Building tailored interventions to promote self-determination in young adults with autism: the TEAM_YOUNG ADULTS program

Abstract

Objectives

Self-determination is crucial throughout life, yet individuals with autism often experience lower self-determination levels than their peers without disabilities. Thus, practitioners need interventions to support self-determination in people with autism, considering their specific characteristics. The objectives of this study are (a) to present the development of a program for the promotion of self-determination in people with autism, the TEAM_YOUNG ADULTS program, and (b) to assess its effectiveness through a pilot study.

Methods

A randomized controlled trial was conducted, with a mixed-methods approach, to assess the effectiveness of the program, combining quantitative and qualitative data. Fourteen participants, aged 17–25 years, were randomly assigned to the intervention group or the waiting list. Instruments were used to measure self-determination in both self-report and proxy-assessment formats. A focus group was also conducted at the end of the intervention to obtain qualitative data.

Results

The experimental group reported an improvement in the three dimensions of self-determination: volitional action, agentic action and action-control beliefs. Facilitators and barriers to change were also discussed.

Conclusions

Our findings show promising results, indicating that the program may be a useful intervention to improve self-determination in young adults with autism. Implications and future lines of research are discussed.

Introduction

The development of self-determination has been related to positive outcomes at an educational level (e.g. Lee et al. 2008); in the transition to adulthood (e.g. Shogren, Lee, et al. 2017); and in the improvement of quality of life (e.g. Lachapelle et al. 2005). Its development has been proven to be important throughout the lifespan, particularly during adolescence and the transition to adulthood. Self-determination-related skills can be promoted through intervention programs and contextualized support systems that enhance and provide opportunities for self-determination development (Morán et al. 2021). Interventions aimed at developing skills and attitudes related to self-determination, such as setting goals and objectives, making decisions, solving problems, self-knowledge, monitoring, and self-assessment skills, among others, must place the person as an active participant, letting them express themselves, choose, and take risks (Field and Hoffman 2000).

According to Causal Agency Theory, self-determination is ‘a dispositional characteristic manifested as acting as the causal agent in one’s life’ (Shogren, Michael, et al. 2015, 257). The construct is composed of three dimensions: volitional action, agentic action, and action-control beliefs. Volitional action involves acting autonomously, being able to set objectives according to one’s preferences and starting the corresponding actions to achieve them. Agentic action refers to the capacity to design a plan to achieve these objectives, the capacity to self-regulate one’s actions, and adjusting to achieve the expected objectives, using such strategies as self-regulation, self-direction, and pathways thinking. Finally, action-control beliefs refer to knowledge about one’s capacities and abilities, as well as limitations, and are linked to the empowerment that one perceives in relation to previous experiences (Wehmeyer et al. 2017).

Although to date, literature on self-determination has focused on people with intellectual disability or learning disorders, recent research has focused on people with autism (Chou et al. 2017; Lee et al. 2010). Studies with children and adolescents with autism and intellectual disability in the Spanish context stress self-determination and social inclusion as priority areas for intervention and research for this population (Morán, Gómez, and Alcedo 2019). People with autism tend to report lower levels of self-determination than other groups with disability (Cheak-Zamora et al. 2019; Chou et al. 2017), as well as in self-regulation, empowerment, and self-realization (Chou et al. 2017). Further, their unique difficulties in social communication and social skills may hinder skills related to self-determination (Cuesta-Gómez, Vidriales-Fernández, and Carvajal-Molina 2016; Sansosti et al. 2017). For this reason, interventions targeting people with autism must be tailored to the unique needs of people with autism, including managing uncertainty about the future, expressing needs and rights, as well as keeping motivation, managing time, or organizing and regulating stress in the academic context (White et al. 2017). Thus, the creation of specific intervention programs to develop self-determination in people with autism urges.

Most research on self-determination in people with autism has focused on adolescence, particularly on the transition period into adulthood. A recent review of studies on self-determination in children and adolescents with autism suggests that research in this field has quadrupled in the last few years (Morán et al. 2021). However, studies of interventions for adults with autism are still scarce (Andrés-Gárriz, Mumbardó-Adam, and Vicente Sánchez 2023). In adulthood, people with autism may experience difficulties in areas such as education, employment, social support, health (both physical and mental), independence, and emancipation (Friedman, Warfield, and Parish 2013), thus becoming a special period of vulnerability. In this context, teaching self-determination skills becomes highly relevant, as the development of such strategies and skills correlates with higher life satisfaction (White, Flanagan, and Nadig 2018). Elemental components of self-determination as autonomy, psychological empowerment, and self-realization have proved to be predictors of better results in employability and independent life (Shogren, Lee, et al. 2017), as well as the development of social relationships and community participation, stress management and developing a positive personal identity (Kim 2019). Despite the aforementioned benefits of self-determination in young adults with autism, they struggle with self-determination-related skills such as self-initiation, self-management skills, empowerment, and self-realization, which are affected by characteristic difficulties present on the autism spectrum (Andrés-Gárriz et al. 2024). Consequently, interventions to develop self-determination-related skills must be tailored to adjust to the specific needs of adults with autism to facilitate the obtention of better outcomes.

To date, different intervention programs have shown positive effects on self-determination development in adults with autism (Andrés-Gárriz, Mumbardó-Adam, and Vicente Sánchez 2023; Vidriales et al. 2017), although they follow a diversity of methodologies and characteristics, both in number of sessions and frequency, and most of them have been created and developed within the US context (Andrés-Gárriz, Mumbardó-Adam, and Vicente Sánchez 2023). In Spain, there are also some, but fewer, good practice guides to work with self-determination (e.g. Ponce 2010). Recently, Mumbardó-Adam et al. (2018) translated into Spanish the Self-Determined Learning Model of Instruction (SDLMI); this model is based on three phases (setting goals, acting, and adjusting goals or plans) to solve problems in adolescents and youth in the educational context (Wehmeyer et al. 2000). Recently, this same model has been used within the Spanish context in a study aiming to develop self-determination in adults with intellectual disability (Álvarez, Vicente Sánchez, and Mumbardó-Adam 2022). In fact, these guides and models offer guidelines to develop self-determination in people with and without disability but again, none of them target people with autism and consider their unique characteristics.

Some initiatives have targeted self-determination in people with autism (e.g. Nadig et al. 2018), but there is not enough empirical evidence to consider them evidence-based practices (Andrés-Gárriz, Mumbardó-Adam, and Vicente Sánchez 2023). Moreover, there are still no intervention tools available in the Spanish context, so intervention experiences to improve the self-determination of youth and adults with autism are scarce. Besides, to the best of our knowledge, there is a lack of interventions with scientific evidence to develop self-determination in youth with autism at a national level, despite the need to provide supports and opportunities for people with autism to develop self-determination. To meet this need and thus provide educators and professionals with an intervention instrument to promote self-determination specifically designed for the Spanish context, the TEAM_YOUNG ADULTS program was created. This program is the first intervention specifically developed for Spanish young adults with autism that is based on the latest self-determination framework, CAT (Shogren, Michael, et al. 2015) and that considers the specific support needs of individuals with autism in our context. Thus, the objectives of this study are to present the TEAM_YOUNG ADULTS program, show how was developed and the programs it is based on, and assess its effectiveness with a pilot test.

Methods

Design

A randomized controlled trial was conducted, in which qualitative data was embedded, using a mixed methods design (Creswell and Plano Clark 2018; Fetters and Molina-Azorin 2020) to assess the effectiveness of the TEAM_YOUNG ADULTS program. The collection of a wide variety of data allowed for a more complex analysis, thus providing a deeper understanding of the program’s effectiveness and as recommended in the field of developmental disabilities (Fàbregues et al. 2022). As this is a complex intervention, a single approach based on quantitative data was considered to be reductionist and would not show changes (Bergman 2008).

Participants

Twenty young adults with autism (13 men, 7 women), ranging in age from 17 to 25 years old, were enrolled in the study. Inclusion criteria consisted of having a diagnosis of autism spectrum disorder, in accordance with DSM-5 criteria (American Psychiatric Association 2013), or of Asperger’s syndrome, according to the DSM-IV (American Psychiatric Association 1994); being between 17 and 30 years of age; and having a verbal comprehension index ≥70 in the Wechsler Adult Intelligence Scale-IV (WAIS-IV; Wechsler 2008). The verbal comprehension index was used as the cut-off point to ensure a good comprehension of the TEAM_YOUNG ADULTS program due to its content and online format. Participants were excluded if they had no spoken language, had a psychotic disorder or behavioral problems that would not allow them to take advantage of the group or that could significantly interfere with the group’s functioning. Finally, 11 participants met the inclusion criteria and thus collaborated in the study (Table 1).

Table 1.

Sociodemographic information for group of participants and baseline evaluation.

	Intervention (n = 6) M(SD), range	Waiting list (n = 5) M(SD), range
Sociodemographic data
Age	19.50 (2.81), 17–25	20.20 (4.38), 18–28
Men (%)	50%	60%
Maximum level of studies (%)
Secondary mandatory education	33.33% (2/6)	60% (3/5)
Highscool (2 final years)	0% (0/6)	20% (1/5)
Vocational training (medium grade)	66.66% (4/6)	20% (1/5)
Academic/employment situation (%)
Studying	66.66% (4/6)	40% (2/5)
Not studying nor working	33.33% (2/6)	60% (3/5)
Medication (%)	50%	20%
Living situation (%)
Living with family members	83.33% (5/6)	100% (5/5)
Residential center of educational action	16.66% (1/6)	0%
Other conditions
ADHD	50% (3/6)	0%
BPD	16.66% (1/6)	0%
PTSD	16.66% (1/6)	0%
Depression and anxiety	0%	20% (1/5)
None	16.66% (1/6)	80% (4/5)
Characteristics
ADOS-2	8.25 (1.50), 7–10	8.00 (1.00), 7–9
SCQ-B	18.50 (10.50), 2–32	18.40 (7.57), 10–27
VCI (WAIS-IV)	111.17 (18.87), 82–130	100.00 (6.93), 94–110
Participants’ self-determination at baseline
AUTODDIS Scale
Volitional action	36.50 (6.83), 30–47	37.40 (9.48), 27–48
Agentic action	31.83 (13.08), 16–51	32.20 (14.48), 21–54
Action-control beliefs	46.00 (11.40), 30–64	47.60 (11.42), 32–64
Total	114.33 (29.53), 76–158	117.20 (31.85), 80–162
SDI:SR	(n = 6)	(n = 3)
Decide	79.00 (17.55), 52–96	67.00 (19.98), 45–84
Act	73.17 (13.79), 53–91	63.00 (11.36), 50–71
Believe	75.17 (13.93), 59–95	62.33 (9.07), 54–72
Total	227.33 (42.87), 164–282	192.33 (38.84), 149–224

Note: ADHD = Attention-Deficit/Hyperactivity Disorder; BPD = Borderline Personality Disorder; PTSD = Post-Traumatic Stress Disorder; ADOS-2 = Autism Diagnostic Observation Scale-2; SCQ-B = Social Communication Questionnaire - Part B; VCI = Verbal Comprehension Index; WAIS-IV = Wechsler Adult Intelligence Scale-IV; SDI:SR = Self-Determination Inventory: Student Report.

Participants were referred by the child and youth mental health center and the adult mental health center of the Hospital of Mataró / Healthcare, from the Healthcare Consortium of Maresme (a specific region of Catalonia, Spain) after signing their informed consent. Regarding the dropout rate, a participant in the experimental group withdrew from the intervention due to incompatibilities with other academic duties. Moreover, two participants in the control group did not respond to their second assessment (once the experimental group’s intervention was over) despite several reminders (see Figure 1).

Figure 1.

Flow diagram of participants.

Instruments

Participants’ data were collected in online one-to-one sessions. Sessions were conducted online due to COVID-19 restrictions. The hetero-reporting questionnaires were sent to the family members by email, and they were answered remotely. Due to the heterogeneity of the autism spectrum, functioning (both at the level of co-occurrent conditions and social functioning) and severity of symptoms were evaluated to create more homogeneous groups that enhance the program’s effectiveness (Spain and Blainey 2015).

Initial assessment

Autism Diagnostic Observation Scale – Second Edition (ADOS-2: Module 4) by Lord et al. (2015). It is a standardized and semi-structured assessment of communication, social interaction and play or imaginative use of materials for people suspected of having autism. It consists of 5 modules corresponding to different ages and language development. It takes between 40 and 60 min to administer. Module 4 for adolescents and adults with fluent language was administered. The instrument has reliability and validity evidence with intraclass correlations ranging from 0.82 to 0.93, a test-retest reliability with correlation coefficients between 0.74 and 0.90 and an excellent internal consistency, with an alpha value higher than 0.90 (Lord et al. 2015). This test was administered by staff accredited for its application and assessment.

Social Communication Questionnaire-Part B (SCQ-B) by Rutter, Bailey, and Lord (2019). This 40-item questionnaire is responded to by parents and provides information about social and communication behaviors, as well as observed interests. It consists of two parts: part A refers to past life, and part B focuses on the behavior over the last three months. For the purpose of the study part B was used.

Wechsler Adult Intelligence Scale-IV (WAIS-IV: Verbal Comprehension Index) by Wechsler (2008). This is a clinical instrument designed to assess general intellectual ability in people between 16 and 89 years and 11 months of age. The main verbal comprehension subtests (Similarities, Vocabulary, and Information) were used to assess the participants’ verbal comprehension.

Pre-post assessments

AUTODDIS Scale by Verdugo et al. (2021). This scale assesses self-determination in youth and adults (11–40 years old) with intellectual disability through external informants. It consists of 46 4-point Likert-type items based on the degree of agreement. These items are arranged into three sections: volitional actions, agentic actions, and action-control beliefs, based on the Causal Agency Theory (Shogren, Michael, et al. 2015). The internal consistency of these sections was excellent (between α = 0.948 to α = 0.962), as well as for the entire scale (α = 0.983), and there is also evidence of its validity.

Self-Determination Inventory: Student-Report (SDI:SR) by Shogren, Lee, et al. (2017). This instrument has been developed within the Causal Agency Theory (Shogren, Michael, et al. 2015), and measures essential characteristics and dimensions of self-determined action (volitional action, agentic action, and action-control beliefs). It is a self-report for youth between 13 and 22 years old with or without disability. The Spanish version by Mumbardó-Adam et al. (2018) was used in this study, which has shown evidence of its internal reliability in the three dimensions of self-determination (α = 0.815–0.911) and validity.

Final assessment

Ad-hoc questionnaire about group assessment

Questionnaire specifically designed for this research that collects information of both the participants’ and their family’s opinion about the content, activities, and facilitators (from one to ten on a liked scale, ten being the highest punctuation). It also includes two open questions to collect possible aspects for improvement and the most important aspects that the program has contributed with. Questions for family members consisted of their satisfaction with the program, whether they would recommend the program to other people in a similar situation and to briefly describe the changes perceived in their children. Questions for the young adults consisted of satisfaction with the program, the materials of the program, the facilitators, the subjective importance of what was done during the program, and the ability to solve problems and achieve goals. They were also asked if they would recommend the group to other young adults in a similar situation, as well as two open questions: (1) what have been the most significant aspects of the group for you?, and (2) what would you improve from the program?

Focus group

For the qualitative assessment of changes emerging after the TEAM_YOUNG ADULTS program, the focus group technique was used with participants in the experimental group. Three blocks of questions were developed to tackle the different changes produced throughout the intervention (see Annex 1). The first block of questions focused on changes in the participants’ volitional actions, with questions about the capacity of choosing activities on their own or starting these behaviors. The second block of questions referred to agentic actions, looking into changes in goal setting and attainment, how they dealt with problems found, and possible solutions. The third block’s objective was to delve into action-control beliefs, with questions about the feeling of capability to face problems, achieve goals, influence of strengths and weaknesses to achieve objectives, and the capacity to ask for help if needed.

Development of the intervention program

The TEAM_YOUNG ADULTS program is an intervention specifically created for youth with autism, which encompasses key aspects from other intervention programs and techniques to develop skills related to self-determination. Particularly, it uses the McGill Transition Support Program (Nadig et al. 2018) and the Self-Determined Career Development Model (SDCDM; Wehmeyer et al. 2003) as references.

The Self-Determined Career Development Model (SDCDM; Wehmeyer et al. 2003) is a career design model to give support to people with disability when establishing goals and solving problems at work. It is based on the Self-Determined Learning Model of Instruction (SDLMI; Wehmeyer et al. 2000) and it is a useful tool that can be used by facilitators to guide people with disability on their way towards the working world. It has three phases: (1) set a goal, (2) take action, and (3) adjust goal or plan. It also offers a series of questions that promote reflection on these phases. Given the characteristics of this model, we thought it could be very useful for our program and for young adults with autism (Table 2). An example of the influence of the SDCDM in the TEAM_YOUNG ADULTS program is session 7 (My goals). In this specific session, the facilitator’s questions from the first phase of the SDCDM (Phase 1: Set a goal) were used to conduct the session and to explain how to create measurable, observable, and specific goals.

Table 2.

Sessions of the TEAM_YOUNG ADULTS program and what program they are based on.

Module	Session	Program the session is based on
Beginning and basic skills	S1: We start!	Session of new creation
	S2: Decide, act, and believe	Session of new creation
	S3: Abilities for the group	(S1) McGill Transition Support Program + adaptation
	S4: Understanding the context	(S4 + S11) McGill Transition Support Program + adaptation
	S5: Good colleagues (and not that good ones)	(S12) McGill Transition Support Program + adaptation
Self-determination	S6: I know myself better	Session of new creation
	S7: My goals	Session of new creation based on the SDCDM
	S8: My plan	Session of new creation based on the SDCDM
	S9: My decisions 1	(S10) McGill Transition Support Program + adaptation
	S10: My decisions 2	(S10) McGill Transition Support Program + adaptation
	S11: What a problem!	(S7) McGill Transition Support Program + adaptation
	S12: Obstacles and barriers	Session of new creation
	S13: Help!	Session of new creation
	S14: Self-advocacy 1	(S8) McGill Transition Support Program + adaptation
	S15: Self-advocacy 2	(S8) McGill Transition Support Program + adaptation
Working with others	S16: Oh, conflicts… 1	Session of new creation based on S13 of the McGill Program
	S17: Oh, conflicts… 2	Session of new creation
	S18: Teamwork	(S15) McGill Transition Support Program + adaptation
Closure	S19: The path that we have walked together	Session of new creation
	S20: Goodbye	Session of new creation

Then, the McGill Transition Support Program (Nadig et al. 2018) is a program consisting of 10 sessions that focuses on the promotion of quality of life and self-determination of youth with autism on the basis of their specific needs, presenting a series of flexible and adaptable modules to the participants in every intervention group. Due to the high compatibility of this program with the philosophy of the program that we wanted to develop, we contacted the authors of the McGill Transition Support Program to carry out its translation and adaptation for the Spanish population and later use parts of the intervention as a reference for the TEAM_YOUNG ADULTS program. Two experts in autism and with a high command of English developed the translation and counter-translation process of this Canadian program to guarantee a reliable translation. Once the translation process was over, the program’s adaptation was carried out, both in formal aspects and in modifications of activities to adjust them to the Spanish population, which were later used as inspiration for the TEAM_YOUNG ADULTS program (Table 2). For example, session 10 (My decisions 2) was based on session 10 of the McGill Transition Support Program (Choice-making). The TEAM_YOUNG ADULTS program offered techniques to make decisions based on the Canadian program and then adapted the situations for practice to the Spanish context.

After the aforementioned adaptations and integration of both programs, the TEAM_YOUNG ADULTS was created considering the needs expressed by people with autism regarding their difficulties in self-determination-related skills (Andrés-Gárriz et al. 2024). An expert evaluation was carried out. Four experts were contacted, two belonging to the clinical field specializing in autism and two from the academic field specializing in self-determination in people with autism and people with disabilities. They were asked about the suitability, clarity, and importance of every module in the program, as well as about aspects of setting and format. After their opinions were collected, their proposed changes were implemented to get the final TEAM_YOUNG ADULTS program.

The online format of intervention was selected, via video conference, to guarantee all participants’ attendance, and due to COVID-19 restrictions, that made the online format more suitable. To ensure a good application of this format, the recommendations in the Guía para la Práctica de la Telepsicología (Practical Guide for Telepsychology) by the Colegio Oficial de Psicólogos de España (Consejo General de Psicología de España 2017) were followed.

As for the number of sessions, we decided to carry out 20 sessions, 10 more than in the McGill Transition Support Program. Although there is no research on the optimal duration of group intervention programs for people with autism, most systematic reviews and meta-analyses agree that 10-week or longer implementations have more probability of treatment improvement (Miller et al. 2014; Spain and Blainey 2015). The decision to implement 20 sessions, a similar number as in interventions in social skills with wider scientific evidence (Gantman et al. 2012; Laugeson et al. 2015) is due to two main reasons. Firstly, people with autism may experience difficulties to process information and implementing knowledge (Wilson et al. 2014), so a larger number of sessions would allow us to tackle issues more deeply and take them to their practical application more easily. Secondly, a longer time would allow participants to have more time to adapt to the professionals and other group peers, with some first introductory sessions so that participants can anticipate the group’s functioning and thus feel more comfortable (Cheak-Zamora and Teti 2015; Solomon, Goodlin-Jones, and Anders 2004).

TEAM_YOUNG ADULTS self-determination intervention program

The TEAM_YOUNG ADULTS program is a group intervention for youth with autism (17–30 years old) with the objective of working on skills related to the development of self-determination. It consists of 20 weekly sessions of about 1 h and 30 min that can be implemented online or in person. Each group is composed of 4–8 participants. The program’s main objective is to develop skills related to self-determination, such as making decisions, setting goals and objectives, planning to achieve them, solving problems, self-advocacy skills, knowledge of strengths and weaknesses and how to use them to achieve objectives, among others. The development of such skills is intertwined, at the same time, with the specific needs that people with autism have in relation to the development of self-determination. For example, difficulties in self-initiating actions, self-regulation and self-realization, and the tendency to getting blocked when facing problems, as well as a lack of empowerment (Andrés-Gárriz et al. 2024). For this reason, during the intervention and the promotion of self-determination, specific activities and forms of presentation were designed to meet the specific support needs of the targeted population. For instance, helping young adults with autism to understand their motivation processes and the impact of their interests when initiating actions towards their goals, supporting the development of self-management skills, working on cognitive flexibility and inhibitory control, and fostering and highlighting the steps that they take towards their goals to enhance their sense of empowerment, among others.

To design the intervention program, a comprehensive treatment model (Odom et al. 2010) was used from a constructivist perspective. With this objective, a naturalistic approach was adapted, following theoretical foundations of the Social Motivation Model (Dawson et al. 2004; Mundy and Crowson 1997) and within the paradigm of Garcia Winner (2008) Social Thinking. The program’s main objective is to improve self-determination in people with autism by developing skills related to this dispositional characteristic. The program is set within the theoretical framework proposed by Schalock and Verdugo (2002), with quality of life being understood as a multidimensional concept, consisting of the same dimensions for all people, being influenced by environmental and personal factors, as well as by interaction, and improving through self-determination, resources, inclusion, and life goals. Regarding self-determination, a key aspect of our group intervention, it is conceptualized within the framework of Causal Agency Theory (Shogren, Michael, et al. 2015). From this perspective, self-determination is understood as a ‘dispositional characteristic manifested as acting as the causal agent in one’s life’ (258).

The program was divided into 4 modules: (a) ‘Beginning and basic skills’, (b) ‘Self-determination’, (c) ‘Working with others’ and (d) ‘Closure’. We decided to start with some initial sessions focusing on working with autism-related difficulties to guarantee that all participants could develop the group with some similar bases (Figure 2). The wider number of sessions (10) was reserved to learn skills related to self-determination, and then the module ‘Working with others’ started, focusing on social skills and conflict resolution. Finally, the ending module dealt with the group’s feedback and perceived changes and was also used as the farewell (see Annex 2).

Figure 2.

Modules and sessions of the TEAM_YOUNG ADULTS program.

As the target goal of this intervention was self-determination, it was considered extremely important to offer participants a space where they could express their own goals and objectives, as well as aspects that they would like to work on. To ensure that participants could express these aspects, a previous individual space was offered at the beginning of the group, and work with the participants’ own objectives and situations was promoted throughout the entire program, to guarantee the usefulness of the intervention. Some authors promote this practice to increase motivation, participation, and personal agency in participants (Atkinson-Jones and Hewitt 2019; Hotton and Coles 2016; Ke, Whalon, and Yun 2018; Solomon, Goodlin-Jones, and Anders 2004; Wehman et al. 2014).

Procedure

The research project was approved by the Ethics Committee at Ramon Llull University and the Ethics Committee at Consorci Sanitari del Maresme (CEIm_76/20). All participants in the study were informed about its characteristics and signed the corresponding informed consent form. Parents signed the consent form if the participant was a minor (under the age of 18 in Spain). This trial was retrospectively registered at www.clinicaltrial.org (NCT05938751) on January 1st, 2023.

Recruitment of the pilot group in this study was carried out two months before the beginning of the program. All the possible participants had a first online individual visit (2 h long approximately) for the explanation and resolution of doubts concerning the program, to sign the informed consent form, and for the initial assessment. Once the initial assessment was conducted with all possible participants, they were randomly distributed into two groups: the experimental group (n = 7), who would develop the program in January 2021; or the control group (n = 7), who would remain in the waiting list until the next group in September 2021. The professionals with autism expertise from the child and youth mental health and adult mental health services at Hospital of Mataró carried out a first selection of participants following the inclusion criteria proposed by the researchers. After the proposal of participants, the first author was in charge of contacting the youth and their families, conducting the initial assessment, and randomizing the groups. The experimental group (EG) completed the TEAM_YOUNG ADULTS program conducted by two facilitators with expertise in group therapy and autism. The intervention was implemented online due to COVID-19 restrictions. The control group (CG) continued receiving the usual service consisting of follow-up (an individual visit every three to one to six months, with an average frequency of two months) conducted by the Mental Health and Addictions Service of Consorci Sanitari del Maresme. When the EG finished the intervention, participants in the CG were offered to carry out the program with the aim of guaranteeing principles of justice, beneficence, and equity (Figure 3).

Figure 3.

Process and temporalization of the project.

To ensure the scientific rigor of the intervention study, it was verified that the program’s application was reliable through checklists for all the sessions. These fidelity checklists considered both the implementation of the program’s sequence and the content. A checklist was prepared for every session taking into account that content varied. The fidelity checklists were consulted before and after the session, to have the structure clear at the beginning and check the application at the end. There were 19 fidelity checklists, as in the last session there was no specific content being taught, with the fidelity of the program for every module reaching 92%.

Measures of attendance and acceptability were only collected for the participants in the group that carried out the TEAM_YOUNG ADULTS program. Attendance was medium-high (79.95%). The participants’ main reasons for not attending some of the sessions were exams, doctor’s appointments, and job interviews. Results in social validity showed a good acceptability of the program, both by the participants and their families.

Data analysis

Regarding quantitative data analysis, firstly central tendency calculations were conducted. The nonparametric Mann-Whitney U test for independent samples was used to contrast changes in self-determination between the experimental group and the control group. The Wilcoxon signed-rank test for dependent samples was used to study changes in the experimental group and in the control group from the initial assessment to the post-intervention assessment. In both cases, a significance level of 95% (p < 0.05) was used. The analysis of results was carried out with the statistical software SPSS 25.0.

Regarding qualitative analysis, a descriptive qualitative research approach was used, and a thematic analysis was conducted following recommendations by Braun and Clarke (2006), to identify, analyze and find the main themes in the data collected via a six-phase procedure: (a) getting familiar with the data, transcribing, reading and rereading data, and posing main ideas; (b) generating initial codes; (c) arranging codes in themes; (d) revising themes and creating a thematic map of the analysis; (e) defining and naming themes, refining the definition of every theme; and (f) writing the analysis report. The results of the analysis were arranged hierarchically: codes were grouped into subthemes, and subthemes in themes.

This analysis was conducted with support from the software program Atlas.ti 9 for Mac (ATLAS.ti Scientific Software Development GmbH 2020). The first author conducted the analysis of codes from the focus group, developing the first three phases of thematic analysis as described by Braun and Clarke (2006). The second author revised the codes (30% of them, 13 out of 39), as well as the themes and subthemes and a thematic map of analysis (phase 4). The emerging codes, themes and subthemes were revised by the first and the second authors together until they reached a total agreement concerning the final coding (phase 5). As in other studies (i.e. Hurd, Evans, and Renwick 2018), a reflective discussion between the two authors was used to ensure a sound agreement on codes and themes. Narrative transparency was guaranteed through direct access to the participants’ perceptions by providing literal quotations from all of them (Trainor and Graue 2014).

Results

Below, the results of this study, whose main objective is to assess the effectiveness of the TEAM_YOUNG ADULTS program to promote self-determination in young adults with autism, are presented. Quantitative data are presented first to contextualize the qualitative data.

Quantitative changes in participants’ self-determination and evaluation of social validity

When comparing results before and after the intervention in the experimental group, although there were no statistically significant differences, there was indeed an increase in post-intervention statistical measures in total scores of self-determination as reported by the young adults with autism (M = 231.67, SD = 46.66) and their families (M = 123.17, SD = 27.5) (Table 3). This increase was observed in all the dimensions of the construct measured with the hetero-reporting instrument (AUTODDIS), and in the dimensions of agentic action and action-control beliefs with the self-report (SDI). In the volitional action dimension measured with the self-report (SDI), there was a slight decrease in post-intervention scores.

Table 3.

Comparison of pre- and post-intervention means of the experimental group.

	AUTODDIS (proxy-report)		SDI (self-report)
	Pre M (SD)	Post M (SD)	Pre M (SD)	Post M (SD)
Volitional action	35.50 (6.83)	39.17 (7.68)	79.00 (17.55)	78.00 (18.92)
Autonomy	20.17 (4.07)	21.34 (2.66)	–	–
Self-initiation	16.34 (3.78)	17.84 (5.64)	–	–
Agentic action	31.83 (13.08)	35.55 (11.64)	73.16 (13.79)	76.67 (16.06)
Self-direction	26.17 (11.55)	29.00 (9.42)	–	–
Self-regulation	5.67 (1.63)	6.34 (2.34)	–	–
Action-control beliefs	46.00 (11.40)	48.67 (9.90)	75.16 (13.93)	77.00 (14.40)
Self-knowledge	14.67 (5.72)	16.00 (5.10)	–	–
Empowerment	31.34 (7.5)	32.67 (6.28)	–	–
Self-determination total	114.33 (29.5)	123.17 (27.50)	227.33 (42.87)	231.67 (46.66)

Regarding post-intervention results between the experimental group and the control group, there were no statistically significant differences. However, the experimental group had higher scores, in comparison with the control group, in the total score of self-determination as reported by the young adults with autism and their families (Table 4). Moreover, in all the dimensions of self-determination measured with the self-report (SDI), and in the dimensions of agentic action and action-control beliefs with the hetero-report (AUTODDIS), participants in the experimental group had higher scores than participants in the control group. In the volitional action dimension (and its subdimensions: autonomy and self-direction) measured with the hetero-report, participants in the control group had slightly higher scores than participants in the experimental group.

Table 4.

Comparison of post-intervention means of the experimental group (EG) and the control group (CG).

	AUTODDIS (proxy-report)		SDI (self-report)
	EG M (SD)	CG M (SD)	EG M (SD)	CG M (SD)
Volitional action	39.17 (7.68)	39.8 (6.61)	78.00 (18.92)	66.80 (17.15)
Autonomy	21.33 (2.66)	21.40 (3.21)	–	–
Self-initiation	17.83 (5.64)	18.4 (3.85)	–	–
Agentic action	35.33 (11.64)	28.4 (10.33)	76.67 (16.06)	68.6 (16.52)
Self-direction	29.00 (9.42)	22.60 (10.26)	–	–
Self-regulation	6.33 (2.34)	5.80 (2.17)	–	–
Action-control beliefs	48.67 (9.91)	45.60 (8.71)	77.00 (14.35)	66.20 (9.12)
Self-knowledge	16.00 (5.10)	15.00 (3.67)	–	–
Empowerment	32.67 (6.28)	30.60 (5.41)	–	–
Self-determination total	123.17 (27.49)	113.80 (24.30)	231.67 (46.66)	201.60 (38.49)

Concerning social validity, participants expressed a good evaluation of the group (MD = 8.2, SD = 1.48), the facilitators (MD = 8.6, SD = 1.34), the importance of the content (MD = 7.6, SD = 1.67) and the ability of the program to help them to think and solve problems (MD = 7.8, SD = 2.17). Participants valued the materials and information of the program as excellent (MD = 9.00, SD = 1.23) and 100% of them would recommend the group to people in a similar situation. Family members also evaluated the program positively, with excellent satisfaction with the program (MD = 9.8, SD = 0.5) and 100% would recommend the group to other young adults in the same situation as their children.

Qualitative changes in participants’ self-determination

Out of the thematic analysis, 5 themes emerged: changes perceived in volitional actions, changes perceived in agentic actions, changes perceived in action-control beliefs, supports and barriers for change.

Changes in volitional actions

Two subthemes were apparent concerning changes in volitional actions: autonomy and self-initiation. Changes in autonomy were perceived, particularly when making their own decisions and establishing a decision process.

I’ve noticed that my decision-making has improved. Well, mostly because my starting point was very low. (Participant 2)

On one hand, I want to leave home and want to be with him [significant other] and live my life with him. On the other hand, I don’t feel so fine, because… my mother is my mother, and my father is my father. […] And when my mom is not fine, she needs me. (Participant 3 when deciding whether to move abroad with her significant other or staying at home with her family)

Regarding self-initiation, participants reported being able to carry out what they have planned, a better quality of decisions, and thinking before acting.

I had always been a person who would think and think about doing something and when I had already made a decision it was too late and basically there was nothing to do. Now I see that I still need time to think but it has been reduced a lot. […] I’ve realized I don’t waste so much time thinking and dwelling in a state of blockage doing nothing. […]. I realize that now I can make… not only make a decision, which was already difficult for me in itself, but I can put it into practice, now it seems to me that this happens in a much more automatic way. (Participant 2)

Changes in agentic actions

Two main subthemes emerged concerning agentic actions: self-direction and self-regulation. Participants reported changes in self-direction regarding planning to achieve their goals and the moment that they were at, as well as self-advocacy skills. It is worth highlighting that all participants took steps towards achieving their goals.

About the goal of going to the nutritionist’s and all, I’ve lost weight, from 90 kg to 77. (Participant 1, regarding her goal: to lose weight)

About spending more time with my family, I can say I’ve improved quite much, because well, before, whatever the reason, basically I couldn’t, and now, […] I’d say that I’ve made some progress in this short time. (Participant 2, regarding his goal: to spend time with his family)

Well, you know… I have some savings. All the money I’ve saved goes to my mum (Participant 3, regarding her goal: to earn money to help her parents)

I’m watching Ladybug in French: with English subtitles! (Participant 4, regarding her goal: to learn French)

And, for example, I would like to learn to play the guitar. Because of my studies I’ve not been able to do it much, but I’ve been able to do it and improve a little, and to learn a little. (Participant 5, regarding his goal: to learn to play the guitar)

Concerning self-regulation, problem solving (looking for alternatives), monitoring and changes in initial goals stand out. Participants pointed at a change in their way of facing problems and thinking about different alternatives to solve them.

I’ve learned more ways of facing my own problems. Because before I was more of facing a problem and banging my head with it. (Participant 2)

Participants also reported having been able to assess the situations that were happening to them and act in consequence. Participant 2 even talked about a change in his initial goal to look for a job. He had found a job right after starting the group but was beginning to consider changing to a job where he could have better conditions.

When I finally managed to process what had happened, I started to be able not to react immediately. (Participant 4 referring to a problem she had during the group sessions)

It’s funny because my goal has changed, it’s not to find a job anymore, but to find another job where I can work sitting down. (Participant 2)

Changes in action-control beliefs

Participants also mentioned changes in action-control beliefs, with three subthemes emerging: control of expectations, empowerment, and self-realization. Regarding control of expectations, participants talked about different assessments of experiences that they had had and their ability to ask for help. For example, Participant 5 had the main goal to pass his pre-university course, which he could not achieve.

Although it didn’t go well [the pre-university course], it was good for me to do it, and I’ve learned a lot of things (Participant 5 referring to his experience in the first year of his pre-university course)

Advice… talk… ask for advice, especially (Participant 5 answering about what had helped him to make the decision to change studies).

Empowerment included such aspects as accepting oneself and trusting one’s own decisions and capacities.

As I have higher self-esteem, all my other capacities also have intensified. That is, maybe I can empathize better with others or be more decisive, etc. It helps with everything, really (Participant 4)

I trust the decisions I make a little more. I don’t doubt so much, when making a decision. (Participant 1)

Finally, the improvement in self-realization was mainly expressed through some better knowledge of their strengths and weaknesses after the intervention.

Getting to know myself a bit better over these months and also see what good and bad aspects I have. The good ones, to reinforce them, and the bad ones, to know they are there and to improve them. (Participant 1 when answering to what had helped her to trust her decisions more)

Supports for change

Beyond changes in self-determination, participants mentioned supports that had facilitated their perceived improvements. Among these supports, three subthemes were identified: people who support, opportunities and exposure to situations, and specific skills training (the TEAM_YOUNG ADULTS program). Regarding people who support, mothers and friends stood out.

Communication with my friends has helped me quite much because at some point they have made me see that some of my ways of communication were not very assertive or not the best ones for the situation (Participant 4)

The opportunities and exposure to situations helped participants to put the skills taught in the group into practice.

Because I have had to come up against more of them and I am a little more experienced (Participant 2 answering to why he feels he is more capable to deal with his problems)

Regarding specific skills training (within the TEAM_YOUNG ADULTS program), participants reported the advantages of training these skills, the advantages of developing the program in group, better adaptation to unexpected situations, and, in the case of Participant 4, how the combination of the group with individual therapy had helped her.

Throughout the group sessions, the advice you’ve been giving us remains in our sub-conscious in the end. And the decisions we are making are based on this. When, for example, solving problems, well, whether you like it or not, it helps quite much. (Participant 1)

Observing other situations and learning from them. Because we’ve learned not only from what happens to us, but using empathy, seeing how another person who is completely different must come up against other situations and how they deal with them. With ways that may be different from what we could think of. (Participant 2)

Barriers for change

Finally, some barriers for change stood out, with two subthemes emerging: (a) barriers to agentic action, and (b) barriers to action-control beliefs. Particularly, barriers to agentic action detected were contextual problems/barriers, not being constant when following the plan to achieve a goal, and cognitive rigidity. Within contextual problems/barriers, people in their environment and unfavorable working conditions, and time needed for academic activities were outstanding.

I’m very busy right now with class and this takes all my time (Participant 4)

For their part, barriers to action-control beliefs were mainly expressed in little awareness of steps made towards the goal. This barrier was only identified in Participant 4, together with cognitive rigidity. In spite of having finished her studies, having written her own CV, having looked for summer jobs in a florist’s, having watched series in French to practice this language, this participant focused more on the steps that she still had to take.

I’m studying, but I haven’t sent CVs yet, and I haven’t looked for how to learn French. (Participant 4)

Integration of quantitative and qualitative results

There were no statistically significant differences between pre- and post-intervention results, or in comparison with the control group. Results showed an increase in the measures of the three dimensions of self-determination based on data provided by the families, and an increase in agentic action and action-control beliefs according to participants. Qualitative data help us understand these improvements better, and also contribute with relevant information regarding volitional action, which is explained by quantitative results.

Firstly, participants reported having taken steps towards achieving their goals, to a greater or lesser extent, in their skills to defend their own rights, in solving problems and monitoring their progress. In the case of one participant, there were changes introduced to his goals after assessing his experience at work.

Secondly, the young adults with autism reported higher acceptance of themselves, higher confidence in their decisions and capacities, knowing how to ask for help when necessary, having the capacity to assess their experiences, and being aware of their strengths and weaknesses. These improvements coincided with the increase in action-control beliefs reported in the questionnaires.

Finally, refining the results from the quantitative analysis of the self-report (SDI), participants perceived improvements in their decision-making and their capacity to carry out what they had planned, as well as the quality of their decisions and thinking before acting. This assessment by participants did not coincide with results obtained in the self-report questionnaire, which did not detect improvements in this dimension, but rather a slight decrease. Our hypothesis is that the qualitative analysis manages to grasp details about the quality of the volitional action that go beyond those from the questionnaire.

Other important data from the qualitative analysis are the influence of supports and barriers for change on results. Firstly, the most significant supports seemed to be people who support (such as family and friends), the opportunities of exposure to situations to develop self-determination and specific skills training (that is, the TEAM_YOUNG ADULTS program). The intervention program allowed participants to get trained in self-determination specific skills, adapt to unexpected events and changes better, and take advantage of the qualities provided by the group format. Secondly, we detected barriers for change that may have hindered a significant improvement in agentic action and action-control beliefs. There were contextual barriers, such as not very favorable people and contexts, lack of perseverance when developing their plan to achieve goals, cognitive rigidity (difficulties in seeing different ways of thinking, adjusting to situations or avoiding dichotomous thinking), and little awareness of steps already taken. These two latter barriers, cognitive rigidity and little awareness of steps taken, were detected in one participant (Participant 4), who despite her progress did not think that she had come closer to her goals.

Discussion

This research addresses the urgent need to dispose of effective interventions in the development of self-determination to facilitate the transition to adulthood of young adults with autism in the Spanish context. We implemented a mixed-methods approach to assess the TEAM_YOUNG ADULTS program, an intervention focusing on the development of skills related to self-determination in young adults with autism. Results highlight the intervention’s acceptance amongst participants and its effectiveness, although the results must be considered with caution due to the small sample size and cannot be generalized. The program’s facilitators conducted it with a high level of fidelity, and the social validity was positive. Although there were no statistically significant changes in quantitative data, qualitative data stressed the participants’ improvement in self-determination dimensions.

Families reported improvements in volitional action, agentic action, and action-control beliefs, although young adults with autism only reported quantitative increases in agentic action and action-control beliefs. The results from the present study could not explain these differences because only quantitative data was collected from caregivers. Therefore, including qualitative data from family members, apart from the young adults with autism, may help to comprehend the differences in the results in future studies. On the other hand, the aforementioned differences in the perception of change align with previous research that stresses the importance of considering insight problems that most people with autism have to be aware of their difficulties in expressing their emotions and mental states (Atkinson-Jones and Hewitt 2019). For this reason, data must be triangulated by using different assessment methods (e.g. self-reports together with proxy-reporting) to ensure data objectivity (Atkinson-Jones and Hewitt 2019; Hotton and Coles 2016; Spain and Blainey 2015). Within self-determination research field, recent studies have highlighted the utility to obtain self-determination assessments through various informants (Trainor and Graue 2014; Mumbardó-Adam et al. 2023; Vicente et al. 2020).

According to qualitative data, changes were observed in the three dimensions of self-determination. With respect to changes in volitional action, participants reported increases both in their autonomy and their self-initiation after intervention. They were more prone to engage in decision-making processes, they highlighted a higher quality of their decisions and a greater capacity to carry out the decisions made, and to think before acting (thus avoiding impulsive behaviors). Even though this progress was not reflected in the participants’ questionnaire results, they were clearly reflected in their family’s assessment and in the qualitative results. These findings highlight that, despite the capacity of people with autism to make decisions, there is a need for more individualized intervention and assessment instruments that delve into the quality of these decisions and the way to implement them. Further research should thus focus on the creation of self-determination measurement instruments specific for people with autism that may be more sensitive to their slightly different aspects of the diverse skills connected to self-determination.

Changes in agentic action showed an increase in self-direction and self-regulation. Participants were able to set goals and planning steps to achieve them. Regarding goal attainment, there were differences among participants. Some of them managed to achieve their goals and even changed or expanded them. Others took some steps towards their goals without actually achieving them, and some had to regulate their actions and even set new goals. On one hand, and according to Causal Agency Theory (Wehmeyer et al. 2017), self-regulation not only implies changing the behavior to attain goals, but also involves the ability to evaluate progress and decide if the goal must be changed. According to this definition, the fact that participants set new goals without achieving former ones also indicates a development in self-determination-related skills. On the other hand, considering that some of the most relevant difficulties of people with autism concerning self-determined behavior are problem solving, goal setting and attainment (Chou et al. 2017; Ruble and Scott 2002), all of them connected to agentic action, more intervention time is possibly needed to see statistically significant changes, in accordance with previous research (Shogren, Michael, et al. 2015). Some of the barriers for change detected in this dimension were precisely cognitive rigidity when considering an achieved goal and the lack of perseverance to develop a plan, both closely linked to the difficulties mentioned by Ruble and Scott (2002), such as problems to engage in goal-directed behaviors and adjust behaviors to environmental demands.

Regarding changes in action-control beliefs, participants expressed they were more able to control expectations, empowerment, and self-realization. Participants reported an increase in the assessment of their experiences, in their capacity to ask for help, as well as in accepting themselves and trusting their own decisions and capacities. They also reported a better knowledge of their strengths and weaknesses. These changes are promising, as action-control beliefs is one of the dimensions with fewer changes in previous interventions (Nadig et al. 2018; Oswald et al. 2018; Shogren et al. 2016; Wehmeyer et al. 2003). As action-control beliefs nurture volitional and agentic action (Mumbardó-Adam, Guàrdia-Olmos, and Giné 2020), interventions aimed at promoting self-determination cannot only focus on producing changes in volitional and agentic actions. Nevertheless, in every change and improvement process, beliefs are usually the aspect with more resistance and that takes more time to change, thus emerging as one of the most relevant future lines of research, scarcely explored up to now.

Beyond changes achieved in self-determination, participants highlighted the importance of considering supports and barriers for change. The main supports perceived were access to opportunities and experiences, the availability of close people to support them, and skills-training through the TEAM_YOUNG ADULTS program. Likewise, the program is perceived by participants as a support that has facilitated skills learning, has helped them to adapt better to unexpected situations, and has offered a group space for participants to share situations and experiences. Regarding the availability of close people to support them, further research on interventions to promote self-determination should contemplate the role of families of young adults with autism to promote enabling environments and act as a support, considering the importance of these supports and their influence in the immediate context of participants. Some studies have already started to consider this kind of interventions, including the figure of coaches to their programs (Oswald et al. 2018). Regarding barriers for change, obstacles towards agentic action and action-control beliefs were detected. Amongst barriers for agentic action, contextual problems and barriers, some working conditions, as well as some people who would hinder the achievement of goals, were identified. Barriers for action-control beliefs were related to the low awareness of the steps taken towards goals, thus possibly affecting empowerment. The identified barriers for change must be considered when working towards self-determination development in young adults with autism. Interventions must endeavor in increasing participants insight and self-awareness and closely work with these young adults’ families and significant others to promote contexts that facilitate and enhance opportunities for self-determined actions (Mumbardó-Adam, Guàrdia-Olmos, and Giné 2020; Vicente et al. 2020).

Qualitative data has allowed an in-depth observation of other aspects about the quality of self-determination-related skills, and thus further research should consider using a combination of quantitative and qualitative data (Bergman 2008; Fàbregues et al. 2022) to better comprehend changes occurred throughout interventions. Regarding the TEAM_YOUNG ADULTS program, results highlight the importance of considering the particular characteristics of autism in the interventions to promote self-determination thus avoiding these characteristics to become barriers for self-determination development (Andrés-Gárriz et al. 2024). Besides, we should consider offering maintenance sessions after the intervention; to provide participants with support towards goal attainment.

Finally, this study is not exempt of limitations that should be considered when interpreting the results. Firstly, and although the objective of this pilot study was to test the program with a first group of participants, the number of participants is small but appropriate for the aim of the study and for the group intervention format. This limitation could also be related to the lack of statistically significant results. Then, we should consider that this study is just a first step towards assessing the effectiveness and acceptance of the TEAM_YOUNG ADULTS program. Further research should replicate the implementation of the program with more participants to add more evidence of its effectiveness. Secondly, we did not include qualitative data on families’ perceptions, which would have enriched the comparison of outcomes between participants and their family members. Inclusion of this data could help to understand the differences in outcomes perceived by each group. Thirdly, another limitation is that our program does not include participants’ contexts, such as families or educators. Further research should work with the family and/or educational or working context of participants to offer more opportunities for change (Mumbardó-Adam, Guàrdia-Olmos, and Giné 2020; Vicente et al. 2020).

Funding Statement

This research has been funded by the Fundació Blanquerna (Blanquerna Foundation) with a contract for predoctoral researchers in training (PIF1920 – COMSAL).

Ethics statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Ramon Llull University and the Hospital of Mataró (CEIm_76/20, October 10 of 2020).

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.