Abstract
Adult intensive care unit (ICU) patients experience significant anxiety, distress, and isolation. Pet therapy or animal-assisted interventions, often delivered by certified therapy dogs, are used in hospitals to improve comfort and engagement, but ICU-specific evidence remains limited. This scoping review maps the recent evidence on pet therapy in adult ICUs, with the aim of describing how it is used, what benefits are reported, what risks and barriers exist, and where the literature remains thin. Peer-reviewed publications from 2015 to 2025 were found through major biomedical databases (including PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, and Scopus) that reported live animal interactions in adult ICUs, along with any relevant reviews and implementation reports. The studies were independently screened according to the inclusion and exclusion criteria, and a total of 15 studies met eligibility criteria. Across the studies, patients showed immediate reductions in self-reported anxiety after visits, families often reported parallel relief, and satisfaction with the experience was high among patients, families, and staff. Pain scores improved modestly in some cohorts, while heart rate, blood pressure, and respiratory rate were generally unchanged, suggesting psychological benefit without physiologic destabilization. Pet therapy programs were feasible in the ICU when supported by clear policies and coordination, although common barriers included infection prevention concerns, scheduling and staffing needs, space constraints, and the limited availability of controlled data. The evidence was insufficient to determine effects on delirium, sedative exposure, length of stay, or mortality. Overall, pet therapy in adult ICUs appears feasible, well-accepted, and safe under structured protocols, with consistent short-term relief of anxiety and perceived distress and potential support for engagement in rehabilitation. Future work should include multicenter controlled studies, formal safety surveillance, standardized outcome sets that include delirium and longer-term psychological recovery, cost and implementation analyses, and evaluation of robotic or virtual animal options when live visits are not possible.