There is a perception in the scientific and general communities that hospitalized children benefit from visits by animals. Animal-assisted interventions (AAI), including animal-assisted therapy and animal-assisted activities, usually involving dogs, are thus employed in pediatric hospitals. However, the actual prevalence of AAI in children's hospitals has been poorly documented in the literature. Furthermore, the evidence base for claims that children in hospital benefit from AAI is limited. There are nine existing research studies in the area, all with methodological challenges that make conclusive statements in either direction about the efficacy of AAI difficult. In this critical review we consider methodological considerations pertinent to evaluations of AAI interventions for hospitalized children. These include: definitions and terminology; cultural attitudes; children's receptivity to animals, including phobia, type of illness and health status of the child, familiar as opposed to unknown animals, and age of the child; animal welfare; zoonoses and allergies; and hospital staff attitudes toward AAI. We highlight the many difficulties involved in conducting research on AAI in pediatric settings. Given the limited information around AAI for hospitalized children, including the risks and benefits and the limitations of existing studies, future research is required. This should take into account the methodological considerations discussed in this review, so that our knowledge base can be enhanced and if and where appropriate, such interventions be implemented and rigorously evaluated. ; Anna Chur-Hansen, Michelle McArthur, Helen Winefield, Emma Hanieh and Susan Hazel