Comorbidity between reading disability and ADHD in a community sample: Implications for academic, social, and neuropsychological functioning

Abstract

To better understand the implications of comorbidity between reading disability (RD) and attention-deficit / hyperactivity disorder (ADHD), a sample of 225 participants with RD without ADHD, 139 participants with both RD and ADHD, and 1,502 children without reading or attentional difficulties was recruited through five large public school districts. In comparison to the group without RD or ADHD, both groups with RD exhibited elevations of comorbid internalizing and externalizing disorders and significant global, academic, and social impairment. However, the group with both RD and ADHD was most impaired on most measures, and analyses of neuropsychological measures indicate that the co-occurrence of RD and ADHD may be due at least in part to weaknesses in cognitive processing speed and working memory that are most severe in the comorbid group. These results indicate that psychoeducational assessments of RD should always screen for ADHD and other emotional and behavioral difficulties, and that when RD and ADHD co-occur interventions are likely to be needed for both disorders.

Lay Abstract

Reading disability (RD) is a common childhood disorder that is defined by significant and unexpected underachievement in reading. In addition to specific difficulties in reading, many individuals with RD also meet criteria for at least one other disorder, a phenomenon known as comorbidity. To better understand the implications of comorbidity between RD and attention-deficit / hyperactivity disorder (ADHD), a sample of 225 participants with RD without ADHD, 139 participants with both RD and ADHD, and 1,502 children without reading or attentional difficulties was recruited through five large public school districts. Results indicated that RD is associated with elevations of emotional and behavioral symptoms, significant academic and psychosocial difficulties. However, the group with both RD and ADHD was most impaired on most of these measures, and analyses of neuropsychological measures indicate that the co-occurrence of RD and ADHD may be due at least in part to weaknesses in cognitive processing speed and executive functions that are most severe in the group with both disorders. These results indicate that psychoeducational assessments of RD should always screen for ADHD and other emotional and behavioral difficulties, and that when RD and ADHD co-occur interventions are likely to be needed for both disorders.

Reading disability (RD) is a common childhood disorder that is defined by significant and unexpected underachievement in reading based on an individual’s age and development (e.g., American Psychiatric Association, 2000). In addition to specific difficulties in reading, the majority of individuals with RD also meet criteria for at least one other disorder, a phenomenon known as comorbidity (e.g., Goldston et al., 2007; Sexton, Gelhorn, Bell, & Classi, 2012; Visser et al., 2020; Wakeman et al., In press; Willcutt & Pennington, 2000b; Willcutt et al., 2013).

It is important to understand the frequency, causes, and implications of comorbidity for several reasons. Individuals with more than one disorder often differ in important ways from individuals with a disorder in isolation, with the comorbid group frequently experiencing greater symptom severity, more extensive and severe functional and neurocognitive impairment, and poorer long-term outcomes (e.g., Purvis & Tannock, 2000; Waschbusch, 2002; Willcutt et al., 2019). High rates of comorbidity also raise practical questions about whether treatment of one disorder will also improve symptoms of the comorbid disorder, or whether treatment approaches need to be modified to account for comorbidity.

The current paper uses a large community sample of children and adolescents to examine the prevalence, functional implications, and cognitive correlates of comorbidity between RD and attention-deficit/hyperactivity disorder (ADHD), a neurodevelopmental disorder that co-occurs with RD in 25 – 40% of individuals (e.g., Sexton et al., 2012; Willcutt & Pennington, 2000a).

Previous studies of the implications of comorbidity between RD and ADHD

Other comorbid disorders.

In addition to comorbidity with ADHD, individuals with RD also exhibit higher rates of internalizing and externalizing symptoms (e.g., Daniel et al., 2006; Fortes et al., 2016; Goldston et al., 2007; Visser et al., 2020; Wakeman et al., In press; Willcutt et al., 2019; Willcutt & Pennington, 2000b). However, several initial studies suggest that associations between RD and externalizing disorders such as oppositional defiant disorder (ODD) and conduct disorder (CD) may be largely restricted to the subset of individuals with comorbid ADHD (e.g., Frick et al., 1991; Maughan, Rowe, Loeber, & Stouthamer-Loeber, 2003; Willcutt et al., 2019; Willcutt & Pennington, 2000b). In contrast, some studies suggest that RD may be independently associated with elevations of anxiety, depression, and suicidal ideation even after controlling for symptoms of ADHD and other disruptive disorders (e.g., Goldston et al., 2007; Maughan et al., 2003; Willcutt et al., 2019; Willcutt & Pennington, 2000b). This pattern is not found consistently in all studies, however, and may differ in males and females (e.g., Goldston et al., 2007; Willcutt & Pennington, 2000b), underscoring the need for additional research using samples with sufficient statistical power to clarify these associations.

Functional impairment.

In comparison to individuals without RD, individuals with RD receive lower grades and report lower academic motivation, are more likely to drop out of school prior to completing high school, and reach lower levels of educational and occupational attainment as adults (e.g., Boetsch, Green, & Pennington, 1996; Daniel et al., 2006; Goldston et al., 2007; Willcutt et al., 2019; Willcutt et al., 2013). In addition to these pervasive academic difficulties, initial studies found that individuals with RD also had lower overall adaptive functioning and experienced more frequent peer rejection and social isolation (e.g., Goldston et al., 2007; Willcutt et al., 2019), although one study suggested that social difficulties may be restricted to the group with comorbid ADHD (Willcutt et al., 2007). Additional research is needed to test whether these initial results replicate in an independent sample and in both males and females, and to extend this work by systematically examining more specific aspects of social and academic functioning.

Neuropsychological studies of RD, ADHD, and their comorbidity

In contrast to early models that suggested that a single primary neurocognitive deficit was necessary and sufficient to explain disorders such as RD and ADHD, more recent studies suggest that RD and ADHD are associated with multiple neuropsychological weaknesses (e.g., Pennington, 2006; Peterson et al., 2017; Willcutt et al., 2010). While difficulties with phonological processing are one of the strongest neuropsychological correlates of reading difficulties and RD (e.g., Vellutino, Fletcher, Snowling, & Scanlon, 2004), recent studies suggest that RD is also independently associated with weaknesses in naming speed, processing speed, and aspects of executive functions (EF) such as working memory (e.g., McGrath et al., 2011; Peterson et al., 2017; Willcutt et al., 2010). Similarly, ADHD is independently associated with weaknesses in processing speed and aspects of EF that include response inhibition and working memory (e.g., Willcutt, 2015; Willcutt, Doyle, Nigg, Faraone, & Pennington, 2005).

These multiple deficit neuropsychological models may also provide a framework to understand the neurocognitive mechanisms that lead to comorbidity between RD and ADHD. For example, initial studies suggest that weaknesses in processing speed or verbal working memory may be domain-general risk factors that increases risk for both RD and ADHD, sometimes leading to comorbidity between the two disorders (e.g., McGrath et al., 2011; Peterson et al., 2017; Willcutt et al., 2010). However, these results come primarily from a single sample and results for executive functions have been mixed, underscoring the need for additional studies of independent samples.

The current study

A community sample of 8 – 14 year old participants with RD only (N = 225), RD + ADHD (N = 149), and neither RD nor ADHD (N = 1,502) completed an extensive battery of measures of internalizing and externalizing psychopathology and global, academic, and social functioning, along with measures of four neuropsychological constructs that are the best candidates for a shared neuropsychological weakness that may help to explain comorbidity between RD and ADHD. This is the first time that this sample has been used to examine the implications of comorbidity between RD and ADHD, and the results expand on earlier research in several important ways. Specific predictions were as follows:

1. We hypothesized that significant comorbidity between RD and externalizing disorders would be explained by comorbidity with ADHD, whereas RD would be independently associated with major depressive disorder and generalized anxiety disorder. We also anticipated that the association between RD and externalizing symptoms would be stronger in males, whereas females with RD would exhibit higher rates of internalizing disorders.

2. We hypothesized that RD would be associated with pronounced difficulties in academic functioning that would not be explained by comorbid ADHD. In contrast, individuals with RD alone were expected to show little or no impairment on measures of social functioning, whereas we expected the group with both RD and ADHD to be impaired on these measures.

3. Based on the pervasive neuropsychological deficits observed in previous studies, we predicted that both groups with RD would exhibit significant deficits on neuropsychological measures of processing speed, sustained attention, and executive functions in comparison to the group without RD or ADHD. We hypothesized that weaknesses in working memory and processing speed would emerge as significant shared weaknesses in RD and ADHD, and that the largest effect sizes for these measures would be observed in the RD + ADHD group.

Method

Participants

The initial screening procedures are described in detail in previous papers (e.g., Creque & Willcutt, 2021; Willcutt, 2012; Willcutt et al., 2011). Teachers and parents of all students between kindergarten and 8th grade in five public school districts (N = 8,238) were invited to complete an initial screening questionnaire, and a subset of participants (N = 1,856) then completed a more extensive assessment that included the measures of word reading, psychopathology, and academic, psychosocial, and neuropsychological functioning that are the focus of the current paper. Study staff conducted a telephone screening interview prior to the individual assessments, and potential participants with a documented brain injury or significant hearing or visual impairment were excluded from the sample.

Operational definitions of RD and ADHD.

The initial screening procedure was completed for all students between kindergarten and 8th grade, but only participants between 8 and 14 years of age were invited to complete the subsequent individual testing session to ensure that all participants had received initial reading instruction. RD was defined by a score below the 10th percentile of the population on the Word Identification subtest of the Woodcock-Johnson Tests of Achievement, Third Edition (Woodcock, McGrew, & Mather, 2001). Parents and teachers rated each child’s ADHD symptoms on the Disruptive Behavior Rating Scale (DBRS; Barkley & Murphy, 1998), and these ratings were then combined using the algorithm from the DSM–IV field trials (Lahey et al., 1994).

Participants were included in the RD + ADHD group if they scored below the reading cutoff score and met full DSM–IV diagnostic criteria for ADHD, including significant functional impairment across multiple settings. Participants in the group with RD without ADHD scored below the reading cutoff score but exhibited no more than five symptoms of inattention or hyperactivity-impulsivity. Finally, participants were included in the comparison group if they did not meet criteria for RD and exhibited no more than two symptoms of inattention or hyperactivity-impulsivity. Because the overall study focused on the comparison of groups with and without RD, children who met criteria for ADHD alone based on the initial screening did not complete the full assessment session, and were therefore not included in the current analyses.

Descriptive and demographic characteristics.

The final sample included 225 participants with RD without ADHD (123 female and 102 male), 139 participants with both RD and ADHD (Ns = 54 female and 85 male), and 1,502 children without reading or attentional difficulties (Ns = 782 female and 720 male). Males with RD were more likely than females with RD to meet criteria for comorbid ADHD (45% vs. 30%; OR = 1.9, p < .01)

Parent reports of race and ethnicity indicated that 78% of participants were White, 20% were Black or African American, 4% were Native American/American Indian, and 4% were Asian American (participants were allowed to indicate more than one race). Across all races 14% of the sample identified their ethnicity as Hispanic. There were no significant group differences in race or ethnicity.

Other demographic and descriptive characteristics of the sample are summarized in Table 1. The three groups did not differ in age, but maternal education was significantly lower in the groups with RD. As expected based on how the groups were defined, the RD and RD + ADHD groups scored lower than the comparison group on the measure of word reading, along with the measures of general cognitive ability and mathematics achievement. Importantly, the groups with RD with and without ADHD did not differ on the measures of word reading or general cognitive ability, indicating that any differences between the two RD groups on other dependent variables were not simply a reflection of a difference in the severity of the reading deficit in the two groups.

Table 1.

Descriptive characteristics and diagnostic measures in groups with RD, RD + ADHD, and a comparison group without RD or ADHD

				Effect size for the comparison between each pair of groups
	Comparison (N = 1,502)	RD only (N = 225)	RD + ADHD (N = 139)	RD only vs. Comparison	RD + ADHD vs. Comparison	RD + ADHD vs. RD only
	M (SD)	M (SD)	M (SD)	d	d	d
Descriptive characteristics
Age	10.6 (1.9)	10.5 (2.1)	10.6 (1.9)	−0.04	0.01	0.05
Maternal education (years)	15.6 (2.4)	14.9 (2.5)	14.2 (2.6)	0.33*	0.59*	0.25
Paternal education (years)	15.9 (2.6)	15.0 (2.6)	13.8 (2.7)	0.34	0.76*	0.43
General Cognitive Ability
WISC Vocabulary	11.7 (3.0)	8.1 (2.9)	7.7 (2.9)	1.21**	1.34**	0.15
WISC Block Design	11.4 (3.2)	9.3 (3.6)	8.6 (3.5)	0.63**	0.83**	0.19
Academic Achievement
WJ-III Word Identification	106.5 (10.7)	81.8 (7.9)	80.5 (7.6)	2.66**	2.89**	0.18
WJ-III Calculations	106.7 (15.5)	90.8 (14.4)	84.1 (14.2)	1.06**	1.52**	0.47*
ADHD Symptoms
Inattention	0.3 (0.6)	1.7 (1.8)	7.6 (1.6)	1.61**	6.64**	3.51**
Hyperactivity - Impulsivity	0.3 (0.7)	0.9 (1.5)	4.7 (2.9)	0.54*	2.50*	1.71**

^*= P < .01

^**= P < .001

Procedures

All measures were administered in a single session by an examiner who had extensive experience working with children. The testing session lasted approximately two hours, and frequent breaks were provided to minimize fatigue and maximize motivation. If a participant was taking psychostimulant medication, their parent was asked to withhold medication for 24 hours prior to the study (N = 41 in the group with RD + ADHD; results did not change when medication status was included as a covariate in analyses). While the child completed the academic and neuropsychological measures, the child’s parent completed the questionnaires in a separate room.

Measures

Due to space constraints this section includes a brief description of the individual measures, and a more detailed description of the measures is provided in the online supplemental materials.

Measures of psychopathology.

Parents completed the DSM–IV Diagnostic Interview for Children and Adolescents (DICA-IV; e.g., Reich, Welner, & Herjanic, 1997), a structured interview that provides diagnoses of ODD, CD, major depressive disorder (MDD). and generalized anxiety disorder (GAD).

Measures of academic, social, and overall functioning.

As part of the DBRS parents and teachers rated the extent to which each participant experienced overall academic difficulties or had trouble understanding assignments and completing homework. Parents and teachers also estimated the student’s current grades in reading, math, and English / language arts, and indicated whether the participant had received extra academic support through an Individual Education Plan (IEP) or services outside the school system. Social functioning was measured by parent ratings on the Social Skills, Leadership, and Withdrawn Behavior scales from the Behavior Assessment Scale for Children, Second Edition (BASC-II; Reynolds & Kamphaus, 2004), along with parent and teacher ratings of the proportion of children who like, dislike, or ignore the participant (Dishion, 1990). Finally, measures of global adaptive functioning were provided by parent ratings on the Child Global Assessment Scale (Setterberg, Bird, & Gould, 1992) and parent and teacher ratings of the participant’s ability to consistently manage daily responsibilities.

Neuropsychological measures.

Measures of processing speed, sustained attention, working memory, and response inhibition were administered because these cognitive constructs are among those that were most strongly associated with RD and ADHD in previous studies. A response inhibition composite score included stop-signal reaction time from the Stop-signal task (e.g., Logan, Schachar, & Tannock, 1997; Schachar, Mota, Logan, Tannock, & Klim, 2000) and commission errors on a standardized continuous performance test (CPT) that assesses the ability to inhibit inappropriate responses during an extended visual task (Gordon, 1983). Scores from the Sentence Span task (Siegel & Ryan, 1989) and the digits backward component of the WISC-III Digit Span subtest (Wechsler, 1991) were used to create a working memory composite score. A processing speed composite score was derived from scores from the WISC-III Coding and Symbol Search subtests and the Trailmaking Test (Reitan & Wolfson, 1985). Finally, sustained attention was measured by the number of omission errors during the CPT task.

Data Analysis

Data cleaning and transformations.

Standard procedures were used to identify and adjust any observed outliers prior to analyses (e.g., Creque & Willcutt, 2021; Willcutt, Pennington, Olson, Chhabildas, & Hulslander, 2005), and an appropriate transformation was implemented to approximate a normal distribution for variables with skewness or kurtosis greater than three (CPT commission errors and completion time for the Trail Making Test). All measures were age-corrected and standardized based on the overall sample. When multiple measures of a construct were available, composite measures were created by computing and re-standardizing the mean of those scores.

Primary analyses.

Due to the high number of statistical tests, an alpha of .01 was adopted as the threshold for statistical significance. Analyses of variance and covariance and chi-square tests were used to compare means and proportions in the three groups. If the initial analysis revealed a significant main effect of group, planned pairwise comparisons were conducted among the three groups.

Potential confounding variables and analyses of age and biological sex assigned at birth.

Parent education and estimated performance IQ were included as covariates in initial models, but were dropped from final models because they did not have a significant impact on any result. Potential differences in the pattern of results as a function of age or biological sex assigned at birth were examined by testing for Group x Sex (Male vs. Female) or Group x Age (8 – 11 versus 12 – 14 years old) interactions.

Multiple deficit cognitive models.

A final set of analyses were conducted to test multiple deficit neuropsychological models of RD and ADHD. Scores on the four cognitive measures were regressed simultaneously onto reading scores and measures of ADHD symptoms to test which cognitive factors are independently associated with each disorder and which may be a shared weakness that helps to account for comorbidity between RD and ADHD.

Results

Comorbidity with other disorders

In comparison to individuals without RD, individuals with RD were significantly more likely to meet criteria for mathematics disability, ODD, CD, GAD, and MDD (Figure 1). However, pairwise comparisons indicated that RD was only associated with elevated rates of ODD and CD if the participant also met criteria for ADHD. In contrast, both groups with RD were more likely than the group without RD to meet criteria for math disability, MDD, GAD, although the rates of MDD and math disability were higher in the group with RD + ADHD than the group with RD alone.

Figure 1.

Rates of Math Disorder, MDD, GAD, ODD, and CD in groups with RD + ADHD, RD-only, and a comparison group without RD or ADHD. Note: Bars with different letters indicate a significant difference between groups (P < .01).

Global, academic, and social functioning

Table 2 and Figures 2 - 4 summarize analyses that compared the three groups across multiple domains of functioning. Because the outcome measures in Figures 2 - 4 are dimensional, the bars in each figure represent the effect size for the comparison between the groups with RD or RD + ADHD and the group without RD or ADHD (Cohen’s d; Cohen, 1988)

Table 2.

Grades and Services Received in Groups with RD only, RD + ADHD, and a comparison group without RD or ADHD

				Effect size for the comparison between each pair of groups
	Comparison (N = 1,502)	RD only (N = 225)	RD + ADHD (N = 139)	RD only vs. Comparison	RD + ADHD vs. Comparison	RD + ADHD vs. RD only
Grades	M (SD)	M (SD)	M (SD)	d	d	d
Reading	3.6 (0.7)	2.4 (1.0)	1.7 (1.0)	1.46***	2.23***	0.64***
Mathematics	3.5 (0.71)	2.7 (1.0)	2.0 (1.1)	0.94***	1.73***	0.73***
English / Language Arts	3.5 (0.7)	2.5 (1.1)	1.9 (1.1)	1.11***	1.81***	0.55***
Services Received	N (%)	N (%)	N (%)	Odds Ratio	Odds Ratio	Odds Ratio
Learning Disability Diagnosis	53 (4%)	89 (40%)	84 (60%)	17.9**	41.8**	2.3**
IEP / Special Education Services	99 (7%)	86 (38%)	103 (74%)	8.8**	40.5**	4.6**
Any Reading Assistance	166 (11%)	151 (67%)	119 (86%)	16.4**	47.9**	2.9**
Any Math Assistance	125 (8%)	76 (34%)	82 (59%)	5.6**	15.9**	2.8**

^*= P < .01

^**= P < .001

Figure 2.

Adaptive and academic functioning in groups with RD + ADHD versus RD-only. Note: Bars indicate the effect size (Cohen’s d) for comparisons between the groups with RD and the comparison group without RD or ADHD. The RD + ADHD and RD-only groups differed significantly from the comparison group on all measures (P < .01). Bars with different letters indicate a significant difference between the RD + ADHD and RD-only groups (P < .01).

Figure 4.

Performance on the neuropsychological measures in groups with RD + ADHD versus RD-only. Note: Bars indicate the effect size (Cohen’s d) for comparisons between the groups with RD and the comparison group without RD or ADHD (P < .01)). The RD + ADHD and RD-only groups differed significantly from the comparison group on all measures. Bars with different letters indicate a significant difference between the RD + ADHD and RD-only groups (P < .01).

Academic functioning and services received.

In comparison to the group without RD, both groups with RD had lower grades in all academic subjects and were less likely to complete homework and other daily academic responsibilities, but each of these effects was significantly larger in the group with RD + ADHD versus the group with RD alone (Table 2 and Figure 2). Similarly, both groups were more likely than the comparison group to have received academic services, but the group with RD + ADHD was most likely to have received an LD diagnosis, special education services, and reading or mathematics assistance (Table 2).

Social and adaptive functioning.

Both groups with RD exhibited greater social and overall adaptive impairment than the group without RD (Figure 2 and Figure 3), but impairment was greatest in the group with RD + ADHD for measures of overall social functioning and social skills, aggressive social behavior, and tendency to be disliked by peers (Figure 3). In contrast, both RD groups were more likely to be shy, withdrawn, and socially isolated, and there were no differences between the groups with and without comorbid ADHD on these measures.

Figure 3.

Social functioning in groups with RD + ADHD versus RD-only. Note: Bars indicate the effect size (Cohen’s d) for comparisons between the groups with RD and the comparison group without RD or ADHD. The RD + ADHD and RD-only groups differed significantly from the comparison group on all measures (P < .01). Bars with different letters indicate a significant difference between the RD + ADHD and RD-only groups (P < .01).

Neuropsychological functioning

All four neuropsychological composite measures were significantly correlated with individual differences in reading and ADHD symptoms (Table 3). Similarly, both RD groups performed more poorly than the group without RD or ADHD on WISC-III Block Design and the measures of processing speed, working memory, inhibition, and sustained attention (Figure 4), and the RD + ADHD group scored lower than the group with RD alone on the measures of verbal working memory, inhibition, and processing speed.

Table 3.

Relations between measures of cognitive functioning and measures of reading and ADHD symptoms

	Correlations				Simultaneous Multiple Regression Analyses Predicting Reading and ADHD symptoms
	Process Speed	Inhibition	Working Memory	Sustained Attention	Processing Speed	Inhibition	Working Memory	Sustained Attention
Diagnostic Measures
Word Reading	.40**	.35**	.28**	.29**	.31 [.23, .39]**	.08 [.00, .16]	.52 [.43, .59]**	.09 [.03, .15]*
Inattention	.42**	.38**	.29**	.28**	.31 [.24, .38]**	.18 [.10, .26]**	.23 [.15, .31]**	.05 [−.01, .11]
Hyperactivity/Impulsivity	.32**	.33**	.20**	.14**	.23 [.15, .31]**	.17 [.11, .23]**	.09 [.00, .17]	.00 [−.06, .06]

^*= P < .01

^**= P < .001.

Note: Overall results were the same when RD and ADHD group status was predicted in a multiple logistic regression model.

When the four composites were regressed onto reading and ADHD symptoms in separate linear regression models (Table 3), only weaknesses in processing speed and working memory were significantly associated with both RD and ADHD (similar results were obtained in parallel logistic regression models predicting RD or ADHD status). In contrast, inhibitory difficulties were uniquely associated with ADHD but not RD, and sustained attention was associated with RD and not ADHD.

Impact of age and sex

Age.

While most results did not vary as a function of age, a significant Age x Group interaction indicated that grades in reading were more impaired in younger children with RD (8 – 11 years old) than in older children with RD (p < .01). No other Age x Group interactions were significant.

Biological sex assigned at birth.

Significant Sex x Group interactions indicated that females with RD were more likely than males with RD to meet criteria for MDD (23% vs. 11%). Further, females with both RD and ADHD were most likely to meet criteria for MDD (31%) and were most impaired on measures of social skills and isolated and withdrawn behaviors (all P <.01). In contrast, males in both RD groups were more likely than females with RD to meet criteria for conduct disorder (20% vs. 8%), had lower reading grades, and were more disruptive in the classroom. No other Sex x Group interactions were significant.

Discussion

Approximately 40% of children with RD also met criteria for ADHD in the current community sample, a rate of comorbidity that is similar to previous studies (e.g., Willcutt & Pennington, 2000a). In this section we discuss the implications of this common comorbidity for our understanding of the adaptive, academic, social, and neuropsychological functioning of individuals with RD, then highlight several key implications of these results for clinicians and educators working with individuals with RD.

Other comorbidity and functional impairment

Other comorbid disorders.

Consistent with our initial predictions and previous studies (e.g., Frick et al., 1991; Willcutt & Pennington, 2000b; Willcutt et al., 2013), comorbidity between RD and both ODD and CD was restricted to the subset of participants who also met criteria for ADHD, whereas individuals with RD had higher rates of math disability, GAD, and MDD even after comorbidity with ADHD was taken into account. However, while ADHD did not fully explain the association between RD and MDD, the group with comorbid RD + ADHD exhibited significantly higher rates of MDD than the group with RD alone, and this pattern was especially pronounced among females with RD and ADHD.

Academic functioning.

In comparison to the group without RD or ADHD, groups with RD only and RD + ADHD received significantly lower grades in all subjects and had greater difficulty completing assignments and homework, and effect sizes were large for all comparisons (d = 0.9 – 2.2). Nonetheless, all of these academic difficulties were once again more severe in the group with both RD and ADHD than the group with RD alone, and the comorbid group was more likely than the group with RD alone to have received a diagnosis of a learning disability, special education services, and additional academic assistance outside of the school setting. These patterns are consistent with our initial predictions and results from previous studies (Willcutt et al., 2007), underscoring the need to take comorbid ADHD into account to maximize the academic outcomes of children with RD.

Social functioning.

In contrast to our initial prediction that any social difficulties associated with RD would be explained by comorbid ADHD, the group with RD alone exhibited significant difficulties in all measured domains of social functioning. However, effects were much larger in the group with RD + ADHD (d = 1.1 – 1.4) than the group with RD alone (d = 0.3 – 0.6) on measures of overall social functioning, social skills, and aggressive behavior, and females with both RD and ADHD were more likely to be isolated and withdrawn than females without RD.

Conclusions regarding comorbidity and functional impairment.

Overall, these results indicate that RD is associated with comorbid internalizing and externalizing disorders and significant global, academic, and social impairment. However, the presence of comorbid ADHD is associated with more severe difficulties in nearly all of these domains, and may entirely explain the comorbidity between RD and externalizing disorders such as ODD and CD. Comprehensive psychoeducational assessments of RD should also screen for ADHD (and other emotional and behavioral difficulties that often co-occur with RD and ADHD), and should carefully consider both academic and social functioning when developing recommendations for interventions or accommodations.

Neuropsychological functioning

When measures of processing speed, working memory, response inhibition, and sustained attention were included simultaneously as predictors in multivariate regression models, individual differences in reading and RD status were associated with weaknesses in processing speed and working memory. Parallel analyses indicated that processing speed and working memory are also independently associated with symptoms of ADHD and ADHD group status, suggesting that shared cognitive weaknesses in these domains may help to explain comorbidity between RD and ADHD. In contrast, difficulties with inhibition were associated with ADHD but not RD, and difficulties with sustained attention were associated with RD but not ADHD.

These findings are consistent with studies of other samples that found that slow processing speed is a shared weakness in multiple deficit neuropsychological models of RD, ADHD, and other disorders such as math disability (e.g., Peterson et al., 2017; Willcutt et al., 2010). Studies of working memory have been somewhat less consistent, with some studies suggesting that reading difficulties are not strongly associated with working memory after controlling for weaknesses in phonological processing or verbal reasoning (e.g., Peterson et al., 2017). Future studies that include a more comprehensive neuropsychological battery will provide a useful extension of the current research to clarify the nature of the shared weakness in executive functions in RD and ADHD.

Implications for Clinical Diagnoses, Interventions, and Accommodations

RD and ADHD are independently associated with important aspects of adaptive, academic, and social impairment. Therefore, when RD and ADHD co-occur they should be conceptualized as distinct but related disorders that are each likely to require intervention (Fletcher, Shaywitz, & Shaywitz, 1999). Despite the voluminous literatures describing treatment of RD and ADHD in isolation, few intervention studies of either disorder have directly examined the potential impact of comorbidity. An initial study suggested that treatment of reading difficulties may not be as effective for children with concurrent attentional difficulties (Rabiner & Malone, 2004), but a second study of a larger sample demonstrated that children with both RD and ADHD benefitted from treatments that specifically targeted both reading and ADHD (Denton, Tamm, Schatschneider, & Epstein, 2020). Overall, much more work is needed to fill the important gap in the literature on the potential impact of comorbidity with ADHD or other disorders on the effectiveness of interventions for RD.

While measures of neuropsychological functioning are not used for clinical diagnoses of RD or ADHD, individual differences on these measures may potentially help to optimize strategies for accommodations for students with RD in school settings and in the home environment. For example, an individual with RD who has significant executive function difficulties might need to be seated in an area of the classroom where distractions are minimized, or may benefit from a structured behavioral and organizational system that provides consistent reminders about tasks to be completed. Similarly, an individual with RD with severe processing speed difficulties may require extended time to complete assignments so that slow processing speed does not compromise their ability to demonstrate mastery of the material.

Limitations and Future Directions

Despite important strengths such as a large sample size and extensive battery of measures of academic, social, and neuropsychological functioning, the current results should also be interpreted in the context of several important limitations.

Other comorbidities.

The current results provide important insights regarding the impact of comorbid ADHD on the functioning of individuals with RD, but budget constraints precluded us from including a group with ADHD without RD in the current study. Future studies may provide a useful extension of the current research by including a group with ADHD alone or groups with RD and other comorbid disorders such as anxiety and depression.

Sample composition.

Although the current sample is representative of the school districts from which it was drawn, the majority of participants were non-Hispanic Caucasians, and relatively few children were from lower SES environments. These factors may limit the generalizability of the current results, and underscore the need for future research on the implications of comorbidity in more diverse populations with RD.

Neuropsychological Measures.

The current results add to a broader literature that suggests that shared weaknesses in processing speed and working memory may contribute to comorbidity between RD and ADHD. However, interpretation of these results is not always straightforward, as slow processing speed could potentially be associated with RD and ADHD for different reasons. For example, processing speed weaknesses could reflect general slow processing across all trials in groups with RD, but intermittent lapses in attention on a subset of trials in groups with ADHD. Similarly, the verbal working memory measures in the current study may be more strongly associated with language abilities but have less of a working memory load than more complex and difficult working memory tasks. Future studies that include a more extensive battery of neuropsychological measures would provide a useful extension of the current results.

Conclusions

RD and ADHD co-occur significantly more often than expected by chance, and individuals with both RD and ADHD exhibit increased impairment on measures of global, academic, social, and neuropsychological functioning in comparison to individuals with RD alone. Analyses of neuropsychological measures suggest that comorbidity between RD and ADHD may be due at least in part to slower processing speed and weaknesses in working memory, a specific dimension of executive functions. Future studies should continue to extend and refine neuropsychological models of RD, ADHD, and other related disorders, and test the potential impact of this common comorbidity on the outcomes of interventions for RD.