- Email: [email protected]
Effect of service dogs on salivary cortisol secretion in autistic children
Psychoneuroendocrinology (2010) 35, 1187—1193
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n
Effect of service dogs on salivary cortisol secretion in autistic children `ve Arsenault-Lapierre b, Ste ´phanie Fecteau a, Robert Viau a, Genevie ¨l Champagne a, Claire-Dominique Walker c, Sonia Lupien b,* Noe a
´bec, Canada Fondation MIRA, 1820 rang Nord Ouest, Sainte-Madeleine, Que Center for Studies on Human Stress, Mental Health Research Centre Fernand Seguin, Hospital Louis-H. Lafontaine, ´ de Montre ´al, 7331 Hochelaga, Montre ´al, Que ´bec, Canada Universite c ´bec, Canada Douglas Mental Health University Institute, McGill University, Que b
Received 1 June 2009; received in revised form 3 February 2010; accepted 4 February 2010
KEYWORDS Salivary cortisol; Autism; Cortisol Awakening Response; Service dogs
Summary Children with Autism Syndrome Disorders (ASDs) exhibit social, communicative, and behavioral deficits. We know that human interaction with dogs, which is thought to serve as a social catalyst, results in a decrease of cortisol levels in healthy adults. Introducing service dogs to children with ASD is an attractive idea that has received growing attention in recent decades. However, no study has measured the physiological impact of service dogs on these children. Therefore, the goal of our study was to assess the effects of service dogs on the basal salivary cortisol secretion of children with ASD. We measured the salivary cortisol levels of 42 children with ASD in three experimental conditions; prior to and during the introduction of a service dog to their family, and after a short period during which the dog was removed from their family. We compared average cortisol levels and Cortisol Awakening Response (CAR) before and during the introduction of the dog to the family and after its withdrawal. We found that the introduction of service dogs translated into a statistically significant diminished CAR. Before the introduction of service dogs, we measured a 58% increase in morning cortisol after awakening, which diminished to 10% when service dogs were present. The increase in morning cortisol jumped back to 48% once the dogs were removed from the families ( p < 0.05). However, service dogs did not have an effect on the children’s average diurnal cortisol levels. These results show that the CAR of children with ASD is sensitive to the presence of service dogs, which lends support to the potential behavioral benefits of service dogs for children with autism. # 2010 Elsevier Ltd. All rights reserved.
1. Introduction
* Corresponding author. Tel.: +1 514 251 4015x2337; fax: +1 514 251 4014. E-mail address: [email protected] (S. Lupien).
Autism Spectrum Disorders (ASDs) are a neurodevelopmental disorder classified under the Pervasive Developmental Disorder (PDD) diagnosis of the DSM-IV-TR (APA, 2000). ASD typically refers to autism, Asperger syndrome and Pervasive
0306-4530/$ — see front matter # 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2010.02.004
1188 Development Disorder Not Otherwise Specified (PDDNOS). It is characterized by symptoms spanning the realm of social interactions, communication, and restricted or repetitive behaviors. Although the prevalence of ASD has risen over the years, making autism one of the most common neurodevelopmental disorders, this increase may be due to modifications to the diagnostic criteria and a better understanding of these disorders (Fombonne, 2003). Despite the improved understanding of ASD, whereby children are now placed in regular schools and enjoy better supervision, these children still face exclusion from their social environments due to the social, communicative, and behavioral nature of their disorder. Rehabilitation intervention, such as zootherapy, has been proven useful in helping individuals with visual, physical, intellectual, or psychiatric deficiencies (Eddy et al., 1988; Mader et al., 1989; Barker and Dawson, 1998; Camp, 2001; Champagne et al., 2004; Viau and Champagne, 2004). Therapeutic dogs are now used to help children with ASD (Redefer and Goodman, 1989; Martin and Farnum, 2002). Dogs in general and therapeutic dogs have been found to serve as catalysts for social interactions (Mader et al., 1989; McNicholas and Collis, 2000). Davis et al. (2004) established that assistance dogs could help autistic children to learn about living beings, feelings, and needs. To this effect, the MIRA Foundation trains service dogs to act as moderators between autistic children and their social environment, in an effort to improve social behavior (Fecteau, 2008). Recent qualitative research has shown that service dogs, such as those trained by the MIRA Foundation, have beneficial effects on autistic children and their families, which can be observed in improved daily routine (Burrows et al., 2008). Conversely, researchers noted that not all children have the same reaction when interacting with service dogs. Less is known about the effects of service dogs on autistic children at a physiological level. Cortisol, a stress hormone that can be measured in saliva samples (Kirschbaum and Hellhammer, 1989), could serve as a good proxy measure to assess potential positive effects of service dogs on autistic children. Odendaal and Meintjes (2003) found that healthy adults secrete less plasma cortisol when bonding with dogs, which suggests that these animals have a calming effect on their human companions. To our knowledge, no study has examined the effects of service dogs on cortisol secreted by autistic children. Autistic children’s cortisol secretion is contentious. While some have measured lower morning cortisol levels and higher evening cortisol levels in children with autism compared to healthy children (Corbett et al., 2008), a few other studies have reported that basal levels of cortisol in children with ASD do not differ from levels found in healthy children (Tordjman et al., 1997; Jansen et al., 1999, 2000). In a review, Lam et al. (2006) examined studies on plasma basal cortisol levels in autistic children and adults, and reached the same conclusion. Their findings suggest that autistic individuals show an abnormal stress response rather than chronic hyper arousal, as would be reflected by elevated basal cortisol levels. To further support their argument, Lam et al. (2006) mentioned two studies where autistic individuals were tested with dexamethasone (DEX) and showed abnormal suppression of cortisol. This suggests an abnormal Hypothalamic—Pituitary—Adrenal (HPA) axis integrity in
R. Viau et al. autistic individuals and leads to believe that a dynamic measure of cortisol secretion may reveal more relevant information about cortisol secretion in autistic children. The Cortisol Awakening Response (CAR) might provide different additional information about the factors modulating cortisol secretion. The CAR corresponds to a rise in cortisol secretion immediately after awakening. In healthy adults, this increase ranges from 50% to 160% and is thought to result from distinct modulatory mechanisms, different from those responsible for the diurnal cycle (Clow et al., 2004). In particular, CAR has been associated with psychosocial markers of stress and disease, making it a distinct element of the HPA axis (Fries et al., 2009). In their systematic review, Chida and Steptoe (2009) found that job and general life stress were positively correlated to CAR. They also reported weaker associations between positive psychological states and reduced CAR. In a case study, Stalder et al. (2009) showed that CAR is positively associated with arousal and expectations of the day ahead, and inversely associated with happiness from the preceding day. However, these results pertained only to adults. One epidemiological study of salivary cortisol in healthy children reported CAR levels of 30%, which is considerably less than in adults (Rosmalen et al., 2005). CAR has never been examined in autistic children. Due to its dynamic nature compared to diurnal basal cortisol levels, it could provide a very useful marker in studying the effect of service dogs on the physiological stress response of children with ASD. Thus, the main goal of this exploratory study was to measure the effects of service dogs on the basal salivary cortisol secretion, as measured by averaged diurnal secretion and CAR. The second goal of this study was to determine if any change in cortisol secretion would be associated with changes in the child’s behavior, as reported by his or her parents.
2. Methods Fifty-seven children diagnosed with ASD were recruited from the MIRA Foundation web site (http://www.mira.ca) and from various national and provincial organizations on autism. The selection process was based on diagnostic evaluations performed by a team of independent professionals. Only children with a diagnostic of autism, Asperger syndrome or PDDNOS were included in this study. Children were excluded from the study if they were taking oral steroids or if any member of their family was allergic to dogs. Other treatment or therapeutic interventions prescribed by public or private organizations prior to the study were kept unchanged during the study. Since the MIRA Foundation requires that the service dog should not be left alone for more than 4 h per day, the study needed families where one of the parents did not work outside the home. First, parents were given instructions on how to collect saliva samples. Then, they attended a three-day training session on how to interact with the dog and how to introduce it to their child. All dogs were provided by the MIRA Foundation where they received proper training and behavioral evaluation. The service dog training, which teaches obedience commands, lasts three months and is divided into several blocks. The main goal of the training is to teach service dogs to remain calm in order to ensure safety should their environment
MIRA, austim and cortisol become chaotic. Selected dogs were deemed to adapt easily to various environments, to be even-tempered, to use staircases comfortably, to easily manage their insecurities, and to be calm and respectful, in an attempt to increase the child’s autonomy and to facilitate the parents’ task. Parents were appointed to issue basic dog obedience commands and to hold the service dog’s harness. It was suggested that their child could be attached to the harness, at the parents’ discretion. Each dog was also carefully matched with the child’s temperament and needs before being placed in a family. Once the study was completed, the dogs were offered to the families. The research protocol for the study was approved by the Ethics Committees of the Douglas Institute and of McGill University. The regular MIRA Foundation protocol was followed, under the supervision of a team of veterinarians, to ensure satisfactory animal welfare and proper allocation of dogs to families. The study comprised three experimental conditions, during which the child’s behavioral parameters and basal cortisol levels were measured at various times during the day. These three phases consisted of a two-week period prior to the introduction of the dog to the family (PRE), followed by four weeks with the dog (DOG), and two weeks after the dog had been removed from the family (POST). During these three phases, an 11-item questionnaire was distributed to parents to assess changes in their child’s behavior. This questionnaire was a semi-quantitative openended questionnaire, which focused on the child’s problematic behaviors. An initial behavioral assessment was made two weeks prior to the introduction of the service dogs (PRE). The parents were asked questions such as: ‘‘What does your child do when: (a) he/she is alone? (b) he/she is with other children? and (c) he/she is anxious?’’ or ‘‘Describe how he/she manifests his/her (a) sadness, (b) pain, (c) happiness, and (d) needs’’. Based on these observations, an initial score was generated by counting disruptive behavioral incidents. In each subsequent week, parents were asked to evaluate whether there had been changes in any of their child’s behaviors. Based on this assessment, 2, 1, or 0 points were subtracted from the initial score if a noticeable change, a small change, or no change was noticed. Salivary cortisol was collected with a salivette (Sarstedt, Germany) or with sterile Q-Tips for children who were prone to swallowing the cotton swab used in the salivettes. Parents were asked to place the cotton swabs, or the QTips, back in the plastic containers provided with the salivettes, and to store these in their freezer until samples were analyzed. Radioimmunoassays for salivary cortisol concentrations were performed at the Douglas Institute according to methods previously described (Tu et al., 2006). The assay’s detection limit was 0.01 mg/dL, and intra- and inter-assay coefficients of variance were 4.0% and 4.6%, respectively. Basal cortisol levels were measured once weekly, three times a day: upon awakening (A), 30 min after awakening (B), and at bedtime (C). Parents were asked to sample their child’s saliva on the same day each week. Average cortisol levels were then calculated for each of the following periods: two weeks prior to the introduction of the dog (PRE), four weeks with the dog (DOG), and two weeks after the service dog was removed from the family (POST). The CAR was computed and averaged for
1189 each condition. The formula used for CAR was (B A)/A. Outliers, cortisol levels measured at three standard deviations above the group mean, were replaced by the mean (this accounted for less than 5% of the entire cortisol measurements). Finally, ANOVAs with repeated measures on conditions (3 levels: PRE, DOG, and POST) were performed to assess changes in autistic children, first using the averaged cortisol samples (3 levels: A, B, and C), and then using the averaged CAR. An ANOVA with repeated measures on conditions (3 levels: PRE, DOG, and POST) was performed to assess changes in behavior. In addition, Pearson correlations were performed on the number of disruptive behaviors reported by parents, the average diurnal cortisol, and CAR.
3. Results Of the 57 children recruited, 6 were excluded from the study because they were diagnosed with other disorders, such as Williams Syndrome, anxiety, trisomy 21, epilepsy or Childhood Disintegrative Disorder, or because they had undergone neurosurgery prior to the study. Nine other children did not complete the study. In total, 42 children and their families participated in this study. Thirty-four were diagnosed with autism, 2 with Asperger syndrome, and 6 with PDDNOS. There were 37 boys and 5 girls aged, on average, 7.1 3.1 years old (ranging from 3.6 to 14.8 years). The average annual family income was $69,500 $33,500 (ranging from $12,000 to $150,000). This range may be a direct consequence of family status: 6 children came from single-parent families. While 7 children did not have any siblings, 24 had 1, 10 had 2, and 1 child had 4. The basal cortisol levels of these children did not differ according to their diagnostic group, their gender, or the structure of their family (one-way ANOVA, all p > 0.05). In addition, the children’s cortisol levels showed no correlation with their age or with their family’s income level (Pearson correlations, all p > 0.05). Information from questionnaires completed by parents (Fig. 1) suggests that children reacted positively to service dogs, as the number of problematic behaviors reported by parents decreased after the introduction of the dogs. An ANOVA with repeated measures showed that the number of problematic behaviors prior to receiving the service dogs was significantly higher than during the subsequent two conditions of the study (F(2,82) = 106.0, p < 0.001). However, these scores did not correlate with any cortisol measures (average or CAR) at the corresponding PRE, DOG or POST conditions (all p > 0.05). As demonstrated in Fig. 2, normal diurnal rhythms in salivary cortisol secretion were observed at each experimental condition, where morning cortisol levels peaked immediately after awakening and declined to a trough around bedtime. The ANOVA with repeated measures revealed a statistically significant main effect of samples (F(2,42) = 53.6, p < 0.01), where the levels of cortisol in samples A and B were different from the levels in sample C ( p < 0.05). This confirms normal diurnal patterns of cortisol secretion in these children. There were no effect of conditions and no interaction of samples by conditions (F < 1), which suggests that cortisol levels in each condition (PRE, DOG, POST) were the same. This was true for each sample (A, B, or C).
1190
R. Viau et al.
Figure 1 The change in problematic behaviors of children with Pervasive Developmental Disorder as reported by parents prior to (PRE), during (DOG) and after (POST) introducing a service dog to the family. Bars are standard errors of the mean and units are arbitrary. * denotes significant statistical difference ( p < 0.05). The questionnaire is a semi-quantitative, open-ended questionnaire focused on problematic behaviors as reported by parents of autistic children. An initial score of such behaviors was established during the two weeks prior to the introduction of the service dog. Then, each week, based on the parents’ account of initial behaviors, 2, 1, or 0 points were subtracted from the initial score if a change, a small change, or no change, respectively, was noticed.
Figure 2 The diurnal cortisol profiles of children with ASD in the three experimental conditions: prior to (PRE; diamonds), during (DOG; squares), and after (POST; triangles) the introduction of service dogs to these children’s families. Sample A represents cortisol levels as measured upon awakening, sample B refers to cortisol levels measured 30 min after awakening, and sample C refers to cortisol levels measured at bedtime. Bars represent standard error of the means and values are in mg/dL.
Some difficulties were experienced when collecting saliva from some children. Consequently, we discontinued collecting the second set of samples (sample B, after awakening) for the last 8 families recruited. These 8 children for whom we did not calculate a CAR did not differ from the others in terms of demographic characteristics or average cortisol levels. We were able to compute a CAR for 34 children. In contrast with the average cortisol secretion, we found a significant effect of condition on the children’s CAR (Fig. 3). The CAR was found to be considerably reduced when dogs were present, plunging from an increase of 58% prior to the introduction of the dogs to 10%. The CAR rose to 48% when dogs were removed from the families. A statistical main effect of conditions (F(2,48) = 5.2, p = 0.01) was found when CAR was entered into an ANOVA with repeated measures. Fig. 2 shows that the decrease in CAR after introducing the dogs appears to be due to an increase in awakening cortisol (sample A), but an ANOVA with repeated measures on sample A only did not yield significant results (data not shown). We were interested in learning whether the reduction in CAR was due to an adaptation to the dog’s presence or not. To
this effect, we divided the conditions in two-week periods: weeks 1—2 prior to introducing the dog, weeks 3—4 immediately after introducing the dog, weeks 5—6 with the dog, weeks 7—8 after removing the dog from the family. We then conducted an ANOVA with repeated measures (Fig. 3), which revealed a gradual decline in CAR, followed by a return to a higher CAR after the dogs were removed from the families (F(3,72) = 3.6, p = 0.02). Least significant difference posthoc analyses showed that results from weeks 1 to 2 (prior to the introduction of the service dogs) were higher than those from weeks 3—4 to 5—6 (with the dogs), and that results from weeks 5 to 6 (the last two weeks with the dogs) were lower than those from weeks 7 to 8 (the two-week period after the removal of the dog).
4. Discussion In this study, we assessed the effects of service dogs on the basal cortisol and on the Cortisol Awakening Response (CAR) of autistic children. We found that the introduction of service dogs had a significant effect on the CAR of autistic children,
MIRA, austim and cortisol
Figure 3 Upper panel represents the Cortisol Awakening Response (CAR) in children with ASD prior to (PRE), during (DOG) and after (POST) the introduction of service dogs to the families. The lower panel represents the CAR in children with ASD per week. WK1-2 is the mean CAR of the first two weeks, prior to the introduction of the dogs; WK3-4 is the mean CAR of the first two weeks while the dogs were present; WK5-6 is the mean CAR of the two following weeks with the dogs; and WK7-8 is the mean CAR of the two weeks after the removal of the dogs from the families. Bars represent standard errors of the means, units are percentage of increases. * denotes significant statistical difference ( p < 0.01).
reducing it from a 58% increase prior to the introduction of the service dogs to 10%. Their CAR jumped back to 48% once dogs were removed from the children’s environment. This reduced CAR was not simply driven by an increase of cortisol concentration in the first salivary samples, since an ANOVA on morning samples did not reach statistical significance. At this time, it is difficult to attribute a precise functional significance to the changes in CAR observed in these children, because few studies have evaluated this parameter in a young population. We know that healthy children have a CAR of 30%, which is considerably lower than the range of 50—150% found in healthy adults (Rosmalen et al., 2005). In adults, CAR has been shown to be affected by various psychological factors (Clow et al., 2004; Fries et al., 2009). More precisely, recent single-case and review studies suggest that general stress is correlated with an increased CAR, and that a positive psychological state is correlated with a reduced CAR (Chida and Steptoe, 2009; Stalder et al., 2009). It is possible that the changes observed in the CAR of these autistic
1191 children were related to the psychological factors linked to the presence of service dogs. Interestingly, in a qualitative study examining the effects of service dogs in families of autistic children, Burrows et al. (2008) found that parents reported that their child was calmer or happier when the dog was present. However, these explanations remain hypothetical, as we did not measure the children’s happiness or subjective stress. Alternatively, the changes in the CAR of autistic children could be explained by changes in their sleep patterns. We know that autistic children suffer from disturbed sleep patterns (see Johnson and Malow, 2008 for review). Although we did not measure the quality of sleep, the wake-up times, or episodes of sleep disturbance in our study, anecdotal data from parents allow us to think that the presence of the service dogs might have encouraged improved sleep. This is congruent with Burrows et al. (2008) who found that besides improving the parents’ sleep, some autistic children loved sleeping with their dog and fell asleep faster than before. Although previous studies have failed to demonstrate a significant effect of sleep disturbances on CAR in healthy individuals (Dettenborn et al., 2007), Backhaus et al. (2004) found that subjective sleep disturbances in healthy adults are associated with lower morning cortisol levels. Further research is needed to understand the relationships between the service dog’s presence, sleep patterns, and the cortisol secretion of autistic children. The elimination of the second set of saliva samples (sample B, taken 30 min after awakening) from the protocol during the course of the study after reports of difficulties in saliva sampling bears some interesting questions. This 30min lapse is within the stress response time (cortisol secretion peaks 30—60 min after stress exposure). Children who were reluctant to give saliva samples upon awakening might have shown a stress response, which may have been reflected in the second sample. If this were true, the reduction in CAR while service dogs are present suggests that the dogs may have simply facilitated the saliva sampling and reduced the cortisol stress response of the children. This would bring further support to the psychological effect of service dogs on autistic children. Alternatively, parents participating in the Burrows et al.’s (2008) study reported that the service dogs used in their study served as an anchor, preventing their children from bolting. This could also explain the results presented here, as it was suggested that the child be attached to the service dog’s harness. However, we do not know which families followed this suggestion. Should the second saliva samples be a measure of stress response, anchoring the children to facilitate the cortisol sampling should diminish this reaction. Children with ASD may be more resistant to changes in routine (Kanner, 1943). Furthermore, the introduction of the service dogs might have significantly altered the family routine and structure. If the reduction in CAR was due to a change in the family structure after introducing the dog, thereby resulting in a stress adaptation, we should have seen a sharp decline immediately after the introduction of the service dogs, followed by a slow return to previous CAR levels. Instead, we found that the reduction in CAR seemed more pronounced during the last two weeks with the dogs, which suggests that the dog’s presence was responsible for this effect, as opposed to an adaptation to change.
1192 We did not find differences in average cortisol levels between conditions. More specifically, the basal salivary cortisol levels of autistic children before, during and after the introduction and removal of service dogs did not differ statistically. This could be explained by the fact that the use of dynamic measures of cortisol secretion may be more appropriate to reveal differences with healthy children. Previous studies indicate that autistic children exhibit normal basal cortisol levels (Tordjman et al., 1997; Jansen et al., 1999, 2000). Conversely, other studies indicate that autistic children are under-responsive to psychological stressors, which means that they secrete less cortisol than healthy children do when exposed to psychological stressors (Jansen et al., 1999, 2000, 2003). Finally, autistic children have been found to be non-suppressors of the DEX test (Lam et al., 2006). Any interpretation would remain speculative, as we did not include a control group in our study. It would have been interesting to find that autistic children secreted basal salivary cortisol levels comparable to that of healthy children, while exhibiting a different CAR. In addition, it may be useful to study whether the CAR of healthy children is influenced by service dogs in the same way as autistic children. Our study is the first one to report interventioninduced changes in autistic children’s cortisol secretion. We believe that the within-subject design of this study provides some degree of reliability in our interpretation that service dogs have an effect on the CAR of children with ASD. As a secondary goal, we wanted to examine if changes in cortisol secretion would be reflected in the child’s behavior. In addition, since all children do not have the same reaction to animals, it was deemed important to show that service dogs had a beneficial effect on the children studied. We found that the frequency of problematic behaviors reported by parents decreased when service dogs were present. This effect persisted when the dogs were removed from the family for two weeks. However, no correlations were found between CAR or average cortisol levels, and the number of disruptive behaviors reported by parents. Due to the qualitative nature of the initial behavioral assessment, the resulting range of problematic behaviors was quite large. This might explain why behavior did not change when the service dog was removed from the family. Parents reported that the service dog had a positive effect on their child’s behavior, as seen through a decrease in the frequency of self-stimulation episodes, repetitive behaviors and tantrums, as well as an increase in the child’s tolerance to some noises, such as household appliances. These behavioral changes may reflect the parents’ satisfaction with the experiment more than an actual modification in their child’s behavior. Due to extraordinary circumstances, it was impossible to perform further study on the type, nature, and scale of the behavioral changes observed in these children. An unbiased third-party’s evaluation may prove to be more appropriate than the parents’ comments. This will be done in a future study. Furthermore, the questionnaire did not measure perceived stress or its effect on the children before and after the introduction of service dogs. Given what is known about the psychological effects on CAR and the psychological effect of service dogs on autistic children, we believe that the relationship between a reduced CAR and reduced disruptive behaviors occurs through the psychological effect of service dogs on autistic children.
R. Viau et al. In summary, our exploratory study showed that service dogs have a beneficial effect in reducing CAR and the number of disruptive behavioral incidents in children with ASD. Literature on the benefits of service dogs is expanding and points to definitive benefits for families of children diagnosed with ASD. Our study brings further support to parental reports on the benefits of service dogs by showing physiological effects (reduced CAR) in autistic children, which may correspond to the physiological effects observed in adults who experience positive psychological states. Additional studies are needed to determine whether these physiological changes are correlated to measures of perceived stress in these children, for instance.
Role of funding source This study was founded by La Fondation MIRA, Quebec, Canada. La Fondation MIRA had a role in study design and in the collection of data, but had no role in the analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest None to declare.
Acknowledgement This article is published in memory of Dr. Robert Viau, the instigator of this study.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Association, Washington, DC (Text Revised). Backhaus, J., Junghanns, K., Hohagen, F., 2004. Sleep disturbances are correlated with decreased morning awakening salivary cortisol. Psychoneuroendocrinology 29 (9), 1184—1191. Barker, S.B., Dawson, K.S., 1998. The effects of animal-assisted therapy on anxiety ratings of hospitalized psychiatric patients. Psychiatr. Serv. 49 (6), 797—801. Burrows, K.E., Adams, C.L., Spiers, J., 2008. Sentinels of safety: service dogs ensure safety and enhance freedom and well-being for families with autistic children. Qual. Health Res. 18 (12), 1642—1649. Camp, M.M., 2001. The use of service dogs as an adaptive strategy: a qualitative study. Am. J. Occup. Ther. 55 (5), 509—517. Champagne, N., St-Pierre, E., Viau, R., 2004. A longitudinal study and practice on the blind youth and the guide dog. , http:// www.mira.ca. Chida, Y., Steptoe, A., 2009. Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis. Biol. Psychol. 80 (3), 265—278. Clow, A., Thorn, L., Evans, P., Hucklebridge, F., 2004. The awakening cortisol response: methodological issues and significance. Stress 7 (1), 29—37. Corbett, B.A., Mendoza, S., Wegelin, J.A., Carmean, V., Levine, S., 2008. Variable cortisol circadian rhythms in children with autism and anticipatory stress. J. Psychiatry Neurosci. 33 (3), 227—234. Davis, B.W.N., O’Brien, K., Patronek, S., MacCollin, G.M., 2004. Assistance dog placement in the pediatric population: benefits,
MIRA, austim and cortisol risks, and the recommendations for future application. Anthrozoos 17 (2), 130—145. Dettenborn, L., Rosenloecher, F., Kirschbaum, C., 2007. No effects of repeated forced wakings during three consecutive nights on morning cortisol awakening responses (CAR): a preliminary study. Psychoneuroendocrinology 32 (8—10), 915—921. Eddy, J., Hart, L.A., Boltz, R.P., 1988. The effects of service dogs on social acknowledgments of people in wheelchairs. J. Psychol. 122 (1), 39—45. Fecteau, S., 2008. Inte ´gration d’un chien d’assistance dans la famille d’un enfant pre ´sentant un TED: la contribution de la the ´orie de l’attachement parent-enfant. Universite ´ de Sherbrooke, Sherbrooke, Que ´bec. Fombonne, E., 2003. Epidemiological surveys of autism and other pervasive developmental disorders: an update. J. Autism Dev. Disord. 33 (4), 365—382. Fries, E., Dettenborn, L., Kirschbaum, C., 2009. The cortisol awakening response (CAR): facts and future directions. Int. J. Psychophysiol. 72 (1), 67—73. Jansen, L.M., Gispen-de Wied, C.C., Jansen, M.A., van der Gaag, R.J., Matthys, W., van Engeland, H., 1999. Pituitary—adrenal reactivity in a child psychiatric population: salivary cortisol response to stressors. Eur. Neuropsychopharmacol. 9 (1—2), 67—75. Jansen, L.M., Gispen-de Wied, C.C., Van der Gaag, R.J., ten Hove, F., Willemsen-Swinkels, S.W., Harteveld, E., et al., 2000. Unresponsiveness to psychosocial stress in a subgroup of autistic-like children, multiple complex developmental disorder. Psychoneuroendocrinology 25 (8), 753—764. Jansen, L.M., Gispen-de Wied, C.C., van der Gaag, R.J., van Engeland, H., 2003. Differentiation between autism and multiple complex developmental disorder in response to psychosocial stress. Neuropsychopharmacology 28 (3), 582—590. Johnson, K.P., Malow, B.A., 2008. Sleep in children with autism spectrum disorders. Curr. Neurol. Neurosci. Rep. 8 (2), 155—161. Kanner, L., 1943. Autistic disturbances of affective contact. Nerv. Child 2, 217—250.
1193 Kirschbaum, C., Hellhammer, D.H., 1989. Salivary cortisol in psychobiological research: an overview. Neuropsychobiology 22 (3), 150—169. Lam, K.S., Aman, M.G., Arnold, L.E., 2006. Neurochemical correlates of autistic disorder: a review of the literature. Res. Dev. Disabil. 27 (3), 254—289. Mader, B., Hart, L.A., Bergin, B., 1989. Social acknowledegments for children with disabilities: effects of service dogs. Child Dev. 60 (6), 1529—1534. Martin, F., Farnum, J., 2002. Animal-assisted therapy for children with pervasive developmental disorders. West. J. Nurs. Res. 24 (6), 657—670. McNicholas, J., Collis, G.M., 2000. Dogs as catalysts for social interactions: robustness of the effect. Br. J. Psychol. 91 (Pt 1), 61—70. Odendaal, J.S., Meintjes, R.A., 2003. Neurophysiological correlates of affiliative behaviour between humans and dogs. Vet. J. 165 (3), 296—301. Redefer, L.A., Goodman, J.F., 1989. Brief report: pet-facilitated therapy with autistic children. J. Autism Dev. Disord. 19 (3), 461—467. Rosmalen, J.G., Oldehinkel, A.J., Ormel, J., de Winter, A.F., Buitelaar, J.K., Verhulst, F.C., 2005. Determinants of salivary cortisol levels in 10—12 year old children; a population-based study of individual differences. Psychoneuroendocrinology 30 (5), 483—495. Stalder, T., Hucklebridge, F., Evans, P., Clow, A., 2009. Use of a single case study design to examine state variation in the cortisol awakening response: relationship with time of awakening. Psychoneuroendocrinology 34 (4), 607—614. Tordjman, S., Anderson, G.M., McBride, P.A., Hertzig, M.E., Snow, M.E., Hall, L.M., et al., 1997. Plasma beta-endorphin, adrenocorticotropin hormone, and cortisol in autism. J. Child Psychol. Psychiatry 38 (6), 705—715. Tu, M.T., Lupien, S.J., Walker, C.D., 2006. Diurnal salivary cortisol levels in postpartum mothers as a function of infant feeding choice and parity. Psychoneuroendocrinology 31 (7), 812—824. Viau, R., Champagne, N., 2004. Les effets du chien dans la vie des humains: recension d’e ´crits. , http://www.mira.ca.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n
Effect of service dogs on salivary cortisol secretion in autistic children `ve Arsenault-Lapierre b, Ste ´phanie Fecteau a, Robert Viau a, Genevie ¨l Champagne a, Claire-Dominique Walker c, Sonia Lupien b,* Noe a
´bec, Canada Fondation MIRA, 1820 rang Nord Ouest, Sainte-Madeleine, Que Center for Studies on Human Stress, Mental Health Research Centre Fernand Seguin, Hospital Louis-H. Lafontaine, ´ de Montre ´al, 7331 Hochelaga, Montre ´al, Que ´bec, Canada Universite c ´bec, Canada Douglas Mental Health University Institute, McGill University, Que b
Received 1 June 2009; received in revised form 3 February 2010; accepted 4 February 2010
KEYWORDS Salivary cortisol; Autism; Cortisol Awakening Response; Service dogs
Summary Children with Autism Syndrome Disorders (ASDs) exhibit social, communicative, and behavioral deficits. We know that human interaction with dogs, which is thought to serve as a social catalyst, results in a decrease of cortisol levels in healthy adults. Introducing service dogs to children with ASD is an attractive idea that has received growing attention in recent decades. However, no study has measured the physiological impact of service dogs on these children. Therefore, the goal of our study was to assess the effects of service dogs on the basal salivary cortisol secretion of children with ASD. We measured the salivary cortisol levels of 42 children with ASD in three experimental conditions; prior to and during the introduction of a service dog to their family, and after a short period during which the dog was removed from their family. We compared average cortisol levels and Cortisol Awakening Response (CAR) before and during the introduction of the dog to the family and after its withdrawal. We found that the introduction of service dogs translated into a statistically significant diminished CAR. Before the introduction of service dogs, we measured a 58% increase in morning cortisol after awakening, which diminished to 10% when service dogs were present. The increase in morning cortisol jumped back to 48% once the dogs were removed from the families ( p < 0.05). However, service dogs did not have an effect on the children’s average diurnal cortisol levels. These results show that the CAR of children with ASD is sensitive to the presence of service dogs, which lends support to the potential behavioral benefits of service dogs for children with autism. # 2010 Elsevier Ltd. All rights reserved.
1. Introduction
* Corresponding author. Tel.: +1 514 251 4015x2337; fax: +1 514 251 4014. E-mail address: [email protected] (S. Lupien).
Autism Spectrum Disorders (ASDs) are a neurodevelopmental disorder classified under the Pervasive Developmental Disorder (PDD) diagnosis of the DSM-IV-TR (APA, 2000). ASD typically refers to autism, Asperger syndrome and Pervasive
0306-4530/$ — see front matter # 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2010.02.004
1188 Development Disorder Not Otherwise Specified (PDDNOS). It is characterized by symptoms spanning the realm of social interactions, communication, and restricted or repetitive behaviors. Although the prevalence of ASD has risen over the years, making autism one of the most common neurodevelopmental disorders, this increase may be due to modifications to the diagnostic criteria and a better understanding of these disorders (Fombonne, 2003). Despite the improved understanding of ASD, whereby children are now placed in regular schools and enjoy better supervision, these children still face exclusion from their social environments due to the social, communicative, and behavioral nature of their disorder. Rehabilitation intervention, such as zootherapy, has been proven useful in helping individuals with visual, physical, intellectual, or psychiatric deficiencies (Eddy et al., 1988; Mader et al., 1989; Barker and Dawson, 1998; Camp, 2001; Champagne et al., 2004; Viau and Champagne, 2004). Therapeutic dogs are now used to help children with ASD (Redefer and Goodman, 1989; Martin and Farnum, 2002). Dogs in general and therapeutic dogs have been found to serve as catalysts for social interactions (Mader et al., 1989; McNicholas and Collis, 2000). Davis et al. (2004) established that assistance dogs could help autistic children to learn about living beings, feelings, and needs. To this effect, the MIRA Foundation trains service dogs to act as moderators between autistic children and their social environment, in an effort to improve social behavior (Fecteau, 2008). Recent qualitative research has shown that service dogs, such as those trained by the MIRA Foundation, have beneficial effects on autistic children and their families, which can be observed in improved daily routine (Burrows et al., 2008). Conversely, researchers noted that not all children have the same reaction when interacting with service dogs. Less is known about the effects of service dogs on autistic children at a physiological level. Cortisol, a stress hormone that can be measured in saliva samples (Kirschbaum and Hellhammer, 1989), could serve as a good proxy measure to assess potential positive effects of service dogs on autistic children. Odendaal and Meintjes (2003) found that healthy adults secrete less plasma cortisol when bonding with dogs, which suggests that these animals have a calming effect on their human companions. To our knowledge, no study has examined the effects of service dogs on cortisol secreted by autistic children. Autistic children’s cortisol secretion is contentious. While some have measured lower morning cortisol levels and higher evening cortisol levels in children with autism compared to healthy children (Corbett et al., 2008), a few other studies have reported that basal levels of cortisol in children with ASD do not differ from levels found in healthy children (Tordjman et al., 1997; Jansen et al., 1999, 2000). In a review, Lam et al. (2006) examined studies on plasma basal cortisol levels in autistic children and adults, and reached the same conclusion. Their findings suggest that autistic individuals show an abnormal stress response rather than chronic hyper arousal, as would be reflected by elevated basal cortisol levels. To further support their argument, Lam et al. (2006) mentioned two studies where autistic individuals were tested with dexamethasone (DEX) and showed abnormal suppression of cortisol. This suggests an abnormal Hypothalamic—Pituitary—Adrenal (HPA) axis integrity in
R. Viau et al. autistic individuals and leads to believe that a dynamic measure of cortisol secretion may reveal more relevant information about cortisol secretion in autistic children. The Cortisol Awakening Response (CAR) might provide different additional information about the factors modulating cortisol secretion. The CAR corresponds to a rise in cortisol secretion immediately after awakening. In healthy adults, this increase ranges from 50% to 160% and is thought to result from distinct modulatory mechanisms, different from those responsible for the diurnal cycle (Clow et al., 2004). In particular, CAR has been associated with psychosocial markers of stress and disease, making it a distinct element of the HPA axis (Fries et al., 2009). In their systematic review, Chida and Steptoe (2009) found that job and general life stress were positively correlated to CAR. They also reported weaker associations between positive psychological states and reduced CAR. In a case study, Stalder et al. (2009) showed that CAR is positively associated with arousal and expectations of the day ahead, and inversely associated with happiness from the preceding day. However, these results pertained only to adults. One epidemiological study of salivary cortisol in healthy children reported CAR levels of 30%, which is considerably less than in adults (Rosmalen et al., 2005). CAR has never been examined in autistic children. Due to its dynamic nature compared to diurnal basal cortisol levels, it could provide a very useful marker in studying the effect of service dogs on the physiological stress response of children with ASD. Thus, the main goal of this exploratory study was to measure the effects of service dogs on the basal salivary cortisol secretion, as measured by averaged diurnal secretion and CAR. The second goal of this study was to determine if any change in cortisol secretion would be associated with changes in the child’s behavior, as reported by his or her parents.
2. Methods Fifty-seven children diagnosed with ASD were recruited from the MIRA Foundation web site (http://www.mira.ca) and from various national and provincial organizations on autism. The selection process was based on diagnostic evaluations performed by a team of independent professionals. Only children with a diagnostic of autism, Asperger syndrome or PDDNOS were included in this study. Children were excluded from the study if they were taking oral steroids or if any member of their family was allergic to dogs. Other treatment or therapeutic interventions prescribed by public or private organizations prior to the study were kept unchanged during the study. Since the MIRA Foundation requires that the service dog should not be left alone for more than 4 h per day, the study needed families where one of the parents did not work outside the home. First, parents were given instructions on how to collect saliva samples. Then, they attended a three-day training session on how to interact with the dog and how to introduce it to their child. All dogs were provided by the MIRA Foundation where they received proper training and behavioral evaluation. The service dog training, which teaches obedience commands, lasts three months and is divided into several blocks. The main goal of the training is to teach service dogs to remain calm in order to ensure safety should their environment
MIRA, austim and cortisol become chaotic. Selected dogs were deemed to adapt easily to various environments, to be even-tempered, to use staircases comfortably, to easily manage their insecurities, and to be calm and respectful, in an attempt to increase the child’s autonomy and to facilitate the parents’ task. Parents were appointed to issue basic dog obedience commands and to hold the service dog’s harness. It was suggested that their child could be attached to the harness, at the parents’ discretion. Each dog was also carefully matched with the child’s temperament and needs before being placed in a family. Once the study was completed, the dogs were offered to the families. The research protocol for the study was approved by the Ethics Committees of the Douglas Institute and of McGill University. The regular MIRA Foundation protocol was followed, under the supervision of a team of veterinarians, to ensure satisfactory animal welfare and proper allocation of dogs to families. The study comprised three experimental conditions, during which the child’s behavioral parameters and basal cortisol levels were measured at various times during the day. These three phases consisted of a two-week period prior to the introduction of the dog to the family (PRE), followed by four weeks with the dog (DOG), and two weeks after the dog had been removed from the family (POST). During these three phases, an 11-item questionnaire was distributed to parents to assess changes in their child’s behavior. This questionnaire was a semi-quantitative openended questionnaire, which focused on the child’s problematic behaviors. An initial behavioral assessment was made two weeks prior to the introduction of the service dogs (PRE). The parents were asked questions such as: ‘‘What does your child do when: (a) he/she is alone? (b) he/she is with other children? and (c) he/she is anxious?’’ or ‘‘Describe how he/she manifests his/her (a) sadness, (b) pain, (c) happiness, and (d) needs’’. Based on these observations, an initial score was generated by counting disruptive behavioral incidents. In each subsequent week, parents were asked to evaluate whether there had been changes in any of their child’s behaviors. Based on this assessment, 2, 1, or 0 points were subtracted from the initial score if a noticeable change, a small change, or no change was noticed. Salivary cortisol was collected with a salivette (Sarstedt, Germany) or with sterile Q-Tips for children who were prone to swallowing the cotton swab used in the salivettes. Parents were asked to place the cotton swabs, or the QTips, back in the plastic containers provided with the salivettes, and to store these in their freezer until samples were analyzed. Radioimmunoassays for salivary cortisol concentrations were performed at the Douglas Institute according to methods previously described (Tu et al., 2006). The assay’s detection limit was 0.01 mg/dL, and intra- and inter-assay coefficients of variance were 4.0% and 4.6%, respectively. Basal cortisol levels were measured once weekly, three times a day: upon awakening (A), 30 min after awakening (B), and at bedtime (C). Parents were asked to sample their child’s saliva on the same day each week. Average cortisol levels were then calculated for each of the following periods: two weeks prior to the introduction of the dog (PRE), four weeks with the dog (DOG), and two weeks after the service dog was removed from the family (POST). The CAR was computed and averaged for
1189 each condition. The formula used for CAR was (B A)/A. Outliers, cortisol levels measured at three standard deviations above the group mean, were replaced by the mean (this accounted for less than 5% of the entire cortisol measurements). Finally, ANOVAs with repeated measures on conditions (3 levels: PRE, DOG, and POST) were performed to assess changes in autistic children, first using the averaged cortisol samples (3 levels: A, B, and C), and then using the averaged CAR. An ANOVA with repeated measures on conditions (3 levels: PRE, DOG, and POST) was performed to assess changes in behavior. In addition, Pearson correlations were performed on the number of disruptive behaviors reported by parents, the average diurnal cortisol, and CAR.
3. Results Of the 57 children recruited, 6 were excluded from the study because they were diagnosed with other disorders, such as Williams Syndrome, anxiety, trisomy 21, epilepsy or Childhood Disintegrative Disorder, or because they had undergone neurosurgery prior to the study. Nine other children did not complete the study. In total, 42 children and their families participated in this study. Thirty-four were diagnosed with autism, 2 with Asperger syndrome, and 6 with PDDNOS. There were 37 boys and 5 girls aged, on average, 7.1 3.1 years old (ranging from 3.6 to 14.8 years). The average annual family income was $69,500 $33,500 (ranging from $12,000 to $150,000). This range may be a direct consequence of family status: 6 children came from single-parent families. While 7 children did not have any siblings, 24 had 1, 10 had 2, and 1 child had 4. The basal cortisol levels of these children did not differ according to their diagnostic group, their gender, or the structure of their family (one-way ANOVA, all p > 0.05). In addition, the children’s cortisol levels showed no correlation with their age or with their family’s income level (Pearson correlations, all p > 0.05). Information from questionnaires completed by parents (Fig. 1) suggests that children reacted positively to service dogs, as the number of problematic behaviors reported by parents decreased after the introduction of the dogs. An ANOVA with repeated measures showed that the number of problematic behaviors prior to receiving the service dogs was significantly higher than during the subsequent two conditions of the study (F(2,82) = 106.0, p < 0.001). However, these scores did not correlate with any cortisol measures (average or CAR) at the corresponding PRE, DOG or POST conditions (all p > 0.05). As demonstrated in Fig. 2, normal diurnal rhythms in salivary cortisol secretion were observed at each experimental condition, where morning cortisol levels peaked immediately after awakening and declined to a trough around bedtime. The ANOVA with repeated measures revealed a statistically significant main effect of samples (F(2,42) = 53.6, p < 0.01), where the levels of cortisol in samples A and B were different from the levels in sample C ( p < 0.05). This confirms normal diurnal patterns of cortisol secretion in these children. There were no effect of conditions and no interaction of samples by conditions (F < 1), which suggests that cortisol levels in each condition (PRE, DOG, POST) were the same. This was true for each sample (A, B, or C).
1190
R. Viau et al.
Figure 1 The change in problematic behaviors of children with Pervasive Developmental Disorder as reported by parents prior to (PRE), during (DOG) and after (POST) introducing a service dog to the family. Bars are standard errors of the mean and units are arbitrary. * denotes significant statistical difference ( p < 0.05). The questionnaire is a semi-quantitative, open-ended questionnaire focused on problematic behaviors as reported by parents of autistic children. An initial score of such behaviors was established during the two weeks prior to the introduction of the service dog. Then, each week, based on the parents’ account of initial behaviors, 2, 1, or 0 points were subtracted from the initial score if a change, a small change, or no change, respectively, was noticed.
Figure 2 The diurnal cortisol profiles of children with ASD in the three experimental conditions: prior to (PRE; diamonds), during (DOG; squares), and after (POST; triangles) the introduction of service dogs to these children’s families. Sample A represents cortisol levels as measured upon awakening, sample B refers to cortisol levels measured 30 min after awakening, and sample C refers to cortisol levels measured at bedtime. Bars represent standard error of the means and values are in mg/dL.
Some difficulties were experienced when collecting saliva from some children. Consequently, we discontinued collecting the second set of samples (sample B, after awakening) for the last 8 families recruited. These 8 children for whom we did not calculate a CAR did not differ from the others in terms of demographic characteristics or average cortisol levels. We were able to compute a CAR for 34 children. In contrast with the average cortisol secretion, we found a significant effect of condition on the children’s CAR (Fig. 3). The CAR was found to be considerably reduced when dogs were present, plunging from an increase of 58% prior to the introduction of the dogs to 10%. The CAR rose to 48% when dogs were removed from the families. A statistical main effect of conditions (F(2,48) = 5.2, p = 0.01) was found when CAR was entered into an ANOVA with repeated measures. Fig. 2 shows that the decrease in CAR after introducing the dogs appears to be due to an increase in awakening cortisol (sample A), but an ANOVA with repeated measures on sample A only did not yield significant results (data not shown). We were interested in learning whether the reduction in CAR was due to an adaptation to the dog’s presence or not. To
this effect, we divided the conditions in two-week periods: weeks 1—2 prior to introducing the dog, weeks 3—4 immediately after introducing the dog, weeks 5—6 with the dog, weeks 7—8 after removing the dog from the family. We then conducted an ANOVA with repeated measures (Fig. 3), which revealed a gradual decline in CAR, followed by a return to a higher CAR after the dogs were removed from the families (F(3,72) = 3.6, p = 0.02). Least significant difference posthoc analyses showed that results from weeks 1 to 2 (prior to the introduction of the service dogs) were higher than those from weeks 3—4 to 5—6 (with the dogs), and that results from weeks 5 to 6 (the last two weeks with the dogs) were lower than those from weeks 7 to 8 (the two-week period after the removal of the dog).
4. Discussion In this study, we assessed the effects of service dogs on the basal cortisol and on the Cortisol Awakening Response (CAR) of autistic children. We found that the introduction of service dogs had a significant effect on the CAR of autistic children,
MIRA, austim and cortisol
Figure 3 Upper panel represents the Cortisol Awakening Response (CAR) in children with ASD prior to (PRE), during (DOG) and after (POST) the introduction of service dogs to the families. The lower panel represents the CAR in children with ASD per week. WK1-2 is the mean CAR of the first two weeks, prior to the introduction of the dogs; WK3-4 is the mean CAR of the first two weeks while the dogs were present; WK5-6 is the mean CAR of the two following weeks with the dogs; and WK7-8 is the mean CAR of the two weeks after the removal of the dogs from the families. Bars represent standard errors of the means, units are percentage of increases. * denotes significant statistical difference ( p < 0.01).
reducing it from a 58% increase prior to the introduction of the service dogs to 10%. Their CAR jumped back to 48% once dogs were removed from the children’s environment. This reduced CAR was not simply driven by an increase of cortisol concentration in the first salivary samples, since an ANOVA on morning samples did not reach statistical significance. At this time, it is difficult to attribute a precise functional significance to the changes in CAR observed in these children, because few studies have evaluated this parameter in a young population. We know that healthy children have a CAR of 30%, which is considerably lower than the range of 50—150% found in healthy adults (Rosmalen et al., 2005). In adults, CAR has been shown to be affected by various psychological factors (Clow et al., 2004; Fries et al., 2009). More precisely, recent single-case and review studies suggest that general stress is correlated with an increased CAR, and that a positive psychological state is correlated with a reduced CAR (Chida and Steptoe, 2009; Stalder et al., 2009). It is possible that the changes observed in the CAR of these autistic
1191 children were related to the psychological factors linked to the presence of service dogs. Interestingly, in a qualitative study examining the effects of service dogs in families of autistic children, Burrows et al. (2008) found that parents reported that their child was calmer or happier when the dog was present. However, these explanations remain hypothetical, as we did not measure the children’s happiness or subjective stress. Alternatively, the changes in the CAR of autistic children could be explained by changes in their sleep patterns. We know that autistic children suffer from disturbed sleep patterns (see Johnson and Malow, 2008 for review). Although we did not measure the quality of sleep, the wake-up times, or episodes of sleep disturbance in our study, anecdotal data from parents allow us to think that the presence of the service dogs might have encouraged improved sleep. This is congruent with Burrows et al. (2008) who found that besides improving the parents’ sleep, some autistic children loved sleeping with their dog and fell asleep faster than before. Although previous studies have failed to demonstrate a significant effect of sleep disturbances on CAR in healthy individuals (Dettenborn et al., 2007), Backhaus et al. (2004) found that subjective sleep disturbances in healthy adults are associated with lower morning cortisol levels. Further research is needed to understand the relationships between the service dog’s presence, sleep patterns, and the cortisol secretion of autistic children. The elimination of the second set of saliva samples (sample B, taken 30 min after awakening) from the protocol during the course of the study after reports of difficulties in saliva sampling bears some interesting questions. This 30min lapse is within the stress response time (cortisol secretion peaks 30—60 min after stress exposure). Children who were reluctant to give saliva samples upon awakening might have shown a stress response, which may have been reflected in the second sample. If this were true, the reduction in CAR while service dogs are present suggests that the dogs may have simply facilitated the saliva sampling and reduced the cortisol stress response of the children. This would bring further support to the psychological effect of service dogs on autistic children. Alternatively, parents participating in the Burrows et al.’s (2008) study reported that the service dogs used in their study served as an anchor, preventing their children from bolting. This could also explain the results presented here, as it was suggested that the child be attached to the service dog’s harness. However, we do not know which families followed this suggestion. Should the second saliva samples be a measure of stress response, anchoring the children to facilitate the cortisol sampling should diminish this reaction. Children with ASD may be more resistant to changes in routine (Kanner, 1943). Furthermore, the introduction of the service dogs might have significantly altered the family routine and structure. If the reduction in CAR was due to a change in the family structure after introducing the dog, thereby resulting in a stress adaptation, we should have seen a sharp decline immediately after the introduction of the service dogs, followed by a slow return to previous CAR levels. Instead, we found that the reduction in CAR seemed more pronounced during the last two weeks with the dogs, which suggests that the dog’s presence was responsible for this effect, as opposed to an adaptation to change.
1192 We did not find differences in average cortisol levels between conditions. More specifically, the basal salivary cortisol levels of autistic children before, during and after the introduction and removal of service dogs did not differ statistically. This could be explained by the fact that the use of dynamic measures of cortisol secretion may be more appropriate to reveal differences with healthy children. Previous studies indicate that autistic children exhibit normal basal cortisol levels (Tordjman et al., 1997; Jansen et al., 1999, 2000). Conversely, other studies indicate that autistic children are under-responsive to psychological stressors, which means that they secrete less cortisol than healthy children do when exposed to psychological stressors (Jansen et al., 1999, 2000, 2003). Finally, autistic children have been found to be non-suppressors of the DEX test (Lam et al., 2006). Any interpretation would remain speculative, as we did not include a control group in our study. It would have been interesting to find that autistic children secreted basal salivary cortisol levels comparable to that of healthy children, while exhibiting a different CAR. In addition, it may be useful to study whether the CAR of healthy children is influenced by service dogs in the same way as autistic children. Our study is the first one to report interventioninduced changes in autistic children’s cortisol secretion. We believe that the within-subject design of this study provides some degree of reliability in our interpretation that service dogs have an effect on the CAR of children with ASD. As a secondary goal, we wanted to examine if changes in cortisol secretion would be reflected in the child’s behavior. In addition, since all children do not have the same reaction to animals, it was deemed important to show that service dogs had a beneficial effect on the children studied. We found that the frequency of problematic behaviors reported by parents decreased when service dogs were present. This effect persisted when the dogs were removed from the family for two weeks. However, no correlations were found between CAR or average cortisol levels, and the number of disruptive behaviors reported by parents. Due to the qualitative nature of the initial behavioral assessment, the resulting range of problematic behaviors was quite large. This might explain why behavior did not change when the service dog was removed from the family. Parents reported that the service dog had a positive effect on their child’s behavior, as seen through a decrease in the frequency of self-stimulation episodes, repetitive behaviors and tantrums, as well as an increase in the child’s tolerance to some noises, such as household appliances. These behavioral changes may reflect the parents’ satisfaction with the experiment more than an actual modification in their child’s behavior. Due to extraordinary circumstances, it was impossible to perform further study on the type, nature, and scale of the behavioral changes observed in these children. An unbiased third-party’s evaluation may prove to be more appropriate than the parents’ comments. This will be done in a future study. Furthermore, the questionnaire did not measure perceived stress or its effect on the children before and after the introduction of service dogs. Given what is known about the psychological effects on CAR and the psychological effect of service dogs on autistic children, we believe that the relationship between a reduced CAR and reduced disruptive behaviors occurs through the psychological effect of service dogs on autistic children.
R. Viau et al. In summary, our exploratory study showed that service dogs have a beneficial effect in reducing CAR and the number of disruptive behavioral incidents in children with ASD. Literature on the benefits of service dogs is expanding and points to definitive benefits for families of children diagnosed with ASD. Our study brings further support to parental reports on the benefits of service dogs by showing physiological effects (reduced CAR) in autistic children, which may correspond to the physiological effects observed in adults who experience positive psychological states. Additional studies are needed to determine whether these physiological changes are correlated to measures of perceived stress in these children, for instance.
Role of funding source This study was founded by La Fondation MIRA, Quebec, Canada. La Fondation MIRA had a role in study design and in the collection of data, but had no role in the analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest None to declare.
Acknowledgement This article is published in memory of Dr. Robert Viau, the instigator of this study.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Association, Washington, DC (Text Revised). Backhaus, J., Junghanns, K., Hohagen, F., 2004. Sleep disturbances are correlated with decreased morning awakening salivary cortisol. Psychoneuroendocrinology 29 (9), 1184—1191. Barker, S.B., Dawson, K.S., 1998. The effects of animal-assisted therapy on anxiety ratings of hospitalized psychiatric patients. Psychiatr. Serv. 49 (6), 797—801. Burrows, K.E., Adams, C.L., Spiers, J., 2008. Sentinels of safety: service dogs ensure safety and enhance freedom and well-being for families with autistic children. Qual. Health Res. 18 (12), 1642—1649. Camp, M.M., 2001. The use of service dogs as an adaptive strategy: a qualitative study. Am. J. Occup. Ther. 55 (5), 509—517. Champagne, N., St-Pierre, E., Viau, R., 2004. A longitudinal study and practice on the blind youth and the guide dog. , http:// www.mira.ca. Chida, Y., Steptoe, A., 2009. Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis. Biol. Psychol. 80 (3), 265—278. Clow, A., Thorn, L., Evans, P., Hucklebridge, F., 2004. The awakening cortisol response: methodological issues and significance. Stress 7 (1), 29—37. Corbett, B.A., Mendoza, S., Wegelin, J.A., Carmean, V., Levine, S., 2008. Variable cortisol circadian rhythms in children with autism and anticipatory stress. J. Psychiatry Neurosci. 33 (3), 227—234. Davis, B.W.N., O’Brien, K., Patronek, S., MacCollin, G.M., 2004. Assistance dog placement in the pediatric population: benefits,
MIRA, austim and cortisol risks, and the recommendations for future application. Anthrozoos 17 (2), 130—145. Dettenborn, L., Rosenloecher, F., Kirschbaum, C., 2007. No effects of repeated forced wakings during three consecutive nights on morning cortisol awakening responses (CAR): a preliminary study. Psychoneuroendocrinology 32 (8—10), 915—921. Eddy, J., Hart, L.A., Boltz, R.P., 1988. The effects of service dogs on social acknowledgments of people in wheelchairs. J. Psychol. 122 (1), 39—45. Fecteau, S., 2008. Inte ´gration d’un chien d’assistance dans la famille d’un enfant pre ´sentant un TED: la contribution de la the ´orie de l’attachement parent-enfant. Universite ´ de Sherbrooke, Sherbrooke, Que ´bec. Fombonne, E., 2003. Epidemiological surveys of autism and other pervasive developmental disorders: an update. J. Autism Dev. Disord. 33 (4), 365—382. Fries, E., Dettenborn, L., Kirschbaum, C., 2009. The cortisol awakening response (CAR): facts and future directions. Int. J. Psychophysiol. 72 (1), 67—73. Jansen, L.M., Gispen-de Wied, C.C., Jansen, M.A., van der Gaag, R.J., Matthys, W., van Engeland, H., 1999. Pituitary—adrenal reactivity in a child psychiatric population: salivary cortisol response to stressors. Eur. Neuropsychopharmacol. 9 (1—2), 67—75. Jansen, L.M., Gispen-de Wied, C.C., Van der Gaag, R.J., ten Hove, F., Willemsen-Swinkels, S.W., Harteveld, E., et al., 2000. Unresponsiveness to psychosocial stress in a subgroup of autistic-like children, multiple complex developmental disorder. Psychoneuroendocrinology 25 (8), 753—764. Jansen, L.M., Gispen-de Wied, C.C., van der Gaag, R.J., van Engeland, H., 2003. Differentiation between autism and multiple complex developmental disorder in response to psychosocial stress. Neuropsychopharmacology 28 (3), 582—590. Johnson, K.P., Malow, B.A., 2008. Sleep in children with autism spectrum disorders. Curr. Neurol. Neurosci. Rep. 8 (2), 155—161. Kanner, L., 1943. Autistic disturbances of affective contact. Nerv. Child 2, 217—250.
1193 Kirschbaum, C., Hellhammer, D.H., 1989. Salivary cortisol in psychobiological research: an overview. Neuropsychobiology 22 (3), 150—169. Lam, K.S., Aman, M.G., Arnold, L.E., 2006. Neurochemical correlates of autistic disorder: a review of the literature. Res. Dev. Disabil. 27 (3), 254—289. Mader, B., Hart, L.A., Bergin, B., 1989. Social acknowledegments for children with disabilities: effects of service dogs. Child Dev. 60 (6), 1529—1534. Martin, F., Farnum, J., 2002. Animal-assisted therapy for children with pervasive developmental disorders. West. J. Nurs. Res. 24 (6), 657—670. McNicholas, J., Collis, G.M., 2000. Dogs as catalysts for social interactions: robustness of the effect. Br. J. Psychol. 91 (Pt 1), 61—70. Odendaal, J.S., Meintjes, R.A., 2003. Neurophysiological correlates of affiliative behaviour between humans and dogs. Vet. J. 165 (3), 296—301. Redefer, L.A., Goodman, J.F., 1989. Brief report: pet-facilitated therapy with autistic children. J. Autism Dev. Disord. 19 (3), 461—467. Rosmalen, J.G., Oldehinkel, A.J., Ormel, J., de Winter, A.F., Buitelaar, J.K., Verhulst, F.C., 2005. Determinants of salivary cortisol levels in 10—12 year old children; a population-based study of individual differences. Psychoneuroendocrinology 30 (5), 483—495. Stalder, T., Hucklebridge, F., Evans, P., Clow, A., 2009. Use of a single case study design to examine state variation in the cortisol awakening response: relationship with time of awakening. Psychoneuroendocrinology 34 (4), 607—614. Tordjman, S., Anderson, G.M., McBride, P.A., Hertzig, M.E., Snow, M.E., Hall, L.M., et al., 1997. Plasma beta-endorphin, adrenocorticotropin hormone, and cortisol in autism. J. Child Psychol. Psychiatry 38 (6), 705—715. Tu, M.T., Lupien, S.J., Walker, C.D., 2006. Diurnal salivary cortisol levels in postpartum mothers as a function of infant feeding choice and parity. Psychoneuroendocrinology 31 (7), 812—824. Viau, R., Champagne, N., 2004. Les effets du chien dans la vie des humains: recension d’e ´crits. , http://www.mira.ca.