A Meta-Analysis on the Effectiveness of Intervention in Children With Primary Speech and Language Delays/Disorders: Focusing on China and the United States
Clinical Psychology & Psychotherapy
Zhang, Z., Xu, Q., et al. (2021).
Clinical Psychology & Psychotherapy, 28(3), 585-605.
This meta-analysis investigates the short-term and long-term effects of various interventions in with primary speech and language delays or disorders who live in China or in English-speaking countries. This meta-analysis also explores whether certain factors (e.g., participant characteristics, intervention characteristics) moderate the intervention effect.
Humanities & Social Science Program,Ministry of Education in China; Open Program of Jiangsu Provincial Key Constructive Laboratory of Special Children's Impairment and Intervention (China)
Between 2002 and 2018
Randomized controlled trials
45 in the systematic review; 44 in the meta-analysis
Overall, speech and language interventions demonstrated a moderate immediate effect size (<em>g</em>) on language outcomes in children with primary speech and language delays or disorders (<em>g</em> = 0.70); however, long-term effects were indicated to be small and not as reliable (<em>g</em> = 0.23).
Findings suggested that language interventions can improve various speech and language outcomes in children and adolescents with primary speech and language delays or disorders. Language interventions demonstrated a small effect size (<em>g</em>) on the following outcomes in children with primary speech and language delays or disorders:
<ul>
<li>expressive syntax (<em>g</em> = 0.46)</li>
<li>receptive vocabulary (<em>g</em> = 0.40)</li>
<li>receptive syntax (<em>g</em> = 0.35)</li>
</ul>
Language interventions demonstrated a medium effect size on the following outcomes:
<ul>
<li>expressive vocabulary (<em>g</em> = 0.62)</li>
<li>expressive phonology (<em>g</em> = 0.72)</li>
<li>receptive phonology (<em>g</em> = 0.57)</li>
</ul>
Results indicated that interventions that lasted 21 or more weeks had the largest effect size (<em>g</em>) of 0.78, followed by interventions that lasted 10 to 20 weeks (<em>g</em> = 0.63). These findings suggested that a longer duration of intervention could be the most effective for children with primary speech and language delays or disorders.
Interventions provided in Chinese were indicated to have significantly larger effects (<em>g</em> = 1.03) than interventions provided in English (<em>g</em> = 0.44); however, differences in the effects may be due to differences in language structures and the variable quality of the included studies.
This review focused on four types of interventions: the naturalistic model, the didactic model, the hybrid model, and other models (e.g., traditional Chinese medicine). The hybrid model (<em>g</em> = 0.80) and naturalist model (<em>g</em> = 0.73) were reported to have larger effect sizes than the didactic model (<em>g </em>= 0.48). While child-centered interventions were suggested to be more effective for certain children with primary speech and language delays or disorders, findings indicated that the differences between the effect sizes of the models were not statistically significant.
Results indicated that interventions delivered by speech-language pathologists had the largest effect size (<em>g </em>= 0.80); however, there were no statistical differences between speech-language pathologists, caregivers (<em>g</em> = 0.56), and trained interveners (<em>g</em> = 0.48) as the provider of the intervention.
Phonology-based interventions (<em>g</em> = 1.05) and multidomain-based interventions (<em>g</em> = 0.68) were reported to have a greater effect on outcomes than vocabulary-based interventions (<em>g </em>= 0.37); however, these findings were limited due to an insufficient number of included studies investigating other domains.
