Understanding the implications of ADHD diagnoses patterns requires information about the predictors and outcomes of receiving a diagnosis of ADHD in childhood. This study investigated whether receiving a diagnosis of ADHD is linked to better or worse longitudinal socio-economic outcomes for Irish children with high hyperactivity/inattention symptoms.
Data were used from Growing Up in Ireland – the National Longitudinal Study of Children (GUI),* which is an ongoing longitudinal study that monitors the physical, psychological and social development of a nationally representative sample of Irish children. The analysis presented here focuses on a cohort of 9-year-olds recruited to the study†; two waves of data from this cohort were collected when the children were aged 9 years (between 2007‒2008) and aged 13 years (between 2011‒2012). Trained interviewers administered questionnaires to the children and their primary caregiver (in 97.8% of cases this was the child’s mother). Teachers used both self-completed and computer-assisted personal interviewing techniques. For this study, two groups were constructed to compare 9-year-olds who held a formal diagnosis of ADHD (Diagnosed ADHD group) with children who showed high hyperactivity/inattention symptoms but had not received a diagnosis (Undiagnosed ADHD group).‡
In total, 71 children (74.8% male) were included in the Diagnosed ADHD group and 582 (72.1% male) were included in the Undiagnosed ADHD group; 7915 children served as Non-ADHD Controls (49.4% male).
Comparison of the Diagnosed ADHD and Undiagnosed ADHD groups showed there were no significant differences in the distribution at Wave 1 for gender (72.1% vs 74.8%, respectively; p = 0.64); two-carer households (1‒2 children: 31% vs 23.4%, respectively; ≥3 children: 40.2% vs 41.4%, respectively; p = 0.25 each); social class (p = 0.91); parental Irish citizenship (92.4% vs 96.1%, respectively; p = 0.3); parental education (p=0.85); or equivalised income (p = 0.63).
On a scale from 1 (very healthy) to 4 (almost always unwell), parents were asked to rate their child’s general health in the past year. In Wave 1, the Diagnosed ADHD group (mean [standard deviation (SD)] 1.65 [0.59]) scored significantly (p < 0.001) worse than the Undiagnosed ADHD group (mean [SD] 1.36 [0.53]); however, this difference was not present at Wave 2 (p = 0.63).
From the Diagnosed ADHD group, 13.4% had been diagnosed in the last 6 months, 14.1% 6‒12 months ago, 19.7% 1‒2 years ago and 52.8% >2 years ago. Parent-rated hyperactivity/inattention scores were significantly higher for the Diagnosed ADHD group vs the Undiagnosed ADHD group (p < 0.001); however, there was no difference between groups for teacher ratings (p = 0.59).
In Waves 1 and 2, 50% and 52.6% of parents of children from the Diagnosed ADHD group had discussed their child’s health with a professional such as a psychologist, psychiatrist or counsellor more than once, compared with 7.1% and 5.8% in the Undiagnosed ADHD group and 2% and 2.3% in the Non-ADHD Control group, respectively. This was statistically significant between the Diagnosed and Undiagnosed ADHD groups for Waves 1 and 2 (p < 0.001 each).
Compared with the Undiagnosed ADHD group, a higher proportion of children in the Diagnosed ADHD group reported receiving support from a psychiatrist (0.5% vs 25.3%), a school psychologist (6% vs 12.4%) and an out-of-school psychologist (2.5% vs 27.8%) in Wave 2. Of the Diagnosed ADHD Group, 41.1% of parents who reported that their child had a mental health condition or disability had received a prescription or any emotional or behavioural problems.
In Wave 1, children from the Diagnosed ADHD group had significantly poorer numerical (p = 0.001) and verbal reasoning (p = 0.017) than the Undiagnosed ADHD group; however, there was no significant difference between groups in either numerical (p = 0.053) or verbal reasoning (p = 0.39) by Wave 2.
For Wave 1, parent-rated Strengths and Difficulties Questionnaire (SDQ) scores showed that when controlling for hyperactivity/inattention symptoms, general health, cognitive ability and service engagement, there was no significant difference in SDQ scores between the Diagnosed and Undiagnosed ADHD groups (p = 0.06). There was also no significant difference between the groups’ self-concepts (p = 0.38).
When controlling for hyperactivity/inattention symptoms, general health, cognitive ability and service engagement in Wave 2, parent-rated SDQ scores significantly differed between groups (p < 0.001). Follow-up univariate analysis confirmed that the Diagnosed group showed poorer scores on the Emotional (p = 0.01), Peer Relationships (p < 0.001) and Prosocial Behaviour (p = 0.02) scales; however, there was no significant difference for Conduct (p = 0.56). Separate univariate tests indicated that the Undiagnosed ADHD group had more favourable self-concept in the domains of Behavioural Adjustment (p = 0.004), Popularity (p = 0.01) and Freedom from Anxiety (p = 0.04). The groups did not differ on Physical Appearance & Attributes (p = 0.74), Intellectual & School Status (p = 0.77) or Happiness & Satisfaction (p = 0.30).
This study had several limitations. The relatively small proportion of children included in the Diagnosed ADHD group may have compromised the power of the analysis to detect differences between study groups at Wave 1, and this may have been attributed to parental under-reporting of recorded ADHD diagnoses. In addition, families with children with behavioural dysfunctions may have been less likely to opt into the study, even though the GUI coverage was high, with approximately one in seven 9-year-olds included nationally. The low prevalence of ADHD in this study may have led to limitations in the statistical power of the analysis; however, prevalence rates may have reflected a culturally conservative approach to childhood psychiatric diagnosis in Ireland.
The authors concluded that children diagnosed with ADHD may fare worse on dimensions of self-concept, social relations and emotional well-being in comparison with their peers who have similar ADHD symptomatology but no diagnosis. Further research should be done to understand whether this association could be triggered by the diagnosis itself or other confounding factors such as the child’s biological, familial or educational background.
*The aim of GUI is to investigate the distribution and longitudinal socio-emotional implications of ADHD diagnosis in an Irish community cohort sample. The analysis has three objectives: (1) to establish the socio-demographic profiles of 9-year-olds with diagnosed and undiagnosed ADHD symptoms in an Irish community sample; (2) to extract relevant information about these two cohorts’ health, service use and cognitive ability; and (3) to determine if the diagnosed and undiagnosed cohorts differed on longitudinal socio-emotional outcomes, controlling for symptom severity and other relevant variables
†Potential participants were identified using a stratified sampling strategy based on the primary school system. This achieved a household response rate of 57% that represented ~15% of the total population of Irish 9-year-olds
‡For children to be included in the Diagnosed ADHD group, parents must have reported that their child had a clinical diagnosis of ADHD at Wave 1. Children in the Undiagnosed ADHD group, had symptoms of ADHD based on the Strengths and Difficulties Questionnaire (SDQ) hyperactivity/inattention scores. Children with a score of ≥9 on either the Wave 1 parent- or teacher-rated hyperactivity/inattention subscale and whose parent had not reported a formal diagnosis of ADHD were included in the Undiagnosed ADHD group
O’Connor C, McNicholas F. What differentiates children with ADHD symptoms who do and do not receive a formal diagnosis? Results from a prospective longitudinal cohort study. Child Psychiatry Hum Dev 2019; Epub ahead of print.