Salivary Cortisol in Children with Autism Spectrum Disorder : Longitudinal Variations in the Context of a Service Dog’s Presence in the Family.

Background: The primary function of service dogs for children with autism spectrum disorders (ASD) is to ensure the safety of the child. Several parents report that the mere presence of the dog encourages the family to follow a lifestyle typical of an average family in addition to providing a source of support. A preceding study suggested that the presence of a service dog in the family induces a significant and continuous decline of cortisol secretion at awakening for children with ASD over a period of four weeks (Viau et al., 2010). This reduction of stress measured by salivary cortisol would confirm parents' perception that their child is calmer when the dog is present (Harwood, 2018). Following this first study, our research group has established a new research design involving a larger sample of families of children with ASD, including a control group, and repeated daily measurements.

Objectives: The aim of the present study seeks to evaluate services dog's impact on a group of children with ASD, based on their well-being measured by salivary cortisol. The study also estimated if age and the severity of symptoms related to autism affected variations in the children stress system.

Methods: The sample was composed of 95 families of children with ASD, aged between 5 and 11 years old (20.7% of girls). 48 of these families received a service dog (certified and trained by the Mira Foundation). The remaining 47 families were recruited from a waiting list and composed the control group (WLC). Autism severity was estimated by the Childhood Autism Rating Scale (Schopler et al., 1988) filled out by an evaluator during the first three weeks of the study after a home visit. Effects of the service dog were evaluated by repeated measures of children's salivary cortisol. These weekly home-based samples were collected twice daily by a parent (wakening and bedtime). The cortisol analyses were based on aggregation of samples over a period of 15 consecutive weeks, yielding five blocks of data each for wakening and bedtime samples; a first block composing the baseline (T0) and four once the family received the dog (T1, T2, T3, T4).

Results: No significant relationship between diurnal cortisol, autism severity and gender was detected. As expected, temporal stabilities in wakening and bedtime cortisol across the last 4 blocks were higher for the control group compared to children who received a dog. Overall, salivary cortisol across the different periods do not differentiate the two groups. However, with a median split based on age, significant differences between the two groups at T1 and T4 are present only among children aged 7 to 11 who received a service the dog (t = 2.1 and 2.3, p < .05).

Conclusions: Cortisol regulation showed no association to autism severity nor gender. Cortisol remained stable for children of the WLC. Effects of the service dog on children's stress were present for a subgroup of children, demonstrating long lasting effects of the service dog for older children.