Attention-deficit/hyperactivity disorder

Attention-deficit/hyperactivity disorder is a childhood-onset neurodevelopmental disorder of hyperactivity, inattention and impulsivity, and it is contested at almost every point: what it costs, what causes it, whether it is a disease at all, how it is diagnosed and counted, how well and how safely its treatments work, and whether it is diagnosed too often or too rarely.

Definition and clinical features

Attention-deficit/hyperactivity disorder is a highly disruptive childhood-onset disorder, characterised by inappropriate hyperactivity, inattention and impulsivity that begins before the age of seven and often persists into adolescence and adulthood. ¹ It presents in adults as well as in children. ² It is one of the most common childhood disorders and the most common neurodevelopmental disorder of childhood, estimated as of 2012 to affect 1.4 to 6% of school children, and probably the most controversial. ¹ In 2012 it was controversial for various reasons: its cause was unknown, its diagnosis subjective, and the long-term efficacy and safety of its treatment unclear. ³

Its major clinical features are severe impulsiveness, lack of concentration and motor hyperactivity, observed in at least two distinct settings, such as home, school and social settings. ¹ It is often accompanied by other psychiatric and developmental disorders, including learning impairments, oppositional defiant disorder, conduct disorder, autism and dyslexia. ¹ These impairing symptoms put children at risk of educational failure and can severely disrupt family, teacher and peer relationships. ¹ It can be a serious neurological illness that enormously affects the daily life of a family. ²

History

It was first described in 1798 by Alexander Crichton and became widely known outside the medical profession in 1846, with the story of Fidgety Phil (Zappelphilipp) in Heinrich Hoffmann's Struwwelpeter. ¹ Scientists and clinicians have struggled to understand its causes ever since. ¹ Children such as Fidgety Phil were chided for being naughty, and questions of cause and possible treatment that were not raised then are now asked under the diagnosis. ⁴

Costs and long-term outcomes

From a public-health perspective its cost is significant in academic underachievement, later unemployment, and the substance abuse and antisocial behaviour that often accompany it. ¹ In England and Wales children with the disorder place a significant cost on health, social and education services, reaching £23 million for initial specialist assessment and £14 million annually for follow-up care. ¹ In 2006 the total annual cost of prescribed stimulants and other drugs for it in England was roughly £29 million, and in 2012 the costs were still rising. ¹ Left untreated, it may persist into adulthood with a variety of behavioural, social and economic problems, including depression and anxiety disorders, antisocial behaviour, poor peer relationships, substance abuse or misuse, learning disabilities, low academic attainment and unemployment. ³

Causes

Given its impact on society, and particularly on children's quality of life, understanding its aetiology and pathophysiology is imperative, yet in 2012 little was known about either its causes or its mechanisms. ⁵ The often heartrending stories of those diagnosed and their relatives underline the need to understand its pathogenesis in order to develop effective preventive and symptomatic interventions. ³ In 2012 substantial progress continued to be made in understanding its aetiology and pathophysiology, particularly from genetic, neurophysiological and neuroimaging studies, and this may help not only to improve its diagnosis and treatment but also to develop and implement preventive strategies in the near future. ³ Because the number of diagnosed cases had recently been increasing, such improvements will have a large impact on health economics globally. ³

Genetic factors

Family, twin and adoption studies provided strong evidence that it is hereditary. ⁵ In twin studies, disease concordance was much higher in identical than in non-identical twins, and inherited factors explained 60 to 90% of the phenotypic variance. ⁵ There is evidence of inherited factors shared with a wide range of psychiatric disorders. ⁵ Prevalence is higher in males than in females, and in 2012 a genetic explanation for this was lacking. ⁵ The high heritability estimates are similar to those for schizophrenia and autism, and, as with those disorders, genetic effects are not 100%, which indicates additional contributions from non-shared environmental factors, epigenetic effects, random events or measurement inaccuracies. ⁵ Like other common medical and psychiatric disorders, it is considered a complex genetic trait, influenced by multiple genes, non-inherited factors and the interaction between them. ⁵

Because heritability is high, much effort has gone into searching for specific susceptibility genes by identifying common DNA variation. ⁵ Whole-genome linkage studies did not point to regions harbouring susceptibility genes, probably because there are no common susceptibility genes of large effect. ⁵ Candidate gene association studies have been more promising, and a small number have consistently withstood replication and meta-analysis, including genes in the dopaminergic pathway long hypothesised to be involved: the 7-repeat allele of the D4 dopamine receptor gene (DRD4), a microsatellite repeat in the D5 dopamine receptor gene (DRD5), and a 480 bp variable number tandem repeat in the dopamine transporter gene (DAT1). ⁵ Considerable sample heterogeneity has been reported for the DAT1 allele, which could result from multiple polymorphisms in that gene. ⁵ There is also evidence that the gene encoding a protein that degrades dopamine (COMT) could modify the phenotype. ⁵ In 2012 genome-wide association studies were at an early stage and had provided some interesting genes to investigate further, but all association studies had failed to find a common gene variant and had not supported the earlier candidate genes, probably because the small effects expected require extremely large samples, and because of sample heterogeneity. ⁵

Another possibility is that the disorder may be better explained by rare genetic variants, such as rare copy number variants. ⁵ A UK study found a significantly higher rate of rare copy number variants in cases than in controls, and an overlap with the copy number variants found in schizophrenia and autism, which further supports the disorder as a neurodevelopmental disorder. ⁵

Molecular genetic studies at best account for less than 5% of the estimated heritability of its symptoms, in large part because the clinical phenotype and the genetic architecture are heterogeneous. ⁵ Future directions therefore include dividing subjects into more homogeneous subgroups for genetic studies and using intermediate phenotypes or endophenotypes, such as magnetic-resonance-based measures of effects on brain structure, which measure simpler traits likely to be influenced by fewer genes. ⁵ Some magnetic resonance imaging studies provide evidence of differences in brain structure or function that may facilitate linkage studies and provide neurobiological mechanisms for how gene variants act on the brain. ⁵

Environmental factors

A number of environmental risk factors have been associated with the disorder. ⁶ The major associations are with maternal prenatal risk (alcohol, smoking, drug use and stress in pregnancy), with pregnancy and birth complications including prematurity and low birth weight, and with environmental exposures including toxins (pesticides, polychlorinated biphenyl and lead) and some viral infections. ⁶ Some studies have found a positive association with an agent, such as low-level prenatal organochlorine exposure with ADHD-like behaviours in childhood, but in 2012 no firm conclusion could yet be drawn for a link to behavioural outcome, except in extreme situations including extreme prematurity, very low birth weight and foetal alcohol syndrome. ⁶ Similarly, despite many studies of diet and its symptoms, in 2012 there was no evidence yet that diet plays a causal role, although some nutritional changes may help relieve some symptoms in diagnosed children. ⁶ Adverse social and family environments have also been associated with the disorder, but by 2012 none had been found to be causal except extreme early deprivation: Romanian orphans studied after their adoption in the UK showed a deprivation-specific ADHD-like behaviour. ⁶ In a study of television and video game use, neither total time spent nor exposure to violent content predicted attention problems or influenced school grades. ⁶

Whether it is a disease

In 2012 there was no single cause, and the identified risk factors were non-specific, as most of them appear to affect a range of different neurodevelopmental and psychiatric phenotypes. ⁷ The lack of common susceptibility genes or loci and the difficulty of a clear-cut diagnosis have led to debate in the scientific and medical community about its aetiology. ⁷

Szasz has argued that the disorder was invented, by psychiatrists to give a medical explanation for antisocial human traits, and not discovered, because behavioural interpretations do not represent a disease. ⁷ Adherents of the social construct theory believe that society has created it through its specific demands on children and its perception of an individual group, and the neurodiversity theory proposes that it is a normal human difference to be tolerated and respected like any other. ⁷ Other critics, among them some psychoanalysts, particularly in Germany, interpret it as the consequence of disturbances in the relationship between the child and the primary attachment figure, usually the mother, a view that usually leads to hot debates at congresses. ⁷

Some believe that it is selected for in evolution, as in the hunter–farmer theory of the disorder developed by Hartmann. ⁷ Building on this theory, one view regards it as a disorder of adaptation and suggests that many emotional and behavioural responses, particularly ones relatively commonplace within a species, may not just be symptoms of disorders but might instead reflect adaptive responses of the organism to environmental demands. ⁷ Another goes further and suggests that it may have evolved because it increases the creativity and inventiveness of the population, and that if the genes behind it are selected for because they foster creativity, it is not a neurological defect but a variant temperament, albeit one that may require intervention. ⁷ A study of Ariaal men in northern Kenya provides evidence for the selection of an associated variant: in them the ADHD-associated allele of the DRD4 gene promotes behavioural or psychological traits that are helpful in some social and ecological contexts and detrimental in others. ⁷ A neuropsychological study concluded that even individually impairing combinations of genes, such as those that may cause the disorder, can carry specific benefits for society and be selected for at that level, rather than being merely genetic coincidences with effects confined to the individual. ⁷

One critic argued in 2008 that the evidence for a genetic or neuroanatomic cause is insufficient and that the disorder is unlikely to exist as an identifiable disease, holding that inattention, hyperactivity and impulsivity are symptoms of many underlying treatable medical, emotional and psychosocial conditions affecting children. ³ Critics have described it as a diagnosis used to label difficult children who are not ill but whose behaviour is at the extreme end of the normal range. ³ In 2012 controversy also continued to grow over the medicines used to treat it, their efficacy, tolerability, safety, long-term effects and abuse potential, and over the social and ethical issues of its diagnosis and treatment. ³ In 2012 several questions remained open:

• whether the arguments against it being a real clinical condition are still valid given the most recent findings in genetics, pathophysiology and neuroimaging
• whether, even if they are, that view helps relieve the suffering of symptoms and impairments in patients
• how an unidentifiable disease can be treated
• whether the impact of undiagnosed cases on children's daily living in family, school and social settings will ever be measured
• what the psychological burden is on parents who believe their children's abnormal behaviour has no biological explanation
• whether the disorder is a by-product of poor parenting or miseducation

³

Diagnosis

As of 2012 there was neither a genetic test nor a brain scan to diagnose the disorder, because it is a heterogeneous collection of behaviours that appear to have different causes and symptoms. ¹ Without an objective diagnostic method, the major difficulty specialists face is deciding where to set the threshold between behaviours and states of mind that require medication or behavioural treatment and differences that can be left alone. ¹

In 2012 it was usually diagnosed using the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), which defines three general subtypes: ⁸

1. Predominantly hyperactive-impulsive: a child who is excessively fidgety and restless, seems always to be on the go, has difficulty waiting and remaining seated, acts immaturely, may not set physical boundaries and may exhibit destructive behaviours.
2. Predominantly inattentive: a child who is easily distracted and forgetful and shows daydreaming, disorganisation, poor concentration and difficulty completing tasks.
3. Combined type.

⁸

There is mounting evidence that the inattentive and combined subtypes are separable disorders with different underlying pathology. ⁸

The DSM-IV diagnostic criteria consist of two dimensions, symptoms and impairment, each with subtype-specific descriptions. ⁸ The distinction between symptom and impairment is often unclear, a symptom-based rating is problematic because judgements of what is normal and abnormal behaviour are subjective, and impairment is ambiguous and depends on the individual challenges and demands patients face in daily life. ⁸ Assessment of behavioural characteristics is therefore subjective and may be interpreted differently by different observers and in different cultures. ⁸ Prevalence is higher in North America than in Europe, where the diagnosis of hyperkinetic disorder in the tenth edition of the International Classification of Diseases (ICD-10) is more commonly used. ⁸ The diagnostic criteria in DSM-IV and ICD-10 are heterogeneous, and a positive diagnosis is three to four times more likely with DSM-IV than with ICD-10. ⁸

Other conditions often coexist with the disorder and make its diagnosis more difficult: as many as a third of diagnosed children have one or more comorbidities, most commonly learning disabilities, oppositional defiant disorder, conduct disorder, anxiety, tics and depressive disorders, and most of these share features with it, such as similar symptoms or age at onset. ⁸

In 2012 there were no biomarkers that could help diagnosis or the assessment of treatment efficacy. ⁸ It is nevertheless important to recognise the limitations of the DSM-IV definitions by adding more objective means of assessment to the diagnostic process. ⁸ One author has pointed to the need to verify the DSM-IV criteria in a more specific way that takes gender, cultural bias and developmental variations into account. ⁸ Some rating scales have been developed specifically to assist diagnosis, score symptom severity and rate improvements in various domains during intervention, in both primary care and clinical trials; the most widely used are the ADHD Rating Scale, the Conners' Parent and Teacher Rating Scales, the Child Behaviour Checklist, the Parent-rated Hyperactivity/Impulsivity Swanson Nolan and Pelham Ratings, and the UPPS Impulsivity Scale. ⁸

Medication

There are pharmacological and non-pharmacological treatments for both children and adults. ⁹ Pharmacological approaches are the most common and typically consist of stimulant medication, such as methylphenidate, dexmethylphenidate, mixed amphetamine salts and lisdexamfetamine dimesylate. ⁹ Non-stimulants such as atomoxetine, clonidine, guanfacine and reboxetine have also been found efficacious, although their efficacy seems slightly lower than that of stimulants. ⁹ Across the substance classes there is a large variety of delivery forms (liquid, sprinkle, tablet, capsule or patch), formulations (active isomers, mixtures of active and less active isomers, or prodrugs) and release forms (immediate-, intermediate- or extended-release). ⁹

Adverse effects are a serious problem that compromises treatment compliance for both stimulant and non-stimulant medications. ⁹ The most common side effects of stimulants are decreased appetite, sleeplessness, headache, abdominal pain and nausea, and those of non-stimulants include decreased appetite, abdominal pain, vomiting, headache, sleepiness and sedation. ⁹ Effects on blood pressure, heart rate and exercise parameters have also been reported for both but usually do not reach clinical relevance. ⁹ For instance, small but statistically significant changes in blood pressure and heart rate were observed at six weeks of treatment with high doses of extended-release methylphenidate in adolescents, without clinically meaningful electrocardiogram changes or serious cardiovascular adverse events. ⁹ Serious cardiovascular adverse events such as vasculopathy have also been reported with stimulant use, although rarely. ⁹ No cytogenetic side effects have been associated with methylphenidate. ⁹ The US prescribing information of the non-stimulant atomoxetine carries a black-box warning for suicidal ideation, based on a meta-analysis showing that the drug is associated with a significantly higher incidence of suicidal ideation than placebo. ⁹

Since these medications are prescribed for long-term treatment, longitudinal safety studies are needed. ⁹ Despite the frequent use of stimulants, for instance, in 2012 the effects of long-term use on children's growth and nutritional status were still unclear. ⁹ A systematic review of the safety information in the summaries of product characteristics of medications licensed in the UK for treating the disorder reported significant differences between those summaries and national prescribing guidelines, partly because the summaries are amended reactively as new information becomes available, and this may confuse clinicians seeking prescribing advice. ⁹

Non-pharmacological treatment

Non-pharmacological treatments are alternatives for patients who cannot or must not take the required medicines to manage their disease adequately, for instance because of contraindications, comorbid conditions such as anxiety and tic disorders, adverse effects, non-responsiveness or reduced efficacy. ¹⁰ Patients at risk of substance misuse should avoid stimulant medication. ¹⁰ These alternatives include psychosocial interventions such as cognitive and behavioural therapies: training parents and teachers in effective behaviour-management techniques, in behavioural parent training programmes, may help reduce the problem behaviours associated with the disorder in children, and cognitive behavioural therapy is commonly used for adults. ¹⁰ Neurofeedback has been proved efficacious, with a large effect on inattention and impulsivity and a medium effect on hyperactivity. ¹⁰ Computerised training of working memory is also beneficial, but in 2012 the consensus was that these non-pharmacological therapies support ongoing pharmacotherapy and should not be regarded as substitutes for it. ¹⁰ That conception is controversial. ¹⁰

Complementary and alternative treatment

Complementary and alternative approaches are also used to ameliorate its symptoms or combat its causes. ¹⁰ They include dietary modifications (diets rich in low glycaemic index carbohydrates, proteins and essential fatty acids), nutritional supplementation (for example with essential fatty acids, vitamin B6, magnesium, zinc, L-carnitine and different amino acids), herbal medicine (for example rhodiola, chamomile and St John's wort), homeopathy and physical exercise, and some of them have proven beneficial in patients with the disorder. ¹⁰ The biological rationale for using them is clear from the possible causes of the disorder and their relationship with diet, but objective assessment of their efficacy is difficult, and the placebo effect adds to the difficulty. ¹⁰ A systematic review of 34 studies published in the Chinese literature found that traditional Chinese medicine may have equal or better effectiveness than methylphenidate, but the quality of the trials does not support any particular recommendation of it for treating children. ¹⁰ Over-the-counter products used in Western medicine include Ginkgo biloba and short-chain fatty acids, but these have not been shown to be significantly superior to placebo or methylphenidate. ¹⁰ In a small placebo-controlled trial, omega-3 and omega-6 fatty acids improved symptoms in a sub-population of children with the inattentive subtype and comorbid neurodevelopmental disorders. ¹⁰

Choosing a treatment

The choice between pharmacological and psychosocial treatment is multifactorial. ¹¹ A study found that at the time of diagnosis parents and children view psychosocial treatment as a more acceptable option than medication, and that medication acceptability is significantly higher among Caucasian than among non-Caucasian parents. ¹¹ Actual experience with medication can increase parent-reported acceptability of medication treatment. ¹¹ Acceptability alone does not predict implementation and adherence, for either psychosocial or pharmacological treatment: the former is influenced by service availability and the feasibility of family attendance, such as time and affordability, and the latter by perceptions of needs and benefits weighed against side effects and costs, patient acceptance and social support. ¹¹ The choice also depends on the type and severity of the symptoms and on the perceived needs of patients and their families. ¹¹ Treatment preferences are often dynamic and context-dependent, as family priorities and values change over time. ¹¹ Pharmacotherapies alone have been noted to be better in the short term, and cognitive and behavioural therapies to deliver the best results in the long term. ¹¹ The medicines have no disease-modifying potential: they bring symptomatic benefit only for as long as the therapy lasts. ¹¹

Over- and under-diagnosis

Whether the disorder is over-diagnosed and children over-medicated is a matter of debate. ² Against the occasional denial that it exists as a clinical entity stands the risk of overdiagnosis and overtreatment, meaning children who are diagnosed but should not be (false positives). ³ Global use of medications for it rose three-fold from 1993 through 2003, while global spending rose nine-fold, to 2.4 billion US dollars in 2003. ³ Use and spending grew in both developed and developing countries, but spending grew more in developed countries. ³ In the United States the number of physician outpatient visits at which the disorder was diagnosed increased by 66%, from 6.2 to 10.4 million, from 2000 to 2010. ³ The recent rise in the rate of diagnosis and of stimulant use raised in 2012 the question of whether diagnostic thresholds are being lowered because societies do not tolerate the behaviours and impairments associated with it. ¹ A review of prevalence studies and research on factors affecting diagnostic accuracy up to 2007 concluded that there does not appear to be sufficient justification for the conclusion that the disorder is being systematically overdiagnosed, and noted that this conclusion is generally not reflected in public perceptions or media coverage. ³

There are also misdiagnosis and underdiagnosis: girls, for instance, are likely to be underdiagnosed because they more often have the inattentive subtype without the disruptive hyperactive behaviour. ³

Social and ethical consequences of the diagnosis

The disorder has served as a case study to illustrate the potential social and ethical consequences of psychiatric diagnosis and treatment, and for macro-level analyses of the corporate, governmental and institutional interactions involved in the uptake of psychiatric diagnoses and drug treatments. ³ One scholar of the disorder's social dimensions has argued that neither bioethics nor sociology has yet fully taken on its complexity, and that there is little attention to it as a lived experience in local contexts, which is particularly problematic because children carry the diagnosis in the midst of complex and highly contested social, political and medical territories. ³ On that account, the consequences of diagnosis for children's overall wellbeing have been neglected in scientific research, and much attention should be drawn to children's perceived experience of living with the disorder. ³