Date of Award


Document Type

Campus Access Dissertation


Epidemiology and Biostatistics



First Advisor

Robert E McKeown


Attention-Deficit Hyperactivity Disorder (ADHD) affects an estimated 5-10% of children and adolescents in the United States and is characterized by excessive inattention, hyperactivity, and impulsivity which cause impairment in multiple settings, including at home, in school, and with friends. Partially due to medication, behavioral modifications, and lifestyle changes, not all children will meet diagnostic criteria in adulthood. However, most children with ADHD are likely to continue to display symptoms into adolescence to an extent inappropriate for their age. The specific aims of this project were to (1) characterize the persistence of ADHD symptomotology from childhood through adolescence, (2) characterize the nature of ADHD-related functional impairment from childhood through adolescence, and (3) determine which ADHD symptoms or combination of symptoms are most responsible for at least moderate functional impairment in six domains of daily life.

This project utilized data from the South Carolina Project to Learn about ADHD in Youth (SC PLAY), a community-based sample from a large suburban school district. After a two-stage screening process, child-parent dyads at high risk of ADHD were invited for interview, as well as a sample of low-risk peers, frequency-matched on sex. Four hundred eighty-one children and their parents were successfully recruited for initial interview. Beginning three years later, all traceable participants were invited for three waves of annual interviews. In total, 353 participants (73.4%) were interviewed at least once between three and seven years after their baseline interview, and 195 participants (41%) completed all four time points.

At each interview, parents completed the Diagnostic Interview Schedule for Children version four (DISC-IV), which is structured according to Diagnostic and Statistical Manual of Mental Disorders version four (DSM-IV) diagnostic criteria. Data on all 18 ADHD symptoms were collected via DISC-IV, as well as information regarding functional impairment in six domains of daily life: (1) getting along with parents/caretakers, (2) participating in family activities, (3) participating in peer activities, (4) academic functioning, (5) relationships with teachers, and (6) distress attributable to symptoms. ADHD medication data and probable diagnoses for nine other common psychiatric disorders were collected via DISC-IV and parent-reported demographic data were collected via questionnaire.

Descriptive statistics were calculated and multivariate logistic regression models were fit using SAS-callable SUDAAN version 10.0. Logistic quantile regression modeling employed bootstrapping and was performed using R's QUANTREG package. All R-based and SUDAAN-based analyses properly accounted for weighting and multistage, stratified study design as well as intra-individual correlation between data collection waves. Statistical weights were used in all analyses to reflect differential sampling as well as non-response.

Prevalence of individual inattentive symptoms ranged from 18% to 45% among those in early childhood and 19% to 41% among those in late adolescence, and most of these were reported in fairly similar proportions across developmental stages. The prevalence of individual hyperactivity-impulsivity symptoms ranged from 17% to 44% among those in early childhood and 6% to 16% among those in late adolescence. After adjustment for sex, ADHD medication use, and presence of psychiatric disorder, inattention symptom count did not differ by age, but the highest 25% of hyperactive-impulsive symptom counts did decrease by age (p<0.01).

The prevalence of at least low functional impairment due to inattention increased from early childhood through late adolescence, while that due to hyperactivity-impulsivity decreased. Children with severe inattention-associated impairment at Time 1 were more likely than those with low or moderate impairment to have at least some inattention-associated impairment later (p=0.05). Counts of domains in which at least moderate functional impairment due to inattention existed were consistent across developmental stages, while those due to hyperactivity-impulsivity tended to decrease as participants aged, much like the pattern found of ADHD symptoms themselves. After adjustment for sex, ADHD medication use, and whether both biological parents lived in the home, the decrease in at least moderate functional impairment due to hyperactivity-impulsivity remained marginally significant (p=0.07).

Most symptoms within each diagnostic category displayed moderate correlation (0.4< ö <0.6) in this sample. While some symptoms only appeared to independently affect one or two domains, other symptoms (such as `fails to follow through on instructions' and `difficulty awaiting turn') were associated with each of the six domains measured.

Characterizing the course of ADHD in the community is a basic but vital piece of understanding the extent of the societal problem. This knowledge is necessary to define prevention strategies at early developmental stages, to pursue appropriate target age groups for behavioral interventions, to inform new educators on what to expect within the classroom, to educate parents in raising affected youth, and to assist clinicians when considering behavioral therapy and decreased dosage of ADHD medication as a patient ages. Findings from this project add substantially to the understanding of the course of ADHD in the community, but further research is required to update the epidemiologic understanding of ADHD prevalence and persistence into and through adulthood.