Oral language interventions can improve language outcomes in children with neurodevelopmental disorders: A systematic review and meta‐analysis

Background Young people who fail to develop language as expected face significant challenges in all aspects of life. Unfortunately, language disorders are common, either as a distinct condition (e.g., Developmental Language Disorder) or as a part of another neurodevelopmental condition (e.g., autism). Finding ways to attenuate language problems through intervention has the potential to yield great benefits not only for the individual but also for society as a whole. Objectives This meta‐analytic review examined the effect of oral language interventions for children with neurodevelopmental disorders. Search Methods The last electronic search was conducted in April 2022. Selection Criteria Intervention studies had to target language skills for children from 2 to 18 years of age with Developmental Language Disorder, autism, intellectual disability, Down syndrome, Fragile X syndrome, and Williams syndrome in randomised controlled trials or quasi‐experimental designs. Control groups had to include business‐as‐usual, waiting list, passive or active conditions. However, we excluded studies in which the active control group received a different type, delivery, or dosage of another language intervention. Eligible interventions implemented explicit and structured activities (i.e., explicit instruction of vocabulary, narrative structure or grammatical rules) and/or implicit and broad activities (i.e., shared book reading, general language stimulation). The intervention studies had to assess language skills in receptive and/or expressive modalities. Data Collection and Analysis The search provided 8195 records after deduplication. Records were screened by title and abstract, leading to full‐text examinations of 448 records. We performed Correlated and Hierarchical Effects models and ran a retrospective power analysis via simulation. Publication bias was assessed via p‐curve and precision‐effect estimate. Main Results We examined 38 studies, with 46 group comparisons and 108 effects comparing pre‐/post‐tests and eight studies, with 12 group comparisons and 21 effects at follow‐up. The results showed a mean effect size of d = 0.27 at the post‐test and d = 0.18 at follow‐up. However, there was evidence of publication bias and overestimation of the mean effects. Effects from the meta‐analysis were significantly related to these elements: (1) receptive vocabulary and omnibus receptive measures showed smaller effect sizes relative to expressive vocabulary, grammar, expressive and receptive discourse, and omnibus expressive tests; and (2) the length of the intervention, where longer sessions conducted over a longer period of time were more beneficial than brief sessions and short‐term interventions. Neither moderators concerning participants’ characteristics (children's diagnosis, diagnostic status, age, sex, and non‐verbal cognitive ability and severity of language impairment), nor those regarding of the treatment components and implementation of the language interventions (intervention content, setting, delivery agent, session structure of the intervention or total number of sessions) reached significance. The same occurred to indicators of study quality. The risk of bias assessment showed that reporting quality for the studies examined in the review was poor. Authors’ Conclusions In sum, the current evidence base is promising but inconclusive. Pre‐registration and replication of more robust and adequately powered trials, which include a wider range of diagnostic conditions, together with more long‐term follow‐up comparisons, are needed to drive evidence‐based practice and policy.