Equine-Assisted Therapy for Children with Autism Spectrum Disorder: a Comprehensive Literature Review

Review Paper


Equine-assisted therapy (EAT) is a therapeutic technique that has been examined for improving physical, emotional, social, cognitive, educational, and behavioral skills in children diagnosed with autism spectrum disorder (ASD). This literature review examines the usefulness of the approach for children diagnosed with ASD. The goal of the search strategy used was to include relevant scientific articles published examining EAT for children diagnosed with ASD. Of the 12 studies examined, 11 researchers demonstrated efficacy for increased physical and social functioning, communication, sensory sensitivity, sensory motivation, self-regulation, adaptive skills, motor skills, improved volition, as well as decreased aberrant behavior and severity of symptoms. The majority of the research documented improvements in functioning; however, these conclusions were qualified by numerous factors that limit the interpretation of the results.


Equine-assisted therapy Autism spectrum disorder 

Although applied behavioral analysis (ABA) is considered the most evidence-based therapeutic approach for children with autism spectrum disorder (ASD), a number of alternative therapies exist, including dietary treatments, art methods, music, and occupational therapy (Lai et al. 2014). Although these methods are frequently used for children, the empirical support of their efficacy is lacking. Equine-assisted therapy (EAT) has recently become a very popular approach for children with ASD. EAT offers additional treatment for parents looking for options to help their child diagnosed with ASD; however, similar to other alternative approaches, questions have been raised concerning the efficacy of this method (Jenkins and DiGennaro Reed 2013). This review examines and analyzes the empirical literature surrounding EAT for children diagnosed with ASD.

The rate of ASD is now 1 in 68 children (Centers for Disease Control 2014; Falco 2014). When a child is diagnosed with ASD, it is important to carefully consider the treatment options available. Given the myriad of treatments applicable, it can be very difficult to select a treatment that will actually help the child. Empirical research provides guidance regarding which interventions are the most appropriate for specific clinical issues. It can help parents to discover which methods are effective as well as allow families to be conversant in which ones are not effective or even detrimental. Professionals in various disciplines have developed many alternative ways of attempting to treat children with ASD, such as EAT. However, for EAT to be a reasonably valid option for children diagnosed with ASD, certain criteria must be evaluated. Evidence-based practice (EBP) includes treatments that are supported by thorough empirical studies utilizing appropriate controls (Jenkins and DiGennaro Reed 2013). EBP is important because it allows clients and clinicians to rely on empirically validated interventions. This paper examines whether or not EAT reaches these standards for children with ASD.

In the sixteenth century, medical professionals began to notice that horseback riding not only improved physical health, but also improved the mind (Weber and Westmoquette 2010). The use of horses to provide therapeutic assistance for patients was further documented in 1901 when Dame Agnes Hunter founded the Orthopedic Hospital in Oswestry (Bustad and Hines 1984). Hunter recognized the unique qualities of the horses in providing companionship, laughter, and restoration of health. Empirical research did not begin until the 1960s; however, research has substantially increased in the last 10 years (Selby and Smith-Osborne 2013). More researchers became interested in EAT and began empirically evaluating the use of EAT for numerous physical and mental benefits. Researchers examined the effects of EAT on self-esteem, depression, empathy, and self-concept (e.g., Bowers and MacDonald 2001; Cawley et al. 1994; Gatty 2001; Poresky 1996). Researchers also began to examine how the unique characteristics of the horse, and of the riding process, could potentially help children who had been diagnosed with ASD. Today, EAT advocates appeal to a number of different theoretical models and various underlying mechanisms of change to explain the hypothesized effects of EAT—ranging from operant models of reinforcement to occupational therapy models of sensory processing.

There are numerous types of therapeutic horsemanship that incorporate specific techniques to treat ASD. Hippotherapy is a type of EAT that includes physical, occupational, or speech therapy treatment strategies that utilize equine movement (Weber and Westmoquette 2010). This type of EAT focuses on posture, balance, and mobility (Holm et al. 2014). It provides a stimulation created from the rhythmic movement of the horse that allows for deep proprioception for the rider (Giagazoglou et al. 2013). Therapeutic horseback riding (THR) is broader than hippotherapy. It focuses on physical, social, learning, sensory, and psychological goals (Holm et al. 2014). It emphasizes attention, sensory management, and communication by teaching riding skills (Ward et al. 2013). It involves holding a horse’s reins, using voice commands to control the horse, and other horsemanship skills (Jenkins and DiGennaro Reed 2013). Equine-Facilitated Experiential Learning (EFEL) incorporates individual psychotherapy and is also used to treat children diagnosed with ASD (Selby and Smith-Osborne 2013). It focuses on the notion that the therapy must make the patient physically and emotionally uncomfortable in order to help them change (Selby and Smith-Osborne 2013). Children with ASD have also been treated with Equine-Facilitated Psychotherapy (EFP), which is a more applied and hands-on approach to facilitate compulsory moves, teamwork in horse care, communication, and self-expression (Vidrine et al. 2002). The current review analyzed studies that used THR, hippotherapy, or were unspecified (general EAT).


The search strategy used in this literature review used key word searches in Psych Info and Google Scholar. Key word searches include “equine therapy for children with ASD,” “hippotherapy,” “equine assisted activities,” “nontraditional therapies for children with ASD,” and “animal assisted therapy for children with ASD.” This strategy maximized chances to discover scientific articles published since 2009 regarding equine-assisted therapy for children with ASD. The goal of this strategy was to include relevant scientific articles published examining EAT for children diagnosed with ASD. In addition to this strategy, references from each of the articles found were further examined to expand the literature base. Each reference was assessed to discover the nature of the article, if it was a scientific journal, and if the source fit the criteria for this literature review. Researchers that specifically examined EAT for children with ASD, and were published in scientific journals, were then analyzed and included in the current literature review. Twelve studies were discovered using this strategy and each was analyzed for the quality of scientific inquiry and the validity of the conclusions.


Dependent Variables and Measures

Various dependent variables and measurements were utilized in the 12 studies to assess the benefits of EAT. Table 1 includes a brief outline of each study analyzed. Bass et al. (2009) measured social functioning and severity of symptoms using the Social Responsiveness Scale (SRS; Contstantino and Gruber 2005) and the Sensory Profile (SP; Dunn 1999). The children in the experimental group demonstrated improvements in sensory seeking, sensory sensitivity, sensory integration, and social motivation. They also witnessed a reduction in sedentary behavior and inattention/distractibility (Bass, Duchowny, and Llabre et al. 2009).
Table 1

A brief outline of each study analyzed





Bass et al. (2009)

-Pre-Posttest Design with Random Assignment and high internal validity

-Control group

-Standardized Questionnaires with high internal consistency

-Structured lesson plan

-Relatively Small sample size

-Low external validity

-Did not indicate if benefits were maintained

Effective for increasing social motivation, sensory seeking, sensory integration, directed attention, sensory sensitivity and reduction in sedentary behavior, inattention/distractibility

Gabriels et al. (2012)

-Control group

-Multiple sources of data

-Pre-Posttest Design with high internal validity

-Low external validity due to small sample size

-Possible Rater bias

-Results were not corrected

-No examination of long-term effects

-Lacked standardized THR intervention protocol

Effective for self-regulation behaviors (irritability, lethargy, stereotypical behavior, hyperactivity) expressive language, motor skills, and adaptive learning skills

García-Gómez et al. (2014)

-Control group

-BASC collects data of children in different contexts

-“Quasi-Control” Test-Retest Design, no random assignment, poor internal validity

-Small sample size, lowering external validity

-No indication that benefits were maintained

Effective for reducing aggressiveness

Holm et al. (2014)

-Single-Subject ABA Design

-Examined behavioral changes magnitude correspondence with dose of THR

-Multiple sources of data in multiple settings

-Standardized measures

-Small sample size, lowering external validity

-Parents were not masked to the general purpose of study

-No control group

Effective for decreased target behaviors nominated by parents such as clapping, chewing, snapping, spontaneous verbal communication, and pounding surfaces

Jenkins and DiGennaro Reed (2013)

-Single-Subject Multiple Baseline Across Participants Design

-Control group

-Low external validity due to small sample size and single-case design

-Only evidence for efficacy is anecdotal

-Failure to replicate

-Used self-report forms

Not effective for response to others’ initiations, or spontaneous initiations. Not effective for off-task behavior, compliance, or problem behavior

Kern et al. (2011)

-Pre-Posttest Design with Follow-Up

-Standardized measures

-No control group

Effective for interactive competencies, reduction in severity of symptoms, improvements on Mood and Tone subscale of questionnaire, and improved quality of life

Lanning et al. (2014)

-Control Group

-Control Group was a Social Circle

-High reliability of measures

-Information indicating what point behavior change occurred

-Appropriate statistical analysis to address issues related to small sample size

-Longitudinal Quasi-Experimental Design with Repeated Measures, no random assignment, low internal validity

-Small sample size

-No knowledge of prior experiences to equines

for parent reports

-Varied lesson plans based on individual

Effective for improvements in physical, emotional, and social functioning

Nelson et al. (2011)

-Single-Subject ABAB Reversal Design, good internal validity

-Direct observation of three boys allowed for detailed explanations for each participant

-Small sample size due to Single-Subject design, low external validity

Effective for increasing social behavior and decreasing aberrant behavior

Taylor et al. (2009)

-Appropriate use of PVQ

-No control

-Small sample size

Effective for increasing volition over time

Van den Hout and Bragonje (2010)

-Prospective Single-Blinded Clinical Trial, high internal validity

-Large sample size, high external validity

-Included analysis of magnitude

-Included differences in terms of severity of ASD

-Scales had strong psychometric properties

-Included examination of concurrent therapies

-Only the parent-rated measure was significant

-No control group

-Majority of participants participating in a concurrent therapy

Demonstrated minor improvements in communication, and large improvements in sociability, and sensory/cognitive awareness

Ward et al. (2013)

-Interrupted Time Series Design examined if behavioral changes were maintained and reinstated

-Single-Group Quasi-Experimental and Interrupted Time Series Design, no random assignment, low internal validity

-Relatively small sample size

-Did not include a control group

Effective for improved social interaction, sensory processing, and decreasing severity of symptoms. The effects were not sustained during breaks, but were reestablished once treatment began

Weber and Westmoquette (2010)

-Great amount of information for further generation of hypotheses

-Gather information about behavior in different environments

-Qualitative Design (semi-structured interviews)

-No proof of cause and effect; no direction comparison between differential reinforcement of alternative behavior with and without equine

-Limitations of interviews

-No standardized measurements

Interviewees reported improvements in physical and psychological developments such as developments in communication, motivation, awareness, interest in surroundings, self-confidence, and reduction in fears or phobias

Gabriels et al. (2012) analyzed self-regulation, adaptive skills, and motor skills for children ages 6–16 diagnosed with ASD. Self-regulation was measured using the Aberrant Behavior Checklist-Community (ABC-C; Aman and Singh 1994). Adaptive skills were analyzed using the Vineland Adaptive Behavioral Scales-Interview Edition (VABS-II; Sparrow et al. 2005). Motor skills were measured using Bruininks-Oseretsky Test of Motor Proficiency (BOT-2, Short Form; Bruininks and Bruininks 2005). The children demonstrated significant improvements in self-regulation (irritability, lethargy, stereotypical behavior, hyperactivity), as well as improvements in motor skills and expressive language when compared to a waitlist control group (Gabriels et al. 2012).

García-Gómez et al. (2014) measured three variables: social, affective, and emotional. These variables were assessed using the multidimensional Behavior Assessment System for Children (for teachers) test battery (BASC-T; Reynolds and Kamphaus 1992). They also measured variables relating to quality of life using an ad hoc parent questionnaire based on the Quality-of-Life Model (Schalock and Verdugo 2002). The BASC-T analysis allowed the clinicians to show significant differences in aggressiveness. The researchers found significance in two of the dimensions of the Quality-of-Life Questionnaire (“interpersonal relations” and “social inclusion”), but the total score on the questionnaire was not significant (García-Gómez et al. 2014).

Holm et al. (2014) evaluated target behaviors nominated by caregivers. The target behaviors included tensing/clenching facial muscles, snapping fingers, spontaneous verbal communication of wants and needs, pounding surfaces with hands, pushing nose, clapping, echolalia/scripting, and mouthing/chewing fingers/hand and non-edible objects. The researchers used standardized measured of change, completed by parents, and the ABC-C (Aman and Singh 1994). Holm et al. (2014) found that 70% of the target behaviors improved during the intervention.

Jenkins and DiGennaro Reed (2013) measured affect (happiness and unhappiness) defined by facial expressions, responses to initiations, spontaneous initiations (unprompted language), off-task behavior (motor or verbalization), compliance (following direction), commands to direct the horse, posture, and problem behaviors using the Child Behavior Checklist (CBCL/6-18; Achenback and Rescoria 2001). The clinicians did not find any systematic changes in any of the measurements. They concluded that THR was not an effective intervention; however, anecdotal observations from the parents indicated language improvements at home. The researchers concluded that THR was a leisure activity rather than a treatment option for children with ASD since the only evidence for efficacy was anecdotal (Jenkins and DiGennaro Reed 2013).

Kern et al. (2011) examined the severity of symptoms using the Childhood Autism Rating Scale (CARS; Schopler et al. 1988) and the Timberlawn Parent–child Interaction Scale (reduced version of the Timberlawn Couple and Family Evaluation Scales, TCFES; Lewis 1989). The results revealed improvements in interactive competencies and a reduction in the severity of the children’s symptoms. The only changes measured in the Timberlawn Parent–child Interaction Scale were in the Mood and Tone subscale (Lewis 1989). The researchers found improved quality of life as measured by the Quality of Life Enjoyment and Satisfaction Questionnaire: General Activities Subscale (Endicott et al. 1993). The only differences in the Sensory Profile were on the High Threshold Auditory Processing. The Treatment Satisfaction Survey (TSS) ratings led the researchers to conclude the parents were very satisfied with the treatment for their children (Kern et al. 2011).

Lanning et al. (2014) assessed changes in health-related quality of life using the Pediatric Quality of Life 4.0 Generic Core Scales (PedsQL; Varni et al. 2001) and measured unique physical and psychosocial concepts using the Child Health Questionnaire (CHQ; Langraf and Ware 2008). The measurements indicated improvements in quality of life domains. Parents noted progression in areas of social functioning, physical functioning, overall mental health and behavior, general behavior, and school functioning. Children in the EAT group had greater improvement than those in the non-equine treatment. The results indicated a high probability that a child diagnosed with ASD would improve in physical and social functioning while participating in an EAT treatment.

Nelson et al. (2011) measured social and aberrant behavior for three boys diagnosed with ASD. Social behavior was defined as imitation of verbal responses or sounds, verbal commands or sounds directed towards the therapist, and unprompted verbal phrases or requests. Aberrant behavior was defined as crying, screaming, hiding or running away, purposefully falling down, whining, and hitting. Behavior was recorded through evaluation and videotapes that were scored using a 6-s interval system recording classification. Nelson et al. (2011) found a significant increase in social behavior and a significant decrease in aberrant behavior.

Taylor et al. (2009) examined volition in three children. Volition includes interests, self-efficacy, and motivation to engage in new activities. The Pediatric Volitional Questionnaire (PVQ; Basu et al. 2008) was used to demonstrate improved volition for the three children (Taylor et al. 2009).

Van den Hout and Bragonje (2010) examined overall change and effects on specific areas of functioning such as communication, sociability, sensory/cognitive awareness, and physical behavior using CARS (Schopler et al. 1988) and the Autism Treatment Evaluation Checklist (ATEC; Rimland and Edelson 1999). The results indicated a relatively small decrease in symptoms, but EAT was significant overall. The researchers found that the overall improvements were relational to the number of EAT lessons and the baseline ASD severity, which indicated that those who received lessons more frequently demonstrated larger improvements. Only a minor percent indicated an increase in communication and physical behavior; the largest effects were in sociability and sensory/cognitive awareness (Van den Hout and Bragonje 2010).

Ward et al. (2013) examined the effect of THR on social communication and sensory processing skills using the Gilliam Autism Rating Scale-2 (GARS-2; Gilliam 2006) and the Sensory Profile School Companion (SPSC; Dunn 2006). They found that children improved in social interaction and sensory processing. The researchers also discovered a significant decrease in the severity of symptoms associated with ASD (Ward et al. 2013).

Weber and Westmoquette (2010) conducted semi-structured interviews, 20–40 min long, of four parents and one caregiver to evaluate the behavior changes of children diagnosed with ASD participating in an EAT treatment. The interviewees recognized physical, social, and psychological developments. For example, the interviewees saw that the children expressed joy and enjoyed EAT which improved the children’s well-being. It was reported that the children’s communication with the parents/caregivers improved as well as general verbal communication and understanding of non-verbal communication. They also informed that the children experienced developments in awareness and self-confidence, as well as a decrease in fears (Weber and Westmoquette 2010).


One study used a qualitative research design to examine EAT for children with ASD (Weber and Westmoquette 2010). This type of study can be useful at early stages of research to help generate further hypotheses by disclosing the richness of the experience. However, this design did not provide any indication of an actual cause and effect relationship between any improvements observed and the EAT treatment (Weber and Westmoquette 2010). The interviews conducted in the study could be revealing and helpful for generation of further research because the interviewees spent a lot time with the children which gave them various chances to observe improvements (Weber and Westmoquette 2010).

Three of the studies used Quasi-Experimental designs which were an improvement compared to qualitative designs. Although these designs could be very useful in generation of subsequent hypotheses, these types of designs are still weak due to lack of random assignment and not specifying a cause and effect relationship (García-Gómez et al. 2014; Lanning et al. 2014; Ward et al. 2013). This lowers the internal validity of the studies which increases the possibility of the presence of confounding variables.

Four of the current studies utilized Single-Subject Designs (Holm et al. 2014; Jenkins and DiGennaro Reed 2013; Nelson et al. 2011; Taylor et al. 2009). This type of design provided proof of cause and effect which increased the internal validity of the study. However, as a result of small sample size, these studies have low external validity. This hinders the ability to generalize to the larger population of interest.

Four of the studies used designs with randomized controlled trials. Kern et al. (2011) used a Pre-Posttest Design with Follow-up. Bass et al. (2009) and Gabriels et al. (2012) used Pre-Posttest Designs. Van den Hout and Bragonje (2010) used a Prospective Single-Blinded Clinical Trial. These studies had high internal validity which meant these researchers could demonstrate a causal relationship between EAT and the improvements observed and measured, free of the confounds of other variables. These designs have high external validity; however, in order for these true experiments to be generalized to the population of interest, a larger sample size is necessary.

One of the main limitations of research on EAT is limited number of participants which lowers the generalizability of any findings. Many of the studies examined had small sample sizes which lowered the external validity of the studies. The use of a larger sample size would have increased the probability that the participants used in the study represent the larger population of interest. Five of the studies had sample sizes between 3 and 8 (García-Gómez et al. 2014; Holm et al. 2014; Jenkins and DiGennaro Reed 2013; Nelson et al. 2011; Taylor et al. 2009). Five of the studies had samples sizes between 13 and 26 (Bass et al. 2009; Gabriels et al. 2012; Kern et al. 2011; Lanning et al. 2014; Ward et al. 2013). Van den Hout and Bragonje (2010) had 60 participants. Weber and Westmoquette’s (2010) qualitative study had five interviewees who were parents or legal guardians of the children. Most of the studies analyzed in the current review used smaller sample sizes which did not permit generalization to the larger target population. The use of a small sample size increases the possibility that the participants may have differed in a meaningful way from other children diagnosed with ASD—such as in their functional level or experiences outside of THR sessions. The use of larger sample sizes in future research could improve the legitimacy of the research surrounding EAT.

Use of Control Group

Only 50 % of the studies used control groups. It is important to use control groups in order to have a comparison level for the variable of interest and to statistically control for differences. However, out of the six that used control groups, five were waitlist control groups. Waitlist comparison can be beneficial in that it allows every individual to have a chance to receive the new treatment at some point. Nevertheless, a limitation of using a waitlist control is the differential expectations between the treatment and control group. Individuals in the waitlist group do not have a reason to expect beneficial changes in behavior. Therefore, the type of control groups utilized in these studies was not completely adequate; consequently, a different type of control is needed. One type that could be employed is an attention placebo control group. This group could act as a baseline to compare the effects of the particular variable or intervention of interest. This control group would experience a treatment that is similar to the experimental treatment in order to more clearly evaluate the novel factor of interest. Although one study used a social circle control group, it still did not adequately compare to the experience of EAT (Lanning et al. 2014). A control group needs to be created that sufficiently mimics the EAT treatment experience in order to empirically examine the effect of EAT.

Type of EAT Used

Researchers only clarified two types of EAT used in the 12 studies. Eight of the studies used THR, one used hippotherapy, and three used the general classification of EAT. The majority of research on EAT did not specify a classification. For the ones that did stipulate a type of EAT, a majority examined therapeutic horseback riding (THR). It may be that the philosophy of THR is more congruent with the skill development needed. Investigation of the various types of EAT would help to create a more substantial base for future research. There is no existing pattern that may indicate that a certain approach is better, only that EAT in general is effective or ineffective. Further research should determine if the categorization is warranted, and if so which type has the most beneficial effect for children diagnosed with ASD.


Each of these studies examined the use of EAT for children with ASD. Out of the 12 studies, 11 researchers showed efficacy in some domain. However, there are various limitations that could attenuate confidence in the results of these studies. Although this is not a meta-analytic examination, EAT appears to have generally positive effects for children diagnosed with ASD. Most of the research has been conducted in the last 4 or 5 years which demonstrates a growing interest in this form of intervention for children diagnosed with ASD.

Overall, EAT appears effective for increasing social motivation, communication, self-regulation, quality of life, volition, awareness, as well as decreasing symptoms and aberrant behavior. The typical trends across the studies appear to be increased social behavior, social motivation, well-being, and decreased symptoms commonly associated with ASD. The improvements in behavior may be explained by appealing to various theoretical models and underlying mechanisms of change.

There are many speculations as to what active mechanisms may be promoting these positive changes in behavior. One possibility is that a child diagnosed with ASD develops heightened sense of body awareness. The child may become more aware of the movements of their arms, legs, or posture while riding a horse. The child may learn how a horse’s movements reflect their own, such as pulling the reins in order to turn a specific way at a specified location. This improved body awareness and motor planning may also be attributed to deep proprioception provided by the rhythm of the horse’s movements.

An additional component may be progression in cognition and language. The act of following directions in a rewarding activity may have stimulated natural development of cognitive concepts. Ward et al. (2013) speculated that the simple existence of the horse was reinforcing and even motivating. This incentive, combined with lessons constructed around evidence-based teaching practices, provided a layout in which the props, routines, task analysis, and scripts helped the children to develop new skills (Ward et al. 2013). García-Gómez et al. (2014) proposed that the riding improved executive control functions. This includes enhancements in attentional control. Bass et al. (2009) suggested the behavioral expectation was isolated, which made it easier for the child to understand what to attend to. They proposed the children could pay attention to the horse as a shared reference for the child and the instructor. Bass et al. (2009) proposed the treatment required the children to socially and actively engage in the activity which helped the children to focus their attention.

One of the most common challenges of children with ASD is impairments in communication (Casenhiser et al. 2014). The improvements noted in research on EAT, for production of speech, may have been due to the child making connections between their verbal commands to the horse and the horse’s behavior. Many children found the lessons enjoyable, and this novel and exciting experience may have helped the child become motivated to communicate. Gabriels et al. (2012) suggested that since the horse responded to commands, the children learned that their communication impacted the horse’s behavior. This notion appeals to a more concrete learning style.

Another important aspect to consider is the emotional bond formed between the child and the horse. The combination of social and physical communication between the child and the horse may have helped the child to relax and acknowledge developing empathy towards the animal. Researchers have suggested their findings of improved physical, emotional, and social functioning were due to the therapeutic effects of the relationship between the child and the horse (Lanning et al. 2014). Other researchers suspected the motion of riding had a calming effect that helped the children to develop new skills and decrease antagonistic behavior (García-Gómez et al. 2014; Ward et al. 2013). This bond may have been strengthened by the behavior and automatic forgiveness of the horse. A horse’s behavior is always an accurate response of what the horse is experiencing. When a horse is frightened or angry, the rider will know right away based on the behavior of the horse. This dependable behavior may have helped the children to form a connection based on trust with the animal. This bond may have served as a bridge to forming positive relationships in the child’s life.

Improvements in body awareness, cognition and language, and forming emotional bonds demonstrated by the children in the lessons are important stepping stones for human interaction. The advances in social and communication skills can be generalized to other aspects of the child’s life. Various researchers have numerous ideas as to what factors of EAT specifically helped the children; however, more research is warranted in order to isolate what factors played a role in the improvement of the behavior and well-being of children diagnosed with ASD.

It is difficult to precisely observe the therapeutic processes involved in EAT for children diagnosed with ASD for numerous reasons. Foremost, it is extremely difficult to control for previous experiences the participants may have had with horses. These experiences could have a negative or positive effect on their behavior during EAT and subsequently affect the results. Second, is that one of the chief characteristics of ASD is a difficulty revealing emotions, especially in environments that require social interaction or communication. Therefore, it is difficult to ascertain what the child is experiencing. Third, it is very difficult to achieve standardized delivery of treatment across participants because much of the treatment is individualized; therefore, each of the participants may have different experiences which can be trying to quantify. Fourth, it can be challenging to control for extraneous variables due to the difficulty of utilizing random assignment in these particular circumstances. This then leads to the possibility of alternative explanations for the effects such as possible social skill increases due to the interaction between participants in a social setting, rather than the unique experience of the horse.

Many of the studies used systematic/structured observation which was somewhat intrusive but allowed for control over the conditions. This raised the internal validity and allowed for a stream of behavior researchers could watch and code. A possible limitation of using observation methods is the Hawthorne effect in which the children may change their behavior because they are aware their behavior is being observed (Landsberger 1958; Steele-Johnson et al. 2000). This then can have effects on the external and internal validity. Another common limitation is the coding scheme. Operational definitions may vary across studies which can create even more difficulty in replication.

Many of the researchers provided respectable psychometrics for the assessments used; however, several did not. Measures that did not have psychometrics could have had issues such as begging the question and interjecting biases. This means that without demonstrating the reliability and validity of the measures used, the only evidence backing the researchers’ claim is their own statements. This can lead to unwarranted conclusions which are no better than anecdotal reports. The use of test-retest allowed for the examination of stability; however, it did not provide any indication of dosage efficacy. An additional limitation of much of the research is a lack of direct observation and measurement of behavior. Many of the studies relied on surveys and standardized assessments.

Many of these limitations are characteristic of most research surrounding ASD treatment options. Limitations specific to research on EAT include small sample sizes due to cost, the associated challenges of data collection in a real life situation, and the use of live animals in research. According to the existing research, most of the designs used small sample sizes. Although a larger sample size is desired, providing the therapy, a trained equine, equipment, and arena access for all individuals can be costly. Even though observation in a more natural environment offers respectable generalization, this type of data collection may lower the internal validity of the study (demonstration of cause and effect relationship). Finally, use of live animals can provide limitations for researchers in terms of concordance with ethical guidelines for animal use. These challenges may be difficult to address in future research.

To create a more substantial evidence base for EAT treatments, future research should focus on methods to eliminate these limitations and provide insight into new factors of EAT. One of these methods could be variations in intensity, duration, or frequency of sessions. Future research should also create better standardization and replicability. Introducing more strictly controlled experiments, with larger sample sizes, could add to the legitimacy of the evidence surrounding EAT. Researchers also need to conduct longitudinal studies that use participants with differing degrees of severity of ASD. This would allow for the examination of magnitude and frequency. It would also allow for follow-up in order to discover if EAT is effective for children diagnosed with ASD.

Future research should utilize more randomized controlled trial research designs to strengthen findings of previous studies. It is necessary for future studies to use control groups in traditional therapy rather than a waitlist control group due to differential expectations. Clinicians should also gather more information regarding concurrent therapies of the participants as well as medication regimens in order to control for confounding variables that could alter the results.

The duration of each of the studies varied from 10 sessions (Van den Hout and Bragonje 2010) to 60–70 sessions (Holm et al. 2014; Weber and Westmoquette 2010). Only one study used a longitudinal design in order to examine the effectiveness of the EAT for children with ASD over a long period of time (Lanning et al. 2014). Lanning et al. (2014) also used an interrupted design and found improvements in the measured dependent variables after 6 weeks; however, additional treatment worsened the children’s behavior. Further Longitudinal Designs are also needed in order to determine what proportions of treatment may be effective or less effective for children diagnosed with ASD. Ward et al. (2013) used a Single-Group Quasi-Experimental and Interrupted Time Series Design and found improvements in the variables of interest; nevertheless, the researchers found the improvements diminished when the treatment was removed. Ward et al. (2013) discovered the improvements were restored as soon as treatment was started again. The use of Interrupted Time Series Designs in future research is warranted in order to determine if the effects of the treatment will be maintained after the treatment is concluded.

The use of quantitative measurement is also lacking. One way to examine the efficacy of EAT is to use video recording and coding to quantify the measurements. Another interesting technique to quantify the results is through the use of a wearable electromyography device (e.g., Funahashi et al. 2014). This device uses physiological signals to detect facial expressions and smiles. This is a respectable method because it is comfortable and also allows the child to move around frequently.


The current manuscript serves as a critical in-depth evaluation of previous research on EAT for children with ASD. The utilization of EAT is increasing and is employed for numerous disabilities and diagnoses. Due to this growth, it is key to discover if this treatment is an evidence-based approach. This paper can serve as a stepping stone for future researchers in order to determine the validity and reliability of this treatment for children with ASD. It provides information regarding the types of EAT used, the research designs employed by previous researchers, the limitations and strengths of the studies, and suggestions to improve the research.

Future inquiry is vital in order to understand why EAT works and what methods of EAT are most sufficient. Although there does not appear to be a pattern in the research examined in this manuscript, it is important to ascertain which types of EAT yield significant results and which types are insufficient. This paper can serve as a guide for future researchers to identify what research designs and research settings are most effective in determining the efficacy of EAT for children with ASD.

This paper also provides a synopsis for clinicians of which behaviors were significantly improved or decreased due to EAT. The majority of the research surrounding EAT for children with ASD yielded significant results ranging from increasing physical and social functioning, communication, sensory sensitivity, sensory motivation, self-regulation, adaptive skills, motor skills, as well as decreasing aberrant behavior and severity of symptoms, and improved volition. However, these conclusions were qualified by small sample sizes, lack of randomization or control groups, low capability of replication, and low standardization, among others. For children diagnosed with ASD, EAT may be a useful adjunct treatment that can address developmental, personal, and social needs.


Compliance with Ethical Standards


This study was funded by the first author, Ayla Mapes, out of personal funds. No grant monies were involved.

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

This article does not contain any studies performed directly by any of the authors with human participants or animals—this is a review article of existing research.

Informed Consent

Informed consent is not applicable for this review article.


  1. Achenback, T. M., & Rescoria, L. A. (2001). Child behavior checklist for ages 6 to 18. Burlington: University of Vermont, Research Center for Children, Youth, and Families.Google Scholar
  2. Aman, M. B., & Singh, N. N. (1994). Aberrant behavior checklist-community. New York: Slossen Educational Publications.Google Scholar
  3. Bass, M. M., Duchowny, C. A., & Llabre, M. M. (2009). The effect of therapeutic horseback riding on social functioning in children with autism. Journal of Autism and Developmental Disorders, 39, 1261–1267. Retrieved from http://paths2learning.com/wp/wp-content/uploads/2012/11/social-functioning.pdf.CrossRefPubMedGoogle Scholar
  4. Basu, S., Kafkes, A., Schatz, R., Kiraly, A., & Kielhofner, G. (2008). The pediatric volitional questionnaire (PVQ) version 2.1. Model of human occupation clearing house. Chicago: University of Illinois at Chicago Press.Google Scholar
  5. Bowers, M. J., & MacDonald, P. M. (2001). The effectiveness of equine-facilitated psychotherapy with at-risk adolescents: a pilot study. Journal of Psychology and Behavioral Sciences, 15, 62–76.Google Scholar
  6. Bruininks, R. H., & Bruininks, B. (2005). Bruininks-Oseretsky test of motor proficiency. Eagan: Pearson Education.Google Scholar
  7. Bustad, L. K., & Hines, L. (1984). Our professional responsibilities relative to human-animal interactions. The Canadian Veterinary Journal, 25, 369–376. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1790657/.PubMedPubMedCentralGoogle Scholar
  8. Casenhiser, D. M., Binns, A., McGill, F., Morderer, O., & Shanker, S. G. (2014). Measuring and supporting language function for children with autism: evidence from a randomized control trial of social-interaction-based therapy. Journal of Autism and Developmental Disorders, 45(3), 846–857. doi:10.1007/s10803-014-2242-3.CrossRefGoogle Scholar
  9. Cawley, R., Cawley, D., & Retter, K. (1994). Therapeutic horseback riding and self-concept in adolescents with special educational needs. Anthrozoös, 7(2), 129–134. doi:10.2752/089279394787001934.CrossRefGoogle Scholar
  10. Centers for Disease Control. (2014). Prevalence of autism spectrum disorder among children aged 8 yearsautism and developmental disabilities monitoring network, 11 sites, United States, 2010. MMWR Surveillance Summaries, 63(2), 1–21.Google Scholar
  11. Contstantino, J. N., & Gruber, C. P. (2005). Social responsiveness scale. Los Angeles: Western Psychological Services.Google Scholar
  12. Dunn, W. (1999). The sensory profile: examiner’s manual. San Antonio: Psychological Corporation.Google Scholar
  13. Dunn, W. (2006). The sensory profile school companion. San Antonio: Pearson.Google Scholar
  14. Endicott, J., Nee, J., Harrison, W., & Blumenthal, R. (1993). Quality of life enjoyment and satisfaction questionnaire; a new measure. Psychopharmacology Bulletin, 29(2), 321–326.PubMedGoogle Scholar
  15. Falco, M. (2014). Autism rates now 1 in 68 U.S. children: CDC. Retrieved from http://www.cnn.com/2014/03/27/health/cdc-autism/.
  16. Funahashi, A., Gruebler, A., Aoki, T., Kadone, H., & Suzuki, K. (2014). Brief report: the smiles of a child with autism spectrum disorder during an animal-assisted activity may facilitate social positive behaviors—quantitative analysis with smile-detecting interface. Journal of Autism and Developmental Disorders, 44(3), 685–693. doi:10.1007/s10803-013-1898-4.CrossRefPubMedGoogle Scholar
  17. Gabriels, R. L., Agnew, J. A., Holt, K. D., Shoffner, A., Zhaoxing, P., Ruzzano, S., et al. (2012). Pilot study measuring the effects of therapeutic horseback riding on school-age children and adolescents with autism spectrum disorders. Research in Autism Spectrum Disorders, 6, 578–588. doi:10.1016/j.rasd.2011.09.007.CrossRefGoogle Scholar
  18. García-Gómez, A., Risco, M. L., Rubio, J. C., Guerrero, E., & García-Peña, I. M. (2014). Effects of a program of adapted therapeutic horse-riding in a group of autism spectrum disorder children. Electronic Journal of Research in Educational Psychology, 12(1), 107–128. doi:10.14204/ejrep.32.13115.
  19. Gatty, C. M. (2001). Psychosocial impact of therapeutic riding: a pilot study. Equine-Facilitated Mental Health Association, 5(2), 8.Google Scholar
  20. Giagazoglou, P., Arabatzi, F., Kellis, E., Liga, M., Karra, C., & Amiridis, I. (2013). Muscle reaction function of individuals with intellectual disabilities may be improved through therapeutic use of a horse. Research in Developmental Disabilities, 34(9), 2442–2448. doi:10.1016/j.ridd.2013.04.015.CrossRefPubMedGoogle Scholar
  21. Gilliam, J. E. (2006). The Gilliam autism rating scale (2nd ed.). Austin: Pro-Ed, Inc.Google Scholar
  22. Holm, M. B., Baird, J. M., Kim, Y., Rajora, K. B., D’Silva, D., Podolinsky, L., Mazefsky, C., & Minshew, N. (2014). Therapeutic horseback riding outcomes of parent-identified goals for children with autism spectrum disorder: an ABA’ multiple case design examining dosing and generalization to the home and community. Journal of Autism and Developmental Disorders, 44(4), 937–947. doi:10.1007/s10803-013-1949-x.CrossRefPubMedPubMedCentralGoogle Scholar
  23. Jenkins, S. R., & DiGennaro Reed, F. D. (2013). An experimental analysis of the effects of therapeutic horseback riding on the behavior of children with autism. Research in Autism Spectrum Disorders, 7(6), 721–740. doi:10.1016/j.rasd.2013.02.008.CrossRefGoogle Scholar
  24. Kern, J. K., Fletcher, C. R., Garver, C. R., Mehta, J. A., Grannemann, B. D., Knox, K. R., … Trivedi, M. H. (2011). Prospective trial of equine-assisted activities in autism spectrum disorder. Alternative Therapies in Health and Medicine, 17(3), 14–20. ISSN:1078–6791.Google Scholar
  25. Lai, M., Lombardo, M., & Baron-Cohen, S. (2014). Autism. The Lancet, 383(9920), 896–910. doi:10.1016/S0140-6736(13)61539-1.CrossRefGoogle Scholar
  26. Landsberger, H. A. (1958). Hawthorne revisited. Ithica.Google Scholar
  27. Langraf, J. M., & Ware, J. E. (2008). The CHQ user’s manual. Boston: HealthActCHQ Inc.Google Scholar
  28. Lanning, B. A., Baier, M., Ivey-Hatz, J., Krenek, N., & Tubbs, J. D. (2014). Effects of equine assisted activities on autism spectrum disorder. Journal of Autism and Developmental Disorders, 44(8), 1897–1907. doi:10.1007/s10803-014-2062-5.CrossRefPubMedGoogle Scholar
  29. Lewis, J. M. (1989). The birth of a family. New York: Brunner/Mazel.Google Scholar
  30. Nelson, K., Axtell, J., Derby, K. M., Moug, R., Berrera, S., & McLaughlin, T. F. (2011). A preliminary analysis of therapeutic horseback riding. International Journal of Social Sciences and Education, 1(4), 644–656. Retrieved from http://www.ijsse.com/sites/default/files/issues/2011/v1i4/paper%2027/paper%2027.pdf.Google Scholar
  31. Poresky, R. (1996). Companion animals and other factors affecting young children’s development. Anthrozoös, 9(4), 159–168. doi:10.2752/089279396787001437.CrossRefGoogle Scholar
  32. Reynolds, C. R., & Kamphaus, R. W. (1992). BASC: behavior assessment system for children. Circle Pines: American Guidance Service.Google Scholar
  33. Rimland, B., & Edelson, M. (1999). The autism treatment evaluation checklist (ATEC). San Diego: Autism Research Institute. Available at http://www.autism.com/ind_atec_report.asp.Google Scholar
  34. Schalock, R., & Verdugo, M. A. (2002). Quality of life for human service practitioners. Washington, DC: American Association on Mental Retardation.Google Scholar
  35. Schopler, E., Reichler, R. J., & Renner, B. R. (1988). The childhood autism rating scale. Los Angeles: Western Psychological Services.Google Scholar
  36. Selby, A., & Smith-Osborne, A. (2013). A systematic review of effectiveness of complementary and adjunct therapies and interventions involving equines. Health Psychology, 32(4), 418–432. doi:10.1037/a0029188.CrossRefPubMedGoogle Scholar
  37. Sparrow, S. S., Cicchetti, D. V., & Balla, D. A. (2005). Vineland adaptive behavior scales (2nd ed.). Circle Pines: American Guidance Service.Google Scholar
  38. Steele-Johnson, D., Beauregard, R. S., Hoover, P. B., & Schmidt, A. M. (2000). Goal orientation and task demand effects on motivation, affect, and performance. The Journal of Applied Psychology, 85(5), 724–738. doi:10.1037/0021-9010.85.5.724.CrossRefPubMedGoogle Scholar
  39. Taylor, R., Kielhofner, G., Smith, C., Butler, S., Cahill, S., Ciukaj, M., & Gehman, M. (2009). Volitional change in children with autism: a single-case design study of the impact of hippotherapy on motivation. Occupational Therapy in Mental Health, 25(2), 192–200. doi:10.1080/01642120902859287.CrossRefGoogle Scholar
  40. Van den Hout, C. M. A., & Bragonje, S. (2010). The effect of equine assisted therapy in children with autism spectrum disorders. Human Movement Sciences: psychomotor therapy. Retrieved from http://static1.squarespace.com/static/55b1504ae4b0824dcbc87ee4/t/55b542c5e4b020bb6b88d49f/1437942469809/The+effect+of+equine+therapy+in+autism.pdf
  41. Varni, J. W., Seid, M., & Kurtin, P. S. (2001). PedsQL 4.0: reliability and validity of the pediatric quality of life inventory version 4.0 generic core scales in healthy and patient populations. Medical Care, 8, 800–812Google Scholar
  42. Vidrine, M., Owen-Smith, P., & Faulkner, P. (2002). Equine-facilitated group psychotherapy: applications for therapeutic vaulting. Issues in Mental Health Nursing, 23, 587–603. doi:10.1080/01612840290052730.CrossRefPubMedGoogle Scholar
  43. Ward, S. C., Whalon, K., Rusnak, K., Wendell, K., & Paschall, N. (2013). The association between therapeutic horseback riding and the social communication and sensory reactions of children with autism. Journal of Autism and Developmental Disorders, 43(9), 2190–2198. doi:10.1007/s10803-013-177.CrossRefPubMedGoogle Scholar
  44. Weber, Y., & Westmoquette, K. (2010). Equine-assisted therapy as a treatment for children with autism spectrum disorders—a qualitative study. Karolinska Institutet, 6–19. Retrieved from http://www.ridterapi-novalis.se/.

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.Department of PsychologyColorado State UniversityFort CollinsUSA

Personalised recommendations