Attention-deficit hyperactivity disorder (ADHD) is a well-established diagnosis in children. However, several studies have indicated that the disorder persists with age: Weiss and others found that disabling symptoms persisted into adulthood in 66% of subjects, and moderate-to-severe symptoms persisted in 51% (1); Gittelman and others found that full diagnostic criteria persisted in 31%, and disabling residual symptoms persisted in a further 5% (2); Mannuzza and others’ reevaluation of Gittelman’s group reported 8% with full diagnostic criteria for ADHD and an additional 3% affected by some ADHD symptoms (3). Others have studied the overall incidence of ADHD in adults: in a community sample, Murphy and Barkley found a 1.3% prevalence of ADHD inattentive type, a 2.5% prevalence of ADHD hyperactive- impulsive type, and a 0.9% prevalence of ADHD combined type (4). Murphy and Barkley emphasized in 2 papers (4,5) that the prevalence of ADHD is underestimated in adults because existing diagnostic criteria are developmentally inappropriate. This has produced debate but no resolution, and the matter requires further research (6).
Since symptoms of ADHD are treated successfully with medications in childhood, it seems logical that adults with similar symptoms could also benefit from such treatment. Wilens and others reviewed studies on the effectiveness of medication in adults with ADHD; these authors examined studies that used stimulant (pemoline or methylphenidate) and nonstimulant (tricyclic antidepressants, antihypertensives, and amino acids) medications (7). The response varied across both groups. The medication used in the greatest number of studies was methylphenidate (of these, 6 studies were double-blind, and 1 was an open trial). The variability of response rates (25% to 78%) was attributed to differences in the studies, in diagnostic criteria, and in stimulant dosage, as well as to variations in study population comorbidity and differing outcome measures. The review showed that few studies using stimulants were double-blind crossover studies comparing medication to placebo. No study used objective measures such as the Continuous Performance Test (CPT) to evaluate treatment response.
The 3 double-blind, placebo–methylphenidate crossover studies described below (8–10) involved adults with ADHD and are similar in method, measures of ADHD symptomatology, and experimentation time; they differ mainly in the dosages of methylphenidate given (with some overlap). They show variable results: Mattes and others reported no significant improvement in attention deficit disorder (ADD) symptomatology with “low/moderate” dosages (8). These authors also reported that only 25% of their subjects improved on an average daily dosage of 0.7 mg/kg and a maximum daily dosage of 60 mg, given in divided doses twice daily. Conversely, Wender and others reported significant improvement in 57% of their ADD group with similar “low/moderate” dosages of methylphenidate (9). In their study, the average daily dosage was 0.6 mg/kg and the maximum daily dosage was 80 mg, given in divided doses 3 times daily (9). Spencer and others reported a response rate of 78% with daily dosages of 1 mg/kg of methylphenidate, given in divided doses 3 times daily (10). These researchers reported significant, albeit lesser, improvement in ADHD symptomatology with lower daily dosages of methylphenidate (specifically, 0.5 mg/kg and 0.75 mg/kg).
It is difficult to compare these 3 studies directly. Wender and others focused on studying the efficacy of methylphenidate in adults with ADHD (9); Spencer and others on whether a dosage–response curve existed for methylphenidate (10); and Mattes and others on medication effect in 2 different groups of subjects, based on childhood histories (8).
In summary, available controlled medication studies in adults are few to date and show conflicting results in response rates and optimal dosage (specifically reference 8, compared with references 9 and 10). More controlled studies are required on the treatment of adults with ADHD.
The purpose of the double-blind, placebo-controlled study reported in this paper was twofold: first, to determine the relative efficacy of low and moderate dosages of methylphenidate in treating adults with ADHD; and second, to evaluate changes in symptoms, using both subjective measures of ADHD symptoms (self-report questionnaires) and objective measures of attention and response inhibition (computerized tests). We believe objective-computerized tests have never been used in any adult ADHD study to date. The low dose (0.4 mg/kg) was found beneficial in clinic patients. The moderate dose (0.6 mg/kg) was found beneficial by Wender and others (9). Both doses had few side effects, yielding better compliance. We hypothesized that methylphenidate would be better than placebo in reducing ADHD symptoms under both objective and subjective tests and that a moderate dosage would be better than a low dosage in reducing symptoms.
Screening and Pretrial Evaluation
A flow chart (Figure 1) shows the evaluation procedure. We screened subjects in the following 4 steps:
Figure 1 Flow chart of evaluation process
We administered the following 2 computer tests to obtain objective measures of attention and impulsivity:
Before starting the trial, each subject was seen by a psychiatrist who described the medication trial, reviewed all prior information on the subject, and interviewed the subject about childhood and current ADHD symptoms, current life situation (that is, work or school and family), and medical and psychiatric history (including drug and alcohol use). Each subject then filled out a baseline side effect form (that is, a review of symptoms scale ranging from mild to moderate to severe), and the psychiatrist performed a focused physical exam (measuring weight, heart rate, and blood pressure). Any concerns raised during screening (that is, an elevated BDI score) were carefully reviewed during the interview to determine whether the subject was suffering from a disorder requiring other immediate treatment. Subjects who did not continue in the study were referred to appropriate resources.
Subjects had to meet the following inclusion-exclusion criteria to participate in the study:
We requested subjects to take their medication 1 hour before testing to ensure a satisfactory level of medication during testing. The dosage was increased to 15 mg 3 times daily for 2 subsequent weeks, after which we asked subjects to return for a reevaluation similar to that undertaken after the first 2 weeks. At the end of this 4-week period, each subject had a minimum 5-day washout. The sequence starting with the lead-in was then repeated with the second “medication” (either methylphenidate or placebo).
Of the 93 people (69 men and 24 women) screened for this study, 55 were not enrolled because their scores were too low (n = 16), they were already on medication (n = 11), they were not interested (n = 8), they were not blind to methylphenidate (n = 5), or they were not appropriate subjects for other reasons (n = 15). A further 8 dropped out after starting because of side effects (n = 1), because they were not blind to methylphenidate (n = 4), because of “too much going on” (n = 1), and for unknown reasons (n = 2). A total of 30 adults (24 men and 6 women) completed the study. Their ages ranged from 17 to 51 years (mean 34 years). Their average IQ was 101. They came from a wide range of educational and socioeconomic backgrounds.
Effects on ADHD Symptoms
Scores on the Adult ADHD Problem Behaviours Scale were significantly lower than baseline for both placebo and methylphenidate; they were significantly lower for methylphenidate than for placebo (mean 0.9 vs mean 1.2, P < 0.005).
Similarly, scores on the Conners’ Adult ADHD Rating Scale were significantly lower than baseline for both placebo and methylphenidate and significantly lower for methylphenidate than for placebo (mean 1.0 vs mean 1.4, P < 0.01).
The CPT commission error rate was lower than baseline for both placebo and methylphenidate and significantly lower for methylphenidate than for placebo (mean 17% vs mean 26%, P < 0.001).
The CPT baseline omission error rate in our group (mean 4.3%) was lower than the “norm” (mean 13.8%) for ADHD measured according to Conners’ scale (18). For placebo, the CPT omission error rate was not significantly different from baseline; however, the omission error rate was significantly lower than baseline for methylphenidate (mean 4.3% vs mean 1.2%, P < 0.0001), and there was a trend toward a significantly lower omission error rate for methylphenidate, compared with placebo (mean 1.2% vs mean 3.8%, P < 0.1).
The stop-signal task reaction time was lower than baseline for both placebo and methylphenidate but did not differ significantly for methylphenidate, compared with placebo.
For this study, we only considered medication responses greater than the relatively high placebo response. We calculated the response rate as follows:
This calculation yielded a response rate (based on improvement on self-report questionnaires scores) of 63%. Overall, 73% of the subjects made fewer commission errors on methylphenidate than on placebo. Therefore response rate was between 63% and 73%.
Overall, methylphenidate appears to be significantly superior to placebo in reducing ADHD symptoms on both self-report questionnaires and on computer tests.
Effects on Other Measures
We compared GAF scores in a subsample of 13 subjects (chosen because a single rater gathered all data). The mean GAF scores were 68.5 at baseline, 69.9 on the highest dosage of placebo, and 75.3 on the highest dosage of methylphenidate. There was no significant difference between baseline and placebo, but there was a highly significant difference between baseline and methylphenidate (P < 0.0001) and a further significant improvement in methylphenidate, compared with placebo (P < 0.01).
Table 4 shows other reported symptoms. Generally, there were no significant differences in other reported symptoms among baseline, placebo, and methylphenidate. On average, over one-quarter of the sample had some physical complaint before starting methylphenidate. The principal complaint with methylphenidate was decreased appetite. Mild-to- moderate appetite decrease was found in 41% of subjects with the higher dosage of methylphenidate, compared with 23% at baseline and 19% with placebo. Despite this, there was no significant weight loss (mean weights were 75.8 kg at baseline, 78 kg with placebo, and 76.4 kg with methylphenidate). Insomnia was a common complaint at baseline (41% of subjects experienced mild-to-moderate insomnia). This improved on placebo (25% of subjects reported mild-to- moderate insomnia) and on methylphenidate (26% of subjects reported mild-to-moderate insomnia). Headache was also a common complaint: at baseline, 25% of subjects complained of headache, compared with 35% on placebo and 21% on methylphenidate. A small proportion of subjects cited irritability, light-headedness, and motor ticks as symptoms, but these symptoms seemed unrelated to any study condition.
This double-blind crossover study found that, compared with placebo, low or medium dosages of methylphenidate reduced the symptomatology of ADHD in adults, as measured by both subjective and objective tests. This finding supports the efficacy of methylphenidate for ADHD symptoms in adults and indicates that low and moderate dosages are effective. It does not resolve whether higher dosages of methylphenidate further reduce ADHD symptoms but suggests that some subjects may not require high dosages.
Methylphenidate and placebo were tried for 1 month each. We chose this short period because stimulants can be adequately evaluated within a month in children and to encourage greater participation and compliance. The significant placebo response may last as long as 4 weeks. This underlines the importance of using a double-blind placebo trial in evaluating the response to ADHD medication. Longer trials may be required to adequately assess methylphenidate’s efficacy on global function and to decrease the placebo effect.
The subjects in this study were neither comorbid for other psychiatric illnesses nor did they have pure ADHD. Overall, ADHD symptomatology was in the moderate range; few were severely affected. This may explain the greater placebo response found in this study, compared with Spencer and others’ study (10), where subjects may have had more severe ADHD symptoms or more comorbidity.
The greatest effect of methylphenidate on computerized tests was to lower the commission error score of the CPT (73% of the group showed an improvement with methylphenidate over placebo). This confirms that methylphenidate reduces ADHD symptoms. This positive effect on attention and response inhibition suggests that methylphenidate may also improve work performance in adults with ADHD. The stop-signal task is relatively new and not discriminatory in this treatment study with methylphenidate. Perhaps a higher dosage of methylphenidate would be required to get a significant improvement on this test.
Non-ADHD symptoms improved with methylphenidate: it reduced symptoms of anxiety and showed a trend toward reducing depressive symptoms, compared with placebo. There was also more improvement with methylphenidate than with placebo on obsessive–compulsive scores and a trend toward improvement on SCL-90-R hostility scores.
Overall, rating scale scores of anxiety and depression in subjects were low. Methylphenidate may directly relieve non-ADHD symptoms, but it seems more plausible that this improvement resulted from increased general well-being secondary to reduced ADHD symptoms. Anxiety may decrease if a subject’s performance at work improves.
The study population was varied, with a 4:1 preponderance of male subjects over female subjects—a ratio similar to that reported for children but different from other adult studies with a higher proportion of women. There is no clear explanation for this. It may be a close approximation to the true ratio or a chance sampling within a small population. Our referral sources may have tended to refer men with ADHD symptoms while categorizing women in more depressive diagnoses. One expects more women to be referred in future, as community referral sources become more aware of adults with ADHD.
The low and moderate dosages used in this study are well tolerated, seem equally effective, and are comparable with those given to children. Side effects were minimal, and some symptoms were common in many individuals before the study. Adults with ADHD may suffer from greater somatic complaints or be more aware of their physical malaise. Although the increase in systolic blood pressure measured with methylphenidate does not seem clinically significant, it indicates methylphenidate’s effect on the cardiovascular system. This suggests that blood pressure should be monitored with the administration of methylphenidate.
Study Limitations and Future Directions
Subject evaluation did not include reports on symptomatology from spouses, friends, coworkers, or employers. This information is harder to obtain for adults than for children. Such reports could have further confirmed our data. However, computer test results confirmed subjective data measured by self-report, suggesting that our data are valid.
The sample size (30 subjects) was relatively small. A larger sample would clarify the effects of methylphenidate on comorbid symptoms and establish whether methylphenidate has a differential response on attention, impulsivity, and hyperactivity. Our sample was not large enough to identify an order effect.
Improvement of ADHD symptoms seems to be a clear response to methylphenidate. Although encouraging, this finding is difficult to extrapolate to individuals with ADHD and comorbid psychiatric symptoms. The efficacy of methylphenidate on populations with comorbidity needs more research and was not addressed.
This study was short-term, which stands out most clearly in the placebo-response finding. It would be worthwhile to determine the duration of placebo response and to determine whether the beneficial effects of methylphenidate continue over time in adults, as in children.
Methylphenidate was superior to placebo in alleviating ADHD symptomatology (using both subjective and objective measures) in 30 adults diagnosed with ADHD. Methylphenidate was better than placebo at alleviating symptoms of anxiety and showed a trend toward significance in alleviating depressive symptoms. There was no significant difference between the 2 dosages (10 mg 3 times daily and 15 mg 3 times daily) of methylphenidate. Methylphenidate was well tolerated, with no significant clinical effect on physiological measures (that is, blood pressure, pulse, and weight) and minimal self-reported side effects.
Many questions remain to be investigated in the medication treatment of adults with ADHD. Further studies should address dosage response, medication treatment effects over a longer term, populations with comorbidity, and functional impairment of work performance and personal relationships.
Funding and Support
This research was supported by an FRSQ grant for the study of adults with ADHD.
The authors acknowledge the support of the Department of Psychiatry, McGill University, and the Research Institute, Montreal Children’s Hospital. The authors also thank Dr Eric Ochs and the Department of Pharmacy, Montreal Children’s Hospital, for their help during this project.
1. Weiss G, Hechtman L, Milroy T, Perlman T. Psychiatric status of hyperactives as adults: a controlled prospective 15-year follow-up of 63 hyperactive children. J Am Acad Child Adolesc Psychiatry 1985;24:211–20.
2. Gittelman R, Mannuzza S, Shenker R, Bonagura N. Hyperactive boys almost grown up. Arch Gen Psychiatry 1985;42:937–47.
3. Mannuzza S, Klein RG, Bessler A, Malloy P, LaPadula M. Adult outcome of hyperactive boys. Arch Gen Psychiatry 1993;50:56–76.
4. Murphy K, Barkley RA. Prevalence of DSM-IV symptoms of ADHD in adults licensed drivers: implications for clinical diagnosis. J Atten Disord 1996;1:147–61.
5. Barkley RA, Murphy K, Kwasnik D. Psychological adjustment and adaptive impairments in young adults with ADHD. J Atten Disord 1996;1:5–54.
6. Shaffer D. Attention deficit hyperactive disorder in adults. Am J Psychiatry 1994;151:633–8.
7. Wilens TE, Biederman J, Spencer T, Prince J. Pharmacotherapy of adult attention-deficit hyperactivity disorder: a review. J Clin Psychopharmacol 1995;15:270–9.
8. Mattes JA, Boswell L, Oliver H. Methylphenidate effects on symptoms of attention deficit hyperactive disorder in adults. Arch Gen Psychiatry 1984;41:1059–63.
9. Wender PH, Reimherr FW, Wood D, Ward M. A controlled study of methylphenidate in the treatment of attention deficit disorder, residual type, in adults. Am J Psychiatry 1985;142:547–52.
10. Spencer T, Wilens T, Biederman J, Faraone SV, Ablon JS, Lapey K. A double- blind, crossover comparison of methylphenidate and placebo in adults with childhood-onset attention deficit hyperactive disorder. Arch Gen Psychiatry 1995;52:434–43.
11. Wender PH, Reimherr FW, Wood DR. Attention deficit disorder (‘minimal brain dysfunction’) in adults. Arch Gen Psychiatry 1981;38:449–56.
12. Conners CK, Erhardt D, Sparrow E. Adult ADHD Rating Scale. In: Mariush ME, editor. The use of psychological testing for treatment. planning and outcome assessment. Hillsdale (NJ): Erlbaum; 1994.
13. Barkley RA. Adult ADHD Problem Behaviours Scale. University of Massachusetts Medical Center. New York: Guilford Press; 1991.
14. Derogatis LR, Lazarus L. SCL-90-R, brief symptom inventory, and matching clinical rating scales. In: Mariush ME, editor. The use of psychological testing for treatment planning and outcome assessment. Hillsdale (NJ): Erlbaum; 1994.
15. Hamilton M. The assessment of anxiety states by rating. Br J Med Psychol 1959;32:50–5.
16. Beck AT. The Beck Depression Inventory. San Antonio (TX): Harcourt Brace Jovanovich; 1978.
17. Wechsler D. WAIS-R. The Psychological Corporation. San Antonio (TX): Harcourt Brace Jovanovich; 1981.
18. Conners CK. The Conners Continuous Performance Test. North Tonawanda (NY): Multi Health Systems; 1993.
19. Schachar RJ, Logan GD. Impulsivity and inhibitory control in normal development and childhood psychopathology. Dev Psychol 1990;26:710–20.
20. Wilkinson L. SYSTAT 6.1. Chicago (IL): SPSS Inc; 1996.
Manuscript received August 2002, revised, and accepted December 2002.
Previously presented as a poster at the Canadian Academy of Child Psychiatry; September 1996; Quebec (QC). Previously presented as a poster at the American Academy of Child Psychiatry; October 1996; Philadelphia (PA). Previously presented as a paper at the 11th annual conference of the Royal Australian and New Zealand College of Psychiatrists; October 1998; Sydney (NSW).
1. Assistant Professor of Psychiatry and Pediatrics, McGill University, Montreal, Quebec; Consultant Child and Adolescent Psychiatrist, Queen Alexandra Centre for Children’s Health, Vancouver Island Health Authority, Victoria, British Columbia.
2. Professor of Psychiatry and Pediatrics, Director of ADHD Research, and Clinical Director of Adolescent Services, Montreal Children’s Hospital, Montreal, Quebec.
3. Professor of Psychiatry and Pediatrics, McGill University, Montreal, Quebec; Clinical Director of the Anxiety Disorder Clinic and of Adolescent Services, Montreal Children’s Hospital, Montreal, Quebec.
4. School Psychologist, North Clackamas School District, Portland, Oregon.
Address for corespondence: Dr L Hechtman, Department of Psychiatry, Montreal Children’s Hospital, 4018 Ste Catherine Street West, Montreal, QC H3Z 1P2
1 | 2
Archives in English | Archives
RCP en français