The following reports are based on a symposium held in New York City at the annual meeting of the American Academy of Child and Adolescent Psychiatry.

Psychiatrists increasingly face the challenge of diagnosing attention-deficit/ hyperactivity disorder (ADHD) in adult patients. Despite the recent introduction of new clinical rating scales for adult ADHD, a number of controversies continue to complicate the diagnosis of this disorder. The DSM-IV criteria for ADHD, for example, were developed for children and adolescents and cannot always be applied to adults. In addition, the relationship between self-rated symptoms of ADHD and reports of symptom severity provided by parents, siblings, or other informants remains uncertain, as does the relationship between childhood and adult symptoms of ADHD.

At the annual meeting of the American Association of Child and Adolescent Psychiatry, investigators discussed the development of several clinical rating scales for adult ADHD and described the ways in which these scales can be used to accurately evaluate symptoms.

THE CONNERS ADULT ADHD RATING SCALE

Although several behavioral rating scales of ADHD are available for pediatric use, few analogous scales have been developed for use in adults. C. Keith Conners, PhD, Professor Emeritus of Psychiatry and Behavioral Sciences at Duke University Medical Center, described the development and validation of a recently-developed clinical rating scale, the Conners Adult ADHD Rating Scale (CAARS).

Dr. Conners and colleagues began their efforts by identifying nine symptom domains thought to encompass the range of ADHD manifestations in adults. Three domains—inattention/concentration, hyperactivity/restlessness, and impulsivity/self-control—were based on the core symptoms of ADHD as it typically appears in children. Other symptom domains that the researchers believed to be of potential interest in adult patients included executive function (eg, self-regulation, prioritization of work, awareness of time); memory; difficulties with self-image; interpersonal problems; learning problems (especially during adulthood); and mood disturbances (eg, irritability, frustration, or overreaction to stressful events).

The researchers created a pool of rating scale items for each of these nine domains. In developing the items, the researchers drew on several sources, including the diagnostic criteria for ADHD specified in the DSM-IV; clinical experience with self-referred adults and with the parents of children with ADHD; publications and instruments from other researchers, such as the Wender Utah Rating Scale; and popular books on the subject of ADHD in adulthood. The final list comprised 93 questions, roughly 10 items per symptom domain studied.

The investigators administered the questions to 167 adults with ADHD and 839 normal subjects. A factor analysis of the responses revealed four distinct factors, which were labeled Inattention/Cognitive Problems, Hyperactivity/ Restlessness, Impulsivity/Emotional Lability, and Problems with Self- Concept. According to Dr. Conners, the Inattention/Cognitive factor is analogous to the inattention that characterizes ADHD in children, although in adults it encompasses a variety of cognitive problems, including difficulties with executive functions and with starting and completing tasks. The Hyperactivity/Restlessness factor encompasses not only motor hyperactivity but also feelings of inner restlessness, distractibility, risk taking, and a tendency to become bored easily. The Impulsivity/Emotional Lability factor resembles childhood impulsivity but also includes impulsive verbal outbursts, "hot temper," stress intolerance, irritability, and labile mood.

Dr. Conners noted that these four factors were identified in both male and female patients. Among subjects ages 18 to 29, men exhibited significantly higher levels of ADHD symptoms than did women; symptom levels declined with increasing age in both sexes, although the rate of decrease was greater in men than in women. Mean symptom scores of men and women converged around age 29 and were more or less equivalent at older ages.

To assess the validity of the CAARS scale, the researchers determined how well the test predicted whether an individual would satisfy the diagnostic criteria for a DSM-IV diagnosis of ADHD. By these criteria, the CAARS produced a false-positive rate of about 13% and a false-negative rate of 18%; the overall correct classification rate was 85%. "To put those numbers in perspective, there are very few laboratory tests that will give you an overall classification rate better than 80% or 90%," Dr. Conners noted. However, he cautioned that the misdiagnosis rate of nearly 15% suggests that adults who are suspected of having ADHD cannot be adequately evaluated using the CAARS alone.

IMPAIRMENT OF EXECUTIVE FUNCTION: ADHD IN A NUTSHELL?

Thomas E. Brown, PhD, Assistant Clinical Professor of Psychiatry at Yale University School of Medicine, in New Haven, Connecticut, described the development of an alternative approach to the diagnosis of ADHD in adults—the Brown Attention Deficit Disorder Scale—and discussed the use of this test, in combination with other standardized measures, to evaluate suspected cases of ADHD. One version of this test has been developed for use with adults and another for adolescents; both emphasize the role of executive function impairment in ADHD. There are no items measuring hyperactivity on this scale; these symptoms can be assessed using the DSM-IV or the CAARS.

Dr. Brown noted that adults who are being evaluated for ADHD present with a variety of complaints—including concentration problems, disorganization or distraction, and difficulty initiating tasks, working consistently, and attending to details—and many of the deficits involve impairment of executive functioning. Using interviews with a number of people who had been evaluated and received a diagnosis based on existing criteria, he and his colleagues attempted to identify a core group of symptoms that affect most adults with ADHD and that these individuals describe as particularly troublesome.

One striking aspect of ADHD in adult patients, Dr. Brown continued, is that there is often a pronounced situational variability in the severity of symptoms. "Every patient that I've ever seen with attention deficit disorder has a few domains of activity in which they are able to concentrate perfectly well and where they experience very little impairment on those cognitive functions which are highly problematic for them in many other areas," he said. This suggests that these individuals usually do not have impairments in individual cognitive functions but rather in how these functions are managed and organized at a higher level. "ADHD is essentially a name for developmental impairment of executive function," he explained.

According to Dr. Brown, there are five important symptom clusters that are consistently reported by adults with ADHD. These are: (1) problems getting organized, prioritizing, and activating tasks; (2) difficulty sustaining focus and attention, especially with reading; (3) trouble sustaining alertness, effort, processing speed, and motivation; (4) issues related to affect, such as preventing anger from getting out of control (a problem not included in the DSM-IV definition of the disorder); and (5) deficits in working memory (the ability to hold something in mind while doing something else) and memory retrieval. "These clusters are picking up on what constitutes a decent description of executive function," Dr. Brown said.

DOWNPLAYING HYPERACTIVITY

Dr. Brown conceded that this conceptualization of adult ADHD has been the subject of some controversy because it places relatively little emphasis on hyperactivity. However, he said, these symptom clusters represent the types of problems that adults with ADHD are most troubled by when they seek help for their disorder. Further, this emphasis on executive functioning may explain why many people do not appear to develop ADHD symptoms until the middle of elementary school or later: Because the use of executive function and the demands made upon it increase with maturation, deficits might not become apparent until later in life.

When evaluating an adult for ADHD, standardized tests can be an important source of information. Dr. Brown described the results of a study that examined a group of 176 adults, ages 16 to 69, who were seeking treatment for ADHD. All of these individuals met DSM-IV criteria for ADHD, 55% with inattentive type and the remainder with combined type. The patients completed the full Wechsler Adult Intelligence Scale-Revised (WAIS-R), as well as the logical memory subtest of the Wechsler Memory Scale (in which the clinician reads two short text passages and the patient is asked to recite them in as close to the original wording as possible) and the Verbal Comprehension Index (in which patients undergo IQ testing, and three subscales that are most sensitive to concentration are removed and examined separately as a concentration index).

The results support the view that impairment of executive function is an important component of ADHD. Eighty-seven subjects (49%) exhibited a discrepancy of at least one standard deviation between their WAIS-R verbal IQ score and their score on the concentration measure; among normal subjects, this magnitude of discrepancy was noted in only 21%. Similarly, a discrepancy of at least one standard deviation between the verbal IQ and prose memory test scores was noted for 66% of ADHD subjects and only 15% of controls. Thus, although not a definitive measure of ADHD for all patients, a discrepancy between verbal IQ and measures of concentration or memory is one factor that should be considered when evaluating an adult who may have ADHD.

DO ADULT SYMPTOMS MATCH MEMORIES OF CHILDHOOD SYMPTOMS?

One important unresolved issue in adult ADHD is whether the disorder is always a continuation of a childhood disorder or whether ADHD can develop in adulthood without childhood precedent. Brian Greenfield, MD, Assistant Professor of Psychiatry at McGill University in Montreal, discussed the results of recent work examining the relationship between adult self-reported symptoms of ADHD and retrospectively recalled childhood symptoms in the same individuals.

There is good reason to suspect that there may be a relatively poor relationship between ADHD symptoms as they actually occur during childhood and later recollection of those symptoms in adulthood. Dr. Greenfield cited a longitudinal study by Daniel Offer, MD, and colleagues at Northwestern University Medical School, Chicago, in which a cohort of mentally healthy individuals was evaluated during adolescence with respect to a number of psychosocial issues and reinterviewed several years later, in adulthood. The investigators found a large discrepancy between events experienced during adolescence and the subsequent recollection of those events, a result that calls into question the usefulness of retrospective questionnaires to evaluate psychiatric disorders having their origins in childhood and adolescence, Dr. Greenfield said. However, he noted that there may be opposing trends in this population. On the one hand—and in line with Dr. Offer's observations—the executive dysfunction seen in adult ADHD patients may put them at increased risk of forgetting their childhood symptoms. On the other hand, these symptoms may have been so severe that patients might be expected to have little difficulty recalling them accurately.

Thus, in a new study, Dr. Greenfield and his colleagues compared current self-reported symptoms of ADHD with retrospective self-reports of ADHD symptom severity in childhood. A total of 140 subjects ages 18 to 60 were tested using several measures of ADHD symptoms, each consisting of 18 to 25 questions, as well as the 93-item Conners scale. The subjects received two test batteries, one for symptoms in adulthood and one for childhood symptoms.

The investigators compared adult and childhood ratings of ADHD both for the same test and for different measures. In addition, symptoms were evaluated two different ways—as either a categorical variable (ie, each patient was classified as either having or not having ADHD on the basis of rating scale scores) or as a continuous variable.

In general, correlations using categorical diagnoses were rather weak between adult ratings of ADHD and retrospective ratings of childhood ADHD, especially for the comparisons between different rating scales. For example, the correlation between scores on the Conners rating scale in adulthood and Wender rating scale for childhood symptoms was only 0.15. Although the correlations were substantially higher when the same rating scale was used for childhood and adult symptoms, and better still when ADHD was treated as a continuous variable, Dr. Greenfield suggested that overall the findings raise doubts regarding the validity of retrospective recall of childhood symptoms and of the importance of childhood symptoms in diagnosing ADHD in adults.

CONCORDANCE BETWEEN SELF-RATINGS AND OTHER RATINGS

The symposium's next panelist, Lily Hechtman, MD, Professor of Psychiatry at McGill, discussed the importance of obtaining a complete diagnostic evaluation of adults suspected of having ADHD, including reports of symptoms from family members or other informants. Adults with ADHD frequently have poor insight into their symptoms, she noted, and may have a variety of comorbidities; these can present a considerable diagnostic challenge because it can be difficult to tell whether symptoms are caused by ADHD or by comorbid substance abuse, depression, or anxiety. "Because of these difficulties," Dr. Hechtman said, "we have engaged in a very extensive diagnostic workup for these patients," including the use of several standardized tests (eg, the Wender and Conners scales); these are filled out by both the patient and an informant, and with regard to childhood as well as current symptoms.

Other components of the diagnostic evaluation include an adult ADHD rating scale developed by Russell Barkley, MD, of the University of Massachusetts Medical School; the Conners Continuous Performance Test; the Stop Signal Task, which gauges the ability to inhibit behavior; and various measures of comorbidity, including the Symptom Checklist-90, the Beck Depression Inventory, and the Hamilton Rating Scale for Anxiety. The assessment also includes IQ testing; measures of freedom from distractibility, such as the Digit Span Test; a semistructured interview for other comorbidities; and a marital adjustment scale for patients who are married or in a long-term relationship.

In a recent study, Dr. Hechtman and colleagues compared the severity of self-reported ADHD symptoms—for both adulthood and childhood—with reports provided by parents, siblings, or other informants. Most of the 131 patients were self-referred. Three clinicians reviewed all available patient information and arrived at a consensus diagnosis for each subject; ADHD was diagnosed in all but 27 cases.

The retrospectively recalled severity of childhood ADHD symptoms by the patients was significantly greater than the symptom severity reported by the informants for both the Wender and Barkley rating scales, Dr. Hechtman and colleagues found. The correlations between "self" and "other" symptom ratings were about 0.7. For the Wender scale, the patient and informant ratings of childhood symptoms correlated only weakly with the consensus diagnosis, though the correlations did reach statistical significance. For the Barkley scale, only the correlation between self-ratings and consensus diagnosis was statistically significant; the correlation between the informant ratings and the consensus diagnosis was not.

For current symptom severity, self-reported ratings were greater than those reported by informants, regardless of the scale used. Once again, the correlation between self-ratings and the consensus diagnosis was higher than the correlation between the informant rating and the consensus diagnosis for all of the measures examined; correlation coefficients were generally in the 0.3 to 0.5 range for these comparisons.

Thus, for both ratings of current symptom severity and retrospective recall of symptoms during childhood, the self-ratings were more highly correlated with the consensus diagnosis of ADHD than were ratings by informants such as parents or siblings. Dr. Hechtman noted that this may reflect the fact that these were primarily self-referred patients who were seeking help for ADHD and who may have been more aware of their symptoms than are many patients with ADHD.

In conclusion, Dr. Hechtman noted that a thorough clinical evaluation is essential in making the diagnosis of ADHD in adults. Standardized tests are helpful but cannot substitute for clinical judgement. Finally, at least in self-referred adults, self-reports of symptom severity provide information that is equal or superior to that provided by parents, siblings, or other family members.

—Mark Bowes, PhD

Suggested Reading
1. Searight HR, Burke JM, Rottnek F. Adult ADHD: evaluation and treatment in family medicine. Am Fam Physician. 2000;62:2077-2086, 2091-2092.
2. Spencer T, Biederman J, Wilens TE, Faraone SV. Adults with attention-deficit/ hyperactivity disorder: a controversial diagnosis. J Clin Psychiatry. 1998;59(suppl 7):59-68.

Return to table of contents