Early Autism Intervention Components Deliverable by Non-specialists in Low- and Middle-Income Countries: A Scoping Review.

Introduction: The past decade has seen key advances in early intervention for autistic children in high-income countries, with most evidence based on specialist delivery of interventions. The care gap seen in low- and middle-income countries (LMIC) remains close to 100%. A key challenge in addressing this care gap concerns the paucity of specialists available to deliver services. Task-sharing provides an important potential solution; there is a need to identify interventions that are suitable for scaled-up delivery through task-sharing in low-resourced settings. We aimed to conduct a scoping review to identify studies which reported autism intervention delivered by non-specialists within LMIC and, using established frameworks, specify intervention components with evidence of successful non-specialist delivery. Methods: A scoping literature search, conducted within four databases, generated 2,535 articles. Duplicates were removed, followed by screening of titles and abstracts, with 10% double-rated for reliability. 50 full text articles were then screened independently by two raters. Articles were included if studies: (a) were conducted in LMIC; (b) included samples of autistic children (age < 10); (c) evaluated psycho-social interventions delivered by non-specialists; (d) reported child outcomes; and (e) were peer-reviewed full-texts in English. Two established frameworks - @Practicewise and NDBI-Fi framework - were then used to ascertain the commonly delivered components of these interventions. Results: Two studies met the inclusion criteria. Both studies evaluated parent-mediated interventions delivered by non-specialists in South Asia. Through the two frameworks, we identified elements and techniques that had been delivered successfully by non-specialists. Conclusion: There is evidence from two acceptability and feasibility trials that non-specialists can be trained to deliver some intervention elements and techniques within parent-mediated interventions, with good fidelity and acceptability and evidence of effectiveness. The review points up the lack of a widespread evidence base in this area and need for further research in low resourced settings, including well-powered trials and mechanistic analyses to identify active ingredients. A focus on the pre-requisites for non-specialist delivery is critical to reduce inequity and provide universal health coverage within resource-constrained health systems.