Functional Status in Children With ADHD at Age 6–8: A Controlled Community Study
OBJECTIVES: To examine the functional status (mental health, academic performance, peer problems) of a community-based sample of children who have attention-deficit/hyperactivity disorder (ADHD) and non-ADHD controls, and to investigate gender and subtype differences.
METHODS: Children aged 6 to 8 years were recruited through 43 Melbourne schools, using a 2-stage screening (parent and teacher Conners 3 ADHD index) and case confirmation (Diagnostic Interview Schedule for Children, Version IV; [DISC-IV]) procedure. Outcome measures were mental health disorders (DISC-IV), academic performance (Wide Range Achievement Test 4), and peer problems (Strength and Difficulties Questionnaire). Unadjusted and adjusted linear and logistic regression were used to compare ADHD and non-ADHD controls.
RESULTS: A total of 179 children who have ADHD and 212 non-ADHD controls were recruited. Compared with controls, children who had ADHD had higher odds of externalizing (odds ratio [OR], 11.0; 95% confidence interval [CI], 5.6–21.6; P < .001) and internalizing (OR, 2.9; 95% CI, 1.2–7.2; P = .02) disorders; poorer reading (effect size, −0.66) and mathematics (effect size, −0.69) performance; and more peer problems (P < .001). Boys and girls who had ADHD were equally impaired. Only 17% of children in our ADHD group had been previously diagnosed. Previous diagnosis was higher in the Combined group and for boys.
CONCLUSIONS: In their second year of school, children who had ADHD performed worse than controls across all functional domains, yet only a minority had been formally diagnosed with ADHD. Findings highlight the need for earlier diagnosis and intervention.
- ADHD —
- attention-deficit/hyperactivity disorder
- AMD —
- adjusted mean difference
- ASD —
- autism spectrum disorder
- CI —
- confidence interval
- DISC-IV —
- Diagnostic Interview Schedule for Children, Version IV
- DMDD —
- Disruptive Mood Dysregulation Disorder
- DSM —
- Diagnostic and Statistical Manual of Mental Disorders
- OR —
- odds ratio
- SEIFA —
- Socioeconomic Indexes for Areas Disadvantage Index
What’s Known on This Subject:
Children who have attention-deficit/hyperactivity disorder (ADHD) attending clinical services have poorer outcomes in adolescence on a range of measures. However, it is unknown how early in development these impairments appear, particularly for community-ascertained samples.
What This Study Adds:
At age 6 to 8 years, children in the community with ADHD have significantly poorer mental health, academic performance and social function compared with control children. Children who have impairing ADHD symptoms should be referred early for assessment and intervention.
Attention-deficit/hyperactivity disorder (ADHD) is associated with social difficulties,1 academic underachievement,2 and mental health comorbidities.3,4 Current evidence regarding functional impairments in children who have ADHD is predominantly drawn from clinical samples, with a resulting over-representation of boys and children who have more severe ADHD,5 and exclusion of those who have the Inattentive subtype.6 Population-based studies are scarce, and have often used designs with limited capacity to describe the typical functional status (eg, ADHD diagnosis not confirmed,7 wide age range at recruitment).8 We address these gaps using baseline data from a community cohort study of children aged 6 to 8 years, who were screened and confirmed as meeting Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria for ADHD, compared with children confirmed to not have ADHD.
Little is known about when functional deficits emerge for children who have ADHD, or which patient subgroups and characteristics carry the highest risk for impairments. There is some evidence that children who have ADHD are at risk for poorer functioning from early life. Children who have ADHD exhibit more behavioral and social problems than controls at preschool age,9 and poorer language and cognitive performance compared with controls in early elementary school.10 Some impairments become more marked as the skill deficits that accompany ADHD are exposed by increasingly complex environmental demands. For example, older children who have ADHD have more social problems than younger children who have ADHD.11
Several studies have examined relationships between ADHD symptoms and functional status in community samples. Rodriguez et al reported a significant negative association between ADHD symptoms and academic skills in a large population sample (n = 13 087; age range, 7–12 years).12 Merrell and Tymms found an inverse relationship between teacher ratings of ADHD symptoms and academic achievement in a large (n = 4148) representative English sample.13 ADHD symptoms were associated with peer dislike in a Swedish study of 12-year-old schoolchildren (n = 635).14 Limitations of these studies included not confirming ADHD diagnosis nor ascertaining comorbidities or recruiting controls for comparison.
Much ADHD research has excluded children who have the Inattentive subtype. Although children who have the Combined and Inattentive subtypes of ADHD demonstrate similar neuropsychological deficits,15 the Inattentive subtype tends to be diagnosed later, include more girls, and have more learning disabilities.16 Weiss et al found that children who have Inattentive ADHD had less functional impairment than those who have the Combined subtype, although their academic achievement was relatively poor and they had more internalizing comorbidities.16 Others have found that the Combined subtype carries a higher risk for both externalizing and internalizing problems.17 Studies of gender differences in the functional status of children who have ADHD have also produced mixed results. Clinical samples of girls who have ADHD are reported to be similarly impaired compared with boys, but girls who have ADHD from non-referred samples have been found to have relatively less inattention, internalizing behavior, and peer aggression than boys.18
The Children’s Attention Project,19 a longitudinal community-based cohort study of children who have ADHD and controls, was designed to address these limitations. In this article we report baseline data on the functional status of recruited children aged 6 to 8 years who underwent comprehensive phenotyping. We aimed to:
Compare comorbid internalizing disorders, externalizing disorders, academic performance, and peer problems between children who have ADHD and non-ADHD controls.
Compare the same outcomes across ADHD subtypes (Combined, Inattentive, and Hyperactive/Impulsive) and between boys and girls.
We hypothesized that children who have ADHD would have poorer functioning across all domains.
Baseline data were collected across 2 consecutive years (2011–2012). Study approval was granted by the Human Research Ethics Committees of the Royal Children’s Hospital, Melbourne (#31056), and the Victorian Department of Education and Early Childhood Development (#2011_001095). Our study methodology has been previously described.19
Study Design and Recruitment
Participants were recruited from 43 government elementary schools in metropolitan Melbourne, Australia. Schools were recruited via the Victorian government Department of Education regions selected for representation of diverse socioeconomic communities. Participants were recruited via a 2-stage screening and case-confirmation procedure.
Stage 1: Screening for ADHD. The Conners 3 ADHD Index20 was distributed to the parents of all second grade children (age 6–8 years) in participating schools. This grade was chosen as it is a developmental stage at which children who have ADHD often begin to manifest functional impairments. Earlier sampling would result in missed cases, and later sampling would reduce opportunities to capture early influences on developmental trajectories. On receipt of parental consent, the child’s teacher completed the same measure.20
Boys scoring ≥75th percentile and girls scoring ≥80th percentile on the ADHD Index on both parent and teacher report (based on age and gender), as well as any child reported by the parent as having an ADHD diagnosis, were defined as screening positive for ADHD. A higher cut-point was used for girls, as our pilot data showed that this resulted in better correspondence with diagnostic confirmation.
Children were defined as screening negative for ADHD if their score on the ADHD Index was <75th percentile (boys) or <80th percentile (girls) by both parent and teacher report and there was no parent-reported diagnosis of ADHD.
Stage 2: Diagnostic confirmation and baseline data collection. Each positive screen was randomly matched on gender and school to a negative screen. Parents of children screening positive and the matched negative screens were invited to participate in the longitudinal study, which at baseline includes diagnostic interviews with parents, direct child assessments, and detailed parent and teacher surveys. Assessments were administered by trained research assistants with at least a 4-year degree in psychology, blinded to child screening status.
Children who had intellectual disability, severe medical conditions, genetic disorders, moderate-severe sensory impairment, and neurologic problems were excluded, as were children whose parents had insufficient English to complete the interviews or questionnaires.
ADHD case confirmation and mental health comorbidities were assessed using the National Institute of Mental Health’s DISC-IV21 conducted face-to-face with parents. Children were classified as having an internalizing disorder if they met criteria for separation anxiety disorder, social phobia, generalized anxiety disorder, post-traumatic stress disorder, obsessive-compulsive disorder, hypomania, or manic episode, and an externalizing disorder if they met criteria for oppositional defiant disorder or conduct disorder. We adapted the method of Copeland22 to derive a proxy for DSM-V Disruptive Mood Dysregulation Disorder (DMDD) using items from the DISC-IV. To be classified as having DMDD, children needed to have temper tantrums occurring most days and to be often “grouchy or irritable and often in a bad mood” (Criteria A–D). Symptoms needed to be present for at least 12 months (Criterion E), and to be present at school, assessed using the “often loses temper” question from the teacher Strengths and Difficulties Questionnaire (see below; Criterion F).
Academic achievement was assessed using the word reading and math computation subtests of the Wide Range Achievement Test 4.23 Word reading measures letter and word decoding through letter identification and word recognition, whereas math computation measures the ability to perform basic mathematical computations through counting, number identification, oral problem-solving, and written problem calculation. Age-based standard scores were derived for all measures (normative mean [SD] = 100 ).
Social functioning was assessed by using the 5-item parent- and teacher-reported peer problems scale of the Strengths and Difficulties Questionnaire.24 Established clinical range cut-points were used to analyze these data categorically.24
Other sample characteristics examined were previous diagnoses (ADHD, autism spectrum disorder [ASD]) and medication use by parent report. Cognitive level was assessed using the Vocabulary and Matrix Reasoning subtests (normative mean [SD] = 50 ) of the Wechsler Abbreviated Scale of Intelligence.25 Neighborhood socioeconomic disadvantage was measured by the Socioeconomic Indexes for Areas Disadvantage Index (SEIFA) for the child’s postcode of residence (mean [SD] = 1000 ; higher scores reflect less disadvantage).26 We also measured primary carer age, education, single parent status, and clinical levels of mental health problems (Kessler 6; cut-point ≥13 for clinical problem).27
Sample characteristics were compared between cases and controls using t tests for continuous variables and χ2 tests for categorical variables. Linear (for continuous outcomes) or logistic regression (for binary outcomes) was used to compare the functioning of children who had ADHD to controls on each outcome domain, using unadjusted and adjusted models (Aim 1). A priori confounders were child gender, parent education level (did not complete high school, completed high school, completed university degree), clinical levels of parent mental health difficulties, and single parent status. In addition, internalizing and externalizing comorbidities were adjusted for when these were not the outcome examined. Effect sizes were calculated by standardizing outcome variables to have a mean of 0 and an SD of 1.
To quantify functional impairments, we created a yes/no variable for impairment in each domain of interest: internalizing disorder, externalizing disorder, <25th percentile on either word reading or math computation, and clinical range on either parent- or teacher-reported peer problems. We used descriptive statistics and χ2 tests to compare the number of functional impairments (0–4) for children who had ADHD and controls.
Linear and logistic regression were used to compare functioning across ADHD subtypes and by gender for each outcome (Aim 2). Predictors of previous ADHD diagnosis (ADHD subtype, gender) were examined using logistic regression. Analyses were completed using Stata 13.0 software (Stata Corp, College Station, TX).
Of 5992 eligible children, 3734 (62%) had complete parent and teacher screening data (Fig 1). Children who had complete screening data were from relatively more advantaged areas (SEIFA) than those who had missing data. Of those screening positive, 65% (267/412) were eligible and consented to participate in the longitudinal study and 179 met criteria for ADHD. Of the matched negative screens, 55% (231/412) consented and 212 did not meet criteria for ADHD on the DISC-IV. Children screening negative who met criteria for ADHD, along with children screening positive who did not meet criteria for ADHD, are being followed as a high-risk group. Consenting positive screens were from relatively more advantaged areas compared with non-consenting positive screens (P = .02); there were no differences in social advantage for consenting versus non-consenting negative screens (P = .11).
Sample Characteristics (Table 1)
The ADHD cohort included 93 Combined, 64 Inattentive, and 22 Hyperactive-Impulsive subtype children. Children who had ADHD had lower cognitive functioning and were more likely to have parent-reported ASD compared with controls. Children who had ADHD were more likely to live in single-parent families, and their primary carers were less likely to have completed high school and had higher levels of psychological distress than primary carers of controls.
Functional Differences Between Children Who Had ADHD and Non-ADHD Controls (Table 2)
Children who had ADHD were more likely to meet criteria for both externalizing (adjusted OR, 11.0; 95% CI, 5.6 to 21.6; P < .001) and internalizing (adjusted OR, 2.9; 95% CI, 1.2 to 7.2; P = .02) disorders than controls. The most prevalent comorbid mental health disorders in the ADHD group were oppositional defiant disorder (54.2%), DMDD (22%), separation anxiety disorder (13.4%), and conduct disorder (10.1%). Children who had ADHD had poorer standard scores than controls for word reading (effect size, −0.67) and math computation (effect size, −0.69). Peer problems were higher in ADHD than control children by both parent and teacher report (effect size, 0.92 and 0.69, respectively).
Number of Impairments
Children who had ADHD were more likely than controls to have multiple impairments (χ2 = 169.8; P < .001). The proportions with 2, 3, or 4 impairments were 30.2% vs 5.7%, 24.6% vs 0.9%, and 11.7% vs 0%, respectively. Only 11% of children who had ADHD had no functional impairments, compared with 57% of non-ADHD controls. Within the ADHD group, children who had Combined subtype were more likely to have multiple impairments than those who had Inattentive subtype (81.7% vs 43.8% with 2 or more impairments), but there were no differences by gender.
Differences Between Boys and Girls Who Had ADHD
In adjusted analyses, girls (reference group) and boys who had ADHD did not differ at the 5% level of significance with respect to rates of internalizing (adjusted OR, 1.1; 95% CI, 0.5 to 2.4; P = .86) or externalizing disorders (adjusted OR, 0.8; 95% CI, 0.3 to 1.7; P = .50). Similarly, there were no significant gender differences in word reading (adjusted mean difference [AMD] 1.2; 95% CI, −4.6 to 7.0; P = .69; effect size, 0.07) or math computation (AMD, 4.0; 95% CI, −0.8 to 8.8; P = .10; effect size, 0.26). Boys who had ADHD had more parent-reported peer problems than girls who had ADHD (AMD, 0.7; 95% CI, 0.03 to 1.4; P = .04; effect size, 0.36), although there was little difference by teacher report (AMD, 0.1; 95% CI, −0.16 to 0.8; P = .82; effect size, 0.04).
Differences by ADHD Subtype (Table 3)
Children who had Combined subtype had higher rates of internalizing (OR, 3.2; 95% CI, 1.4 to 7.3) and externalizing disorders (OR, 3.4; 95% CI, 1.7 to 7.0) than children who had Inattentive subtype, however, these relationships attenuated in adjusted analyses. There was little difference in academic performance or social functioning by subtype.
Predictors of Previous Diagnosis
Only 31 (17%) children in the ADHD group had a previous diagnosis of ADHD and 23 (13%) were taking medication for ADHD. Previous diagnosis was more frequent in boys than in girls (22% vs 5%; OR, 5.2; 95% CI, 1.6 to 17.5; P = .007) and in those who had Combined type (24%) compared with those with Inattentive (9%) and Hyperactive-Impulsive (14%) type (overall P = .05).
This community-based study identified multiple functional impairments in a cohort of children aged 6 to 8 years who had ADHD. Children who had ADHD were found to be performing well below their non-ADHD peers on all domains assessed. They had a high burden of mental health comorbidities, were performing markedly worse academically, and had poorer peer relations.
Rates of identification of mental health comorbidities may be lower in clinical practice than in research studies.28 In contrast to many previous studies that have described mental health symptoms using parent report on questionnaires, we report robust mental health comorbidities ascertained by structured parent interview. Our findings suggest that comorbidities are identifiable by early primary school age if symptoms are inquired about systematically. This is important, as children who have ADHD and comorbidities respond optimally to a combination of medication and behavioral interventions.29 Our finding of a high prevalence of ASD in our ADHD cohort is consistent with recent literature,30 and is of interest given that DSM-V now permits the concurrent diagnoses of ADHD and ASD.31 Consistent with previous research,32 peer problems were elevated in our ADHD group, with large effect sizes observed. Peer problems in children who have ADHD do not generally respond to ADHD treatment,33 and require specific intervention.
Differences in academic performance between children who had ADHD and controls in this study were striking, equating to nearly a full SD for both literacy and numeracy. DuPaul et al found that children aged 3 to 5 years who had ADHD (n = 58) scored approximately 1 SD lower than controls (n = 36) on pre-academic skills.9 Masetti et al tracked academic achievement over 8 years in 125 children who had symptoms of ADHD at age 4 to 6 years.34 They found that children who had Inattentive subtype had lower reading, spelling, and mathematics scores than matched non-referred controls and children who had other subtypes. By contrast, in our study, subtype differences in academic performance were not detected. This may be a function of the younger age of our sample, as the inter-subject discrimination of academic measures becomes more sensitive with age.
The academic performance findings are likely to have multiple contributing factors. Firstly, the ADHD sample had lower estimated IQ scores than controls. Furthermore, a range of executive function deficits have been identified in children who have ADHD, with inconsistent findings across studies. Language problems are common in children who have ADHD and are strongly associated with poor academic functioning.35 Finally, learning disorders are known to co-occur with ADHD in up to 45% of cases.36 These children are doubly disadvantaged in relation to learning, as attention problems identified in kindergarten-age children have been shown to predict reading difficulties in later primary school.37
We did not find significant differences by ADHD subtype in academic or social functioning, nor differences in mental health function by subtype after adjusting for confounders. We found, however, that the overall burden of impairments was higher in the Combined subtype group, suggesting that this subgroup has the most difficulties within the ADHD population at this stage of life.
Boys and girls who had ADHD in our study had similar rates of mental health comorbidities, in contrast with most previous research. Two previous meta-analyses on gender differences in children who have ADHD have found that girls are less likely to exhibit externalizing comorbidities, although they presented conflicting findings in relation to the risk for comorbid internalizing disorders.18,38 The lack of gender differences in academic performance or peer problems is consistent with previous studies.
Few children identified with ADHD in this non-referred sample had received a previous diagnosis of ADHD. These findings are consistent with data from a study of help-seeking among children identified with ADHD symptoms in a Californian elementary school district.39 Despite similar impairment profiles girls were less likely to have been diagnosed than boys, highlighting this gap in access to care.
This study design had a number of strengths. We recruited and carefully phenotyped a large number of children in a narrow age band, and so were able to clearly document function at this developmental stage. We sampled boys and girls across the sociodemographic spectrum, included all subtypes, and did not exclude children who had comorbidities. Finally, we examined a broad range of functional outcomes (to the extent that 43% of controls had at least 1 impairment).
Our study also had some limitations. There were potential sample biases, as families excluded because of incomplete screening data were relatively socially disadvantaged compared with participating families, and the rate of consent in our control group was lower than in our cases. Furthermore, our measure of peer functioning was brief.
Sonuga-Barke has suggested that in early life children who have ADHD might comprise heterogeneous subgroups with different developmental pathways and risk profiles.40 A better understanding of the factors that influence impairments is needed to develop targeted interventions to improve outcomes for children who have ADHD. Community-based ADHD studies overcome some of the limitations of previous research in this field.
We acknowledge all research assistants, students, and interns who contributed to data collection for this study. We would also like to thank the many families, teachers, and schools for their participation in this study. Some study data were collected and managed using Research Electronic Data Capture (REDCap) electronic data capture tools hosted at MCRI. REDCap is a secure, web-based application designed to support data capture for research studies.
- Accepted July 28, 2014.
- Address correspondence to Daryl Efron, MD, FRACP, Royal Children’s Hospital, Flemington Rd, Parkville, Victoria 3052, Australia. E-mail:
Dr Efron conceptualized and designed the study and drafted the initial manuscript; Dr Sciberras contributed to the conception and design of the study, carried out the analyses, reviewed and revised the manuscript, and provided critical input; Drs Anderson and Hazell contributed to the conception and design of the study, reviewed and revised the manuscript, and provided critical input; Dr Ukoumunne contributed to the conception and design of the study, reviewed and revised the manuscript, and provided statistical input; Dr Jongeling contributed to the conception and design of the study and reviewed and revised the manuscript; Mr Bisset and Ms Schilpzand coordinated the study, collected study data, and reviewed and revised the manuscript; Dr Nicholson contributed to the conception and design of the study, reviewed and revised the manuscript, and provided critical input; and all authors approved the final manuscript as submitted.
FINANCIAL DISCLOSURE: Dr Hazell or his employer has received payment from Shire for participation in advisory boards and Eli Lilly and Shire for speaker’s bureau; the other authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: The Children’s Attention Project is funded by an Australian National Health and Medical Research Council (NHMRC) project grant (100852). This project has also received funding from the Collier Foundation and Murdoch Childrens Research Institute (MCRI). Dr Efron’s position is funded by a Career Development Award from MCRI. Dr Sciberras is funded by an NHMRC Early Career Research Fellowship (1037159). Dr Anderson is supported by an NHMRC Practitioner Fellowship (607333). Dr Ukoumunne is supported by the Peninsula Collaboration for Leadership in Applied Health Research and Care, a collaboration between the University of Exeter, University of Plymouth, and National Health Service South West, funded by the National Institute for Health Research. MCRI is supported by the Victorian Government's Operational Infrastructure Support Program.
POTENTIAL CONFLICT OF INTEREST: Dr Hazell's department has received revenue from Eli Lilly, Janssen, and Shire; he also serves on a Shire advisory board. The other authors have indicated they have no potential conflicts of interest to disclose.
- Booster GD,
- Dupaul GJ,
- Eiraldi R,
- Power TJ
- Weiss M,
- Worling D,
- Wasdell M
- Shaffer D,
- Fisher P,
- Lucas CP,
- Dulcan MK,
- Schwab-Stone ME
- Wilkinson GS,
- Robertson GJ
- Wechsler D
- Census of Population and Housing
- American Psychiatric Association
- Andrade BF,
- Tannock R
- Sciberras EMK,
- Efron D,
- Bisset M,
- et al
- DuPaul GJ,
- Gormley MJ,
- Laracy SD
- Gershon J
- Copyright © 2014 by the American Academy of Pediatrics