Predicting and Evaluating Treatment Response: Evidence Toward Protracted Response Patterns for Severely Impacted Students With Dyslexia.

Abstract

Great strides have been made in the development of effective methods of instruction for children with dyslexia. However, individual response to treatment varies, and weaknesses persist for some students with dyslexia despite otherwise effective instruction. Continued efforts are needed to support the prospective identification of poor response, particularly in routine intervention settings. The current study addressed whether indicators of dyslexia risk as outlined by hybrid diagnostic models predict response in children who received Tier 3 dyslexia intervention in their schools. The program's efficacy has been previously documented in remediating reading abilities in children with dyslexia. Data were examined from 115 elementary-age children who received routine Tier 3 dyslexia intervention in their schools. Logistic regression revealed powerful effects of preintervention fluency and gender in predicting response, with weaker effects of decoding and rapid naming. Attention-deficit/hyperactivity disorder status also played a role in predicting response. Phonological awareness and listening-reading comprehension discrepancy did not predict response. Profile analyses indicated near- and far-transfer of skill for the adequate response group, whereas growth in the poor response group was limited to near-transfer. Findings support a continuum of severity that may be associated with less robust growth and generalization over the course of the intervention.