The aim of the study was to investigate the interaction process between child and dog and how it possibly affects children's wellbeing during Animal Assisted Activity.

Children have reported negative feelings such as fear and anxiety when being cared for in hospital and various kinds of complementary treatment can alleviate this. Different complementary treatments, including interaction with a dog, can create positive emotions and the treatment has been reported to have both physiological and psychological beneficial effects. However, there is a lack of studies describing children's interaction with a dog.

This is an observational study, analysed from field notes with qualitative content analysis using a deductive approach. Children (n = 49) aged 3–18 years of age at a paediatric hospital voluntarily participated in the study.

The results are reported on a six-level scale that describes the child-dog interaction: 1. Passive interaction, 2. One-way non-spoken communication, 3. Facilitating the interaction, 4. Interaction by activity encouragement, 5. Interaction initiated by the child, and 6. Interaction through deepened interplay. All children attained level five. Eighty-nine per cent attained level six and these children interacted fully, having a two-way deepened interplay with the dog. Further, when the interaction proceeded to a deepened interplay this affected the children positively both physically and emotionally.

Structured Animal Assisted Activity with a dog that includes an introduction, an active part and a relaxing part is a suitable model to offer children in paediatric hospital care since the children attained a child-initiated interaction or interaction through deepened interplay.