stability, and companionship, especially in the absence of strong external social support. Conversely, those with insecure attachment styles, such as attachment anxiety or avoidance, may experience more complex and sometimes detrimental effects. For example, individuals with attachment anxiety may rely heavily on animals for emotional support but also experience heightened distress, perceived burdensomeness, and thwarted belongingness, which contribute to their increased risk for SI. Similarly, individuals with avoidant attachment styles may struggle to derive a sense of belonging or support from their pets due to their self-reliant nature and distrust of attachment relationships. These dynamics highlight the need for individualized assessment when considering the therapeutic potential of pets or AAT, as the human-animal bond can act as either a protective or exacerbating factor depending on the individual's attachment characteristics and mental health context.

Strengths and limitations

This scoping review offers several strengths. It is the first PRISMA-compliant review on the influence of pet ownership and AAT in SI and suicide deaths. The review included a broad scope of interventions (pet ownership and AAT) and suicide-related outcomes (ideation, attempts, and death; despite yielding no results for attempts), which allowed for the identification of trends and gaps across different populations and contexts. The title/abstract and full-text screening, along with the data extractions, were performed independently by two reviewers, which enhances the reliability of the study. By including records in any language, we ensured that valuable information was not overlooked to provide a comprehensive review of the available literature.

Despite its strengths, this review has several limitations. One key limitation is the methodological diversity of the included studies. Few experimental studies specifically examined AAT, and none addressed the impact of pets directly. Most of the studies were observational or descriptive in nature, limiting the ability to draw causal conclusions between pet ownership or AAT and reductions in suicidality. Furthermore, the mixed quality of the studies is a significant limitation, as many of the included studies were of low quality, which undermines the strength of the evidence. Another limitation is the skewed representation of certain populations, types of animals, and suicide outcomes. AAT studies predominantly focused on veterans with PTSD, which may limit the generalizability of findings to other high-risk groups, such as adolescents, the elderly, or individuals with other mental disorders. Similarly, most studies focused on dogs as the primary animal, which leaves a gap in understanding the role of other animals in suicide prevention efforts. Most articles focused on SI, with only three examining suicide deaths, and those three were all in the context of pet ownership. Additionally, there was a lack of standardization in both the interventions and the ways SI was measured across studies. This variability makes it difficult to compare results across studies. Finally, the mental health status and sample sizes of participants were often insufficiently documented, preventing the ability to draw transdiagnostic insights regarding how different mental disorders may affect the impact of pet ownership or AAT on SI and deaths.

Conclusion

This scoping review provides an overview of the evidence regarding the role of pet ownership and AAT in SI and suicide deaths. While experimental and observational studies offer inconclusive results on the impact of pet ownership and AAT on SI and deaths, qualitative studies consistently highlighted user-perceived positive outcomes. Despite some promising findings, the methodological limitations, including the predominance of descriptive studies, a lack of standardization, and mixed quality of studies, highlight the need for further research to clarify the potential of pet ownership and AAT in suicide prevention. Future research should focus on conducting high-quality with stringent methodology, large-scale, experimental studies that include diverse populations and investigate the effects of different animal species. Control of confounding variables through regression analysis should be conducted where possible. A greater focus on suicide-related outcomes other than ideation (i.e., attempts and deaths) would be fruitful. Standardizing intervention protocols and outcome measures will be essential for making meaningful comparisons across studies, ultimately faciliating the clinical integration of animal-assisted approaches in suicide prevention and informing policy decisions. Until more definitive evidence emerges, the use of pets and AAT in suicide prevention should be considered on a case-by-case basis, taking into account the individual's mental health condition, capacity to care for a pet, and attachment style.

Funding sources

None.

Ethics approval and consent to participate

Not applicable. No human or animal participants.