Following a challenging and stimulating day in the office a few years ago, I received a phone call from Dr. James Pen-in telling me that the American Academy of Pediatrics was interested in the development of a practice guideline on attention-deficit /hyperactivity disorder (ADHD) with a focus on primary care pediatrie practice. Jim asked if I would co-chair the committee responsible for the development of the guideline. This would be the first practice guideline from the Academy of Pediatrics on a neurobehavioral condition. Aware of the uncertainties among many pediatricians about the precision of the diagnosis and treatment options in children with ADHD, as well as wide practice variations, I accepted the invitation.
Colleagues and friends counseled me against participation in the project. There were a variety of reasons given: not enough scientific evidence for a practice guideline; uncertainty about diagnostic criteria for most behavioral conditions; or not wanting more medications prescribed for these children. A few of my colleagues believed that diagnosis and management of children with ADHD was best left to child psychologists and psychiatrists.
Their points of view did not dissuade me from the importance of the project. Like most primary care pediatricians, I had seen children with ADHD in my practice for many years. Although I felt comfortable with the diagnostic process and treatment of children with the condition, I knew that my practice was not grounded in the same scientific foundation as treating children with other common chronic conditions such as asthma and chronic otitis media with effusion.
In this article, I describe the process that led to the two AAP clinical practice guidelines: Diagnosis and Evaluation of the Child with ADHD, and Treatment of the School-aged Child with ADHD.1'2 The AAP directed us to ensure a multidisciplinary approach, to focus the guidelines on practical issues in the diagnosis and treatment in primary care pediatrie practice, and to develop an evidence-based document consistent with other AAP practice guidelines. Most important, the guidelines were to be developed with the goal of changing practice patterns in a manner that reflected the best available evidence.
An outstanding multidisciplinary group of clinicians with clinical and research experience in ADHD was assembled. They represented general pediatricians, developmental-behavioral pediatricians, child and adolescent psychiatrists, child psychologists, child neurologists, specialists in neurodevelopmental disabilities, family practitioners, educators, and experts in clinical research methodology and data analysis. The committee also invited consultants to clarify specific diagnostic and treatment issues.
It was the consensus of the committee members that most of the scientific data on diagnosis and treatment of patients with ADHD was limited to school-aged children (between 6 and 12 years old). Studies of practice patterns suggested that most of the children with ADHD seen by primary care pediatricians were also in this age group. In addition, it was apparent that although there was an emerging literature in ADHD among adolescents and a sparse literature on ADHD in preschool children, we would be on firm scientific grounds if we concentrated our recommendations on the school-aged child.
We agreed that the guidelines should be divided into 2 parts, one on diagnosis and one on treatment. The types of evidence available were somewhat different in the two areas and, using this format, we planned to publish a diagnosis guideline while working on the treatment guideline.
My first reaction to the requirement from the AAP that this should be an evidence-based guideline was "Are not all guidelines evidence based? Isn't that the way we have always practiced medicine based on the best available data found in the medical literature?" In the process of answering the question, the committee read many articles on evidence-based medicine and had the benefit of several committee members who were knowledgeable about its significance in the context of clinical medicine. Briefly stated, the principles of evidence-based medicine, which have informed practice guidelines among all medical specialties since 1990, include recommendations that are: 1) accurate based on the best available scientific, randomized, controlled studies; 2) accountable with rational explanations consistent with available data; 3) predictable in a manner that will ensure that clinicians will be able to use the recommendations effectively in their practice; 4) defensible through clear reasoning based on evidence; and 5) usable in the context of current clinical practices. 3
What became apparent, as we asked questions about diagnosis and treatment, is that an evidence-based guideline has important requirements. Most significant is its reliability on randomized, controlled clinical trials that meet predetermined criteria for consideration. I was surprised to discover how many published studies in reputable, peer-review journals did not meet stringent evidenced-based criteria. It is this process that assures the best scientific foundation to our clinical practice.
We reasoned that if we adhered to the principles of evidence-based guidelines and demonstrated a sound scientific foundation for the recommendations, primary care pediatricians would begin to change their practice patterns. In addition, the guidelines were based on evidence generated from primary care pediatrie practices whenever possible rather than data from subspecialty, referral centers.
THE PREVALENCE OF ATTENTIONDEFICIT/HYPERACTIVITY DISORDER
Children with behavior patterns similar to what we now call ADHD have been described in the medical literature for more man 100 years. The definition and name of the condition changed several times during this period. In contemporary clinical practice and research, children with ADHD are identified with three core behavioral symptoms: hyperactivity, impulsivity, and inattention. The degree to which each symptom manifests in an individual child may vary significantly.
The diagnostic format found in almost all research studies in the past 40 years has been informed by the Diagnostic and Statistical Manual of Mental Health Disorders, Fourth Edition (DSMIV)* published by the American Psychiatric Association and now in its fourth edition. Recognizing the significant limitations in evidence-based studies designed to determine diagnoetic criteria in a behavioral condition, the committee concluded that the DSM-IV contains the best available criteria for ADHD (Table 1). A critical element of these criteria is the requirement for documentation of at least 6 of 9 behaviors in the hyperactive/impulsive domain and/or in die inattentive domain. Importantly, these behaviors must occur "inappropriately often for the developmental age of the child/' be present for at least a 6-month period, present prior to 7 years of age, and have a significant impairment on learning and /or social interactions in more than one setting (eg, home and school).
DSM-IV Criteria for Attentlon-Deflclt/Hyperactlvlty Disorder*
We reviewed the literature on the prevalence of ADHD, using the DSM criteria, in school-aged children. In most textbooks and journal articles, a figure of 3% to 5% is mentioned. We discovered that this is an extrapolated prevalence from those children referred to tertiary care centers and psychiatric clinics. When we evaluated epidemiology studies that used community samples (ie, evaluating all children in a school system or in general pediatrie, community-based clinics), the prevalence in 11 studies was 4% to 12% with most of the larger studies showing a prevalence of 6% to 9%.5 The male predominance of about 3:1 was consistent with prior studies.
DOCUMENTATION OF ATTENTIONDEF1C1T/HYPERACTIVITY DISORDER BEHAVIORS
Methods for the accurate documentation of ADHD behaviors and other DSM criteria were a critical question for the practice guideline committee. We concluded that the studies supported the recommendation that the assessment of ADHD requires reliable evidence directly obtained from parents (or caregivers) and the classroom teacher (or other school professional who has observed the child in the classroom) regarding the core symptoms of ADHD, the duration of symptoms, the degree of functional impairment, and coexisting conditions.
The committee was interested in the question, "What direction can be found in the literature to ensure accuracy and efficiency in obtaining this information?" To answer this question, as well as the prevalence data discussed above, and information about co-existing mental health and learning conditions, the committee worked with Technical Resources International, Washington, DC, under the auspices of the Agency for Healthcare Research and Quality, to develop the evidence for these topics.5
The literature review concluded that behavior questionnaires for parents and teachers that are specific for the diagnosis of ADHD can be used in office practice. These behavior scales ask questions about each of the 18 behaviors in the DSMIV criteria for ADHD. There are several published forms available. Our literature review did not support the use of broadband or global behavior rating scales that are not specific for ADHD. However, some clinicians find these broadband scales useful for determining other conditions that may coexist with ADHD (oppositional behaviors, anxiety, and depression).
Although the evidence-based literature supports the use of ADHD specific rating scales, we could not find evidence to suggest that other forms of data gathering were not equally effective. We concluded that behavior symptoms may be obtained from parents or guardians using 1 or more methods including open-ended questions, focused questions about specific behaviors, semistructured interview schedules, questionnaires, and rating scales. Whatever the method of data gathering, the clinician must record relevant behaviors of inattention, hyperactivity, and impulsivity from the DSM-/V.
The diagnostic guideline emphasizes that, "As data are gathered about the child's behavior, an opportunity becomes available to evaluate the family environment and parenting style. In this way, behavioral symptoms may be evaluated in the context of the environment that may have important characteristics for a particular child." Information from the child's teacher or other school personnel who have observed the child in the classroom may be obtained from a verbal narrative, written narrative, questionnaire, or rating scale. The rating scale has the advantage of assessing all 18 ADHD behaviors; the teacher narrative has the advantage of providing a description of the child that gives the clinician more insight into a particular child's behavior and learning style and the classroom experience.
The literature review of school-aged children with ADHD found that as many as one third of the children have a coexisting condition that includes mental health disorders (Table 2) and learning disabilities.6 These associated conditions are found frequently in children with ADHD seen in the offices of primary care clinicians and not limited to children cared for by psychiatrists and psychologists. Motor disabilities, problems with parent-child interactions, and family violence should also be considered. In the initial evaluation, clinicians should ask general and specific questions to screen for these conditions.1
Prevalence of Selected Coexisting Conditions In Children With AttentlonDeflclt/Hyperactlvlty Disorder
ADHD is a neurobehavioral disorder without a biological marker. Similar to all other behavioral disorders, accurate assessment is dependent on an accurate history. In the development of the diagnostic guideline, we reviewed the literature on laboratory tests mat have been recommended by some clinicians as a part of a diagnostic evaluation for ADHD. We concluded that there is not sufficient evidence to support the routine use of other diagnostic tests to establish the diagnosis of ADHD; the strength of the evidence for this recommendation was strong. Laboratory tests including blood lead, thyroid hormone levels, brain imaging studies, electroencephalography, and continuance performance tests were not useful in the evaluation of a child with ADHD.
TREATHBNTS SUPPORTED BY EVIDENCE-BASED STUDIES
The treatment guideline comprises 5 specific recommendations based on a comprehensive review of evidence. Each recommendation and the rationale used by the committee to establish its importance is discussed.
Primary care clinicians should establish a management program that recognizes ADHD as a chronic condition.
There are 3 chronic conditions seen in high prevalence in primary care pediatrics practices: asthma; otitis media with effusion; and ADHD. Most work in primary care is consumed with acute illness and preventative health supervision visits. The management of a child with ADHD requires a different approach - a paradigm shift in thinking about clinical disorders in primary care practice. Pediatricians are familiar with the chronic disease model of care and will find it useful to adapt this model when caring for a child with ADHD.
The treatment guideline outlines die main principles of care for children with a chronic condition, all of which apply to children with ADHD:
* Providing parents and child with information about the condition;
* Updating and monitoring family knowledge and understanding on a periodic basis;
* Counseling about family response to the condition;
* Developmentally appropriate education of child about ADHD, with updates as the child grows;
* Availability to answer family questions;
* Ensuring coordination of health and other services;
* Helping families set specific goals in areas related to child's condition and its effects on daily activities;
* Linking families with other families with children who have a similar chronic condition as needed and when available.
The practical implications of this recommendation are found in the treatment guideline: "Studies of children and adults with several chronic conditions indicate better adherence to treatment plans, improved health and disease status measures, and higher levels of satisfaction in the context of a comprehensive treatment plan with specific goals, follow-up activities, and monitoring." 2
The treating clinician, parent, and child, in collaboration with school personnel, should specify appropriate target outcomes to guide management.
The development of a list of 3 to 6 target outcomes agreed on by the parents, child, and teacher can be of enormous help following the diagnosis of a child with ADHD. Target outcomes bring precision to the discussion of treatment goals with parents and child, and direction for planning the follow-up process of a child with a chronic condition. Target outcomes should reflect key symptoms and the specific impairments these symptoms cause. The goals should be realistic, attainable, and measurable.
In the child with ADHD, examples of target outcomes include:
* Improving verbal and written communication; completing homework, with emphasis on volume of work, efficiency, completions, and accuracy; increasing independence in self-care or homework; improving self-esteem; reducing the degree of supervision at school or in the community; improving social relations with adults, family or peers; enhanced safety in the community.
* Improving core symptoms of ADHD by increasing attention; decreasing hyperactivity; and curbing impulsivity.
* Reducing associated symptoms such as anxiety; depression; oppositional defiant behaviors; and conduct disturbance.
The clinician should recommend stimulant medication and/or behavior therapy as appropriate to improve target outcomes in children with ADHD.
The process of developing this recommendation was among the most instructive and humbling learning experiences since my pediatrie residency 30 years ago. Treatment recommendations reflect the precision required for valid studies, the importance of recognizing evidencedbased from non-evidence- based studies, and the challenges of synthesizing and interpreting a large body of evidence.
Two sources of literature reviews and a recently published large randomized, controlled study were used to support the recommendation that both stimulant medication and /or behavior therapy are effective in children with ADHD. The committee partnered with the Agency for Healthcare Research and Quality5 and the Evidence-based Practice Center at McMaster University, Hamilton, Ontario, Canada.7 In addition, the committee reviewed the comprehensive, multimodal treatment study of children with ADHD (A review of therapies for attentiondeficit /hyperactivity disorder) by the Canadian Coordination Office for Health Technology Assessment.8 Finally, the committee reviewed the findings of the recently completed multimodal treatment of children with ADHD (MTA), a 14month randomized controlled trial of treatment of school-aged children with ADHD.9
More than 150 randomized controlled clinical trials of school-aged children with ADHD support the benefit of stimulant medications (methylphenidate and dextroamphetamine). Documented benefits are seen not only in core ADHD symptoms (hyperactivity, impulsivity, and inattention), but in many cases improvement is seen in a child's ability to follow rules and improve relationships with peers and parents.
Although most of the medication studies were short term, the MTA study demonstrated a persistent effect of decreasing core ADHD symptoms for at least 14-months. Since the publication of the guideline, this positive result has been extended to a 24-month follow-up period.
Twenty-two head-to-head studies showed no differences comparing methylphenidate with dextroamphetamine (or different forms of these stimulants) in their effect on core ADHD behaviors. Approximately 70% of school-aged children with ADHD respond to either methylphenidate or dextroamphetamine. Among the 30% non-responders to maximum dose without side effects, about half will respond to the other stimulant. The use of stimulants does not require hematological, biochemical or electrocardiograph monitoring.
Current evidence supports the use of only 2 other medications for children with ADHD: tricyclic antidepressants and bupropion. However, there are significantly fewer randomized, controlled studies of these drugs compared to stimulants.
A practical guide to the initiation and maintenance of stimulant medications in children with ADHD, consistent with the recommendations of the AAP evidence-based treatment guideline, was published recently.10
The recommendation to include behavior therapy is based on empirical support of the efficacy of parent training in behavior therapy and teacher training in classroom interventions.11 The guideline carefully defines behavioral therapy as a program with "specific interventions that have a common goal of modifying the physical and social environment to alter or change behavior ... [it] involves providing rewards for demonstrating the desired behavior (eg, positive reinforcement) or consequences for failure to meet the goals (eg, punishment). Repetitive application of the rewards and consequences gradually shapes behavior."2
Pediatricians should differentiate behavior therapy from other forms of psychological interventions designed to change a child's emotional status such as cognitive therapy, psychotherapy, and play therapy. These interventions may be useful with some of the coexisting conditions found in some children with ADHD (eg, oppositional behavior, anxiety and depression), but they have not been shown to be effective with the core ADHD behaviors.
Well-designed studies that support die use of behavior therapy characteristically include 8 to 12 weekly group sessions with a trained therapist. With a focus on the child's behavior and difficulties in family relationships, these programs attempt to improve a parent's understanding of a child's behaviors and teach specific skills of behavior modification - techniques for giving commands, reinforcing adaptive and positive social behaviors, and decreasing or eliminating inappropriate behavior. Although the committee recognized that this kind of program is not available in many communities, it is the only format that has been studied and demonstrated to be effective. It is the gold standard for behavior modification programs.
Most pediatricians are neither trained to provide a formal behavior modification program for children with ADHD, nor have office personnel available to carry out such a program. When a referral is not available or when the target behaviors are not too severe, behavior modification can be initiated by pediatricians, an office nurse, or other office personnel with training and experience in behavior modification. The treatment guideline provides a template for effective behavioral techniques (Table 3).
For many years, pediatricians and other clinicians that treated children with ADHD recommended a multimodel approach, including medication, parent and child education about ADHD, behavior management, and classroom accommodations. The evidence-based review concluded that most studies comparing behavior therapy with stimulants alone show a much stronger effect from stimulants than from behavior therapy. The MTA study showed that combined treatment (medication and behavior management) had similar effects on core ADHD behaviors as seen in the children treated with medication alone.8 However, there were potential benefits in those in the combined group, including improved scores on some academic measurements, conduct, and anxiety symptoms. In addition, parents and teachers of those children who received the combined treatment were more satisfied with the treatment plan.1243 These benefits of combined therapy may be extraordinarily significant for some children with ADHD.
When the selected management for a child with ADHD has not met target outcomes, clinicians should evaluate the original diagnosis, use of all appropriate treatments, evaluate adherence to the treatment plan, and the presence of coexisting conditions.
Implied in this recommendation is the framework found in the first recommendation - that the management of a child with ADHD should be carried out in the context of chronic care. The guideline points out that continuing lack of response to treatment may be due to: unrealistic target symptoms that need révaluation; lack of information about the child's behavior that may require better communication with the school and /or the family; an incorrect diagnosis - always a possibility to consider when an evidenced-based treatment is ineffective; a coexisting condition affecting treatment - especially an undiagnosed anxiety disorder or depression or a learning disability that is not recognized or not receiving proper intervention; lack of adherence to the treatment regimen (this may require skilled interviewing in order to verify); or a true treatment failure which is defined in the guideline as: a) lack of response to 2 or 3 stimulant medications at maximum dose without side effects or at any dose with intolerable side effects; b) inability of behavioral therapy or combination therapy to control the child's behavior; or c) the interference of a coexisting condition. In these situations, most pediatricians should consider referral to a mental health specialist.
Effective Behavioral Techniques for Children With Attentlon-Deflclt/Hyperactlvlty Disorder*
The clinician should periodically provide a systematic follow-up for the child with ADHD. Monitoring should be directed to target outcomes and adverse effects by obtaining specific information from parent, teacher, and the child.
This may have been the most challenging recommendation for the committee in that there is not sufficient evidenced in the health services literature to guide pediatricians on critical issues such as the frequency of office visits, the most effective method of charting to ensure adherence, proven ways to communicate effectively with school personnel in order to monitor progress, and how to adapt a pediatrie office practice that is formulated to see acute sick visits and health supervision visits to the diagnosis and management of children with ADHD. Although the strength of the evidence was only fair, the following recommendations were made on the basis of experience and consensus:
* Periodic monitoring of a child with ADHD should be guided by a plan that includes obtaining information about target behaviors, educational output, and medication side effects. The frequency of monitoring through office visits and phone calls depends on the degree of dysfunction and adherence to the treatment plan. There are no controlled studies that document the appropriate frequency of follow-up visits. It is useful to separate the titration phase (the initial establishment of the most effective medication regimen and monitoring side effects) from the maintenance phase. The former period, which may last a few weeks or up to 1 or 2 months, usually requires frequent communication with parent, teacher, and the child.
* The medical record should be organized as a flow sheet or in progress notes to systematically record monitoring data, including phone calls and medication refills.
* A system for periodic communication with parents and teachers should be established at the initiation of a treatment plan.
* The monitoring plan should consider normal developmental changes in behavior over time, educational expectations that increase with each grade, and changes in the home and school environment.
The committee recognized that implementation of the diagnostic and treatment guidelines requires an intensive education effort through publications and continuing medical education programs. As an extension of the guideline development, the American Academy of Pediatrics is developing of a tool kit that will enable pediatricians to diagnose and manage children with ADHD efficiently and productively in a primary care office setting. The tool kit will include specific suggestions for the creative use of office personnel, office records, communication with school personnel, and monitoring effects of medication (and side effects), behavior management, and target behaviors. Examples of behavior monitoring forms, flow sheets and letters to schools will be included in the tool kit.
Pediatricians are encouraged to read the full text of the ADHD Diagnosis and Treatment Guidelines published in Pediatrics.1'2 Pediatricians, offices, and clinics should be able to adapt the guidelines to their individual practice while maintaining sufficient standardization to reverse contemporary variations in practice patterns.14 Group discussions among pediatrie colleagues and office personnel can serve as continuing education on a local level as well as a forum for sharing ideas and experiences in working with children with ADHD.
As a profession that strives to practice in the best interest of children, the scientific evidence for both the diagnosis and management of school-aged children with ADHD provides pediatricians with a solid foundation for quality care in this important area. The guidelines lay out the scientific principles for current state-ofthe-art practice.
1. American Academy of Pediatrics, committee on quality improvement and subcommittee on attentiondeficit /hyperactivity disorder. Diagnosis and evaluation of the child with attention-deficit/hyperactivity disorder. Pediatncs. 2000;105:1158-1170.
2. American Academy of Pediatrics, committee on quality improvement, subcommittee on attention-deficit /hyperactivity disorder, 2001. Clinical practice guideline: treatment of the school-age child with attention-deficit/hyperactivity disorder. Pediatrics. 2001;108:1033-1044.
3. Eddy DM. Clinical decision making: from theory to practice. Designing a practice policy. Standards, guidelines, and options. Journal of the American Medical Association. 1990263:3077-3084.
4. Diagnostic and Statistical Manual of Mental DisordersAth ed. Washington, DC: American Psychiatric Association; 1994.
5. Green M, Wong M, Atkins D, et al. Diagnosis of attentiondeficit/hyperactivity disorder: technical review No. 3 (AHCPR publication 99-0050). Rockville, MD: US Dept of Health and Human Services, Agency for Health Care Policy and Research; 1999.
6. Lindsay RL. School failure /disorders of learning. In: Bergman AB, ed. 20 Common Problems in Pediatrics. New York, NY: McGraw HUl; 2001:319-336.
7. Jadad AR, Boyle M, Cunningham C, et al. Treatment of attention deficit/hyperactivity disorder. Evidence report/technology assessment No. 11 (AHRQ publication 00-E005). Rockville, MD: US Dept of Health and Human Services, Agency for Healthcare Research and Quality; 1999.
8. Miller A, Lee S, Raina P, et al. A review of therapies Jor attention-deficit/hyperactivity disorder. Ottawa, Ontario: Canadian Coordinating Office for Health Technology Assessment; 1998.
9. A 14-month randomized clinical trial of treatment strategies for attention deficit hyperactivity disorder. The MTA Cooperative Group. Multimodal Treatment Study of Children with ADHD. Arch Gen Psychiatry. 1999^6:1073-1086.
10. Wender EH. Managing stimulant medication for attention-deficit /hyperactivity disorder. Pediatr Rev. 2001;22:183-190.
11. Pelham WE, Wheeler T, Chronis A. Empirically supported psychosocial treatments for attention deficit hyperactivity disorder. J Clin Child Psyckol. 1998;27:190-205.
12. Conners CK, Epstein JN, March JS, Angold A, Wells KC, Klaric J, et al. Multimodal treatment of ADHD (MTA): an alternative outcome analysis. J Am Acad Child Adolesc Psychiatry. 2000;40:159-167.
13. Wells KC, Epstein JN, Hinshaw SP, Conners CK, Klaric J, Abikoff HB, et al. Parenting and family stress treatment outcomes in attention deficit hyperactivity disorder (ADHD): an empirical analysis in the MTS study. J Abnorm Child Psychol. 2000;28:543-553.
14. Reiff MI. Attention-deficit /hyperactivity disorders. In: Bergman AB, ed. 20 Common Problems in Pediatrics. New York, NY: McGraw HUl; 2001:265-300.
DSM-IV Criteria for Attentlon-Deflclt/Hyperactlvlty Disorder*
Prevalence of Selected Coexisting Conditions In Children With AttentlonDeflclt/Hyperactlvlty Disorder
Effective Behavioral Techniques for Children With Attentlon-Deflclt/Hyperactlvlty Disorder*