Secondary Logo

Journal Logo

Original Article

Unique Associations Between Specific Attention-Deficit Hyperactivity Disorder Symptoms and Related Functional Impairments

Zoromski, Allison K. PhD*,†; Epstein, Jeffery N. PhD*,†; Ciesielski, Heather A. PhD*,†

Author Information
Journal of Developmental & Behavioral Pediatrics: June/July 2021 - Volume 42 - Issue 5 - p 343-354
doi: 10.1097/DBP.0000000000000904

Current diagnostic criteria for attention-deficit hyperactivity disorder (ADHD) require the presence of (1) developmentally inappropriate levels of inattention (IA) and/or hyperactivity/impulsivity (HI) in more than one setting with several ADHD symptoms apparent before the age of 12 years and (2) symptom-related impairment in at least one domain of functioning.1 It is well documented that ADHD symptom profiles across children with ADHD are quite heterogeneous.2 Current Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition nosology attempts to capture this heterogeneity by using ADHD presentation types (i.e., inattentive [ADHD-IA; 6 or more of 9 inattentive symptoms in youth], hyperactive-impulsive [ADHD-HI; 6 or more of hyperactive/impulsive symptoms in youth], and combined [ADHD-C; both 6 or more inattentive of 9 and 6 or more of 9 hyperactive/impulsive symptoms in youth] presentations) (see Table 1 for the list of ADHD symptoms in each symptom domain). However, even within these presentation types, the constellation of symptoms and the severity of symptoms are very diverse across patients.

Table 1. - Descriptive Statistics
Parent Teacher
Mean SD N Mean SD N
Covariates
 Age 8.18 1.74 8689
 ODD 1.33 0.80 8689 0.40 0.57 8689
 Internalizing 0.91 0.67 8689 0.53 0.51 8689
Inattention
 Makes careless mistakes 2.12 0.80 8689 1.99 0.82 8689
 Difficulty sustaining attention 2.37 0.71 8689 2.29 0.80 8689
 Does not seem to listen 1.94 0.83 8689 1.63 0.90 8689
 Does not follow through 2.02 0.83 8689 1.84 0.94 8689
 Difficulty organizing 1.93 0.88 8689 1.92 0.91 8689
 Avoids tasks 1.99 0.91 8689 1.73 0.96 8689
 Loses things 1.71 0.98 8689 1.48 0.95 8689
 Is easily distracted 2.40 0.75 8689 2.37 0.78 8689
 Is forgetful 1.89 0.90 8689 1.69 0.91 8689
Hyperactivity
 Fidgets or squirms 2.04 0.98 8689 1.84 1.03 8689
 Leaves seat 1.72 1.01 8689 1.44 1.06 8689
 Runs about or climbs 1.33 1.08 8688 0.78 0.98 8689
 Difficulty playing quietly 1.25 1.00 8689 1.16 0.99 8689
 Is “on the go” 1.60 1.12 8689 1.17 1.08 8689
 Talks too much 1.70 1.04 8688 1.59 1.04 8689
Impulsivity
 Blurts out answers 1.45 1.03 8689 1.29 1.08 8689
 Difficulty waiting 1.63 1.02 8689 1.30 1.06 8689
 Interrupts or intrudes 1.41 1.11 8689 1.37 1.05 8689
Impairment
 School 3.48 1.04 8590
 Reading 3.36 1.23 8461 3.53 1.16 8469
 Writing 3.63 1.01 8561 3.92 0.96 8436
 Mathematics 3.28 1.14 8460 3.48 1.07 8355
 Relationship with parents 2.36 1.15 8633
 Relationship with siblings 2.68 1.16 7699
 Relationship with peers 2.90 1.05 8641 3.37 0.90 8649
 Participation in activities 2.96 1.09 8097
 Classroom behavior 3.64 1.18 8456
 Organizational skills 4.02 0.89 8602
ODD, oppositional defiant disorder.

Similarly, the impairment profiles across children with ADHD are quite heterogeneous in both the breadth of impairment and the severity of impairment across impairment domains (e.g., academics and social). Indeed, the range of ADHD-related impairments is broad. For example, children with ADHD often exhibit impairments in the school setting such as lower academic achievement than peers without ADHD, impairments in social settings (e.g., shy and passive behavior in social situations, disliked by peers, and ignored by peers), and impairments in adaptive functioning.2 In addition, for children with ADHD, impairments are evident during participation in organized activities. For example, in sports, children with ADHD often exhibit greater aggression, emotional reactivity, and injury than peers without ADHD.3

Examining the relations between ADHD symptoms and ADHD-related impairment has the potential to inform our understanding of why some patients with ADHD struggle with specific impairments and some do not. Most of the research on this topic has examined how the 2 ADHD symptom domains (i.e., IA and HI) relate to the various domains of ADHD-related impairments. This research has primarily taken 2 forms: (1) comparisons of impairment profiles across the ADHD presentations/subtypes and (2) examinations of the relations between each ADHD symptom domain and specific areas of impairment.

ADHD Presentations and Impairment

The results of a meta-analysis2 examining differences in impairment between children with different ADHD subtypes indicate that within the academic functioning domain, children with ADHD-C and ADHD-IA are more impaired based on parent and teacher ratings of academic achievement than children with ADHD-HI. Within the social impairment domain, children with ADHD-C have more social problems, are more disliked by peers, and exhibit less prosocial behavior than children with ADHD-IA or ADHD-HI. However, children with ADHD-IA tend to be more shy or passive than children with ADHD-C and are more often ignored by peers than children with ADHD-HI. In regard to family functioning, some studies have indicated that children with ADHD-C are more disruptive to family activities compared with children with ADHD-IA or ADHD-HI.4 However, other studies have not found subtype differences in family functioning-related variables.5

ADHD Symptom Domains and Impairment

In studies examining the relationship between the 2 ADHD symptom domains and impairment, a similar pattern has been noted. For example, several studies have found a strong relationship between the inattentive symptom domain and academic impairments.6,7 Garner et al.6 found that the inattentive symptom domain predicted parent- and teacher-rated impairment in reading, writing, and math after controlling for relevant covariates (i.e., sex, age, medication status, learning disability, internalizing symptoms, and oppositional symptoms). Moreover, the relation between the HI symptom domain and ratings of academic impairment tends to be either nonsignificant7 or significant but considerably smaller than relations between the IA domain and ratings of academic impairment.6

Although it seems that the HI symptom domain relates highly to social impairments,7 this relationship must be qualified by the impact of comorbid disorders on social functioning. Specifically, the HI domain is a strong predictor of impairment in peer and teacher relationships when comorbidities are not controlled for.7 However, it is well established that oppositional defiant disorder (ODD) symptoms (i.e., loses temper, touchy or easily annoyed, angry or resentful, argues with adults, actively defies, deliberately annoys others, blames others, and spiteful or vindictive) are prevalent among children with ADHD (e.g., Ref. 8) and that these symptoms have a large impact on social functioning.9 Once oppositional defiant symptoms are controlled for, the ADHD symptom domains account for very little variance in social relationships with parents, siblings, or peers of youth with ADHD.6

In studies that have examined the relationship between the ADHD symptom domains and impairments related to participating in sports or organized activities, both ADHD symptom domains have been shown to have moderate relationships to aggression, emotional reactivity, and injury in the context of sports.3 However, Garner et al.6 again found that the ADHD symptom domains accounted for minimal variance in parent-rated impairment in organized activities after controlling for oppositional symptoms.

Individual ADHD Symptoms and Impairment

Although research on the relation between ADHD symptom domains and ADHD presentation types provides some insight into the potential causes and correlates of ADHD-related impairments, the use of the 2 ADHD symptom domains or the ADHD presentation types fails to capture the full heterogeneity in the number, constellation, and severity of symptoms across individuals. That is, there is wide variability across patients, even within the ADHD presentation types, for ADHD symptom profiles, and it may be that specific ADHD symptoms within the ADHD symptom domains may specifically be related or not related to the various ADHD-related impairments. Research focusing on the relationship between individual symptoms of ADHD and specific impairment is sparse. Mota and Schachar10 assessed the sensitivity and specificity of specific ADHD symptoms in predicting global impairment in school-age children using both teacher and parent ratings of symptoms and impairment. Their results indicated that some ADHD symptoms were better predictors of global impairment than others, and those symptoms varied by rater. Specifically, the HI symptom “blurts out” was the most efficient (i.e., maximum sensitivity and specificity) predictor of global impairment when parent ratings were examined, but when teacher ratings were examined, the HI symptom “leaves seat” was the most efficient predictor of global impairment. This discrepancy in the relationship between symptoms and impairment for parent and teacher ratings underscores the importance of understanding the relationships between individual symptoms and impairment using both parent and teacher data separately. These findings provide some insight into the relationship between symptoms and impairment; however, a global measure of impairment was used, and thus, they were unable to assess the relationship between ADHD symptoms and specific domains of impairment.

Zoromski et al.7 examined the relationship between individual ADHD symptoms and various domains of impairment in an elementary school setting using teacher ratings. The IA symptoms “difficulty sustaining attention,” “avoids tasks,” and “difficulty organizing” were the symptoms most strongly associated with academic impairment. The HI symptoms “leaves seat” and “interrupts or intrudes” and the IA symptom “avoids tasks” were the symptoms most strongly associated with peer impairment. Finally, the HI symptoms “leaves seat” and “interrupts or intrudes” and the IA symptoms “does not seem to listen,” “difficulty sustaining attention,” and “difficulty organizing” were most strongly associated with classroom functioning. Some ADHD symptoms (e.g., “blurts out answers” and “easily distracted”) did not relate to any ADHD impairment domains. Zoromski et al.7 used only teacher ratings and had a limited range of impairment domains (i.e., those that occurred in the school setting). Furthermore, neither of the previous studies controlled for relevant demographic (e.g., sex11) and clinical characteristics (e.g., comorbidities12 and medication status13) although these variables have been shown to be related to both ADHD symptom and impairment presentation.7,10 For example, gender differences in the frequency of suspension for boys and girls have been found.11 In addition, stronger correlations between symptoms and impairment have been found for adults than children.14 Given that it is anticipated that individuals will encounter more opportunities for impairment with age (e.g., children in kindergarten have a much lower demand placed on their organization skills than fifth graders), it is reasonable that the association between ADHD symptoms and impairment may change within the childhood years.

Present Study

The aim of the current study was to conduct a preliminary exploration of the relationship between individual ADHD symptoms and functional impairment in elementary school-age children using the Vanderbilt ADHD Diagnostic Rating Scales.15 This study addresses the limitations of other studies (e.g., Mota and Schachar10) by using parent and teacher ratings of symptoms and impairment, includes demographic and clinical characteristics that have previously been shown to be related to ADHD symptoms and impairment, and explores the relationship between individual ADHD and a wide range of impairment domains in a much larger sample than has been previously used. Areas of impairment assessed in the current study include peer, sibling, and parent relationships, overall school performance, performance in specific subject areas (math, writing, and reading), classroom behavior, and participation in organized activities or teams. We expected to find a pattern in which specific ADHD symptoms related to specific domains of impairment. To our knowledge, no prior studies have assessed the relationship between individual ADHD symptoms and specific domains of impairment using parent ratings. Thus, hypotheses regarding the relationships between individual ADHD symptoms and various domains of impairment are based on the findings of Mota and Schachar.10 We hypothesize that the 2 ADHD symptoms that they found were related to parent ratings of global impairment (i.e., “not listening” and “blurts out”) will be related to multiple parent-rated domains of impairment. Specifically, we hypothesized that based on prior findings7,10, teacher ratings would suggest that specific IA symptoms (i.e., “difficulty sustaining attention,” “avoids tasks,” and “difficulty organizing”) would relate to academic domains, and specific HI symptoms (i.e., “leaves seat” and “interrupts or intrudes”) and an IA symptom (“avoids tasks”) would predict peer impairment. Moreover, specific ADHD symptoms in the IA (i.e., “does not seem to listen,” “difficulty sustaining attention,” and “difficulty organizing”) and HI domains (i.e., “interrupts or intrudes,” “leaves seat”) would predict classroom behavioral functioning.

METHOD

Participants

This study is a retrospective data analysis using data gathered from an attention-deficit hyperactivity disorder (ADHD) web portal software. The ADHD web portal is a platform in which parents, teachers, and health care providers all mutually input information about the patient, after which information is scored, interpreted, and formatted in a report style that is helpful to the health care provider in their assessment and treatment of patients with ADHD.16 Health care providers initiate an ADHD evaluation by registering families on the portal to facilitate the collection of parent and teacher rating scales. All measures were collected for the provider's assessment of ADHD. Users agreed that deidentified ratings inputted into the ADHD web portal can be used for research purposes. This project was approved by the local Institutional Review Board.

Participants were selected from a national sample of patients age 5 to 12 (n = 8689 patients; mean age = 8.18, SD = 1.74; 67.8% male) registered on the ADHD web portal between October 2008 and October 2015. To be included in this study, patients had to have at least one completed parent and one completed teacher rating. 7.9% of the sample had a parent-reported learning disability, 1.4% of the sample had a parent-reported autism or pervasive developmental disability diagnosis, and 7.7% of parents reported that their child took medication at the time they were assessed for ADHD.

Measures

Vanderbilt ADHD Diagnostic Rating Scales15

Parents (Vanderbilt ADHD Diagnostic Parent Rating Scale) and teachers (Vanderbilt ADHD Diagnostic Teacher Rating Scale) respectively rated symptoms of ADHD (18 items), oppositional defiant disorder (ODD) (Parent: 8 items; Cronbach's α = 0.92; Teacher: 10 items; Cronbach's α = 0.92), and internalizing disorders (7 items; Cronbach's α = 0.87; Teacher: 7 items; Cronbach's α = 0.86) on a numeric rating scale with anchors of “never” (0), “occasionally” (1), “often” (2), and “very often” (3). Total symptom scores (TSS) are created for each ADHD symptom domain (Inattention [IA]: 9 items; Hyperactivity/Impulsivity: 9 items) by summing the 9 item scores in each domain. Parents rated impairment across 8 domains (school, reading, writing, math, participation in activities and relationships with parents, siblings, and peers) and teachers rated impairment across 8 domains (reading, writing, math, relationships with peers, following directions, classroom behavior, assignment completion, and organizational skills; following directions, assignment completion, and organization skills were not used for the present study given their overlap with the ADHD symptom items) on a numeric rating scale with anchors of “Excellent” (1), “Above Average” (2), “Average” (3), “Somewhat of a Problem” (4), and “Problematic” (5). ADHD symptom and impairment ratings were used at the item level. Mean scores for the ODD and internalizing items were calculated for parents and teachers to be used as covariates (see Table 1 for descriptive statistics). Parents invited their child's teacher to complete the ratings on the portal.

Demographic Questionnaire

The demographic questionnaire is a short form completed by parents to provide information regarding their child's development, previous diagnosis, and treatment history. From this measure, parental report of age, sex, whether the child had been diagnosed with Learning Disorders or Learning Disabilities (yes or no), whether the child had been diagnosed with autism or pervasive developmental disorder (yes or no), and whether the evaluation was based on a time when the child was on medication (yes or no) were used in the current study.

Statistical Analyses

Given previous literature indicating demographic and clinical characteristics including age,17 sex,11 comorbidity (internalizing and oppositional symptoms),12 presence of a learning disorder,18 autism or pervasive developmental disorder diagnosis,19 and medication status13 may contribute to functional impairment, bivariate correlations were conducted between potential covariates representing each of these factors and the various impairment domains (Table S1 and Table S2, Supplemental Digital Contents 1 and 2, http://links.lww.com/JDBP/A285). These were conducted for parent-rated and teacher-rated impairment ratings separately. In addition, bivariate correlations were assessed between each individual ADHD symptom and domain of impairment (Table S1, Table S2, Supplemental Digital Contents 1 and 2, http://links.lww.com/JDBP/A285). Again, these were conducted separately for parent and teacher ratings.

Hierarchical regression was used to investigate which ADHD symptoms predicted each domain of impairment. Parent and teacher ratings were modeled separately. For each model, potential covariates that had a significant (p < 0.05) bivariate correlation with the impairment domain were entered simultaneously in the first step. Then, because of the sample size and the likelihood of finding significant relationships, symptoms that accounted for a minimum of 2% of the variance (i.e., at least a small effect size) in the impairment domain based on their bivariate correlation were entered into the model in the second step using stepwise entry. Although this approach has limitations,20 we chose it because we needed an approach that would allow us to identify the symptoms most associated with impairment from a group of symptoms that were correlated with each other. Hierarchical regression is well-suited to address this type of question. Finally, post hoc analyses were conducted to assess whether age moderated the relationship between an ADHD symptom and impairment. For every symptom that was a significant predictor of impairment in the primary analyses, a regression model was run that included the covariates including age, the ADHD symptom, and the ADHD symptom by age interaction for the corresponding domain of impairment. A significant age × ADHD symptom interaction would indicate that the relationship between a specific ADHD symptom and a specific ADHD-related impairment was moderated by age.

RESULTS

The range for parent and teacher ratings on the inattention (IA) and hyperactivity/impulsivity (HI) symptom domain scores on the Vanderbilt attention-deficit hyperactivity disorder (ADHD) Diagnostic Rating Scale15 was 0 to 27 (parent-reported IA TSS M = 18.37, SD = 5.48; parent-reported HI TSS M = 14.67, SD = 7.16; teacher-reported IA TSS M = 16.94, SD = 6.29; teacher-reported HI TSS M = 11.94, SD = 7.79).

Academic Impairments

Results of regression models across parent and teacher ratings of academic functioning (Table 2) indicated that the covariates accounted for between 4% and 7% of the variance in academic functioning domains. LD status had a significant positive association with impairment across all academic domains. When ADHD symptoms were added after accounting for the covariates, ADHD symptoms accounted for between 10% and 24% of the variance in academic functioning domains. The IA symptoms “makes careless mistakes” and “avoids tasks” significantly predicted academic impairment across areas of academic impairment and across both parent and teacher rating models. “Difficulty organizing” was also a strong significant predictor on both parent and teacher ratings of writing impairment. Across parent and teacher ratings of academic functioning, most age by symptom interactions were nonsignificant (Table 2). For parent-rated overall academic impairment, and parent- and teacher-rated impairment in math, there was a significant and positive age by symptom interaction for “careless mistakes.” This interaction suggests that the relation between “careless mistakes” and overall academic impairment as well as specific math impairment is stronger at older ages. In addition, there were significant negative age by symptom interactions for “difficulty organizing” and parent- and teacher-rated impairment in writing and for “careless mistakes” and parent-rated impairment in writing. These interactions suggest that the relationships between these specific symptoms and writing impairment are stronger at younger ages.

Table 2. - Results of Regression Models Examining Individual ADHD Symptoms Predicting Academic Impairment
School Reading Writing Mathematics
Parent β Teacher β Parent β Teacher β Parent β Teacher β Parent β
Covariates
 Age −0.09*** −0.08*** −0.09*** −0.09*** 0.01 0.03**
 Sex −0.02 0.06*** 0.09*** −0.18*** −0.17***
 Med status −0.01 −0.02* −0.02* 0.00 0.01
 LD 0.12*** 0.16*** 0.16*** 0.15*** 0.14*** 0.15*** 0.13***
 ODD 0.03 −0.13*** −0.08***
 Internalizing −0.03a 0.05*** −0.04a 0.05*** −0.02 0.06*** −0.02
 Autism/PDD 0.01 0.00
  R2 0.04*** 0.06*** 0.04*** 0.07*** 0.04*** 0.07*** 0.07***
Inattention
 Makes careless mistakes 0.28***b 0.26*** 0.23*** 0.35*** 0.26***c 0.27***b 0.28***b
 Difficulty sustaining attention −0.04a
 Does not seem to listen −0.08a −0.05a
 Does not follow through −0.07a −0.06a 0.04*
 Difficulty organizing 0.16***c 0.09***c 0.05**
 Avoids tasks 0.15*** 0.23*** 0.18*** 0.22*** 0.16*** 0.21*** 0.15***
 Loses things −0.08a −0.05a −0.12a
 Is easily distracted −0.05a
 Is forgetful 0.09*** −0.05a
Hyperactivity
 Fidgets or squirms −0.09a
 Leaves seat
 Runs about or climbs
 Difficulty playing quietly 0.03**
 Is “on the go”
 Talks too much
Impulsivity
 Blurts out answers
 Difficulty waiting
 Interrupts or intrudes
  R2 0.17*** 0.20*** 0.14*** 0.31*** 0.16*** 0.25*** 0.18***
  ΔR2 0.13*** 0.14*** 0.10*** 0.24*** 0.12*** 0.17*** 0.11***
*p < 0.05,**p < 0.01, ***p < 0.001.
aIndicates instances in which variables were statistically significant but in the opposite direction of their bivariate correlations. This was determined to be due to suppression in post hoc analyses.
bThe symptom by age interaction effect was significant and positive.
cThe symptom by age interaction effect was significant and negative.
ADHD, attention-deficit hyperactivity disorder; LD, learning disorder/learning disability, ODD, oppositional defiant disorder; PDD, pervasive developmental disorder.

Relational Impairments and Impairments in Organized Activities

Across parent and teacher ratings of relational functioning (parents, peers, and siblings; Table 3) and participation in organized activities (Table 4), covariates accounted for between 15% and 34% of the variance in functioning across these models. Medication status, oppositional symptoms, and internalizing symptoms were positively associated with impairment across all models. Of note, oppositional symptoms were the strongest predictor across parent-rated relationships with parents, parent-rated relationships with siblings, and both parent- and teacher-rated relationships with peers. Autism or pervasive developmental disorder diagnosis was positively associated with worse parent- and teacher-rated relationships with peers and participation in organized activities. After accounting for these covariates, ADHD symptoms accounted for between 0% and 6% of the variance in functioning in these areas. The HI symptom “difficulty playing quietly” was significant across all models. Across parent and teacher ratings of relational functioning and participation in organized activities, the majority of age by symptom interactions were nonsignificant (Tables 3 and 4). For parent-rated impairment in organized activities, there was a significant negative age by symptom interaction for “leaves seat,” which indicated that the strength of the symptom-impairment relationship was stronger at younger ages.

Table 3. - Results of Regression Models Examining Individual ADHD Symptoms Predicting Relational Impairment
Relationship with Parents Relationship with Siblings Relationship with Peers
Parent β Parent β Teacher β Parent β
Covariates
 Age 0.09*** −0.01 −0.01
 Sex −0.06*** −0.05*** 0.00 0.02
 Med status 0.06*** 0.05*** 0.05*** 0.06***
 LD 0.01
 ODD 0.51*** 0.51*** 0.39*** 0.27***
 Internalizing 0.07*** 0.03* 0.10*** 0.12***
 Autism/PDD −0.01 0.06*** 0.07***
 R2 0.31*** 0.29*** 0.34*** 0.20***
Inattention
 Makes careless mistakes
 Difficulty sustaining attention 0.07***
 Does not seem to listen 0.04*** 0.08*** 0.03**
 Does not follow through
 Difficulty organizing
 Avoids tasks
 Loses things 0.04***
 Is easily distracted
 Is forgetful
Hyperactivity
 Fidgets or squirms
 Leaves seat −0.05a
 Runs about or climbs
 Difficulty playing quietly 0.05*** 0.05** 0.10*** 0.09***
 Is “on the go” −0.08a −0.08a
 Talks too much −0.11a
Impulsivity
 Blurts out answers −0.10a
 Difficulty waiting 0.13*** 0.14***
 Interrupts or intrudes −0.03a 0.18***
  R2 0.31*** 0.30*** 0.40*** 0.22***
  ΔR2 0.004*** 0.001*** 0.06*** 0.02***
*p < 0.05,**p < 0.01, ***p < 0.001.
aIndicates instances in which variables were statistically significant but in the opposite direction of their bivariate correlations. This was determined to be due to suppression in post hoc analyses.
bThe symptom by age interaction effect was significant and positive.
cThe symptom by age interaction effect was significant and negative.
ADHD, attention-deficit hyperactivity disorder; LD, learning disorder/learning disability, ODD, oppositional defiant disorder; PDD, pervasive developmental disorder.

Table 4. - Results of Regression Models Examining Individual ADHD Symptoms Predicting Impairment in Behavior in Classroom and Organized Activities
Classroom Behavior Participation in Activities
Teacher β Parent β
Covariates
 Age −0.04*** −0.15***
 Sex 0.07*** 0.03*
 Med status 0.00 0.06***
 LD −0.01 0.01
 ODD 0.14*** 0.13***
 Internalizing 0.14***
 Autism/PDD 0.00 0.08***
 R2 0.30*** 0.15***
Inattention
 Makes careless mistakes
 Difficulty sustaining attention
 Does not seem to listen 0.05***
 Does not follow through
 Difficulty organizing 0.09***
 Avoids tasks
 Loses things 0.04***
 Is easily distracted 0.05**b
 Is forgetful −0.05a
Hyperactivity
 Fidgets or squirms 0.07***b
 Leaves seat 0.14***b 0.04**c
 Runs about or climbs
 Difficulty playing quietly 0.08***b 0.08***
 Is “on the go” −0.10a
 Talks too much 0.22***b
Impulsivity
 Blurts out answers 0.07***b −0.05a
 Difficulty waiting 0.10***b 0.15***
 Interrupts or intrudes 0.12***b
  R2 0.65*** 0.19***
  ΔR2 0.35*** 0.04***
*p < 0.05,**p < 0.01, ***p < 0.001.
aIndicates instances in which variables were statistically significant but in the opposite direction of their bivariate correlations. This was determined to be due to suppression in post hoc analyses.
bThe symptom by age interaction effect was significant and positive.
cThe symptom by age interaction effect was significant and negative.
ADHD, attention-deficit hyperactivity disorder; LD, learning disorder/learning disability, ODD, oppositional defiant disorder; PDD, pervasive developmental disorder.

Impairments in Classroom Behaviors

For teacher ratings of classroom behavior, covariates accounted for 30% of the variance (Table 4). The covariates accounting for the greatest amount of variance in classroom behavior were oppositional symptoms, followed by being male and age. After controlling for covariates, ADHD symptoms accounted for 35% of the variance in functioning related to classroom behaviors. The strongest predictor of teacher-related impairment related to classroom behavior was the HI symptom “talks too much.” For teacher-rated impairment in classroom behavior, there were significant positive age by symptom interactions for “is easily distracted,” fidgets or squirms,” “leaves seat,” “difficulty playing quietly,” “blurts out answers,” “talks too much,” “difficulty waiting,” and “interrupts or intrudes,” which indicated that the impact of these symptoms on impairment is stronger at older ages (Table 4).

DISCUSSION

Overall, our findings indicate a pattern whereby specific attention-deficit hyperactivity disorder (ADHD) symptoms are associated with specific areas of ADHD-related impairment. That is, beyond established relations between the 2 primary ADHD symptom domains (i.e., inattention [IA] and hyperactivity/impulsivity [HI]) and ADHD-related impairments, specific symptoms within the 2 ADHD symptom domains seem to be uniquely related to specific areas of ADHD-related impairment. Moreover, this pattern of ADHD symptom–ADHD impairment findings demonstrated convergence across parent and teacher ratings. The model in which ADHD symptoms accounted for the most variance in impairment was in the area of classroom behavior, suggesting that ADHD symptoms are more directly associated with this domain of behavior compared with the academic and relational domains.

For academic impairment, the IA symptoms “avoids tasks” and “makes careless mistakes” were significant predictors of impairment across a variety of academic domains including parent and teacher ratings of reading, writing, math, and overall school performance. It is notable that these symptoms accounted for more of the variance than the parent-reported presence of a prior learning disability. The significant association between “avoids tasks” and academic impairment was also reported by Zoromski et al.7 Moreover, Mota and Schachar10 found that “avoids tasks” predicted teacher-rated global impairment. “Makes careless mistakes” was not found to be related to teacher-rated academic impairments in prior studies.7 For both these symptoms, the full symptom description orients the rater specifically to academic tasks. The full symptom description for “makes careless mistakes” on the Vanderbilt ADHD Diagnostic Parent Rating Scale (VADPRS) is “Does not pay attention to details or makes careless mistakes with, for example, homework,” and on the Vanderbilt ADHD Diagnostic Teacher Rating Scale (VADTRS) is “Fails to give attention to details or makes careless mistakes in schoolwork.” The symptom “avoids tasks” refers to academics on the teacher VADTRS (“Avoids tasks (e.g., schoolwork, homework) that require sustained mental effort”) but does not do so on the parent VADPRS (“Avoids, dislikes, or does not want to start tasks that require ongoing mental efforts”). Nevertheless, both parent and teacher ratings on this item predicted academic impairments. Perhaps the wording of these items or the types of activities that these symptoms connote (e.g., homework situations for parents when rating “avoids, dislikes, or does not want to start tasks that require mental effort”) prime parents and teachers to focus on the child's academic performance, which might explain the reported associations between these items and parent and teacher ratings of academic impairments. In addition, if task avoidance is occurring, the opportunity to observe the manifestation of other ADHD symptoms (e.g., difficulty sustaining attention, careless mistakes, and following through) may be limited, which underscores the importance of intervening to address task avoidance.

Continuing with the academic domain, “difficulty organizing” predicted parent- and teacher-rated writing impairment and teacher-rated math impairment. Similarly, Zoromski et al.7 found that this symptom predicted teacher-rated academic impairment, and Mota and Schachar10 found that it predicted global ratings of impairment. The relationship between organizational skills and academic impairments is well-established. For example, organization and planning predicts GPA and homework problems (e.g., Langberg et al.21). Surprisingly, although a previous study found that “difficulty sustaining attention” predicted teachers' ratings of academic impairment,7 our study did not find any associations between this symptom and any parent- or teacher-rated domains of impairment. It is possible that this discrepancy might have occurred because in the present study, teachers rated impairment in specific academic areas (reading, writing, and math), whereas the previous study had teachers rate global academic impairment.7 Indeed, previous studies have demonstrated stronger relationships between ADHD symptom dimensions and global ratings of impairment than between ADHD symptom dimensions and specific measures of impairment14; thus, the present findings might have been influenced by the way impairment was measured.

Across the relational domains of impairment, we found, as has been reported elsewhere,6 oppositional symptoms are the strongest predictors of impairments in parent and peer relationships with children with ADHD. Indeed, the association between oppositional defiant disorder (ODD) symptoms and social functioning has been clearly documented. For example, parent-rated ODD symptoms during childhood and adolescence predict poor social functioning and romantic relationships in early adulthood even after controlling for symptoms of ADHD, conduct disorder, anxiety, and depression.22 Despite the large portion of variance accounted for by oppositional symptoms across relational domains, there were a few HI symptoms that accounted for additional, although modest, variance. “Difficulty playing quietly” was a significant predictor across all relational domains. The HI symptom “difficulty waiting” predicted parent- and teacher-rated impairment in relationships with peers. “Difficulty waiting” was also the strongest predictor of impairment in organized activities. It is possible that “difficulty waiting” was specifically associated with impairment in teacher-rated relationship with peers and organized activities in domains because children may encounter more situations in these domains in which waiting their turn is required (e.g., waiting in line at school and waiting turns in games). Generally, these findings align with the past research. “Interrupts or intrudes” has been shown to predict teacher-rated impairment in peer relationships7 and parent- and teacher-rated global impairment.10 Similarly, “difficulty waiting” predicted parent- and teacher-rated global impairment, and “difficulty playing quietly” predicted parent-rated global impairment.10 Furthermore, impulsivity has been shown to negatively predict peer relationships. Specifically, teacher-rated impulsivity (e.g., impatience, ability to wait, and acting without thinking) is associated with low preference ratings on peer sociometrics for children while controlling for sex, age, and cognitive abilities.23

Previous research also indicates that “leaves seat” and “avoids tasks” are strong predictors of teacher-rated impairment in peer relationships,7 but this was not supported in the present study. It is possible that differences in findings across studies may be due to controlling for ODD symptoms, and there may be shared variance between “leaves seat,” “avoids tasks,” and ODD symptoms.7 It should be noted that children may exhibit ODD-like behaviors (i.e., loses temper, touchy or easily annoyed, angry or resentful, argues with adults, actively defies, deliberately annoys others, blames others, and spiteful or vindictive) to avoid difficult or arduous tasks at home or school without having the “pattern of angry/irritable mood, argumentative/defiant behavior, or vindictiveness” that is characteristic of children with ODD.1 In addition, the present sample included children who were going through the process of an ADHD assessment and the prior study7 used a normative sample of students.

Previous studies found that individual IA symptoms may be related to peer relationships.7 Notably, the present study found that teachers' ratings had a stronger association between individual IA symptoms and relational impairments than parents' ratings. Indeed, “does not seem to listen,” “difficulty sustaining attention,” and “loses things” were all significantly associated with teacher-rated impairment in peer relationships. Alternatively, for parents, the only IA symptom that was significantly associated with relational impairment was “does not seem to listen,” which significantly predicted impairment in relationships with parents. Historically, there have been mixed findings regarding the relationship between IA and social impairment. For example, one study demonstrated that teacher-rated HI had a stronger association with social problems than IA.24 However, a study conducted in Taiwan found that parent ratings of IA symptoms, but not HI symptoms, were associated with lower peer acceptance and fewer dyadic friendships.25 Although it is possible that there may be cultural differences in perceptions of IA, the discrepancies between parent and teacher ratings may in part be due to the different contexts in which they observe the children. Teachers likely have more opportunities to observe children interacting with their peers than parents and thus may be more aware of more subtle social behaviors, whereas parents may only be more aware of the socially intrusive behaviors (i.e., the HI symptoms). Indeed, teachers' ratings tend to evidence more concerns with IA symptoms, whereas parents' ratings tend to evidence more concerns with all symptoms.10

Finally, we found that “leaves seat” and “talks too much” were the strongest predictors of teacher-rated impairment in classroom behavior. Similarly, Mota and Schachar10 found that teacher-rated “leaves seat” was the individual ADHD symptom that best discriminated children who were impaired from those who were not. Consistent with the findings of the present study, prior research has indicated that the behaviors elementary teachers perceive to be most disturbing in the classroom involve “breaking classroom rules,” which would likely include leaving seat and talking too much.26 In addition, both “inattention” and “classwork organization” were rated highly by teachers when asked to rate how disturbing they were in the classroom.26 This is consistent with the present findings that “easily distracted” and “loses things” had strong associations with teachers' ratings of impairment in classroom behaviors. Finally, it is notable that the amount of variance accounted for by ODD symptoms in predicting classroom behavior was much lower than the amount of variance accounted by ODD symptoms in relational impairment. This may have been because teachers perceive ODD symptoms to be more problematic in student to peer or student to teacher interactions, whereas hyperactivity symptoms (e.g., “leaves seat” and “talks too much”) are disruptive to the classroom as a whole.

The Association of Symptoms and Impairment with Age

Most interactions between age and symptoms were nonsignificant. This is likely due to the sample comprising elementary school-age children. In the models predicting parent-rated overall academic impairment and parent- and teacher-rated math impairment, the strength of the symptom-impairment relationship for “careless mistakes” seemed to be stronger at older ages. This may be because as the complexity of instructions and academic tasks increase there are greater opportunities for these types of errors or that they are more noticeable because careless errors are more typical for younger children. In the models predicting writing impairment, parent- and teacher-rated “difficulty organizing” and parent-rated “careless mistakes” were more impactful for younger children. This was an odd set of findings that is difficult to interpret, especially in the context of parent and teacher rating of problems with “writing.” Indeed, writing has multiple aspects (e.g., clarity of handwriting, spelling, punctuation, and creativity). It is possible that parents and teachers are reflecting on the quality of handwriting in young children when they rate problems with “writing.” Perhaps “disorganization” and “careless mistakes” are considered reflections of poor quality of handwriting in young children. It would be interesting to study associations between the ADHD symptoms and writing using a more fine-grained, well-defined assessment of multiple aspects of writing skills to better understand these relations. Finally, multiple HI symptoms affected teacher-rated classroom behavior more strongly at older ages. This may be because these symptoms are increasingly atypical as children develop, and thus, when these behaviors do occur, they may stand out more and be viewed as more problematic and more disruptive to the classroom environment.

Symptoms Lacking a Strong Relationship with Impairment

Several ADHD symptoms did not predict any domains of impairment. In particular, 2 HI symptoms, “runs about or climbs” and “is on the go,” were not significantly associated with impairment across any of the impairment domains for parent or teacher ratings. Of note, mean scores for teacher ratings on these symptoms were lower than the means of any other teacher-rated symptoms, and mean scores for parent ratings on these symptoms were also low in comparison to most other parent rated symptoms. Thus, it is possible that the lack of significant impairment associations for these symptoms may be due to limited variance in symptom ratings (i.e., a floor effect). Prior research has also not found relations between these symptoms (i.e., “runs about or climbs” and “is on the go”) and ADHD-related impairments in elementary school-age children.7,10 Although “runs about or climbs” and “is on the go” do not seem to predict ADHD-related impairments, they have demonstrated high positive predictive power for ADHD diagnostic status.27 Past decisions to retain specific ADHD symptoms in the diagnostic criteria have been based on the results of factor analyses of the symptoms as well as studies examining predictive power27 without determining whether the individual symptoms are associated with impairment. Reliance on factor analytic and psychometric properties of the symptoms may have resulted in the inclusion of some symptoms that have limited associations with ADHD-related impairment.7 It may also be that these symptoms may have stronger associations with impairment in different age groups (e.g., preschoolers, adolescents, and adults), when alternative domains of impairment (e.g., occupational functioning) are measured, or possibly had we measured the domains of impairment differently (e.g., peer sociometrics as a measure of social functioning).

Limitations

It is important to interpret these findings in the context of the limitations of the present study. First, because of the method with which these data were collected using the web portal, certain relevant demographic information (e.g., race and socioeconomic status) was unavailable. This limits the extent to which we can discuss the potential generalizability of these findings. Furthermore, we did not have data regarding whether children had Individualized Education Plans or were receiving behavioral intervention, and it is possible that these factors could have independently influenced ratings of symptoms or impairment. In addition, when caregivers completed ratings through the ADHD web portal, they identified themselves as the primary caregiver but did not identify their relationship to the child (e.g., mother or father), so it remains unclear whether mother versus father report moderated the relationship between individual ADHD symptoms and impairment. The use of either teacher ratings or parent ratings as both the independent and dependent variables on a single measure has the potential to create associations between measures because of common method bias (i.e., common variance across items because of the shared measurement method) rather than reflecting associations between the constructs intended to be measured. Furthermore, although the Vanderbilt ADHD Diagnostic Rating Scales have important strengths compared with measures used in the past studies of the relationship between symptoms and impairment including the provision of domain-specific ratings of impairment, these domains only comprise single items. Universal measures of impairment (e.g., peer sociometric ratings, number of disciplinary referrals, and grades) were unavailable for the current sample, and it is possible that the patterns of findings may have differed if they had been used. The Vanderbilt ADHD Diagnostic Rating Scales also only include items to screen for internalizing disorders and oppositional symptoms rather than the full set of Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition symptoms, and learning disability and autism or pervasive developmental disorder diagnoses were based on parent report rather than formal evaluation. Finally, although our predictors accounted for significant variance in parent and teacher ratings of a variety of child impairment, there was variance that remained unaccounted for in our models. This may be due to factors that we did not assess in the present study such as instructional quality, adverse childhood events, behavioral interventions, or parental discord.

Clinical Implications

The findings from the present study provide an improved understanding of the relationships between individual ADHD symptoms and impairment in children and have the potential to help providers understand what domains of functioning children are likely to be impaired based on symptom presentation. Furthermore, this enhanced understanding of the relationship between symptoms and impairment may facilitate targeted early intervention efforts to address impairments that are highly related to specific symptoms a child has before the impairments become problematic. For example, given that “avoids tasks” was a strong predictor of impairment across academic domains, classroom-based strategies to address this symptom may be especially important. Specifically, the use of small group instruction,28 which has been shown to be associated with increased on task behavior in children with ADHD, may help decrease task avoidance. In addition, these findings underscore the importance of behavioral classroom management interventions that can address avoidance of work and provide external motivation for work completion such as a Daily Report Card, an evidence-based intervention for ADHD in which children have specific goals that they are working toward and receive frequent feedback about during the day that is typically linked to home or school-based rewards.29 Finally, the apparent association between organization and academic impairment, especially in the subject area of writing, continues to support the use of organizational skills interventions, which have shown benefits for children with ADHD (e.g., Evans et al.29).

Regarding social functioning, the strength of the association between oppositional symptoms and impairment might be an important consideration when attempting to address social impairments in children with ADHD. Social skills training interventions for social impairment in youth with ODD have not been shown to be effective to address difficulties with social functioning.30 In addition, there is some evidence that children with ADHD and comorbid ODD benefit less from social skills training interventions.31 Thus, it may be important for social skills interventions for children with ADHD to target ODD symptoms. Furthermore, given that evidence-based interventions to address ODD rely heavily on caregiver involvement (e.g., Eyberg et al.32) and the most promising social functioning interventions for children with ADHD have included components that teach parents or teachers to foster generalization of skills learned, including parents and/or teachers as part of the social skills intervention may be critical to their success.30

REFERENCES

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5 ed. Washington DC: Author; 2013.
2. Willcutt EG, Nigg JT, Pennington BF, et al. Validity of DSM-IV attention deficit/hyperactivity disorder symptom dimensions and subtypes. J Abnorm Psychol. 2012;121:991–1010.
3. Clendenin AA, Businelle MS, Kelley ML. Screening ADHD problems in the sports behavior checklist: factor structure, convergent and divergent validity, and group differences. J Atten Disord. 2005;8:79–87.
4. Graetz BW, Sawyer MG, Hazell PL, et al. Validity of DSM-IVADHD subtypes in a nationally representative sample of Australian children and adolescents. J Am Acad Child Adolesc Psychiatry 2001;40:1410–1417.
5. Gadow KD, Nolan EE, Litcher L, et al. Comparison of attention-deficit/hyperactivity disorder symptom subtypes in Ukrainian school children. J Am Acad Child Adolesc Psychiatry 2000;39:1520–1527.
6. Garner AA, O'Connor BC, Narad ME, et al. The relationship between ADHD symptom dimensions, clinical correlates, and functional impairments. J Dev Behav Pediatr. 2013;34:469–477.
7. Zoromski AK, Owens JS, Evans SW, et al. Identifying ADHD symptoms most associated with impairment in early childhood, middle childhood, and adolescence using teacher report. J Abnorm Child Psychol. 2015;43:1243–1255.
8. Kadesjo C, Hagglof B, Kadesjo B, et al. Attention-deficit-hyperactivity disorder with and without oppositional defiant disorder in 3- to 7-year-old children. Dev Med Child Neurol. 2003;45:693–699.
9. Frankel F, Feinberg D. Social problems associated with ADHD vs. ODD in children referred for friendship problems. Child Psychiatry Hum Dev. 2002;33:125–146.
10. Mota VL, Schachar RJ. Reformulating attention-deficit/hyperactivity disorder according to signal detection theory. J Am Acad Child Adolesc Psychiatry 2000;39:1144–1151.
11. Bauermeister JJ, Shrout PE, Chavez L, et al. ADHD and gender: are risks and sequela of ADHD the same for boys and girls? J Child Psychol Psychiatry 2007;48:831–839.
12. Ostrander R, Herman K, Sikorski J, et al. Patterns of psychopathology in children with ADHD: a latent profile analysis. J Clin Child Adolesc Psychol. 2008;37:833–847.
13. Langberg JM, Becker SP. Does long-term medication use improve the academic outcomes of youth with attention-deficit/hyperactivity disorder? Clin Child Fam Psychol Rev. 2012;15:215–233.
14. Barkley RA, Cunningham CE, Gordon M, et al. ADHD symptoms vs. impairment: Revisited. ADHD Rep. 2006;14:1–9.
15. Wolraich ML, Lambert W, Doffing MA, et al. Psychometric properties of the Vanderbilt ADHD diagnostic parent rating scale in a referred population. J Pedia Psychol. 2003;28:559–567.
16. Epstein JN, Langberg JM, Lichtenstein PK, et al. The myADHDportal. com Improvement Program: an innovative quality improvement intervention for improving the quality of ADHD care among community-based pediatricians. Clin Pract Pediatr Psychol. 2013;1:55.
17. Biederman J, Mick E, Faraone SV. Age-dependent decline of symptoms of attention deficit hyperactivity disorder: impact of remission definition and symptom type. Am J Psychiatry 2000;157:816–818.
18. Capano L, Minden D, Chen SX, et al. Mathematical learning disorder in school-age children with attention-deficit hyperactivity disorder. Can J Psychiatry 2008;53:392–399.
19. Factor RS, Ryan SM, Farley JP, et al. Does the presence of anxiety and ADHD symptoms add to social impairment in children with autism spectrum disorder? J Autism Developmental Disord. 2017;47:1122–1134.
20. Whittingham MJ, Stephens PA, Bradbury RB, et al. Why do we still use stepwise modelling in ecology and behaviour? J Anim Ecol. 2006;75:1182–1189.
21. Langberg JM, Dvorsky MR, Evans SW. What specific facets of executive function are associated with academic functioning in youth with attention-deficit/hyperactivity disorder? J Abnorm Child Psychol. 2013;41:1145–1159.
22. Burke JD, Rowe R, Boylan K. Functional outcomes of child and adolescent oppositional defiant disorder symptoms in young adult men. J Child Psychol Psychiatry 2014;55:264–272.
23. Gomes L, Livesey D. Exploring the link between impulsivity and peer relations in 5- and 6-year-old children. Child Care Health Dev. 2008;34:763–770.
24. Gaub M, Carlson CL. Gender differences in ADHD: a meta-analysis and critical review. J Am Acad Child Adolesc Psychiatry 1997;36:1036–1045.
25. Tseng WL, Kawabata Y, Gau SS, et al. Symptoms of attention-deficit/hyperactivity disorder and peer functioning: a transactional model of development. J Abnorm Child Psychol. 2014;42:1353–1365.
26. Safran JS, Safran SP. Teachers' judgments of problem behaviors. Except Child. 1987;54:240–244.
27. Frick PJ, Lahey BB, Applegate B, et al. DSM-IV field trials for the disruptive behavior disorders: symptom utility estimates. J Am Acad Child Adolesc Psychiatry 1994;33:529–539.
28. Hart KC, Massetti GM, Fabiano SV, et al. Impact of group size on classroom on-task behavior and work productivity in children with ADHD. J Emot Behav Disord. 2010;19:55–64.
29. Evans SW, Owens JS, Wymbs BT, et al. Evidence-based psychosocial treatments for children and adolescents with attention deficit/hyperactivity disorder. J Clin Child Adolesc Psychol. 2018;47:157–198.
30. Mikami AY, Smit S, Khalis A. Social skills training and ADHD-what works? Curr Psychiatry Rep. 2017;19:93.
31. Antshel KM, Remer R. Social skills training in children with attention deficit hyperactivity disorder: a randomized-controlled clinical trial. J Clin Child Adolesc Psychol. 2003;32:153–165.
32. Eyberg SM, Nelson MM, Boggs SR. Evidence-based psychosocial treatments for children and adolescents with disruptive behavior. J Clin Child Adolesc Psychol. 2008;37:215–237.
Keywords:

ADHD symptoms; impairment; academic functioning; social functioning

Supplemental Digital Content

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.