BackHome
Refresh: OffView Personal Schedule
Individual Submission Summary
Share...

Direct link:

A Meta-analytic Review of Reading Intervention Treatment Effects for ADHD

Wed, April 7, 3:15 to 4:15pm EDT (3:15 to 4:15pm EDT), Virtual

Abstract

Introduction: An estimated 33% to 80% of children with attention-deficit/hyperactivity disorder (ADHD) demonstrate academic difficulties (Mayes & Calhoun, 2006). Meta-analytic evidence indicates academic achievement discrepancies between children with and without ADHD are greatest in reading (Frazier et al., 2007). Nevertheless, there is limited understanding on the efficacy of reading interventions for ADHD, and limited research examining mechanisms that may bolster reading treatment effects for this high-risk population (DuPaul & Eckert, 1998). The present review addresses these gaps in the literature by examining (a) if extant reading interventions are effective for children with ADHD, and (b) if so, whether treatment effects are comparable to those seen in children without ADHD. Additional analyses were conducted to investigate treatment, participant, and study characteristics as moderators of treatment-related gains.

Methods: Literature search procedure followed PRISMA. Studies were included if participants were diagnosed or had well-described ADHD symptoms, at least one objective measure of reading achievement, and reported sufficient information to calculate Hedge’s g. Studies were excluded if participants were second language English learners or diagnosed with intellectual disability. Among 3,307 studies identified on PsycINFO, ERIC, and ProQuest Dissertations/Theses, 11 studies met the inclusion criteria, which yielded 45 effect sizes.

Results: Due to the large number of potential moderators and variance in several key study variables, a random effects model was estimated and resulted in a moderate effect size of g=0.53 (95%CI=0.44-0.62) with an I2 estimate of heterogeneity of 14.20%. A Q-test for this model indicated high levels of heterogeneity (Q=104.71, p<0.0001; I2=60.07%) providing evidence that testing for moderators was appropriate. Robust variance estimation was used to test the effects of clustering on the outcome, and resulted in no change to the estimated effect size (g=0.53), but did increase the overall confidence interval of said effect (95%CI=0.19-0.87). Funnel plots and Egger’s tests were used to address publication bias, finding no signs of bias (z=0.71, p=0.48).

Meta-regression was used to test possible moderators, including treatment outcome (e.g., reading comprehension versus decoding) and type (e.g., reading intervention only versus reading plus behavioral intervention), minutes of intervention, medication status, diagnosis method, and study year. All moderators were non-significant except diagnostic method (p=0.046), indicating that studies applying gold standard multi-informant/assessment ADHD diagnostic procedures yielded significantly smaller effect sizes (i.e., g=.33; small to moderate effect) than studies that used less rigorous methods such as single informant procedures or prior ADHD diagnoses only.

Discussion: Reading interventions produce moderate improvements in reading skills for children with ADHD. Surprisingly, these improvements were comparable to meta-analytic treatment effects reported for struggling readers without ADHD (e.g., g=0.53 verses g=0.49-0.54; Scammacca et al., 2007; Hall & Burns, 2018). The finding that medication status did not moderate intervention efficacy was consistent with prior studies that medication minimally affects academic achievement (e.g., Langberg & Becker, 2012). We hypothesize that the reduced efficacy found in studies with more rigorous diagnostics was due to better matched control groups and methods to measure study effects. Taken together, results indicate extant reading interventions are as efficacious for children with as without ADHD.

Group Authors

Florida State University

Authors