$$XID RFA MH9203 MH-92-03 P1O1 ***************************************** COOPERATIVE AGREEMENT FOR A MULTI-SITE TREATMENT STUDY OF ATTENTION- DEFICIT HYPERACTIVITY DISORDER/ATTENTION DEFICIT DISORDER NIH GUIDE, Volume 21, Number 9, March 6, 1992 RFA: MH-92-03 P.T. 34, AA; K.W. 0404004, 0720010, 0414014 National Institute of Mental Health Application Receipt Date: May 19, 1992 PURPOSE The National Institute of Mental Health (NIMH) announces a Request for Applications (RFA) for "Multimodal Treatment Study of Attention-Deficit Hyperactivity Disorder and Attention Deficit Disorder (ADHD/ADD)." This program is a five-year multi-site collaborative study intended to address a range of issues concerning ADHD as emphasized by the Institute of Medicine study "Research on Children and Adolescents with Mental, Behavioral, and Developmental Disorders," the NIMH "National Plan for Research on Child and Adolescent Mental Disorders," the "Healthy People 2000" objectives in the area of mental disorders in children and adolescents (6.3), and the existing literature. Foremost among these are treatment issues subsumed by the question: Under what circumstances and patient characteristics (e.g., comorbid conditions, family socioeconomic background, metabolic/nutritional status, physical/neurological findings, neurophysiological/neuropsychological status, gender, age) do which treatment combinations (specific medication, behavior therapy, parent training, school-based intervention), have what impacts (improvement, stasis, deterioration) on what domains of child functioning (cognitive, academic, behavioral, neurophysiological, neuropsychological, peer relations, family relations), for how long (short vs. long term), to what extent (effect sizes, normal vs. pathological range), and why (processes underlying change)? These questions form the scaffolding for a multi-site, multimodal treatment study of ADHD/ADD and its comorbid conditions. During the first year, the Principal Investigators (PIs) of the collaborating sites will develop a common protocol from their pooled applications. This common protocol will be designed to (1) maximize the potential of the cross-site data set to address the treatment questions outlined above and (2) support major studies focusing on related issues of the assessment, comorbidity, etiology, and natural history of ADHD with its comorbid conditions. In year five, data will be analyzed, scientific reports will be prepared for publication as warranted by initial study findings, and the scientific merit of examining long-term outcomes will be evaluated. HEALTHY PEOPLE 2000 The Public Health Service (PHS) is committed to achieving the health promotion and disease prevention objectives of "Healthy People 2000," a PHS-led national activity for setting priority objectives. Of the 300 such objectives, 170 concerning children and adolescents have been collected into "Healthy Children 2000." This RFA, Cooperative Agreement for a Multi-site Treatment Study of ADHD/ADD, is related to the priority area of research on appropriate biomedical and psychosocial interventions. A copy of "Healthy People: 2000" (Full Report: Stock Number 017-001-00474-0, or the Summary Report: Stock Number 017-001-00473-1) may be obtained from the Superintendent of Documents, Government Printing Office, Washington, DC 20402-9325; telephone 202-783-3238. ELIGIBILITY REQUIREMENTS Applications may be submitted by domestic and foreign, public and private, non-profit and for-profit organizations such as universities, colleges, hospitals, laboratories, units of State and local governments, and eligible agencies of the Federal Government. Applications are encouraged from women and minority investigators, who are still underrepresented in mental health research. MECHANISM OF SUPPORT The cooperative agreement (U01) mechanism used to support this program differs from investigator-initiated research grants. Although awardees will be primarily responsible for the conduct of the study, there will be (1) collaboration among the participating sites and (2) programmatic involvement of NIMH staff above and beyond the levels required for the traditional program management of grants. Applicants are expected to recommend a design with specific methods, instruments, psychosocial treatments, and psychopharmacological protocol and, during year 1, will negotiate a common design/protocol through the steering committee mechanism explained below under "Study Organization and Oversight." Applicants should realize that the final actual common protocol is likely to be different from any one submitted protocol; even though a grant is awarded partly on the basis of a proposed protocol, the protocol proposed by an awardee is not likely to be the final one collegially decided upon by the steering committee. Period of Support Applications may request support for a period of up to five years. It is anticipated that NIMH will accept competing continuation applications after the initial award period. Application Receipt and Review Schedule To qualify for funding in Fiscal Year 1992, applications must meet the following one-time only special schedule. Initial Advisory Funding Receipt Date Review Council Review Date May 19, 1992 July 1992 Sept 1992 Sept 1992 Applications received after the receipt date will be returned to the applicant without review. FUNDS AVAILABLE It is anticipated that up to $2.5 million per year will be available to support up to six new cooperative agreement awards under this RFA during fiscal year 1992, subject to the availability of funds. Funding in future years will depend upon annual appropriations. RESEARCH OBJECTIVES Background Clinical Features of the Syndrome: ADHD/ADD is among the most common of childhood Axis I mental disorders. It has been estimated to afflict between two percent and nine percent of children nationwide, accounts for one-third to one-half of all referrals for child mental health services, and comprises a large proportion of economic cost and human suffering caused by childhood mental disorders. The core clinical features of ADHD, many of which can be detected as early as three years of age and persist through the school years, include inappropriate activity levels, low frustration tolerance, impulsivity, poor organization of behavior, distractibility, and especially inability to sustain attention and concentration. There has been considerable debate over the years about the appropriate definitional boundaries of hyperactivity and about the scientific legitimacy of its status as a distinct clinical syndrome. There has never been controversy, however, about whether or not a significant number of children suffer from the core clinical symptoms described above or about the social and academic impairments and psychiatric comorbidities described below. Associated Functioning Deficits: Tragically, the core clinical symptoms of ADHD reflect impairments in precisely the domains of functioning that are central to mastery of virtually all of the major developmental milestones of childhood. It is, therefore, not surprising that a majority of diagnosed children tend also to perform poorly in school, despite normal intelligence, and suffer significant social and emotional impairments in the formation and maintenance of relationships with classmates, peers, parents, and teachers. Comorbidity: It has been known for some time that ADHD is characteristically comorbid with other childhood mental disorders, especially conduct and oppositional defiant disorders. More recently, investigators have examined the comorbidity of ADHD with mood and anxiety disorders and learning problems. A review of clinical studies suggests substantial comorbid prevalence for conduct/oppositional- defiant disorders (30-50 percent), mood disorders (widely disparate reports averaging about one-third), anxiety disorders (about 25 percent), and learning disorders (20-25 percent). Although the definitive community epidemiological research on child and adolescent mental disorders has not yet been completed in the United States, a large community-based New Zealand study reported that 47 percent of children with ADHD diagnoses had a coexisting conduct or oppositional disorder, while 26 percent had a coexisting anxiety/phobic disorder, and 18 percent had two or more comorbid conditions. Similar patterns of comorbidity with ADHD have been reported from epidemiological studies in Puerto Rico and Canada. These comorbid conditions and associated social and academic impairments provide evidence of the heterogenous nature of the disorder, add to its clinical complexity, and have significant implications for its etiology, course, and treatment. Long-Term Prognosis: Although ADHD is classified as a childhood disorder and is typically identified in the early school years, it has been estimated that 70 percent of afflicted children continue to experience the full syndrome in adolescence. Children diagnosed in childhood have been found as adolescents to suffer continuing overactivity, poor school performance, and significant home, school, and community behavior problems, such as temper tantrums, defiance, police contacts, and peer rejection. Moreover, it has been estimated that as many as two-thirds of hyperactive adolescents suffer serious discipline problems at school, with high rates of suspension and expulsion and chronically low levels of self-esteem. Although few long-term follow-up studies into adulthood have been reported, those available converge on a portrait of continuing deficits in many domains of functioning. Compared to matched normal controls, young hyperactive adults have been shown to suffer significantly higher levels of impulsivity and restlessness, nonmedical drug use, court referrals, incarceration, and personality disorders. At the diagnostic level, follow-up studies of hyperactive children into young adulthood suggest that approximately half continue to have mental disorders, including ADD, antisocial disorder, and drug use disorder. Treatments Stimulant Treatment: It has been estimated that between two and three percent of all elementary school children in North America receive some form of pharmacological intervention for hyperactivity. Although controlled studies support the efficacy of a variety of medications (including antidepressants, clonidine, and neuroleptics) for hyperactivity, by far the most widely prescribed and thoroughly studied have been the psychostimulants, especially dextroamphetamine and methylphenidate. Since these are generally agreed to constitute the "first-line" psychopharmacology for ADHD, it is anticipated that the multi-site study will focus on them. Antidepressants are generally acknowledged as an established second-choice category, and have even been advocated by a few authors as first-choice drugs. Unfortunately, serious safety concerns have been reported for some antidepressants. Further, the antidepressants appear less predictable and specific and in the estimation of many psychopharmacologists probably less efficacious than the stimulants for the majority of ADHD children. The popularity of stimulant drugs stems from their demonstrated short-term efficacy, compared to placebo conditions, in dramatically reducing a range of core hyperactivity symptoms such as task-irrelevant activity and classroom disturbance, with associated increases in compliance and sustained attention. Positive effects of stimulants have also been shown on parent-child interactions, problem-solving activities with peers, aggressive behavior, and performance in a variety of controlled laboratory tasks, including paired-associate learning, cued and free recall, auditory and reading comprehension, and arithmetic computation. But these stimulant effects have been less reliable in bringing about associated improvements, at least of an enduring nature, in social-emotional and academic problems, such as antisocial behavior, poor peer and teacher relationships, and school failure. Several factors may contribute to limitations in documented stimulant efficacy, all of which have implications for subsequent treatment studies: o First, many tests of the effects of stimulant medication have been short-term studies lasting only a few weeks or months. Many of the social and academic impairments associated with hyperactivity, however, may require significantly longer periods for gradual improvements to materialize. The long-term effects of stimulants remain in doubt. o Second, some hyperactive children may not respond favorably to stimulants even in the short run. Earlier reports estimated the prevalence of non-response to range from 10 to 40 percent, depending on populations studied, criteria for clinical improvement, and whether more than one stimulant was tried. A recent controlled study concluded that well over 90 percent of hyperactive children respond favorably to either methylphenidate or dextroamphetamine if both are tried and if dosages are carefully titrated; the most common type of "nonresponse" was intolerable side effects. o Third, some evidence has suggested a possible dose-response interaction with domain of functioning. An example is the claim that the higher dosages necessary for maximal effects on teacher-rated classroom comportment may impair rather than improve learning abilities in certain hyperactive children and that lower stimulant doses may be necessary for learning improvements in these children. Though the hypothesis of differing drug-response thresholds for learning and behavior is controversial, in more general terms the unexamined dose-response relationships may obscure important individual differences across children that have implications for stimulant treatment effectiveness and interactions with psychosocial treatments. o Fourth, the possibility of state-dependent learning remains a nagging question. Could some children's performance gains resulting from stimulants fail to carry over effectively to the unmedicated state? Though the preponderance of evidence suggests that state-dependent learning is not a problem for low doses of stimulants, this is a complex question related to the dose-response issue. o Fifth, the magnitude of stimulant benefit may not be consistent across age groups or mental age/IQ groups. o Sixth, the high levels of comorbidity characteristically associated with ADHD may have important implications for differential stimulant effectiveness in subgroups of hyperactive children. For example, one study found a significant interaction of methylphenidate effect with comorbid anxiety and a trend of greater placebo response in the anxious hyperactive subjects than in the nonanxious. o Seventh is the possibility of a differential effect by comorbidity and/or domain of function between the two major stimulants. Since there is little overlap in the reported nonresponse rate, it appears that different patient characteristics determine nonresponse for methylphenidate than for d-amphetamine. There is reason to suspect that one such characteristic is comorbidity. Furthermore, the differential general efficacy (regardless of comorbidity) of the two major stimulants remains in question. The characterization of methylphenidate as the "drug of choice" may not be warranted since the available published studies that directly compared it to dextroamphetamine in the same subjects failed to show an advantage for methylphenidate, although individual children often responded better to one stimulant than the other. o Eighth is the concern that for some children stimulant benefit may possibly be offset by maladaptive psychological attributions of failure and success. The limited controlled data available suggest that while most preadolescent medicated children attribute success to their own efforts and failure to the pill (a normal, adaptive, self-enhancing attributional style), a subgroup seems to attribute success to the pill and failure to lack of ability, leading to a cessation of effort/motivation. o Ninth, the possibility of a linkage between differential drug response and minor physical anomalies, neurological "soft signs," neurophysiological aberrations, or metabolic/ nutritional status has not been adequately explored. o Finally, a study of stimulant effects needs to consider the issue of prior medication. At any given age, a history of earlier medication suggests more serious symptomatology, earlier manifestation, and/or socioeconomic and familial variables such as parental concern/attitudes and access to health care. Therefore, merely excluding subjects with a prior medication history would bias the sample. If a withdrawal period from prior medication is too short, carryover effects could confound the baseline measures. If too long, sample attrition may bias the sample against the more serious or more drug-responsive cases. The nature of dependent variables is an important consideration: While behavior and cognition usually revert to baseline within a week after stimulant withdrawal, some stimulant-induced biochemical changes have been reported not to wash out for over two weeks. Should only drug-naive subjects have the affected biochemical tests, or should the tests be confined to those tests presumed unaffected by drug carryover, or should the washout period be prolonged enough to collect all desired data on every subject? Further, while most ADHD patients are medicated at some point, the majority do not continue medication after several years. Previous studies have generally not addressed these issues. There is an emerging consensus concerning an important yet limited role for stimulant medication in the treatment of hyperactivity. It may be the necessary and sufficient treatment for a subset of ADHD children. For others, it may be contraindicated. For still others, stimulant medication in isolation should not be expected to yield gains beyond its immediate effects on attention, impulsivity, and activity levels. It may be most effective in normalizing and stabilizing the primary functioning characteristics of these ADHD children, whose behavior and learning problems must then be addressed through a range of psychosocial treatments. Psychosocial Treatments: Many of the research results and considerations raised above have given rise to a newer phase of research focused on the use of psychosocial treatment modalities, alone and in combination with stimulants. Psychosocial interventions that have been systematically explored include classroom-based behavior modification, social skills and cognitive training, and parent training/home-based interventions. Controlled studies of stimulant medication, psychosocial treatments, and their combination have suggested that combined approaches may yield more favorable results than single treatment modalities. Some studies have demonstrated that token reinforcement systems may "normalize" aggressive and other off-task behaviors in the classroom. A number of cognitive-behavioral interventions have been shown to produce both increased self-control and the use of specific coping strategies in hyperactive children; in some cases, these effects were neither enhanced by the addition of, nor produced independently by, stimulant medication. Other child-focused interventions, such as self-control procedures, have produced desired treatment effects in experimental classrooms, but a critical problem has been the lack of evidence that such interventions generalize to other settings (e.g., regular school classrooms) or across behavioral domains. And still other interventions, such as interpersonal problem-solving skills therapy, have failed to facilitate interpersonal competence in either medicated or unmedicated children. Similarly, cognitive therapy and social skills training have produced only weak and variable effects, with little evidence to date that they have a significant impact on the academic performance or social behavior of hyperactive children. Parent training in child behavior modification has been shown to improve both the school and home behavior of hyperactive children; however, only with medication were there also reductions of impulsivity and inattention. Home-based treatments with parental involvement, coordinated with school interventions, are thought by many workers in this field to be an essential component of treatment effectiveness; they may increase the salience of school interventions for many children and facilitate the generalization of treatment effects across settings and behavioral domains. While there is promising evidence for the clinical utility of a variety of psychosocial interventions in the treatment of hyperactive children, most psychosocial treatment studies have not attended to important individual differences in the comorbidities and functioning impairments of hyperactive children. Multimodal Treatments: It is clear from both the well-developed literature on stimulant medication and the developing literature on psychosocial treatments that no single treatment in isolation is likely to yield clinically significant, long-term, cross-domain therapeutic gains in an unselected, and, therefore, heterogeneous group of hyperactive children. Within any such group, there are likely to be subgroups with significantly different patterns of comorbidity, family backgrounds, and functioning deficits -- all of which have important implications for children's treatment needs. These considerations have given rise to an interest in multimodal treatment strategies. More closely approximating the ideal of sound clinical practice, one multimodal strategy differs from traditional research approaches by tailoring stimulant medication and/or psychosocial interventions to the particular needs of individual ADHD children and their families. A prototypical multimodal approach begins with an intensive evaluation of each child's impairments and strengths in each domain of functioning, as well as those of his/her family. On the basis of these evaluations, specific treatment plans are developed for each child's needs and are administered through an intensive, coordinated team approach. Multimodal treatment strategies may hold considerable promise for matching combined treatments to the needs of ADHD children and their families. Although impressive early findings have been reported from studies of multimodal treatments, enthusiasm has been tempered by concerns over possible recruitment bias, the absence of no-treatment control groups, failure to use blinded assessment procedures, and the uncontrolled assignment of subjects to treatment conditions. Studies to date have experienced significant limitations of sample size and methodology, precluding meaningful comparisons of treatment combinations and interactions among treatment combinations, comorbidity patterns, and child/family characteristics. Other Treatments: Numerous other treatments (e.g., elimination diets, nutritional supplementation, perceptual stimulation/training, sensorimotor integration, biofeedback, progressive relaxation, hypnosis, detoxification, antifungal therapy) have been advocated/tried. It is not anticipated that the multi-site study will focus on these because they either have insufficient controlled research evidence of efficacy or seem to benefit only a small proportion of ADHD children. However, baseline measures relevant to such treatments could be considered for inclusion in the common protocol. The Need for a Multi-Site Study under a Cooperative Agreement Considerable research into treatment strategies will be needed to establish answers to the manifold question posed earlier: Under what circumstances (e.g., comorbid conditions, family socioeconomic background, age, gender, metabolic/nutritional status, physical findings, neurophysiological/neuropsychological status) do which treatment combinations (specific medication, behavior therapy, parent training, school-based intervention) have what impacts (improvement, stasis, deterioration) on what domains of child functioning (cognitive, academic, behavioral, physical, peer relations, family relations), for how long (short vs. long term), to what extent (effect sizes, normal vs. pathological range), and why (processes underlying change)? The very nature of a multimodal treatment approach is such that answers to these questions will require adequate numbers of ADHD subjects with varying comorbid profiles, functioning deficits, physical/physiological status, and family characteristics. This will require coordination of multiple sites in a unified investigation. Such coordination is best done through a cooperative agreement, which invokes substantial involvement of NIMH collaborators. Accordingly, NIMH staff will have substantial involvement in the collegial negotiation of a common protocol, the assuring of uniform cross-site training, the development of monitoring systems to ensure cross-site fidelity to the protocol, the development of centralized data management and analysis, and the enhancing/facilitating of communication and cooperation among the sites. Program Description Goals/Objectives and Research Questions. The goal of this program is to implement a five-year multi-site collaborative treatment study of ADHD/ADD and its associated comorbidities and social/emotional and academic impairments. Examples of appropriate hypotheses that might be tested are stated below in question form. The main questions are expected to concern synergism of stimulant and psychosocial treatment, interaction of treatment type with comorbidity pattern and socioeconomic status (SES), differential stimulant efficacy, and differential efficacy of psychosocial treatment prescriptions: o Differential Effects of Stimulant Treatment and Interactions with Comorbidity and SES: How do comorbidity and socioeconomic status affect drug response including placebo response? Is there a difference in efficacy between stimulants? Do they have a differential affect by comorbidity (i.e., interaction of stimulant type with comorbidity)? Do the stimulants differentially affect different domains of functioning (i.e., interaction of stimulant type and domain of function)? o Differential Effects of Psychosocial Treatments and Interactions with Comorbidity and SES: Is tailored multicomponent psychosocial treatment better than packaged (fixed battery) psychosocial treatment of the same intensity? Are multicomponent psychosocial treatments better than single-modality? Do comorbidity and SES affect treatment response? Do comorbidity and SES interact with the type of treatment? o Additive/Synergistic Effects: How much do psychosocial treatments add to stimulant benefit? Does continuing stimulant medication add to the efficacy of ongoing long-term psychosocial treatments? Does psychosocial treatment allow more successful weaning from successful stimulant treatment? If so, is this additive/synergistic effect affected by comorbidity and socioeconomic status? o Is there a differential treatment response by gender? Does gender interact with SES or comorbidity in treatment response? o How does a comorbid learning disorder affect treatment response? Does learning disorder interact with Axis I comorbidity and/or SES? o Is the distinction "with" or "without" hyperactivity associated with different comorbid patterns and/or differential treatment response? o Are baseline metabolic/nutritional status (e.g., minerals, heavy metals, dietary habits), physical/neurological examination findings (e.g., "soft signs" and minor anomalies), or neuropsychological findings associated with differential treatment response? o Is each treatment, systematically closely monitored, truly better than no contact? Is the superiority of treatment over no contact affected by comorbidity and SES? o Does parental restriction of treatment options (by refusal of drugs or psychosocial treatment) affect treatment outcome? Does parental choice/restriction of treatment interact with comorbidity and SES? Is it only the less seriously disordered children who afford their parents the luxury of refusing treatment options? How does the parental decision to obtain treatment (presumably motivated by severity of the child's symptoms) bias sample composition (e.g., comorbidity) and prognosis? Program Schedule. Following the selection of participating sites, the PIs participating in the cooperative agreement will meet with participating NIMH staff to establish appropriate levels of coordination and refine the governing structure (see "Study Organization and Oversight" below). The first year of the collaborative effort will be devoted mainly to developing a common protocol from the selected applications, developing the necessary training and implementation procedures, finalizing manuals to ensure cross-site consistency in study execution, and hiring and training assistants/therapists. The common protocol will be designed to (1) maximize the potential of the cross-site data set to address the manifold treatment questions, and (2) support major studies focusing on related issues of the assessment, comorbidity, etiology, validity, and natural history of ADHD with its comorbid conditions. Domains of assessment should include formal psychiatric assessments of probands and parents, and assessments of all probands in the domains of neurological, intellectual, cognitive, academic, behavioral/ psychosocial, and metabolic/nutritional/physiologic functioning. Project years two through four will be devoted to the implementation of the protocol developed during the first project year. Most of the subjects should be entered into the study during year two to allow at least two years of treatment and follow up. Year five of the project will be used mainly to analyze the results, prepare scientific reports for publication, and prepare data tapes for the public domain. Methodological Considerations Applicants are expected to propose resolutions of the complex design, recruitment, sampling, assessment, and treatment issues involved, including: o Design. Applicants should consider the relative advantages and disadvantages of parallel group vs. within-subject crossover designs and combinations/hybrids (e.g., one imbedded in the other or one preceding the other) for the various hypotheses under study. Which design best suits any intended pharmacological comparison? Which best suit the psychosocial treatment comparison? What is an optimal compromise? If a crossover is proposed, how will the issue of carryover effects be resolved? Should some aspects of the study be "unbundled" into one or more preliminary brief studies before longer- term treatment? If so, what is an appropriate duration for such preliminary treatment trials? What design will facilitate the optimal individualized drug treatment being compared with various psychosocial treatments, alone and in combination? Multimodal treatment studies typically suffer from severe sampling bias, partly because of the large proportion of subjects who decline one treatment but would be willing to participate in the other. Nonetheless, it is important to study how these children differ from full consenters in treatment response and perhaps in other ways. Should consent be a sequential process with options for partial consent, or should consent be global and obtained on an "all-or-none" basis? What design will ensure maximal sample retention of no-contact controls and/or those who consent to one treatment modality (drug or psychosocial) but not the other, as well as those who give full consent for random treatment assignment? If an age range wider than two to three years is sampled, how will the design control for age in treatment response and participation/compliance? If mentally retarded subjects are included, how will the design control for mental age? How many variables can be managed with adequate stratification at the site level without compromising cell size? o Subject Recruitment and Selection. Applicants must describe the anticipated demographics of the proposed sample and discuss its population representativeness. One consideration in the selection of applicant sites will be overall cultural, socioeconomic, and racial/ethnic representativeness of the total sample. Although it would not be possible for each site to recruit a nationally representative sample, each site's sample should at least represent a broad cross-section of the ADHD/ADD population in its metropolitan area or region. Centers in geographic proximity (within 1-2 hours travel) are encouraged to form a consortium and apply as one site for the purpose of broadening sample representativeness and accessing adequate subject numbers. If such a consortium is formed, arrangements should be documented and a single principal investigator must be agreed on. Issues of volunteer bias, referral bias (e.g., clinical samples versus school-identified samples, whose condition may not be as significant or severe as those who seek treatment), and subject attrition must be addressed. Subjects should be recruited with careful attention to the range of comorbidity, especially such Axis I comorbidity as other disruptive behavior disorders (conduct and oppositional-defiant disorders), depression, and anxiety. Applicants must demonstrate access to an adequate pool of eligible subjects from multiple sources. In view of the anticipated program schedule, the need for over 700 total subjects (suggested by preliminary power analyses), and the expectation of funding four to six sites, approximately 120-180 subjects will need to begin the study at each site within a 15-month period (in order to allow at least two years of treatment follow-up); applicants must state how many subjects they can reasonably access in what time frame. Estimates must be included for the proportion of subjects who are expected to consent to each proposed treatment condition. Estimates must also be included for the proportions of subjects expected to have each comorbid condition/combination. Exclusion/inclusion criteria for subjects must be well articulated. It is anticipated that all subjects will be in elementary school. o Sample Retention. Applicants must propose and justify strategies and procedures (e.g., sequential consent process for assessment, stimulant treatment, and psychosocial treatments) to facilitate retention of subjects who decline one or more treatment modalities and to study their response to any treatment for which they do consent. o Diagnosis and Assessment Across Multiple Domains Using Multiple Information Sources. Significant concerns have been raised among hyperactivity researchers about the validity of the DSM-III-R criteria for ADHD and many workers in this area have noted that the empirical evidence most closely fits the original DSM-III criteria. The advent of DSM-IV will bring a number of changes to the ADHD diagnostic criteria that will more closely approximate original DSM-III criteria. Each application must discuss and justify the proposed choice of diagnostic criteria (DSM-III, DSM-III-R, DSM-IV, or combinations or alternatives). Applicants must specify validated diagnostic interviews and other procedures and instruments for collecting symptom information from parents, children, and their teachers. Criteria for combining information from multiple sources must be specified. The method for psychiatric assessment of parent as well as child must be specified. o Withdrawal Period from Previous Medication. An appropriate preliminary withdrawal period from any previous psychopharmacotherapy must be proposed and justified in view of washout times of various drugs in common use. Applicants must specify how the history of prior treatment will be handled in the overall design. o Treatments. Given the emphasis of this study on the differential effectiveness of various treatments by comorbidity, SES, and sex, applicants should consider inclusion of at least two drugs and three psychosocial treatments. These are not required to be compared simultaneously: for example, the drugs could be compared in a preliminary within-subjects crossover before assignment to psychosocial treatment conditions with the best-response maintenance drug. It is anticipated that medication alternatives will include methylphenidate and dextroamphetamine because the vast majority of hyperactive children will respond to at least one of these when properly titrated. Applicants will be expected to recommend specific drugs, dosages, procedures for administration, management, and verification of medication dose and compliance. The choice of multiple fixed doses vs. clinical titration and the choice of sustained-release vs. regular tablets must be discussed and justified. Applicants must specify criteria for determining inadequate or negative drug response based on standardized assessment instruments and clinical procedures/techniques. Side effects in both responders and nonresponders should receive special attention. The types and formats of psychosocial treatments must be specified, including a child-directed treatment, a parent-directed treatment, and a school-oriented intervention. The issue of relative efficacy of tailored vs. packaged/fixed battery of the same intensity must be addressed. The procedure/algorithm for individually prescribing tailored psychosocial treatment must be specified. Procedures for insuring uniform administration of psychosocial and drug treatments across and within sites must be proposed. o Randomization/Stratification and Blinding Procedures. Strategies, techniques, and procedures for randomization and/or stratification of subjects should be proposed and justified. Since not all variables can be stratified, applicants should justify their choices and priorities. Which variables are best managed by inclusionary/exclusionary criteria? Which ones are best left to chance without stratification? Applicants must propose, justify, and discuss alternatives for enhancing the blindness of evaluations/ratings, considering such issues as parents' detailed knowledge of treatments, teachers' knowledge of school- oriented intervention, the possibility of keeping the medication- managing psychiatrist blind, and the videotaping of child behaviors for blind rater assessments. For psychopharmacologic components/phases, standard psychopharmacologic blinding procedures should be specified. o Data Management and Analysis. Although management and analysis of the pooled data will be centralized at NIMH, applicants must propose an optimal plan for this, and describe plans and resources for site-level data management. Key issues to be addressed include: statistical handling of dropouts during treatment; differences among drugs; differences among psychosocial treatment conditions; synergism or additive effect of drug and psychosocial treatments; effect of patient characteristics (e.g., comorbidity); interactions. SPECIAL REQUIREMENTS Study Organization and Oversight Although awardees will be primarily responsible for the conduct of the study, there will be (1) collaboration among the participating sites and (2) substantial programmatic involvement and participation of NIMH staff above and beyond the levels required for the traditional program administration of individual research grants (R01). Applicants are expected to recommend a design and specific methods, instruments, and psychosocial treatments while recognizing that during year one, awardees, in conjunction with NIMH collaborators, will negotiate a common protocol. Collaborating Site Research Teams. Each research site funded under this program will be a collaborative research team. A consortium formed to ensure sufficient subjects and sample representation will be considered one site and must have one designated PI. It is anticipated that the staff needs at each site will include, but not be limited to, the following personnel: o Principal Investigator (PI). The PI will have overall responsibility for the research program at his/her site. Also, the PI will be responsible for integrating the site's program goals and procedures with those recommended for use by the collaborative program. The PI represents the site in collaborative decision-making on the overall scientific steering committee and is responsible for maintaining communication with NIMH staff. The PI must be able and prepared to devote a substantial, significant percentage of time and effort to this project, including extensive travel, especially the first year, for steering committee meetings and cross-site training and fidelity monitoring. o Co-Investigator(s) (Co-I). At each site the Co-I takes a lead role second only to that of the PI. The Co-I shares all responsibilities (outlined above) with the PI for the research program at his/her site. In a consortium arrangement, it might be desirable for the Co-I to be at a center/institution other than the PI's, or even to have multiple Co-Is, one taking the lead at each center. Multidisciplinary expertise is essential for the conduct of this complex study, including strong clinical expertise, extensive knowledge of diagnostic and nosologic issues, child development, psychopharmacology, psychosocial treatments, assessment and neuropsychology, statistics, epidemiology, and data management. Expertise in evaluating psychosocial treatments is important, preferably with experience in ensuring fidelity of treatment across settings. Therapists must be trained and receive sufficient monitoring and supervision to ensure fidelity to the treatment protocol. The research team at each participating site should show evidence of appropriate depth and breadth to accomplish the objectives of the collaborative program. To facilitate the multidisciplinary breadth of the team, it might be helpful in some instances to have the PI and Co-I (as defined above) from different disciplines. Role of NIMH Staff. The participation of NIMH staff in this cooperative agreement will ensure the broader scientific and public health goals of the collaborative agreement and bring a stabilizing influence to the collaborative process. In addition, the participating NIMH staff members, associated with the Child and Adolescent Disorders Research Branch and other branches of the Division of Clinical Research, will bring to the collaboration a wide range of scientific and clinical expertise in the areas of assessment, taxonomy, and diagnosis, and in the treatment of hyperactivity and related childhood conditions. At least one NIMH staff member will participate in all project meetings and committees. Specifically, NIMH collaborators will participate in development of the study plan, facilitation of cross-site training, quality control, coordination of the project across sites, data analysis and interpretation, and preparation of publications, but will not participate in activities directly involving human subjects. NIMH staff who have significant involvement in design/protocol development and/or data analysis/interpretation may cooperate with awardees as coauthors of resulting publications. In this regard they will be subject to the publication/authorship policies governing all participants. In addition, publications involving NIMH staff require internal NIMH clearance. Steering Committee. The steering committee will be the primary decision-making body of this collaborative multi- site study. Several NIMH staff will be members of the steering committee but cannot hold the position of chair. Membership will be composed of: 1. Chief, Child & Adolescent Disorders Research Branch, Division of Clinical Research, NIMH 2. Other NIMH collaborators to be designated by the Division of Clinical Research, NIMH 3. Principal Investigator (PI) from each site 4. One Co-Investigator from each site 5. A statistical consultant designated by NIMH 6. Other consultants as determined by the committee Voting and Governance: NIMH will have one vote on the steering committee regardless of the number of representatives present. Each site will have one vote regardless of whether one or both representatives are present. A quorum will require at least one NIMH representative and at least one representative from at least three-fourths of participating sites. Membership on the steering committee becomes effective for sites upon receipt of notice of award. All decisions will be made by majority vote of a quorum, with an attempt for consensus when possible. Since the use of a common standardized protocol is essential, all participating sites must agree to carry out the study design, procedures, and policies developed by this committee. Any committee member who considers a committee decision unacceptable may appeal to the arbitration procedure described below. The committee is expected to meet very frequently during the first year to develop, standardize, and coordinate the common protocol. In subsequent years the meeting frequency will probably decrease. The meetings will usually be held in the Washington, DC metropolitan area. The steering committee may establish additional by-laws and/or subcommittees for specific tasks, such as a data management committee to coordinate data coding, entry, and analysis, and an implementation committee to ensure uniformity of procedures across sites. Each such subcommittee will include at least one representative from each site and NIMH. No committee/subcommittee may be chaired by NIMH staff. Responsibilities of steering committee members include: 1. Collaboratively finalizing the study plan, including design, assessment instruments, component protocols, and detailed implementation procedures 2. Abiding by and directing the study design determined by the steering committee 3. Monitoring the study 4. Facilitating the pooling of data for common analysis and for eventual release to the larger scientific community (see "Public Domain" below) 5. Reporting study results 6. Abiding by all scientific, practical, and policy decisions of the steering committee and the provisions and intent of this RFA Management and analysis of the pooled data will be centralized at NIMH under procedures approved by the steering committee. The steering committee will develop policies on publication and authorship. Publication policies will be written and authorship will be decided using procedures developed by the steering committee. PIs may publish results from a single site but must obtain prior approval from the steering committee. The quality of publications will be the responsibility of authors; no NIMH clearances will be required except for NIMH staff who may serve as coauthors. Arbitration Procedure. If a steering committee decision is not acceptable to an awardee, the awardee may, within 30 days of notification of the decision, request a review by an arbitration panel composed of one arbitrator nominated by the awardee, one nominated by NIMH, and one chosen by the first two nominees. This panel will make a decision within 60 days of the request. Failure to comply with the decision of the panel may result in termination of support for the awardee by NIMH. This arbitration procedure in no way affects the awardee's right to appeal an adverse action in accordance with PHS regulations at 42 CFR Part 50, subpart D, and HHS Grant Administration Regulations at 45 CFR Part 74, section 304 and HHS regulations at 45 Parts 16 and 75. External Data Monitoring Board. During the first year, as part of the protocol development process, four external scientific advisors who have no project involvement will join a senior NIMH scientist on a monitoring board chaired by the Director, Division of Clinical Research. These external advisors will be selected from eminent scientists knowledgeable of adult and child psychopathology and treatment studies. The monitoring board will periodically (at least annually) review study progress and will provide input and recommendations to NIMH. While it will have no supervisory role, its input and annual review of the project are expected to help maintain the highest quality of research by facilitating transfer of new knowledge from adult treatment studies to this project. Public Domain of Data. It is expected that the data from this study will be collected and stored in a form suitable for alternative analyses. The data will first be used by the collaborators. However, since the data set will represent a potential national scientific resource, NIMH intends that it be made available to the larger research community as soon as feasible. NIMH will determine when all or parts of the collected data should be made available to the larger community. SPECIAL INSTRUCTIONS TO APPLICANTS REGARDING IMPLEMENTATION OF ADAMHA POLICIES CONCERNING INCLUSION OF FEMALES AND MINORITIES IN CLINICAL RESEARCH STUDY POPULATIONS Applications for Alcohol, Drug Abuse, and Mental Health Administration (ADAMHA) grants and cooperative agreements are required to include both females and minorities in study populations for clinical research, unless compelling scientific or other justification for not including them is provided. This requirement is intended to ensure that research findings will be of benefit to all persons at risk of the disease, disorder, or condition under study. For the purpose of these policies, clinical research involves human studies of etiology, treatment, diagnosis, prevention, or epidemiology of diseases, disorders or conditions, including but not limited to clinical trials; and minorities include U.S. racial/ethnic minority populations (specifically: American Indians or Alaskan Natives, Asian/Pacific Islanders, Blacks, and Hispanics). ADAMHA recognizes that it may not be feasible or appropriate in all clinical research projects to include representation of the full array of U.S. racial/ethnic minority populations. However, applicants are urged to assess carefully the feasibility of including the broadest possible representation of minority groups. Applications should include a description of the composition of the proposed study population by gender and racial/ethnic group, and the rationale for the numbers and kinds of people selected to participate. This information should be included in the form PHS 398 in Section 2, A-D of the Research Plan AND summarized in Section 2, E, Human Subjects. Applications should incorporate in their study design gender and/or minority representation appropriate to the scientific objectives of the work proposed. If representation of females or minorities in sufficient numbers to permit assessment of differential effects is not feasible or is not appropriate, the reasons for this must be explained and justified. The rationale may relate to the purpose of the research, the health of the subjects, or other compelling circumstances (e.g., if in the only study population available there is a disproportionate representation in terms of age distribution, risk factors, incidence/ prevalence, etc., of one gender or minority/majority group). If the required information is not contained within the application, the review will be deferred until it is complete. Peer reviewers will address specifically whether the research plan in the application conforms to these policies. If gender and/or minority representation/justification are judged to be inadequate, reviewers will consider this as a deficiency in assigning the priority score to the application. All applications/proposals for clinical research submitted to ADAMHA are required to address these policies. ADAMHA funding components will not award grants that do not comply with these policies. APPLICATION PROCEDURES Applicants are to use the grant application form PHS 398 (rev. 9/91). The number and title of this RFA, Cooperative Agreement for a Multi-site Treatment Study of ADHD, MH-92-03, must be typed in item number 2 on the face page of the PHS 398 application form. Application kits containing the necessary forms and instructions may be obtained from business offices or offices of sponsored research at most universities, colleges, medical schools, and other major research facilities. If such a source is not available, the following office may be contacted for the necessary application material: Grants Management Branch National Institute of Mental Health 5600 Fishers Lane, Room 7C-05 Rockville, MD 20857 Telephone: (301) 443-4414 The signed original and four legible copies of the completed application must be sent to: Division of Research Grants National Institutes of Health Westwood Building, Room 240 Bethesda, MD 20892** Additionally, it is requested that one copy be sent to: Division of Extramural Affairs National Institute of Mental Health 5600 Fishers Lane, Room 9-105 Rockville, MD 20857 Applications submitted for the special receipt date of May 19, 1992, must be complete and contain all information needed for initial and Advisory Council review. No subsequent addenda will be accepted unless specifically requested by the Scientific Review Administrator of the review committee. No site visits will be made. Applicants are reminded that the combined number of pages for Sections 1 through 4 of the Research Plan must not exceed 30 pages. Applications exceeding this limitation will be returned. Applicants from institutions that have a General Clinical Research Center (GCRC) funded by the NIH National Center for Research Resources may wish to identify the GCRC as a resource for conducting the proposed research. In such a case, a letter of agreement from either the GCRC program director or the Principal Investigator must be included with the application. If the applicant has an approved assurance covering the research (Multiple Project Assurance for human subjects), the applicant must provide with the application the certification of ethical IRB approval. The review and approval should occur prior to submission of the application. Terms and Conditions of Award Cooperative agreement funds may be used only for expenses clearly related and necessary to conduct the research outlined under this program including direct costs that can be specifically identified with the project and allowable indirect costs of the institution. Grants must be administered in accordance with the "PHS Grants Policy Statement (revised October 1, 1990)," which is available from any office of sponsored research and from the Superintendent of Documents, U.S. Government Printing Office, Washington, DC. Federal regulations at 42 CFR Part 52, "Grants for Research Projects," and 45 CFR Parts 74 and 92, generic requirements concerning the administration of grants, are applicable to these awards. Application Characteristics Applicants must follow the guidelines for a research grant in completing the application (PHS 398 application kit, revised 9/91). The application must include a design and protocol in the same detail as an R01 application to be used as one of the starting points in the year 1 development of the common protocol. The protocol should address, for example, the issues outlined under "Methodological Considerations" and the objectives of this RFA. Because of the expected complexity of the design, applicants may use 5 additional pages, up to 30 pages total, for sections 1 through 4. Budget Considerations. The first year, focused mainly on common protocol development and hiring and training of therapists and assessment staff, is expected to have a much lower budget than the years of assessment, treatment, and data gathering. The $2.5 million per year mentioned under "AVAILABILITY OF FUNDS" is the average total costs per year available for all sites, including indirect costs, over the five years. This estimate includes a reasonable allowance for purchased baseline metabolic, nutritional, and medical tests on urine, blood, and other tissue/fluid and compensation to subjects, parents, and teachers for cooperating with data collection. Collaborative Aspects. Because collaboration is so essential to the success of this study, it is important to include in the application evidence of good working relations in the applicant team, including any on-site consortium arrangement, and between the applicant team and cooperating referral sources. It is also useful to list any past research experience involving collaboration between the applicant team and other research teams or between the PI and investigator(s) at another university/center/institution. Protection of Human Subjects Applicants must demonstrate adequate procedures for minimizing risks and ensuring the safety and confidentiality of human subjects. The Department of Health and Human Services has developed additional regulations for protection of children involved as research subjects, including parental consent. A copy of these regulations (45 CFR 46, Protection of Human Subjects) is available from the NIMH staff listed below. During year one, the collaborators will need to develop common elements for use in a consent form that can then be adapted to individual sites-specific IRB requirements. While it is not anticipated that subjects would be compensated for participating in treatment, applicants should consider fair compensation/reimbursement for assessment time, inconvenience, lost time from work of the parents, transportation and parking cost, and possible discomfort or stress of tests. The following questions must be addressed in the application: What are appropriate but not coercive incentives for completing assessments? Should subject compensation/reimbursement vary by site in view of differing local costs of living and differing IRB standards or be uniform across sites? What are the implications for cross-site comparability in retention? REVIEW CONSIDERATIONS Applications received under this announcement will be reviewed for scientific and technical merit by an initial review group (IRG) consisting primarily of non-Federal experts. Final review is by the Mental Health Advisory Council. Notification of the initial review outcome will be sent to the applicant by the NIMH. Only applications recommended for approval by Council may be considered for funding. Criteria for scientific/technical merit review of research applications include the following: o potential contributions in areas covered by the objectives and scope of this announcement and responsiveness to the issues outlined above o scientific merit of the research design, approaches, and methodology o qualifications and experience of the Principal Investigator and proposed staff o adequacy of the conceptual and theoretical framework for the research o evidence of familiarity with relevant research literature o feasibility of the research plan o adequacy of proposed procedures for protecting human subjects and for ethical use of animal subjects o appropriateness of the budget, staffing plan, and timeframe to complete the project o adequacy of the existing and proposed facilities and resources o evidence of appropriateness of collaborative arrangements o plans for dissemination of research findings o appropriate inclusion of women and minorities in the study population AWARD CRITERIA Factors considered in determining which research applications will be funded include: scientific and technical merit as determined in the initial review, Council recommendations, responsiveness to the goals of this announcement, significance of the topic under study to NIMH priorities as announced in this or other special announcements or guidelines, program balance, public health significance, and availability of funds. INQUIRIES Potential applicants are encouraged to seek pre-application consultation from: L. Eugene Arnold, M.D. Child and Adolescent Disorders Research Branch Division of Clinical Research National Institute of Mental Health Parklawn Building, Room 10-104 5600 Fishers Lane Rockville, MD 20857 Telephone: (301) 443-5944 FAX: (301) 443-4045 Inquiries about grants management issues may be directed to: Stephen J. Hudak Chief, Grants Management Section National Institute of Mental Health Parklawn Building, Room 7C-05 5600 Fishers Lane Rockville, MD 20857 Telephone: (301) 443-4456 Potential applicants may obtain a copy of the "NIMH National Plan for Research on Child and Adolescent Mental Disorders" from: Information Resources and Inquiries Branch National Institute of Mental Health Parklawn Building, Room 15C-05 5600 Fishers Lane Rockville, MD 20857 Telephone: (301) 443-4513 AUTHORITY AND REGULATIONS This program is described in the Catalog of Federal Domestic Assistance Number 93.242. Awards are made under the authorization of the Public Health Service Act, Title IV, Part A (Public Law 78-410, as amended by Public Law 99- 158, 42 USC 241 and 285) and administered under PHS grants policies and Federal Regulations 42 CFR Part 52 and 45 CFR Part 74. This program is not subject to the intergovernmental review requirements of Executive Order 12372 or Health Systems Agency review.