Differences in patterns of social communication and interaction and associated challenges with complying with mainstream scholastic demands and expectations are hallmarks of autism spectrum conditions (henceforth autism) (ICD-11; World Health Organization, 2019/2021). Understanding autism as a neurodivergent condition and the extreme end of a trait (Constantino and Charman, 2016; Pellicano et al., 2018) has caused many countries to mandate inclusive education for autistic students (e.g., Roleska et al., 2018; van Kessel et al., 2019). Yet, autistic learners are at higher risk of social exclusion and adverse outcomes in school settings. Notably, students with social challenges specifically rarely reach their full academic potential (Domitrovich et al., 2017), which may contribute to negative long-term vocational outcomes (Clarke et al., 2021). School-aged neurodivergent learners are also more likely to experience loneliness and victimization (Locke et al., 2010; Zeedyk et al., 2014), which may contribute to mental health issues and lower quality of life (Demir et al., 2012) as well as school absenteeism (Anderson, 2020). Successful inclusive school settings require accommodation of learning environments for neurodivergent students (Kim et al., 2018; Pellicano et al., 2018), yet research indicates that regular school staff are not well prepared to practice inclusion of neurodivergent students (Bölte et al., 2021b; Leifler et al., 2022a). In addition, autistic learners may not incidentally acquire or develop socio-communicative understanding, skills, or strategies over time by mere proximity to non-autistic peers. Given this mismatch, environmentally and individually, targeted support strategies are crucial to combat risks of adverse outcomes for these students.

Social skills training aims to expand the socio-communicative understanding and behavioral repertoire through a variety of approaches and strategies, including behavioral skills training, social stories, antecedent interventions and reinforcement, peer mediation, and video modeling (Radley et al., 2020). It may positively affect social behaviors in adolescent autistic learners with average cognitive abilities (McKeithan and Sabornie, 2020), autistic learners with co-occurring complex communication needs and intellectual disability (Babb et al., 2021), and autistic learners in inclusive school settings (Dean and Chang, 2021). Group-based social skills training formats may offer added benefits including real-time practice of taught skills with peers in a structured environment (Tachibana et al., 2018) and can serve more students with better cost-effectiveness (Myhr and Payne, 2006; Tucker and Oei, 2007) for some autistic learners.

Though scarce in educational settings, in clinical settings, manualized, group-based social skills training (c-SSGT) is one of the most widely applied and evaluated psychosocial interventions for school-aged verbal autistic children and adolescents in the average range of intellectual functioning (Hall et al., 2018). Here, a comprehensive set of evidence-based strategies for autistic youth (Hume et al., 2021), such as modeling, behavioral practice, and direct instruction, are often integral parts of a peer-group setting with one or several group leaders. In a meta-analysis of 18 randomized control trials comprising 745 participants, c-SSGT showed an overall positive medium-sized effect (g = 0.51) on social competence outcomes (Gates et al., 2017). The hitherto largest randomized controlled trial in autism compared c-SSGT KONTAKT™ to waitlisted controls in regular clinical settings (Choque Olsson et al., 2017). Manualized, 12 (short) or 24 (long) weekly sessions include group discussions and practical social activities—implemented through learning and cognitive behavioral principles—with children (8–12 years) or adolescents (13–18 years). Socio-communicative competence improved primarily for girls (d = 0.33) and adolescents (d = 0.40) following short KONTAKT™ (Choque Olsson et al., 2017) while large effects were found for children and adolescents (d = 0.82) following long KONTAKT™ (Jonsson et al., 2019).

Reviews of SSGT stress the importance of considering SSGT in naturalistic school environments (Gates et al., 2017). Benefits may include added flexibility by which support may be accessed as well as possible spin-off effects in terms of the qualifications of the school-staff (Leifler et al., 2022b). Moreover, while c-SSGT manuals typically include activities aiming to promote generalizability, e.g., homework assignments, parent participation, and social excursions outside the clinic (Choque Olsson et al., 2017; Jonsson et al., 2019), the main intervention setting is detached from the regular, everyday environment. School-based SSGT (s-SSGT) may further facilitate the generalizability of training outcomes beyond the intervention setting by reducing demands on skills transfer, which may be particularly challenging for autistic learners (Neely et al., 2016). Indeed, adapting c-SSGT to s-SSGT is feasible and may improve teacher-rated socio-communicative function, family frequency ratings of hosted and invited get-togethers with peers, and adolescent-rated social competence (Laugeson et al., 2014; Dean et al., 2020).

Notably, however, qualitative studies indicate that even participants with little gains in terms of standardized socio-communicative measures experience socially relevant and meaningful outcomes (Choque Olsson et al., 2016; Afsharnejad et al., 2022). Few standardized scales are designed to assess defined natural contexts where social skills are performed (i.e., generalization). Moreover, by definition, standardized measures are not able to consider individual priorities and personally important areas of development. Goal Attainment Scaling (GAS; Kiresuk et al., 1994) is a goal-setting procedure that has been applied in a variety of care and support settings. As an added component in SKOLKONTAKT™, it has the potential to address the individual variability and priorities of autistic learners (Bishop et al., 2016) as well as define generalizability to important naturalistic contexts while detailing outcomes at the group level. GAS has been shown to be a feasible and safe measure in school-aged autistic learners (Ruble et al., 2012, 2022) and as a goal-setting procedure, and it may motivate participants to engage with the intervention (Locke and Latham, 2019).

Given the potential of SSGT in a school setting, our research center has adapted c-SSGT KONTAKT™ to ‘SKOLKONTAKT™’ (Eng. ‘SCHOOL-KONTAKT): an s-SSGT aimed at students with formally documented social communication and interaction challenges at school and following several principles. First, while studies investigating the effects of social skills training at school have often applied research staff as intervention providers (Dean and Chang, 2021), using qualified school staff may be crucial in terms of contextualizing the training. Staff-led training is consistent with regular schooling—for the student as well as for the sustainability of the intervention (Watkins et al., 2017). Second, intervention development was conducted with the aim to enhance stakeholder acceptability of intervention procedures, i.e., the social validity, to reduce barriers for effective implementation (Fleury et al., 2014). Findings from our previous qualitative multi-perspective interview study have informed the social validity of SKOLKONTAKT™ compared with other socially engaging group activities at school (Leifler et al., 2022b). Given adequate support and resource planning, school staff considered SKOLKONTAKT™ as a well-placed and motivational intervention for their school. Participating youths, school leaders, and SKOLKONTAKT™ facilitators experienced an improved social climate and culture at school as well as social behavior development among students, such as knowing how and daring to engage in small talk with peers, making one’s voice heard, and managing social situations. Little to no such experiences were reported for other socially engaging group activities.

The present study aims to add quantitative data to these findings to investigate acceptability, feasibility, and preliminary efficacy outcomes of manualized SKOLKONTAKT™ in comparison to non-specific effects of engaging in social group activities with peers for autistic learners and students with formally documented socio-communicative needs. Moreover, understanding of the extent of skill transfer beyond the intervention (i.e., generalization) was examined by use of goal attainment outcomes.

The aim of the present study was to investigate the acceptability, feasibility, safety, and preliminary efficacy of SSGT SKOLKONTAKT™ for autistic learners and students with formally documented social communication challenges. Despite the COVID-19-related challenges in school and for delivery of the intervention, the majority of SKOLKONTAKT™ participants completed the training. Careful consideration of unwanted effects by group and individual monitoring and measures suggests that the training is a safe with limited or mild/manageable side effects within the school setting. Few studies on SSGT have previously collected data on side effects. However, systematic understanding of side effects may not only optimize intervention safety but also intervention efficacy by the ability to intervene at appropriate times (Afsharnejad et al., 2021). Most dropouts in SKOLKONTAKT™ occurred early in the intervention, which underlines the importance of appropriate a priori training inclusion procedures to evaluate among other things students’ motivation and mental health, as indicated by drop-out rationales in the current study. While more participants dropped out from SKOLKONTAKT™ than controls, reasons were comparable and not necessarily related to intervention. In addition, school absence was higher in the control groups than in SKOLKONTAKT™, and school attendance and SKOLKONTAKT™ attendance were positively associated, while school attendance and control intervention were not. SKOLKONTAKT™ facilitated aspects of social communication skills, improved life quality from the primary perspective of students—notably including experiences of peer relationships and social support and physical well-being—in comparison to other social group-based activities with peers.

Moreover, personal social and school-related goals were more likely attained following SKOLKONTAKT™. The present study also clarified relevant themes of goal-formulation using GAS, reflecting students desire to connect with peers, feel included, and participate in classroom education as well as find strategies to manage organizational and routine demands integrated into on-site mainstream schooling—findings supported by qualitative student report elsewhere (Goodall, 2018). Importantly, students’ goals often detail the collaboration and relations between themselves and school staff or other students, such as developing strategies to ask for help in the classroom or to initiate desired peer interactions. This emphasizes the interaction between school accommodations and student functioning in how well the student is adjusted at school (Bölte et al., 2021a). Goal formulations according to GAS may contribute to a sense of self-empowerment in relation to social, desirable goals by operationalizing self-efficacious strategies to attain them. Indeed, qualitative student reports suggest that goal-formulation according to GAS was experienced as one of the most meaningful components of SKOLKONTAKT™ (Leifler et al., 2022b). Indeed, this quantitative study of SKOLKONTAKT™ aligns with our earlier multi-perspective qualitative interview study of SKOLKONTAKT™, which demonstrated that it is perceived being a socially valid intervention strategy in mainstream school-settings (Leifler et al., 2022b). The present study further detailed the development of listed prepared goals and corresponding Goal Attainment Scales for use in SKOLKONTAKT™, i.e., the Goal catalogue. Findings suggest using the Goal catalogue might increase not only the quality of goal-formulations but also reduce time consumption of goal-setting procedures. Demonstrations of acceptability and efficiency in the present study suggest that this tool and these procedures are a feasible further adaptation of KONTAKT™ to SKOLKONTAKT™ and will therefore be retained as part of the SKOLKONTAKT™ manual.

Compared to clinical training with KONTAKT™, findings of this study indicate selected and specific effects on social skills questionnaire items rather than overall improvement of social communication skills. A major difference between KONTAKT™ and SKOLKONTAKT™ is that school staff are in charge of the training, not experienced clinicians, which may impact of social communication outcomes, especially if SKOLKONTAKT™ is administered for the first time. Indeed, adequate training, supervision, and resources (i.e., funding, staff, and preparation) is crucial for successful implementation as evidenced by stakeholder interviews (Leifler et al., 2022b). Another change for students that may have contributed to the found patterns of results in compared with clinical delivery of KONTAKT™, was moving the intervention to an online format following pandemic-related social distancing protocols at school. Like KONTAKT™, SKOLKONTAKT™ activities have two main foci. One focus is on increasing social awareness and problem-solving skills (e.g., by activities like group discussions and homework on functional analysis on social situations), and the other is on direct practice of social skills (e.g., recognition and expression of emotions and other group activities). Perhaps intervention components relating to becoming more socially aware and problem-solving in SKOLKONTAKT™ were less affected by the change to online format than aspects in which participants practice social skills more directly, e.g., using others social behaviors to modulate one’s own in conversations when practicing conversation in one group activity. If so, one might not expect clear effects on general social skills but on abilities to understand and problem-solve unique social concerns, as better reflected in our secondary outcomes goal attainment and quality of life. Indeed, superior effects on quality of life and the attainment of social meaningful goals indicate that SKOLKONTAKT™ has added value, even when contrasted with other social activities of comparable lengths and group size. Critiques of social skills training curricula suggest that such programs may contribute to a sense of inauthenticity by training autistic people in “masking” behaviors (Bottema-Beutel et al., 2018). While the authors of the present study encourage increased attention to further exploring active, effective, and safe components of social skills training curricula, the current findings might also indicate that these fears might be unfounded following SKOLKONTAKT™ intervention. Rather than generic social skills behavior effects, the main effects were on well-being and personal meaningful social outcomes. Indeed, benefits in terms of personalized, socio-communicative goals and reduced anxiety in comparison to active controls have been previously reported for the English adaptation of c-SSGT KONTAKT™, the model of SKOLKONTAKT™, for Australian adolescents (Afsharnejad et al., 2021).

Limitations of the study include the small sample size. Additionally, the exploratory and uncorrected nature of parts of the data analysis increase the likelihood of false positive findings, particularly in the analysis of the primary (although preliminary) outcome measure. While overall effects were statistically insignificant, we found significant improvements for social skills on the item level. While interpretation of these effects should be made with all necessary caution it is noteworthy that there were some similarities between responders, including blinded teacher ratings. This might lend support to true, common effects of SKOLKONTAKT™ on single aspects of socio-communicative skills. Moreover, SKOLKONTAKT™ may not be readily available in international contexts as is. The needs-based, rather than diagnosis-based, system of providing educational support in Swedish schools requires careful consideration of socially valid inclusion criteria of students in SKOLKONTAKT™. Among other cultural factors to be considered, inclusion criteria might need further specification in any international use of SKOLKONTAKT™. While scarce in SSGT literature generally (Gates et al., 2017), moderator analysis participant characteristics predicting outcomes may further elucidate inclusion criteria and guide facilitators of the program.

In conclusion, SKOLKONTAKT™ was feasible and safe in an inclusive educational setting in Sweden. We identified specific associations with goal-attainment and student-perceived quality of life in comparison to social activity controls. While the present study could not provide support for the main effects on primary outcome measures, specific aspects of common socio-communicative skills may improve following SKOLKONTAKT™. Future large-scale studies may investigate the adequate primary outcomes of SKOLKONTAKT™, focusing on the social problem-solving and self-empowerment of autistic and other learners requiring social support at school.

The datasets presented in this article are not readily available because Raw data cannot be made immediately publicly available, as this was not specified in ethical approval, and parts of it (e.g., Goal Attainment Scale), are not completely anonymous. Thus, raw data can currently only be provided by the authors upon legitimate request and additional procedures (e.g., ethical amendment; additional anonymization of verbal data). The authors can provide without delay additional details on the data in processed form. Requests to access the datasets should be directed to anna.fridell@ki.se.

The studies involving human participants were reviewed and approved by Swedish Ethical Review Board. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

AF and SB made contributions toward the scope, structure, methods, and key findings of this research manuscript and contributed to the formal descriptive analysis, tables, methodology, writing—original draft preparation, and writing—review and editing. AF, CC, AB, and SB made contributions of the content and structure of the social skills group training and contributed to the conceptualization. AF, CC, and AB were active in the pilot and randomized controlled trial. AF and CC contributed to the data curation. CC contributed to the randomization. AF contributed to the blinding and goal assessment. SB contributed to the supervision and formal inferential analysis. CC, AF, and SB contributed to the validation. All authors contributed to the article and approved the submitted version.

This study was supported by Marcus and Amalia Wallenbergs Minnesfond.

SB receives royalties from Hogrefe publishers for KONTAKT™ materials.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.