Stakeholder informed development of the Emotion Awareness and Skills Enhancement team-based program (EASE-Teams)

Abstract

Emotion dysregulation underlies psychiatric symptoms and impedes adaptive responses in autistic individuals. The Emotion Awareness and Skills Enhancement (EASE) program was the first mindfulness-based intervention designed to target emotion dysregulation in autistic adolescents (12-17 years old). This project partnered with stakeholders to adapt EASE for autistic adolescents and adults with co-occurring intellectual disability and autistic elementary-aged children, groups who often benefit from caregiver support in treatment. Over three adaptation phases, we: 1) elicited stakeholder and expert feedback to adapt the original EASE program for autistic individuals with intellectual disability; 2) redesigned the adapted manual and expanded the target age range following a small “micro-trial” with a sample of autistic adolescents and adults with intellectual disability (n=6); and 3) demonstrated feasibility and acceptability of a caregiver-client team-based approach (EASE-Teams) in a sample of 10 autistic individuals with and without intellectual disability (ages 7-25) and their caregivers. EASE-Teams was both acceptable and helpful to families. Significant improvements were noted in participant emotion dysregulation, psychiatric symptoms, and caregiver stress from their child’s dysregulation. Findings suggest that EASE-Teams may be appropriate for heterogeneous developmental and cognitive needs. Future research to establish efficacy and refine EASE-Teams with community providers is warranted.

Lay Abstract

Difficulties with emotion regulation – or emotion dysregulation (ED) - contribute to psychiatric symptoms and flexibility in autistic individuals. In prior work, the Emotion Awareness and Skills Enhancement (EASE) program was developed to address ED using a mindfulness-based approach. The present study partnered with autistic individuals and their caregivers (“stakeholders”) as well as clinicians with expertise in autism spectrum disorder (ASD) and ED to adapt EASE for use with autistic individuals who need more support: specifically autistic adolescents and young adults with intellectual disability (ASD-ID) and autistic elementary-aged children without ID (ASD-E). The study was conducted in three phases and included: 1) stakeholder and expert feedback to adapt the original EASE program for ASD-ID individuals; 2) redesigning the adapted manual after using it with several ASD-ID individuals; and 3) demonstrating that the newly developed team-based approach (EASE-Teams) was feasible to use as a treatment of ED, was acceptable to families, and was successful in reducing ED in a sample of 10 autistic individuals with and without ID (ages 7-25). Families found EASE-Teams to be both acceptable and helpful. Additionally, autistic individuals showed improvement in ED and psychiatric symptoms, such as irritability, anxiety, and depression. Additionally, caregivers had reductions in stress related to their child’s behavior after participating in EASE-Teams. This project shows promise for an ED intervention that can be tailored to developmental and cognitive capacity needs, although future work is needed (e.g., demonstrating effectiveness in a large trial, refining treatment for use in community clinics).

Background

Emotion dysregulation (ED), difficulty managing the intensity and duration of emotional reactions, is highly prevalent among autistic individuals across the lifespan (Conner et al., 2021). Research suggests that autistic individuals are predisposed to ED and it is a transdiagnostic mechanism that underlies co-occurring psychiatric conditions (e.g. depression, anxiety), externalizing problems (e.g. aggression, explosive behavior), and suicidality (Conner et al., 2020; Herrington & Schultz, 2010; Mazefsky & White, 2014; Richey et al., 2015; Weiss et al., 2014). This predisposition for clinically elevated ED is present in autistic individuals with and without intellectual disability (ID), defined as individuals experiencing significantly lower intellectual functioning and adaptive behavior than average levels (Bertelli, 2019). Further, clinically elevated ED in autism is associated with psychiatric hospitalization and use of crisis services (e.g. emergency room and police contact) and lower family quality of life (Conner et al., 2021; Nuske, Hedley, Tseng, Begeer, & Dissanayake, 2017). As such, ED is an important target mechanism to improve outcomes for autistic individuals and their families across the lifespan.

Despite this, psychosocial treatments for ED are lacking for autistic individuals, especially those with limited cognitive and verbal abilities (Beck, Conner, Breitenfeldt, et al., 2020). Most evidence-based psychosocial treatments in autism use cognitive behavior therapy (CBT) to target anxiety in verbal autistic adolescents and adults without ID (White et al., 2018, 2020). Unfortunately, these treatments do not meet the needs of most autistic individuals and their families, who often seek treatment for multiple psychiatric diagnoses and complex problems related to ED (Beck et al., 2020; Mazefsky et al., 2013; White et al., 2021). Instead, behaviors associated with ED are cited by community agencies and providers as barriers to treatment and reasons for referring autistic individuals elsewhere (Maddox et al., 2020).

Furthermore, it is estimated that between 30-40% of autistic individuals have co-occurring intellectual disability (ASD-ID), with great variability in verbal capacity (Christensen et al., 2018; Van Naarden Braun et al., 2015), yet there are virtually no psychosocial treatment options available for this population. Thus, the state of science does not reflect the real-world heterogeneous presentations of the autism community, leaving providers, individuals, and families without options.

There is some preliminary evidence for CBT-based treatments targeting ED in school-aged autistic children (e.g., Beaumont et al., 2015; Swain et al., 2019). Importantly, the creation of treatment programs that are geared to specific age groups or presentations reduces flexibility and adaptability for future implementation (Damschroder, 2019). There is a critical need to develop treatment options targeting ED that can be utilized with autistic individuals across a wider range of ages and cognitive and verbal abilities.

Mindfulness-based interventions (MBIs) offer a promising approach to target ED while remaining flexible to the heterogenous needs across autistic individuals (Shaffer et al., 2019; White et al., 2018). MBIs teach individuals to cultivate awareness of emotions through repeated and routinely scheduled meditations, which are easily tailored to support an individual’s cognitive and verbal needs, and experiential exercises to practice new skills in applied settings (Beck, Conner, White, et al., 2020). Further, evidence suggests that MBIs improve emotion regulation capacity in other populations (Gu et al., 2015). Preliminary evidence supports the feasibility of MBIs with autistic children, adolescents, and adults with and without ID (Beck, Conner, White et al., 2020; Conner et al., 2019; Conner & White, 2018; Singh & Hwang, 2020), and caregivers of autistic individuals (Hartley et al., 2019;), suggesting that this may be an ideal therapeutic approach for autism across intellectual functioning and the lifespan.

In response to the urgent need for treatments targeting ED in ASD, we developed The Emotion Awareness and Skills Enhancement (EASE) Program (Conner et al., 2019). EASE was designed to improve emotion regulation in verbal and autistic adolescents without ID. EASE was developed in partnership with caregivers of autistic adolescents and adults, mindfulness experts, and autism-specialized interventionists. EASE is an MBI that explicitly addresses the common ED profile in ASD, characterized by a tendency to react intensely without awareness or forethought (Mazefsky & White, 2014). EASE consists of sixteen 50-60-minute individual therapy sessions with minimal caregiver involvement (see Conner et al., 2019). EASE was piloted with 20 verbal autistic 12- to 17-year-olds without ID across two sites with specialized research clinicians. This pilot established feasibility and acceptability of EASE, with significant, large-effect size improvements in emotion regulation at post-treatment (d=0.67-2.64; Conner et al., 2019).

The Present Study

The objective of this study was to modify EASE to meet the needs of a heterogeneous population of autistic individuals in order to maximize future research and intervention adaptability. Two phases of iterative manual development were completed prior to conducting a pilot study. The initial aim of the study and focus of the first two phases was to partner with stakeholders to adapt EASE for ASD-ID adolescents and adults. Over the course of manual development, the research team received multiple requests from caregivers and community providers seeking ED treatment for autistic elementary-aged children. Hypothesizing that the modifications made for ASD-ID adolescents and adults would also be appropriate for younger ages, we then expanded the pilot trial (Phase 3) to include autistic elementary-aged children. Figure 1 depicts the design and phases of the present study.

Figure 1.

Study phases and trail screening and retention

ED= Emotion Dysregulation

NVIQ = Nonverbal IQ <50 measured by the Leiter

*=Expended age range to 7-25 due to stakeholder request

The specific aims of the present study were to partner with stakeholders from the autistic community to: 1) Adapt EASE for autistic individuals with clinically elevated ED who may benefit from a higher level of caregiver support in treatment, such as those with intellectual impairment and/or limited verbal ability (ASD-ID) or elementary age-autistic children with significant ED impairment (ASD-E); and 2) Establish acceptability, feasibility, and proof-of-concept for decreasing ED in a pilot sample of ASD-ID and ASD-E participants and their caregivers. It was hypothesized that the adaptation would be feasible and acceptable to participants and would show proof-of-concept that ED can be improved over time. Secondary analyses explored the impact of the treatment on participant mental health and caregiver outcomes (caregiver stress and self-efficacy).

Methods

Participants and Procedures

Phase One: Initial Manual Prototype

First, EASE developers specified the core components of EASE to retain through the adaptation process. The EASE core components were identified based on our pilot, proof-of-concept trial (Conner et al., 2019), a dimensional model of ED in autism (Mazefsky et al., 2013; Mazefsky, Yu, et al., 2018; Mazefsky et al., 2020), and previous work studying mindfulness in autism spectrum disorder (Beck, Greco, Terhorst, et al., 2020; Conner & White, 2018). The research team then conducted panels with two stakeholder groups: caregivers of autistic individuals and an expert panel of clinicians, researchers, and allies. The family panel included four caregivers of autistic adolescents and adults (including two caregivers of individuals with ASD-ID, one nonverbal). One of the parents was employed by an ASD service provider in an advocacy role and had in-depth experience with a range of caregiver perspectives, and two parents had adolescents who completed EASE in the randomized controlled trial (IQ > 80). The advisory expert panel consisted of a psychiatrist, doctoral-level speech language pathologist, and two psychologists. We also engaged in individual consultation meetings with experts in mindfulness interventions for adults with ID.

In the two-hour family panel meeting, we conducted four discussions related to adapting EASE for ASD-ID individuals with varying levels of verbal ability: 1) The level of caregiver involvement in treatment; 2) The use of communication aides and tools to support learning; 3) Strategies for encouraging motivation; and 4) General feedback on the acceptability of the EASE approach in this population. The feedback from the family panel was shared with the advisory expert. Researchers elicited feedback from the expert panel on the same four topics discussed with the caregiver group. Using the feedback from the family stakeholder panel and expert panel, the research team developed the first iteration of the adapted manual. The adaptation process did not follow a specific adaptation framework but was informed by Movsisyan et al.’s guidance for planned intervention adaptations (Movsisyan et al., 2019).

Phase Two: Micro-trial & Manual Redesign

Six adolescents and young adults with ASD-ID (mean age: 19.33, Range: 13-25; 5 males) enrolled in a pilot trial with intention to complete 16 individual therapy sessions. Participants had a mean non-verbal IQ of 78.67 (SD = 12.63). Four participants were White and two were Black. Recruitment was conducted via advertisements through community agencies and advocacy groups. Inclusion criteria included: (1) ASD diagnosis based on expert clinical opinion and supported by Autism Diagnostic Observation Schedule-2; (2) age 12-25 years old; (3) Intellectual Disability designation in medical or educational records; (4) Nonverbal IQ>50 based on the Leiter International Performance Test-3; (5) caregiver available and willing to participate; and (6) clinically significant ED (t-score ≥ 60 on the Emotion Dysregulation Inventory [EDI] Reactivity, general norms). Exclusion criteria included: (1) Current suicidality or homicidality that warranted more immediate care; (2) Active psychosis; and (3) Receiving concurrent psychotherapy treatment for emotional issues. Medication use had to be stable for 4 weeks with no planned changes.

Participants completed weekly individual therapy sessions with trained research interventionists, consisting of three licensed professional counselors or clinical psychologists with PhDs and expertise in ASD. Interventionists were trained by reviewing the revised manual, several team practices in EASE-Teams meditation practices, and weekly supervision with a PhD specializing in mindfulness delivery in ASD. Each supervision group included tape review and regular practice leading EASE-Teams meditations. Interventionists recorded content and delivery suggestions for future revision on weekly fidelity forms and elicited feedback from caregivers on the acceptability of the intervention.

Approximately 10 weeks after the start of the trial, all in-person research was halted due to the COVID-19 pandemic. A total of 21 sessions across the six participants had been completed (one participant completed seven sessions, one completed four, three completed three, and one completed only the first session). Interventionists provided extensive feedback on fidelity, operational facilitators and barriers, and suggestions for further revision on a weekly basis throughout this period, which was applied to redesign the manual. During this time, we sought IRB approval for virtual clinical sessions. Participants utilized this option in the phase three feasibility pilot trial.

Phase Three: Open Feasibility Pilot of EASE-Teams

Enrollment of pilot participants resumed in September 2020 after a 5-month hiatus. Three participants from the original micro-trial restarted treatment with the revised manual. During the COVID-19 shutdown, the research team received multiple requests from caregivers and community providers seeking ED treatment ASD-E. Hypothesizing that the modifications made for ASD-ID adolescents and adults would also be appropriate for younger ages, we enrolled ASD-E in the trial. In the end, six ASD-ID participant-caregiver dyads (ages 15-23; all male) and four ASD-E participant-caregiver dyads (ages 7-10; two male) were enrolled in the feasibility pilot. Table 1 displays participant demographics and characteristics.

Table 1.

Phase 3 Pilot Study Participant Demographic Information

	ASD-ID (N = 6)	ASD-E (N = 4)
Age: M (SD)	18.67 (2.88); Range = 15-23	8.00 (1.41); Range = 7-10
Sex: N (%)	6 male (100%)	2 male (50%)
Race: N (%)	6 White (100%)	3 White (75%), 1 Asian (25%)
IQ1: M (SD)	69.40 (12.68); Range = 57-87	105.50 (4.95); Range = 102-109
Auditory Comprehension Age	8.29 (1.77)	7.63 (2.36)
Equivalent (TACL): M (SD)
Caregiver Sex: N (%)	5 female (83.33%)	4 female (100%)
Caregiver Education	2 some college, 2 college degree, 2 graduate degree	2 college degree, 2 graduate degree
Household Income2	2 less than $51K	1 less than $51K
	2 $51K-100K	1 $51K-100K
	2 more than $100K	1 more than $100K

¹- IQ only available for five of the ASD-ID participants and two of the ASD-E participants. Three ASD-ID participants had non-verbal IQs based on administration of the Leiter-3 during eligibility, the remaining two ASD-ID participants and two ASD-E participants had Full Scale IQs from prior year WISCs administered by their schools.

²- Household income unavailable for one ASD-E participant

Due to COVID-19 safety precautions, inclusion criteria (see Phase 2) were adapted to minimize in-person testing. ADOS-2 was replaced by caregiver-report of a professional ASD diagnosis and a score of > 12 on the caregiver-completed Social Communication Questionnaire - Lifetime version (Rutter et al., 2003). Additionally, IQ inclusion criterion (> 50 NVIQ on the Leiter-3) was replaced with a requirement for age equivalent > 5.5 years on the Test for Auditory Comprehension (TACL-4; Carrow-Woolfolk & Allen, 2014). Lastly, ASD-E children were included in the study. All other inclusion and exclusion criteria remained consistent.

Interventionists included six masters and doctoral level clinicians from four professional backgrounds (i.e., professional counseling, clinical psychology, social work, and special education). Both in-person and virtual sessions (via Zoom) were offered. Four of the ASD-ID participants conducted all sessions virtually. The remaining two ASD-ID participants and all of the ASD-E participants did a combination of in-person and virtual sessions, with approximately 60% on average of their sessions being conducted in-person. Measures were completed pre- and post-treatment, with emotion dysregulation and fidelity ratings completed weekly.

Measures

Demographics

Caregivers filled out a questionnaire that included information about their child’s gender identity, sex, ethnicity, and race and their own sex, education level, and household income.

Treatment Fidelity

After each session, the interventionist completed a form that assessed participant homework completion, rapport, participant and caregiver involvement, suggestions for session revision, delivery challenges, and adherence to session objectives.

Treatment Satisfaction

Caregivers and participants (when willing/able) completed a 10-item treatment satisfaction scale on the overall impact of the intervention. It included six items on a 5-point scale (1 = not helpful, 5 = very helpful) and open-ended questions on aspects found to be most useful, not useful, lessons learned, and suggestions for improvement.

A semi-structured interview on satisfaction with treatment and study procedures was also completed with the caregiver-participant dyad at post-treatment.

Primary Outcome Measures

The Emotion Dysregulation Inventory (EDI; Mazefsky, Day, et al., 2018; Mazefsky et al., 2020; Mazefsky, Yu, et al., 2018) is a validated, change-sensitive measure of ED designed for use with autistic populations. The EDI includes two scales: Reactivity (EDI-R; intense and poorly regulated negative emotional reactions) and Dysphoria scale (EDI-D; minimal positive affect and unease). Caregivers filled out the 30-item EDI at pre- and post-treatment and the short form EDI (13 items) weekly. Higher scores on the EDI indicate higher levels of ED.

The Clinical Global Impression scale (CGI; (Guy, 1976) was completed by a reliably trained, independent rater. , . The CGI-Improvement scale was completed at post-treatment to measure overall change in ED during the treatment.. Improvement is rated on a 7-point scale from very much worse (7) to very much improved (1).

Secondary Outcome Measures: Participant

All of the following measures were completed by caregivers about their child: 1) Aberrant Behavior Checklist (ABC;(Aman et al., 1985b, 1985a): Analyses utilized the Irritability subscale, which taps various aspects of behavioral dysregulation with higher scores indicating more irritability; 2) PROMIS anxiety and depression short forms (Irwin et al., 2010, 2012): Each includes eight items that capture potential change in participant symptoms of anxiety and depression, with higher scores indicating greater severity; 3) The Child Behavior Checklist (CBCL; (Achenbach & Rescorla, 2001) or Adult Behavior Checklist (ABCL; Achenbach & Rescorla, 2003): Analyses utilized the Internalizing, Externalizing, and Total problems subscales, with higher scores indicating more problems in these areas.

Secondary Outcome Measures: Caregiver

All of the following measures were completed by caregivers about themselves: 1) The Parenting Stress Index (PSI-4 SF; Abidin, 2012): Analyses utilized the Parental Distress (PD), Parent-Child Dysfunctional Interaction (P-CDI), and Difficult Child (DC) subscales, and the Total Stress scale, with higher scores indicating higher stress levels; 2) The Difficult Behavior Self-Efficacy Scale (DBSES; (Hastings & Brown, 2002; Hastings & Symes, 2002): Five items related to feelings of self-efficacy (e.g., confidence, control, satisfaction) in managing challenging behaviors, with higher scores indicating greater self-efficacy.

Data Analysis

Paired t-tests were run on primary and secondary outcome measures of interest. Between group comparisons between the ASD-ID and ASD-E subgroups were not completed given the small sample size.

Community Involvement Statement

Family members of autistic people initiated the direction of this study by contacting our research team in search of non-behaviorally oriented treatments to support autistic family members with ED. We received requests to provide our ED treatment to young adults with ASD-ID and later from family members and community providers of ASD-E. We experienced an increase in these requests during the COVID-19 pandemic, influencing our decision to expand the age range of EASE-Teams, ultimately creating one treatment for a wide range of developmental, verbal, and cognitive needs. Autistic people, family members, and advocates provided feedback in the iterative adaptation process.

Results

Phase One: Initial Manual Prototype

Defining EASE core components

EASE is primarily a mindfulness-based intervention designed to target ED in autistic individuals (see Figure 2). EASE teaches present-moment awareness of changes in emotional intensity rather than labeling discrete emotions and uses mindfulness meditations which engage the parasympathetic nervous system to combat emotional and physiological arousal. EASE emphasizes non-judgmental acknowledgement of intense emotions (e.g., all emotions are OK) and distress tolerance (e.g., staying in control, but not avoiding strong negative emotions). Finally, EASE emphasizes repetition of newly learned mindfulness practices during individualized social challenges that elicit emotional reactions.

Figure 2.

Core components of EASE

Family Stakeholder Panel

The family panel discussed caregiver involvement in treatment. Caregivers shared experiences of children not speaking on their own behalf when a caregiver was present in medical or therapeutic sessions and the need for a treatment that promotes autonomy. Pros and cons were considered for: caregiver present only at the start or end of session, caregivers attending entire sessions initially then phasing out as the treatment progressed, or caregivers calling therapists for a midweek check-in. Caregivers collectively expressed the belief that caregiver presence in therapeutic sessions should be limited, but no consensus on the amount and format was reached.

The family panel suggested a caregiver session prior to the start of treatment to review communication strengths, strategies for optimal communication, and difficulties related to ED. The group suggested having several communication aides as options for participants with heterogeneous needs. Differing opinions arose related to facilitating motivation. Some caregivers expressed that many families utilize behavioral reward systems, incorporated into daily routines and tasks, while others thought that reward systems impede intrinsic motivation.

The family panel also provided feedback on EASE treatment modifications. Caregivers emphasized the importance of the non-judgement of strong emotions core component and suggested the phrasing ‘strong emotions are OK’. Caregivers shared examples of their children believing that strong emotions are bad because of the externalizing behaviors that often accompany them.

Expert Review Panel

Subject matter experts discussed the complexity of balancing caregiver involvement and participant autonomy. The speech language pathologist with extensive experience working with ASD-ID agreed with focus group members that participants are more likely to disengage and talk less with a caregiver present. Other experts emphasized that caregivers were critical to therapeutic progress due to generalization of skills and reliable reporting of weekly progress. Overall, it was agreed that a combined approach (involving some caregiver involvement and opportunities for participant autonomy) would be needed to address this complex issue. Experts suggested that EASE revisions incorporate extensive repetition of strategies embedded into already established family routines.

Phase One Adaptation

Treatment structure.

The Phase 1 adaptation resulted in a 16-week treatment with approximately 45-50 minutes of content per session, similar to original EASE.

Caregiver involvement.

Caregiver involvement was expanded from original EASE, with framing as a caregiver-assisted (rather than individual) treatment. This expansion included two initial caregiver-only sessions covering treatment overviews and gathering specific information on ED and communication supports. Autistic participants began treatment with caregivers in session 3, with the intention to promote autonomy by phasing out caregiver presence beginning in the fourth participant session.

Treatment content and delivery.

Adaptations were made to session content while retaining the EASE core components. Consistent with original EASE, sessions began and ended with mindfulness meditation practice. Psychoeducational material was reduced and included exclusively in the initial caregiver-only sessions. Instead, simplified key phrases (‘all emotions are OK’) and meditations were repeatedly practiced in session with participants. The Noticing My Emotions Scale to support emotional awareness was simplified (e.g., 3 intensity levels instead of 10) and individualization was encouraged (e.g., colors, decorations, labels).

The original EASE Change module, which is based on a combination of Mindfulness-based Cognitive Therapy and traditional CBT, was significantly altered. Three sessions dedicated to learning mindful distancing and reframing of maladaptive thoughts were removed and replaced with the creation of a Helper Thought session. Helper Thoughts are individualized and simplified adaptive thoughts for participants and caregivers to repeatedly practice. Examples of Helper Thoughts include: “I’m still learning” and “Only a little longer.”

The Distraction module from original EASE, consisting of physical mindful distraction techniques, was renamed to Change Focus to reduce confusion with traditional distraction strategies used often with ASD-E and ASD-ID.

Finally, one additional community practice session (individually tailored sessions used to practice skills outside of clinic in applied social contexts) was added for a total of five across the treatment.

Mindfulness Meditations.

Mindfulness meditations were simplified and shortened (from 5-10 minutes to ~1-5 minutes). Two meditations were removed, leaving three simplified meditations: two breathing awareness practices (one short and one longer) and a walking meditation. Physical aides, such as using a Hoberman Sphere® to demonstrate intake and exhale of breath, were presented as options to help participants learn the new skill, with intent to phase out over treatment. The original EASE mini retreat, which consists of an entire session dedicated to reviewing meditation practices, was replaced with shorter practices multiple times per session to support mastery of practices with increased dosage and repetition.

Phase 2: Micro-trial and redesign into EASE-Teams

Micro-trial feedback

During the micro-trial, interventionists recorded delivery challenges and elicited feedback from caregivers and participants on the utility and acceptability of the adapted intervention. Suggested changes revolved around two main themes: 1) The need for more caregiver support both in and out of therapeutic sessions; and 2) The need for further simplification of session content and materials.

First, most participants had difficulty reporting on ED and practice in the prior week, making it difficult for interventionists to build on personal examples. In addition, participants had difficulty transferring content and skills learned in session to the home environment without considerable caregiver support. Finally, interventionists noted that caregivers were frequently involved in ED challenges that arose during the week, and caregivers reported that these challenges were also dysregulating for them personally.

Despite substantial simplification in phase 1, interventionists noted concerns regarding both the complexity and amount of material. Participants displayed waning interest/attention and cognitive fatigue during sessions. Interventionists also noted a need for simplified language in both describing skills and mindfulness exercises.

Phase 2 Redesign

Micro-trial feedback was applied to redesign the treatment into EASE-Teams. The specifics of the redesign are detailed in Table 2 and below, and Figure 3 shows the content and structure of the final treatment.

Table 2.

Iterative treatment adaptation across pilot trials

	Original EASE	Phase 1 Adaptation	Phase 2 Adaptation/Phase 3 Pilot: EASE-Teams
Population	ASD without ID	ASD with ID	ASD with or without ID
	Age 12-21	Age 12-25	Age 7-25
	IQ > 80	Nonverbal IQ >50	Receptive Language > 5.5 years age equivalent
Treatment Structure	16 weekly sessions	16 weekly sessions	16 weekly sessions
	45-50 mins content	45-50 mins content	25-30 mins content
			10-15 mins personalized activity
Caregiver Involvement	Individual therapy	Caregiver-assisted	Team-based
		2 caregiver-only sessions at start	2 caregiver-only sessions at start
	Caregiver joins for 5 min at end of sessions	Caregiver in entire session for first module, phased out over treatment	Caregiver in every session during content delivery
Treatment Content and Delivery	Discussion based	Minimal discussion with skill practice	Repeated concrete skill practice and scheduling
	Psychoeducation: ED & mindfulness	Psychoeducation limited to caregiver-only sessions	Psychoeducation limited to caregiver-only sessions.
	Noticing my emotions: 10 point numeric scale	Noticing my emotions: 3 level individualized scale	Noticing my emotions: 3 level individualized scale
	Conceptual cognitive model	Simplified cognitive model reviewed	Cognitive model further simplified
	CBT reframing & MBCT distancing	Helper thoughts	Helper thoughts
	New content weekly	New content weekly	Systematic process to learn new skills (see Figure 2)
	4 community practice sessions	5 community practice sessions	5 community practice sessions
Mindfulness Meditations	6 unique practices	3 unique practices	2 unique practices
	5-20 minutes	1-5 minutes	1-5 minutes
		Reduced verbal instruction	Minimal verbal instruction
		External aide available	Encourage without aide
	All practices reviewed in a “mini-retreat” in session 6	All practices reviewed in a “mini-retreat” in session 6	“Mini-retreat” removed

Notes. ASD = autism spectrum disorder; ID = intellectual disability; ED = emotion dysregulation; CBT = cognitive behavioral therapy; MBCT = mindfulness-based cognitive therapy

Figure 3.

Session content and structure for EASE-Teams

Treatment Structure.

Sessions were redesigned to dedicate 25 minutes to learning and practicing content followed by 10-15 minutes of personalized activity with the interventionists (alone, if desired/appropriate). This personalized time was designed to dedicate space within each session to build rapport and meet participants’ independent needs.

Caregiver Involvement.

The two caregiver-only sessions were retained. The caregiver-assisted design was changed into a team-based approach where the participant and caregiver work together as a team in therapy to learn and practice new skills in session and at home. Caregivers were active participants in all sessions - learning and using the strategies as well as modeling strategy use during their own ED. Thus, caregivers were also expected to engage personally in skill building. In addition, a handout with key components of EASE-Teams was created to share with other caretakers and providers to encourage consistent language and approach across settings s.

Treatment Content.

Simplified key phrases and ‘helper thought’ adaptations were retained. The number of skills introduced was reduced (1 skill per session) and revised to be taught in a more concrete way (see Figure 4). Session structure changed to: (1) introduce a new skill, (2) interventionist-led practice of that skill (several times), and (3) participant-caregiver practice of the new skill with interventionist feedback. Interventionists and caregivers created a plan to integrate newly learned skills into already-established family routines. The plan was refined until the caregiver and participant successfully and routinely practiced the skill. The interventionist then worked with caregiver and participant to utilize this skill in moments of ED (concretely labeled “STOP and ____” with the skill inserted after STOP) during community sessions.

Figure 4.

Systematic process for introducing and learning EASE-Teams skills with caregiver-participant team

Mindfulness Meditations.

Meditation practices were simplified into one short awareness of breathing practice and one walking meditation practice. Meditation language was further simplified and consisted of minimal and repeated verbal instruction (e.g., “feeling breathing in and feeling breathing out” repeated over 2 minutes). While almost all the micro-trial participants preferred to use the Hoberman Sphere®, caregivers expressed challenges practicing the meditations without the aide. Thus, the redesign included encouraging participants to first use their own body to feel breathing (e.g., hand on belly or chest) and occasionally introducing an external aide when interventionist deemed necessary.

Phase 3: EASE-Teams Pilot Trial

Feasibility

Treatment Adherence and Fidelity.

All ten participants completed the sixteen-session treatment (100% adherence). Sessions were 50 minutes long on average (SD = 10.27; range = 25-70 minutes). Interventionist ratings indicated a high degree of caregiver involvement (M = 3.84; SD = 0.41 on a 1-4 scale) and participant involvement (M = 3.57, SD = 0.66) in session, and good therapeutic relationships with participants (M = 3.67, SD = 0.60). Interventionist ratings of caregiver buy-in to treatment, rated at the start of therapy, were also high (M = 3.67, SD = 0.50). In weekly questionnaires about at-home practice, caregivers indicated that their child was using mindfulness strategies mostly every day on average (M = 1.89), and caregivers also reported using the mindfulness strategies themselves mostly every day on average (M = 2.10), indicating excellent engagement in the team-based approach.

Fidelity forms indicated that all session objectives were met in 90% of sessions. Interventionists noted challenges utilizing mindfulness strategies during instances of the participant dysregulation during session. In particular, both caregivers and interventionists reflected on the difficulty of not using behavioral approaches to manage disruptive behaviors during dysregulation (i.e., throwing marker at therapist) and instead using the EASE-Teams mindfulness-based awareness and calming strategies. Over the course of the pilot, both caregivers and interventionists noted that this challenge was improved by playing audio recordings together, and for longer periods of time, during dysregulation rather than having the caregiver or interventionist verbally lead the meditation.

Participant Satisfaction.

All caregivers filled out a satisfaction survey at the end of treatment. On a 1 (low) to 5 (high) scale, caregivers rated the helpfulness (M = 4.5, SD = 0.71), impact (M = 4.3, SD = 0.67), benefits to caregiver (M = 3.8, SD = 1.03), acceptability (M = 4.8, SD = 0.63), and relevance (M = 4.7, SD = 0.48) of the program highly. Nine out of the ten caregivers indicated they would be “very likely” (M = 4.8, SD = 0.63) to recommend the program to another caregiver. Seven participants also responded to the survey questions (typically with caregiver help). Participants also rated the helpfulness (M = 4.57, SD = 0.53), impact (M = 4.28, SD = 0.75), acceptability (M = 4.57, SD = 0.79), and relevance (M = 4.57, SD = 0.79) of the program highly. They were willing to recommend the program to a peer (M = 4.42, SD = 0.98), with five of the seven participants indicating they would be very likely and the other two participants indicating they would be somewhat likely.

Caregivers most frequently mentioned the following impacts from the treatment: 1) participant now able to engage in enjoyable activities more/longer (e.g., playing games with friends/family); 2) participant able to complete important life tasks more easily (e.g., traveling without support, going to doctor); 3) improved awareness of emotions and emotion intensity; and 4) having multiple strategies that help participant to calm or stay in control. One caregiver emphasized the impact of increased awareness in his daughter, stating:

“Beforehand she felt ‘out of control’ and ‘I don’t know what’s happening to me’. She would be doubly upset, and so now I feel like she goes ‘okay, this is how I am expressing emotions and it’s okay.”

Several caregivers noted that the team-focused aspect had been beneficial in helping their whole family develop new skills to effectively manage ED as a family unit. For example, one caregiver noted that the whole family now used the Noticing My Emotions Scale to communicate about emotional intensity. Another caregiver stated:

“My wife has never done any mindfulness stuff, and so, the meltdowns were the hardest for her. And she is able to handle half of them now. And she is able to calm her down, and she has never been able to do that before. […] And so, that has been a huge help because I have someone else that I can go and say, ‘I need to tag you in because I am empty.’“

Caregivers also commented that the team-based approach to learning mindfulness meditation and awareness strategies was useful and different from previous behavioral treatments. One mother noted:

“Initially we would start trying to focus on our breathing all three of us. And he would be like ‘nope, not doing it’ and [INTERVENTIONIST] would just go ahead and start doing it anyways for us. Then I noticed that [.] he would say, ‘When are we going to do our breathing together?’ So, I think that was really great. I learned that from [INTERVENTIONIST] -- giving him that space to be able to make his own decision and come to it himself once he exposed enough to it. […] He started meditating very regularly at bedtime now.”

Several other caregivers expressed an appreciation for the focus on ED and improving quality of life rather than focusing on behavioral treatment of autism symptoms. As one caregiver, who is herself autistic, said,

“In a lot of places, the focus is on changing the behavior, not finding the cause or soothing methods, or not accepting. A lot of times autism spaces come across as ‘how can we make your kid not autistic,’ instead of ‘how can we make your child’s existence better’ […] and how can we find what they need and what will help them…”.

When asked about suggestions for improvement or future work, four caregivers brought up the idea of a social support or group aspect to the therapy as an opportunity to practice skills in a social environment, normalize challenges, and make connections with others. Two caregivers also mentioned a desire for ongoing support after the end of therapy to help families maintain gains. And one caregiver suggested that some acceptance-based and psychoeducational content on ED in autism for participants that prefer that learning modality and to “take the shame away from the child.”.

Proof of Concept

Table 3 displays descriptive statistics of pre- and post-treatment scores as well as pairwise t-tests evaluating change from pre- to post-treatment for all primary and secondary measures. Figure 5 displays individual-level changes for all participants with different lines per subgroup.

Table 3.

Pre-treatment and Post-treatment Scores on Outcome Measures in Phase 3 Pilot Trial

	Pre-treatment		Post-Treatment
Variable	Mean (SD)	Range	Mean( SD)	Range	t (df)	p-value
EDI Reactivity Theta	0.43 (0.69)	−0.52-1.45	−0.37 (0.72)	−1.14-0.68	5.52 (9)	<.001
EDI Dysphoria Theta	0.07 (1.1)	−1.36-2.03	−0.74 (0.68)	−1.36-0.8	3.41 (9)	0.008
ABC irritability	18.5 (10.44)	7-39	13.9 (10.72)	5-38	3.21 (9)	0.011
PROMIS Anxiety Theta	1.05 (1.41)	−1.02-3.37	0.35 (1.38)	−1.47-3.2	2.85 (9)	0.019
PROMIS Depression Theta	0.86 (1.15)	−0.82-2.55	−0.01 (0.95)	−1.38-1.26	3.48 (9)	0.007
CBCL/ABCL Externalizing Problems t-score	64.1 (7.69)	56-79	62.9 (7.77)	53-79	1.81 (9)	0.104
CBCL/ABCL Internalizing Problems t-score	66.6 (12.08)	51-84	63 (11.95)	45-83	2.19 (9)	0.056
CBCL/ABCL Total Problems t-score	69.4 (8.73)	59-83	66.1 (8.08)	53-79	3.3 (9)	0.009
Parent Difficult Behavior Self-Efficacy Scale	19.7 (3.47)	12-26	23.2 (4.44)	14-28	−1.87(9)	0.095
PSI Difficult Child t-score	69.33 (8.29)	54-78	64.3 (6.4)	53-71	3.12 (8)	0.014
PSI Parent Child Dysfunction t-score	62.22 (6.38)	55-71	62.4 (7.95)	46-73	−1.43 (8)	0.189
PSI Parental Distress t-score	55.11 (6.51)	44-62	55.8 (6.09)	41-62	−0.52 (8)	0.619
PSI Total Stress	63.22 (6.02)	54-71	61.7 (6.07)	51-68	0.23 (8)	0.827

Note. One participant is missing PSI scores at pre-treatment

Figure 5.

Individual pre-treatment and post-treatment Emotion Dysregulation Inventory scores for participants in the Phase 3 Pilot trial. solid lines indicate ASD-ID participants and dashed lines indicate ASD-E participants.

Primary Measures.

EDI.

Eight of 10 participants (80%) showed more than a 0.5 SD improvement in EDI-R scores, and six of the 10 participants (60%) showed more than a 0.5 SD improvement in EDI-D scores. On average, participants showed significant improvement in both EDI-R/D, improving 0.79 SDs in Reactivity (p<.001) and 0.82 SDs in Dysphoria (p = .008).Figure 5 shows individual pre/post theta scores for EDI-R/D for all participants with clinical cutoffs for clinically elevated EDI-R/D. Theta scores were derived based on ASD norms with clinical cutoff designations at 1SD above the general population mean (i.e. > −.364 for Reactivity and > .222 for Dysphoria; Mazefsky et al., 2020). Both ASD-ID and ASD-E participants showed improvement (Figure 5).

Clinical Global Impression Scale.

We were unable to obtain an independent rating interview from one participant. Three of the remaining nine participants (33.3%) were rated as much improved and six (66.6%) of participants as minimally improved.

Secondary Measures.

Participant Mental Health.

Caregivers reported significant improvement in participant irritability (p = .01), anxiety (p = .01), and depression (p = .007). On the CBCL/ABCL, caregivers reported non-significant change in internalizing (p = .06) and externalizing (p = .10) problems, and significant change in total problems (p = .01).

Caregiver Outcomes.

Caregivers reported a non-significant improvement in difficult behavior self-efficacy (p = .10). On the PSI, caregivers reported significant improvement in stress related to child characteristics (p = .01), but did not show improvement in parental distress, parent-child dysfunctional interaction, or total stress.

Discussion

Treatments that are responsive to community needs and stakeholder input in the development process improve future usability and adoption in real world contexts (Concannon et al., 2014). Stakeholder input influenced the direction of this iterative study, beginning with the original development of EASE, adapting EASE for autistic individuals with ID, and expanding the age range to also include EASE-E with severe ED. Results of this study suggest that this stakeholder-informed treatment, EASE-Teams, is acceptable and feasible for a wide range of developmental and cognitive needs.

Perhaps most importantly, we learned from conversations with stakeholders, micro-trial participants and their caregivers that ED is a family process. Caregivers reported feeling unsure about how to help or respond to dysregulation, while also feeling overwhelmed themselves. This is consistent with research indicating high levels of caregiver stress, greater depression, and lower family quality of life, among caregivers whose children have emotional and behavioral problems (Falk et al., 2014; Nuske et al., 2018). In response, the new team-based approach emphasized caregiver(s) and participant jointly learning skills to use together in moments of dysregulation. This team-based approach is distinct from a more hierarchical caregiver-implemented model of treatment in which the caregiver acts solely as a support person in helping the participant implement skills (e.g., parent management approaches). Instead, participants and caregivers were framed as equal teammates in the treatment, and caregivers were encouraged to practice skills individually during their own moments of ED. Further, we learned that the importance of integrating practice into already established family routines so that participants and caregivers were able to master newly learned skills.

Based on outcomes of the Phase 3 pilot trial, this team-based approach was acceptable and feasible to families, many of whom specifically mentioned finding this aspect of the treatment helpful. Furthermore, we found preliminary evidence for impacts of the treatment of caregiver outcomes—specifically improvement in caregiver feelings of self-efficacy in managing their child’s behavior and reductions in parenting stress related to their child’s behavior. We also found statistically significant changes in ED and caregiver reports of participant improved irritability, depression, and anxiety. This is consistent with what was seen in the open pilot trial of the original EASE, conducted with autistic adolescents and adults without ID (Conner et al., 2019), and provides further support for the importance of targeting underlying mechanistic processes, such as ED, rather than focusing only on specific diagnostic outcomes (e.g., anxiety). While the statistically significant changes are promising especially given the small sample size, independent ratings of clinical improvement painted a more conservative picture. Most participants were rated as minimal improvement with some rated as substantial improvement. Adequately powered efficacy testing is needed to determine the clinical impact of this treatment. While some empirically supported treatment options targeting ED in ASD are emerging (Beck, Conner, Breitenfeldt et al., 2020), these are largely geared towards verbal autistic individuals without ID and are developed for specific ages, thus limiting their reach and flexibility. Although our original intent was to fill a gap in treatment options for ASD-ID individuals specifically, repeated stakeholder requests encouraged us to broaden our reach and create a treatment that could be used flexibly to meet the needs of a diverse range of autistic individuals. The pilot study provides preliminary evidence that EASE-Teams was both acceptable and feasible for autistic individuals across a range of developmental, cognitive, and language abilities. Further, our focus on accepting all emotions was noted by several caregivers as a strength of this treatment in contrast to behaviorally focused programs targeting autism symptoms only, which often focus on reduction of negative behaviors and less frequently attend to the corresponding emotions. This is in line with qualitative research showing that parents of autistic individuals name emotion-related outcomes (e.g., distress, anxiety) as the most important treatment and research targets (McConachie et al., 2018).

The iterative design of this adaptation study allowed us to collect important feedback from interventionists on delivery challenges and implementation determinants. Interventionists and caregivers reported difficulty leading mindfulness practices during moments of dysregulation, often triggering externalizing behaviors from the participants. Both found that using recordings and practicing together was more helpful than verbal prompts or guidance for newly learned regulation skills. Further, interventionists new to mindfulness noted challenges using this approach and our team is working to develop additional supports for interventionists to test formally in future hybrid trials and increase ease of access to mindfulness recordings to use during dysregulation.

Several methodological features should be considered when interpreting these findings. The small sample size limits generalizability and the results may not be stable in other samples. Additionally, several participants were lost during the 5-month COVID-19 hiatus, but the hiatus resulted in an unplanned micro-trial that ultimately strengthened the development of the EASE-Teams treatment. Future studies might consider using rapid prototyping methods during initial development of psychosocial interventions (Lyon & Koerner, 2016). Further, COVID-19 complications impacted participants’ and families’ experience (e.g., masks, social distancing, last minute switch to virtual sessions, and general unpredictability), which could have impacted the results.

This study did not include community-based providers and agency leaders as a stakeholder in the adaptation process. Future work will include discussing optimal treatment length (e.g. 16 sessions) and funding considerations (e.g. insurance coverage) as well as any additional implementation barriers that may be identified prior to testing in the community with community-based providers. Future studies with larger samples are needed to determine when a team-based approach or individual therapy is most appropriate for participants. Ultimately, we plan to conduct a large-scale effectiveness-implementation Hybrid Type 1 trial that combines EASE and EASE-Teams into one simplified and flexible modular ED treatment for autistic individuals. In addition, we plan to work with community providers to ensure that this treatment is relevant, scalable, and usable.

In conclusion, the EASE-Teams adaptation demonstrated preliminary feasibility, and acceptability among a heterogeneous group of young children and adolescents with ASD and their primary caregivers. Data and feedback from caregivers suggest that the intervention fills a critical gap by improving ED in children, adolescents, and adults with ASD and ASD-ID, and targeting ED symptoms that caregivers often find most distressing and disruptive to family life. Further refinement and adaptation, as well as more widespread dissemination, has the potential to improve individual and family outcomes through a novel/understudied mechanism (i.e.emotion regulation) in the ASD intervention literature.