|
|||||||||||||
|
|||||||||||||
This is an intermediate level course. After completing this course, mental health professionals will be able to:
NOTE: This course is adapted from my recently published textbook with Kevin Murphy and Mariellen Fischer entitled ADHD in Adults: What the Science Says (New York: Guilford; Guilford.com, 2008).
The history of ADHD is extensive for the childhood stage of the disorder and is discussed in detail in the earlier textbook by Barkley (2006). Far less exists concerning the history of ADHD in adults, largely because ADHD was widely held to be a disorder strictly of childhood for most of the past century. While popular interest in the possibility that adults can have attention deficit hyperactivity disorder (ADHD) most likely originated with the bestseller, Driven to Distraction, published in 1994 by psychiatrists Edward Hallowell and John Ratey, clinical and scientific papers acknowledging the existence of an adult version of this disorder date back at least 40 years and possibly even a century ago.
In his series of three published lectures to the Royal College of Physicians, George Still (1902) described 43 children in his clinical practice having serious problems with sustained attention and in the moral control of their behavior. By the latter symptom, Still meant the regulation of behavior relative to the moral good of all. He viewed the latter construct as a conscious comparative process in which one evaluates both the present and likely future consequences of one’s actions for both one’s self and for others prior to choosing a course of action. Most of his cases were not just inattentive and lacking in forethought but also were quite overactive. He proposed that the immediate gratification of the self was the “keynote” quality of these and other attributes of the children. In addition, among all of them, passion (or heightened emotionality) was the most commonly observed attribute and the most noteworthy. Still noted further that a reduced sensitivity to punishment characterized many of these cases for they would be punished, evenly physically, yet engage in the same infraction within a matter of hours. Still believed that the major “defect in moral control” so typical of these cases was relatively chronic. While it could arise from an acquired brain defect secondary to an acute brain disease, and might remit on recovery from the disease, in most cases it was chronic. Here then likely arises the first mention that ADHD may persist into adulthood thereby logically opening the door to the possibility that adults can possess this same pattern of symptoms dating back to childhood.
Acknowledging that George Still may have first signaled the possibility that adults could have ADHD as an outcome of its chronic childhood course, the first papers on actual adults having ADHD seem to date to the late 1960s. At that time, the disorder was known as Minimal Brain Damage or Dysfunction (MBD) and its likely existence in adults arose from three sources. The first of these was the publication of several early follow-up studies demonstrating the persistence of symptoms of hyperactivity/MBD into adulthood in many cases (Mendelson, Johnson, & Stewart, 1971; Menkes, Rowe, & Menkes, 1967). The second source was the publication of research showing that the parents of hyperactive children were likely to have been hyperactive themselves and to suffer in adulthood from sociopathy, hysteria, and alcoholism (Cantwell, 1975; Morrison & Stewart, 1971). Later papers would further confirm this familial association of hyperactivity in which the biological parents of such children were also abnormal in their attention, impulse control, and activity levels (Alberts-Corush, Firestone, & Goodman, 1986). This early work began to suggest that children with ADHD symptoms were likely to have parents with ADHD symptoms, logically implying that ADHD could therefore exist in adults.
The third source of evidence implying the existence of ADHD in adults was the publication of studies on adult patient samples that were believed to have hyperactivity or MBD. The first of such papers appears to have been that of Harticollis (1968) who focused on the results of neuropsychological and psychiatric assessments of 15 adolescent and young adult patients (ages 15–25) seen at the Menninger Clinic. The neuropsychological performance of these patients suggested evidence of MDB or moderate brain damage. Their behavioral profile suggested many of the symptoms that Still initially identified in his own child cases, particularly impulsiveness, overactivity, concreteness, mood lability, and proneness to aggressive behavior and depression. Some of the cases appeared to have demonstrated this behavior uniformly or consistently since childhood. Using psychoanalytic theory, Harticollis speculated that this condition arose from an early and possibly congenital defect in the ego apparatus in interaction with busy, action-oriented, successful parents. In short, an inborn error of cognition interacts with a particular pattern of child rearing by parents to result in the condition of MBD.
A year later Quitkin and Klein (1969) described two behavioral syndromes in adults that may be related to MBD. The authors studied 105 patients at the Hillside Hospital in Glen Oaks, New York, for behavioral signs of “organicity” (brain damage). They searched for behavioral syndromes that might be considered soft neurological signs of CNS impairment, as well as the results of electroencephalogram (EEG), psychological testing, and clinical presentation and history that might differentiate these patients from other types of adult psychopathology. They selected those cases having a childhood history that suggested CNS damage, including early hyperactive and impulsive behavior they believed may reflect the likelihood of such damage. These cases were further sorted into three groups based on current behavioral profiles: those having socially awkward and withdrawn behavior (N = 12), those having impulsive and destructive behavior (N = 19), and a “borderline” group that did not fit neatly into these other two groups (N = 11). Their results indicated that nearly twice as many of these “organic” groups had EEG abnormalities and impairments on psychological testing indicating organicity, as did the control group. Noteworthy for our purposes here was their finding that an early history of hyperactive–impulsive–inattentive behavior was highly predictive of placement in the adult impulsive–destructive group, implying a persistent course of this behavioral pattern from childhood to adulthood. Of the 19 patients in the impulsive–destructive group, 17 had received a clinical diagnosis of Character Disorder (primarily emotionally unstable types) as compared to only five in the socially awkward group (which were of the schizoid and passive dependent types).
These results were in conflict with the widely held belief at the time that hyperactive–impulsive behavior tended to wane in adolescence. Instead, the authors argued that some of these children continued into young adulthood with this specific behavioral syndrome. Quitkin and Klein (1969) also took issue with Harticollis’ psychoanalytic hypothesis that demanding and perfectionistic child rearing by parents was causal of or contributory to this syndrome given that their impulsive–destructive patients did not uniformly experience such an upbringing. In keeping with Still’s original position that family environment could not account for this syndrome, the authors hypothesized “that such parents would intensify the difficulty, but are not necessary to the formation of the impulsive–destructive syndrome” (p. 140) and that the “illness shaping role of the psychosocial environment may have been overemphasized by other authors” (p. 141). Treatment with a well structured set of demands and educational procedures as well as with phenothiazine medication was thought to be indicated.
The first paper to focus specifically on adult cases defined as MBD, as opposed to the earlier and more general concept of “organicity,” may have been that by Shelley and Reister (1972). These authors described 16 cases seen at an Air Force training base psychiatric clinic (ages 18-23) because of difficulties coping with their military basic training. These patients were described as having marked difficulties concentrating, being emotionally labile, fearing their loss of impulse control, and showing marked irritability as well as anxiety and self-depreciation. Problems with poor motor skills and sluggish reaction or response timing were noteworthy. While EEG and neurological exams were normal for gross findings of hard neurological signs, all showed evidence of “soft” signs of “neurointegrative disturbances” such as motor clumsiness, poor balance, confused laterality, and poor coordination. Psychological testing also revealed evidence of perceptual-motor problems and motor incoordination and timing. On history, 14 of the 16 cases reported difficulties with temper tantrums and low frustration tolerance as children, with 12 (75%) reporting behavior consistent with hyperkinetic behavior syndrome, among other early behavioral symptoms. Over the ensuing 30 years, these problems with motor development and coordination have been well documented in children with ADHD (Barkley, 2006).
The following year, Anneliese Pontius (1973) summarized her clinical observations of more than 100 adult cases of MBD. She proposed that many cases of adult MBD demonstrated hyperactive and impulsive behavior and that their disorder likely arose from frontal lobe and caudate dysfunction. Such dysfunction would lead to “an inability to construct plans of action ahead of the act, to sketch out a goal of action, to keep it in mind for some time (as an overriding idea) and to follow it through in actions under the constructive guidance of such planning” (p. 286). Moreover, if adult MBD arises from dysfunction in this frontal-caudate network, it should also be associated with an inability “to reprogram an ongoing activity and to shift within principles of action whenever necessary” (p. 286). She went on to show that indeed adults with MBD demonstrated such deficits indicative of dysfunction in this brain network. Such observations would prove quite prophetic. Two decades later, research demonstrated reduced size in the prefrontal–caudate network in children with ADHD (Castellanos, Giedd, Marsh, Hamburger, Vatuzis, et al., 1996; Filipek, Semrud-Clikeman, Steingard, Renshaw, Kennedy, & Biederman, 1997). And theories of ADHD argued that the neuropsychological deficits associated with it involved the executive functions, such as planning, the control of behavior by mentally represented information (working memory), rule-governed behavior, and response fluency and flexibility, among others (Barkley 1997a, 1997b).
Morrison and Minkoff (1975) would subsequently argue that adult patients with explosive personality disorder or episodic dyscontrol syndrome might well be the adult outcome of the hyperactive child syndrome. By 1976, Mann and Greenspan proposed that adults having MBD constituted a distinct diagnostic entity (adult brain dysfunction), which they illustrated with two of their clinical cases. They believed that MBD adults shared a basic impairment in attention, and that they were also likely to manifest problems with hyperactivity, impulsiveness, depression, and anxiety. They recommended the use of Leon Eisenberg’s (1973) behavior questionnaire for hyperactive child syndrome as part of the diagnostic workup; this was a rating scale actually developed by C. Keith Conners who at the time was working with Eisenberg and which would later become a mainstay of the evaluation of hyperactive children (Barkley, 1981). Mann and Greenspan also found that these symptoms were responsive to antidepressant medication (imipramine) or stimulants, echoing the same suggestion made earlier by Hans Huessy (1974) in a letter to the editor of a journal that both antidepressants and stimulants may be the most useful medications for the treatment of these hyperkinetic or MBD adults.
Around this time, the first truly scientific evaluation of the efficacy of stimulants with adults having MBD was conducted by Wood, Reimherr, Wender, and Johnson (1976). They used a double-blind, placebo-controlled method to assess response to methylphenidate in 11 of 15 adults with MBD followed by an open trial of pemoline (another stimulant), and the antidepressants imipramine and amitriptyline. The authors found that eight of the 11 tested on methylphenidate had a favorable response whereas 10 of the 15 tested in the open trial showed a positive response to either the stimulants or antidepressants. Others in this decade and into the next would also make the case for the existence of an adult equivalent of childhood hyperkinesis or MBD and the efficacy of using stimulants and antidepressants for its management (Gomez, Janowsky, Zetin, Huey, & Clopton, 1981; Mann & Greenspan, 1976; Packer, 1978; Pontius, 1973; Rybak, 1977; Shelley & Reister, 1972). Yet, it would not be until the 1990s that both the lay public and the professional field of adult psychiatry would begin to seriously recognize the adult equivalent of childhood ADHD on a more widespread basis and to recommend stimulant or antidepressant treatment in these cases (Barkley, 1994, 1998; Spencer, Wilens, Biederman, Faraone, Ablon, & Lapey, 1995; Wender, 1995). Even then, there remained skeptics concerning whether or not adults could have ADHD or should be treated for it (Shaffer, 1994).
Gomez and colleagues later published a study of 100 adult psychiatric patients of which 32% showed signs of childhood hyperactivity, attention deficits, and impulsivity compared to just 4% of a control group (Gomez et al., 1981). They reported that 20% also had symptoms consistent with adult hyperkinetic syndrome compared to none of their control group. The highest incidences of these symptoms were found in cases that traditionally would have been diagnosed as character disorder (47% had childhood and current signs of hyperkinetic syndrome). This study suggests that a sizeable minority of adults evaluated at a psychiatric clinic were likely to have a childhood history of hyperkinetic syndrome and that perhaps one in five currently did so at clinical evaluation.
Another historical development worth noting here is the initial development of diagnostic criteria for ADHD in adults. Besides initiating the first scientifically based study of medication treatment for ADHD noted earlier, Paul Wender was also the first to offer explicit criteria for the manner in which the diagnosis of ADHD in adults should be made. At the time, this position was at odds with prevailing clinical opinion that children outgrew the disorder. Wender (1995) recognized that diagnostic criteria proposed for the syndrome of childhood hyperactivity (as in DSM-II, American Psychiatric Association, 1968) or the later Attention Deficit Disorder (as in DSM-III, American Psychiatric Association, 1980), were not developmentally appropriate for adult patients. While both volumes recognized that ADHD might be a residual condition in some adults and was permitted to be diagnosed as such, the widespread existence of the full disorder in adults was not recognized at the time nor was explicit criteria provided for doing so. Based on his empirical work, Wender developed an approach for diagnosis of ADHD in adults (Wender, 1995) that was subsequently used in a number of research projects, especially medication trials. Patients and an additional informant, preferably a parent, are interviewed to assess retrospectively the childhood diagnosis of ADHD. Evidence is also obtained for ongoing, continued impairment from hyperactive and inattentive symptoms. Seven symptoms were proposed to characterize the phenomenology of adult ADHD, namely 1) inattentiveness; 2) hyperactivity; 3) mood lability; 4) irritability and hot temper; 5) impaired stress tolerance; 6) disorganization, and 7) impulsivity. Known as the “Utah criteria,” these diagnostic guidelines required a retrospective childhood diagnosis, ongoing difficulties with inattentiveness and hyperactivity, and at least two of the remaining five symptoms. Wender also developed a rating scale, the Wender Utah Rating Scale (WURS), to aid in the retrospective diagnosis of childhood ADHD (Ward, Wender, & Reimherr, 1993). The WURS is a self-completed report of retrospective childhood behavioral symptoms. To its credit, the Utah approach to adult ADHD established the need for retrospective childhood diagnosis, careful elucidation of current symptoms, and the routine use of third party informants of childhood and adult behavior. These stipulations have become standard practice for many clinicians and investigators, including ourselves.
Though clearly an advance in the diagnosis of ADHD in adults in view of their being no such guidelines prior to 1993 for doing so, Wender’s approach would later be argued to be problematic for ongoing research and clinical work (McGough & Barkley, 2004). With subsequent editions of the DSM, the Utah criteria have diverged further and further from current clinical conceptualizations of ADHD. It is difficult to apply knowledge derived from the study of child ADHD diagnosed by DSM to adults identified with alternative diagnostic schemes like the Utah criteria. By design, the Utah criteria include only individuals with lifelong inattention and hyperactivity, and therefore exclude patients with the predominately inattentive ADHD subtype. Conversely, although symptoms of irritability and hot temper were included in early conceptions of childhood ADHD, substantial research shows this dimension of behavior to be semi-independent of ADHD symptoms, to have different associated impairments, to be more closely associated with problems in the social environment, and to predict different developmental outcomes than do the symptoms of ADHD (Hinshaw, 1987; Loeber, Burke, Lahey, Winters, & Zera, 2000). Permitting symptoms of irritability and hot temper to qualify one for the disorder creates an automatic confound of ADHD with oppositional defiant disorder, conduct disorder, and possibly the mixed mood dysphoric form of bipolar disorder. Likewise, the inclusion of symptoms of mood lability without further clarification may further confound the delineation of this disorder from other mood disorders in adulthood. The Utah criteria exclude the diagnosis of ADHD with coexisting major depression, psychosis, or severe personality disorder. While these restrictions can be useful in research studies of medication response, they will fail to diagnose significant numbers of patients who are clearly impaired and would benefit from treatment. Studies indicate that a significant minority of children and adults with ADHD are likely to have major depression or dysthymia (20-27%) and personality disorders (11-24%) by adulthood (Barkley, 2006; Fischer, Barkley, Smallish, & Fletcher, 2002; Murphy & Barkley, 1996). Likewise, adults who self-refer to clinics specializing in adult ADHD may have even higher rates of anxiety disorders and depression than do children with ADHD followed to adulthood (Murphy & Barkley, 1996a; Shekim, Asarnow, Hess, Zauha, & Wheeler, 1990). A further problem with the criteria was the initial lack of adequate norms for adults on the WURS so as to more precisely determine an appropriate, empirically based cutoff score for developmental deviance of symptoms than those that were based on the developers’ clinical experience. For these reasons, the Utah criteria have declined in use among investigators and clinicians in favor of using the more current DSM-IV criteria. Later and better-constructed scales with adult norms – such as those developed for adults by Conners, Erhardt, and Sparrow (1998), or others that were more closely aligned with the DSM symptom lists, such as those by Brown (1996) and ourselves (Barkley & Murphy, 1998, 2006) – would offer DSM-based scales as alternatives to the WURS for clinical practice.
A further watershed moment in the history of adults with ADHD was the development of a nonstimulant medication, atomoxetine (Strattera®), by the Eli Lilly Company that would be investigated in thousands of adults with ADHD in several randomized, placebo-controlled trials. Initially, this drug was studied by the company as an antidepressant in approximately 1,200 adults. In this trial, neither atomoxetine nor desipramine was significantly different from the placebo, and so atomoxetine was shelved as an antidepressant. However, largely based upon the demonstrated efficacy of the tricyclic antidepressants in treating ADHD, mainly in children (see Spencer, 2006), a proof of concept study in adults with ADHD was encouraged by John Heiligenstein with the Eli Lilly Co. and subsequently initiated at Massachusetts General Hospital. In this study, Spencer and colleagues, using a double-blind placebo controlled design, demonstrated that atomoxetine was well tolerated and significantly more effective than placebo in reducing clinical symptoms of ADHD (Spencer, Biederman, Wilens, Prince, Hatch, Jones et. al., 1998). These initial positive findings led to the performance of two large multi-site trials of atomoxetine in adult ADHD, evaluating more than 536 adults with ADHD, that likewise proved the drug to be efficacious for ADHD management in adults (Michelson, Adler, Spencer, Reimherr, West, Allen et al., 2003). These studies are the largest ever done in evaluating a medication for adults with ADHD. More would follow.
Atomoxetine would be the first new drug developed for the management of ADHD in more than 25 years and would be the first ever approved for treatment of ADHD in adults by the US Food and Drug Administration. Later, stimulants (methylphenidate, mixed amphetamine salts) would eventually be studied more thoroughly as well for adults with ADHD (Spencer, Biederman, Wilens, Doyle, Surman, Prince, 2005; Spencer, Biederman, Wilens, Faraone, Prince, Gerard, et al., 2001) and receive similar FDA approval for use in this age group. New delivery systems have also been recently developed that permit greater sustained therapeutic action across the day than did immediate release preparations. These include osmotic pumps (Concerta®), variable timed-release pellets (Focalin XR®, Metadate CD®, Ritalin LA®, Adderall XR®, et al.), and skin patches (Daytrana®), besides the earlier available but clinically disappointing wax-matrix sustained release formulation of methylphenidate (Ritalin SR®). As of this writing, a new nonabusable formulation of a mixed amphetamine compound has received FDA approval for use with ADHD (Vyvanse®) in which the pills must be dissolved in stomach acid before the amphetamine compound can be activated and available for absorption through the gut.
Psychological treatments for adults with ADHD, though numerous in clinical practice, have to date not received much serious scientific scrutiny. This remains a glaring gap in the clinical scientific literature on the disorder in adults. Fortunately, this may be changing. Safren and colleagues (Safren, Perlman, Sprich, & Otto, 2005) have developed a group-delivered cognitive behavioral therapy program for adults with ADHD as a supplement to their medication treatment. Initial results from a small-scale study of this manualized therapy have been favorable in showing significant benefits from this treatment beyond those achieved by medication alone (Safren, Otto, Sprich, Winett, Wilens, & Biederman, 2005). More recently, Ramsay and Rothstein (2007) have developed a cognitive-behavioral therapy program for ADHD in adults. More such research is to be encouraged given that medication treatments are hardly likely to address all of the domains of impairment associated with ADHD, much less its frequent comorbid disorders, such as anxiety, depression, and learning disabilities.
Given the growing acceptance of ADHD as a legitimate disorder in adults by the time of this writing, one can rightly ask just how prevalent ADHD is among adults. We address that issue next.
An initial attempt at estimating prevalence was conducted by these authors in 1996 when we surveyed 720 adults (ages 17-84) in the central Massachusetts region who were renewing their driver’s licenses (Murphy & Barkley, 1996b). At that time, all drivers had to renew their licenses in person making this a fairly reasonable way of obtaining a representative sampling from the general adult population of this area, or at least of its adult drivers. These adults completed two rating scales based on the DSM-IV symptom list for ADHD; one scale was for current functioning and the second for their recall of their own behavior as children at ages 5-12 years. When we required that adults meet the DSM symptom thresholds for clinical disorder (6 of 9 symptoms rated as “often” or “more frequent”) for both current and childhood functioning (a rather stringent criterion), the prevalence for ADHD (all types) was 4.7%.
This figure is well within the 3-5% estimate conjectured above from the follow-up studies of children. But it is certainly higher than those found in three previous studies of college students. Weyandt, Linterman, and Rice (1995) reported a prevalence of 7% of 770 students reporting significant current symptoms of ADHD (+1.5 SD above the mean) while this figure fell to 2.5% if this threshold was imposed for both current and childhood functioning. DuPaul and colleagues found that 2.9% of men and 3.9% of women in their US college sample (N=799) met symptom thresholds (6 of 9) for current functioning (DuPaul, Schaughency, Weyandt, Tripp, Kiesner, Ota, & Stanish, 2001). In a third study (Heiligenstein, Conyers, Berns, & Smith, 1998), a prevalence of 4% of 448 students was reported based on current symptoms, very similar to those of the DuPaul et al. study. No data were available, however, in the DuPaul and Heiligenstein studies concerning childhood ADHD symptoms but it is likely, given the results of our own study and that of Weyandt and colleagues above, that their figures would be reduced by at least 50% or more if significant childhood symptoms also had been required for determining prevalence of disorder. This would likely reduce the prevalence figures for those two studies closer to 1.5 and 2% – again, very similar to those of Weyandt et al. Studies of college students would likely result in underestimates of true adult ADHD prevalence given the significant adverse impact ADHD has on educational functioning and eventual attainment such that the vast majority (approximately 80% or more) of children growing up with ADHD do not attend college (Barkley, 2006).
All of these prevalence figures may be an underestimate, however, given that they are based on the DSM symptom list and symptom thresholds that may be developmentally inappropriate and too severe, respectively, for use with adults. In addition, the more symptoms of ADHD one may have, the more one may underestimate the occurrence and severity of those symptoms, as shown above, further limiting this estimate. Yet another problem with this research could render these figures as overestimates. That is because no imposition of a criterion for having evidence of impairment in major life activities was used in any of these studies although it is required as part of the DSM diagnostic criteria for this disorder. In studies of children, when such a criterion is imposed, it can result in a significant reduction in prevalence (see Barkley, 2006). Our own research is further limited by restricting its sample to central Massachusetts, while Weyandt et al.’s sample was chiefly Washington State and DuPaul et al.’s samples were also regionally limited to three localities, making it difficult to extrapolate these figures to other regions of the United States. Indeed, it appears that a large part of the Weyandt et al. study sample (N=444) may have been included in the DuPaul et al. study that may limit the status of the DuPaul et al. study as a replication of the Weyandt et al. study.
The latter limitation has now been overcome as a consequence of a far larger study of prevalence of ADHD in an adult general population sample (Kessler, Adler, Barkley et al., 2006). In that study, a screen for adult ADHD was included in a probability sub-sample (n = 3,199) of 18-44 year-old respondents in the National Comorbidity Survey Replication (NCS-R), a nationally representative household survey that used a lay-administered diagnostic interview to assess a wide range of DSM-IV disorders. Blinded clinical follow-up interviews of adult ADHD were carried out with 154 NCS-R respondents, over-sampling those with a positive screen. This study found an estimated prevalence of adult ADHD to be 4.4%. Lifetime prevalence has been estimated by Kessler and colleagues to be 8.1% (Kessler, Berglund, Demler, Jin, & Walters, 2005). This figure for current prevalence is highly similar to that found in the more regionally limited study by Murphy and Barkley (1996b) (4.7%) and falls well within the midrange of the prevalence estimate inferred above from childhood longitudinal studies (3.3-5.3%). The Kessler et al. study also found that ADHD was significantly correlated with being male, being previously married, being unemployed, and being of Non-Hispanic White ancestry. ADHD was also noted to be highly comorbid with many other DSM-IV disorders and was associated with substantial role impairment. The majority of cases determined to be ADHD in this study had been untreated specifically for their ADHD, although many had obtained treatment for other comorbid mental and substance disorders. We will return to many of these issues of comorbidity, impairments, and treatment history later in this course as our own results relate to them. For now, it is worth noting that Kessler et al (2006) concluded that efforts were needed to increase the detection and treatment of ADHD in adults and that more research was required to determine whether effective treatment would reduce the onset, persistence, and severity of disorders that co-occur with adult ADHD – a call for more research with which we heartily agree.
To summarize, it appears from both childhood follow-up studies and, more directly, from studies of adult general population samples that the prevalence of ADHD in adults in the United States is approximately 5%. The US Census Bureau (www.census.gov) estimates that in 2005, there were 295,507,134 people in the country of which 221,868,077 were 18 years of age or older. Based on this figure and the 5% estimated prevalence for ADHD in adults, it seems likely that at least 11,093,403 adults in the US probably have ADHD. This is a sizeable number, making it imperative that the mental health, medical, and educational professions as well as employers become more aware of the existence of this disorder and its treatments. It also makes it essential that we understand as much about the expression of this disorder in adulthood – along with its comorbid psychiatric disorders and impairments in major life activities – if we are to be able to better understand the disorder and to be able to manage it and its consequences more effectively.
This section provides a quick review of the current diagnostic criteria for ADHD as set forth in the DSM-IV (American Psychiatric Association, 2000). These criteria are shown in Table 1. Readers should understand that despite the criticisms and limitations I will level at these criteria below, especially as they pertain to the diagnosis of adults, they are the most empirically-based, rigorously tested, and logically coherent criteria of their time for the diagnosis of ADHD, especially in children. For more on the development of these criteria, see the paper by Lahey and colleagues (Lahey, Applegate, McBurnett, Biederman, Greenhill, Hynd, Barkley, et al., 1994) and others (Applegate, Lahey, Hart, Biederman, Hynd, Barkley et al., 1997), and the earlier field trial for DSM-III-R (Spitzer, Davies, & Barkley, 1990).
|
Table 1. DSM-IV Criteria
for ADHD
|
Code based on type:
Coding note: For individuals (especially adolescents and adults) who currently have symptoms that no longer meet full criteria, “In Partial Remission” should be specified. |
|
From American Psychiatric Association (2001). Copyright 2001 by American Psychiatric Association. Reprinted by permission. |
The discussion that follows is largely taken from Barkley’s earlier paper on this same subject with James McGough (see McGough & Barkley, 2004). In contrast to the Wender Utah criteria for diagnosing adults with ADHD discussed above (see also Wender, 1995) that gained some popularity in the early 1990s, Biederman and colleagues showed the utility of adhering to the DSM-based approach for the diagnosis of adults and demonstrated that those so identified had patterns of impairment, comorbidity, and neuropsychological functioning similar to children with ADHD (Biederman, Faraone, Knee, & Munir, 1990; Biederman, Faraone, Spencer, Wilens, Norman, Lapey, Mick, Lehman, & Doyle, 1993). In their research, structured diagnostic interviews for adult psychopathology are supplemented with modules from child structured diagnostic interviews, specifically the child module for ADHD. Patients report retrospectively on the occurrence and associated impairment of each of the ADHD symptoms from the DSM occurring during their childhood. Where positive responses occur, further inquiries are made as to current DSM symptoms and related impairment. As with Wender’s criteria, the diagnosis of adult ADHD based on this approach is contingent upon the retrospective recall of the patient, which is then used to infer a diagnosis of ADHD in childhood. In our clinical and research practices, we have followed a similar approach to assess childhood and current functioning (Barkley, Murphy, & Bush, 2001; Murphy & Barkley, 1996a; Murphy, Barkley, & Bush, 2001). However, we developed our own structured interviews and rating scales for doing so (Barkley & Murphy, 1998; 2006), considering that none existed at the time specifically for the diagnosis of ADHD in adults and that the child diagnostic structured interviews were not developed or intended for use with adults. Despite this successful adaptation of the DSM for identification of adult ADHD patients, significant limitations to this approach deserve notice. The DSM-IV criteria are noted below in bold type along with specific difficulties that arise from each criterion as they may be applied to the diagnosis of adults.
A. Either (1) or (2): (1) six (or more) of the following symptoms of inattention have persisted for at least 6 months to a degree that is maladaptive and inconsistent with developmental level; (2) six (or more) of the following symptoms of hyperactivity-impulsivity have persisted for at least 6 months do a degree that is maladaptive and inconsistent with developmental level. ADHD symptoms listed in DSM-IV are based on symptoms from prior DSMs along with items chosen from empirically-derived behavior rating scales that load on these same factors or dimension, expert clinical opinion, and a field trial testing the psychometric properties and utility of the item pool (Lahey et al., 1994). Symptoms eventually selected for inclusion significantly correlated with parent and teacher ratings of impairment, and differentiated clinically diagnosed ADHD from non-ADHD disorders in a sample of 380 clinically referred children ranging from four to 17 years of age. Field trial results further suggested that a threshold of six of nine hyperactive-impulsive symptoms or six of nine inattentive symptoms optimally predicted significant impairment and a clinically validated diagnosis of ADHD. It also produced the best inter-judge reliability.
Despite the commendable empirical basis to the development of DSM-IV, its shortcomings for use in diagnosing adults with ADHD are readily apparent. Symptoms were identified for application to children and by a workgroup concerned only with childhood disruptive disorders. Unlike Wender’s Utah criteria, there was no attempt to test symptoms more developmentally representative of the adult stage of the disorder. Moreover, no adults were included in the DSM field trial. Several DSM-IV symptoms, such as “runs and climbs excessively” or “has difficulty playing…quietly” are clearly inappropriate and without face-validity for use with adults. There is little evidence to suggest that current DSM symptoms designed for use with children best characterize adults with ADHD. This was one of the major purposes of the present UMass Study and a secondary aim of the Milwaukee Study – to evaluate current DSM symptoms when used with adults and to examine additional items that potentially may have greater applicability to and validity for detecting the adult stage of the disorder.
There is also a concern about whether the DSM symptom list represents the current conceptualization(s) of ADHD as accurately as it could or should. Since the late 1980s and early 1990s, ADHD has been conceptualized as involving a disorder of behavioral inhibition (Barkley, 1997; Nigg, 2001; Quay, 1988), and earlier conceptualizations certainly made it part of the trinity of symptom complexes (Douglas, 1972) even as far back as DSM-II (American Psychiatric Association, 1968). Yet if one examines the symptom lists in DSM-IV, the greatest weight is given to inattention (9 symptoms), followed by hyperactivity (6 symptoms), with the remaining three symptoms thought to reflect impulsiveness. Most of those impulsive items reflect principally verbal behavior. The words “impulsive” or “poorly inhibited” are not even mentioned in the symptom list despite being viewed currently as a core feature, if not THE core feature of this disorder. This has proved to be a glaring oversight because the symptom of “makes decisions impulsively” and others related to it (acts before thinking, has difficulty waiting for things, etc.) are among the most discerning symptoms for distinguishing ADHD from other psychiatric disorders as well as the general non-disordered population (Barkley, Murphy, & Fischer, 2008). DSM-V will need to give greater balance to the triad of symptoms if it is to reasonably represent current conceptualizations of what is disordered in ADHD.
Similarly, there is no scientific basis for establishing nine symptoms as the best list length or six symptoms as the most appropriate threshold for adult diagnosis – it is a threshold based on and meant for children. I studied the prevalence of DSM-IV symptoms in a sample of adults in Massachusetts (Murphy & Barkley, 1996a) and found that a threshold of six symptoms required for diagnosis of ADHD was 2 to 4 standard deviations above the normal adult mean for that sample, in essence representing the 98th to 99.9th percentile of the adult population. The use of this threshold could therefore literally define adult ADHD nearly out of existence or, at best, consign it to a statistical rarity. Likewise, this threshold was 3.5 standard deviations above the mean for the control group followed from childhood to adulthood in the Milwaukee Study of ADHD children at their age 21 year follow-up (Barkley, Fischer, Smallish, & Fletcher, 2002). That study concluded that approximately one-third of individuals with severe symptoms (98th percentile) failed to meet the DSM criteria for clinical diagnosis. Our earlier study of a general population sample in Massachusetts (see above) reported that cutoff scores of four for current symptoms of hyperactivity or inattention would be more commensurate with the 93rd percentile often recommended as being clinically significant in childhood diagnosis of ADHD. Others have recommended this same threshold to identify college students with sufficient difficulties to warrant treatment (Heiligenstein, Conyers, Berns, Miller, & Smith, 1998; Kooij, Buitellar, van den Oord, Furer, Rijnders, & Hodiamont, 2004). This suggests that DSM-IV criteria are overly restrictive when extrapolated to adults and fail to identify a substantial minority of adults who suffer clinically meaningful levels of ADHD symptoms.
Clinicians should also take note of the fact that the majority of children who participated in the field trial were male (Lahey et al., 1994) by an average of at least 2:1. Many studies have found that males show higher frequencies of these symptoms than do females in general population samples (e.g., DuPaul, Power, Anastopoulos, & Reid, 1998). If symptom thresholds are based predominantly on males as they were in the DSM field trial, then it is likely that those thresholds are more severe for females relative to the general female population than it would prove to be for males relative to the male population. This would make it more difficult for a female to obtain the diagnosis than a male despite equal levels of symptom deviance (developmental inappropriateness) relative to others of the same sex. Given that no adults were studied in the field trial, it is open to question as to whether the symptom thresholds are equally valid (sensitive) at detecting females as males with ADHD in adulthood. The general population study by Murphy and Barkley (1996a) suggested that symptoms in adults were as frequent in females as in males but that this was not true for retrospective recall of childhood symptoms in which the more typical pattern of greater male frequency of symptoms was evident.
B. Some hyperactive-impulsive or inattentive symptoms that caused impairment were present before age 7 years. DSM-IV requires the onset of symptoms producing impairment prior to age seven for diagnosis of ADHD, essentially at 6 years of age or earlier. ADHD has always been characterized as a disorder of early childhood, with the implication that ADHD-like symptoms arising later in life represent some other condition. But the use of a precise age of onset criterion was not introduced until DSM-III in 1980. Even then, it was not based on any sound scientific rationale or empirical basis. In its defense, it was commonplace at that time to include such a criterion in research studies of ADHD in children but that was done mainly to insure the presence of a developmental disorder (early onset of disorder) and to permit replication of the study by other researchers (Barkley & Biederman, 1997).
The age of onset criterion for symptoms and associated impairment poses particular difficulties for the diagnosis of adult patients. Unlike the assessment of children, the clinical evaluation of adults is highly dependent on patient self-report. Adults likely have a limited recall of the exact time course and nature of symptoms (Barkley et al., 2008). Adults are also likely to have a limited recall of the domains of childhood impairments related to those symptoms within the developmental time frame associated with so precise a childhood onset of age 7 as is required for diagnosis in DSM-IV. Moreover, many adults who present for clinical care are unable to provide independent evidence of the disorder, either through retrospective parental report or records of academic functioning. Adults do not typically come to clinical evaluations with their parents in order to provide the customary evidence for judging symptom onset as is done in children. While it is implicit in the use of the DSM-IV criteria with children that the judgments of someone besides the patient (child) will provide this evidence, it is not made explicit. And so it is typically not required when adults are being evaluated even though that requirement is assumed to exist in child cases. A further problem here is the much greater time span over which these adults (or even their parents, if interviewed) must retrospectively recall their childhood behavior relative to the time span upon which parents of children with ADHD must reflect. Add to this the likelihood that ADHD may create a positive illusory bias in adults concerning their possible impairment – as it does in children with ADHD (Knouse, Bagwell, Barkley, & Murphy, 2005) – that could possibly diminish self-awareness of symptoms and impairments. This gives a further reason to question the reliance upon adult patients for establishing the age of onset of their symptoms and associated impairments.
Given the lack of empirical evidence supporting the age of onset criterion, as well as practical difficulties in demonstrating impairment prior to age 7 in older adolescents and adults, some have argued that the criterion should be abandoned or redefined to include the broader period of adolescence (ages 12-14) (Barkley & Biederman, 1997). The suggestion is not as radical as it may first appear. Consider that most mental disorders do not have a criterion requiring such an explicit age of onset for symptoms or impairment, if they have any at all. This state of affairs applies as well to other developmental disorders known to commonly arise in childhood, such as the specific developmental disorders (no age of onset), mental retardation (onset before 18 years), tic disorders and Tourette's syndrome (onset before 18 years), and Asperger's disorder (none). Even though such disorders are considered to be as much or even more "developmental" in nature as ADHD, and that most cases have their onset in childhood, an age of onset is either not required for them or is quite broadly construed (e.g., before age 18 years). A case can now easily be made that ADHD is just as deserving of such liberal treatment concerning this diagnostic criterion, as are the other developmental disorders. A further point in favor of broadening or abandoning an age of onset requirement for ADHD rests in the fact that there never existed a compelling rationale or empirical foundation for inserting this diagnostic criterion into the DSM. This criterion did not exist in DSM-II for the predecessor of this disorder (hyperkinetic reaction) and its insertion into DSM-III was based solely on committee consensus alone without benefit of an empirical field trial. Its retention across DSM-III-R and IV appears to have been based more on a sense of tradition than on any empirical foundation for its diagnostic validity or utility.
But the greatest evidence against using so precise a criterion for age of onset as age 7 years is that the DSM-IV field trial not only failed to find evidence in support of this specific age as being diagnostically useful, it even found substantial evidence arguing against its retention (Applegate et al., 1997). That study found that using the onset of age 7 years for this purpose reduced the classification accuracy of the remaining diagnostic criteria when compared to using older ages of onset (8, 9, or higher). It also reduced the inter-judge reliability for the diagnosis significantly, and failed to show any association with the types of impairments examined in that study. Unfortunately, DSM-IV had been published before these aspects of the field trial analyses were completed and published so the age of onset criterion was retained solely by default.
Lastly, no evidence is available in the literature to our knowledge that suggests that onset of ADHD symptoms at or after age 7 years results in a qualitatively or even quantitatively different disorder than cases of ADHD having the earlier recommended symptom onset. The fact that some prior studies have reported a mean age of onset of initial ADHD symptoms as occurring between ages 3-4 years does not automatically argue for inclusion of a precise age of onset criterion into a diagnostic system for ADHD; it only suggests that ADHD, like retardation, appears to be "developmental," arising early in many cases. Although the range of symptom onsets around this mean is substantial, the reliability of parental identification of this precise onset is questionable, and the diagnostic or conceptual significance of a precise age of onset remains unexamined and unjustified (Barkley & Biederman, 1997). Furthermore, the early age of onset found in most studies of children with ADHD may be due, in part, to method artifact – it is by virtue of studying clinic-referred children who, almost by definition if not by default, have developed their symptoms in childhood. Moreover, because they are children who do not self-refer to clinics on first appearance of their problem behavior, their symptom onset is often well before the decision to refer for mental health services is finally reached by the family, teachers, or primary care professionals. The DSM-IV field trial also demonstrated that a significant percentage of children meeting all other criteria for the disorder failed to demonstrate symptom onset prior to age seven, particularly those with the Inattentive Type (Applegate et al., 1997). Yet, in favor of broadening the criteria, that same field trial found that all cases of ADHD used in that study had developed their disorder by the more generous adolescent age of onset of 14 years (Applegate et al., 1997). To continue to argue that the 7-year age of onset criterion must be applied to adults for a diagnosis of ADHD to be valid just because the DSM-IV so stipulates is an empirically indefensible justification smacking more of ritual or dogma than of supporting data.
C. Some impairment from the symptoms is present in two or more settings (e.g., at school [or work] and at home. Problematic here, obviously, is that adults are involved in far more numerous and important adaptive settings or domains of major life activities than this criterion stipulates. Not only are those settings specified here too global to be of much good to the clinician evaluating domains of impairment (e.g., “home”) but they ignore many more domains of major life activities that are not only more specific but also comprise important domains of adult adaptive functioning. General functioning within the larger organized community (e.g., participation in government or formally organized community groups, cooperation with others living in the same neighborhood, abiding by laws, driving), financial management (e.g., banking, credit, contracts, debt repayment), parenting and child-rearing (e.g., protection, sustenance, financial and social support, appropriate education, discipline), marital functioning, and routine health maintenance activities are additional domains of major life activities in which symptoms may produce impairment that would not be evident in children. Current criteria fail to reflect these potential areas of impairment. This, too, represents another major aim of the present book, which is to examine the major domains of important life activities in adulthood and the manner in which ADHD in adults affects them.
D. There must be clear evidence of clinically significant impairment in social, academic, or occupational functioning. ADHD is conceptualized as a pervasive disorder of functioning, and not as a reaction limited to a specific stressor or to a specific circumscribed setting. In childhood, it is expected that ADHD leads to some measurable impairment in multiple settings, most notably at home, at school, and with peers. DSM-IV briefly specifies substitution of “work” for “school,” but fails to reference the full range of adult impairments as noted above. Both children with ADHD followed to adulthood as well as clinic-referred adults with ADHD have demonstrated decreased educational achievement, poorer occupational functioning, a greater propensity for antisocial activities and drug use/abuse, higher divorce rates, poorer personal health choices, earlier parenthood, and increased driving risks compared with non-ADHD adults (Barkley, 2006; Barkley, Fischer et al., 2004, 2006; Biederman, Faraone, Spencer, Wilens, Norman, et al., 1993; Fischer & Barkley, 2006; Murphy & Barkley, 1996b). Although we review each of these domains in detail along with our results concerning them later in this course, the point here is that DSM-IV provides an insufficient array of major life activities for judging adult impairments. It also provides no guidance for differentiating among various domains of functioning necessary to meet diagnostic criteria. Based on our results, we will make further recommendations as to the other domains that should be noted in DSM beyond just home, school, and work settings.
Conversely, many adults have adopted lifestyles that minimize self-reported dysfunction across multiple domains. For example, an adult with significant occupational impairment might live alone; no longer attend school; stay with less responsible part-time, unskilled, or entry level work; involve himself recreationally with the more liberal social mores of the performing arts such as rock music or acting; be content with few or no friends, or those with equally elevated antisocial or drug abuse tendencies; or might have minimal insight into the full range of his dysfunction. Clinicians who strictly adhere to the requirement for multiple domains of impairment might fail to treat patients who would clearly benefit from treatment but who demonstrate impairment in only a single domain. These clinicians may fail to appreciate this form of adult niche-picking that renders the patient’s symptoms less impairing by virtue of the individual associating with these less culturally mainstream and more avant-garde, asocial, or frankly antisocial arenas of life.
Recent controversy exists over the definition of impairment due to an enormous increase in requests for special accommodations in employment and high-stakes academic testing under the Americans with Disabilities Act. Further specification of the meaning of impairment is necessary in DSM-V so as to avoid misunderstandings among clinicians and public agencies. Some clinicians assess impairment based on comparison of deficits relative to a person’s intellectual level, much like had been done in the earlier history of defining learning disabilities as being significant discrepancies between IQ and some specific area of academic achievement, such as reading. Others believe impairment is based on how well an individual functions relative to their specialized peer group, particularly if they are unusually intelligent or well-educated, such as fellow gifted individuals or peers in medical or law school. Still others have argued that impairment should mean serious dysfunction in the performance of major life activities (e.g., family, marital, social, or occupational functioning) that are required of society in general. More to the point, this view holds that impairment should be defined as being relative to the norm or average person, as required by the Americans with Disabilities Act (ADA), and not relative to some narrow, highly specialized and accomplished subset of adults or to an estimate of one’s general cognitive ability, such as IQ (see Gordon & Keiser, 1998; Murphy & Gordon, 2006). We prefer the latter view of defining impairment because of a number of factors: its consistency with scientific views of valid mental disorders (harmful dysfunctions that are failures of severe deficiencies in mental adaptations (Wakefield, 1992)); its consistency with the ADA, with associated court rulings, and with the legislative intent behind the ADA (granting protections and accommodations to subnormally functioning individuals); and simple fairness or justice – individuals should not be viewed as disordered and granted special protections, accommodations, disability financial benefits, or other societal privileges when they are not below the average of the population at large. It is inherently unfair to grant advantages to those who are not actually subnormal. Also, this latter view of impairment respects the fact that one’s intelligence is not an indicator of functioning in all avenues of adult life nor are disparities between IQ and some other measure of adaptive functioning. Future DSM committees should make the criteria for impairment clearer as to the domains it encompasses and the comparison group to be used for its determination.
E. The symptoms are not better accounted for by another mental disorder. Current DSM symptoms were selected in part for their ability to differentiate ADHD from other psychiatric disorders (Lahey et al., 1994). However, without adult studies there is no evidence to suggest that childhood ADHD symptoms similarly differentiate ADHD adults from other adult psychiatric conditions. Unlike the Utah criteria that exclude significant comorbidities from the adult ADHD diagnosis, DSM-based studies report high rates of co-occurring psychopathology with adult ADHD – findings corroborated in both our projects later in this text. ADHD in adults is described frequently as having coexisting oppositional defiant disorder, conduct disorder, antisocial personality disorder, and psychoactive substance use disorders as we noted above. More controversial are increased rates of anxiety, depression, and possibly bipolar disorder (Faraone & Biederman, 1997; Faraone, Biederman, Mennin, Wozniak, & Spencer, 1997; Shekim, Asarnow, Hess, Zaucha, & Wheeler, 1990). Several ADHD symptoms (i.e. concentration difficulties, restlessness, increased speech, acting “on the go,”) are also symptoms of other disorders, particularly anxiety, depression, and mania. It is unclear if these DSM symptoms adequately differentiate ADHD from other adult disorders, or how other disorders might manifest when co-occurring with ADHD.
DSM-IV Subtypes. ADHD subtypes (i.e. Inattentive, Hyperactive-Impulsive, and Combined) are based on the DSM field trial that suggested ADHD symptoms clustered around two partially distinct factors of inattentive and hyperactive-impulsive symptoms (Lahey et al., 1994). Subtyping based on any approaches using two lists would create three groups of entities by mere default and not based on clinical or conceptual significance. There is ongoing controversy as to whether ADHD subtypes so formed represent manifestations of the same or different conditions. Some researchers have found evidence that DSM subtypes breed true (Hudziak, Heath, Madden, Reich, Bucholz, Slutske et al., 1998; Levy, Hay, McStephen, Wood, & Waldman, 1997), while other investigators report a lack of relationship in subtypes among family members (Faraone, Biederman, & Friedman, 2000; Smalley, McGough, Del’Homme, NewDelman, Gordon, Kim, Liu, & McCracken, 2000; Willcutt, Pennington, & DeFries, 2000). The question of subtypes is particularly vexing with adults. Evidence suggests that hyperactive-impulsive symptoms decrease significantly over development, earlier in development, and significantly more over development than do inattentive symptoms (Biederman, Mick, & Faraone, 2000; Hart, Lahey, Loeber, Applegate, & Frick, 1995; Mick, Faraone, & Biederman, 2004), at least as they are currently described in the DSM symptom list. Yet this could easily be the result of using symptom descriptions most applicable to early childhood that fail to capture the expression of the same construct at later developmental stages. Results from the field trial reveal that most children in the Hyperactive-Impulsive type develop the Combined type during early school years, while those in the Inattentive type tend to remain there (Lahey et al., 1994). In many cases patients who might have been diagnosed with the Combined type as youths, will appear as Inattentive types in adulthood simply by virtue of the aforementioned expected decline in their hyperactivity with age. That decline eventually may reach a point where fewer than six symptoms of hyperactivity-impulsivity exist necessitating their reclassification from Combined to Inattentive type. DSM provides no guidance as to whether an adult subtype should be assigned based on one’s child, adolescent, or adult symptom presentation. Once a child has been diagnosed with Combined Type, should they always be considered Combined Type even if they no longer manifest a sufficient number of Hyperactive/Impulsive (HI) symptoms as adolescents or adults? Should the predominantly Inattentive Type be reserved only for “true” or “pure” Inattentive Types who have never had any developmentally inappropriate symptoms of hyperactivity or impulsivity? Or is it appropriate to assign the Inattentive Type label to those who met criteria for Combined Type as children and no longer have a sufficient number of HI symptoms as adults? These issues have relevance to investigations of potential biological differences between types and to practitioners who clinically diagnose patients. The current DSM subtype classification has, in fact, never been validated in adults with ADHD and has insufficient empirical evidence at present to justify its use after childhood, if even then.
DSM Residual Categories. In earlier versions of DSM, the category ADHD, residual type was reserved for adults who met criteria for ADHD in childhood and continued to have significant symptoms and impairment that fell below threshold for full diagnosis. This category has been replaced in DSM-IV with ADHD, in partial remission. The DSM-IV category of ADHD Not Otherwise Specified is employed when patients have impairment from symptoms of inattention or hyperactivity/impulsivity, but fail to meet all criteria for ADHD. These diagnoses are default categories that are useful when clinicians are required to provide a diagnosis. Neither of these categories has defined or validated criteria, and reliability is insufficient to support research efforts. Adult patients would be better served with developmentally appropriate and well-validated diagnostic criteria. Hence, the justification for examining potentially new items for diagnosing ADHD in adults in the present book along with evaluation of the symptom thresholds and age of onset criteria.
As noted earlier, no precise age of onset of symptoms producing impairment should be required for a diagnosis of ADHD. An upper boundary of age 16 has been recommended by a recent book of mine involving extensive research on the issue (Barkley et al., 2008). Typically, we require corroboration of ADHD symptoms and impairment from someone else who knows the person well, such as parents, siblings, or spouses/partners, as part of a clinical diagnosis of ADHD.
The clinician conducting the interview should exercise her judgment as to whether the patient’s reports on these matters could be considered to be realistic or have some veracity. As a consequence, a few individuals may be clinically diagnosed as having ADHD by the clinician despite their not meeting diagnostic criteria strictly by their own initial self-report. Others who meet criteria based solely on their self-report may not be granted the clinical diagnosis of ADHD. The latter may have other disorders that interfere with attention, such as anxiety disorders or depression, but do not have the cardinal features of ADHD (chronic ADHD symptoms from childhood). A moment’s reflection will show several reasons for why this necessarily has to be the case.
First, the self-reported ADHD-like symptoms may be clinically judged to be better accounted for by the presence of another diagnosis (such as dysthymia, depression, anxiety, substance abuse, marital problem, a situational stressor, etc.). This is a requirement of the DSM criteria for ADHD that may often go overlooked in research studies that select their ADHD group merely by rating scales or solely on self-reported information. This criterion can only be executed via clinical judgment and knowledge of differential diagnosis and cannot be incorporated into some mindless algorithm that relies exclusively on self-report.
Second, the symptoms patients endorse and/or the associated impairments they allege may not rise to the level of being clinically significant in the clinician’s judgment. DSM criteria require that symptoms be developmentally inappropriate and lead to impairment, both of which inherently involve clinical judgment. For example, a patient may endorse 14 of the 18 symptoms but the examples of the symptoms given were judged to be clinically trivial and the impact they had in producing clinically significant impairment is either minor or nonexistent. Likewise, another patient may give evidence of having no real impairment other than an internal perception that he was somehow not working up to his potential or not being as successful or effective as he thought he should be. In other words, there is no other historical corroborative evidence in these reports that the behavior of which they complain is actually a symptom (abnormal) or that the impairment claimed is so interfering with their functioning that it resulted in being well below the average-person standard discussed above. For example, in the second patient’s case, despite his reported symptoms, he had suffered no problems in school, had received no prior psychological treatment, had received no accommodations for a disorder at school or at work, was happily married, demonstrated no occupational impairment, or failed to manifest convincing social or daily adaptive impairment that in the clinician’s judgment was significant and a consequence of ADHD. In some cases, these are what one might call “ADHD wannabes” who are self-diagnosed before coming into the clinic, typically based on reading a popular trade book on ADHD in adults or hearing media accounts of the disorder and believing themselves to have it. In short, to be eligible for the diagnosis of ADHD, patients have to have a sufficient number of DSM-IV symptoms that in the clinician’s judgment produced clinically significant real-world functional impairment in major life activities.
Third, for the clinician to render an ADHD diagnosis, he needs to see fairly compelling evidence of an onset of symptoms sometime during childhood or adolescence, a chronic (unremitting) and pervasive pattern of ADHD symptoms, and impairment that could be reasonably attributed to ADHD. The clinician should not simply record mere self-reported symptom counts or statements of impairment relying solely on a judgment-free algorithm. It will be clear that some patients do not have a good perspective on what constitutes impairment. For example, a patient may have denied having any significant impairment yet a closer look at her history and school records may show substantial struggles in school achievement and deportment, in conduct in the community (delinquency), in her job performance or social relationships, or in just managing daily major responsibilities. However, she might have simply chalked it up to "I just hated school" or her job, or her friend or partner, etc., rather than viewing it as stemming from any sort of disorder.
Finally, there has to be convincing evidence that the symptoms actually developed and produced impairment sometime during childhood or adolescence. Consistent with criticisms raised above about the DSM criteria for ADHD, when many patients are asked the question about onset, they have a hard time specifying an exact age. They use phrases such as "as long as I can remember,” "always," "forever," etc. Others give evidence of a very poor memory of their childhood and cannot remember when they first noticed problems yet they may have given the clinician other information that helped attach an age to the onset of symptoms producing impairment (such as getting suspended or held back in first grade). In addition, some might say their impairment began in high school yet during the interview or from inspecting school records, it becomes clear that the impairment had begun much earlier. Hence, differences could exist in self-reported perceptions of onset versus a clinician’s determination of onset based on the totality of information received during the assessment.
All of these issues should be kept in mind when making the diagnosis of ADHD. The DSM-IV should be viewed as providing rough guidelines for making this diagnosis in adults, and requires sound clinical judgment to implement the guidelines properly in adult cases. Interview and rating scale forms that the clinician may find helpful in conducting this evaluation can be found in my clinical workbook with Kevin Murphy (Barkley & Murphy, 2006).
My colleagues and I recently published a large study examining a large item pool of potentially better symptoms for identifying ADHD in adults (Barkley et al., 2008). Our analyses found that six items reflecting executive function deficits along with three from the DSM-IV list made a better item set than does the DSM-IV as it is currently structured. These new items are shown in Table 2.
|
Table 2. Newly Proposed Criteria
for ADHD in Adults
|
|
|
© Copyright by Russell A. Barkley, Ph.D. From Barkley, R. A., Murphy, K. R., & Fischer, M. (2008) ADHD in Adults: What the Science Says. New York: Guilford |
In my new book on ADHD in adults (Barkley et al., 2008), my colleagues and I reported on the results of two large studies of adults with ADHD. One, the UMass Study, compared a large sample (148) of adults clinically diagnosed with ADHD against a clinical control group of patients with other psychiatric disorders seen in the same clinic (97) and a general population control group (109). We also reported on the adult follow-up of a large sample of ADHD and control children followed to a mean age of 27 in Milwaukee, Wisconsin, often using the same or similar measures as in the UMass Study. We evaluated the functioning of these adults in both projects in numerous major life activities. Our conclusions are reported here.
Across all of our results, one thing seems abundantly clear – ADHD in adults is a significantly impairing disorder. It is associated with numerous difficulties in virtually every domain of major life activity studied here. Whether one studies functioning in education, occupation, social relationships, sexual activities, dating and marriage, parenting and offspring psychological morbidity, crime and drug abuse, health and related lifestyles, financial management, or driving, ADHD can be found to produce diverse and serious impairments. Indeed its impairments are more substantial than are those seen in other disorders most likely to present to outpatient mental health clinics, such as anxiety disorders, dysthymia, and major depression, among others. This is obvious in the numerous differences we found between adults with ADHD and our clinical control group. The disorder also deserves its status as one distinct from other forms of psychopathology or developmental disabilities. Its symptoms and impairments are not simply due to general psychopathology. They stand out from other forms of psychopathology in numerous respects. Statements to the effect that ADHD is not a valid disorder but rather a myth created by pharmaceutical companies or mental health professionals for shear mercenary commercial gain, or that it is indistinct from the other disorders with which it may be associated are not only wrong, they are egregiously so. Numerous differences that emerged in the context of these two studies between those with ADHD and general population controls, and between those with ADHD and clinical control groups, make such assertions moribund. To continue to make such statements in the face of such overwhelming evidence to the contrary is to either show a stunning scientific illiteracy or planned religious or political propaganda intended to deceive the uninformed or unsuspecting general public.
Global Reports of Impairment. As we stated earlier, the symptoms of ADHD are the behavioral expressions associated with this disorder – they are the actions demonstrated by those having the disorder that are believed to reflect that disorder (i.e., inattention, distractibility, impulsive responding, hyperactivity, poor executive functioning, etc.). In contrast, impairments are the consequences that ensue for the individual as a result of these cognitive-behavioral expressions. Symptoms are actions of an individual (cognition/behavior), and impairments are the consequences of those actions (outcomes or social costs). The term impairment refers to deficits that are relative to the functioning of the normal population or “average person,” and not to an intra-personal disparity or comparison to some highly specialized or high functioning peer group.
When examining the various domains of major life activity specifically in our interviews, we found that with the exception of dating or marriage, the ADHD group in the UMass Study showed a significantly greater percentage as being impaired in most domains than was the case for either of the control groups. The domain most affected by adult ADHD was education followed by home responsibilities and occupational functioning and then, to a lesser extent, dating/marriage and social activities. Community activities such as participating in clubs, sports, or organizations were the least likely to suffer impairment due to ADHD. The Milwaukee Study found a somewhat different pattern of impairment, where current ADHD at age 27 was associated with a somewhat lower likelihood of being impaired in any particular domain. Home and occupational domains were the most likely to be impaired as was money management and daily responsibilities. Unlike the UMass Study, the educational domain was not as likely to be self-reported as impaired in those hyperactive children retaining ADHD at follow-up. This is interesting in so far as the actual evidence for educational impairment was found to be far greater in the hyperactive children grown up than in the clinic-referred adults having ADHD.
We also collected information from retrospective reports on the childhood domains most likely to be impaired. For the ADHD group in the UMass Study, education or the school setting was far and away the domain most likely to be adversely impacted by ADHD (over 90%) followed by daily chores and responsibilities (75%). The same was true for the ratings provided by significant others about the childhood impairments in these groups. While a smaller proportion of each group was rated as being impaired in the reports of others compared to self-reports, a higher proportion of the ADHD group was rated as impaired in each of the 8 domains from childhood than was the case for either control group. Again, the educational setting was the domain in which more of the ADHD group had been affected relative to the other domains surveyed. The Milwaukee Study found that many domains were reported to be impaired but the domain of social (peer) interactions was the one most likely to be associated with the ADHD group in childhood.
Our findings further indicated that the reports of clinic-referred adults who are in their 30s or older about themselves or those provided by others who know them well concerning ADHD symptoms are likely to be impressively correlated with reports within each of these sources about degree of impairment (rs = .70-.80). These relationships are strong whether they pertain to current functioning or to recall of childhood functioning. Such severity, especially at clinically elevated levels (4 or more symptoms), is highly likely to be associated with risk of impairment in one or more major life activities (100% are impaired). The results for children followed to adulthood are somewhat lower but still show significant relationships between severity of ADHD and the severity and pervasiveness of impairments.
Comorbidity. There is compelling evidence that ADHD increases the liability for certain other psychiatric disorders. More than 80 percent of our ADHD groups had at least one other disorder, more than 50% had two other disorders, and more than a third had at least three other disorders, these being markedly higher than in our control groups in both studies. As in the prior literature on children and adults with ADHD, we found a markedly elevated risk for oppositional defiant disorder (ODD), and to a lesser extent for (conduct disorder (CD), in our clinic-referred ADHD group and in our hyperactive children as adults. Current ADHD was especially associated with a childhood history of ODD.
The internalizing disorders of major depressive disorder, dysthymia, and anxiety disorders are more likely to occur in ADHD cases referred to clinics over that risk seen in a community control group. But MDD and anxiety disorders are also significantly elevated in non-ADHD clinical control cases seen at the same ADHD clinic and thus may not be as specifically linked to ADHD as to general outpatient psychopathology. Even so, both epidemiological studies in children and adults find some association between ADHD and depression that make it unlikely that our findings of a limited association are purely due to referral bias. Nevertheless, the relationships that do exist are not as strong when comparisons to other clinical samples are used than when comparisons to community samples are studied. The Milwaukee Study did not find an elevated risk for MDD specifically in those with persistent ADHD into adulthood but did find an elevated risk for depressive personality disorder and for mood disorders more generally, both of which suggest some link between ADHD and the level of depressive symptoms, even if not with full syndrome MDD. Neither study found any elevated risk for OCD, bipolar disorder, or schizophrenic spectrum disorders.
Both the ADHD groups in our studies showed a greater risk for alcohol use disorders while the clinic-referred adults (but not the hyperactive children grown up) also showed a greater risk for cannabis use disorders compared to community controls. Alcohol use disorders and risk for any drug use disorder may be specifically linked to ADHD though the level and type of drug use disorders probably have more to do with comorbid CD and antisocial personality disorder as well as local access to specific drugs than to ADHD, per se.
We also studied comorbidity dimensionally, using the SCL-90-R and the Young Adult Child Behavior Checklist. Adults with ADHD (whether clinic-referred or children grown up) showed elevations on all scales of the SCL-90-R psychological maladjustment relative to community controls and on most of the scales relative to the clinical control group. Our findings are consistent with all but one prior study in the literature on adults with ADHD using this instrument. There is clearly greater maladjustment of all types associated with ADHD than in clinical or community comparison groups. Such findings imply that ADHD is a more severe psychological disorder than many outpatient disorders seen in the same clinics.
Concerning the risk of suicidal ideation and attempts, we found that the ADHD group in the UMass Study had only a slight but not significant increase in risk over the two control groups in both ideation (25% vs. 15-16%) and attempts (6% vs. 2-4%) prior to 18 years of age. After age 18, however, both the ADHD and clinical control groups reported elevated rates of suicidal thinking (27-29%) over that seen in the community control group (6%). The ADHD group specifically also reported a greater risk of suicide attempts relative to the community group (8% vs. 1%). The Milwaukee study also found an elevated risk of suicidal thinking and attempts in the hyperactive groups, particularly before 18 years of age, and an ongoing risk of greater ideation (but not attempts) going forward to follow-ups at ages 21 and 27. The two hyperactive subgroups did not differ in these risks, indicating that persistent ADHD into adulthood was not the major determinant of such risks. The greater risks of ideation and thinking reported here were largely mediated by the presence of MDD and, to a lesser extent, dysthymia but were not especially related to the presence of comorbid CD.
Education. Clinically diagnosed adults with ADHD share some of the same types of academic difficulties in their histories, as do children who were hyperactive and followed over development. However, their intellectual levels are higher, their high school graduation rates are higher, more are likely to have attended college, and their likelihood of having achievement difficulties or learning disabilities is considerably less in most respects than that seen in children with ADHD followed to adulthood. This higher level of intellectual and academic functioning in clinic-referred adults with ADHD makes sense given that they are self-referred to clinics in comparison to children with ADHD. This fact makes it much more likely that these adults have employment, health insurance, and a sufficient educational level to be so employed and insured. They could also be expected to have a sufficient level of intellect and self-awareness to perceive themselves as being in need of assistance for their psychiatric problems and difficulties in adaptive functioning. Children with ADHD brought to clinics by their parents are less likely to have these attributes by the time they reach adulthood. They are not as educated, are having considerable problems sustaining employment, are more likely to have had a history of aggression and antisocial activities, and are not as self-aware of their symptoms as adults having ADHD who self-refer to clinics.
The educational careers of the ADHD groups were checkered with adversities. More of the adults with ADHD reported having been retained in grade, received special education, and been diagnosed with learning disabilities or behavior disorders while in compulsory schooling than adults in either of the two control groups. These risks were even higher in the children with ADHD followed to adulthood. High class rankings and grade point averages were significantly lower in the ADHD groups than in our control groups. Among those participants who had attended college, more of the ADHD group had unsatisfactory grades and had withdrawn from more classes as reflected on their college transcripts than did the two control groups. On tests of educational achievement given in our projects, the ADHD groups were poorer in their arithmetic, spelling, and reading and listening comprehension skills than were adults in the control groups. We also found adults with ADHD to have a higher comorbidity with specific learning disabilities replicating the substantial literature on children with ADHD. This risk was even higher in the children with ADHD followed to adulthood than in the clinic-referred adults with ADHD. All of this leads us to conclude that of all domains of major life activity adversely affected by ADHD in adults, the domain of education is the most pervasively affected and affects more such adults.
Occupational Functioning. Both clinic-referred adults with ADHD and children growing up with the disorder experienced significant problems in their occupational histories. In the UMass Study, adults with ADHD were rated by a clinician as functioning at a lower level overall than adults in the other groups function. They were also found to have experienced a number of problems in a higher percentage of their previous jobs than adults in the two control groups. These problems were related to getting along with others, demonstrating behavior problems, being fired, quitting out of boredom, and being disciplined by supervisors, all of which were more frequent in the work histories of the adults with ADHD than in either of the control groups. The Milwaukee follow-up study found much the same results, except that growing up as a child with ADHD was associated with lower job status and fewer current working hours per week regardless of its persistence into adulthood. Even so, the persistently ADHD group experienced even more difficulties in their current workplace functioning than did either the H-ADHD or the control group. This was also true in comparison to the clinic-referred adults with ADHD whereas children with ADHD that persists to adulthood have a far greater percentage of jobs in which they are fired or experience disciplinary actions than do clinic-referred adults with the disorder. We corroborated these problems through employer ratings in both studies (UMass Study currently, Milwaukee Study at age 21). The ADHD groups were rated as having significantly more symptoms of inattention in the workplace, and as being more impaired in performing assigned work, pursuing educational activities, being punctual, using good time management, and managing daily responsibilities. Both projects provide direct evidence that ADHD has an adverse impact on workplace functioning not only via self-reports but also corroborated through employer-blinded ratings.
Drug Use. Prior research shows that children with ADHD followed to adulthood carry an elevated risk for later substance use and abuse as well as for many forms of antisocial activities and their legal consequences (arrests, jail). In both instances, it is the presence of CD in childhood or adolescence that greatly elevates these risks and, in some cases, accounts for them entirely. However, ADHD does convey some elevated risk for nonviolent activities, such as drug use, possession, or sale, and may convey an elevated risk for tobacco and alcohol use even in the absence of CD. What little research exists on clinic-referred adults with ADHD likewise suggests a greater likelihood of drug use disorders and antisocial personality disorder. However, prior studies have not examined rates of drug use or specific forms of antisocial activities in as much detail as has the literature on children with ADHD followed to adulthood. We further explored these risks in both of our studies.
The UMass Study found that adults with ADHD were likely to have been past or current smokers; to be users of marijuana, cocaine, LSD, or prescription drugs; and to have been treated for previous alcohol and drug use disorders than was the case in the community control group. However, the clinical control group also showed some elevated risks for some drug use problems, primarily past tobacco use and current marijuana use. The ADHD group differed from the clinical group chiefly in having more members who had tried cocaine and LSD. Like past studies of children with ADHD grown up, we found that the presence of CD appears to account for the significantly higher frequency of some drugs used by the ADHD group relative to the control groups. We believe that the presence of childhood CD may not account for whether an adult with ADHD ever tries a particular substance at least once, but it does seem to contribute to the frequency with which they may continue to use that drug.
The findings from the Milwaukee longitudinal study largely corroborated the UMass study in finding a greater risk for being a smoker, using alcohol, getting drunk, or using illegal prescription drugs among both hyperactive (childhood ADHD) groups at age 27. It also found a greater frequency of caffeine use for those groups than for the control group. However, it is largely being referred and diagnosed as ADHD in childhood that is related to risk for later substance use and abuse than whether ADHD is persistent to age 27. While the ADHD groups in both projects appear to be more likely to abuse both legal and illegal substances, the children with ADHD growing up may carry a greater risk for using alcohol and tobacco while clinic-referred adults with ADHD seem more likely to use marijuana, cocaine, and LSD.
Worth reiterating are the findings from the Milwaukee Study that found no evidence that treatment with stimulants in childhood was associated with increased drug use or abuse in any category of illegal drugs. In fact, some evidence showed that being treated with stimulants as a child reduced the likelihood of using certain drug types, such as the use of speed (amphetamines) or the illegal use of prescription drugs. These findings are consistent with the vast majority of other research on this issue, further solidifying the conclusion that childhood stimulant treatment is NOT associated with risk for later drug use or abuse, no matter how critics, fringe religious zealots, and the popular media wish to portray this issue.
Antisocial Behavior and Its Consequences. Children with ADHD are routinely found in follow-up studies to be more likely to engage in antisocial activities, to be arrested, and to be jailed as they grow up than are non-ADHD children. We found the same in both projects. More clinic-referred adults with ADHD had engaged in shoplifting, stealing without confronting a victim, breaking and entering, assaults with fists, carrying an illegal weapon, being arrested, and being sent to jail. In addition, more adults with ADHD had sold drugs illegally in comparison to the community control group; the clinical control group did not differ from either of these two groups on this outcome. The most common forms of antisocial activity for the adults with ADHD were shoplifting (53%) followed by assaulting someone with their fists (35%), and selling illegal drugs (21%). While much of this risk for antisocial behavior was mediated by the presence of a childhood history of CD (established retrospectively), even the non-CD subset of those having ADHD still committed more antisocial activity than the control groups. Likewise, the Milwaukee Study continued to find markedly higher proportions of the hyperactive group to have committed various antisocial acts than the control group. In most cases, this elevated risk was not related to whether the ADHD had persisted to age 27. This suggested to us that while ADHD in both projects is clearly associated with a greater risk of antisocial activity than in control groups, when it occurs in children and leads to early clinic referral and diagnosis it is associated with an even greater risk for later antisocial acts, arrests, and being jailed than is seen in clinic-referred adults with ADHD.
Both projects found that lifetime criminal diversity and arrest frequency were likely to be predicted by childhood hyperactivity or ADHD as well as earlier levels of CD symptoms. The Milwaukee Study also found that diversity of teen drug use also makes an independent contribution to both of these outcomes. This implies a spiraling effect over time between teen antisocial behavior and teen drug use in which each contributes to the maintenance and increase in the other at later developmental stages. We also found much the same for level of education such that it makes an independent contribution to crime diversity and arrest frequency beyond that made by earlier severity of antisocial behavior. Such analyses inform us that it is not severity of ADHD that is associated with crime and arrest rates (although childhood hyperactivity makes a small contribution to risk (about 7-8%). Additional and greater contributions are made by childhood conduct problems, teen antisocial activity and drug use, and education, while persistence of ADHD across development did not contribute to these outcomes.
Health and Related Lifestyles. Lifestyle and personality factors, especially conscientiousness, were noted earlier to be significant contributors to human longevity. Half or more of all deaths in the U.S. are related to drug use, diet, exercise, sexual behavior, driving, and risk-taking in general. Adults with ADHD are more likely to possess these high risk characteristics leading us to speculate that they will be in poorer health or at least carry significantly higher risks going forward for coronary artery disease, cancer, and accidental death, among other causes of death. We found good evidence to support such concerns. The adults with ADHD in the UMass Study had a higher percentage of individuals reporting problems in sleep, social relationships, family interactions, tobacco use, non-medical drug use, medical/dental care, motor vehicle safety, work and leisure, and emotional health than did the community control group. But illicit drug use, driving, and emotional health are areas in which adults with ADHD differed specifically from other clinic-referred adults who do not have ADHD. Similar, though not identical, elevations in lifestyle and health risks were also detected in the Milwaukee Study, with eating habits, sleep, social relations, tobacco use, non-medical drug use, and emotional health also being concerns for a significantly greater percentage of those hyperactive children having persistent ADHD to age 27 (H+ADHD group) compared to the community control group. Even those hyperactive children whose ADHD did not persist to adulthood had more members reporting concerns about sleep and tobacco use than the control group.
The Milwaukee Study took a detailed look at current health, medical history, and risk for future coronary heart disease (CHD). It found that the hyperactive group had a significantly greater risk for injury, nonsurgical hospitalizations, and poisonings and experienced more such events than the community group in their medical histories. A significantly greater number of current health complaints was evident in the medical histories of the persistently ADHD cases relative to those who no longer had ADHD at follow-up. The former also had more such medical complaints than the community group. These complaints were associated with elevated levels of somatization, depression, and phobic anxiety, implying that they may be more indicative of psychiatric rather than medical problems. We also identified a slight but significantly higher-risk lipid profile and a greater risk for CHD over the next 5 and 10 years, mainly for the H+ADHD group compared to the community control group. That group also reported less regular exercise than the other groups, while both hyperactive groups were more likely to be smokers and consumers of alcohol than the control group. If such trends continue over the next decade, it will become more evident that individuals with persistent ADHD into adulthood carry a higher future risk of heart disease (and possibly cancer) than the general population.
Money Management. Individuals who are more impulsive, have a penchant for immediate gratification and discount future consequences, and are generally poorer at self-regulation can be expected to have problems managing their finances. Given that these characteristics typify adults with ADHD, we hypothesized that those adults would have considerable problems managing money. Our hypotheses were largely borne out. The adults with ADHD in the UMass Study had a higher proportion of its members reporting problems with managing money, saving money, buying on impulse, nonpayment of utilities resulting in termination, missing loan payments, exceeding credit card limits, having a poor credit rating, and not saving for retirement. Relative to the normal control group, the adults with ADHD appear to be having relatively pervasive problems with the management of their finances. Several areas of money management were specifically elevated in the ADHD group more than in both the clinical and community control groups – deferred gratification (saving and putting money away for retirement), impulse buying, and meeting financial deadlines (nonpayment of utilities resulting in their termination). On all six frequency measures of money management, the adults with ADHD reported more difficulties more often than did the adults in our community control group. Money difficulties were also more common in the ADHD than in the clinical control group in at least four of these six areas – missing rent payments, missing utility payments, missing loan payments, and having more total money problems. Numerous financial problems were also associated with the hyperactive group in the Milwaukee Study, though these were most frequent in that group whose ADHD had persisted until age 27. Both studies have found a clear, robust, and specific relationship of adult ADHD to a diversity of financial problems, regardless of how adult ADHD patients were ascertained (clinic-referred or children followed to adulthood).
Driving Risks. Driving has probably been the most thoroughly studied major life activity affected by the disorder in the extant literature. Our results add more evidence to this burgeoning evidence of significant and pervasive risks. Such risks do not seem to be due to the common comorbid disorders associated with ADHD. Clinic-referred adults with ADHD were more likely to have had their licenses suspended or revoked, to have driven without a valid drivers’ license, to have crashed while driving, to have been at fault in such a crash, and to have been cited for speeding and even reckless driving compared to the community control group. Several of these risks were also documented on official DMV records. The ADHD group also had more license suspensions/revocations, more crashes, more speeding citations, and were held to be at fault in more such crashes than either the clinical or community control adults. On the DMV record, the adults with ADHD again had more speeding citations and more total citations. Similar though less robust differences were evident between the hyperactive and control groups in Milwaukee, perhaps in part because they are younger and have had less driving experience than adults in the UMass Study. However, like the UMass Study, children who were hyperactive experienced a higher risk for frequent crashes, a greater risk for reckless driving and more citations for such driving, and a greater risk for license suspensions and revocations. Driving risks were associated not only with ADHD severity, but also other factors such as age, more diverse criminal activity, poorer credit ratings, greater hostility (road rage?), and low levels of anxiety (fearlessness?), depending upon the driving outcome being predicted.
Sex, Dating, Marriage, and Offspring. Two prior studies have documented a riskier sexual lifestyle in teens and adults with ADHD, one of which was the Milwaukee Study at age 21 follow-up. We continued to identify this area as one of concern and greater medical and public health attention for those with ADHD. Childhood ADHD is associated with earlier initiation of sexual activity and intercourse, more sexual partners, more casual sex, and more partner pregnancies, or female pregnancies if the woman has ADHD. These risks are further elevated by higher levels of conduct problems but such problems do not account for the separate contribution made by ADHD. The Milwaukee Study continued to find evidence for concern in this domain of life activity. Children growing up with ADHD are more likely to become pregnant (if female) or impregnate others (if males), are more likely to be parents by ages 21 and 27, and are more likely to contract a sexually transmitted disease by age 21 than are community control children followed over this same time.
In keeping with the previous longitudinal studies of hyperactive children, and with the inconsistent results of past studies of clinic-referred adults, we did not find differential rates of marriage or higher rates of divorce among the children growing up with ADHD or among the clinic-referred adults with ADHD. We did find some evidence that women with ADHD were less likely to be married at the time of our studies. We also found a greater incidence of marital dissatisfaction in both groups of adults with ADHD across these projects as well as poorer quality of dating relationships among those hyperactive children with persistent ADHD to adulthood that are still single and dating. The spouses of those adults with ADHD were also significantly less satisfied in the marriage than were spouses of the community group in the UMass Study. Such findings, however, may not be specific to adults with ADHD or their spouses because we also found such marital dissatisfaction in our clinical control adults.
Prior studies have also found elevated risks for ADHD among the offspring of adults with ADHD, ranging from 43-57% of their children. We also found such an elevated risk, with 22-43% of the offspring of adults with ADHD falling in the clinically elevated range on either the inattention, hyperactive-impulsive, or combined symptom lists from the DSM-IV. These prevalence figures were certainly higher than those found in our community control adults. But we also used a clinical control group of adults, and found that their offspring also carried significantly elevated risks for symptoms of inattention than did the children of the community group, implying that offspring risk for inattention is also found in non-ADHD clinic-referred adults. In contrast, risk for hyperactive-impulsive behavior, oppositional defiant disorder, and conduct disorder appear to be specifically elevated in the offspring of the adults with ADHD. ODD appeared to be the most common psychological morbidity to be found in the offspring of adults with ADHD, occurring in nearly half of them (48%). Such findings are quite consistent with, and supportive of, the strong genetic predisposition to ADHD and its high familial transmission as demonstrated in numerous prior behavior genetic studies of biological family members and twins. Yet our research also evaluated a wider array of dimensions of children’s psychological maladjustment than the earlier studies of offspring morbidity. Here as well, we found that the children of adults with ADHD showed greater symptoms of both externalizing (ADHD, oppositional disorder, conduct disorder) and internalizing (depression, somatization, atypicality) problems than did children in the clinical and community control groups. There is a wider range of offspring psychological morbidity associated with ADHD in parents than is the case for either parents who do not have ADHD, whether clinically-referred or not.
The parents in both the ADHD and clinical control groups reported higher rates of parenting stress on the specific scales of parent, child, and parent-child domains than did community control parents, and did not differ from each other on these scales. However, the parents with ADHD reported more total stress in their family lives than did parents in either of the control groups. Parenting stress, particularly in those domains associated with the child or with parent-child interactions, was primarily predicted by the extent of child ODD symptoms consistent with prior research in children with ADHD. Degree of child inattentiveness was also a contributor to some domains of parenting stress suggesting that it also makes some contribution to the degree of perceived stress reported by parents. However, when parental characteristics of ADHD, depression, and anxiety were considered, our results showed that parental depression might make a large and consistent contribution to all domains of parenting stress. Child ODD symptoms continued to contribute to stress beyond that made by parental depression. Parental level of ADHD did not contribute to levels of parenting stress.
Neuropsychological Functioning. Substantial research exists on the neuropsychological performance of adults with ADHD, especially on measures thought to index the executive functions. The executive functions typically include the components of inhibition, resistance to distraction (interference control), verbal and nonverbal working memory, fluency (verbal and nonverbal), sense and use of time, and planning, among others. All of these have been found to be impaired in prior studies, though planning was less consistently identified. The UMass study documented an excess of omission errors and greater reaction time variability on a continuous performance test in our adults with ADHD relative to both control groups. The findings are consistent with some prior studies showing such problems with inattention while further identifying them as relatively more impaired and specific in ADHD in adults. Errors of commission were also greater in the ADHD group than the community group but seemed less specific to the ADHD group given the presence of greater errors in the clinical control group as well. Further analyses found that reaction time variability (inattention) and commission errors (inhibition) were the two best CPT measures for distinguishing the ADHD group from the community control adults while inattention was the only measure of much utility in identifying the ADHD group from the clinical control adults. Recent research shows that high levels of response variability may be specifically associated with ADHD in adults and typify an abnormal error pattern associated with impaired vigilance that is not merely an exaggeration of normal variability. We, therefore, recommended that high response variability may be a useful phenotype for behavioral genetic neuro-imaging studies of adults with ADHD.
Interference control or resistance to distraction, assessed using the Stroop Word-Color Task, was found to be impaired in adults with ADHD in both the UMass and Milwaukee Studies. But these differences in the UMass Study were only between the ADHD and community group. Such findings suggest that this form of inhibition may not be very specific to ADHD, at least among adults with the disorder. Even so, in the Milwaukee Study, it distinguished the hyperactive children with persistent ADHD to age 27 from those whose ADHD had not persisted.
Prior studies have had little success in identifying problems on the Wisconsin Card Sort Task as being reliably associated with ADHD, in either children or adults having the disorder. Our study also found no differences among out groups on any WCST measures. This further supports conclusions that response flexibility or set shifting is not a problem in adults with ADHD.
Few prior studies have examined fluency or generativity in adults with ADHD. Those that have concentrated mainly on verbal fluency measures and have shown mixed results. One study examining design or nonverbal fluency found greater problems with perseverative responses on this task. Both of our studies used a measure of design fluency. The UMass Study found the adults with ADHD to generate fewer responses on this task than did either control group of adults. The Milwaukee Study found that both groups of children who had ADHD in childhood had deficits on this test at adult follow-up regardless if their ADHD had persisted to this age or not. However, we did not find more perseverative design responses in either of our studies. Both studies suggest some problems with nonverbal fluency may be associated with ADHD in adults and that such a deficit persists into adulthood in children with the disorder, even if they no longer meet diagnostic criteria for ADHD.
Among the most reliable findings in past studies have been deficits in verbal working memory as indexed by digit span tasks, among others. Problems with verbal learning have been mixed as have past studies using other verbal working memory tasks besides digit span. Our own studies support the earlier findings of problems with digit span tasks in adults having ADHD, and that they are relatively specific to this disorder compared to our two control groups in the UMass Study. The Milwaukee Study once again found that clinic-referred children with ADHD followed to adulthood also continue to have deficits in this measure of verbal working memory even if their level of ADHD symptoms is no longer sufficient to be diagnosed as such.
The UMass Study used an extensive battery of verbal and nonverbal learning and memory tasks. In general, we found no evidence of difficulties with immediate verbal learning or recall in such tasks as paragraph, word list, or word pair learning using the UMass Study groups. Our UMass Study did find deficits associated with ADHD in adults in retention of information over time such that they were more impaired on delayed recall trials after several intervening tasks had to be performed. These retention problems were more evident in free recall and were improved somewhat by cueing of the likely response category. We believe these findings are consistent with a verbal working memory disorder in adults with ADHD similar to that previously shown in children with the disorder, and find it to be relatively specific to the disorder relative to both our control groups.
To summarize, ADHD in adults is associated with some relatively specific EF deficits not seen in community or clinical control adults. These are most likely to be in measures of inattention, inhibition (interference control), nonverbal and verbal working memory, and design fluency. But no deficits in verbal learning, immediate recall, set shifting, or tower planning were evident here.
Our numerous findings that were specific to the ADHD groups (Barkley et al., 2008) overwhelmingly support the validity of this disorder in adults in keeping with prior research and reviews on this issue (Spencer, Biederman, Wilens, & Faraone, 1994; Wilens, Faraone, & Biederman, 2004). For the most part, children with ADHD that persists into adulthood are remarkably similar to clinic-referred adults who may be newly diagnosed with ADHD in adulthood. Yet some noteworthy differences were evident here. Most of them found the clinic-referred adults with ADHD to be less impaired with one exception, and that was in risk for comorbid psychiatric disorders and especially anxiety and depression. Children growing up with ADHD report significantly lower risks for these disorders. Yet those same children as adults report greater impairment in their educational careers and occupational functioning than do clinic-referred adults with the disorder. Children with ADHD grown up also seem to have greater difficulties with antisocial behavior, frank antisocial personality disorder, and possibly drug use than do clinic-referred adults with the disorder. Both groups display significant problems in their health and lifestyle, money management, and driving. The Milwaukee Study also began to document evidence that the disorder may pose risks for future cardiovascular disease and shorter life expectancy if current findings of greater body mass index, smoking, alcohol use, and lower HDL cholesterol and regular exercise continue over the next decade of their lives.
Our findings not only have significant implications for the design of better diagnostic criteria and for the validity of the disorder itself, but also for clinicians who would evaluate and manage this disorder in adults. The myriad impairments we have identified as being associated with ADHD will call for a variety of psychiatric, psychological, educational, and occupational interventions and accommodations to more effectively assist these adults in the management of their disorder and in the reduction of these impairments associated with it. For a detailed discussion of treatments for ADHD in adults, including counseling and medications, see the text by Barkley (2006).
Specifically, clinicians need to be aware of and specifically assess for the high comorbidity of ADHD with other psychiatric disorders, particularly dysthymia, depression, oppositional defiant disorder, conduct disorder, alcohol use disorders, and drug use disorders more generically. Noteworthy is that the elevated risk for suicidal ideation and attempts associated with the disorder is driven largely by comorbid mood disorders and not so much by ADHD specifically. Such comorbid disorders and psychological problems are highly likely to require separate treatment approaches than those aimed at the management of ADHD symptoms and their related impairments. Thus, a package of treatments is more likely to be put in place for most adults with ADHD to address this complexity of clinical presentation than is any single drug or psychological therapy.
Our studies lead us to believe that ADHD in adults, particularly when seen in clinic-referred adults, is therefore likely to require poly-pharmacy more than is the case for childhood ADHD given these higher risks for comorbid mood and anxiety disorders. While anti-ADHD drugs, such as stimulants and nonstimulant norepinephrine reuptake inhibitors, are clearly indicated for such cases, they are unlikely to address the risk for mood disorders evident here that are likely to require separate medical (i.e., antidepressant) and psychological (i.e. cognitive-behavioral) treatments in their own right. The elevated risk for anxiety disorders in both clinic-referred ADHD adults and children with pe
ContinuingEdCourses.Net is approved by the American Psychological Association (APA) to sponsor continuing education for psychologists. ContinuingEdCourses.Net maintains responsibility for this program and its content.
ContinuingEdCourses.Net is approved by the Association of Social Work Boards (ASWB) to offer continuing education for social workers, through the Approved Continuing Education (ACE) program. ContinuingEdCourses.Net maintains responsibility for its courses. ASWB provider #1107.
ContinuingEdCourses.Net is approved by the National Board for Certified Counselors (NBCC) as an NBCC-Approved Continuing Education Provider (ACEP) and may offer NBCC-approved clock hours for events that meet NBCC requirements. ContinuingEdCourses.Net solely is responsible for all aspects of the program. NBCC provider #6323.