Population-Based Screening for Language Delay: Let’s Talk STEPS Study

Abstract

This study evaluates the reliability of two differently constructed screening instruments for language delay. Only few studies have addressed the question of early identification at population level. Data for this article were drawn from a Finnish cohort study, entitled the Steps to the Healthy Development and Well-Being of Children (the STEPS study, N = 9.936). The MacArthur Communicative Development Inventories (CDI-T), based on parent reported vocabulary, was used for screening at 24 months. At 36 months, child’s language skills were screened with the Fox Language Inventory (FLI) carried out by a clinical nurse. The Renfrew Word Finding Vocabulary Test and Reynell Developmental Language Scales III (language comprehension), served as outcome measures at 36 months. Receiver operating characteristic-analysis (ROC) was used to examine the cost and benefit of the two screening methods in decision making at 36 months. We found that expressive vocabulary at 24 months, can already foretell later language development. However, to reach even better predictivity, screening based on a structured language battery and age point of 36 months would be a valuable addition to clinical assessment. Further studies are needed to address to what extend early screening is able to predict atypical language during later preschool-years.