Attention-deficit/hyperactivity disorder
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
PRIMER
Attention-deficit/hyperactivity disorder
Stephen V. Faraone1,2, Philip Asherson3, Tobias Banaschewski4, Joseph Biederman5,
Jan K. Buitelaar6, Josep Antoni Ramos-Quiroga7–9, Luis Augusto Rohde10,11,
Edmund J. S. Sonuga-Barke12,13, Rosemary Tannock14,15 and Barbara Franke16
Abstract | Attention-deficit/hyperactivity disorder (ADHD) is a persistent neurodevelopmental disorder
that affects 5% of children and adolescents and 2.5% of adults worldwide. Throughout an individual’s
lifetime, ADHD can increase the risk of other psychiatric disorders, educational and occupational failure,
accidents, criminality, social disability and addictions. No single risk factor is necessary or sufficient to cause
ADHD. In most cases ADHD arises from several genetic and environmental risk factors that each have a
small individual effect and act together to increase susceptibility. The multifactorial causation of ADHD is
consistent with the heterogeneity of the disorder, which is shown by its extensive psychiatric co-morbidity,
its multiple domains of neurocognitive impairment and the wide range of structural and functional brain
anomalies associated with it. The diagnosis of ADHD is reliable and valid when evaluated with standard
criteria for psychiatric disorders. Rating scales and clinical interviews facilitate diagnosis and aid screening.
The expression of symptoms varies as a function of patient developmental stage and social and academic
contexts. Although there are no curative treatments for ADHD, evidenced-based treatments can markedly
reduce its symptoms and associated impairments. For example, medications are efficacious and normally
well tolerated, and various non-pharmacological approaches are also valuable. Ongoing clinical and
neurobiological research holds the promise of advancing diagnostic and therapeutic approaches to ADHD.
For an illustrated summary of this Primer, visit: http://go.nature.com/J6jiwl
Attention-deficit/hyperactivity disorder (ADHD; also associated with ADHD. These studies have created relia-
known as hyperkinetic disorder) is a common dis- ble and valid measurement tools for screening, diagnosis
order characterized by inattention or hyperactivity– and monitoring of treatment. Likewise, rigorous clinical
impulsivity, or both. The evidence base for the diagnosis trials have documented the safety and efficacy of ADHD
and treatment of ADHD has been growing exponentially treatment, and it is now clear which ADHD treatments
since the syndrome was first described by a German work, which do not and which require further study. In
physician in 1775 (REF. 1) (FIG. 1). In 1937, the efficacy this Primer, we discuss the evidence base that has created
of amphetamine use to reduce symptom severity was a firm foundation for future work to further clarify the
serendipitously discovered. In the 1940s, the brain was aetiology and pathophysiology of ADHD and to advance
implicated as the source of ADHD-like symptoms, which diagnostic and therapeutic approaches to this disorder.
were described as minimal brain damage in the wake of
Correspondence to S.V.F.
e-mail: sfaraone@
an encephalitis epidemic. In 1980, the third edition of Epidemiology
childpsychresearch.org the Diagnostic and Statistical Manual of Mental Disorders Age-dependent prevalence of ADHD
Departments of Psychiatry (DSM) created the first reliable operational diagnostic ADHD is a common disorder among young people
and of Neuroscience and criteria for the disorder. These criteria initiated many worldwide. In 2007, a meta-analysis of more than 100
Physiology, State University
programmes of research that ultimately led the scien- studies estimated the worldwide prevalence of ADHD in
of New York (SUNY) Upstate
Medical University, Syracuse,
tific community to view ADHD as a seriously impair- children and adolescents to be 5.3% (95% CI: 5.01–5.56)2.
New York 13210, USA; ing, often persistent neurobiological disorder of high Three methodological factors explained this variabil-
K.G. Jebsen Centre for prevalence that is caused by a complex interplay between ity among studies: the choice of diagnostic criteria, the
Neuropsychiatric Disorders, genetic and environmental risk factors. These risk fac- source of information used and the inclusion of a require-
Department of Biomedicine,
University of Bergen,
tors affect the structural and functional capacity of brain ment for functional impairment as well as symptoms for
5020 Bergen, Norway. networks and lead to ADHD symptoms, neurocognitive diagnosis. After adjusting for these factors, a subsequent
deficits and a wide range of functional impairments. meta-analysis concluded that the prevalence of ADHD
Article number: 15020
doi:10.1038/nrdp.2015.20
We now have many large and well-designed epi does not significantly differ between countries in Europe,
Published online demiological, clinical and longitudinal studies that have Asia, Africa and the Americas, as well as in Australia3.
6 August 2015 clarified the features, co-morbidities and impairments Although other meta-analyses have found either lower or
NATURE REVIEWS | DISEASE PRIMERS VOLUME 1 | 2015 | 1
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Author addresses
symptoms; and third, enabling ADHD to be diagnosed
in the presence of an autism spectrum disorder. The
1
Departments of Psychiatry and of Neuroscience and Physiology, State University third change is consistent with the reconceptualization
of New York (SUNY) Upstate Medical University, Syracuse, New York 13210, USA. of ADHD in DSM‑5 as a neurodevelopmental disorder
2
K.G. Jebsen Centre for Psychiatric Disorders, Department of Biomedicine, rather than a disruptive behavioural disorder. Overall,
University of Bergen, 5020 Bergen, Norway.
these new criteria have yielded an increase in ADHD
3
Social Genetic and Developmental Psychiatry, Institute of Psychiatry Psychology
and Neuroscience, King’s College London, London, UK.
prevalence, which is insubstantial for children but is likely
4
Department of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of to have had a more considerable effect on diagnosis rates
Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany. in adults13,14.
5
Department of Psychiatry, Harvard Medical School, Massachusetts General Hospital,
Pediatric Psychopharmacology Unit, Massachusetts General Hospital, Boston, Sociodemographic factors
Massachusetts, USA. Alongside age, other factors such as sex, ethnicity and
6
Radboud University Medical Center, Donders Institute for Brain, Cognition and socioeconomic status are also important when consider
Behaviour, Department of Cognitive Neuroscience and Karakter Child and Adolescent ing the prevalence of ADHD. In children and adolescents,
Psychiatry University Centre, Nijmegen, The Netherlands. ADHD predominantly affects males and exhibits a male-
7
ADHD Program, Department of Psychiatry, Hospital Universitari Vall d’Hebron,
to-female sex ratio of 4:1 in clinical studies and 2.4:1 in
Barcelona, Spain.
8
Biomedical Network Research Centre on Mental Health (CIBERSAM), Barcelona, Spain.
population studies2. In adulthood, this sex discrepancy
9
Department of Psychiatry and Legal Medicine, Universitat Autònoma de Barcelona, almost disappears14, possibly owing to referral biases
Barcelona, Spain. among treatment-seeking patients or to sex‑specific
10
ADHD Outpatient Program, Hospital de Clinicas de Porto Alegre, Department of effects of ADHD over the course of the disorder.
Psychiatry, Federal University of Rio Grande do Sul, Porto Alegre, Brazil. Larsson and colleagues 15 found that low family
11
National Institute of Developmental Psychiatry for Children and Adolescents, income predicted an increased likelihood of ADHD in
Sao Paulo, Brazil. a Swedish population-based cohort study of 811,803
12
Department of Psychology, University of Southampton, Southampton, UK. individuals. However, this finding does not necessar-
13
Department of Experimental Clinical and Health Psychology, Ghent University, Ghent, ily support the conclusion that socioeconomic status
Belgium.
increases the risk of ADHD because the disorder
14
Neuroscience and Mental Health Research Program, Research Institute of The Hospital
for Sick Children, Toronto, Canada.
runs in families and leads to educational and occupa-
15
Department of Applied Psychology and Human Development, Ontario Institute for tional underattainment. Underemployment could in
Studies in Education, University of Toronto, Toronto, Ontario, Canada. turn lead to the over-representation of socioeconomic
16
Radboud University Medical Center, Donders Institute for Brain, Cognition and disadvantage among families affected by ADHD16.
Behaviour, Departments of Human Genetics and Psychiatry, Nijmegen, The Netherlands. Finally, although the true prevalence of ADHD does
not vary with ethnicity, some studies have inconsistently
associated ethnicity with ADHD owing to referral pat-
higher prevalence rates, these presented important limita- terns and barriers to care that disproportionately affect
tions, such as the exclusive use of DSM criteria to diagnose particular ethnic groups17–19.
ADHD and the use of simulated prevalence rates4,5. In
addition, there is no evidence, worldwide, of an increase Mechanisms/pathophysiology
in the real prevalence of ADHD over the past three Genes and environment
decades3. Despite the fact that both overdiagnosis and Genetic epidemiology. ADHD runs in families, with
underdiagnosis are common concerns in medicine, the parents and siblings of patients with ADHD show-
common public perception that ADHD is overdiagnosed ing between a fivefold and tenfold increased risk of
in the United States might not be warranted6. developing the disorder compared with the general
ADHD also affects adults. Although the majority population20,21. Twin studies show that ADHD has a
of children with ADHD will not continue to meet the heritability of 70–80% in both children and adults22–25,
full set of criteria for ADHD as adults, the persistence of with little or no evidence that the effects of environ
either functional impairment 7 or subthreshold (three or mental risk factors shared by siblings substantially influ-
fewer) impairing symptoms into adulthood is high8. For ence aetiology 26. Environmental risk factors play their
instance, on the basis of a meta-analysis of six studies, greatest part in the non-shared familial environment
Simon and colleagues9 found the pooled prevalence of and/or act through interactions with genes and DNA
ADHD to be 2.5% (95% CI: 2.1–3.1) in adults. In addi- variants that regulate gene expression — such as those
tion, studies in older adults have found prevalence rates in promoters, untranslated regions of genes or loci that
in the same range10,11, and prospective longitudinal stud- encode microRNAs.
ies support the notion that approximately two-thirds of Although ADHD is a categorical diagnosis, results
youths with ADHD retain impairing symptoms of the from twin studies suggest that it is the extreme and
disorder in adulthood7 (FIG. 2). impairing tail of one or more heritable quantitative traits27.
Recent alterations to diagnostic criteria have had an The disorder is influenced by both stable genetic factors
impact on ADHD prevalence measures in both young and those that emerge at different developmental stages
and adult populations. In 2013, DSM‑5 (REF. 12) included from childhood through to adulthood28. Thus, genes con-
three important changes: first, increasing the age of onset tribute to the onset, persistence and remission of ADHD,
from 7 years to 12 years; second, decreasing the symptom presumably through stable neurobiological deficits as
threshold for patients ≥17 years of age from six to five well as maturational or compensatory processes that
2 | 2015 | VOLUME 1 www.nature.com/nrdp
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Weikard describes ADHD syndrome in a German textbook DSM-II describes hyperkinetic reaction Omega-3 is a weak but effective
treatment
Douglas’ neurocognitive model of ADHD
Hoffman cartoons of Still describes DSM-5 extends age of onset
‘Fidgeting Philip’ and ‘defect of moral to 12 years and adjusts criteria
control’ in ADHD-like symptoms Twin studies for adults
‘Johnny Head-in-the-Air’
The Lancet described as ‘minimal document high
brain damage’ heritability CBT for adult ADHD
Bourneville, Boulanger, Paul-Boncour US FDA approves DSM-III Rare genomic insertions and
and Philippe describe ADHD symptoms methylphenidate operationalizes deletions discovered
as ‘mental instability’ in French medical for depression diagnostic
and educational literature and narcolepsy criteria Molecular polygenic
background confirmed
1775 1798 1845 1887 1901 1910 1930s 1940s 1950s 1960s 1970s 1980s 1985 1990s 1995 2000s 2010s
Sagvolden describes an ADHD rat model
Bradley shows that benzedrine
reduces hyperactivity Similar correlates of ADHD found in boys and girls CDC describes
Crichton describes ADHD as a
ADHD syndrome in Kramer–Pollnow syndrome Neuroimaging documents structural and serious public
a Scottish textbook discovered Parent training functional brain anomalies health problem
treatments
DSM-IV refines criteria Long-acting
Methylphenidate stimulants
Prediagnostic era ADHD in adults recognized as a valid disorder developed
indicated for
Minimal brain dysfunction era behavioural
Attention-deficit disorder era disorders Co-morbidity with anxiety, mood or autism spectrum Non-stimulants
ADHD era in children disorders and executive dysfunction confirmed approved
Nature Reviews | Disease Primers
Figure 1 | The history of attention-deficit/hyperactivity disorder. Attention-deficit/hyperactivity disorder (ADHD)
‘syndromes’ have been described in the medical literature since the eighteenth century, but the growth of systematic
research required the development of operational diagnostic criteria in the late twentieth century. This schematic
outlines selected important developments in the history of ADHD research. CBT, cognitive–behavioural therapy;
CDC, Centers for Disease Control and Prevention; DSM, Diagnostic and Statistical Manual of Mental Disorders.
influence development. The inattention and hyperactiv- and twin studies that found significant co‑aggregation
ity or impulsivity that characterize ADHD are separate of ADHD with depression37, conduct problems39 and
domains of psychopathology, with a genetic correlation schizophrenia40. Furthermore, combined GWAS of
of around 0.6, reflecting substantial genetic overlap but ADHD, autism spectrum disorders, depression, bipolar
also genetic influences that are domain specific29. Shared disorder and schizophrenia identified four genome-wide
genetic factors also account for the co‑occurrence of significant loci shared by these disorders37.
ADHD with emotional dysregulation — an independ- In addition to the common-variant studies, rare
ent source of impairment in ADHD30,31. Family and twin (prevalence of 500,000 base pairs
disorders, including conduct disorder and problems32, in length compared with 7.5% of individuals without
cognitive performance33, autism spectrum disorders34 the disorder. The rate of large CNV carriage was even
and mood disorders35,36. higher (42.4%) in those with both ADHD and an IQ
below 70 ± 5 (which, along with poor adaptive function-
Molecular genetics. On the basis of data from genome- ing, defines intellectual disability)42. These findings have
wide association studies (GWAS), approximately 40% of been replicated43, and together these studies implicate
the heritability of ADHD can be attributed to numer- genes at 16p13.11 along with the 15q11–15q13 region in
ous common genetic variants37. In polygenic risk score ADHD. The 15q11–15q13 region contains the gene that
analysis, the genetic signals attributed to common vari- encodes the nicotinic α7 acetylcholine receptor subunit,
ants derived from a discovery sample are used to predict which participates in neuronal and nicotinic signalling
phenotypic effects in a second sample. The polygenic risk pathways. Finally, ADHD-associated CNVs also span
for clinically diagnosed ADHD predicts ADHD symp- several glutamate receptor genes, which are essential
toms in the population more broadly 38, confirming the for neuronal glutamatergic transmission44, and the gene
conclusion from twin studies that the genes determining encoding neuropeptide Y, which is involved in signal-
the diagnosis of ADHD also regulate the expression of ling in the brain and autonomic nervous system45. CNVs
subclinical levels of ADHD symptoms. In addition, these associated with ADHD also occur in schizophrenia
analyses have confirmed earlier evidence from family and autism42.
NATURE REVIEWS | DISEASE PRIMERS VOLUME 1 | 2015 | 3
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
100 evoke ‘hostile’ styles of parenting, and genes linked to
ADHD might explain the association of parental vari-
ables, such as maternal smoking during pregnancy,
75 with offspring who have ADHD51,52. One notable study
71% investigated maternal hostility while controlling for
Persistence (%)
65% genetic effects by studying children adopted at birth
50 and children conceived through in vitro fertiliza-
tion and their genetically unrelated rearing mothers53.
The study found a role for genetically influenced early
25 Functional impairment child behaviour on the hostility of biologically unrelated
Impairing symptoms 15% mothers, which in turn was a predictor of subsequent
Full diagnostic criteria ADHD symptoms developed by the children. Another
0 study followed Romanian adoptees who had experi-
0 5 10 15 20 25 30 enced severe early maternal deprivation in orphanages
Mean age at follow-up (years) before adoption. It showed a dose-dependent relation-
Figure 2 | The age-dependent decline and persistence of attention-deficit/ ship between length of deprivation and risk of develop-
Nature Reviews | Disease Primers
hyperactivity disorder throughout the lifetime. Follow‑up studies have assessed ing ADHD-like symptoms54. Other environmental risk
children with attention-deficit/hyperactivity disorder (ADHD) at multiple time points factors that have been associated with ADHD include
after their initial diagnosis. Although they document an age-dependent decline in ADHD prenatal and perinatal factors, such as maternal smok-
symptoms, ADHD is also a highly persistent disorder when defined by the persistence of ing and alcohol use, low birth weight, premature birth
functional impairment7 or the persistence of subthreshold (three or fewer) impairing
and exposure to environmental toxins, such as organo-
symptoms8. By contrast, many patients remit full diagnostic criteria7.
phosphate pesticides, polychlorinated biphenyls, zinc
and lead55,56. Animal models have also contributed
Although GWAS that investigated common genetic much to the study of environmental risk factors57–59.
variants (FIG. 3) have not identified specific ADHD genes Similar to genetic risk factors, the effects of any one
at genome-wide levels of significance46, intriguing results environmental risk factor are small and could reflect
have emerged from meta-analyses of studies of candi- either small effects in many cases or larger effects in a
date genes involved in the monoamine neurotransmit- few cases. Furthermore, rather than being specific to
ter systems47. These systems had been implicated in the ADHD, these environmental risk factors are associated
pathophysiology of ADHD by the mechanisms of action with several psychiatric disorders29.
of drugs used in clinical management. Methylphenidate In addition to the main effects of the environment,
and amphetamine target the sodium-dependent dopa- the high heritability of ADHD suggests that gene–
mine transporter (encoded by SLC6A3), atomoxetine environment (G × E) interactions might be the main
targets the sodium-dependent noradrenaline trans- mechanism by which environmental risk factors increase
porter, and both extended-release guanfacine and the risk of ADHD. For example, a variant of 5‑HTTLPR —
extended-release clonidine target the α2A-adrenergic a polymorphic region located in the promoter of SLC6A4
receptor. Within the monoamine systems, the strongest — is involved in the hyperactivity and impulsivity dimen-
evidence of ADHD association is for variants in the genes sions of ADHD in interaction with stress60. Although
encoding the D4 and D1B dopamine receptors47. The some early studies identified other G × E effects, none
association of the SLC6A3 gene variant is equivocal47, has been reliably reproduced. Future success in this area
possibly owing to age-related effects48. Other genes that requires the use of large data sets, such as those emerg-
show possible associations with ADHD include SLC6A4 ing from the use of national databases in Denmark and
(which encodes the sodium-dependent serotonin trans- Sweden, which can combine large-scale genetic studies
porter), HTR1B (which encodes 5‑hydroxytryptamine with recorded data on exposure to environmental risks.
receptor 1B (also known as serotonin receptor 1B)) and Another approach to identify environmental risk fac-
SNAP25 (which encodes synaptosomal-associated pro- tors in ADHD is to focus on the detection of epigenetic
tein 25)47. Owing to methodological issues, a cautious changes, such as DNA methylation, which are revers-
approach must be taken to the interpretation of candi ible changes in genomic function that are independent
date gene studies. Nevertheless, the role of the dopa- of DNA sequence. Epigenetics provides a mechanism
mine, noradrenaline, serotonin and neurite outgrowth by which environmental risk factors alter gene func-
systems is supported by genome-wide association study- tion. However, as epigenetic changes are highly tissue
based gene-set analyses reporting that, as a group, genes specific, they are difficult to study in ADHD because of
regulating these systems were associated with ADHD limited access to brain tissue. Studies must, therefore,
and hyperactivity or impulsivity 46,49,50. rely on peripheral tissues such as blood, the epigenetic
profile of which partly overlaps with that of brain tis-
Environmental risk factors. Identifying environmental sue. Environmental toxins and stress can all induce epi
causes of ADHD is difficult because environmen- genetic changes, thus the identification of genes that
tal associations might arise from other sources, such show epigenetic changes linked to ADHD, or in response
as from child or parental behaviours that shape the to environmental risk factors, might in the future pro-
environment, or they might reflect unmeasured third vide new insights into the mechanisms involved in the
variables. For example, children with ADHD might pathogenesis of ADHD61.
4 | 2015 | VOLUME 1 www.nature.com/nrdp
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Brain mechanisms anticipation of reward than in controls75. ADHD is also
Cognition. ADHD is characterized by deficits in multi- associated with hyperactivation in somatomotor and vis-
ple, relatively independent, cognitive domains. Executive ual systems74, which possibly compensates for impaired
functioning deficits are seen in visuospatial and verbal functioning of the prefrontal and anterior cingulate cor-
working memory, inhibitory control, vigilance and tices76. A single dose of methylphenidate (a stimulant)
planning 62,63. Studies of reward dysregulation show that markedly enhances activation in the inferior frontal cor-
patients with ADHD make suboptimal decisions64, pre- tex and insula bilaterally — which are key areas of cogni-
fer immediate rather than delayed rewards65 and over- tive control — during inhibition and time discrimination
estimate the magnitude of proximal relative to distal but does not affect working memory networks77. By con-
rewards66. Other domains impaired in ADHD include trast, long-term treatment with stimulants is associated
temporal information processing and timing 67; speech with normal activation in the right caudate nucleus dur-
and language68; memory span, processing speed and ing the performance of attention tasks78. Resting-state
response time variability 69; arousal and activation70; MRI studies have shown that ADHD is associated with
and motor control71. Although most patients with less-pronounced or absent anti-correlations between the
ADHD show deficits in one or two cognitive domains, default-mode network (DMN) and the cognitive control
some have no deficits and very few show deficits in all network, lower connectivity within the DMN itself and
domains72. In addition, across the lifespan of patients lower connectivity within the cognitive and motivational
with ADHD, deficits in cognitive control, reward sensi loops of the frontostriatal circuits79.
tivity and timing have been shown to be independent Along with functional changes, a range of struc-
of one another 73, and it is currently unclear whether tural brain alterations are also associated with ADHD.
cognitive deficits cause ADHD symptoms and drive the For example, ADHD is associated with a 3–5% smaller
development of the clinical phenotype 72 or reflect total brain size than unaffected controls80,81 that can be
the pleiotropic outcomes of risk factors. attributed to a reduction of grey matter 82. Consistent
with genetic data that support a model of ADHD as the
Structural and functional brain imaging. Several brain extreme of a population trait, total brain volume cor-
regions and neural pathways have been implicated relates negatively with ADHD symptoms in the general
in ADHD (FIG. 4). Functional MRI studies in patients with population83. In patients with ADHD, meta-analyses
ADHD that used inhibitory control, working memory have documented smaller volumes across several brain
and attentional tasks have shown underactivation of fron- regions, most consistently in the right globus pallidus,
tostriatal, frontoparietal and ventral attention networks74. right putamen, caudate nucleus and cerebellum84,85.
The frontoparietal network mediates goal-directed In addition, a meta-analysis of diffusion tensor imaging
executive processes, whereas the ventral attention net- studies showed widespread alterations in white matter
work facilitates reorientation of attention towards salient integrity, especially in the right anterior corona radiata,
and behaviourally relevant external stimuli. In reward- right forceps minor, bilateral internal capsule and left
processing paradigms, most studies report lower activa- cerebellum86. Both structural and functional imaging
tion of the ventral striatum of patients with ADHD in findings are very variable across studies, suggesting that
the neural underpinnings of ADHD are heterogeneous,
which is consistent with studies of cognition.
22q11.2 deletion syndrome 16p13.11 Monoamine Just as the prevalence of ADHD is associated with
15q11–15q11 13 region containing systems
Jacobsen syndrome genes age (FIG. 2), so too are many changes in the brains of
(deletions of the end of 11q) nicotinic α7 acetylcholine
receptor subunit gene Neurite patients with ADHD7. Some brain volumetric alterations
Turner syndrome (X0) outgrowth
Klinefelter syndrome (XXY) Rare point mutations expected
genes observed in childhood normalize with age82,85, whereas
from sequencing studies
other measures remain fixed. For example, a longitudi-
nal MRI study found lower basal ganglion volumes and
High
Rare reduced dorsal surface area in adolescents with ADHD
chromosomal compared with controls, and this difference did not
anomalies
change as patients aged87. Furthermore, for ventral striatal
surfaces, control individuals showed surface area expan-
Rare and low
Effect size
sion with age, whereas patients with ADHD experienced
frequency copy
number variants a progressive contraction of the surface area. The as-yet-
unknown process underlying this contraction might
explain abnormal processing of reward in ADHD87.
Common variants
explain ~40% ADHD is also associated with delayed maturation of the
of heritability cerebral cortex. In one study, the age of attaining peak
cortical thickness was 10.5 years for patients with ADHD
Low
Very rare Rare Low Common
and 7.5 years for unaffected individuals; this delay was
Allele frequency most prominent in the prefrontal regions that are impor-
Figure 3 | Genetics of attention-deficit/hyperactivity disorder. Common variants tant for executive functioning, attention and motor plan-
Nature Reviews | Diseasedisorder
explain approximately 40% of the heritability of attention-deficit/hyperactivity Primers ning 88. The development of cortical surface area was
but, compared with rarer causes, individual common variants have much smaller effects also shown to be delayed in patients with ADHD, but
on the expression of the disorder. ADHD was not associated with altered developmental
NATURE REVIEWS | DISEASE PRIMERS VOLUME 1 | 2015 | 5
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
a Dorsolateral Parietal b Ventral anterior Dorsal anterior c d
Nigrostriatal
prefrontal cortex cortex cingulate cortex cingulate cortex Mesocortical Locus Cortex Thalamus
Caudate coeruleus
nucleus
Dorsal
Mesolimbic anterior
Nucleus
Ventromedial accumbens Substantia nigra cingulate Basal
prefrontal tegmentum cortex ganglia
Putamen
cortex Noradrenergic Executive control Pons
Amygdala Cerebellum Dopaminergic Corticocerebellar
e f g
Orbitofrontal Frontal Medial view Lateral view Lateral
cortex cortex parietal
cortex
Medial Medial
Ventromedial prefrontal prefrontal
prefrontal Ventral cortex Posterior cortex
cortex striatum cingulate Medial
cortex temporal lobe
Figure 4 | Brain mechanisms in attention-deficit/hyperactivity disorder. a | The cortical regions
Nature (lateral
Reviews view) ofPrimers
| Disease the
brain have a role in attention-deficit/hyperactivity disorder (ADHD). The dorsolateral prefrontal cortex is linked to working
memory, the ventromedial prefrontal cortex to complex decision making and strategic planning, and the parietal cortex
to orientation of attention. b | ADHD involves the subcortical structures (medial view) of the brain. The ventral anterior
cingulate cortex and the dorsal anterior cingulate cortex subserve affective and cognitive components of executive
control. Together with the basal ganglia (comprising the nucleus accumbens, caudate nucleus and putamen), they form
the frontostriatal circuit. Neuroimaging studies show structural and functional abnormalities in all of these structures in
patients with ADHD, extending into the amygdala and cerebellum. c | Neurotransmitter circuits in the brain are involved in
ADHD. The dopamine system plays an important part in planning and initiation of motor responses, activation, switching,
reaction to novelty and processing of reward. The noradrenergic system influences arousal modulation, signal-to-noise
ratios in cortical areas, state-dependent cognitive processes and cognitive preparation of urgent stimuli. d | Executive
control networks are affected in patients with ADHD. The executive control and corticocerebellar networks coordinate
executive functioning, that is, planning, goal-directed behaviour, inhibition, working memory and the flexible adaptation
to context. These networks are underactivated and have lower internal functional connectivity in individuals with ADHD
compared with individuals without the disorder. e | ADHD involves the reward network. The ventromedial prefrontal
cortex, orbitofrontal cortex and ventral striatum are at the centre of the brain network that responds to anticipation and
receipt of reward. Other structures involved are the thalamus, the amygdala and the cell bodies of dopaminergic neurons
in the substantia nigra, which, as indicated by the arrows, interact in a complex manner. Behavioural and neural responses
to reward are abnormal in ADHD. f | The alerting network is impaired in ADHD. The frontal and parietal cortical areas and
the thalamus intensively interact in the alerting network (indicated by the arrows), which supports attentional functioning
and is weaker in individuals with ADHD than in controls. g | ADHD involves the default-mode network (DMN). The DMN
consists of the medial prefrontal cortex and the posterior cingulate cortex (medial view) as well as the lateral parietal
cortex and the medial temporal lobe (lateral view). DMN fluctuations are 180 degrees out of phase with fluctuations in
networks that become activated during externally oriented tasks, presumably reflecting competition between opposing
processes for processing resources. Negative correlations between the DMN and the frontoparietal control network are
weaker in patients with ADHD than in people who do not have the disorder.
trajectories of cortical gyrification89. Remission of ADHD widespread deviations in cortical thickness persist in
has been associated with normalization of abnormalities many adults with ADHD. Findings include both cortical
as measured by activation during functional imaging thinning (in the superior frontal cortex, precentral cor-
tasks90, cortical thinning 91 and structural and functional tex, inferior and superior parietal cortex, temporal pole
brain connectivity 92–95. and medial temporal cortex)89,96 and cortical thickening
Although these data could be taken to suggest that the (in the pre-supplementary motor area, somatosensory
age-dependent decline in the prevalence of ADHD might cortex and occipital cortex)97. More work is needed to
be due to the late development of ADHD-associated brain determine how developmental changes in patterns of
structures and functions, most patients with ADHD do cortical thickness predict developmental changes in
not show complete developmental ‘catch up’. Indeed, ADHD symptom expression.
6 | 2015 | VOLUME 1 www.nature.com/nrdp
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Summary Children and adolescents
Neurocognitive, neuroimaging and genetic theories The diagnosis of ADHD relies on clinical symptoms
of ADHD have shifted from single-cause or single- reported by patients or informants (including rela-
pathway models to models that delineate causes that tives and teachers), which is standard for all psychiatric
lead to ADHD through several molecular, neural and disorders12. National clinical guidelines and practice
neurocognitive pathways33,98–102. These approaches have parameters for ADHD, developed over the past dec-
received clear support from aetiological studies indi- ade, show good consensus and the potential to enhance
cating that most cases of ADHD arise from a ‘pool’ of evidence-based clinical practice105. Diagnosis is based
genetic and environmental risk factors. Most of these on information from a detailed clinical interview, which
risk factors have only a small effect on causal pathways. remains the ‘gold standard’. Diagnosticians ask about
Cumulative vulnerability increases ADHD trait scores, each ADHD symptom, the age of onset and resultant
and our current model suggests that ADHD emerges functional impairments. A clinical interview aims to
when these exceed a certain threshold. In most cases, no establish whether symptoms are more extreme, persis-
single factor is necessary or sufficient to cause ADHD. tent and impairing than expected for the developmen-
However, in some patients, rare genetic variants41,42 or tal level of the patient. Validated rating scales (TABLE 1)
environmental risk factors — for example, psychosocial help with such decisions, as they enable informants to
deprivation54 — might have a major influence. quantitatively rate the behaviour of the patient at home,
The multifactorial causation of ADHD leads to a at school and in the community.
heterogeneous profile of psychopathology, neuro Several factors present challenges to clinicians aiming
cognitive deficits and abnormalities in the struc- to determine whether a diagnosis of ADHD is appro-
ture and function of the brain. Many cases probably priate. For instance, cultural and ethnic differences can
involve dysregulation of the structure and function hinder diagnosis owing to variability in attitudes towards
of the frontal–s ubcortical–c erebellar pathways ADHD, willingness to report symptoms or the accept-
that control attention, response to reward, salience ance of the diagnosis. For example, a literature review
thresholds, inhibitory control and motor behaviour. suggested that African-American youths had more
A meta-a nalysis of peripheral biomarkers in the ADHD symptoms than Caucasian youths but were
blood and urine of drug-naive or drug-free patients diagnosed with ADHD only two-thirds as often, possi-
with ADHD and unaffected individuals found several bly owing to parent beliefs about ADHD and the lack of
measures — specifically, noradrenaline, 3‑methoxy‑4‑ treatment access and use106. In addition, patient age can
hydroxyphenylethylene glycol (MHPG), monoamine be an issue. Developmental changes can internalize or
oxidase (MAO) and cortisol — to be significantly modify some symptoms. For example, the hyperactivity
associated with ADHD56. Several of these metabolites of childhood might be experienced as inner restless-
were also related to response to ADHD medication ness in adolescence, and distractibility could manifest
and symptom severity of ADHD. These results sup- as distracting thoughts. Accordingly, self-reports from
port the idea that catecholaminergic neurotransmitter adolescents are useful, but patients can sometimes
systems (discussed in further detail in the following lack insight into their own difficulties. Furthermore,
section) and the hypothalamic–pituitary–adrenal although younger children can provide useful informa-
axis are dysregulated in ADHD. Finally, genetic and tion, especially about internalizing symptoms107, parents
clinical studies also implicate other systems, including remain the main source of information for this group of
the serotonergic, nicotinic, glutamatergic and neurite patients. Parents can report on symptoms during school
outgrowth systems. recesses and vacations when teacher reports are not
available. Although parent reports show good concur-
Diagnosis, screening and prevention rent and predictive validity 108,109, information from other
The diagnostic process for ADHD assesses the inatten informants such as teachers, when available, is valuable
tive and hyperactive–impulsive symptom criteria for for documenting ADHD in other settings, for predicting
ADHD, evidence that symptoms cause functional prognosis and for increasing the confidence of diagno-
impairments and age of onset before 12 years. Although ses110–112. Finally, diagnosticians can also inquire about
ADHD is associated with other features such as execu- other medical conditions associated with symptoms
tive dysfunction62 and emotional dysregulation31,103, of ADHD, such as seizure disorders, sleep disorders,
these are commonly observed in other disorders and hyperthyroidism, physical or sexual abuse and sensory
are not core diagnostic criteria for ADHD12. To assist impairments113, as these can confound diagnosis.
diagnosis, several open access assessment tools have Although screening for ADHD is theoretically feasi
been created for use in both children (TABLE 1) and adults ble given the availability of parent and self-reported
(TABLE 2), and excellent, well-normed (standardized) scales (TABLE 1), the few studies that have investigated the
commercial scales are available104. Importantly, patient use of early screening for ADHD have yielded inconsist-
age is relevant when assessing standard diagnostic cri- ent findings. For example, a 6-year longitudinal study
teria, such as those of the DSM or the International suggested that a parent-rated questionnaire might help
Statistical Classification of Diseases and Related Health with early detection, prediction and treatment plan-
Problems (ICD), owing to changes in the expression ning 114. However, owing to a lack of accurate predic-
of ADHD symptoms and impairments throughout an tors of onset, attempts at early prevention of ADHD
individual’s lifetime (FIGS 2,5). currently rely on population-level efforts to mitigate
NATURE REVIEWS | DISEASE PRIMERS VOLUME 1 | 2015 | 7
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
the effects of environmental risk factors for the dis screening programmes in primary care, parent training
order. Primary prevention strategies optimize maternal programmes, and specific games and play-based pro-
health during pregnancy by reducing extreme stress and grammes to enhance self‑regulation when symptoms
psychosocial adversity, eliminating smoking, alcohol are identified115,116.
and drug use and reducing risk factors for preterm birth
and low birth weight. Secondary prevention approaches Adults
that detect symptoms of ADHD at an early stage — Over the past 40 years, clinical, family, treatment,
for example at infancy or preschool age — include longitudinal and population studies have generated
Table 1 | A selection of open access resources for assessing attention-deficit/hyperactivity disorder in childhood
Approach Comments Websites
Interviews
Schedule for Affective • A semi-structured diagnostic interview http://www.psychiatry.pitt.edu/node/8233
Disorders and • Evaluates past and current psychopathology in children and
Schizophrenia in adolescents, according to DSM‑IV and DSM-III criteria
School Age Children • Translations in many languages
(K‑SADS) • A DSM‑5 version is imminent
Diagnostic Interview • A structured diagnostic that uses DSM‑IV to assess psychopathology http://www.cdc.gov/nchs/data/nhanes/
Schedule for Children in children and adolescents limited_access/interviewer_manual.pdf
(DISC) • Translations in many languages
Child and Adolescent • A semi-structured interview that evaluates current psychopathology https://devepi.duhs.duke.edu/capa.html
Psychiatric Assessment in children and adolescents
https://devepi.duhs.duke.edu/pubs/
(CAPA) • Based on DSM‑IV criteria
papachapter.pdf
• Versions for youths and preschool-aged children
• Spanish and Portuguese translations
Development and • For clinicians and trained non-clinicians http://www.dawba.com/b0.html
Well-Being Assessment • Uses a prespecified set of questions and probes for impairment
(DAWBA) • Generally used together with the SDQ
• Translations in many languages
Parent Interview for • A semi-structured interview focused on diagnostic criteria for ADHD, http://www.sickkids.ca/MS-Office-Files/
Child Symptoms (PICS) ODD and CD in children and adolescents Psychiatry/17145-Administration_Guidelines_
• Addresses symptoms of other psychiatric disorders PICS6.pdf
• Has been updated for DSM‑5 criteria
http://www.sickkids.ca/pdfs/Research/
• Includes the TTI, which assesses symptoms of ADHD, ODD and CD in
Tannock/6013-TTI-IVManual.pdf
school, with screening questions for other psychopathology
• Dutch translation
Child ADHD • A structured telephone interview for teachers Available from the authors254
TTI (CHATTI) • Focuses on DSM‑IV criteria for ADHD in school
• Only available in English
Scales
Vanderbilt ADHD • Versions for a parent or caregiver and teacher http://www.nichq.org/childrens-health/adhd/
Diagnostic Rating • Part of the American Academy of Pediatrics ADHD Toolkit resources/vanderbilt-assessment-scales
Scales (VARS) • Spanish translation
Swanson, Nolan and • A rating scale for symptoms of ADHD and ODD Short scale (26‑item) available from: http://www.
Pelham (SNAP)-IV • Can be completed by a teacher, parent or caregiver caddra.ca/pdfs/caddraGuidelines2011SNAP.pdf
Rating Scale • Sensitive to changes related to treatment
Scoring guidelines available from: http://
• Portuguese, Spanish and French translations
www.caddra.ca/pdfs/caddraGuidelines2011
SNAPInstructions.pdf
Full scale (90‑item) available from:
http://www.adhd.net/snap-iv-form.pdf
Scoring guidelines available from:
http://www.adhd.net/snap-iv-instructions.pdf
Strengths and • Versions for a teacher, parent or caregiver http://www.adhd.net/SWAN_SCALE.pdf
Weaknesses of ADHD • Based on DSM‑IV criteria
Symptoms and Normal • Unusual in that the items are positively worded and it covers both
Behavior Scale (SWAN) strengths as well as weaknesses in ADHD and ODD symptoms
• Spanish and French translations
SDQ • Brief measure of emotional, ADHD, conduct and relationship problems http://sdqinfo.org
• Versions for a parent, caregiver or teacher and a self-report
• Translations in many languages
ADHD, attention-deficit/hypersensitivity disorder; CD, conduct disorder; DSM, Diagnostic and Statistical Manual of Mental Disorders; ODD, oppositional defiant
disorder; SDQ, Strengths and Difficulties Questionnaire; TTI, Teacher Telephone Interview.
8 | 2015 | VOLUME 1 www.nature.com/nrdp
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Table 2 | A selection of open access resources for assessing attention-deficit/hyperactivity disorder in adulthood
Approach Comments Websites
Interviews
Diagnostic Interview for Adult • A structured diagnostic interview for ADHD in adults according to http://www.divacenter.eu/
ADHD, second edition (DIVA 2.0) DSM‑IV DIVA.aspx
• A new version based on DSM‑5 criteria is in press
Adult (ACDS) v1.2 • A semi-structured interview of current symptoms of ADHD in adults Available from the author (Lenard
• Provides age-specific prompts for rating both childhood and adulthood Adler) at: http://www.med.nyu.edu/
symptoms biosketch/adlerl01
Scales
Adult ADHD Self-Report Scale • Developed by WHO to measure ADHD symptoms in individuals http://www.hcp.med.harvard.edu/
(ASRS) >18 years of age ncs/asrs.php
• An 18‑item version covers all DSM‑IV symptoms of ADHD
• A 6‑item version is a screening tool validated for adolescents and adults
• The 6‑item version (ASRS-Telephone Interview Probes for Symptoms;
ASRS-TIPS) uses semi-structured interview probes for examples of
ADHD symptoms
• Both versions have been translated into many languages
Adult ADHD Investigator • Incorporates suggested prompts for each ADHD item Available from Lenard Adler at:
Symptom Rating Scale (AISRS) • Descriptors for each ADHD item are explicitly defined http://www.med.nyu.edu/
• Takes context into account biosketch/adlerl01
Wender Utah Rating Scale (WURS) • Developed to retrospectively diagnose childhood ADHD in adults Available from the authors255
ADHD, attention-deficit/hypersensitivity disorder; DSM, Diagnostic and Statistical Manual of Mental Disorders.
very strong evidence that ADHD frequently persists satisfactory grades, a university student with ADHD
into adulthood, although its presentation changes with might need to work twice as hard as peers with the same
age23,117–119 (FIG. 5). Nevertheless, ADHD in adults is still aptitude to focus attention or to organize school work.
undertreated120, leading to international efforts to edu- If that restricts the student’s social life or causes other
cate clinicians (TABLE 3) and to drive changes to DSM. problems, it might be viewed as impairing. Nonetheless,
DSM‑5 provides guidance about the differential expres- ADHD can be reliably diagnosed in these patients124.
sion of ADHD symptoms throughout the patient’s Finally, in adults with ADHD, hyperactive–impulsive
lifetime. For instance, in contrast to young children, symptoms usually become internalized, such as feeling
adults with many impairing ADHD symptoms do not restless, and deficient emotional self-regulation125 and
typically climb on tables, have boundless energy or executive dysfunction126 become increasingly promi-
run around in a place where one should remain still. nent. Although deficient emotional regulation and
Hyperactivity in adulthood is often experienced as a executive dysfunction are not diagnostic for ADHD,
feeling of inner restlessness — an internal ‘motor’ that they are highly characteristic of the disorder in adults
never stops — which makes it difficult for the indivi and could indicate the need for specific treatments,
dual to relax 121. By adopting symptom descriptors of such as cognitive–b ehavioural therapy, to improve
this sort, DSM‑5 is easier to apply to adults compared organizational or emotional self-regulation skills.
with its predecessors.
Despite these differences in symptom presentation, Heterogeneity of ADHD
the diagnostic process for adults parallels the process for Patients with ADHD show marked variation in profiles
youths in regards to documenting symptoms, impair- of symptoms, impairments, complicating factors, neuro
ment and onset of the disorder on the basis of a clinical psychological weaknesses and underlying causes127.
interview with the patient and, when available, reports Accordingly, effective partitioning of this heterogeneity
from informants. This process is aided by the avail- to refine diagnostic approaches and to provide tailored
ability of structured diagnostic interviews, such as the and targeted treatments remains an important research
Conners’ Adult ADHD Diagnostic Interview 122, along goal. To address this aim, DSM‑5 recognizes three
with rating scales for patients and informants, including presentations: predominantly inattentive, predomi-
the Adult Self-Report Scale (TABLE 2)123. nantly hyperactive–impulsive and combined. These
In adulthood, additional domains of impairment presentations are no longer deemed ‘subtypes’, as in
emerge and can include difficulties related to occupa- prior versions, because they can change over time128.
tion, marriage and parenting. Patients with high intelli Moreover, even within presentations, patients greatly
gence also present with a unique set of challenges. In differ in symptom profiles. For instance, the predomi-
these individuals, impairment can be assessed relative nantly inattentive presentation applies to individuals
to their aptitude. Some of these patients go to great with a wide range of inattention and can include sub-
lengths to accommodate their symptoms, which itself threshold hyperactive–impulsive symptoms. Although
indicates impairment to the degree that it causes distress common in population samples, inattentive ADHD
or displaces other activities. For example, to achieve is less common in the clinic, which suggests that
NATURE REVIEWS | DISEASE PRIMERS VOLUME 1 | 2015 | 9
© 2015 Macmillan Publishers Limited. All rights reservedPRIMER
Behavioural disinhibition, emotional ability Full expression of ADHD, Inattention persists and
and emergence of diagnosis in preschool years psychiatric co-morbidity, hyperactive–impulsive symptoms wane
school failure, peer
Prodrome: hyperactivity; and speech, language rejection and Smoking Substance abuse, low self-esteem and
and motor coordination problems neurocognitive dysfunction initiation social disability
In utero Childhood Adolescence Adulthood
Genetic predisposition Psychosocial influences, chaotic family environments, peer influences and mismatch with school and/or work environments
Fetal exposures Different genetic risk factors affect the course of ADHD at different stages of the lifespan
and epigenetic
changes Frontal–subcortical–cerebellar dysfunction via structural and functional brain abnormalities and downregulation of
catecholamine systems that regulate attention, reward, executive control and motor functions
Clinical progression
Aetiology Persistence of cortical thickness, default-mode
Pathophysiology network and white matter tract abnormalities
Naturecases.
Figure 5 | Developmental course of attention-deficit/hyperactivity disorder in persistent Reviews | Disease
Although noPrimers
single
sequence of events describes the pathway from in utero to adulthood, this figure describes key developmental events,
with boxes spanning their approximate onset along with hypotheses about the timing of the biological underpinnings
of aetiological events and pathophysiological expression. ADHD, attention-deficit/hyperactivity disorder.
population screening for marked inattention should be The heterogeneity of ADHD has implications for
considered, especially in female children and adults, in both research and practice. In research, the diluting
which this pattern might be particularly impairing 129. effect of heterogeneity reduces effect sizes in ADHD
Persistent inattention — even at subthreshold levels — is case–control comparisons and renders biomarkers
a key predictor of poor academic outcomes130. that are identified on the assumption that ADHD is
Psychiatric co-morbidity is another clinically impor- pathophysiologically homogeneous obsolete. Clinically,
tant dimension of ADHD heterogeneity. At one extreme, heterogeneity means that tests — either neuropsycho
a small proportion of clinic-referred individuals are free logical or tests of other underlying processes — that
of co-morbidity; at the other end, some patients have a focus on only one domain will be of very limited diag-
complex pattern of multiple problems, including com- nostic value. However, such assessments could help to
munication disorders, intellectual disabilities131, sleep identify specific targets for therapeutic and educational
disorders132, specific learning disabilities131, mood dis- interventions that are aimed at remediating particular
orders131, disruptive behaviour 131, anxiety disorders131, areas of impairment and weakness. For instance, indivi
tic disorders131, autism spectrum disorders131,133 and sub- duals with working memory deficits might respond
stance use disorders131,134,135. Consideration of a patient’s favourably to working memory training 138.
co-morbidity profile is important, as it will influence
treatment planning. Management
Pathophysiological heterogeneity might be impor- By educating patients and families, clinicians can cre-
tant clinically — although new research is required to ate a framework that increases treatment adherence,
determine whether subtyping on the basis of genetic, proactively plans for continuity of treatment through-
environmental, neurobiological or neuropsycho- out the lifetime of the patient and effectively integrates
logical factors will improve diagnostic and treatment pharmacological and non-pharmacological approaches.
approaches. In this regard, the largest body of evidence Education includes information about the causes of
relates to cognition. Objective tests indicate that sev- ADHD, its associated morbidity, the potential for a
eral distinct deficit profiles exist. For example, only a compromised course, the rationale for treatments and
minority of patients show a deficit in executive func- plans for key life transitions139. This education sets
tion136, which was once thought to be the core deficit in the stage for managing ADHD within a chronic care
ADHD. Other patients, who are clear of such deficits, paradigm that uses shared decision making to bol-
have problems in non-executive cognitive processes, ster treatment adherence and prepare patients for
which include those involved in basic memory and developmental challenges140.
temporal processing, motivational processing (delay There are geographic variations in the sequencing
tolerance or reinforcement processing) and cognitive of pharmacological and non-pharmacological treat-
energetic regulation72,73,137. Four cognitive ADHD sub- ments. For example, in the United States pharmaco
types were revealed in a study based on a community logical treatment is typically the first approach,
of children with or without ADHD70; however, whether whereas in Europe medication is usually reserved for
these subtypes predict treatment response or course severe cases or for milder cases that do not respond to
remains unclear. non-pharmacological treatments141.
10 | 2015 | VOLUME 1 www.nature.com/nrdp
© 2015 Macmillan Publishers Limited. All rights reservedYou can also read
