Kindergarten Screening for Early (Grade 1) and Late-Emerging (Grade 4) Dyslexia Risk.

Abstract

Pediatric primary care offers a unique opportunity for early dyslexia screening, yet validated prereading measures capturing the full spectrum of developmental risk are absent from standard assessments, representing a critical gap in preventive care. Establishing predictive validity of these measures is essential before clinical implementation. To examine whether and to what extent deficits in 4 cognitive-linguistic domains in kindergarten estimate risk for both early-emerging (grade 1) and late-emerging (grade 4) dyslexia. This prospective longitudinal cohort study was conducted at 128 kindergartens and 60 public elementary schools representing diverse socioeconomic backgrounds in northern Israel from May 2019 to June 2023. Hebrew-speaking children who demonstrated age-appropriate nonverbal abilities (as assessed with the Raven Progressive Matrices) were included. Exposures included 4 cognitive-linguistic measures, (1) phonological awareness, (2) rapid automatized naming, (3) letter knowledge, and (4) morphological awareness, assessed in kindergarten (May to June 2019). The primary outcome was dyslexia risk in grades 1 (January to March 2020) and 4 (February to June 2023), defined as word-reading fluency at or below the 10th percentile. Cognitive-linguistic and word-reading fluency composite scores were computed by averaging raw scores from 2 to 3 tasks. Logistic regression was used to estimate associations of cognitive-linguistic measures with dyslexia risk, which were expressed as odds ratios (ORs) with 95% CIs. Among 515 Hebrew-speaking children (285 girls [55.3%]; mean [SD] age, 5.9 [0.4] years), kindergarten deficits in letter knowledge (OR, 4.75; 95% CI, 2.04-11.04; P < .001) and phonological awareness (OR, 4.17; 95% CI, 2.05-8.47; P < .001) were associated with a 4- to 5-fold increased risk of dyslexia in grade 1. Deficits in letter knowledge (OR, 3.57; 95% CI, 1.53-8.31; P = .003), morphological awareness (OR, 2.56; 95% CI, 1.09-5.97; P = .03), and rapid automatized naming (OR, 2.39; 95% CI, 1.05-5.43; P = .04) were associated with a 2.4- to 3.6-fold increased risk of dyslexia in grade 4, independently of grade 1 risk (OR, 4.98; 95% CI, 2.22-11.13; P < .001). In this cohort study of 515 children followed from kindergarten through grades 1 and 4, distinct patterns of cognitive-linguistic deficits in kindergarten were associated with 2- to 5-fold increased risk for early- and late-emerging dyslexia. These findings provide a foundational evidence base to support the integration of developmentally sensitive screening into pediatric surveillance frameworks, facilitating early identification and a shift from reactive to preventive care, including for children at risk of late-emerging dyslexia.