Impulsiveness means action without reflection—often described as a failure to ‘stop and think’. The term covers premature, unprepared and poorly timed behaviours—such as interrupting others, and giving too little time to appreciate what is involved in a school task or a social situation.

Inattentiveness means disorganized and forgetful behaviour: short-sequence activities, changing before they are completed, with a lack of attention to detail and a failure to correct mistakes. All these are behavioural observations, not psychological constructs. At a cognitive level, ‘Attention deficit’ is a rather poor descriptor; the performance of affected children does not fade with time on a task any more than that of ordinary people and the presence of irrelevant information (‘distractors’) does not worsen their performance disproportionately to that of other people.⁽³⁾ There are cognitive changes of a different kind (see ‘Aetiology’ below); but the diagnosis of inattentiveness depends on descriptions and observations of behaviour rather than on tests of performance.

Many other behavioural changes characterize some children with ADHD. They are, for instance, often irritable and their emotions can flash very rapidly when provoked. They may sleep badly (and this in turn can contribute to poor concentration). They can be aggressive to other people and non-compliant to authority. They can also be charming, humourous, inquisitive and intuitive. None of these, however, are either constant in ADHD or confined to those with ADHD. They are worth noting, but they do not make the diagnosis.

The behaviours of ADHD are continuously distributed in the population (see ‘Epidemiology’). The level that is considered normal or acceptable will vary from one culture to another and from one rater to another. To be diagnosed, they should be excessive not only for the child’s age but also for the developmental level; and this demands considerable familiarity with the usual range of variation. The diagnostician will acquire this in the course of training and experience; experienced teachers will be excellent judges; but inexperienced or overstressed parents may identify the problems at a low level of hyperactive behaviour, or suppose that an abnormal level is only to be expected in childhood. It is usually helpful to obtain a detailed behavioural account rather than rely on an overall judgment of ‘overactivity’ or ‘failure to concentrate’. Contradiction between sources may occur, and leads to arguments between parents and teachers. This may be due to different expectations, the emotional relationship of raters with the child, or children behaving very differently in contexts that vary in the demands placed on the children. The clinician needs to understand the full context of the way involved adults describe the child.

It is commonplace for children whose problems meet the criteria for ADHD to show other patterns of disturbance as well. This is often, confusingly, called ‘comorbidity’ – confusing because it assumes that the other pattern is a distinct disorder, which is only one of the explanations for coexistent problems. Clinicians need to understand the relationships for two reasons; so that they do not make or miss the diagnosis of ADHD; and so that they can make good strategies for treating ADHD in the presence of other disorders and other disorders in the presence of ADHD (see ‘Treating Complex cases’, below).

The commonest association and the best researched is with conduct and oppositional disorders. Nearly half the children with hyperactive behaviour in a community survey showed high levels of defiant and aggressive conduct as well; but the associations of the two problems were different, with hyperactivity (but not conduct problems) being associated with delays in motor and language development.⁽⁸⁾ Genetic research indicates higher heritability for hyperactivity than for defiant and aggressive behaviour; but there are some genetic influences that are common to both.⁽⁹⁾ Hyperactivity is more responsive to stimulant medication than are less hyperactive forms of conduct problem.⁽¹⁰⁾

When both ADHD and conduct problems are present, then the combined diagnosis (‘hyperkinetic conduct disorder’ in ICD-10) shows the associations of both disorders. ADHD is therefore not to be diagnosed by the absence of conduct disorder features, but by the clear presence of the core problems of inattentivess and disorganization.

A different kind of differential is presented by children with Tourette disorder. Their motor restlessness may indeed represent the coexistence of ADHD, but can result directly from tics. If a child’s tics are very frequent and there are a large number of them, then their repetitive and stereotyped nature may not be apparent and they may be seen simply as restless fidgetiness. Again, direct observation of the pattern of overactivity is the key. When there is doubt, videorecording the child and subsequent slow-motion review may make repetitive patterns evident.

Children with autism have clear and characteristic impairments of language, communication, and social development. Spectrum disorders, however, can raise diagnostic challenges. Children with ADHD alone often show language delays (usually of an expressive nature with over-simple utterances, by contrast with the receptive difficulties and idiosyncratic patterns of autism). Their attention difficulties may make them unresponsive to the overtures of others in a way that can simulate the social obliviousness of people in the spectrum of autism, and they are often friendless—not because of lack of interest in others but because of the capacity of hyperactive behaviour to irritate other people. Indeed, attention problems can extend to perseverativeness on certain activities such as video games that may be mistaken for the restricted interests of autism. All these factors can lead to ADHD being mistaken for autism, but the reverse can happen too. There are other reasons for overactive behaviour in autism. First, stereotyped patterns of driven overactivity can be seen: they are not disorganized or impulsive and are often made worse by change and novelty (which usually reduce the overactivity of ADHD). Second, episodic bursts of extreme activity can be seen and may be best regarded and treated as catatonic. Third, akithisia may result from neuroleptic medication, or irritable
restlessness from anticonvulsants, and it will be necessary to establish a clear history that ADHD has been a persistent trait.

Reactive attachment disorder (RAD) may share with ADHD a disinhibited style of relating to other people (an unreserved but shallow making of social contact). Children with RAD, however, tend to be controlling rather than disorganized, and vigilant rather than inattentive; and inattention and impulsiveness are not cardinal features of RAD; so it is not difficult to recognize both patterns when present in an individual child. The confusion in practice often comes from theoretical misconceptions. Those caring for neglected or abandoned children may consider that the diagnosis of ADHD cannot be accurate because the cause of the children’s problems is clearly to be found in their early deprivation. The causal pathway may indeed be that of neglect (though genetic inheritance and fetal exposure to toxins also need considering); but ADHD is a descriptive category, not an explanatory one. If the pattern of ADHD is present it still needs recognizing—not least because the cause of the ADHD behaviour does not seem to determine the response to stimulant medication, and children who have encountered neglect or abnormal early attachment may still have their ADHD problems reduced by medication.⁽¹⁰⁾

Both ADHD and manic conditions are characterized by overactivity, overtalkativeness, a sensation of whirling thoughts, and often by irritable mood. The distinction is made by the presence in bipolar disorder of episodicity, euphoria, and grandiosity. A suggestion that these distinguishing features are not in fact present in childhood bipolar disorder has naturally led to great overlap between the expanded childhood bipolar diagnosis and ADHD with poor emotional regulation, and further research will be needed to clarify whether there is a distinction.

In all these differential diagnoses, the principle is to establish that the child shows not only overactive behaviour, but the specific pattern of ADHD. Experienced judgement may be required, and the practice of diagnosing on the basis of questionnaire scores alone risks overidentification.

In adult life, there are still more possibilities for misdiagnosis. The commonest reasons for uncertainty are in distinguishing from atypical bipolar disorder and the effects of substance misuse. ‘Personality disorder’ is sometimes applied; and indeed ADHD shares with personality disorders a long-standing trait quality, but can also be a more precise way of describing the difficulties presented. Differentiation from the normal range of variation can be difficult in the absence of clear standards. The task of the diagnostician is harder when adults are presenting for the first time if only self-report is available; the self-description of hyperactivity may be a form of self-depreciation.

Questionnaire ratings by parents or teachers are very useful for screening purposes, and in group studies they give a fairly good discrimination between people with a clinical diagnosis of ADHD and controls from the ordinary population.⁽¹¹⁾ Many are available^(12,13) and the most famous are those from Conners, which yield several different scoring systems; and derivatives such as the Iowa Conners, the SWAN and SNAP scales.⁽¹⁴⁾ They do however leave a fair number of individuals misclassified, and are not suitable as the sole means of establishing a diagnosis. A detailed interview with parents establishes what actual behaviours are the basis for ratings, allows professional judgement to be included, and remains the most informative single method.

Interview with the child is valuable for the observation of attention and social interaction that it yields, and for understanding a child’s view of their predicament. Children, however, are not good witnesses about their own concentration and impulse control, and even affected adults are not good at describing themselves in these terms. The experience of ADHD is usually one of suffering the reactions evoked from other people, or an experience of repeated failure. Adults often describe an experience of whirling and interrupted thoughts (in the absence of manic features); and some children will say the same, especially if treatment has enabled them to make a comparison with another way of being.

Assessment needs not only to distinguish ADHD from related disorders, but to consider whether the ADHD pattern may result from remediable causes. The anamnestic history is by far the most productive investigation. It should include whether hearing problems have been excluded by previous testing (and, if not, an expert assessment should be arranged), and any injuries or diseases are potentially damaging the brain. The strengths and weaknesses of the family environment need to be assessed; they may dictate the choices of treatment. Physical examination should be sufficient to detect congenital anomalies, skin lesions, and motor abnormalities that can be the pointers to a neurological cause. Psychometric assessment is desirable whenever there are problems at school, both to generate an idea of developmental level against which the ‘developmental inappropriateness’ of behavioural symptoms can be judged, and to detect barriers to learning that may be the reason for inattentiveness. Special physical investigations are not routinely necessary. EEG often yields evidence of immaturity, but this does not advance assessment much and is not routinely indicated. It is valuable in the investigation of epilepsy and in the rare cases when deterioration of function suggests the possibility of a degenerative disorder. Blood tests should be planned only on the basis of history and examination, but may include tests of thyroid function, lead (in high-lead areas) chromosomal integrity (including fragile-X probe) when there is other evidence of developmental delay, and specific DNA tests when there is clinical suspicion of a phenotype such as that of Williams syndrome.