The study team conducted a single-site, two-arm randomized controlled trial to evaluate feasibility of the Community Autism Peer Support (CAPS) program. Forty autistic adults aged 18 to 30 were randomized evenly to CAPS (n = 20) or services-as-usual alone (n = 20). Assessments occurred at baseline, 3 months, and 12 months. The trial registration number is NCT05702658. Ethical approvals were granted by Drexel University IRB and City of Philadelphia IRB. The funder was the National Institute of Mental Health (R34MH130830). Data are slated for deposit in the National Institute of Mental Health Data Archive repository.
Recruitment produced a cohort the authors described as demographically and diagnostically diverse. The intervention targeted young autistic adults pursuing self-determined goals related to independent living and community participation. Intervention delivery relied on trained autistic peer specialists who provided individualized support; the source did not report specific training curricula, session frequency, or dosing parameters in the preprint, and those implementation details were not provided.
Most participants assigned to CAPS remained engaged for extended periods, the authors observed. Engagement was sustained across the 12-month follow-up window for a majority of the CAPS group. That retention signal addresses a common practical barrier for community interventions aimed at young autistic adults: sustaining participation long enough to influence real-world outcomes. The manuscript did not provide exact numbers for retention at each time point in the preprint, so precise attrition rates and per-protocol counts are not reported here.
Although the trial was explicitly described as a feasibility study and therefore underpowered for confirmatory hypothesis testing, investigators reported a consistent pattern of results that favored the CAPS condition across all measured outcomes. The largest positive signals were observed for overall and social quality of life. Secondary favorable changes were seen for internalized stigma, unmet needs, social functioning, loneliness, and positive autistic identity. The authors noted these domain-level signals without reporting formal statistical significance thresholds or p-values in the preprint; specific effect sizes and inferential statistics were not provided in the source document.
This pattern—favorable change across multiple psychosocial measures—was interpreted by the study team as supportive evidence for the model’s promise. The consistency of directionality across outcomes, rather than any single definitive outcome result, formed the core efficacy signal in this feasibility trial.
The authors framed CAPS as a community-based model in which trained autistic peer specialists deliver individualized support for self-directed goals. The program attracted young autistic adults with varied backgrounds, and the peer-delivered format appeared feasible to implement at a single site. The preprint confirmed IRB approvals and participant consent procedures were completed and documented. Competing interests were declared none.
Full datasets generated or analyzed during the study will be made available in the National Institute of Mental Health Data Archive repository, according to the preprint. The article was posted as a medRxiv preprint on October 20, 2026. The preprint does not include complete implementation manuals, standardized training content for peer specialists, or step-by-step fidelity measures; those materials were not reported in the source document.
The authors explicitly stated that the study was underpowered to detect statistical differences because feasibility, implementation, and acceptability were the primary aims. That limitation constrains interpretation of the favorable outcome signals; the observed patterns require confirmation in larger, adequately powered trials. The trial was single-site, which limits generalizability across settings and regions. The preprint did not provide detailed demographic breakdowns, exact retention counts, or numerical effect estimates in the posted manuscript, so readers cannot independently evaluate magnitude of change from the preprint alone.
Investigators recommended continued development of CAPS as a person-centered approach to addressing outcomes important to autistic adults. The preprint did not specify concrete next regulatory milestones, plans for multicenter scaling, or timelines for a definitive efficacy trial. Those operational steps remain unanswered in the source and will need to be addressed in subsequent reports.
The study provides feasibility evidence and consistent directional outcome signals favoring an autistic-delivered peer specialist model for community living and participation. However, replication in larger, multi-site trials with prespecified outcome analyses, detailed reporting of training and fidelity, and transparent effect estimates will be necessary before efficacy claims can be confirmed.