The therapeutic benefits of psychostimulants were first reported in 1937. Now, more than 60 years later, stimulant therapy remains the single most effective therapy for treatment of children and adolescents with attention-deficit/hyperactivity disorder (ADHD).
This article focuses on the various new medications for treatment of ADHD. It is not intended to be a comprehensive review of pharmacotherapy for ADHD. For example, although tricyclic antidepressants are occasionally useful to treat ADHD in children, these medications are not discussed because they are not new. The reader is referred to several other resources for a more complete discussion of established stimulant and non-stimulant treatment options.1-2
In the past 2 years, there has been a proliferation of new methylphenidate preparations. The Table lists the various new stimulants approved for treatment of ADHD. Although the active ingrethent is the same in many of these preparations, the products vary in dose, delivery system, and duration of action.
Concerta. Concerta is an extended-release tablet that uses a novel delivery system, oral osmotic therapeutic system, to achieve a 12-hour duration of action (Figure). Following ingestion of the Concerta tablet, there is an immediate release of methylphenidate (MPH) in the outer covering of the tablet. Following this initial bolus, there is a progressive 8-hour release of MPH by virtue of an osmotic pump, which displaces the MPH from within the capsule as water enters the capsule through a semi-permeable covering.
Recent work by Swanson et al. had suggested that children perform better when their MPH level is rising rather than maintained at a level or steady state.3 The relative deterioration in response when MPPH levels are held constant was attributed to tachyphalaxis or acute tolerance to MPH. Thus, the Concerta tablet was designed with 2 separate drug sub-compartments, each with a different drug concentration; this ensures that mere is an increasing concentration of medication in the bloodstream in the afternoon.
Concerta is now available in 4 different strengths. The 18 mg tablet is comparable to 5mg of immediate-release (IR) MPH Unice daily (tid); the 36 mg tablet is comparable to 10 mg twice daily (bid); and the 54 mg tablet parallels a 15 mg tablet tid. The FDA recently approved a fourth tablet strength, 27 mg, which parallels 7.5 mg tid and provides greater dosing flexibility. Although the Physicians' Desk Reference states that the maximal daily dose for Concerta is 54 mg, mis dose is below the comparable IR maximal daily dose of 60 mg. Because adolescents and adults frequently require 20 or 25 mg tid, many pediatricians commonly prescribe 72 mg or more per day in older patients.
Figure. OROS Mechanism. Oral Osmotic Therapeutic Systems resemble a conventional tablet in appearance. The OROS system for Concerta consists of a semipermeable tablet shell surrounding an osmotically active drug core. The tablet shell is covered by a drug overcoat. The core itself is divided into three compartments: two active layers containing the drug in different concentrations and a push layer containing pharmacologically inert but osmotically active polymer ingrethents. In an aqueous environment such as the gastrointestinal tract, the exterior drug overcoat dissolves within approximately one hour, providing an immediate release of 22% of the entire dose of methylphenidate. After the drug overcoat is dissolved, water permeates through the tablet shell into the tablet core. As the osmotically active polymer excipients expand, methylphenldate is released through the laser-drilled orifice in a delivery rate designed to provide efficacy for 12 hours. Inert ingrethents of the tablet remain intact during gastrointestinal transit and are eliminated in the fèces as an insoluble shell.
Because Concerta was the first of the new preparations introduced within the past 2 to 3 years, several studies are now available regarding clinical efficacy and patient satisfaction. In a randomized, double-blind, placebo-controlled study by Wolraich et al.,4 treatment with Concerta and IR methylphenidate given tid were comparable with respect to efficacy (effect sizes of 1.05 and 1.02, respectively) and no serious adverse events were noted with either medication condition. In another placebo-controlled study using a doubleblind, double-dummy protocol so mat there was no difference in dosing convenience, parents more frequently expressed a preference for Concerta compared to tid IR MPH.5
Concerta is the only ADHD medication approved by the FDA with a 12-hour duration of action; thus, it is especially helpful in treating children and adolescents who need afterschool medication coverage as well. Although the medication lasts 12 hours on average, the medication may have a somewhat longer duration of benefit in some children and somewhat shorter efficacy in other children. However, unlike several of the other long-acting stimulant preparations, Concerta's absorption and pharmacokinetics are not adversely affected by a high-fat breakfast.6
It should be noted that in the recently completed MTA study that assessed the efficacy of medication and psychosocial interventions together and separately, children treated in the medication group improved considerably more than did those medicated by physicians in their community. One of several key differences between these two treatment groups was that all the children in the medication group received medication for 12 hours, whereas many of the children treated in a community setting only received medication for school. Although Concerta was not given to subjects in the medication treatment group, many of these children were treated with a 12-hour methylphenidate regimen (IR MPH given tid).
When Concerta was introduced 2 years ago, concerns were surfacing about rare reports of teenagers diverting their IR MFH to friends or others, in some instances with the intention of snorting the medication for a psychoactive effect. Because Concerta is an extended-release preparation that generally enables ADHD patients to be treated effectively with once-a-day dosing, it eliminates the need for children and adolescents to assume responsibility for mid-day dosing, thus reducing the potential for diversion. Similarly, the Concerta tablets cannot be easily crushed, and attempted crushing will result in a pasty compound that cannot be snorted for psychoactive effects.7
Metadate CD. Metadate CD is a new sustainedrelease MPH product that is designed to provide medication efficacy "throughout the school day." Metadate CD is not designed to provide extended coverage after school. Its use is appropriate if full day coverage is not being considered or in combination with some other formulation to cover the afterschool time.
Metadate CD capsules are reported to have a biphasic release, using bead technology. Thirty percent of the dose is rapidly released and 70% of the dose is continuously released. The manufacturer reports that Metadate leads to 2 peaks (at 1.5 and 4.5 hours following ingestion). However the pharmacokinetic curve for a 20 mg dose presented in the package insert is relatively flat. Further study of behavioral effects during the course of a day is required to clarify the actual effects.
At present, Metadate CD is only available as a 20 mg tablet. The manufacturer recommends starting children taking a 20 mg dose in the morning and then increasing in weekly 20 mg increments to a maximum daily dose of 60 mg. Metadate CD has the advantage that it can be given as a sprinkle over food (eg, applesauce). Although this is especially helpful when treating younger children who cannot swallow a pill, the 20 mg dose may be too strong for some. (Although not recommended in the package insert, I have often counseled parents to "titrate" medication using a sprinkle dosing approach. I instruct families to empty the entire capsule into 3 or 4 teaspoons of applesauce or another suitable pureed food, stir, and then give the child the appropriate fraction of the food. Parents must be clearly instructed to carefully discard all unused medication each morning.) Metadate CD's pharmacokinetic curve is affected by food; a high-fat meal leads to approximately a 1-hour delay in the early peak from the rapid-release bead.
Metadate CD is dispensed as a 30-capsule dose pack, with 6 rows of 5 tablets. Although this may facilitate tracking of medication administration for children who only take the medication on weekdays, it will likely prove less helpful for children who take the medication daily.
Metadate ER. Metadate ER is a branded generic version of Ritalin SR, the original sustained release MPH preparation that uses a wax-matrix tablet design. Ritalin SR has always been viewed as an imperfect sustained-release formulation. Compared to immediate-release MPH given bid (breakfast and lunch), Ritalin SR had a somewhat delayed onset of action, diminished efficacy after lunch, and highly variable (generally diminished) duration of action. Metadate ER tablets presumably offer the same benefits and limitations as did Ritalin SR tablets. Metadate ER is less expensive than Ritalin SR and is also available in 10 mg tablets. With the subsequent release of Metadate CD and the more recent release of Ritalin LA, there should be little demand for Metadate ER, because the newer products were specifically designed to provide more sustained coverage throughout the 8-hour school day. Nonetheless, given the individual variability of response to medication, there will likely always be a small number of children for whom Metadate ER may be the preferred sustained-release preparation. Also, in select instances Metadate ER can be used to supplement treatment with an extendedrelease preparation. For example, if a child seems to need a higher dose of MPH for school than after school or if side effects warrant a dose reduction later in the day, Metadate ER can be effectively paired with Concerta to achieve oncea-day dosing in a "sculpted" fashion.
Methylin and Methylin ER (Methylin 5, 20, 20mg; Methylin ER 10, 20mg). Methylin and Methylin ER are other methylphenidate preparations that are promoted as brand-name products yet have the cost profile of generic products. Methylin and Methylin ER are designed as "branded generics" to compete with Ritalin and Ritalin SR. Methylin's manufacturer, Mallinckrodt, has emphasized that these products are lactose-free and dye-free, implying that they would be better tolerated by children with certain food sensitivities. However, the amount of lactose in other methylphenidate preparations is exceedingly small. There are no published reports linking lactose intolerance gastrointestinal symptoms to other methylphenidate preparations. The 10 mg Methylin ER preparation may have some clinical utility in the same fashion that Metadate ER 10mg does. However, because Methylin ER is designed to parallel the pharmacokinetics of Ritalin SR, it is not likely to be die best long-acting preparation for children who need consistent coverage throughout the school day.
Focalin. Focalin is one of the newest stimulant preparations to be approved by the FDA. Whereas Ritalin and all other available methylphenidate preparations contain racemic or d,lmethylphenidate, Focalin is the d-isomer of methylphenidate. The development of this singleisomer form of ^/-methylphenidate was predicated on the fact that there is a substantial difference in clinical efficacy and metabolism of the two stereoisomers. The /-isomer is more rapidly metabolized and degraded following oral administration and has little, if any, pharmacologie activity.8 The d-isomer is thought to be the clinically active enantiomer. In a double-blind, parallel-group dose titration study sponsored by Novartis comparing dex-MPH (Focalin), d,l-MPH, and placebo, investigators identified a greater percentage of children on Focalin as scoring "much improved" or "very much improved" compared to Ritalin or placebo (66%, 47%, and 20%, respectively)
At present, Focalin is available only as an immediate-release (IR) formulation. This medication would likely be most helpful whenever a short-acting preparation is needed, to maximize dosing flexibility, to complement therapy with a longer-acting preparation, or to minimize side effects. Focalin should be considered specifically when a longer IR preparation is needed and /or a child has not responded optimally to IR racemic MPH. Focalin is generally prescribed at half the usual methylphenidate dose. Thus, with young children, one would likely start at 2.5 mg per dose and then increase to 5 or 10 mg as needed. Preliminary clinical data suggest that Focalin may last somewhat longer than IR racemic MPH, with benefit noted 5 to 6 hours post dose. The manufacturer recommends that subsequent doses be given at least 4 hours apart. Although many families and pediatricians prefer once-aday dosing during the school week, Focalin and other IR MPH preparations can be helpful on weekends and other days when some children do not need 8-hour or 12-hour dosing. A long-acting version of Focalin is being developed but will likely not be available for at least 2 years.
Ritalin LA. Ritalin LA is another new long-acting preparation designed to provide once-daily dosing for the school day. Although a sustainedrelease Ritalin preparation has existed for many years (Ritalin SR), this formulation has often been noted to be less effective or to have shorter duration of action in some patients compared to bid dosing with IR MPH.
As with Metadate CD and Adderall XR, Ritalin LA employs bead technology to provide an improved long-acting (sustained-release) preparation with a biphasic release. Whereas Metadate CD capsules have 30% rapid-release beads and 70% continuous-release beads, Ritalin LA has 50% immediate-release beads and 50% modifiedrelease beads. Approximately 4 hours after administration, the MPH in modified-release beads diffuse out of the beads through pores in the outer polymer coating. This technology leads to a bimodal release profile that is more evident compared to 20 mg Metadate CD, but has smoothed peaks and troughs compared to IR MPH given bid.
Adderall XR. Adderall XR is a longer-acting version of Adderall, die recently introduced mixed amphetamine salts preparation that was originally marketed under the name Obitrol. Adderall gained considerable popularity in the late 1990s as an effective alternative to MPH for treatment of ADHD. The clinical duration of action of IR Adderall is highly variable, ranging from 5 hours to 8 hours, depending on dose and other factors. Adderall XR is designed to parallel a 4-hour bid regimen of IR Adderall. Adderall XR has a drug delivery system similar to Ritalin LA. Fifty percent of the active ingrethent in Adderall XR is in immediate-release beads, with the other 50% in beads that begin to release approximately 4 hours later. Adderall XR is described as a oncea-day treatment for ADHD; preliminary studies suggest that its duration of benefit extends beyond the traditional 8-hour school day, more likely 10 to 12 hours. Food may significantly affect Adderall XR's absorption. Compared to a fasted condition, a high-fat breakfast resulted in 44%, 57%, and 65% reductions in drug exposure for the first 4, 6, and 8 hours following administration in normal adults.6
Most Adderall studies to date have compared it to MPH; one recent study compared a single morning dose of IR Adderall with dextroamphetamine tablets and spansules in a double-blind fashion.9 Methodological constraints precluded the authors' declaring one amphetamine best overall. This study did confirm Adderall's efficacy, with most of its benefits noted in die morning. Not surprisingly, the dextroamphetamine tablet had a more immediate effect than the spansule, but the spansule was more effective after lunch. Adderall had a similar number of side effects as the dextroamphetamine preparations, though sleep seemed to be less affected by the Adderall.
At present, stimulants are the only medications that have an ADHD indication. Although several non-stimulants have been used with varying success in patients with ADHD, none have FDA approval at this time. Nonetheless, some of these medications can be considered for use "off label" when treating children and adolescents with ADHD that is refractory to stimulant therapy.
Guanfacine. Guanfacine is an alpha-2 agonist that is an anti-hypertensive similar to donidine, but longer acting and less sedating. Recent studies have suggested that guanf acine is helpful for treating children with ADHD and co-morbid tic disorders. As a second-line treatment for ADHD, the alpha-2 agonists have previously been shown to reduce hyperactivity; interestingly, a recent study by Scahill et al.10 demonstrated that guanfadne also improved attention span. The typical daily therapeutic dose for guanfaäne is 0.5 to 4 mg (30 to 100 meg/kg) divided bid. Side effects may include sedation (though less than donidine), insomnia, irritability, and dry mouth. Although hypotension and rebound hypertension are frequently suggested as concerns with guanf acine, no clinically or statistically significant impact on blood pressure or pulse were noted in one recent report.11
Bupropion. Bupropion is a novel-structured anti-depressant that has both indirect dopamine agonist and noradrenergic effects. Bupropion has been used recently as a second-line medication for ADHD in children. To date, bupropion SR has been studied primarily in adult ADHD patients and adolescents with ADHD and a comorbid depressive disorder.12'13'14 In each of these studies, bupropion was superior to placebo. In a parallel-groups design comparing bupropion, methylphenidate, and placebo in adults with ADHD, both medications were superior to placebo and there was no difference between the two drugs with respect to response rates. At this time, bupropion is seldom prescribed by pediatricians, but it is likely the non-stimulant that psychiatrists now prescribe most commonly for treating ADHD in youth. Studies still need to be done examining the efficacy of bupropion in grade school children with ADHD. Although bupropion may be especially helpful for treating children with ADHD and depression, pediatricians should bear in mind that low self-esteem is common in children with ADHD (especially among older patients, those in middle school and high school). Bupropion SR is generally taken twice daily, up to 200 mg in the morning and 100 to 200 mg in the evening, with a maximum daily dose of 400 mg. Common side effects include agitation/restlessness, anxiety and insomnia initially; additional potential side effects include headaches, difficulty urinating, constipation, and, possibly, reduced seizure threshold.
Medications On the Near Horizon
Atomoxetine. Atomoxetine (originally referred to as tomoxetine) is a novel compound that enhances noradrenergic function through highly selective blockade of the pre-synaptic norepinephrine transporter. Atomoxetine has been studied in children with ADHD in 3 double-blind, placebo-controlled studies (2 dose-titration and 1 dose response), with additional data garnered in a large open-label study. Results from the dose-response study were recently published.15 Atomoxetine was shown to be effective in reducing hyperactivity and improving attention in children with ADHD.16 One-year follow-up data on 19 children (ages 7 to 15 years) enrolled in an earlier open-label study documented that the clinical efficacy was sustained and that the medication was well tolerated. None of the 19 responderá withdrew due to side effects, though decreased appetite was one of the more common side effects reported.
Lilly has announced that it anticipates receiving FDA approval in spring 2003 and that it will market the drug under the brand name Strattera.
Methypatch. Noven Pharmaceuticals, a developer of advanced transdermal drug delivery technologies and prescription transdermal products, has completed a phase IH clinical trial for a transdermal methylphenidate preparation. This product, MethyPatch, is designed to be applied daily to children with ADHD. In a double-blind study comparing transdermal MPH with placebo, the primary outcome measure was teacher ratings of attention and behavior using the IOWA Conners' Rating Scale. MethyPatch led to a statistically significant improvement (p < .001) in patient attention and behavior compared to placebo. Noven expects to submit its new drug application to the FDA in summer 2002 and win approval by the second half of 2003.
WHICH MEDICATION IS BEST?
Most children and adolescents with ADHD will likely benefit from each of the stimulant medications approved by the FDA for treatment of ADHD. Past studies indicate that the majority of youth respond well to both amphetamine and MPH, although some children may respond better to one than the other.17 Just as there is no preferred medication, there is no single drug preparation that meets the needs of all children. Whereas many children with ADHD, combined type likely would most benefit from an extendedrelease preparation, not all children with ADHD need to be treated for 12 hours. The child with ADHD, inattentive type, may not need medication coverage after school if he or she is able to get his or her work done in a quiet home environment. Assuming medication is not needed for afterschool activities or to address social skills, this child would likely only need an 8 to 9 hour preparation, such as Ritalin LA or Metadate CD, or possibly Dexedrine spansule.
Pediatricians should feel comfortable using more than one medication in a specific child. For example, a child may need a 12-hour preparation for weekdays and only require IR MPH on weekends when homework or studying is being done. Similarly, as noted earlier, dosing regimens can be individualized (sculpted) as needed by combining different stimulant preparations in the same day.
Obstacles to Optimal Care
Research lag. Prior to the release of Adderall in the mid-1990s, pediatricians treating children with ADHD had a very narrow selection of medications, all of which had been on the market for more than a decade. This static period enabled pediatricians to garner considerable first-hand experience with the few approved medications available and, more important, gave researchers ample time to study the relative efficacy of these medications independently in different patient populations. In addition, it allowed researchers to compare the medications objectively against each other.
Although there will likely be many independent head-to-head comparisons of MPH and amphetamine preparations conducted by pediatricians, most of the studies published to date that examine the safety and efficacy of these new medications have been sponsored by their respective manufacturers.
Many of these industry-sponsored studies incorporate sound research methodology; however, there have been a few recent instances where industry-sponsored studies comparing stimulants had such significant design flaws in their methodology that they no longer represent direct comparisons. Pediatricians need to scrutinize all published studies to verify that the methodology is appropriate.
Given the time required for study design, Institutional Review Board approval, subject recruitment, data collection /analysis, journal review, and publication, it is unlikely that independent studies of ADHD medications would appear within 18 months of a new medication's release. Thus, at a time when pediatricians are trying to assess the relative merits of each new medication released objectively, there is a paucity of such information.
For example, are there significant differences between Ritalin LA and Metadate CD, the two newest sustained-release preparations designed to last for the entire school day? Swanson's MPH acute tolerance data suggests that an "ascending profile" is most effective. Because Ritalin LA appears to have a modestly ascending pharmacokinetic curve whereas Metadate CD has a relatively flat curve, one would expect children to do less well after lunch in school on Metadate CD compared to Ritalin LA. Questions such as these will undoubtedly be answered in the near future.
Access to Medication
Insurance-related formulary restrictions often limit medication options for physicians and families. There is often a delay before new stimulant medications are included on each managed care company's list of approved drugs, and some companies choose not to recognize some preparations altogether. It is difficult for physicians to track formulary-specific medication options, and it is cumbersome for both families and physicians to seek prior authorizations or waivers on a caseby-case basis. To the extent that children with ADHD generally need to continue to take medications for several years, it is not feasible for parents to assume the cost of these medications.
Variability of response. Children do not read psychopharmacology texts and therefore, they do not always respond to stimulant medications as expected. In general, methylphenidate and amphetamine preparations are equally effective for the treatment of ADHD. Although many children respond equally well to both types of stimulants, some children respond better to one stimulant medication than the other. Unfortunately, this cannot be predicted. Thus, pediatricians should feel comfortable prescribing several different medications for ADHD. Pediatricians must also remember that not only is there inconsistency between medications regarding efficacy, there is also variability within medications. Thus, although studies indicate that a medication has an 8-hour duration of benefit on average, this medicine may last only 6.5 or 7 hours in some children and 9 hours in others. This variability in clinical response means that pediatricians, despite their best clinical judgment in selecting a medication to best meet a patient's needs, may still need to adjust doses and occasionally try more than one medication to achieve success.
The development of several new stimulant preparations provides pediatricians and their patients with considerable treatment advantages with respect to dosing flexibility. These newer products are useful additions to the pediatrician's armamentarium. To the extent mat the various new long-acting MPH products have somewhat distinct pharmacokinetic curves and durations of action, pediatricians could ideally choose among these various preparations to find the stimulant medication that best meets the needs of the patient with a minimum of side effects. These newer stimulant preparations enable children and adolescents to avoid mid-day dosing at school, mus facilitating confidentiality and adherence. Longer-acting preparations also reduce the need for afterschool dosing and reduce the potential for diversion or abuse by latchkey teens.
In its recent clinical practice guidelines for treatment of ADHD, the American Academy of Pediatrics clearly affirms that primary care pediatricians should assume responsibility for diagnosis and treatment of ADHD in school-aged children,2 Moreover, for the first time they have acknowledged that treatment with stimulant medication may be part of an initial treatment plan, and that not all children require psychosocial interventions. The expanded number of stimulant preparations greatly facilitates dosing flexibility when treating children with medication. The likely prospect of transdermal preparations and non-stimulant preparations such as atomoxetine will further expand the pediatrician's treatment armamentarium. Pediatricians should become familiar with the potential advantages of the various new preparations so that they can develop a treatment plan that is effective and with minimal side effects and inconvenience.
1. American Academy of Child and Adolscent Psychiatry. Practice parameters for the use of stimulant medications in the treatment of children, adolescents, and adults. JAm Acad Child Adúlese Psychiatry. 2002;41:26S-49S.
2. American Academy of Pediatrics. Clinical practice guideline: treatment of the school-aged child with attentiondeficit/hyperactivity disorder. Pediatrics. 2001;108:1033-1044.
3. Swanson J, Gupta S, Guinta D. Acute tolerance to methylphenidate in the treatment of attention-deficit hyperactivity disorder in children. Clin Pharmacol Ther. 1999;66:29-305.
4. Wolraich ML, Greenhill LL, Pelham W, et al. Randomized controlled trial of ORQS methylphenidate once a day in children with attention-deficit /hyperactivity disorder. Pediatrics. 2001;108:883-892.
5. Robb AS, Stein MA. Parent Preference for OROS MPH qd over MPH tid. Poster presentation at annual meeting, American Psychiatric Association, 2002.
6. Auiler JF, Liu K, Lynch JM, Gelotte CK. Effect of food on early drug exposure from extended-release stimulants: Results from the Concerta, Adderall XR food evaluation (CAFÉ) study. Curr Med Res Opin. 2002; 18:311-316.
7. Jaffe SL. Failed attempts at intranasal abuse of Concerta. / Am Acad Child Adúlese Psychiatry. 2002;41:5.
8. Novaras, Focalin Product Monograph, 2002.
9. James RS, Sharp WS, Bastain TM, et al. Double-blind, placebo-controlled study of single-dose amphetamine formulations in ADHD. / Am Acad Child Adolesc Psychiatry. 2001;401268-276.
10. Scahill L, Chappell PB, Kim YS, et al. A placebo-controlled study of guanfacine in the treatment of children with tic disorders and attention deficit hyperactivity disorder. Am ] Psychiatry. 2001;158:1067-1074.
11. Horrigan JP, Melkpour AH, Barnhill LJ. The long-term safety of guanfacine for children and adolescents with psychiatric disorders. New Clinical Drug Evaluation Unit; 2000.
12. Kuperman S, Perry PJ, Gaffney GR, et al. Buproprion SR vs methylphenidate for attention-deficit hyperactivity disorder in adulte. Ann Clin Psychiatry. 2001;13:129-134.
13. Daviss WB, Bentivoglio P, Racusin R, et al. Bupropion sustained release in adolescents with co-morbid attentiondeficit /hyperactivity disorder and depression. / Am Acad Child Adolesc Psychiatry. 2001;40:307-314.
14. Wilens TE, Spencer TJ, Biederman J, et al. A controlled clinical trial of buproprion for attention deficit hyperactivity disorder in adults. Am J Psychiatry. 2001;158: 282-288.
15. Spencer T, Biederman J, Heiligenstein J, et al. An openlabel, dose-ranging study of atomoxetine in children with attention deficit hyperactivity disorder. / Child Adolesc Psychopharmacol. 2001; 11:251-265.
16. Michelson D, Paries D, Wernicke J, et al. Atomoxetine ADHD Study Group. Atomoxetine in the treatment of children and adolescents with attention-deficit /hyperactivity disorder: a randomized, placebo-controlled, doseresponse study. Pediatrics. 2001;108:E83.
17. Elia J, Borcherding B, Rapoport J, Keysor C. Methylphenidate and amphetamine treatments of hyperactivity: are there true non-responders? Psychiatry Res. 1991;36:141-155.