William B. Carey, MD
Remarks to the American Enterprise Institute Panel on Ritalin and ADHD
Washington, DC
June 18, 2001


Today we are discussing one of the most important contemporary issues in the health care of children, the diagnosis of Attention-Deficit/Hyperactivity Disorder (ADHD) and its treatment with methylphenidate (Ritalin). Some have charged that there is a conspiracy between the drug companies and the American Psychiatric Association (APA) to invent a diagnosis that can be profitably utilized to market their specific pharmaceutical products. I do not have any evidence to support or refute that claim, and, therefore, leave that matter to those involved in the contention.

What I can do is to spend my few minutes today calling attention to the certain problems and by suggesting some ways to go about solving them. The diagnosis of ADHD has dramatically increased in the last two decades and consequently prescriptions for Ritalin and other stimulants have grown at an alarming rate. We should take a very close look at what is going on, because much of this epidemic appears to have happened without a sound scientific basis. Furthermore, abuse of these increasingly available drugs by children and adults has become a major public health concern. I wish to stress at the outset that I come to this session as an advocate for children, a pediatrician and grandparent who is eager to help all of us do a better job.

I should explain that for over 30 years my chief research interest has been the normal temperament variations of children: what they are, how they matter for parents and for childrenís health, behavior, and school performance, and how best to manage them. (Carey & McDevitt 1995). What has drawn me to this controversial area of ADHD is the clear evidence that the diagnostic systems now promoted by the mental health establishment ignore this broad range of normal behavior in children, with the resulting tendency to overdiagnose annoying but normal traits and ascribe them to malfunction of the brain (Carey 1999, 2000).


What can we agree about? Abundant evidence supports the conclusion that 1-2% of children are so pervasively overactive or inattentive that those qualities by themselves severely impair chances for normal living and make these children very hard for any caregivers to manage A better-designed diagnosis of ADHD, like the more restrictive one used in the United Kingdom, could properly be used to describe these children (Taylor 1994). For that relatively small group medication may be a rational choice as a part of a larger plan. Yet, even for them it is generally not known whether the behaviors are due to abnormal brains, adverse environments, or both. These children come from various backgrounds, such as premature birth, drug exposure during pregnancy, families undergoing a variety of stresses, and severe deprivation in substandard foster or institutional care.

The problem-

My main concern is not for that 1-2% but for the rest of the 3-5% (or up to 10-15 %) of American children who today are being diagnosed with ADHD and usually being treated with stimulants inappropriately. These presently overdiagnosed children probably have a variety of other conditions: normal but annoying temperament variations, undiagnosed learning disabilities, adjustment disorders, anxiety, depression, fatigue, effects of abuse and neglect, and various other problems interfering with expected functioning.

An example- Let me illustrate with the story of a five-year-old boy, whom I saw as a consultation several months ago. His parents came to me with the report that the presumably competent teacher had informed them that he had ADHD and should be treated with Ritalin. The parents doubted this non-professional analysis and wanted another opinion. The family pediatrician was not sure and sent them on to me. My review of his behavior revealed clearly that he was not overactive, not impulsive, and not distractible. He did have some challenging temperament traits including shyness, slow adaptability, and not as sunny a disposition as one might like. About 10% of normal children have this ìspirited,î challenging, or ìdifficultî behavioral style, which makes them hard for adults to manage, but they are not necessarily dysfunctional (and this child was not) and they do not fit the existing criteria for ADHD. The teacher mistook his difficult but normal temperament for the popular diagnosis of ADHD. I believe that such misinterpretations are very common. The child is now doing well with a different teacher.

In my opinion there are three principal problems that have contributed to this confusion.

I) The ADHD diagnosis is seriously flawed. Dr. Wiener has already mentioned the criteria. How are they inadequate?

The DSM-IV system fails to distinguish the supposedly abnormal ADHD behaviors of activity and inattention from the same behaviors which are normal temperament variations. It assumes that these behaviors are signs of an underlying, pre-existing malfunction of the brain, although they are just as likely, if not more so, to be annoying normal traits in neurologically intact children that sometimes lead to social or academic dysfunction through stressful interactions with the environment. In other words, the current diagnostic system is part of the problem today because it confuses cause and effect. It also fails to distinguish adequately inattention due to other clinical problems, such as learning disabilities or anxiety, thus contributing to the ìcomorbidityî problem.

There is no clear evidence that 6 out of 9 or any specific number of the ADHD behaviors are related to abnormal brain function. These ìcut-pointsî are arbitrary and unsupported by empirical data. Some preliminary brain imaging studies have shown inconsistent differences in children with the diagnosis but there is no proof that they are abnormalities. The findings may just be traces of the predisposing normal temperament traits rather than an indication of dysfunction. The same cautions must be applied to our appraisal of the reports of genetic factors in the causation of ADHD.

The present ADHD diagnostic criteria completely ignore the role of the environment and interactions with it as causes of the behaviors and the likelihood of referral. The problem is seen as being entirely in the child. Not only is this assumption generally unsupportable by the evidence but this also means that the true complexity of the situation is likely to be overlooked and the management oversimplified.

The diagnostic criteria and the questionnaires used to assess them are highly subjective and impressionistic. How is one to rate reliably such vague items as ìoften talks excessivelyî or ìoften has difficulty awaiting turn?î They probably measure parent or teacher discomfort or perplexity as much or more than the actual behavior of the child. Thus, the observer is being allowed to determine what is abnormal. And yet this is the flimsy basis on which a categorical diagnosis of brain malfunction is being made.

Several other problems with the ADHD diagnosis as it stands today include: 1) the fact that the most important factors in children getting the diagnosis are probably either learning problems or low adaptability (which some call poor executive functions); 2) its lack of cultural and historical perspective; and 3) the small practical usefulness and possible harm from the label.

II) The widespread misapplication of the present ADHD label. Whether or not one approves of the ADHD criteria as they stand today, there is strong evidence that at the practical level they are not being faithfully applied in most cases. Recently two comprehensive studies, one of over 400 pediatricians throughout the country (Wasserman et al. 1999) and one of family physicians and pediatricians in western North Carolina (Angold et al. 2000) demonstrated that the accepted diagnostic criteria were used less than half of the time. We have no evidence that other surveys would be more reassuring.

III) The nonspecific effects of Ritalin and other stimulants. Many professional persons and members of the general public still believe that, if stimulant medication leads to an improvement in the childís behavior, that is solid proof of the diagnosis of ADHD and good reason to continue the drug. What they evidently do not understand is that, as with other cerebral stimulants such as caffeine, the effect of improved attention is experienced by almost all who take it, including completely normal children. The popular practice of ìa trial of Ritalinî for diagnosis is, therefore, irrational (Diller 1998).

The solution.

Let us direct our energies toward specific solutions of these real problems in order to advance our knowledge and enhance the well-being of our children.

A better diagnostic system. The DSM Task Force (which does not include even one pediatrician) must recognize the full range of normal variations of behavior and offer more precise criteria for defining abnormality. Disorders should be defined in terms of disorders (social problems, school underachievement, etc.), not in terms of possibly predisposing risk factors like activity and attention span. Brain malfunction should not be asserted or implied unless there is solid proof of it.

Better research. The fact that the diagnosis is so unclear demonstrates that we need a great deal more study to provide a better understanding of these problems. It is simply not true that, as certain prominent spokespersons claim, we have almost conclusive scientific knowledge now and that any who doubt this are guilty of promoting ìmythologyî or ìfairy tales.î

Broader education. Professional persons and the public need much clearer and more detailed instruction as to what we really do and do not know. Ignorance of the range of normal and the nonspecificity of Ritalin effects are but two of the current widespread knowledge deficits.

Better evaluations. Children with problems in school performance need thorough evaluations of their physical, developmental, cognitive, and behavioral status (temperament and adjustment) and their environmental settings and interactions with them. In particular, an adequate psychoeducational examination should be included. Five or 10 minutes in the physicianís office for a Ritalin prescription is an unacceptable current practice.

Better treatment. The management of the child must be individualized for his or her unique needs. One treatment does not fit all.


The main point of my remarks has been that the best use of our resources would be better research, revising the ADHD diagnosis itself, applying it with greater care, and better education of professionals and the public. I do not know if the alleged conspiracy exists. However, there will surely be a failure of responsible citizenship if those of us who are aware of the problems I have described continue to tolerate them without any attempt at correction. Our children deserve much better than what they are getting at present.  


Angold A, Erkanli A, Egger HL, Costello EJ. Stimulant treatment for children: A community perspective. J Am Acad Child Adolesc Psychiatry. 2000;39:975-984.

Carey WB. Is ADHD a valid disorder? Invited presentation on November 16, 1998, at NIH Consensus Development Conference on Diagnosis and Treatment of Attention Deficit Hyperactivity Disorder, Bethesda, MD. Longer version in press.

Carey WB. Problems in diagnosing attention and activity. Pediatrics. 1999;103:664-667. Commentary.

Carey WB. What the Multimodal Treatment Study of Treatment of Children with Attention-Deficit/Hyperactivity Disorder did and did not say about the use of methylphenidate for attention deficits. Pediatrics. 2000;105:863-864. Commentary.

Carey WB, McDevitt. Coping with Childrenís Temperament. New York. Basic Books. 1995.

Diller L. Running on Ritalin. A Physician Reflects on Children, Society, and Performance in a Pill. New York. Bantam. 1998.

Taylor E. Syndromes of attention deficit and overactivity. In: Rutter ML,Taylor E, Hersov L (Eds.) Child and Adolescent Psychiatry. Third Edition. Oxford, UK: Blackwell Scientific. 1994:294.

Wasserman RC et al. Identification of attentional and hyperactivity problems in primary care. Pediatrics. 1999;103:e38.

Go to "Things to do at this website"