Attention-Deficit Hyperactivity Disorder Across the Lifespan

In 1902, George Still (1) published one of the first clinical descriptions of what today would be recognized as attention-deficit hyperactivity disorder (ADHD). The first treatment of impulsive, hyperactive, and disruptive behavior with stimulant medication was reported in the 1930s (2). Virginia Douglas (3) first suggested that “attention deficits,” rather than hyperactivity, were the core symptoms of the disorder. Since then, the criteria have undergone refinements leading to the current terminology of ADHD in DSM-5 (Box 1).

ADHD is a neurodevelopmental disorder in which the ability to attend and/or control impulses (including inhibiting motor activity when appropriate) (a) is significantly less than that of a typically developing individual of a given age; (b) causes impairment in the individual’s academic, employment, or social functioning; and (c) is not better accounted for by some other medical or psychiatric condition. Box 1 presents the specific criteria, including the definition of the three presentations of ADHD: inattentive, hyperactive-impulsive, and combined. The only major change in the criteria from DSM-IV is the change in the age of onset, which now requires some symptoms to be present before age 12 years, rather than age 7 years. The individual symptom descriptions have been broadened to also include behaviors typical of an adolescent or adult, rather than just a young child.

Epidemiology

Looking at ADHD among children, a review and meta-analysis suggested a worldwide prevalence rate of 5.3% (4), but recent data suggest that the rate of diagnosis is increasing. The 2011 National Survey of Children’s Health researchers performed telephone interviews of nearly 100,000 parents of children ages 4–17 years (5). Eleven percent of children had received a diagnosis of ADHD at some point in their lives, 8.8% were reported as having a current diagnosis of ADHD, and 6.9% were taking medication for ADHD. Epidemiological studies suggest a prevalence of 4.4% for ADHD among adults (6).

Comorbidity

Both oppositional defiant and conduct disorders, as well as anxiety disorders, affect 25%−33% of children with ADHD; learning and language disorders affect another one-quarter of children with ADHD (7). Many children with ADHD will have two or more comorbid disorders, complicating clinical management. Compared with children with ADHD alone, those with oppositional defiant/conduct disorders show more severe symptoms of impulsivity, higher rates of aggression, a greater prevalence of learning disorders, and a greater propensity to develop both antisocial personality and substance abuse during their adolescent years (8). Compared with those with ADHD alone, individuals with comorbid anxiety (without oppositional defiant/conduct disorders) show lower levels of impulsivity on laboratory measures of attention as well as a greater tendency to respond to psychosocial interventions (9–11).

In the Multimodality Treatment of ADHD study (12), 11% of the sample also met criteria for major depressive disorder. How depression affects the clinical expression of ADHD has not been explored; family studies suggest that ADHD and major depressive disorder might share genetic factors (13). Because treatment of depression rarely leads to remission of ADHD symptoms, it seems unlikely that major depressive disorder masquerades as ADHD to any significant degree. If a child with ADHD meets the full DSM-5 criteria for major depressive disorder, this most likely represents the full syndrome of major depressive disorder and not a state of demoralization as a result of the ADHD symptoms (14). As individuals mature into adulthood, the comorbidity of mood and anxiety disorders remains problematic (6).

Etiology and Risk Factors

Approximately 71%−90% of the variance in ADHD traits is found to be attributable to genetics (15). Heritability estimates include the effects of gene-environment interactions; thus, the high heritability rates in ADHD do not minimize the effect of environment. Genome-wide association studies to date have not yet revealed any gene variant that passes the very high statistical threshold for genomewide significance (16). Maternal smoking during pregnancy and prenatal/perinatal adversity have been established as risk factors for ADHD (17–19). Children with ADHD exposed to smoking during pregnancy have more severe behavioral problems, lower IQ, and poorer neuropsychological test performance than nonexposed children with ADHD, even when controlling for income level, ethnicity, maternal age, and maternal alcohol use (20). Severe head injury can result in ADHD, even when preinjury ADHD diagnosis is controlled for (21, 22).

Pathophysiology

Cognitive Deficits

Recent research has suggested multiple cognitive deficits of children with ADHD (23, 24). Working memory deficits robustly delineate individuals with ADHD from those in a control group, particularly central executive working memory deficits such as updating of working memory, manipulating information in working memory, and mental manipulation of temporal order (25). Differential response to reward and delay aversion have also been proposed as key deficits in ADHD (23, 26). That is, children with ADHD either cannot tolerate a delay to wait for an anticipated reward or are hyporesponsive to the reward, making their behavior difficult to shape by the normal reinforcements and punishments in the environment. Despite these findings, many children with ADHD show no deficits on either neuropsychological or laboratory testing of cognition (27).

Neuroanatomical Findings

A meta-analysis of anatomical MRI studies showed reduced volume of multiple brain regions in persons with ADHD versus controls (28). Over the last several decades, the National Institute of Mental Health intramural program has performed MRI scans of children with ADHD and matched controls in a longitudinal design (29, 30). These studies have shown a global cortical maturational delay in the development of both cortical surface area and cortical thickness relative to control participants. In control groups, surface area rises through childhood, peaks around age 12 years, and then declines as pruning progresses. A study by Shaw et al. (29) showed that individuals with ADHD had less cortical surface area than those in the control group at study entry and were delayed approximately 2 years in reaching peak area, with the effect more pronounced on the right cortex. Cortical thickness followed a similar pattern (peaking in early adolescence and then declining). Individuals with ADHD have a thinner cortex at baseline than controls and prune later; interestingly, decreased pruning (resulting in thickness more similar to controls) is associated with remission of ADHD symptoms in adulthood (30).

Functional Neuroimaging

Functional studies in ADHD have been performed principally using tasks assessing response inhibition (such as the Go/No Go or Stop Signal Task), as well as tasks measuring attention. In a meta-analysis of these data, Hart et al. (31) found that relative to a control group, patients with ADHD showed reduced activation during response inhibition in the right inferior frontal cortex, supplementary motor area, and anterior cingulate cortex (ACC), as well as in striatothalamic areas. For attention tasks, patients with ADHD showed reduced activation relative to the control group for attention in the right dorsolateral prefrontal cortex, posterior basal ganglia, and thalamic and parietal regions. Another meta-analysis by Rubia et al. (32) of 14 functional magnetic resonance imaging (fMRI) data sets involving 212 children showed that acute methylphenidate treatment in ADHD consistently enhanced right inferior frontal cortex/insula activation.

Recent fMRI work has shifted from the study of discrete regions activated by tasks to resting state networks (33). A key area of study is the default mode network consisting of the precuneus/posterior cingulate cortex, the medial prefrontal cortex, and lateral/inferior parietal cortex. This circuit is active in task-irrelevant, “daydreaming” activities and is turned off when the attention circuits are activated. A recent large-scale, resting-state fMRI study compared 481 control participants with 276 adolescents with ADHD (34). Sripada et al. (34) found ADHD to be associated with a lack of anticorrelation between the default mode network and both the ventral attention (bottom-up) and frontoparietal (top-down) networks.

As noted earlier, patients with ADHD have difficulty delaying gratification and often seem unresponsive to reward. For patients in a control group, anticipation of a reward leads to increased dopamine in the ventral striatum; on fMRI, these can be detected as an increase in signal in the ventral striatum. Plichta and Scheres (35) reviewed fMRI studies comparing patients with ADHD with control subjects using reward tasks; they found that individuals with ADHD had hyporesponsiveness of the ventral striatum during anticipation of reward. This would suggest that individuals with ADHD are not as motivated as control participants by ordinary rewards. Family members often note that both children and adults with ADHD are hard to get motivated for day-to-day tasks.

Relative to control groups, adolescents with ADHD have greater amygdala activity when looking at subliminal angry faces and greater amygdala-cortical connectivity (36). Children with ADHD who have high ratings of emotional lability were found to have greater connectivity between the amygdala and ACC than those in a control group or children with ADHD who were not emotionally labile (37). These two studies suggest that patients with ADHD (particularly those with mood dysregulation) have an abnormally increased influence of the limbic system on cortical functions.

Course and Prognosis

Results of the Montreal Children’s Hospital Study, a longitudinal study of hyperactive children into adulthood, revealed that 50% still showed symptoms of inattention, impulsivity, low self-esteem, and social skills deficits (38). Barkley et al. (39) and Fischer et al. (40) followed up with 158 children with ADHD 8–10 years after diagnosis and compared them on a large number of functional measures with a control group of 81 children. Not only did 83% of the children with ADHD still meet criteria for the disorder in their late teen years, but they were more likely than the control group to have car accidents (41), to smoke cigarettes (39), and to have failed a grade or been suspended (40). Biederman et al. (42) compared boys with ADHD with matched controls 16 years after initial diagnosis. Individuals with ADHD had significantly higher rates of mood, antisocial, anxiety, and addictive disorders; moreover, they had more psychosocial impairments than did the boys in the control group. The Multimodality Treatment of ADHD study showed that one-third of a sample of 7- to 9-year-old children still met criteria for ADHD 8 years later and that the patients with ADHD fared poorly on multiple outcome measures, including lower academic progress, more psychiatric hospitalizations, and more police contacts and arrests (43). Type of treatment (medication versus psychosocial) in the first year of the study did not predict outcome, and nearly two-thirds of the sample was not taking medication at the 8-year follow-up. Substance abuse also was more common in patients with ADHD than in controls (44). Despite the fact that ADHD is a chronic condition, the Multimodality Treatment of ADHD study showed that most individuals receive no or sporadic treatment through adolescence, limiting the potential effect on long-term outcomes.

Adults with a childhood history of ADHD have higher-than-expected rates of antisocial behavior (45), injuries and accidents (46), employment and marital difficulties, health problems, teen pregnancies, and children out of wedlock (47). A 33-year prospective follow-up of 135 boys with ADHD and matched control participants was recently completed (48). There were 15 deaths in the ADHD group relative to five deaths in the control group; 10 of the 15 deaths in the ADHD group were due to suicide, homicide, or accident. Risky sexual behavior, traffic tickets, and accidents were more prevalent in the ADHD group, with comorbid conduct disorder predicting poor outcome.

Clinical Evaluation

Interview

Table 1 illustrates the major aspects of the clinical interview across the lifespan. The central theme of the interview process is to identify and quantify the specific symptoms of ADHD, keeping in mind the various developmental expressions of these symptoms. When possible, documentation of early life impairment should be obtained from adults presenting with ADHD. The clinician should screen for all of the major comorbidities (anxiety, depression, and disruptive behaviors) and rule out psychosis or severe mood disorder, which would require primary treatment.

Psychological Testing

Achievement and IQ testing to rule out learning disabilities is not mandatory before making a diagnosis of ADHD. Such testing may yield the most valid results after the symptoms of inattention have been controlled. If a child’s academic performance does not improve with control of ADHD symptoms or if the developmental history or mental status examination yields evidence of language or motor delays, then IQ and achievement testing should be performed, because the child most likely has a comorbid language/learning disorder.

Medical Evaluation and Laboratory Tests

The patient’s medical history should be obtained, as well as a physical examination within the last year. Patients with a history of cardiovascular disease or significant cardiac symptoms should have consultation with a primary care physician or cardiologist prior to starting stimulant medication. There is no need for cardiac screening (i.e., ECG) of otherwise healthy individuals.

Pharmacotherapy

Stimulants

The pharmacotherapy of ADHD with both stimulants and nonstimulants is well established by hundreds of double-blind placebo-controlled trials dating back to the 1960s (49). Dosing and titration of these agents is shown in Table 2. Response rates to stimulants can be as high as 90% when both methylphenidate and amphetamine are fully titrated (50). Although preschoolers may have a slightly decreased rate of response relative to school-age children (51), adults show a response equally robust to that of school-age children and adolescents (52, 53). Short-term side effects of stimulants include decreased appetite, insomnia, and headaches, whereas mood changes and tics are much rarer and may be idiosyncratic (49). Long-term effects on height, up to about 2 cm, have been found in long-term studies (54). A nationwide survey found a small but significantly increased rate of cardiovascular side effects (broadly defined), but no deaths, among patients with ADHD treated with stimulants, a finding consistent with an earlier large-scale study (55, 56).

Atomoxetine

This nonstimulant agent is a noradrenergic reuptake blocker that has some indirect agonism on dopamine. Numerous studies (57) show that atomoxetine is superior to placebo in the treatment of ADHD for children and adolescents; however, in one major study, atomoxetine was less efficacious than long-acting methylphenidate (58). Some patients may not see its full therapeutic effect until after a month of treatment (59).

Alpha-Agonists

The alpha-2 receptor agonists clonidine and guanfacine have varied effects on noradrenergic function. In recent years, long-acting versions of both agents have been the focus of studies either as monotherapy or as an add-on to stimulant treatment (60). Both alpha-agonists are superior to placebo for treatment of ADHD and oppositional defiant symptoms; adding extended-release versions of clonidine or guanfacine to a stimulant significantly improves ADHD symptoms over the addition of placebo (61, 62).

Psychosocial Interventions

Behavior therapy is the only psychosocial intervention validated in the treatment of ADHD (63). General principles of behavior therapy in the treatment of ADHD are as follows: information about the nature of ADHD, learning to attend more carefully to the child’s misbehavior and to when the child complies, establishing a home token economy, using time out effectively, managing noncompliant behaviors in public settings, using a daily school report card, and anticipating future misconduct. Occasional booster sessions often are recommended.

Conclusions

The clinical presentation, epidemiology, and acute pharmacotherapy of ADHD are now well established. Behavior therapy is an important adjunct treatment, particularly for partial responders or those with comorbid conditions. At least one-third of children with ADHD continue to meet criteria for the disorder as adults, with many more suffering some degree of impairment. Although genetics plays a major role in the disorder, specific genes for ADHD have not been discovered. Neuroimaging is not yet useful as a diagnostic marker or in tracking treatment. Future research is needed to further understand the interaction of genetic and environmental factors involved in both etiology and outcome; such understanding may lead to new pharmacological and psychosocial interventions.