Development and Evaluation of a Parent-led Intervention Module for Early Autism Spectrum Disorder (PRIME-ASD) in Odisha, India: Protocol for a Multiphase Mixed-methods Study Including an Open-label Randomised Controlled Trial

Background. Parent-mediated interventions improve parent-child interaction and show smaller but consistent benefits for child social communication. Existing Indian adaptations were developed in other linguistic and service contexts and have not been co-designed with Odia-speaking families or aligned with the District Early Intervention Centres (DEIC) through which Odisha delivers early developmental services.

Development. PRIME-ASD was developed in earlier phases of a multiphase mixed-methods study: a systematic review and meta-analysis of randomised trials of parent-mediated interventions (registered with PROSPERO); in-depth interviews and focus groups with caregivers, providers and administrators; and a modified Delphi process (8-12 experts; consensus ≥80%) with content validation (I-CVI ≥0.78; S-CVI/Ave ≥0.80), face validation with 6-8 parents and field testing with 3-5 parent-child dyads. This record covers only the randomised controlled trial.

Design. Two-arm, parallel-group, open-label, assessor-blinded, superiority randomised controlled trial with 1:1 allocation and an embedded qualitative study, reported according to SPIRIT 2013 and CONSORT. Allocation uses a computer-generated sequence of randomly permuted blocks of variable size [2, 4 and 6], prepared by an independent statistician and concealed in sequentially numbered, opaque, sealed envelopes opened after the baseline assessment.

Intervention. PRIME-ASD comprises 12 weekly individual coaching sessions of about [60] minutes (didactic coaching, demonstration, coached caregiver-child practice with feedback and home-goal setting), weekly telephone follow-up of about [15-20] minutes and daily home practice of about [20-30] minutes across six domains aligned with the ISAA. Sessions are delivered in person at AIIMS Bhubaneswar, with synchronous video sessions when travel is not possible. Fidelity is monitored with attendance logs, home-practice checklists, an interventionist fidelity checklist and supervisor rating of [20%] of audio-recorded sessions.

Comparator. Treatment as usual (TAU): routine outpatient psychiatric review, psychoeducation and counselling, medication where clinically indicated and referral to available therapy services. TAU is not restricted in either arm, and co-interventions are recorded at each assessment.

Sample size. Assuming a standardised mean difference of 0.45 in ISAA score, two-sided α = 0.05 and 80% power, 79 children per arm are required; allowing for 15% attrition by 24 weeks, 94 per arm (188 in total) will be recruited.

Analysis. Intention to treat. The 12-week ISAA total score will be compared with analysis of covariance adjusted for baseline score [and any stratification factors]. Secondary outcomes will be analysed with linear mixed-effects models (arm, time, arm-by-time and baseline score) and logistic regression for response, without adjustment for multiplicity. Missing data will be handled by multiple imputation by chained equations, with complete-case, per-protocol and tipping-point sensitivity analyses. No interim efficacy analysis is planned.

Embedded qualitative study. After the 12-week assessment, approximately [15-20] intervention-arm caregivers, sampled on adherence and ISAA change, and the interventionists will be interviewed; findings will be merged with trial results in joint displays.