A Mixed Methods Investigation of Equine-assisted Occupational Therapy for Children With Autism

The proposed study will implement a mixed-methods investigation of equine-assisted
occupational therapy (EAOT) for children with autism spectrum disorder (ASD). The
quantitative strand will consist of a multiple-baseline single-case design investigating the
effects of EAOT on social functioning, self-regulation, and occupational performance of eight
children with ASD. The qualitative strand will consist of interviews with the occupational
therapists before, during, and after the 10-week intervention, focused on developing EAOT's
theory of change.

PARTICIPANTS Invesitgators will distribute fliers to local schools and ASD organizations to
recruit eight participants for the study. Investigators will screen children for inclusion
through a four-step process. First, they will call interested participants to ensure
eligibility for the study, provide basic information, and answer any questions. Next,
investigators will mail interested participants a packet that includes the SCQ, the ABC-C,
and Hearts and Horses Enrollment Packet, along with a cover letter and instructions to mail
back the completed forms, which will include a physician's signature stating the child is
medically cleared to ride horses. Next, eligible participants will attend a screening at
Hearts and Horses Therapeutic Riding Center to ensure they meet all PATH, Intl standards, and
can ride a horse for 10 minutes while following safety rules. Finally, eligible participants
will attend a final screening visit that will include administration of the ADOS-2, Leiter-3,
and Adaptive Behavior Assessment System, Third Edition (ABAS-3). This final screening visit
will take place at Colorado State University for most participants, unless a participant can
provide documentation of previous ADOS scores, in which case the screening visit can take
place at Hearts and Horses if the family would prefer. Eight participants who meet
established criteria will be enrolled in the study. If a child already had ADOS scores,
investigators will ask the caregiver to bring the clinical report, or if they do not have it
investigators will ask the caregiver to sign a HIPAA release form that would allow them to
obtain the clinical report from the hospital or clinic that administered the ADOS.

SCREENING MEASURES. Social Communication Questionnaire (SCQ). This parent-report measure is a
quick, 10-minute screen for ASD. A score of 15 or above indicates the individual likely has
ASD. The SCQ has strong discrimination between children with and without ASD and is widely
used for entry into ASD research studies.

Autism Diagnostic Observation Schedule, Second Edition (ADOS-2). The ADOS- 2 is a play-based
assessment of communication, social interaction and behaviors, and is often used to confirm a
clinical diagnosis of ASD. The ADOS has strong predictive validity for diagnosis of ASD.
Participants who meet clinical-cutoffs for ASD will be included in the study. The ADOS-2
requires extensive training to administer reliably; an investigator who has obtained
research-level reliability will administer the ADOS-2 to potential participants.

Leiter International Performance Scales, Third Edition (Leiter-3). The Leiter-3 is a
completely nonverbal measure of intelligence and cognitive abilities that is widely used with
the ASD population. The Brief IQ score can be used reliably in place of the Full IQ score
when a minimal IQ estimate is needed; therefore, this study will use the Brief IQ version of
the Leiter-3. Participants who have an IQ below 55 will be excluded from the study. A
registered and licensed occupational therapist will administer the Leiter-3.

Aberrant Behavior Checklist- Community (ABC-C). The ABC-C is a valid and reliable behavior
rating scale that measures the extent of problem behaviors in children and adults with
development disabilities. Any adult who knows the child well, in this case the caregiver, can
complete the 58-item checklist. The irritability and hyperactivity subscales will be used as
outcome measures in the study. Therefore, to avoid a ceiling effect as described above,
children included in this study must have a combined score ≥ 11 on the irritability and
hyperactivity subscales of the ABC-C, as established by Gabriels et al. (2015).

PATH, Intl. Standards. PATH, Intl provides a list of contraindications for participants who
should not participate in equine-assisted interventions. In order to ensure participants meet
all PATH, Intl standards, they will complete Hearts and Horses enrollment packet and process.
See attachments for the enrollment packet. Amongst other things, the packet includes a form
that requires a physician signature that the child does not have any of the listed
contraindications, and is medically cleared to participate in equine-assisted interventions.
Furthermore, each participant will attend a screening visit at Hearts and Horses to ensure
participants meet all other standards.

Adaptive Behavior Assessment System (ABAS-3). The ABAS-3 is a norm-referenced rating scale
that measures ten domains of adaptive behaviors and can be completed by a caregiver. The
ABAS-3 was chosen because it measures adaptive behavior in a variety of domains, therefore
providing a well-rounded picture of the child. The ABAS-3 will not be used to screen for
inclusion, but rather will be completed by the caregiver prior to the occupational therapy
evaluation in order to provide the therapist with information regarding areas of strengths
and weaknesses in the child's everyday functioning, which will help to guide the evaluation
and goal-setting.

THE INTERVENTION The intervention will occur at Hearts and Horses Therapeutic Riding Center,
which follows guidelines set forth by PATH, Intl to ensure safety of participants; for
example, instructors are PATH-certified, trained side-walkers will be present during the
intervention, and participants will wear helmets. The occupational therapists who will
deliver the intervention currently provide EAOT to children with ASD at Hearts and Horses on
a consistent basis. They are certified to provide equine-assisted interventions by PATH,
Intl.

Prior to the intervention, each participant and caregiver will participate in an evaluation
with an occupational therapist to determine current level of functioning and family
priorities. Before the evaluation, caregivers will be asked to fill out two forms and bring
them to the evaluation: a questionnaire about their child's adaptive functioning and the
sensory profile. During the evaluation, occupational therapists will conduct a
semi-structured interview with the parent and child. Next, the occupational therapist will
collaborate with the child and family to create occupation-focused goals, a process that will
be facilitated by the administration of the Canadian Occupational Performance Measure (COPM).
Occupational therapists will help to guide the family in setting occupation-focused goals
that are realistically attainable with this 10-week intervention.

Guided by the evaluation, the occupational therapist will individualize the intervention to
address each participant's unique goals, accommodate for individual learning styles (e.g.
minimize use of verbal cues), and incorporate each child's special interests or powerful
motivators. Despite this individualization, common components will be present across all
intervention sessions. Each session will be 45-60 minutes long, and the child will be mounted
on the horse for at least 25 minutes. The session will begin with a visual schedule of the
day's activities, which can then be referred to throughout the session to facilitate
transitions from one activity to another. There will be a one-to-one ratio between the child
and occupational therapist to ensure the intervention is individualized to each child's
needs. Additionally, trained volunteers will serve as horse leaders and sidewalkers during
mounted portions of the intervention; every effort will be made to pair each child with the
same horse and volunteers for the duration of the 10-week intervention. Two EAOT sessions
will occur simultaneously in the same arena, and the occupational therapists will facilitate
social interaction between participants. As described below in the quantitative data
collection section, participants will be paired into groups of two so that the session that
is occurring simultaneously is with a child with autism with a similar level of adaptive
functioning. Activities will be structured to provide positive reinforcement for
communication, and the occupational therapist will manipulate equine movement to provide
graded sensory stimulation and optimal arousal. Unmounted, or "ground activities" will be
customized to address each child's goal. For example, if a child has a goal related to
sequencing, the occupational therapist may design an activity focused on sequencing the steps
of grooming the horse. A graduate research assistant will videotape 30% of intervention
sessions, and an undergraduate research assistant blinded to the purpose of the study will
use a checklist to assess intervention fidelity of the videotaped sessions.

DATA COLLECTION: It is likely that improvements made during EAOT will be maintained after the
intervention is withdrawn, therefore a multiple-baseline design is the optimal single-case
experimental design. After the occupational therapy evaluation, participants will be paired
into groups of two based on similar levels of adaptive functioning and/or similar or
complementary treatment goals. Pairs of participants will then be randomized to a 5-week, or
7-week baseline phase. The baseline condition will be a no-treatment waiting list, during
which caregivers will fill out the ABC-C on a weekly basis, retrospectively reporting on
their child's behavior for the previous week. Caregivers will also complete a brief
questionnaire that asks about any changes in medication, therapy, or routines that week. In
addition, caregivers will rate their child's target behaviors on a VAS everyday. All of these
measures (ABC-C, parent questionnaire, and VAS) will be administered online using a Qualtrics
survey that is emailed or texted to them, depending on parent preference.

The 10-week intervention phase will consist of weekly, 45-minute EAOT intervention sessions,
described above. Throughout the intervention phase, caregivers will continue completing the
ABC-C and brief questionnaire weekly; caregivers will also continue providing VAS ratings
daily. In addition, caregivers will complete the intervention credibility scale, the SRS-2,
and the COPM on the first week and the 10th week of the intervention phase, in order to
provide a pre and post assessment of caregivers' attitude toward treatment, and child social
functioning and occupational performance. Three months after the intervention, caregivers
will complete the SRS-2, COPM, and the ABC-C a final time to determine maintenance of effects
DATA ANALYSIS: Visual inspection will be the primary method of data analysis. The Co-PI will
plot the dependent variables (y-axis) against time (x-axis) for each participant. Next, she
will inspect the data for changes in mean (average rate of performance), and trend (direction
of change over time), of the dependent variables between baseline and intervention
conditions. If the data demonstrate positive findings, the plotted graphs will be further
inspected for evidence of non-effect, such as delayed latency to change, positive trends
during the baseline phase, or variable data with overlap between phases. As a secondary
method of analysis, investigators will calculate the non-overlap of all pairs (NAP) of the
ABC-C, and VAS scores between baseline and intervention phases, which will yield an effect
size and a measure of statistical significance . Additionally, investigators will perform a
Wilcoxon matched-pairs signed-rank test to compare the scores on the SRS-2 and COPM before
and after the intervention. For these calculations, the a priori α level will be set to .05,
and Cohen's d will be calculated to generate an effect size. All calculations will be
performed in the latest version of IBM SPSS statistics.

QUALITATIVE STRAND: CONCEPTUAL DEVELOPMENT OF THEORY PARTICIPANTS: Two occupational
therapists have been recruited to participate in this study. Both occupational therapists
helped to screen the children with autism to determine if they met all PATH, Intl standards
for participation in equine-assisted therapy. Second, they have both helped to conduct
occupational therapy evaluations to determine the current level of functioning and therapy
goals for the children with autism. Third, they will provide the equine-assisted occupational
therapy to children with autism. Finally, they will participate in interviews about the
theory guiding the intervention they are delivering.

DATA COLLECTION: Qualitative data will consist of interviews with the occupational therapists
during and after the 10-week intervention. During the course of the 10-week intervention, the
Co-PI will interview both occupational therapists using a semi-structured interview guide.
Questions during the first interview will focus on the occupational therapists' theoretical
rationale for how the intervention is designed. During the intervention phase, the
investigators will observe and take fieldnotes, and a research assistant will videotape 30%
of the intervention sessions. Fieldnotes, videotapes, and transcriptions of previous
interviews will guide the creation of a new semi-structured interview guide, once again aimed
at elucidating the rationale guiding the choice of intervention components, and the theory of
change underlying each component. Overall during the intervention phase, investigators will
conduct 5 hours of interviews with each occupational therapist. After the 10-week
intervention has concluded and quantitative data have been analyzed, the Co-PI will conduct
one final hour-long interview with both occupational therapists aimed at explaining the
quantitative results. The therapists may have insights concerning why certain outcomes were
found and not others, why children responded differently to the intervention, or why trends
in the data changed throughout the course of the 10-week intervention phase.

DATA ANALYSIS: InvestigatorsI will transcribe audio-recorded interviews and perform
theoretical thematic analysis with constant comparison using the qualitative data analysis
software program, Nvivo. First, interviews will be coded according to these start codes,
derived from work on theory-guided interventions: 1- "problems" experienced by participants
2- intervention components designed to address problems, 3- theories of change and
mediational processes, and 4- intended outcomes. Next, investigators will conduct open
coding, whereby the text will be broken down, examined, conceptualized, and categorized. Data
will be sub-categorized within existing start-codes as appropriate, and new categories may
emerge from the data. After open coding, investigators will engage in axial coding, seeking
to find links and relationships between and within categories and sub-categories.
Investigators will bring emerging analyses to the practitioners for occassional member
checks, in order to ensure that ongoing analyses are consistent with their intended meanings.
Finally, selective coding will allow final themes to emerge, and will elucidate relationships
between themes.

OPTIONAL FOLLOW-UP QUALITATIVE STRAND: CAREGIVER PERCEPTION OF CHANGE PARTICIPANTS: On the
last day of therapy, caregivers of children with autism will be asked if they would like to
participate in an optional follow-up interview. These are the same eight caregivers who have
already been participating in the study by filling out daily and weekly questionnaires about
their child. If they choose to participant in the optional follow-up interview, caregivers
will be given a new consent form explaining this additional portion of the study.

DATA COLLECTION: Interviews will be scheduled sometime during the Spring semester, likely
January or February as the participants' schedules allow. Participants can choose if they
would like the interview to occur at Hearts and Horses Therapeutic Riding Center in Loveland,
at Colorado State University, or at a different location of their choosing such as their
home. While in-person interviews are preferred, phone interviews will be an option for
caregivers who would prefer not to meet in person. Interview questions will center around the
caregivers' perceived impact of the therapy on their child. For instance, "How, if at all, do
you believe participating in equine-assisted therapy affected your child?". Interviews will
be audio-recorded.

DATA ANALYSIS: Recorded interviews will be transcribed either by a professional transcription
service. Transcriptions will be uploaded into NVivo, a qualitative analysis software.
Investigators will conduct open coding, whereby the text will be broken down, examined,
conceptualized, and categorized. Data will be sub-categorized within existing start-codes as
appropriate, and new categories may emerge from the data. After open coding,investigators
will engage in axial coding, seeking to find links and relationships between and within
categories and sub-categories. Finally, selective coding will allow final themes to emerge,
and will elucidate relationships between themes. After initial analysis is complete,
participants will be asked to participate in a follow-up interview or phone call in a place
of their choosing, in order to conduct a member check, and ensure that emerging analyses are
in line with participants' intended meaning.

Inclusion Criteria:

- aged 5-14

- have an ASD diagnosis by a community provider

- score ≥15 on the Social Communication Questionnaire

- score ≥11 on the Aberrant Behavior Checklist-Community

- meets clinical cut-offs for ASD on the Autism Diagnostic Observation Schedule, Second
Edition

- score ≥55 on the Leiter International Performance Scale, Third Edition

- can tolerate a helmet

- can participate in 10-minutes of riding while following safety rules

- meets physical, mental, and emotional standards set forth by the Professional
Association of Therapeutic Horsemanship, International

Exclusion Criteria:

- behavioral issues that could interfere with safety

- 2 hours or more of previous experience with equine-assisted interventions in the last
6 months

- weight exceeding 200 pounds