Long Acting Medications for the Hyperkinetic Disorders

Tobias BanaschewskiDavid CoghillParamala SantoshAlessandro ZuddasPhilip AshersonJan BuitelaarMarina DanckaertsManfred DopfnerStephen V. FaraoneAribert RothenbergerJoseph SergeantHans-Christoph SteinhausenEdmund J.S. Sonuga-BarkeEric Taylor

Long-acting medications for thehyperkinetic disorders

A systematic review and European treatment guideline

Accepted: 13 March 2006Published online: 5 May 2006

j Abstract A systematic review ofpublished and unpublished dataon the use of long-acting medica-tions in ADHD and hyperkineticdisorder is reported, giving effectsizes and numbers-to-treat forextended-release stimulant prepa-rations and atomoxetine (ATX). Apanel of experts from severalEuropean countries used the re-view to make recommendationsabout the use of these drugs inpractice, and conclusions are re-ported: (1) Long-acting prepara-tions should be available and used;(2) They should not replace short-acting drugs (which will be theinitial treatment for many childrenfor reasons of cost and flexibilityof dosing). Individual clinicalchoice is needed. (3) Both ATXand extended-release preparationsof stimulants should be available.The choice will depend upon thecircumstances, and detailed rec-ommendations are made.

j Key words guidelines review stimulants atomoxetine ADHD

J. BuitelaarDept. of PsychiatryAcademic Centre for Child & AdolescentPsychiatryUMC St RadboudNijmegen, The Netherlands

M. DanckaertsDept. Child & Adolescent PsychiatryU. Z. GasthuisbergLeuven, Belgium

M. DopfnerPsychiatry and Psychotherapy ofChildhood & AdolescenceUniversity of CologneKoln, Germany

S.V. FaraoneMedical Genetics Research Program andDept. of Psychiatry and BehavioralSciencesSUNY Upstate Medical UniversitySyracuse (NY), USA

J. SergeantDept. of Clinical NeuropsychologyFaculty of Psychology & EducationVrije UniversitatAmsterdam, The Netherlands

H.-C. SteinhausenDept. of Child & Adolescent PsychiatryUniversity of ZurichZurich, Switzerland

E.J.S. Sonuga-BarkeDept. of PsychologyUniversity of SouthamptonSouthampton, UK

E. TaylorDept. of Child & Adolescent PsychiatryInstitute of PsychiatryKings College LondonLondon, UK

ORIGINAL CONTRIBUTIONEur Child Adolesc Psychiatry (2006)15:476495 DOI 10.1007/s00787-006-0549-0

ECAP549

Priv. Doz. Dr. T. Banaschewski (&)A. RothenbergerDept. of Child & Adolescent Psychiatry/PsychotherapyUniversity of GoettingenVon-Siebold-Str. 537075, Goettingen, GermanyTel.: +49-551/39-6727Fax: +49-551/39-8120E-Mail: [email protected]

D. CoghillDept. of PsychiatryUniversity of DundeeDundee (SCO), UK

P. SantoshDept. of Psychological MedicineHospital for ChildrenGreat Ormond StreetLondon, UK

P. SantoshMaudsley Hospital, Institute of PsychiatryLondon, UK

A. ZuddasChild NeuropsychiatryDept. NeuroscienceUniversity of CagliariCagliari, Italy

P. AshersonMRC Social Genetic & DevelopmentalResearch CentreInstitute of PsychiatryKings College LondonLondon, UK

Introduction

j Background

A group of European physicians and psychologistsrecently published guidelines for the treatment ofhyperkinetic disorders [71]. There have been severaldevelopments in the field since then; one has been theadvent of newly licensed drugs and drug preparationswhose effect lasts through a substantial part of theday. This can overcome the inconveniences of theshort-acting stimulants. A non-stimulant has beenlicensed, and often is in use. All, however, entail in-creased costs, and the choice among them can beconfusing. This paper provides a review of their use.

The previous guidelines for the treatment ofhyperkinetic disorders are still valid: a comprehensivetreatment programme should include measures otherthan medicinessuch as psychological, educationaland social interventions. Treatment should be initi-ated under the supervision of a specialist in childhoodbehavioural disorders. Diagnosis should be made inaccordance with DSM-IV criteria or the guidelines inICD-10 and should be based on a full history andevaluation of the patient. Pharmacotherapy is notindicated for all children with this syndrome and thedecision to use any of the drugs must be based on athorough assessment of the severity, impact anddevelopmental appropriateness of the childs symp-toms. The current paper should be seen as a supple-ment to these recommendations.

j Methodology of the review and guidelinedevelopment

A group of experienced academic clinicians andclinical researchers with special interests in thehyperkinetic disorders was identified through theEuropean Network for Hyperkinetic Disorders(EUNETHYDIS). Systematic reviews by NICE [30]and SIGN [54] were recent and available, and formedthe basis for identifying published papers on meth-ylphenidate, dexamfetamine and atomoxetine (ATX).Faraones meta-analyses [12, 14, 15] were also re-ferred to. Pharmaceutical companies with long-actingdrugs in their portfolios were asked to provideunpublished as well as published data and posters(but information in commercial confidence was notincluded). The companies also presented at a meetingin September 2005 held in London. The calculation ofeffect sizes was carried out using a standard meth-odology. Where companies provided calculations ofeffect sizes using the standard methodology thesewere employed. If companies did not provide calcu-lations using the standard methodology they werecalculated by the authors.

The review and guideline development process wasiterative. The identification of clinical questions wascarried out in a meeting of all the authors, and aworking group (TB, DC, SP, AZ) made a systematicreview of published and unpublished clinical trialsincorporating long-acting preparations; a quantitativereview of the relevant data is reported below. Furthermeetings critiqued these reviews and their conclu-sions. A quantitative methodology for resolving dis-agreements had been anticipated (including anassessment of competing-interest information); but inthe event, conclusions were unanimous. Drafts of thepaper were exchanged and discussed iteratively. Thefinal document is subscribed to by all the authors. Thepaper includes both a narrative summary of conclu-sions and a scientific examination of the data in twoseparate sections.

Narrative review and recommendations

j Description of the drugs

Stimulants

j Methylphenidate hydrochloride

Methylphenidate hydrochloride (MPH) is a mildcentral nervous system (CNS) stimulant with moreprominent effects on mental than motor activities. Itsuse in the treatment of hyperkinetic disorder andother forms of ADHD is well established and has beenreviewed by EUNETHYDIS for European Child andAdolescent Psychiatry [71]. It is licensed in mostEuropean countries as part of comprehensive treat-ment programmes in children (over 6) and adoles-cents when remedial measures alone proveinsufficient. Ritalin LA (Novartis), Equasym XL(UCB), Concerta XL (JanssenCilag) and MedikinetRetard (Medice) all provide a mixture of immediate-and extended-release methylphenidate; they differ inthe physics of the delayed-release system and in theproportion of immediate to delayed. Figure 1 showsthe pharmacokinetic (PK) profile over time of thesedifferent formulations; the actions on behaviourparallel the concentrations in the blood.

j Concerta XL

Pharmacology: Concerta XL employs an osmoticpump system (OROS), which has been designed tohave a12-h duration of effect. An 18 mg tablet in-cludes a shell of 4 mg immediate-release (i.e. 22%). Inclinical trials, Concerta XL has been shown to be aseffective as 3-times-a-day methylphenidate [48, 80].However, a 20% higher dose of Concerta XL was re-quired in the Pelham et al. [48] study.

T. Banaschewski et al. 477Long-acting medications for the hyperkinetic disorders

Time course of action: After oral administration ofConcerta XL to adults, plasma methylphenidate con-centrations rise and reach an initial maximum after12 h, due to the capsule drug overcoat dissolving.Following this, the concentrations increase gradually,due to slow release from the capsule functioning as anosmotic pump designed to deliver MPH with anascending profile for about 10 h after administration[69]. The action on behaviour is stated to last about12 h [69]; but clinical experience suggests that (aswith other ER preparations) there is considerableinter-individual variation.

Clinical aspects: Concerta XL is usually available in18, 36 and 54 mg strengths. The tablets are adminis-tered orally once daily in the morning, must beswallowed whole with the aid of liquids, and must notbe chewed, divided or crushed. Concerta XL may beadministered with or without food. Dosage should beindividualised according to the needs and responsesof the patient, often by starting with 18 mg andadjusting in 18 mg increments at approximatelyweekly intervals. The maximum dose is stated as54 mg; but this takes no account of the patients sizeand the guidelines group consider higher doses up to

% IR % ER % IR % ER MPH IR BID

100%

CONCERTA XL

22% 78%

EQUASYM XL

30% 70%

MEDIKINET RETARD

50% 50%

RITALIN LA

50% 50%

ADDERALL XR

50% 50%

Fig. 1 MPH and amphetamine plasma levels over time with differentpreparations and their IR/ER proportions Note: curves for MPH IR BID,Medikinet retard provided by Medice, for Adderall XR provided by Shire; forRitalin LA see: Markowitz et al. [39]; for Concerta XL and Equasym XL see:

Gonzalez et al. [21]; all curves are adapted to a common time-scale. Doses ofdifferent products are not equivalent so there is no common Y-axis and directcomparisons should not be made for the absolute levels

478 European Child & Adolescent Psychiatry (2006) Vol. 15, No. 8 Steinkopff Verlag 2006

a maximum of 2 mg/kg/day; or 108 mg total dailydosewhichever is the smaller, to be acceptable andrecommended in some cases. It is possible to titratechildren straight on to Concerta XL without theirneeding to be first titrated on to immediate-releasestimulant medication [68]. Some clinicians, however,prefer to establish the optimal dose by first titratingwith immediate-release.

j Equasym XL

Pharmacology: In Equasym XL, the capsules compriseboth immediate- and extended-release beads suchthat 30% of the dose is provided by the immediaterelease component and 70% of the dose is providedby the extended-release component.

Time course of action: In line with its differentdosing profiles, Equasym XL has a shorter duration ofaction than Concerta XL (approximately 8 h) but alarger effect in the 45 h immediately after dosinggiven similar total daily doses [70]. As with ConcertaXL there is variation between individuals in the timecourse of action.

Clinical aspects: At the time of writing, EquasymXL is only available in the UK. The drug isadministered once daily in the morning, beforebreakfast. Our recommended starting dose is 10 mgonce daily. The companys advice is that dosagemay be adjusted in weekly 20 mg increments to amaximum of 60 mg/day; our experience is in linewith this, but there may be a limited number ofcases who benefit from higher doses and the max-imum for them is 2 mg/kg/day; or 100 mg totaldaily dosewhichever is the smaller.

j Medikinet retard

Pharmacology: This is another proprietary formula-tion of methylphenidate, made in Germany and li-censed only there at the time of writing, mixing 50%immediate with 50% extended, coated for gastric acidresistance.

Time course of action: The effective action is atleast 7 h and the effects of once daily Medikinet Re-tard are comparable to the effects of twice daily Rit-alin [11]. As with Equasym XL the dosing profilepredicts a larger immediate, and smaller delayed,action than Concerta XL, with individual differencesin time course being observed.

Clinical aspects: The drug is administered oncedaily in the morning, after breakfast. If titration isdone with this drug our recommended starting doseis 10 mg once daily. The companys advice is thatdosage may be adjusted in weekly 10 mg incre-ments. The maximum dose is as for Equasym XL(above).

j Ritalin LA

Pharmacology: Ritalin LA includes a mixture ofimmediate- and extended-release beads using theproprietary SODAS (Spheroidal Oral Drug AbsorptionSystem) drug delivery system. It has about 5050proportions of immediate and delayed release. It islicensed in a small number of European countries aswell as the USA.

Time course of action: As with Equasym XL andMedikinet Retard the dosing profile predicts a largerimmediate, and smaller delayed, action than ConcertaXL and some evidence supports this [36].

Clinical aspects: Recommended doses are as forEquasym XL.

One characteristic of Ritalin LA, Medikinet Retardand Equasym XL, is that the capsules can be openedand the contents sprinkled on a suitable foodstuffwithout any loss of activity. In most cases, formula-tions are not licensed for this use and this practice isnot recommended by the manufacturers. However,sprinkling can be useful for children who have dif-ficulties swallowing. It has also been suggested thatgiven this characteristic these formulations could beeasier to abuse, although there is no evidence for thisview.

j Adderall XR

Pharmacology: Adderall XR, like its IR predecessorAdderall, is composed of four amphetamine salts:amphetamine aspartate, amphetamine sulfate, dex-amfetamine1 saccharate, and dextroamfetamine sul-fate (this last is the active ingredient of Dexedrine).These four component salts are metabolised at dif-ferent rates. Adderall XR utilises a beaded deliverysystem in which the first type of bead dissolvesimmediately and the second type 4 h later. The drugis not licensed in Europe, but clinicians mayencounter it in immigrants already using it.

Time course of action: Maximum plasma concen-tration is achieved in 7 h, compared to regular Ad-derall IR (immediate-release) that reaches maximumplasma concentration within 3 h. The duration ofaction is around 810 h [42].

Clinical aspects: Doses for both immediate- andextended-release form come in 5, 10, 15, 20, 25, and30 mg increments.

1The spelling of dexamfetamine reflects the changes made inRecommended International Non-proprietary Names (rINN) andNew British Approved Names (BAN) and follow EuropeanDirective 92/27/EEC. Amphetamine has not yet followed theconvention. We are therefore following this inconsistent rule ofspelling to help people using electronic retrieval. Trade rather thangeneric names of drugs have been used to distinguish the differentpreparations conveniently.


Two other extended-release stimulants, Ritalin-SRand Focalin XR (dexmethylphenidate extended re-lease) will not be covered in this review. Ritalin-SR islikely to be replaced in the market by the more clin-ically effective Ritalin LA. Focalin XR is currently notavailable in Europe.

Adverse effects and misuse of stimulants

All stimulants, whether immediate- or extended-re-lease, share some disadvantages. They can be bothabused by the patient and diverted to an illegal rec-reational market. They, therefore, represent a poten-tial public health risk. They are sometimes ineffective.The adverse effects (insomnia, appetite loss, growthretardation, abdominal pain, and occasionally dys-phoria, tics or agitation) can sometimes be unac-ceptable. The issue of long-term growth retardationhas received much attention recently but the issue iscomplex and current data are conflicting [70, 78] anddo not allow the effects to be determined definitively.Nevertheless, the group recommend that childrensheight and weight should be monitored as a matter ofcourse and followed on growth charts. Stimulants arecontraindicated in several circumstances, most ofthem uncommon in childhood: schizophrenia, severedepression, hyperthyroidism, cardiac arrhythmias,moderate to severe hypertension, angina pectoris,glaucoma, previous hypersensitivity, or concomitantuseor use within the last 2 weeksof monoamineoxidase (MAO) inhibitors. Caution is advised in: pa-tients with motor tics, patients with known drugdependence or history of drug dependence or alco-holism, pregnancy and breast-feeding, anorexianervosa, or a history of suicidal tendency. People withpre-existing cardiac abnormalities may be at specialrisk. Strategies for dealing with side effects includemonitoring, dose adjustment of the stimulant,switching medication, and adjunctive pharmacother-apy to treat the side effects (see [71]).

Stimulants were introduced at a time when drughazards were less rigorously monitored than is nowthe case. Strattera has been developed under strictersurveillance, and as a result some uncommon adverseevents have been notified to regulatory authorities (seebelow). This in turn has led to a need to reconsider thepossibility that stimulants too could have relation-ships with cardiac arrhythmias, liver failure, suicidalthoughts and seizures. For example, an unpublishedreport provided to a panel of the (US) Food and DrugAdministration described seven reported cases ofsudden death in children taking methylphenidate,relative to about 10 million prescriptions. In presentknowledge, we think it is unlikely that this is greaterthan the rate in the untreated population, but also thatthe true figures of adverse reactions are probably

greater than reported. The main practical implicationsare to include physical examination before prescrip-tionto seek for cardiovascular abnormalities such asraised blood pressure or heart murmurs (ECG is op-tional) and enquire about symptoms such as syncopeon exercise, with cardiological evaluation if a warningsign is found; and to include enquiry about seizures,signs of liver damage, or heart irregularities in moni-toring patients. These recommendations are the sameas those for immediate-release drugs, and do not differfor different stimulants or preparations. Routine bloodtests are not recommended by us (though they are bythe companies); blood dyscrasias can occur as a rareevent, as with most medicines, but there is no sug-gestion either that the risk is big enough to justifymonitoring or that it could be detected by infrequenttests.

Stimulants are controlled substances. Nevertheless,a meta-analysis of long-term naturalistic studies ofstimulant treatment in ADHD concluded that stimu-lant treatment does not increase the risk of substanceabuse [79].

Limited information is available regarding theinappropriate use of stimulants in patients withADHD. Data on different forms of misuse (e.g., forcognitive enhancement by healthy people) anddiversion of stimulants are few. Recently, Wilens et al.[77] evaluated the prevalence and correlates of stim-ulant diversion and inappropriate use in young NorthAmerican adults with ADHD using a self-reportquestionnaire. It was based on a relatively small cross-sectional sample. They found, rather alarmingly, thatout of 98 subjects with ADHD who were takingmedication, 11% said that they had sold their pre-scribed medication, and 22% had misused medi-cation in the sense of escalating dose without properauthorisation.

The 2004 US National Survey on Drug Use andHealth also found evidence that a minority of patientsmay sell or misuse the medication [53] and a surveyof 10,904 randomly selected college students in the USfound that the life-time prevalence of non-medicalprescription stimulant use was 6.9% [41]. Data forEurope are lacking. Little evidence was found ofmisuse/diversion in subjects with ADHD who did notalready have substance abuse or conduct problems.All of the medications misused/diverted were imme-diate-release preparations of stimulants, suggestingthat extended-release stimulants are less likely to beused inappropriately or diverted. This in turn sug-gests that IR formulations of MPH should not beprescribed in drug abusers and high-risk populationsas first choice.

In some European countries, the perceived risk ofdrug abuse has led to stimulants being forbidden onpublic health grounds. We considered that this is a


disproportionate response, causing disadvantage topatients without likely impact on rates of substanceabuse.

Non-stimulants

j Atomoxetine hydrochlorideStrattera

Pharmacology: ATX, brand name Strattera, is aselective noradrenaline transporter blocker. It iscurrently licensed in several countries for the treat-ment of ADHD in children over six and adolescentsand adults who were treated as adolescents. It ismetabolised by the cytochrome enzyme P450 2D6;people with low levels of activity of this enzyme me-tabolise the drug more slowly and may well needlower doses. Drugs inhibiting P450 2D6 (such as flu-oxetine)2 will lead to higher blood levels of Stratteraand may call for dosage adjustment.

Time course of action: The full effect of the medi-cation appears only after 68 weeks of treatment orlonger; but responders usually show some change bythe 4-week point. Once full clinical effectiveness isestablished this appears to persist across the day at afairly consistent level.

Clinical aspects: Like the stimulants, Strattera isindicated as an integral part of a multimodal treat-ment programme for patients with ADHD. Strattera isavailable in 10, 18, 25, 40, or 60 mg capsules (for oraladministration with or without food). In children andadolescents up to 70 kg in body weight, Stratterashould be initiated at a total daily dose of approxi-mately 0.5 mg/kg and increased after a minimum of7 days to a target daily dose of approximately 1.2 mg/kg administered either as a single daily dose in themorning or as evenly divided doses in the morningand late afternoon/early evening. However, a singleevening administration produces lower efficacy andless side effects [26]. The total daily dose in childrenand adolescents should not exceed 1.8 mg/kg or100 mg, whichever is less.

j ModanilProvigil

Pharmacology: This is a chemically distinct drug,established for the promotion of wakefulness, withnon-dopaminergic activating action on frontal cortex.

Time course of action: The effect is apparent fromabout 1 week of therapy.

Tablets are 100 mg. The dose is around 350 mgdaily for children up to the age of 12.

Clinical aspects: Modafinil does not (at the time ofwriting) have a marketing license for ADHD. Theguidelines group does not have sufficient experiencewith it to make detailed recommendation, but it couldpotentially be considered for children refractory toother therapies.

Adverse effects of non-stimulants

A direct comparison of the adverse effects of Stratteraand extended-release stimulants is not really possibleon the basis of published data because Strattera hasbeen introduced at a time of closer regulation andmore explicit reporting; while stimulants have beenused for much longer period with correspondinglygreater chance of rare hazards appearing. For Strat-tera common adverse effects include nausea, sedationand appetite loss; other side effects of Strattera in-clude dry mouth, insomnia, constipation, and moodswings. Additionally, urinary retention and sexualdysfunction have been observed in adult patients.Most of these adverse effects diminish over the firstmonths of treatment. There is no significant differ-ence in adverse events between normal and poormetabolisers of Strattera.

Regulatory agencies, such as the MHRA (Medicinesand Health Care Products Regulatory Agency) in theUK, collect spontaneous reports about problems thathave arisen in people treated with drugs. It can,however, be hard to know whether such problemswould have arisen even if the drug had not been gi-ven. The most common serious events reported havebeen seizures: up to May 2005, there had been 212recorded, from about 2.2 million treated, and thefigure is likely to underestimate the rate. Many,however, were reported from people who were alreadyprone to seizures, or were taking other drugs that cancause fits. The company reckoned that only 4 of the212 events had no obvious cause other than Strattera,and so it is not clear whether the drug can causeseizures. We recommend caution: enquire about sei-zures before prescribing, monitor seizure frequencyin those with epilepsy, and be alert to the possibilityof seizures appearing for the first time.

Suicidal ideation appears rarely0.44% of peopletreated in studiesbut this is significantly more fre-quent than in those given placebo (0%) [4], so pre-scribers should be aware of this association andprepared to detect depression and suicidality and totreat, or refer, if necessary. Similar data for stimulantsare currently lacking. For Strattera, irritability, agi-tation, anger or aggression may also occur in rarecases.

The warnings given for serious idiosyncratic he-patic events are similar to those for stimulants.Strattera therapy should be discontinued in patients

2Useful websites with fuller information about these interactionscan be found at http://medicine.iupui.edu/flockhart/table.htm or athttp://www.medscape.com/druginfo/druginterchecker?src=ads


who present with signs or symptoms of liver injurysuch as jaundice, dark urine, or unexplained, persis-tent flu-like symptoms, and these should be askedabout. Patients who develop symptoms of hepaticdisease should discontinue Strattera. Chemical mon-itoring of hepatic function is, however, not recom-mended [51]. Blood pressure can be increased andshould be monitored throughout therapy. The MHRAhas received reports of QT interval prolongationnotenough to establish a link, but with particular concernafter overdose. We do not think that there is a needfor routine ECG monitoring, but recommend it forpeople already known to have long QTc intervals (or afamily history), or who are also receiving drugs thatcan also produce QT prolongation, or cause electro-lyte disturbances, or inhibit cytochrome P450 2D6.

Unlike stimulants, preliminary data on Strattera donot show any potential for abuse or long-term effectson growth [64]. The following side effects have beenreported for Modafinil: insomnia, headache, nauseaand anorexia as common adverse effects; skin rashes,including erythema multiforme and Steven-Johnson-syndrome as less common ones. It is contraindicatedin hypertension and heart failure.

j Costs of medications

Table 1 indicates the current prices of drugs for allavailable doses, based on those for the UK andGermany. These data should be seen only as aguide; the excess costs of extended-release overimmediate-release may be less when other factorsare taken into account, such as the costs of in-school dosing for short-acting medications given inschool hours [38].

j Recommendations for use of long-actingformulations

Should extended-release preparations be used?

The group was clear that there are benefits for at leastsome children from preparations that can be givenonce daily. They can be seen as superior to placeboand some are equivalent to multiple doses of imme-diate-release methylphenidate (see Part II for evi-dence for this and other recommendations; theguidelines group regard this as established at Grade Aaccording to the SIGN system given in Appendix 1).The main countervailing argument is cost. This iscertainly a consideration for all publicly funded ser-vices and some private practices. In reviewing thepublished literature, NICE found little difference inefficacy between multiple doses of immediate-releasemethylphenidate and single doses of long-acting for-mulations [30]. They also found no good evidence toestablish difference in efficacy between extended re-lease methylphenidate formulations and Strattera interms of the reduction in core symptoms althoughthis view is challenged by recent meta-analysis, sup-ported by our review below, showing a significantlygreater effect of stimulants [15]. This evidence shouldnot be considered conclusive given limitations in theevidence base and dearth of well-designed head-to-head trials. The relative cost-effectiveness of thestimulants is, therefore, largely but not solely deter-mined by the price of the drugs in the absence of gooddata on other indicators such as those relating toQuality of Life (QoL), while efficacy may become anissue when non-stimulants are considered. In currentpractice, dexamfetamine is substantially the leastexpensive; but in our view this would not be preferred

Table 1 Indicative annual cost of licensed drugs (February 2006)

Generic Proprietary UK cost per tabletper year (British NationalFormulary)

Germanya cost per tabletper year (Net wholesaleprice including tax)

Methylphenidate hydrochloride Equasym 30 mgm/day = 273 30 mgm/daya = 390 Ritalin 30 mgm/day = 345 30 mgm/day = 577 Medikinet Not available 30 mgm/daya = 390 Concerta XL 36 mgm/day = 666 36 mgm/day = 1.072 Equasym XL 30 mgm/day = 867 Not availableMedikinet retard Not available 30 mgm/day = 897

Dexamfetamine sulphate Dexedrine 15 mgm/day = 113 Not availableAtomoxetine Strattera Once daily = 1.068 Once dailyb = 1.286

Twice daily = 2.136 Twice dailyc = 2.634)2695

Costs based on UK and German prices. Ritalin LA, which is not available ineither country, costs 447 per tablet per year in Belgiuma German prices for Equasym and Medikinet are based upon 20 mg tablets, forRitalin on 10 mg tablets (20 mg not available). Corresponding prices for 10 mgtablets Equasym or Medikinet are 514

b 40 or 60 mgm/dayc Depending on doseNB prices vary from time to time, do not include costs of monitoring andcotherapies, and are not the major decision in prescribing. The doses cited arenot necessarily equipotent


over MPH because of its perceived presence in theillegal drug scene and by the lack of good evidence onsafety (Good Practice). This aspect needs furtherstudy. Excluding dexamfetamine, the next most cost-effective is immediate-release methylphenidate. Theguidelines group recommend this as the drug of firstchoice for the majority of patients when cost is aconsideration. The extra cost of extended-release ishard to give precisely; it depends on pricing policiesin different countries that may change from time totime. NICE economists attempted a calculation of thecost for units of gain in the QoL of patients beingtreated with Concerta XL, Equasym XL and Strattera[30]. In general, the costs for available treatments fallwithin the range that European society accepts inmedical treatment, and correspondingly all threedrugs types should be available in treatment regimes.Calculations such as this do not give guidance for theclinician in the individual case. For one thing, they areheavily dependent upon very questionable estimatesabout both the number of children who will fail tocomply with a regime and the changes achieved inQoL measures.

Drugs with longer action have other advantages,which may vary as a function of the administrativecontext in different EU countriesespecially, thatthey do not have to be taken at school. The schooldoes not have the expense and risk of maintaininggood administration, and the child does not have thepotential stigma of being seen to take a pill. Compli-ance may be correspondingly better, but data are few.ADHD patients on long-acting preparations may bemore likely to persist on their medication than thoseprescribed IR MPH and have significantly feweremergency room visits and general practitioner visitsper patient, on average, over 1 year [28, 34]. Anindividual decision therefore needs to be made in thelight of the childs circumstances. A little evidence(Grade C) suggests that sustained release medicationsmay be less prone to abuse because they tend to havea slower rate of onset of effect than IR medications[73]. Kollins et al. [32] assessed the effects of sus-tained-release methylphenidate (SRequivalent toextended release), immediate-release methylpheni-date (IR), and placebo among ten healthy volunteers.The IR medication produced increased ratings ofpleasant effects. In contrast, the SR formulation pro-duced only transient effects. These results should betaken with caution until good data emerge on abusepotential. In summary, the guidelines group considerthe key advantages of immediate-release stimulants tobe its lower cost and flexibility of dosage; the keyadvantages of long-acting drugs as a potentialreduction of stigma at school, and improved com-pliance and possibly reduced risk of misuse. Theconclusions were therefore (1) that long-acting

preparations should be available and used; and (2)that they should not entirely replace short-actingdrugs. Individual clinical choice will determine thechoice of formulation used.

Which long-acting drug to use?

The comparative merits of the long-acting prepara-tions have been hotly disputed. Part 3 of this paperprovides a detailed review. Although between-studycomparisons of effect sizes are useful, they must beconsidered cautiously because the trials being com-pared differed not only in the type of active medicationemployed but also in other design features that mayhave affected the results (e.g., outcome measures,raters, doses, experimental design). There are sometrial data based on direct comparisons of Concerta XLand the other drugs. These include data against Equ-asym XL [70] and Strattera [76]: The first highlightingthe different profile of effect across the day and thesecond suggesting that Concerta XL had a significantlylarger effect. This latter trial did, however, excludechildren who previously had a poor response tomethylphenidate. This was for sound ethical rea-sonsit would scarcely have been justifiable torandomise them to a drug known not to help them. Butthe effect could well have been to bias the comparisonin Concerta XLs favour; and a subgroup analysis ofchildren who had not previously received stimulants(the key group for clinicians choosing the first treat-ment) yielded no significant differences between thepreparations. A recent comparison of Adderall XR andStrattera suggested that the mixed amphetamine saltsof Adderall XR had, in the doses administered, thelarger effect on a hyperactivity rating scale [76]. In linewith meta-analysis [15], the review reported heresuggests that the effectiveness of Strattera is likely tobe somewhat less than that of stimulants generally, onthe basis of the data provided by the companiesbutthe studies are heterogeneous in design, doses andsubjects treated. In terms of normalisation rates andnumbers needed to treat, differences are less obvious.Qualitative as well as quantitative considerations,therefore, come into the choice of a long-actingpreparation. For instance, Strattera can have an actionthroughout the day and this may be welcome to fam-ilies who may struggle in mornings and evenings [27].It helps some children who are not helped by meth-ylphenidate (and vice versa).

The balance of adverse effects is different forstimulants and non-stimulantsStrattera is morelikely to produce somnolence, methylphenidateinsomnia; Strattera nausea, methylphenidate anorex-ia. Strattera may, therefore, be particularly usefulwhen adverse effects have limited the value of astimulant. There is possibly a lower abuse potential


for Concerta XL related to properties inherent in theOROS technology compared to other MPH-ER for-mulations (although further research is needed to testthis, see above)3 However, Strattera may be consid-ered as a first choice in populations at risk. It does nothave a greater abuse potential than desipramine [24].

Both Strattera and extended-release preparationsof methylphenidate should be available. Adderall XRhas not been considered in detail in this guidelinebecause of its unlicensed status in Europe. The choicewill depend upon the circumstances:

If treatment is starting with a long-acting prepa-ration, then considerations4 of the anticipated averagecost/dose and efficacy suggest starting with extended-release methylphenidate and proceeding to Stratteraif the former is unhelpful (Grade B for comparativeeffect). The choice of extended-release methylpheni-date will depend upon the profile of action requiredover time, and upon the availability of drug. Stratterais a relatively expensive drug; so a cost-effective pol-icy will usually reserve it for cases where methyl-phenidate is unhelpful or contraindicated. It maynevertheless be preferred as first choice if substanceabuse or comorbid tics are a problem, if there is astrong family preference for a non-stimulant, or if a24-h action is particularly strongly required [27].

If a child has responded well to immediate-releasemethylphenidate, there may still be reasons to shift toextended-releasefor example, to avoid the stigma orinconvenience of repeated dosing. Clearly, an ex-tended-release preparation of methylphenidate willthen be preferred.

If a child has suffered adverse effects on immediate-releasemethylphenidate, then the next stepwill often beto proceed to Strattera (Grade B for difference in ad-verse effect profile; see above). The previous EuropeanGuidelines paper outlines symptomatic actions to betaken to limit the impact of adverse effects [71]. Forinstance, growth effects are not sufficiently differentbetween drugs that one rather than another shouldroutinely be preferred on those grounds. But, if anindividual childs growth is decelerating on stimulantsand cannot be managed conservatively then a switch toStrattera is a rational move.

If a child has failed to respond to immediate-releasemethylphenidate because of lack of efficacy rather

than adverse effect, then the next option to try eitherimmediate release dexamfetamine if available andacceptable (see above and previous guidelines [71] orStrattera). If dexamfetamine fails, or is unacceptableor unavailable, or if the presence of substance misuse,or family choice, contraindicates dexamfetamine,then Strattera should be considered. Grade B evidencesuggests that methylphenidate non-responders havearound a 40% chance of responding to Strattera [43].

Which extended-release preparation ofmethylphenidate?

Clinical practice will often begin with a decision aboutprofile of effect desired across the day. Loss of controlin the evenings or preference for evening control bythe family will indicate the longer acting; insomniamay suggest the shorter-acting one. The smaller initialeffect of Concerta XL may mean that morning controlis achieved at the price of greater overall exposure tomethylphenidate compared to Equasym XL, RitalinLA or Medikinet Retard. Where this is an issue (e.g.,for the family) then Equasym XL, Ritalin LA orMedikinet Retard would be preferred. An alternativestrategy would be to combine Concerta XL with amorning dose of immediate release (see below). Thereis, however, a range of individual variation. EquasymXL is somewhat more expensive (at the time of writ-ing) but has the property, along with Medikinet Re-tard and Ritalin LA, that the capsules can be openedand the contents sprinkled on food. This helps somechildren who have difficulty swallowing; but theopportunity to administer it surreptitiously shouldnot be taken because of the potentially harmful effectsof stealth on trust.

Should combination of treatments be used?

Extended-release preparations were designed to give agood profile of action across the day. Nevertheless,the range of individual variation suggests that somechildren on extended-release formulations may behelped by an additional dose of immediate-release(beginning with 5 mg); either in the morning to assistwith control at the beginning of the day; or in theevening to prevent an unsettling loss of action (whichmay manifest either in disruptive behaviour or ininitial insomnia).

j Other considerations

Comorbidity

Little evidence exists of the specific advantages ordisadvantages of short-acting versus long-actingtreatments in relation comorbidities. In general, more

3However, the IR MPH overcoat of the capsule could still be mis-used.4Since these are determined by price, and price will vary from timeto time and place to place, the prescriber is advised to check thesegeneralisations against current circumstances. When consideringthe total medication costs there is more than just the cost of one pillto consider. Many patients taking one medication require longercoverage, top-up doses and medication for comorbidities or sideeffects. One should not choose a long-acting medication purelybased on price.


RCT data is available for Strattera than other formu-lations.

Tics: Methylphenidate is not necessarily contrain-dicated in people with tics although in individualcases it can be a hazard [47, 72]. Responsive discus-sion with the family is useful to determine the utilitiesthey attach to tic severity against severity of ADHD.In the case of treatment-emergent tics or comorbidtics Strattera is a useful alternative and does notworsen tics and may even improve them [2].

Anxiety: Patients with ADHD and comorbid anxi-ety or disruptive behaviour disorders have as robust aresponse of their ADHD symptoms to stimulants asdo patients who do not have these comorbid condi-tions [45]. A 12-week double-blind, placebo-con-trolled trial [67] found that Strattera significantlyreduced symptoms of both ADHD and anxiety rela-tive to placebo, showing the drug to be efficacious inthe treatment of both conditions with a moderateeffect size (0.5) for anxiety.

Oppositional defiant disorder: A meta-analysis of28 studies found that stimulant effects for aggression-related behaviours in ADHD had effect sizes similar tothose for the core symptoms of ADHD but these ef-fects were smaller for patients diagnosed with conductdisorder [9]. Extended release methylphenidate for-mulations [63, 74, 80], Adderall XR [62] and Strattera[25, 46] have been shown to be effective at treatingcomorbid cases. Data for Strattera are conflicting withregard to effects on symptoms of ODD themselves,with an effect-size of 0.39 in the US-study [46] and alack of significance in a European study [3].

Depression: Preliminary data suggest no effect ofStrattera on depression (unlike stimulants for whichit is a recognised though uncommon complication).Strattera should not be regarded as an anti-depres-sant.

Autism or mental retardation: Both stimulants andRisperidone can be superior to placebo in treatinghyperactivity/impulsivity when it occurs in peoplewith Autism or mental retardation. Risperidone hasnot been considered elsewhere in these guidelines, asit is not approved generally for the treatment ofADHD, but it may well be considered if autism ispresent too. Pending further research, these combi-nations of symptoms are not a specific indication foreither ER stimulants or Strattera. Due to the potentialfor increased side effects cautious dosing for thesecombinations of symptoms is required.

ADHD across the lifespan

Pre-school ADHD: The diagnosis of pre-schoolADHD is deemed problematic by many cliniciansbut can be made reliable [58]. This is despite per-sisting concerns over the difficulty of distinguishing

ADHD from normal developmental variations inactivity and attention [59]. Given concerns aboutthe lack of evidence relating to the long-term effectsof very early initiated extended MPH treatment,drugs should be prescribed very cautiously andevidence-based psychological approaches (such asstructured parent training) should normally takeprecedence in the first instance. Evidence with re-gard to the efficacy of short acting MPH formula-tions was described in the revised Europeanguidelines [71]. While MPH can be efficacious forpre-schoolers, RCT evidence is sparse and whereavailable effects on core symptoms were generallysmaller and side effects more pronounced than forolder children [22, 33]. There is currently no evi-dence, (and no evidence was submitted to theGuidelines group) specifically in relation to theefficacy of effectiveness of long-acting stimulant ornon-stimulant drug formulations in pre-schoolADHD.

Adult ADHD: ADHD often persists beyond theadolescent years, with the majority of cases dis-playing either the operationally defined disorder orthe persistence of some symptoms associated withsignificant levels of academic, occupational and so-cial impairment and a high level of psychiatriccomorbidity [5, 13, 16, 19, 50]. Clinical experiencesuggests that there is a significant subset of indi-viduals who benefit greatly from the appropriate useof stimulants and other drug treatments, but datado not allow accurate estimation of the proportionof individuals with ADHD who require long-termmedical treatment. At the time of writing, mostdrug treatments are not licensed for use within theadult population in any European country. Thismeans that current prescriptions are written off-label. The exception is Strattera, which is licensedfor use in adults but only when treatment wasinitiated in childhood or adolescence. Nevertheless,the guidelines group conclude that both stimulantsand Strattera are effective in adults with ADHD andrecommend their clinical use, both for individualswho started treatment in childhood/adolescence, aswell as for individuals receiving a first time diag-nosis of ADHD in adulthood.

In most cases, lack of licensing in the adultpopulation has not come about from failed licensingapplications, but rather from a historical lack ofinterest of treating ADHD in adults from both cli-nicians and drug companies. It is envisaged thatthis situation will change in the near future sinceseveral of the ER preparations are undergoingEuropean trials in adult ADHD samples to dem-onstrate the safety and efficacy levels required bythe licensing agencies. The treatment of adults islikely to remain a specialist interest in the short-


term, although in the medium term an increasingnumber of adult psychiatrists and adult mentalhealth services are expected to incorporate treat-ment of ADHD into their general practice.

The short-term effectiveness of stimulants in adultlife has been demonstrated with meta-analysis ofavailable data showing similar effect sizes to that seenin child and adolescent samples [17]. They concludedthat the degree of efficacy of MPH, in treating adultswith ADHD, was similar to that reported from meta-analyses of the child and adolescent literature, pro-viding assurance for clinicians that the diagnosis andtreatment of ADHD can be validly applied in adult-hood. The one area where different parameters mayoperate is dosage level. Titration to an effective dose isimportant [17]. The most common dose range used inEuropean adult ADHD clinics for IR methylphenidateis 1020 mg taken 35 times daily, with both higherand lower dosing required in individual cases. Oursuggested maximum is 30 mg for each dose. The use ofextended release stimulants has been less extensivelyinvestigated in the adult ADHD population but thesimilarity in treatment effect sizes for IR MPH sug-gests that the guidelines for their use in older childrenand adolescents should be followed for adults. As withIR preparations titration to a clinically effective dose isrequired. One large trial of Concerta XL [6] involved arandomised, 6-week, placebo-controlled, parallel de-sign study of Concerta XL in 103 adult patients withDSM-IV ADHD. Dosage could be titrated up to1.3 mg/kg/day; the average daily dose used in thisstudy was 72 mg, and the maximum 108 mg/day.

For an adult with a long working day andresponsibilities in the evening, or where markedbehavioural symptoms such as irritability andimpulsivity appear once the effects of medicationwear off alternative treatment may be needed.Careful consideration of the timing of reappearanceof symptoms and the PK profile of each drugshould be considered in deciding whether to opt fora supplement of IR medication or a change toStrattera (see below). Similar treatment effect sizesto those seen in children have also been shownwithin available drug trials of Strattera [1, 44, 56].As in children the possible advantages include re-duced potential for drug misuse, extended actionthroughout the day and a different profile of sideeffects and hazards. The usual daily dose for adultsis 100 mg taken once per day. The same argumentsfor and against the use of extended release prepa-rations in children and adolescents also apply toadults. The main difference here is the way that thetreatment options are evaluated since adult patientsare often capable of providing detailed descriptionsof their personal preferences. The main indicationsfor choosing ER preparations include compliance

issues, stigma or inconvenience in the work setting,individual differences in response to drugs withdifferent PK profiles, and concern over potentialdrug misuse. Depending on the profile of symptomsand associated impairments, individuals often ex-press preferenceseither for drugs that have sus-tained effects, such as Strattera or ER preparations,or for IR preparations, which provide flexibility forsome individuals who do not require or desiretreatment all the time. Some individuals havetrouble with the on-off effects of IR drugs andprefer the smoother effects of ER preparations.Others have no trouble regulating their use of IRmedication and prefer the added control that use ofIR preparations gives them for dose and timing ofdose throughout the day. The choice of drugs withdifferent PK profiles increases the chances of find-ing one that is suited to an individual patientssymptom profile and individual clinical require-ments.

A quantitative comparison of long-actingtreatments

j Pharmacokinetics

Laboratory studies suggest a close relationship be-tween PK and pharmacodynamic (PD) properties;PD studies of both IR & ER stimulants suggest thatthe PD profiles closely mirrors both the IR & ERPK profiles. Thus, methylphenidate plasma levels ofdifferent ER-stimulant preparations measured overtime across the day correspond to their profile ofefficacy [21]. Clinical superiority at any point intime is typically achieved by the formulation withthe highest expected plasma MPH concentration[60, 70]. Thus, when given in roughly equivalentdaily doses the relative efficacy of preparations withdifferent dose delivery and PK profiles will varydepending on the time since dosing. For example,Equasym XL was found to be more effective thanConcerta XL early in day and vice versa late in day[60, 70]. Figure 1 illustrates the typical plasma levelprofiles for stimulants considered in this reviewover time as a function of each formulations rela-tive IR and ER proportions. Bioavailability curvesfor MPH IR BID and Medikinet retard were pro-vided by Medice and for Adderall XR by Shire; thecurve for Ritalin LA is based upon Markowitz et al.[39]; curves for Concerta XL and Equasym XL arebased upon Gonzalez et al. [21]. All bioavailabilitycurves were adapted to a common time scale (X-axis: hours) by the authors and serve as a com-parison of the relative bioavailability of the for-mulations over time, but direct comparisons


between the different preparations should not bemade for absolute plasma levels (Y-axis) as theselevels are dose-dependent. Moreover, it should benoted that PK profiles may show considerable inter-individual variation [29]. Caution should be ob-served when generalising from these aggregatedprofiles to individual patient cases.

Whilst these different delivery profiles do not dis-tinguish any one preparation as better than theother, they do provide the clinician with increasedoptions when choosing which preparation to use for aparticular patient. They also allow clinicians to pro-vide a more flexible and sensitive individualisedadjustment whilst retaining the benefits of an ERpreparation. In much the same way as it is possible toadjust individual doses of IR MPH to ensure thatsymptoms are adequately treated across the day, it isnow also possible to choose an ER preparation whosePK, and thus PD, profile, matches the patients needsand preferences.

j Review methodology: rationale and procedure

Ideally, estimates of the relative efficacy of the differentlong-acting formulations would be based upon pub-lished high quality head-to-head comparisons. How-ever, such direct comparative studies are largelylacking and those available rarely allow to firm con-clusions about the comparative efficacy of the productsto be drawn (e.g., because of exclusion of prior MPH-responders or too short study duration). Therefore, thegroup decided to compare the various efficacy mea-sures of the different available products across studies,by using a common set of standard efficacy indices:

treatment effect sizes (a measure of change) andnumbers needed to treat (a measure of outcome) (seebelow). A working group (TB, DC, SP, AZ) made asystematic review of published clinical trial literaturefor the extended release preparations and unpublishedclinical trial data provided by the companies. The re-viewonly included randomised, double-blind, placebo-controlled studies not excluding medication non-re-sponder a priori and presenting the means and stan-dard deviations of either change scores or endpointscores for both medication and placebo groups.

j A comparison of treatment effect sizes

The calculation of effect sizes standardises the magni-tude of the difference between outcomes in drug andplacebo groups so that a 1-point difference indicatesthat the active treatment and placebo groups differ by 1standard deviation on a particular outcome measure.This allows a direct comparison of treatment effec-tiveness across studies. A commonly used effect sizeindex is the standardised mean difference (SMD)thedifference in outcome scores between drug and placebogroups divided by the pooled standard deviation (of theplacebo and medication group at end of treatment).Ideally, different medications should be compared bycalculating the SMDs based on the same outcome scoreand the same type of rater for each medication. How-ever, study design parameters, outcome measures andthe type of rater often varied among the double-blind,placebo-controlled studies identified for review. Com-paring effect sizes between studies is questionable if thestudies differ substantially in design features thatmightinfluence medication versus placebo differences (e.g.,

Table 2 Mean medication effect sizes (SMDs) on total ADHD symptoms by preparation and type of rater

Parent Teacher Clinician

Total ADHD

SMD Number ofstudies (ratingscales used)


SMD Number ofstudies (ratingscales used) References

Adderall XR 0.9a* 1 1.1b* 1 1.2c** 1 Data on file ShireConcerta XL 1.0d 1 1.0d 1 Wolraich et al. [80]Equasym XL 0.6f 2 0.9d,f 1 1.8e*** 1 Greenhill et al. [23],

Swanson et al. [70],Findling et al. [18]

Medikinet retard 1.0g 1 1.0g 1 0.9e,c*** 1 Dopfner et al. [10, 11]Ritalin LA 1.0h 1 Biederman et al. [7]ATX 0.7i 6 0.7c 11 Data on file Eli LillyModafinil 0.6c 3 0.7c 3 Data on file Cephalon

Rating Scales: a Conners Global Index Scale-Parent Version (CGIS-P) (10 item);b Conners Global Index Scale-Teacher Version (CGIS-T)(10 item); c ADHD-RatingScale; d Inattention/Overactivity with Aggression (IOWA) Conners Scale;e Swanson, Kotkin, Alger, M-Flynn and Pelham (SKAMP) Scale; f Swanson, NolanAnd Pelham (SNAP) IV Questionnaire; g Fremdbeurteilungsbogen-HKS (Germanversion of the ADHD-Rating Scale); h Conners ADHD/DSM-IV ScaleTeacher

Version (CADS-T) (27 item); i Revised Conners Parent Rating Scale (27 item)*Only children studied**Only adolescents studied; a non-linear dose-response relationship betweendose and response was observed (20 mg: 1.2, 30 mg: 0.9, 40 mg: 0.9)***Lab-school study


different types of rater, durations of studies, dosingregimens). Therefore, SMDs for the various ER medi-cations were calculated separately for different types ofrater (parent, teacher, and clinician) and for the fol-lowing outcome measures: total ADHD symptoms,inattentive symptoms, and hyperactive/impulsivesymptoms. This approach was designed to help controlfor potentially confounding factors across studies.

If a study presented data on more than one fixeddose, the most efficient dose (which was usually thehighest dose) to reduce total ADHD symptoms wasused for the calculation of SMDs. As an exception,a non-linear dose-response relationship betweendose and response (SMD) was observed for AdderallXR efficacy in the clinicians ratings of total ADHDadolescents (20 mg: 1.2, 30 mg: 0.9, 40 mg: 0.9).Where there was more than one study dataset

available, SMDs were averaged to a correspondingoverall SMD. Averaging was done by weighting thesingle studies SMDs by the corresponding numbersof subjects included in the study. Company calcu-lations using a standard approach were employedfor Medikinet Retard, Strattera, Adderall XR, Mod-afinil. The authors calculated the SMDs for ConcertaXL based upon the results of a randomised, double-blind placebo-controlled study from a publishedpaper [80]; two other randomised, double-blindplacebo-controlled studies on Concerta XL wereidentified by the systematic review [48, 69]; theseincluded MPH-responders exclusively (i.e., childrenwho previously had a prior poor response to MPHwere excluded) and their SMDs were substantiallyhigher than those of studies investigating unselectedsamples. In order to ensure comparability, results

Table 3 Mean medication effect sizes (SMDs) on inattention symptoms by preparation and type of rater (n = number of studies)


Inattention



SMD Number ofstudies (ratingscales used) References

Adderall XR 1.2a* 1 Data on file ShireConcerta XL No data availableEquasym XL 0.7d 2 0.8b,d 1 Greenhill et al. [23],


Medikinet retard 1.0e 1 1.1e 1 0.8c** 1 Dopfner et al. [10, 11]Ritalin LA 0.9f 1 Biederman et al. [7]ATX 0.6a 10 Data on file Eli LillyModafinil 0.6a 3 0.6a 3 Data on file Cephalon

Rating Scales: a ADHD-Rating Scale; b Inattention/Overactivity withAggression (IOWA) Conners Scale; c Swanson, Kotkin, Alger, M-Flynn and Pel-ham (SKAMP) Scale; d Swanson, Nolan And Pelham (SNAP) IV Questionnaire;e Fremdbeurteilungsbogen-HKS (German version of the ADHD-Rating Scale);

f Conners ADHD/DSM-IV ScaleTeacher Version (CADS-T) (27 item)*Only adolescents studied; SMD at 40 mg was 1.8**Lab-school study

Table 4 Mean medication effect sizes (SMDs) on hyperactivity/impulsivity symptoms by preparation and type of rater (n = number of studies)


Hyperactivity/ImpulsivitySMD Number of studies

(rating scales used)SMD Number of studies

(rating scales used)SMD Number of studies

(rating scales used) References

Adderall XR 1.1a* 1 Data on file ShireConcerta XL No data availableEquasym XL 0.6d 2 0.9b,d 1 Greenhill et al. [23],


Medikinet retard 1.0e 1 1.1e 1 0.7c** 1 Dopfner et al. [10, 11]Ritalin LA 0.9f 1 Biederman et al. [7]ATX 0.6a 10 Data on file Eli LillyModafinil 0.5a 3 0.6a 3 Data on file Cephalon

Rating Scales: a ADHD-Rating Scale; b Inattention/Overactivity with Aggression(IOWA) Conners Scale; c Swanson, Kotkin, Alger, M-Flynn and Pelham (SKAMP)Scale; d Swanson, Nolan And Pelham (SNAP) IV Questionnaire; e Fre-mdbeurteilungsbogen-HKS (German version of the ADHD-Rating Scale);

f Conners ADHD/DSM-IV ScaleTeacher Version (CADS-T) (27 item)*Only adolescents studied**Lab-school study


from these studies were not considered for thecalculation of the SMDs for Concerta XL. SMDcalculations for Equasym XL are based uponrandomised placebo controlled studies provided byUCB [18, 23, 70].

Tables 24 give mean effect sizes (SMDs) for eachpreparation by the type of rater on ADHD symptomdomains and the numbers of studies and rating scales(indexed) onwhich these figures are based upon.Whilethe various long-acting stimulants differ according totheir PK and PD profiles (see above), their SMDs arestrikingly similar and comparable to the figures re-ported across raters for this class of medication in therecent meta-analysis; no significant difference in effectsize was observed for immediate-release stimulantscompared with long-acting stimulants [14, 15]. SMDsof immediate release methylphenidate on core symp-toms of ADHD in children from randomised, double-blind, placebo-controlled trials are between 0.8 to 1 [14,15, 30]; SMDs vary among symptom domains, with thestrongest effects of stimulant medication on measuresof attention, distractibility, and impulsivity and ob-servable social and classroom behaviour. For adultADHD, Faraone et al. [17] calculated a mean SMD of0.9 based on six trials including a total of 140 MPH-treated ADHD adults and 113 placebo-treated ADHDadults. When treatment was optimised to high doses,the SMD for MPH in adults was 1.3.

Overall, our results support conclusions of the re-cent meta-analysis that long-acting stimulants havesimilar SMDs to immediate-release stimulants (evi-dence level Ia) while SMDs for non-stimulants aresomewhat smaller [14, 15] Although it should beborne in mind that effect sizes cannot providedefinitive guidelines for clinical practice, they doprovide additional information for clinicians to con-sider when planning treatment regimens for ADHDpatients [12, 14, 15]. Limitations of these data includethe inability to control for the effects of potentialconfounding variables (differences concerning thedistribution of diagnostic subtypes, gender, and age,design of study (parallel versus crossover; lab-schoolversus naturalistic setting), type of outcome scoreused (change score versus endpoint score), source ofinformation of the clinicians rating,5 dosing method

(fixed dose versus best dose), and use of placebo lead-in (yes/no).

j A comparison of normalisation rates and numbersneeded to treat

Normalisation rates are defined as the proportion ofpatients normalised, e.g., having no problems morethan mild (i.e. Conners scale T-score < 63 orSNAP < 1/item). Once again these were provided bythe companies, except for Concerta XL, for which noformal Normalisation analysis had been completed.The only published paper reporting normalisationdata was Stein et al. [66], which was not an adequatedouble-blind study (appearance of placebo and OROSdifferent). Thus, these figures cannot be compared tothose of the other preparations. From these normal-isation data, Number Needed to Treat (NNT) can becalculated (see below). The NNT corresponds to theexpected number of patients needed to be treated tosee one patient normalise in terms of ADHD symp-toms with medication and would not have normalisedon placebo therapy. This is a common treatmentefficacy outcome measure in contrast to SMD, whichis a measure of change by treatment. NNT is com-puted as follows:

NNT 1=Percent improved on Drug Percent improved on Placebo 100

Confidence intervals (CI) for NNT are calculated asfollows:

CI 1.96Square rootfnormalisation rate*1 normalisation rate))/n placebog fnormalisation rate*(1 normalisation rate))n treated groupg]

Whilst comparisons are somewhat limited by the useof different measures and slightly different definitionsof normalisation (e.g., SNAP scores 1/item versusADHS-RS T-score 63), figures suggest that differ-ences between the ER medications investigated aresmaller using this metric than using the SMD analysis.An exception is the data for Concerta XL. However,these normalisation data may be inflated because ofan inadequate study design [66]6 and results shouldtherefore not be compared directly to the otherproducts figures straightforward. The relative efficacyfor Strattera using this index also seems stronger. Onepossible explanation for the discrepancy betweenmeasures may relate to a relatively larger efficacy of

5It has to be mentioned that clinician ratings were based on dif-ferent sources of information. The Medikinet retard and EquasymXL studies used direct observations (lab-school ratings). In theStrattera trials, clinical investigators were specifically instructed notto use any other information as the basis for the rating exceptparent interviews (most Strattera studies), respectively teacher re-ports (one Strattera study;75. Weiss M, Tannock R, Kratochvil C,Dunn D, Velez-Borras J, Thomason C, Tamura R, Kelsey D, StevensL, Allen AJ (2005) A randomised, placebo-controlled study of once-daily ATX in the school setting in children with ADHD. J Am AcadChild Adolesc Psychiatry 44:647655).

6The appearance of placebo and Concerta XL was different, i.e. itwas not a blinded study.


ER stimulants than Strattera on more severely im-paired patients (which are not normalised by treat-ment). This possibility needs to be investigatedfurther. (Table 5)

It is worth noting that the effect sizes and NNTs forthese medications compare favourably with those re-ported for other psychiatric drugs. For example, theeffect sizes and NNT for antidepressants in the treat-ment of adult depression or obsessive-compulsivedisorder are in the region of 0.5 and 9, respectively [20,61]; and for atypical antipsychotics in the treatment ofschizophrenia are around 0.25 and 20 [35].

j Quality of life

Quality of Life measurement is a desirable part ofclinical trials, but has not often been applied suffi-ciently frequently, or has been applied with too littlepower [8]. Panels of parents have been asked to de-velop utility measures for cost-effectiveness estimates,but the method has not yet been able to give robustcomparisons of drugs [30]. A systematic review istherefore not included in this paper. The Child HealthQuestionnaire is sensitive to the psychosocial deficitsin ADHD [31] and was used to conclude that Strat-tera is superior to placebo in random-allocationdouble-blind trials (n = 747) [40, 49]. Open labelstudies of Strattera with a total of 921 participants[52] suggest an SMD of around 1.0 SD.especially ona psychosocial summary score, a family activitiessubscale and a parent emotional impact subscale. TheQoL Enjoyment and Satisfaction Questionnaire (Q-LES-Q) was used in a long-term open extension studyof Adderall XR; parents rated a 15% increase after1 year on treatment [37]. A range of specific domainsof functioning that are likely to be related to QoLhave been used to show positive effects of extended-release methylphenidate Concerta XL. These includefamily functioning, parental stress, social interac-tions, academic functioning/homework improvement[48, 65, 69].

Recommendations for future study

The database remains rather limited for all issuesother than the comparison of drugs with placebo onsymptom rating scales. The effects on QoL or utilityratings; the head-to-head comparisons betweendrugs; efficacy in subtypes of ADHD (such ashyperkinetic disorder and various comorbidities);effectiveness studies in real-world settings; andeconomic analyses such as cost-effectiveness; allneed development. Long-term safety assessmentsare pressingly needed. Ta

ble

5Normalisationratesan

dnu

mbe

rsne

eded

totreatby

prep

arationan

dtype

ofrater(95%

CIforNNT)

Normalisationan

dNum

bers

need

edto

treatversus

Placeb

o

Parent

Parent

andteache

rTeache

rClinician(Paren

t)a

Clinician(Teacher)1

% norm

alised

activ

emed

% norm

alised

placeb

o

Num

ber

need

edto

treat

(95%

CI)

% norm

alised

activ

emed

% norm

alised

placeb

o

Num

ber

need

edto

treat

(95%

CI)

% norm

alised

activ

emed

% norm

alised

placeb

o

Num

ber

need

edto

treat

(95%

CI)

% norm

alised

activ

emed

% norm

alised

placeb

o

Num

ber

need

edto

treat

(95%

CI)

% norm

alised

activ

emed

% norm

alised

placeb

o

Num

ber

need

edto

treat

(95%

CI)

MPH

IRb

4120

4.8(

0.15

)64

292.9(

0.16

)Ad

derallXR

c51

253.8(

0.14

)54

365.6(

0.15

)Co

ncerta

XLd

66d

14d

1.9d(

0.20

)Eq

uasym

XLb

3920

5.3(

0.15

)58

293.5(

0.16

)Med

ikinet

retard

e49

122.7(

0.18

)72

262.2(

0.19

)Ritalin

LANoda

taavailable

ATXf

42.3

18.5

4.2(

0.07

)45

.522

.54.4(

0.05

)69

.142

.93.8(

0.21

)Mod

afinil

Noda

taavailable

aClinicianratin

gsba

sedsolely

onpa

rent,respectiv

elyteache

rrepo

rts;

clinical

investigatorswerespecifically

instructed

notto

usean

yothe

rinform

ation

bFind

linget

al.[18]:averag

epa

rent

&teache

rSN

APIV

scores

1/ite

mcSp

enceret

al.[63]:SN

APmeanscores

1/ite

mdSteinet

al.[66

]:AD

HDRS

-IVmeanscore

1/ite

m;(54

mgCo

ncerta

XLversus

placeb

o);a

ppearanceof

placeb

oan

dCo

ncerta

XLwas

differen

t,i.e.n

otablinde

dstud

y,thus

thedifferen

cebe

tweenplaceb

oan

dORO

Sno

rmalisationmay

beinfla

ted;

therefore,

thesefig

ures

may

notbe

directly

compa

rableto

theothe

rprod

ucts

data

eDop

fner

etal.[10];Sinzig

etal.[57]:FB

B-HKS

meanscore

1/ite

mfLilly

prov

ided

tableforno

rmalisationratesba

sedon

alltrials;CP

RSAD

HD

Inde

xT-score

63;AD

HD-RSPa

rent

TotalT-Score

63;AD

HD-RSTeache

rTo

talT-Score

63


Summary and Conclusions

The above review has revealed that long-acting for-mulations are effective in the management of ADHDand are a useful addition to the clinical repertoire.Although there is limited evidence, studies suggestthat ER stimulants are equivalent to multiple doses ofIR. While effect sizes are somewhat smaller forStrattera than ER stimulants, numbers-needed-to-treat are more similar and it is also to be consideredan effective treatment of ADHD. Time course for thedifferent formulations differ. Hazards differ for thetwo classes of drugs. Abuse potential appears to below for all the drugs, with Strattera least likely anddexamfetamine most likely licensed drugs to be abu-sedbut more research, especially in relation toadolescent diversion and misuse, is required. Treat-ment in adulthood is appropriate but great caution isrequired in relation to pre-school children. Costs varyand cost-effectiveness calculations are not yet ade-quate to guide practice.

More specifically it is recommended that

Long-acting preparations should be available andused. They should not replace short-acting drugs (whichwill be the initial treatment for many children forreasons of cost and flexibility of dosing). Individualclinical choice is needed.

Both ATX (Strattera) and extended-release prepa-rations of stimulants should be available. Thechoice will depend upon the circumstances:

If treatment is starting with a long-acting prepara-tion, then start with extended-release methylphe-nidate and proceed to Strattera if the former isunhelpful. The choice of extended-release methyl-phenidate will depend upon the profile of actionrequired over time, and upon the availability ofdrug. Strattera may nevertheless be preferred asfirst choice if substance abuse or comorbid tics area problem, if there is a strong family preference fora non-stimulant, if a 24-h action is particularlystrongly required or if there is comorbid anxiety.

If a child has responded well to an immediate-re-lease stimulant, there may still be reasons to shift toextended-releasefor example, to avoid the stigmaor inconvenience of repeated dosing. An extended-release preparation of methylphenidate will then bepreferred.

If a child has suffered adverse effects on immediate-release methylphenidate, then the next step willoften be to proceed to Strattera.

If a child has failed to respond to immediate-releasemethylphenidate, because of lack of efficacy ratherthan adverse effect, then the next option is to trydexamfetamine or Strattera depending on the rel-ative balance of advantages.

j Acknowledgements The authors meeting was funded by equalcontributions from several companies. The authors expenses fortravel were paid but they received no fees. Some had potentialcompeting interests, and these are detailed separately (appendix 2).

Appendix 1

Excerpts from: SIGN 50: A guideline developershandbook Section 6: Forming guideline recommen-dations [55]

These excerpts are included to assist the readerunderstand the grading system used in this guideline.They are not intended as a substitute for the full SIGNguidance and anyone wishing to explore this areafurther should consult the full guidance [55]:

Guideline recommendations are graded to dif-ferentiate between those based on strong evidenceand those based on weak evidence. This judgment ismade on the basis of an (objective) assessment of thedesign and quality of each study and a (perhaps moresubjective) judgment on the consistency, clinical rel-evance and external validity of the whole body ofevidence. The aim is to produce a recommendationthat is evidence-based, but which is relevant to theway in which health care is delivered and is thereforeimplementable.

It is important to emphasise that the grading doesnot relate to the importance of the recommendation,but to the strength of the supporting evidence and, inparticular, to the predictive power of the study de-signs from which that data was obtained. Thus, thegrading assigned to a recommendation indicates tousers the likelihood that, if that recommendation isimplemented, the predicted outcome will beachieved. SIGN grading system

Levels of evidence

1++ High quality meta analyses, systematic reviews ofRCTs, or RCTs with a very low risk of bias

1+ Well conducted meta analyses, systematic reviews ofRCTs, or RCTs with a low risk of bias

1) Meta analyses, systematic reviews of RCTs, or RCTs witha high risk of bias

2++ High quality systematic reviews of case-control orcohort studies, High quality casecontrol or cohortstudies with a very low riskof confounding, bias, or chance and a high probabilitythat the relationship is causal

2+ Well conducted casecontrol or cohort studies with a lowrisk of confounding, bias, or chance and a moderateprobability thatthe relationship is causal

2) Casecontrol or cohort studies with a high risk ofconfounding, bias, or chance and a significant risk thatthe relationship is not causal

3 Non-analytic studies, e.g. case reports, case series4 Expert opinion


On occasion, guideline development groups findthat there is an important practical point that theywish to emphasise but for which there is not, nor istheir likely to be, any research evidence. This willtypically be where some aspect of treatment is re-garded as such sound clinical practice that nobody islikely to question it. These are marked in the guide-line as Good Practice Points, and are indicated . Itmust be emphasised that these are not an alternativeto evidence-based recommendations, and should onlybe used where there is no alternative means of high-lighting the issue.

Potential conflicts of interest

UCB Lilly Janssen/McNeil Medice Shire Cephalon Novartis

A or C Other A or C Other A or C Other A or C Other A or C Other A or C Other A or C Other

Asherson X X XBanaschewski X X X X X X XBuitelaar X X X X X X X XCoghill X X X X X X X XDanckaerts X X X X X X X XDopfner X X X X X X X X XFaraone X X X X X X X X XRothenberger X X X X X X X XSantosh X X X XSergeant X X X X X XSonuga-Barke X X X X X XSteinhausen X X XTaylorZuddas X X X X X X X X

NoteA or C = served in an advisory or consultancy role either personally or for ones employer. This includes advisory boardsOtherPaid public speaking/conference attendance support/unrestricted research support/meeting or conference grants

Grades of recommendation

A At least one meta-analysis, systematic review, or RCT rated as 1++,and directly applicable to the target population; orA systematic review of RCTs or a body of evidence consistingprincipally of studies rated as 1+, directly applicable to the targetpopulation, and demonstrating overall consistency of results

B A body of evidence including studies rated as 2++, directlyapplicable to the target population, and demonstratingoverall consistency of results; or Extrapolated evidence fromstudies rated as 1++ or 1+

C A body of evidence including studies rated as 2+, directlyapplicable to the target population and demonstratingoverall consistency of results; or Extrapolated evidence fromstudies rated as 2++

D Evidence level 3 or 4; or Extrapolated evidence fromstudies rated as 2+

References

1. Adler LA, Spencer TJ, Milton DR,Moore RJ, Michelson D (2005) Long-term, open-label study of the safety andefficacy of atomoxetine in adults withattention-deficit/hyperactivity disor-der: an interim analysis. J Clin Psychi-atry 66:294299

2. Allen AJ, Kurlan RM, Gilbert DL, Cof-fey BJ, Linder SL, Lewis DW, WinnerPK, Dunn DW, Dure LS, Sallee FR,Milton DR, Mintz MI, Ricardi RK, Er-enberg G, Layton LL, Feldman PD,Kelsey DK, Spencer TJ (2005) Ato-moxetine treatment in children andadolescents with ADHD and comorbidtic disorders. Neurology 65:19411949

3. Bangs ME, Hazell P, Danckaerts M,Williams D, Moore RJ, Levine L (2006)Atomoxetine for the treatment ofattention-deficit/hyperactivity disorderand oppositional defiant disorder inchildren and adolescents (accepted fororal presentation). 159th AnnualMeeting of the American PsychiatricAssociation. Atlanta, GA

Appendix 2

Good practice points

jX Recommended best practice based on the clinical experienceof the guideline development group


4. Bangs ME, Tauscher-Wisniewski S,Polzer J, Zhang S, Acharya N, DesaiahD, Allen AJ (2005) Meta-analysis ofsuicide-related events in atomoxetine-treated patients (Poster). 52nd AnnualAACAP Meeting, Toronto

5. Biederman J, Mick E, Faraone SV(2000) Age-dependent decline ofsymptoms of attention deficit hyper-activity disorder: impact of remissiondefinition and symptom type. Am JPsychiatry 157:816818

6. Biederman J, Mick E, Surman C, DoyleR, Hammerness P, Harpold T, DunkelS, Dougherty M, Aleardi M, Spencer T(2005) a randomized, placebo-con-trolled trial of oros methylphenidate inadults with attention-deficit/hyperac-tivity disorder. Biol Psychiatry 59:829835

7. Biederman J, Quinn D, Weiss M,Markabi S, Weidenman M, Edson K,Karlsson G, Pohlmann H, Wigal S(2003) Efficacy and safety of Ritalin LA,a new, once daily, extended-releasedosage form of methylphenidate, inchildren with attention deficit hyper-activity disorder. Paediatr Drugs 5:833841

8. Clarke SA, Eiser C (2004) The mea-surement of health-related quality oflife (QoL) in paediatric clinical trials: asystematic review. Health Qual LifeOutcomes 2:66

9. Connor DF, Glatt SJ, Lopez ID, JacksonD, Melloni RH Jr (2002) Psychophar-macology and aggression. I: a meta-analysis of stimulant effects on overt/covert aggression-related behaviors inADHD. J Am Acad Child Adolesc Psy-chiatry 41:253261

10. Dopfner M, Banaschewski T, Schmidt J,Uebel H, Schmeck K, Gerber WD,Gunther M, Knolke U, Gehrke M,Haler F, Resch F, Brunger M, Ose C,Fischer R, Poustka A, Rothenberger A,Lehmkuhl G (2003) LangzeitwirksamesMethylphenidat bei Kindern mit Au-fmerksamkeits-Hyperaktivitatsstorun-gen. Nervenheilkunde 22:8592

11. Dopfner M, Gerber WD, BanaschewskiT, Breuer D, Freisleder FJ, Gerber-vonMuller G, Gunter M, Hassler F, Ose C,Rothenberger A, Schmeck K, Sinzig J,Stadler C, Uebel H, Lehmkuhl G (2004)Comparative efficacy of once-a-dayextended-release methylphenidate,two-times-daily immediate-releasemethylphenidate, and placebo in alaboratory school setting. Eur ChildAdolesc Psychiatry 13(Suppl 1):I93101

12. Faraone SV, Biederman J (2004) Usingmeta-analysis to draw conclusionsabout ADHD medication effects. 17thECNP Congress. Stockholm

13. Faraone SV, Biederman J, Mick E (inpress) The age dependent decline ofattention-deficit/hyperactivity disor-der: a meta-analysis of follow-up stud-ies. Psychol Med 36:159165

14. Faraone SV, Biederman J, Spencer T,Aleardi M, Pagano C (2003) Using ameta-analysis to draw conclusionsabout ADHD medication effects (Pos-ter). 156th Annual Meeting of theAmerican Psychiatric Association. SanFrancisco, California

15. Faraone SV, Biederman J, Spencer T,Aleardi M, Pagano C, META-ANALY-SIS CTEOMFAU (in press) Comparingthe efficacy of medications for ADHDusing meta-analysis

16. Faraone SV, Biederman J, Spencer T,Wilens T, Seidman LJ, Mick E, DoyleAE (2000) Attention-deficit/hyperac-tivity disorder in adults: an overview.Biol Psychiatry 48:920

17. Faraone SV, Spencer T, Aleardi M,Pagano C, Biederman J (2004) Meta-analysis of the efficacy of methylphe-nidate for treating adult attention-def-icit/hyperactivity disorder. J ClinPsychopharmacol 24:2429

18. Findling RL, Quinn D, Hatch S, Cam-eron S, DeCory H, McDowell M (sub-mitted) Comparison of the clinicalefficacy of twice-daily Ritalin) andonce-daily Equasym XL with placeboin children with attention-deficit/hyperactivity disorder. Data on fileUCB, submitted

19. Fischer M, Barkley RA, Smallish L,Fletcher K (2002) Young adult follow-up of hyperactive children: self-re-ported psychiatric disorders, comor-bidity, and the role of childhoodconduct problems and teen CD. J Ab-norm Child Psychol 30:463475

20. Geddes J, Butler R (2002) Depressivedisorders. Clin Evid 7:867882

21. Gonzalez MA, Pentikis HS, Anderl N,Benedict MF, DeCory HH, Dirksen SJ,Hatch SJ (2002) Methylphenidate bio-availability from two extended-releaseformulations. Int J Clin PharmacolTher 40:175184

22. Greenhill L, Vitiello B, Abikoff H,Kollins S, Wigal S, Swanson J (2004)Outcome results from the NIMH, mul-ti-site, preschool ADHD treatmentstudy (PATS). 51th Annual MCAPmeeting, Washington, DC

23. Greenhill LL, Findling RL, Swanson JM(2002) A double-blind, placebo-con-trolled study of modified-releasemethylphenidate in children withattention-deficit/hyperactivity disor-der. Pediatrics 109:E39

24. Jasinski DR, Faries D, Moore RJ, AllenAJ (2004) Abuse liability assessment ofatomoxetine in a drug-abusing popu-lation. In: CINP; Abstract in Int JNeuropsychopharmacol 7, Suppl. 1,June 2004, p. S310

25. Kaplan S, Heiligenstein J, West S,Busner J, Harder D, Dittmann R, CasatC, Wernicke JF (2004) Efficacy andsafety of atomoxetine in childhoodattention-deficit/hyperactivity disorderwith comorbid oppositional defiantdisorder. J Atten Disord 8:4552

26. Kelsey D, Sutton V, Lewis D, Schuh K,Sumner C, Quintana H (2005) Morn-ing- versus evening-dosed atomoxe-tine: effects on core ADHD symptoms.158th Annual Meeting of the AmericanPsychiatric Association. Atlanta, GA

27. Kelsey DK, Sumner CR, Casat CD, Co-ury DL, Quintana H, Saylor KE, SuttonVK, Gonzales J, Malcolm SK, Schuh KJ,Allen AJ (2004) Once-daily atomoxe-tine treatment for children with atten-tion-deficit/hyperactivity disorder,including an assessment of evening andmorning behavior: a double-blind,placebo-controlled trial. Pediatrics114:e1e8

28. Kemner JE, Wan GJ (2003) Increasedpersistency with once-dailyOROSMethylphenidate: an analysis ofhealthcare claims. 16th US Psychiatric& Mental Health Congress, Orlando, FL

29. Kimko HC, Cross JT, Abernethy DR(1999) Pharmacokinetics and clinicaleffectiveness of methylphenidate. ClinPharmacokinet 37:457470

30. King S, Griffin S, Hodges Z, WeatherlyH, Asseburg C, Richardson G, Golder S,Taylor E, Drummond M, Riemsma R(2004) Methylphenidate, dexamfeta-mine and atomoxetine for the treat-ment of attention deficit hyperactivitydisorder in children (commercial inconfidence information removed);produced by the Centre for Reviewsand Dissemination (CRD). http://www.nice.org.uk/pdf/ADHD_assess-ment_report.pdf

31. Klassen AF, Miller A, Fine S (2004)Health-related quality of life in childrenand adolescents who have a diagnosisof attention-deficit/hyperactivity dis-order. Pediatrics 114:e541e547

32. Kollins SH, Rush CR, Pazzaglia PJ, AliJA (1998) Comparison of acute behav-ioral effects of sustained-release andimmediate-release methylphenidate.Exp Clin Psychopharmacol 6:367374

33. Kratochvil CJ, Greenhill LL, March JS,Burke WJ, Vaughan BS (2004) The roleof stimulants in the treatment of pre-school children with attention-deficithyperactivity disorder. CNS Drugs18:957966


34. Lage M, Hwang P (2004) Effect ofmethylphenidate formulation forattention deficit hyperactivity disorderon patterns and outcomes of treatment.J Child Adolesc Psychopharmacol14:575581

35. Leucht S, Pitschel-Walz G, Abraham D,Kissling W (1999) Efficacy and extra-pyramidal side-effects of the new anti-psychotics olanzapine, quetiapine,risperidone, and sertindole comparedto conventional antipsychotics andplacebo. A meta-analysis of random-ized controlled trials. Schizophr Res35:5168

36. Lopez F, Silva R, Pestreich L, Muniz R(2003) Comparative efficacy of twoonce daily methylphenidate formula-tions (Ritalin LA and Concerta) andplacebo in children with attention def-icit hyperactivity disorder across theschool day. Paediatr Drugs 5:545555

37. Lopez FA, Chandler MC, Biederman J,Spencer T, Mays DA, Michaels MA,Tulloch SJ (2003) Long-term AdderallXR treatment improves quality of lifein ADHD children. 156th AnnualMeeting of the American PsychiatricAssociation, San Francisco, California

38. Marchetti A, Magar R, Lau H, MurphyEL, Jensen PS, Conners CK, Findling R,Wineburg E, Carotenuto I, EinarsonTR, Iskedjian M (2001) Pharmacother-apies for attention-deficit/hyperactivitydisorder: expected-cost analysis. ClinTher 23:19041921

39. Markowitz JS, Straughn AB, Patrick KS,DeVane CL, Pestreich L, Lee J, Wang Y,Muniz R (2003) Pharmacokinetics ofmethylphenidate after oral administra-tion of two modified-release formula-tions in healthy adults. ClinPharmacokinet 42:393401

40. Matza LS, Rentz AM, Secnik K, Swen-sen AR, Revicki DA, Michelson D,Spencer T, Newcorn JH, Kratochvil CJ(2004) The link between health-relatedquality of life and clinical symptomsamong children with attention-deficithyperactivity disorder. J Dev BehavPediatr 25:166174

41. McCabe SE, Teter CJ, Boyd CJ, KnightJR, Wechsler H (2005) Non-medicaluse of prescription opioids among U.S.college students: prevalence and cor-relates from a national survey. AddictBehav 30:789805

42. McGough JJ, Biederman J, GreenhillLL, McCracken JT, Spencer TJ, PosnerK, Wigal S, Gornbein J, Tulloch S,Swanson JM (2003) Pharmacokineticsof SLI381 (ADDERALL XR), an ex-tended-release formulation of Adderall.J Am Acad Child Adolesc Psychiatry42:684691

43. Michelson D (2004) Results

Documents

Long Acting Medications for the Hyperkinetic Disorders