Accepted for/Published in: JMIR Formative Research
Date Submitted: Sep 30, 2025
Date Accepted: Feb 5, 2026
A GPT-powered chatbot-based positive psychology intervention to promote well-being among parents of children with autism spectrum disorder: A pilot study
ABSTRACT
Background:
Parents of autistic children frequently experience elevated stress, depressive symptoms, and reduced well-being. Positive Psychological Interventions (PPIs) can strengthen resilience, and chatbots offer a scalable channel to deliver such skills. However, there is limited evidence on the evaluation of large-language-model (LLM)–guided, PPI-based chatbots for this population.
Objective:
The study evaluated the feasibility and acceptability of a GPT-powered chatbot (“Allie”). It was designed to deliver culturally adapted PPIs to parents of autistic children, and to explore preliminary effects on well-being, depression, stress, and health-related quality of life.
Methods:
We conducted a single-arm, mixed-method pilot with 19 parents with autistic children. They engaged with Allie for two weeks to complete eight structured PPI exercises. Primary outcomes were feasibility (completion, ease of use, practicality) and acceptability (multi-dimension user ratings). Secondary outcomes were the World Health Organization–Five Well-Being Index (WHO-5), the Patient Health Questionnaire-9 (PHQ-9), the Perceived Stress Scale-10 (PSS-10), and the Short Form-12 Health Survey version 2 (SF-12v2) Physical (PCS) and Mental Component Summary (MCS) scores. Outcomes were analyzed using paired t-tests or Wilcoxon signed-rank tests. Optional post-intervention interviews were analyzed using reflexive thematic analysis.
Results:
Seventeen of 19 participants (89.5%) completed all exercises which indicated high procedural feasibility. There were also high ratings in ease of use and practicality (means 4.47/5 and 4.32/5, respectively). Acceptability was favorable (overall satisfaction mean 5.68/7; prompt response time 6.37/7). WHO-5 scores improved significantly from 32.84 to 46.11 (t(18)=2.48; P=.02; Cohen d=0.52). Changes in PHQ-9 (z=−0.49; P=.63; r=.11), PSS-10 (t(18)=−0.82; P=.43; Cohen d=0.12), and SF-12v2 PCS (t(18)=−0.94; P=.36; Cohen d=0.18) and MCS (t(18)=−0.89; P=.39; Cohen d=0.17) were not significant. Qualitative feedback (n=14) described benefits aligned with PPI mechanisms such as greater self-reflection, a more positive orientation, perspective taking, emotional support, and coping skills. However, participants also suggested refinements like more natural conversation (colloquial Cantonese), shorter/ less repetitive outputs, user-chosen sequencing with reminders and progress tracking, multimodal features, and ASD-specific resources.
Conclusions:
This pilot study showed the feasibility, acceptability and improved well-being of the PPI-based, GPT-powered Allie among parents of autistic children, despite no significant short-term change in other outcomes. The findings provided insights into design priorities including personalization, conversational naturalness, multimodal content and ASD-specific guidance. Larger, controlled trials with longer exposure and more diverse samples are needed to establish efficacy and durability Clinical Trial: ClinicalTrials.gov NCT 06438120; https://clinicaltrials.gov/study/NCT06438120
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.