Abstract
Adolescents with autism spectrum disorder (ASD) are vulnerable to declines in social connections and an increase in depression, anxiety, and other co-occurring conditions. This study introduces a novel intervention that matches adolescents and adults with ASD in one-to-one mentoring relationships in an afterschool setting and examines its social validity. In this single-group, mixed-method pilot study, participants were seven adolescent mentees (14–18 years old; 100% male), seven adult mentors (19–33 years old; 71% male), and eight parents of mentees. A combination of project-specific and standardized assessments was used to describe the participants’ perceptions of the program and to assess well-being, self-concept, and social-emotional and behavioral outcomes. Results showed high uptake, program satisfaction, positive ratings of mentoring relationship quality, and desirable pre- to post-test change on several targeted outcomes. This study provides preliminary evidence to support the applicability and utility of a mentoring program for adolescents with ASD by adults with ASD.
Autism spectrum disorder (ASD) is a lifelong developmental condition that affects a person’s ability to understand and communicate with the social and physical world. ASD is characterized by social communication impairments, as well as repetitive patterns of behavior, interests, and activities (American Psychiatric Association, 2013). Currently, 1 in 54 children are identified with ASD nationally (Maenner etal., 2020). Although numerous intervention options exist for young children with ASD, the availability of supports wane by adolescence (Reichow & Volkmar, 2010). Adolescents with ASD face the same developmental changes as those without ASD, such as puberty, egocentricity, and individuation. At the same time, they often experience additional difficulties that coincide with the disorder (e.g., sharp declines in social connections, stalled neurological progress, and a marked increase in depression, anxiety, and other negative outcomes; Cassidy etal., 2014; Kasari & Sterling, 2013; Vasa etal., 2018). Although many adolescents with ASD are seeking social relationships (McGovern & Sigman, 2005), they are also considerably more aware of social isolation and loneliness (Kuusikko etal., 2008). Finding community and belonging is critical for positive youth development, including for youth with ASD.
Adult–youth mentoring relationship, conceptualized as a relationship between a young person and a more experienced adult acting in a nonprofessional capacity to provide support (DuBois & Karcher, 2014), is a familiar and encouraging strategy that may provide necessary support and guidance for navigating adolescence with ASD. Mentoring relationships are regarded as key developmental assets that can promote positive development (Bowers etal., 2016; Theokas & Lerner, 2006). Youth involved in formal mentoring programs, in which youth mentees and adult mentors are matched, supported, and supervised by program staff, have shown improvements in peer and family relationships and social-emotional development (Raposa etal., 2018). Despite the robust literature documenting the impact of youth mentoring interventions, few studies have documented mentoring for youth with ASD. One notable exception was a preliminary study by Curtin et al. (2016), which showed the promise of a 6-month mentoring intervention for 13- to 18-year-old adolescents with ASD with regard to self-esteem, social skills, and quality of life. In this program, college-age mentors were trained to mentor teens with ASD. Additional programs targeting college students with ASD found in the extant literature (e.g., Lucas & James, 2018; Ncube etal., 2019) also showed promise of mentoring to affect social and emotional outcomes. After searching three databases using keywords related to ASD, autism, mentee, mentor, mentoring, and mental health and social-emotional outcomes and excluding studies involving college students as mentees or studies focused on academic or career outcomes, no literature describing programs utilizing mentors with ASD were found.
Theoretically, high-quality mentoring relationships are characterized by mutuality, trust, and empathy (Rhodes etal., 2002). Although perceived similarity (e.g., ASD experiences) is not a prerequisite for forming meaningful mentoring relationships, it may offer a strong foundation for relationship building and a context for building trust. In previous studies, matching mentors and youth based on shared interests (DuBois etal., 2011) or shared dislike of activities (Raposa etal., 2018) was linked to improved program outcomes and relationship length, respectively. Without some mutual understanding of one another’s experiences, the full benefit of this type of relational support may not be realized (Raposa etal., 2018). Relatedly, adults with ASD are often underutilized in their community, despite having wisdom and experience that could be critical to supporting a teen with ASD (e.g., finding a passion or spark, applying to college, applying independent living skills, negotiating peer relationships, coping with depression or suicidal ideation; Kirchner etal., 2016; Solomon, 2020). Considering the reciprocal relationship of mentoring, mentors with ASD may also glean unique benefits. For instance, neurotypical mentors realize benefits in areas such as civic engagement, interpersonal skills, and self-esteem (Goldner & Golan, 2017; Weiler etal., 2013). Serving as a mentor could provide opportunity for social participation, an important avenue by which adults with ASD experience greater quality of life (Tobin etal., 2014).
Youth mentoring, as a relationship-based intervention, also poses potential challenges among mentees and mentors with ASD. Because individuals with ASD may have difficulty with social interactions (Mendelson etal., 2016; Scheeren etal., 2012), unique barriers may exist with building and maintaining the mentoring relationship. Communication, social-cognitive, and emotion regulation deficits have an evident impact on the ability to build relationships (Liss etal., 2001). Individuals with ASD may have difficulty understanding others’ thoughts and feelings thereby affecting their behavior and response in a given situation (Travis etal., 2001). Limited understanding of social norms or ability to read others may affect the quality of the mentoring relationship, especially if conceptualized as personal (e.g., a mentor who perceives mentee’s quietness as dislike). As such, mentoring interventions for adolescents with ASD must take into account the possibility of relationship-related barriers throughout the program design.
In this proof-of-concept, preliminary study, the Autism Mentorship Program (AMP) is presented and evaluated for social validity and preliminary outcomes. AMP is a strengths-based mentoring intervention designed by invested community members, including adults with ASD, for youth with ASD. A relational intervention delivered by adults with ASD is a plausible strategy for promoting positive youth development. Yet, because interpersonal difficulties may arise, the program was iteratively and respectfully developed. This study represents one step toward understanding the capacity of this type of intervention strategy to affect outcomes among adolescents with ASD. The primary aim was to determine social validity, including uptake, program satisfaction, and mentoring relationship quality. A secondary aim included the assessment of pre- to post-test changes on well-being, self-concept, and social-emotional and behavioral outcomes. Results of this study should serve as proof-of-concept for future research on mentoring adolescents with ASD and inform improvements in program design and implementation.
Method
Participants
Participants were seven adolescent mentees, eight parents of mentees (i.e., two parents of one mentee participated), and seven adult mentors (see Figure 1). Adolescent participants were included in the study if they attended the participating high school and provided a clinical or educational evaluation report documenting ASD, no intellectual disability, and verbally fluent language. Mentees and mentors were excluded if they did not speak or understand English, although interpreters were available for parent participants as needed.
Consolidated standards of reporting trials (CONSORT) diagram of study participants.
Adolescent mentee participants
Mentees were adolescents attending a large, urban high school in the U.S. Midwest with a medical diagnosis or designated educational eligibility category of ASD within their Individualized Education Program (ages 14–16 years; 100% male; n = 7). Mentees also reported medical diagnoses of attention-deficit/hyperactivity disorder (ADHD; n = 5), anxiety disorder (n = 4), and other various conditions (e.g., post-traumatic stress disorder; n = 3). Parents reported just over half of the mentees (n = 4) identified as non-Hispanic, while three mentees identified as Hispanic. Parents reported on their child’s race as follows: two White, one American Indian/Alaskan Native, two more than one race, and three “other” specifying Latino/Hispanic/Mexican American in their written response. Parents of mentees reported a range of incomes from <US$10,000 annually (n = 2) to US$120,000–US$139,999 annually (n = 1), with the others falling in between.
Adult mentor participants
Mentors were adult volunteers with ASD (ages 19–33; 71% male; n = 7) from a Midwest metropolitan community who completed the program application process and had a clinical diagnosis of ASD prior to participating in the study. Mentors also reported co-occurring medical diagnoses of anxiety disorder (n = 4), ADHD (n = 3), and other various conditions (e.g., depression; n = 2). Six of the seven mentors attended a local community-based transition program for young adults with ASD. All mentors identified as non-Hispanic (n = 7). Six mentors also identified as White and one chose not to answer. Mentors reported a range of incomes from US$0–US$9,999 (n = 1) to US$60,000–US$69,999 (n = 1), with the others falling in between. Regarding their highest level of education, four mentors reported receiving a high school diploma or general educational development (GED), two reported 1–3 years of college, and one reported an Associate degree. At the time of the study, approximately 71% of mentors were employed either part-time or full-time (n = 5).
Procedures
All procedures involving human participants were approved by the Institutional Review Board and were implemented in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Declaration of Helsinki as revised in 2000. Participant consent and assent was obtained after reviewing and discussing the consent form and study procedures with each participant. Per institutional guidelines, adults with ASD also completed the University of California, San Diego Brief Assessment of Capacity to Consent (Jeste etal., 2007) at the time of consent in order to document understanding of the study and the capacity to give consent. Data were collected from mentees, parents of mentees, and mentors. Participants completed a combination of project-developed questionnaires and standardized assessments prior to the start of the program (i.e., baseline) and following program completion (i.e., post-intervention). Each data collection session required approximately 60–90 min per participant. Participants received a US$15 gift card for each completed data collection activity (i.e., pre-test, post-test).
Intervention
AMP involves one-to-one mentorship of youth with ASD by adults with ASD. The purpose of AMP is to provide a buffer against the isolation, emotional difficulties, and social challenges youth with autism often confront in adolescence. Meeting these goals can be obtained by establishing meaningful connections with mentors who know firsthand the challenges of autism and who can point the way forward for their younger peers at a particularly vulnerable period of development. At the same time, the program seeks to give opportunity to adults with ASD to serve as role models and leaders in their community, where they are highly valued specifically for the insights their autism brings, and play an active role in improving life for the next generation. The program was developed based on the essential elements of effective youth practice (Garringer etal., 2015) and in collaboration with the program’s steering committee, which consists of adults with ASD, parents of children with ASD, special educators, and professionals that work with individuals with ASD, and the local state partnership of youth mentoring. The steering committee met biweekly for about 1 year to develop the key components of AMP, which center on an individualized relationship (i.e., mentoring), and continue to meet regularly to inform research activities and to contextualize results.
The program mentees were students in a local public high school autism program who were encouraged to apply by their special education teachers. They completed an application, including an interest survey, and were then interviewed by the AMP program manager to determine their fit, interests, communication styles, and any potential emotional triggers so they could be well-matched with their mentors. The mentors were primarily recruited from a local 3-year postsecondary, life skills training program for young autistic adults and encouraged to apply. To fill one empty slot, the seventh mentor was recommended to the program by a board member of the local autism society. All mentors completed an application process, which included an interview with the program manager, reference checks, and a background check. Mentors and mentees were then matched based on identified characteristics important to the individual, their interests, and experiences. Mentors were matched with mentees at least 3 years younger (range of age difference: 3–19 years). Mentors, mentees, and their parents received 2 hr of training in general mentoring best practices and autism-specific education prior to the start of the program. Mentors completed this training as one group and mentees and their parents were in a separate group. Training for mentors focused primarily on how to build effective mentoring relationships, communication skills, adolescent development, and program policies and procedures. Training for parents and mentees focused primarily on how to get the most out of a mentoring relationship and program policies and procedures. Mentors also participated in a 2-hr mid-program refresher training and feedback session and mentees had a 1-hr in-service training session.
The program included 12 weekly 60-min mentoring meetings that occurred after school at the high school. Mentoring sessions were semi-structured around shared interests (e.g., a pair could play a game, draw, build with Legos, and/or just talk with each other) with occasional planned group activities, allowing for open conversation and time for larger group interaction with other pairs. Midway through the 12 weeks, based on feedback from mentors and mentees, proposed weekly conversation topics were provided from a list of ideas generated by the mentors (e.g., getting a job). Central to the intervention is the mentoring relationship. As such, mentors and mentees could choose how to spend their time together, so long as the relationship and mentees’ positive development was prioritized. In practice, this could look like building with Legos while the mentor checks in on the mentee’s experience of friendships and/or bullying. Mentoring pairs met in the same room at the high school and familiar school and AMP staff were available during each mentoring session and offered support, as needed. In between weekly sessions, the project manager reviewed participant feedback forms completed at the end of every session and followed up directly with mentors, mentees, or their parents. She also offered reminders, support, and encouragement to prepare for the next session. The end of the program was celebrated with a pizza party where each participant received a certificate of completion and verbal communication thanking them for their particular contributions to the program.
Measures
Measurement for the current study is as follows. All measures were assessed at baseline and post-intervention with two exceptions. Demographic variables were assessed only at baseline and items specific to participants’ experience with the program were assessed only post-intervention.
Demographic questionnaire
A demographic questionnaire was developed to gather information on participant background characteristics. Parents reported demographic information for themselves and their child and mentors self-reported demographic information. Information was gathered regarding participant gender, race, ethnicity, household income, diagnoses, and education level.
Participant well-being and program questionnaire
A questionnaire was developed to gather information on various aspects of mentee and mentor participants’ daily life and experiences in the program. This questionnaire was adapted from a previous study evaluating intervention outcomes for transition-age individuals with ASD (Dawalt etal., 2021). Mentees and mentors answered both quantitative questions (e.g., Likert-type scale) and qualitative questions (e.g., open-ended) about themselves. Mentee and mentor participants were asked to self-report on their quality of life, frequency of negative feelings, pride in autistic identity, self-satisfaction, relationship quality, challenging behaviors, and level of independence. Sample items included: “How satisfied are you with your current level of independence?” and “To what extent do you feel pride in your autistic identity?” Parents responded to parallel questions and reported on their perception of their child’s well-being and daily life. The qualitative questions asked about the experiences of mentees, parents, and mentors with AMP, such as perceived challenges and benefits of the program.
Mentoring relationship quality
Mentees and mentors reported on the quality of their mentoring relationship via the Strength of Relationship scales (Rhodes etal., 2017). Using a five-point scale from strongly disagree to strongly agree, mentors completed 14 items, such as: “My mentee and I are interested in the same things” and “I feel confident handling the challenges of being a mentor.” Similarly, mentees responded to 10 items, such as: “My mentor has lots of good ideas about how to solve a problem” and “When I’m with my mentor, I feel mad.” Negatively worded items were reverse coded such that higher scores indicate higher quality relationships. The scales showed internal consistency and were predictive of relationship length in previous research (Rhodes etal., 2017). This is the first known study to use this scale with participants with ASD.
Self-concept
To assess self-concept, mentees and mentors completed the Piers-Harris Self-Concept Scale, Second Edition (Piers-Harris II). This measure contains 80 true/false statements and six subscales relating to self-concept, including behavioral adjustment, intellectual/school status, physical appearance, freedom from anxiety, popularity, and happiness/satisfaction. Higher scores reflect greater self-concept. The authors report strong retest reliability and internal consistency as well as concurrent validity (Piers, 1984), and it has been used previously in evaluating interventions among ASD samples (e.g., Chang etal., 2014).
Social-emotional and behavioral outcomes
The Achenbach System of Empirically Based Assessment (ASEBA) questionnaires were used to measure youth’s and mentors’ internalizing problems (e.g., withdrawal, somatic complaints, and anxiety/depressive symptoms) and externalizing problems (e.g., delinquent and aggressive behavior). Parents of mentees completed the Child Behavior Checklist (CBCL), mentees completed the Youth Self-Report (YSR), and mentors completed the Adult Self-Report (ASR). The response format is on a three-point scale, with response options ranging from not true, somewhat true, or sometimes true to very true or often true. The psychometric properties of the CBCL, YSR, and ASR are well established (Achenbach & Rescorla, 2001; Achenbach & Rescorla, 2003; Greenbaum & Dedrick, 1998) and have been used previously with participants with ASD (e.g., Arias etal., 2021; Lohr etal., 2017; Pandolfi etal., 2012).
To assess social competence (e.g., cooperation, engagement, assertion, and communication) and problem behaviors (e.g., internalizing, externalizing, bullying, hyperactivity, and inattention), mentees, parents of mentees, and mentors completed the Social Skills Improvement System (SSIS) Rating Scales. Mentees and mentors both completed the 13- to 18-year-old forms as past research supports the use of this scale with young adults with ASD (e.g., McVey etal., 2016). Parents completed the parent-report form. Items were rated on a four-point scale, with response options ranging from never or not true to almost always or very true. Numerous studies have supported the validity and reliability of the SSIS (Gresham etal., 2010; Gresham etal., 2011).
Data Analysis
Data files were imported into SPSS Version 25 for statistical analyses. Descriptive statistics were used to describe the characteristics of participants and to determine program uptake and implementation, as well as mentor- and mentee-reported mentoring relationship quality. For measures that included pre- and post-test assessments, we calculated effect sizes using Hedges’ g to account for the small sample size. The effect size for paired-samples t tests was quantified by calculating the corrected, unbiased Hedges’ g (Hedges & Olkin, 1985). Values above .2 were deemed a small effect, above .5 a moderate effect, and above .8 a large effect (Lakens, 2013).
Results
Social Validity
Attendance rates indicated high uptake of the program. Four mentees attended all 12 sessions, and three mentees attended 11 sessions. Five mentors attended all 12 sessions and two mentors attended 11 sessions. Regarding the implementation of the program, all but one session lasted the full hour, and program activities were implemented as planned. The most frequent activities included conversation, games, art, and building materials, while the least frequent activities included physical activity, puzzles, books, puppets/characters/action figures, and story cubes. On average, mentees rated the quality of their mentoring relationship as 4.00 on a five-point scale, while mentors rated the quality of their relationship at 3.89 on a five-point scale.
One-hundred percent of mentees, parents of mentees, and mentors reported feeling satisfied with the mentoring relationship and 100% of mentees and mentors agreed or strongly agreed their participation was meaningful. All mentees, parents of mentees, and mentors agreed mentees and mentors were well matched and reported learning valuable information and skills during the program. The majority of mentees (85.7%) and mentors (85.7%) reported feeling satisfied with their participation in the program and the level of support they received. Most parents (85.7%) indicated the program helped prepare their child for the future.
Mentees, parents of mentees, and mentors answered open-ended, project-developed questions regarding their perception of the program. First, mentees described quality relationships, progress toward personal goals, learning about people and communication, and minimal challenges throughout the program (with the exception of group activities, which was noted as a difficulty for some). When asked about the biggest take-away from the program, mentees described feeling more comfortable talking with people and seeing a positive trajectory forward due to having an older person to look up to and go to for guidance. Second, mentors reported greater confidence, more patience, and better leadership skills as a result of participating in the program. Mentors noted their mentees showed improvements in communication, including sharing more about themselves over time. A challenge noted by mentors was some difficulty helping mentees stay focused on a single topic or going deeper in the conversation. Finally, parents described improvements with their child’s communication, emotional expression, motivation, attention, confidence, and friendships. When asked about their own experience, the majority of parents indicated satisfactory experiences with the program regarding appropriate implementation and positive opportunities for their family. Some parents desired more communication regarding their child’s progress, engagement in activities, and feedback that could be generalized and encouraged at home. One parent suggested social interactions outside of school.
Preliminary Outcomes
Well-being
Mentees, parents of mentees, and mentors were asked to report on well-being (see Table 1). In comparison of pre-test to post-test means, mentees noted a small increase (g = .34) and parents noted no change (g = .00) in their child’s quality of life. Mentees also indicated a small increase (g = .34) and parents indicated a small decrease (g = −.41) in life satisfaction for their child. Regarding enjoyment in life, mentees reported a moderate increase (g = .70) and parents reported a small increase (g = .27). Moreover, mentees indicated no change (g = .00) and parents indicated a small to moderate increase (g = .47) in the extent to which their child feels pride in their autistic identity. Mentees also noted a small to moderate increase (g = .46) in their self-satisfaction. Mentees reported a small increase (g = .37) and parents reported a moderate to large increase (g = .73) in their child’s relationship satisfaction. Finally, mentees indicated no change (g = .00) and parents indicated a small decrease (g = −.15) in their child’s frequency of negative feelings, such as blue mood, despair, or depression.
Pre- and Post-Test Scores on Participant Well-Being.
Note. Ratings based on a five-point Likert-type scale. Effect size = Hedges’ g. PR = parent report.
In comparison of pre-test to post-test means, mentors noted no change (g = .00) in their quality of life, but a small increase (g = .37) in life satisfaction. Regarding enjoyment in life, mentors reported a small to moderate increase (g = .47). Moreover, mentors indicated a small increase (g = .18) in the extent to which they feel pride in their autistic identity. Mentors also noted a small increase (g = .19) in their self-satisfaction and a moderate increase (g = .52) regarding their relationship satisfaction. Finally, mentors indicated a very large decrease (g = −1.35) in their frequency of negative feelings, such as blue mood, despair, or depression.
Self-concept
Mentees and mentors reported on their self-concept at pre- and post-test (see Table 2). When asked about their overall self-concept (e.g., behavioral adjustment, intellectual and school status, physical appearance and attributes, freedom from anxiety, popularity, happiness, and satisfaction), mentees indicated a moderate increase (g = .64) in overall self-concept following AMP. More specifically, mentees indicated a small increase in behavioral adjustment (g = .33), a very small decrease in intellectual and school status (g = −.10), a very small increase in physical appearance (g = .13), a large increase in freedom from anxiety (g = .87), a large increase in popularity (g = .87), and a moderate increase in happiness (g = .70) following AMP (Piers-Harris II).
Pre- and Post-Test Scores on Participant Self-Concept.
Note. Pre- and post-test means reported as standard scores. Effect size = Hedges’ g.
When asked about their overall self-concept (e.g., behavioral adjustment, intellectual and school status, physical appearance and attributes, freedom from anxiety, popularity, happiness, and satisfaction), mentors indicated a very small decrease (g = −.06) in overall self-concept following AMP. More specifically, mentors indicated a very small decrease in behavioral adjustment (g = −.02), a small increase in intellectual and school status (g = .30), a small decrease in physical appearance (g = −.33), a very small increase in freedom from anxiety (g = .07), a very small decrease in popularity (g = −.07), and a small decrease in happiness (g = −.20) following AMP (Piers-Harris II).
Social-emotional and behavioral outcomes
Mentees, parents of mentees, and mentors were asked to report on social-emotional and behavioral outcomes (see Table 3). Mentees reported a small to moderate decrease in their internalizing problems (g = −.38); however, parents of mentees reported little to no changes in their child’s internalizing problems (g = −.02). Mentees also indicated a large decrease in their externalizing problems (g = −.87), but parents of mentees indicated a small increase in their child’s externalizing problems (g = .23). In the area of social skills, mentees reported a small increase (g = .16), and parents of mentees indicated a very small increase (g = .06) in their child’s social competence following AMP. Furthermore, in the area of problem behaviors, mentees noted a very large decrease (g = −3.62); however, parents of mentees indicated a small to moderate decrease (g = −.32) in their child’s problem behaviors following AMP.
Pre- and Post-Test Scores on Participant Socio-Emotional and Behavioral Outcomes.
Note. Pre- and post-test means reported as standard scores. Effect size = Hedges’ g. PR = parent report.
Mentors reported a small decrease in their internalizing problems (g = −.27) and a small to moderate decrease in their externalizing problems (g = −.43). In the area of social skills, mentors reported a small decrease (g = −.29) in their social competence following AMP. Furthermore, in the area of problem behaviors, mentors reported a moderate to large decrease (g = −.76).
Discussion
Results of the current study show that one-to-one mentoring by adults with ASD in an afterschool setting is a promising intervention strategy for youth with ASD. This study is the first to examine mentoring of youth with ASD by adults with ASD. In this program, adults with ASD were invited to serve as role models and leaders in their community, share their unique insights and experiences, and empower youth with ASD to be successful and thrive. Limited previous literature showed that mentoring young people with ASD is a positive, prosocial service (Curtin etal., 2016; Lucas & James, 2018; Ncube etal., 2019), but no known studies utilized adult mentors with ASD. Pairing youth and adults with ASD in a relational intervention simultaneously posed opportunity and potential challenge. Somewhat surprisingly, few challenges were noted by program participants or parents. In fact, mentees and mentors were largely satisfied with their mentoring relationships, felt their participation in the program was meaningful, and reported learning valuable skills and information during their mentoring sessions.
With some exceptions, initial outcome measures indicated that AMP provided mutual benefit to mentees and mentors participating in the program. Program participants experienced improvements in quality of life, social skills, behavioral regulation, and mental health. The strongest impacts were seen regarding improved interpersonal relationships, reduced problem behaviors, and improved internalizing symptoms for both mentors and mentees, with significant improvements in mental health seen for mentors. Small to moderate improvements in self-satisfaction and life enjoyment were reported for mentors and mentees. Mentees reported significant improvements in self-concept, which were not seen for mentors. Although mentees reported improvements in social skills, mentors reported a small decrease in social competence on the SSIS. Results suggest mentees may experience more benefits from the program compared to mentors; however, additional evaluation is needed to further explore differences in mentor and mentee outcomes and potential factors affecting social and emotional outcomes across a longer intervention period. Inconsistent or minimal changes were seen in regard to quality of life and autistic identity. Notably, mentees and parents of mentees tended to report different levels of change, with mentees generally reporting stronger outcomes than their parents. Parents also reported a small increase in externalizing behaviors on the ASEBA while reporting a decrease in problem behaviors on the SSIS. Additional evaluation with a larger sample is needed to examine whether these differences may be related to limited parental involvement in the intervention itself or other factors related to self-report questionnaires with this population.
The current study has several strengths. AMP is an example of a successful university–community partnership that results in careful planning, development, implementation, and evaluation of programming within a local community. Our team relied heavily on a strength-based model that emphasized community input rather than the traditional “expert” model. As mentioned above, AMP is an innovative mentoring model developed for youth with ASD by adults with ASD and other stakeholders in the ASD community, including parents of children with ASD, special educators, and researchers and clinicians who specialize in mentoring and ASD. Individuals with ASD and their families rarely have direct involvement in the planning and development of interventions meant to support their needs; thus, AMP has the potential to better serve the unique needs of individuals with ASD compared with programming developed without input from individuals on the spectrum.
AMP was launched in a deliberative manner with a focus on serving individuals we believed would be most likely to be affected by participation in AMP. Per consultation with the steering committee and community stakeholders, AMP was initially implemented with adolescents who were early in their high school careers. An additional hypothesis suggests adolescents may be adjusting to the high school setting during the first couple of years, but are early enough in their high school careers to implement some of the guidance received from their mentors before experiencing the next transition of exiting high school and entering young adulthood. Mentors also have the potential to provide support and guidance that will ease the stress and challenges related to this transition period.
The study also has limitations, including a small sample consisting of all male mentees from one high school. Although the mentee participants were more racially and ethnically diverse, all mentors identified as White. Due to the exploratory nature of this study, initial mentee participants were nominated by special educators who knew the individuals and anticipated they would be motivated to participate in the program and able to successfully attend sessions. Mentee participants were determined to meet eligibility for the study through record review and previous clinical and/or educational documentation of ASD and cognitive and language skills; therefore, additional information describing the characterization of mentees was not available across mentee participants. Although the study team was able to evaluate initial outcomes from these 12 sessions, we do not know whether AMP would produce similar results over a longer period of time. Given the selection of a small number of participants and the short duration of the study, results are not necessarily generalizable to other individuals participating in similar programming. To remedy some of these limitations in future research, studies should recruit and assess outcomes within a larger sample and compare short-term versus long-term programming. Future research should consider the implications of near-peer mentoring (e.g., mentoring of an adolescent by an adult only a few years older) versus mentoring from a much older adult. Finally, future research should also prospectively assign eligible mentee participants to intervention and control conditions.
The active and regular engagement of the AMP steering committee is a true strength of AMP and the development of this project. However, we acknowledge the valuable perspectives and input from this steering committee may not represent the viewpoints of other stakeholders outside of this group. Our team sought additional and alternative input via focus groups before and after the initial implementation of AMP. Local steering committees and additional input should be sought by others who are interested in implementing similar mentoring programs to best represent and serve the individuals participating in programming.
Implications
Results of the current study indicate that the AMP shows promise as a community-based intervention to promote positive outcomes for youth and adults with ASD. These initial results have exciting implications for policy and ASD programming. AMP is an example of a partnership formed to meet a specific community need to be identified by the mother of children with ASD. This collaboration grew to include experts from the local community that would help to bring the concept of a specialized mentoring program for ASD to life. The expertise and dedication within this partnership allowed for planning, implementation, and evaluation of AMP over a relatively short period of time. The success of this partnership indicates community–university partnerships are an effective way to combine knowledge and resources to have a direct impact on the community, and these partnerships should be encouraged and supported whenever possible.
AMP has the potential to increase access to programming and build capacity within the community. The need for innovative community supports for individuals with ASD is clear given the shortage of professionals serving individuals with ASD, the lack of provider confidence and training in serving individuals with ASD, and the limited number of services available for adolescents and young adults (Chiri & Warfield, 2012; Mazurek etal., 2019). This results in long waitlists in which families regularly wait 6 or more months to receive clinical services and interventions. Community-based interventions, in contrast, have the potential to reduce waiting times for services by more quickly serving a larger number of families within their own communities and then serving as a supplement to traditional outpatient treatment.
Moreover, AMP can build community capacity by implementing the program within existing school systems. By embedding AMP within afterschool programming, program staff and resources are more readily available. For example, one-third of program staff were special education teachers from the school in which AMP was being implemented. The presence of familiar and supportive staff allowed for quicker rapport building with mentors and mentees and increased mentees’ comfort and participation during sessions. Additional AMP staff were staff from the transition program that the majority of mentors attended, which similarly increased the comfort level of mentors. School administration also offered space for program activities and interpreters for data collection with families who did not speak English. This model also enabled AMP to be delivered at no cost to families.
Capacity is also built by recruiting mentors with ASD from within the local community. In our initial acceptability study and in the current study, adults with ASD have repeatedly mentioned their eagerness to give back to their local ASD community by helping adolescents with ASD navigate some of the challenges they experienced themselves. As reported in the results of this study, participation in AMP has given many mentors a newfound sense of purpose, confidence, and leadership skills, all of which have the potential to increase mentors’ meaningful participation and impact on their local community.
Conclusion
Results of this study and the initial acceptability study suggest that one-to-one mentoring for adolescents with ASD by adults with ASD is an acceptable intervention strategy and has the potential to promote positive outcomes for program participants with regard to well-being, self-concept, and social-emotional and behavioral outcomes. Researchers are continuing to evaluate the outcomes of AMP over longer periods of time with a larger sample size and control condition. Results of program evaluation will inform the continued refinement and expansion of the program.
Footnotes
Acknowledgements
We thank each of our participants and valued community partners who made this study possible.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This study was generously supported by the Clinical and Translational Science Institute at University of Minnesota, the Institute for Translational Research in Children’s Mental Health at the University of Minnesota, and the College of Education and Human Development at the University of Minnesota. This work was also supported by the USDA National Institute of Food and Agriculture, Hatch project MIN-52-109. These funding sources were not involved in study design, the collection, analysis or interpretation of data, the writing of this manuscript, or the decision to submit the article for publication. The content is solely the responsibility of the authors and does not necessarily represent their official views.