Animal-assisted therapy on psychological and physical outcomes: a meta-analysis of randomised controlled trials

Why clinicians should know about it

• Picked for Anesthesiology and Pain Medicine (top studies of the week, 19 July 2026).
• Picked for Health Policy (top studies of the week, 19 July 2026).
• Picked for Public Health, Environmental and Occupational Health (top studies of the week, 19 July 2026).
• Picked for Psychiatry and Mental Health (top studies of the week, 19 July 2026).

Abstract

BACKGROUND: Adults are at heightened risk of anxiety, stress, and depression; animal-assisted therapy (AAT) may serve as an effective approach to promote psychological well-being. In this study, we compared the effectiveness of AAT in improving depression, anxiety, stress, pain, and gait among adults with or without illness. METHODS: We systematically searched six electronic databases (PubMed, CINAHL, Embase, Web of Science, the Cochrane Library, and Scopus) and included all studies published up to August 2024. We used comprehensive meta-analysis software to complete the quantitative synthesis. We reported pooled effect sizes as Hedges' g with corresponding 95% confidence intervals (CIs), after applying a random-effects model. Furthermore, we assessed heterogeneity using Cochran's Q test and the I2 statistic. We applied the Cochrane Risk of Bias 2.0 tool to appraise the methodological quality of the included studies. The synthesis process followed PRISMA guidelines. RESULTS: From the 13,345 studies identified, 35 randomised controlled trials involving 2391 adults were included. Across diverse populations, AAT was associated with reductions in depression (Hedges' g = -0.403; 95% CI = -0.536, -0.271), anxiety (Hedges' g = -0.661; 95% CI = -1.069, -0.253), and stress (Hedges' g = -1.062; 95% CI = -1.849, -0.275) at post-intervention, although substantial between-study variability was observed. CONCLUSIONS: We demonstrated that AAT significantly improves anxiety, depression and stress in adults, but has no meaningful effect on pain or gait. Subgroup and meta-regression analyses indicate that psychological benefits depend on population and intervention characteristics, and are not moderated by age or gender. REGISTRATION: PROSPERO: CRD42024570108.

Abstract as published, via PubMed.