The therapeutic application of human–animal interaction has gained interest recently. One form this interest takes is the use of service dogs as complementary treatment for veterans with Post-Traumatic Stress Disorder (PTSD). Many reports on the positive effect of PTSD Service Dogs (PSDs) on veterans exist, though most are indirect, anecdotal, or based on self-perceived welfare by veterans. They therefore only give a partial insight into PSD effect. To gain a more complete understanding of whether PSDs can be considered an effective complementary treatment for PTSD, a scoping literature review was performed on available studies of PSDs. The key search words were ‘dog’, ’canine’, ‘veteran’, and ‘PTSD’. This yielded 126 articles, of which 19 matched the inclusion criteria (six empirical studies). Recurrent themes in included articles were identified for discussion of methodology and/or results. It was found that results from most included studies were either applicable to human–animal interaction in general or other types of service animals. They therefore did not represent PSDs specifically. Studies which did discuss PSDs specifically only studied welfare experience in veterans, but used different methodologies. This lead us to conclude there is currently no undisputed empirical evidence that PSDs are an effective complementary treatment for veterans with PTSD other than reports on positive welfare experience. Additionally, the lack of development standardization and knowledge regarding welfare of PSDs creates risks for both human and animal welfare. It is therefore recommended that a study on the effect of PSDs be expanded to include evaluation methods besides self-perceived welfare of assisted humans. Future studies could include evaluations regarding human stress response and functioning, ideally conducted according to validated scientific methodologies using objective measurement techniques to identify the added value and mechanisms of using PSDs to assist treatment of PTSD in humans.

The Monash Dog–Owner Relationship Scale (MDORS) is a questionnaire that is used to evaluate the perceived relationship between humans and their dog. This questionnaire was originally only formulated and validated in English, which limits its use among non-English speaking individuals. Although a translation could be made, the translation of questionnaires without additional validation often impairs the reliability of that questionnaire. Therefore, the aim of this study was to validate a translation of the MDORS that is suitable for use among native Dutch speakers. To achieve this, a Dutch translation of the MDORS was made and checked for spelling/grammar mistakes, readability, feasibility, and clarity. A test–retest comparison was subsequently performed on the translation together with a calculation of Cronbach’s alpha score and principal component analysis (PCA). Through the PCA, we found that the three-factor model of the original MDORS was also largely present in the Dutch translation. However, deviations were also found, as several questions did not achieve high PCA scores in their original factor. Therefore, we propose that these questions are excluded from the Dutch MDORS.

Only a few studies have investigated the welfare of animals participating in animal-assisted interventions (AAIs). Most of these studies focus on dogs in therapeutic settings. There are, however, also dogs—service dogs—that are employed to continuously support a single human. Because the welfare of these service dogs is important for the sustainability of their role, the aim of this study was to investigate their stress response to service dog training sessions. To do this, we took repeated salivary cortisol samples from dogs who participated in a training session (n = 19). Samples were taken just after arrival at the training ground, before training, after training, and after a period of free play. Our results showed that mean cortisol levels in all samples were relatively low (between 1.55 ± 1.10 and 2.73 ± 1.47 nmol/L) compared to similar studies. Analysis further showed that samples taken before and after participation in the training’s session did not differ from one another. Mean cortisol levels in both situations were additionally lower than those upon arrival at the training site and after a period of free play. This led to the conclusion that the dogs in our study did not seem to experience training as stressful.

The fourth author in this published Review Article is incorrectly displayed. The correct name of the fourth author is Bas Rodenburg. The original version has been corrected to reflect this change.

No abstract