Equine assisted interventions (EAIs) are recently facing an increasing popularity, and are characterized by a wide diversity of practices. However, information on the welfare of animals involved in this kind of activity is often lacking. Horses are highly susceptible to work stressors related to physical constraints and/or to the need to control emotions while interacting with humans. Considerations of the emotional state of horses involved in EAIs have multiple valences: for the safety of humans and animals involved, for the quality and efficacy of interventions, as well as for ethical reasons. The aim of this unsystematic narrative review is to summarize the different approaches used for the evaluation of horses’ stress responses, investigate their application in the context of EAIs, and discuss some methodological considerations for researchers and practitioners involved in EAI. The sources of information are mostly based on electronic databases (i.e., Medline, Scopus and Google scholar), as well as on hand searches of the references of retrieved literature, and discussions with experts in the field. At present, a few studies have investigated horses’ stress responses during EAIs, and further studies are recommended, with the final aim to derive a reliable multidimensional method for assessing a horse’s reaction during therapeutic programs, ultimately helping professionals to better develop interventions by taking into consideration the animal’s perspective.

Drug addiction is a major care and safety challenge in prison context. Nowadays, rehabilitation and specific therapeutic programs are suggested to improve health and well-being of inmates during their detention time and to reduce substance abuse relapse after release from prison. Among these programs, several studies reported the benefits for inmates coming from animal assisted interventions. In this pilot controlled study, we investigated the efficacy of a dog assisted therapy program addressed to 22 drug addicted male inmates housed in an attenuated custody institute in Italy. The study lasted six months, the treated group (12 inmates) was involved once a week for one hour in 20 dog assisted therapy sessions, whereas the control group (10 inmates) followed the standard rehabilitation program. One week before the beginning and one week after the end of the sessions, all inmates involved were submitted to symptom checklist-90-revised and Kennedy axis V. Inmates involved in the dog assisted therapy sessions significantly improved their social skills, reducing craving, anxiety and depression symptoms compared to the control group. Despite the limitation due to the small number of inmates enrolled and to the absence of follow up, we found these results encouraging to the use of dog assisted therapy as co-therapy in drug addicted inmates rehabilitation programs, and we claim the need of more extensive study on this subject.