Scribd Logo
Upload

Welcome to Scribd!

  • A 14-Month Randomized Clinical Trial of Treatment Strategies For Attention-Deficit:Hyperactivity Disorder.

    Uploaded by

    gene flowers
    23 views14 pages

    Original Title

    A 14-Month Randomized Clinical Trial of Treatment Strategies for Attention-Deficit:Hyperactivity Disorder.

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    23 views14 pages

    A 14-Month Randomized Clinical Trial of Treatment Strategies For Attention-Deficit:Hyperactivity Disorder.

    Uploaded by

    gene flowers

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 14
    Sa A 14-Month Randomized Clinical Trial of Treatment Strategies for Attention-Deficit/ Hyperactivity Disorder The MTA Cooperative Group Background: Previous studies have demonstrated the short-term ellicacy of pharmacotherapy and behavior therapy forattention-deficivhyperactvity disorder (ADHD), but no longer-term (ie, >4 months) investigations have ‘compared these 2 treatments or their combination, Metheds: A group of 579 children with ADHD Com- bined Type, aged 7 to 0.9 years, were assigned t0 14 months of medication management (titration followed by monthly visits); intensive behavioral treatment (pat= cent, school, and child components, with therapist in- volvement gradually reduced over time); the wo com- bined; or standard community care (treatments by ‘community providers). Outcomes were assessed in mul Liple domains before and during treatment and at treat- ‘ment end point (with the combined treatment and medi- ‘cation management groups continuing medication atall, assessment points). Data were analyzed through intent- to-treat random-clfects regression procedures Results: All 4 groups showed sizable reductions in symp- toms over time, with significant differences among them in degrees of change. For most ADHD symptoms, chil- dren in the combined treatment and medication man- agement groups showed significantly greater improve- ‘ment than those given intensive behavioral reatment and community care. Combined and medication manage- ment treatments did not differ significantly on any di- rect comparisons, but in several instances (oppositional aggressive symptoms, internalizing symptoms, teacher- rated social skills, parent-child relations, and reading achievement) combined treatment proved superior to in- tensive behavioral reatment and/or community care while medication management did not. Study medication stral- egies were superior to community care treatments, de- spite the fact that wo thirds of community-treated sub- jects received medication during the study period, Conclusions: For ADHD symptoms, our carefully crafted medication management was superior to behavioral teat- mentand to routine community care that included medi- cation, Our combined treatment did not yield signifi- cantly greater benefits than medication management for core ADHD symptoms, but may have provided modest advantages for non-ADHD symptom and positive fune- oning outcomes Arch Gen Psychiatry. 1999;56:1073-1086 TTENTION-defieiVhyper- activity disorder (ADHD) occurs in 3% to 5% of school-aged children, ac- counts for as many’ as 30% nonresponders, lack of unimodal compari- sons (medication management vs behav- {oral interventions), and inadequate stais- tical power to examine subject factors that predict treatment response.” Aliso the collaborators and investigators for this study ‘appears on page 1077 ‘Downloaded From: http://jamanetwork.com/pdfaccess.ashy?/url=/data/journals/psych/S107/ (o 50%6 of child referrals to mental health services,'*and results in substantial impair- See also pages 1088 and 1097 ‘ment in peer, family, and academic fune- ‘ning Although benefits of short-term treatments (principally stimulants, behav- {or therapy, and both combined) have been well documented," few controlled studies have examined effectiveness beyond 3 months. Two exceptions? suggested that stimulant effects can persist 1 to 2 years when taken faithfully. However, the gen- cralizability and usefulness of thes les are constrained by sample sociodemo- graphic homogeneity, exclusion of stimulant Given public concerns regarding stimulant treatment,!" wide variations in eatment practices,” and lack of evidence toguidelong-term treatmentsof this chronic disorder, in 1992 the National Institute of Mental Health and Department of Edues- tion cosponsored arandomized clinical trial, the Multimodal Treatment Study of Chil- dren With ADHD (MTA). Iisrationale"and methods!” have been detailed previously The MTA Cooperative Group posed 3 questions: How do long-term medica- tion and behavioral treatments compare with one another? Are there additional benefits when they are used together? What is the effectiveness of system: carefully delivered treatments ¥s routine community care? This report constitutes (©1909 American Medical Association, All ights reserved. ‘a University of Wisconsin -Madison User on 04/29/2017 PARTICIPANTS, MATERIALS, AND METHODS RECRUITMENT PROCEDURES AND SAMPLE CHARACTERISTICS ‘etn atau wide ete ‘The presence of comorbid conditions such asoppositional- isi Goby cok lene Bh eta NIA pul tbl’ ad pe scat ipa tant DESIGN In a 4-group parallel design, children were assigned ran- domly to medication management, behavioral treatment, combined treatment, or community carefor + months. (In themethod articles for this study, treatmentassignments were referred to as medication, psychosocial treatment, combined treatment, and community-ireatmentiassessment and refer- ral. To reflect more accurately the actual treatments, we have changed the terminology forall outcome articles to medi cation management, behavioral ireatment, combined treat- ‘ment, and community care.)Rather than testing fixed single treatments, we designed each MTA treatment armasa man- agement strategy, stich that each was sulicienly robust and flexible to stand on its own and to respond to individual p= tient’ clinical needs throughout the study. Power analyses indicated 24 subjects per treatment con- dition per site (80 families at each of 6 sites) for compari- sons of core ADHD symptoms between any’ 2 treatment arms (critical elfect size, 0-4; power, 0.81, with a 5% 2-tailed. test)" Accordingly, 576subjeets were required: 579 were recruited, Sample demographics, mean scores on standard zed Conners teacher and parent scales," comorbidity profile and impairment ratings (Columbia Impairment Seale) are typical of other ADHD samples this age (Table 3) Randomization was done centrally by the National stitute of Mental Health Data Center, Rockville Md, strati- fied by site in blocks of 16 (4 10 each group) Sealed, or- dered envelopes were sent to sites for successive entries, ‘Treatment assignment was concealed wil the family con- firmed agreement to accept randomization, ASSESSMENTS Principal components analyses narrowed theselection of out Come domate from the comprehensive sesessment ba tery eliminating edundant measures Six major outcome dbomains were dented. Within domains, the measures load ing highest from each informant were selected (1) ADHD !yinptome were measired wih inallention and hyperacivi impulsysubcales of pret an cacher completed SNAP ratings (an acronym denoting the names ofthe Insti tments developers)" (2) oppostiona aggressive sympoms tweremescured witha parent and teacher SNAP oppstional dian disorder subeeale; (3) socal skis were mencared wth 4 parent and leacher-completed subscale from the Soll Suille Rating System (SSRS)" (4 internalizing symptoms (ansiety and depression) were measured with an internal ing subscale from parent and teacher-complted SSRS, and children's sell-atings on the Multidimensional An tty Seale for children (MASC)” (3) parent-child relations ‘were measured with compost sales from parent-child ‘lationship questionaire’ and (6) academic achievement twas measured with 3 subscales from the Wechsler Indi ‘dual Achievement Test (reading, math, and spelling). ‘These measures show acceplable peychomtrc properties, and are reviewed extensively elsewhere" subject were ascesed at baseline, and a 3,9, and M4 months (reatment re pont). Baseline and end-point values forthe outcome dlornsins are presented in Fable 4. “The open parent, teacher, and child ratings for do- amains 1 through 5 were augmented by blinded ratings of School-based ADHD and oppositionaV/aggressive symp- tome using the Abikoff classroom Obseratonal System™ ‘nd social shils and peer relations using peer sociometrc procedures. Finaly, we videotaped parent-child interac tions daring standardized laboratory tasks performed by tates Blind to treatment condition, Videotaped interac tions, parental characterises, amily functioning, cogs tive ability, general impairment, and services te, at tudes and barr wl be reported in ater articles. ‘TREATMENT CONDITIONS AND PROTOCOLS ‘The 3 MTA-deliered treatment strategies were chosen for well-established elfcacy (atleast in the short term), port- ability, sufficient intensity to stand alone, and distinction, from each other. For all 3 arms, subjects had up to 8 2d~ ditional sessions provided when needed to address clini- ‘al emergencies or instances of possible study atrition, Lchavioral Treatment Behavioral treatment included parent training, child focused treatment, and a school-based intervention orga nized and integrated with theschool year. The parent train- fing, based on work by Barkley” and Forehand and MacMahon,” involved 27 group (6 families per group) and. (©1909 American Med Association, All rights reserved. ‘Downloaded From: httpi//jamanetwork.com/pdfaccess.ashx?url=/data/journals/psyet +8/5107/ by a University of Wisconsin -Madison User on 04/29/2017 8 individual sessions pet fly. began weekly on ran- domnizaton, concurrent with biweekly teacher consulta tion; both were tapered over me. The same therapst- Consultant condueted parent training and teacher Consultation, with each therapist-consultant having case load of 12 faies ‘The chld-focused treatment was a suummer treat- sment program (STP) developed by Pelham as therapeu ticsunmer camp. The-week,Sdays-per-wee 9-hours- peralay STP employed intensive behavior interventions Ndmintstered by counsclorsfaides, supervised bythe sae teacher-consultants who performed parent taining and teacher consultation Behavioral interventions were deli cred in group-based recreational settings, and included « point system ved to specific rewards, me ou, socal re Inforement modeling group problem solving sportskils, sndsocalsilsrsining Sumner estment program clase rooms provided individualized academic kil practice and Feinforcement of appropriate classroom behavior. “Theschool-baeedreatmenthad? components 101016 sessions of biweekiy teacher consultation focused on case ‘eniormangenntnicand awl dys) of pariime, behaviorally tained, paraprofessional side working recy with he hd (methods aed fom Swanson”), Theaideshad been STP counselors, andthe po- gam continued nthe all classroom, which helped ogen- tralizeSTP gains to classrooms. Throughou!theschool year, Sally report cat inked home and sol. The dally report Cand ava -pageteacher completed check ofthechlds scceseson peti preselected behaviors and wasbrought home dally by the cild wo be reinforced bythe parent with home-based revards (eg, television ime, snacks). “The treatments outlined above consttue the maxi- smum "dose" of behavioral testnent childrens could reerve, given perfect atlendance and compliance. in prac tice, es (given both behavioral and combined teat tment) attended an average of 77% of parent taming ses Sons and 36.2140 possible STP days. The school component averaged 107 teacher consultation vss and 47.6 dae (of 60 possible) of work with classroom aides ites dif si- fiend inthe extent of atendancefor2 components (pa nt taining and lseoom ade), but there were no sig Cant diferencesbetweenbeivioral and combined eaten, clther within or acrose sites, n degree ofalendancel implemenation for any behavioral component (ables aval able rom the aulhorson request). As deserbed in our com- panion report, a summary measure of attendance! Compliance forall behavioral treatment components was treated to treatment outcomes; ikewise attendance didnot mediate any steXtrestment interactions on outcomes” Consistent with the time-limited invelvement of pro- ders in clinical practice, the involvement of our perzon- delim the delivery of the Behavioral treatments was grad al tapered, with the goal that parents would inreasingh tnanage he cild's behavioral restment.nmost cases thera Pistcontact with parentshadbeen reduced t once-monthly Fessonsor stopped aliogeherpriortoend-pointassesment Medication Management Medication management" started with a 28-day, double blind, daily-switch titration of methylphenidate hydro- chloride, using 5 randomly ordered repeats each of pla- cebo, Sg, 10mg. and 15 or 20mg (higher doses or children >25 kg). Each of the doses listed was given at breakfast andlunch with hal-doc (rounded tothe nearest 5 mg) in ihesflemoon, Crosse easofexperenced clinicansbindy reviewed graphs portraying parent and eaherrlingsfre- sponses tocathof the 3 doses andbyconsenstsscectedeach ciisbes doce, teragreementon best dos, he lind wae broken and the agreed on dose not placebo) became the subject’ nal maintenance dose. For subjects nt ob ingan adequate response o ethyiphendateduring ration, Alermate medications were utrated openly in the following order unt asatsfatory one was found dextroamphearine psmoline,tmipramine, and. necessary othersapproved by eros panel, Of 29 subjece assigned tomediction man agement (= 144) and combined eatment (n= 143) forint {lutraton, 18 adn titation: 17 becase they refised the tniremedialioncomponentand whomovedaway. Ana divonal 1 subjects stare ut didnot complet tration: 4 because ofsie lfc, because of dicaliescerating the {ition procedures and4who suppl inadequatedta, This, $Sosubjects (8809) successfily completed tation; oft, 198 (68, 5%) of289 subjects wereasigned oan individually tuted best dow ofmetyphenidatewthaverage nial dose f30-5imgd.Theremaiingtitaton completers were lther openly trated odextoamphetamine (= 26) because ofun- Sitsfactory methyiphendate response or inially gven no tmedication (n= 32) because of robust placebo response (2 ofthese placco responders alle io cooperate urer after ‘uraton) During hal-hour monthly medication maintenance ve ts, pharmacotherapsts provided suppor, encouragement, talpaccal viet not Beha ener When deemed necessary by the clinician or requested by the par ent readings rom atapprovd ist weresuppled. Alter are- fulrevew oparent-anteacherprovided information, phar ‘tacothenpistconld make agora guided doeadjctnens Of a10 mg/d of methylphenidate (or an equipoten amount AThesbectoastlanoter rg). Adame beyond 210 mg/d could be authorized by acrosesite panel experienced pharmacotherapisi In genera dose edu tions were allowed only to address dose-related side fects. By study end, 212 (73.4%) ofthe 289 subjects given aedicaon management and combined treatment wer be ingsuccesulymainained on methylphenidate, 30 104%) on dextroamphetamine, + (14%) on pemolin, 3 (0%) on imipramine, (0.3%) on bproprion, 1 (0.3%) om hao- Peril and 18 3.1%) on no medication (combined teat fnent [14 subjects] and medication management [4 sub- ject) (1%), wth 20 persistently unmedieated (18 since study outset, 2 during maintenance). There were no di ferences between the mediation management and com- bined treatment groups in the proportion of subjects main- tained on the various medications ‘side effects were monitored monthly—not present, ane, moderate or severe—using the paren-completed 13- item Pitsburg ide Elects Rating Sale," reviewed bythe Pharmacotheraps. tend point 245 combined estnent/ Inedication management fales provide information on $idecflects with some eporing more han one: 88 (33.0%) reported no side elects; 122 (9.8%) reported mild side tilecisonly, 28 114) reported moderate ste fects, nd 7 Gon) reported severesibe effects, These gures may over cstimate side elects, because 6 of 11 reported severe side ellecs (pression, wornying ot iriabiy) could have Been due to nonmedieation factors. Continued of next page (©1909 American Med Association, All rights reserved. ‘Downloaded From: httpi//jamanetwork.com/pdfaccess.ashx?url=/data/journals/psyet +8/5107/ by a University of Wisconsin -Madison User on 04/29/2017 Combined Treatment Combined reatment provided wtmentsoullinedabove formedicaionmansgenentandichral eaten ney Atsatonfollowedby monty sedation aitenace par tntgroupandindivtua seston eae consufaon STP, thdheclasroomalde However, toapponinate ial prac ite, we negated the 2 weatment Modliiesnformnaton ‘vc reulrt shared beeen be tacherconslant anda tiacoherpistand used o guide overall deons Manual {Pe gullies determined land whenanadjustnct none tretictshoudbemade vlrveningfs heater Consequcly,therliodalconbinaon wasnt hn saoofthetw uta estment Cotsen Re iteratre"hyueatmentendpontcombined eatment su sete rocivedoner tll daly dwesiethylpenidate O-2 ‘phan mediatonmanagerent subjects OF 7g). Though Stsdileredsignfcanly nal daly melyiphenda doses (Gang of meat doses erst 30.2-413 mg there were 0 sie ireatment interactions otal daly doses (reament guphslio.peolsice=32P = Ose texnent ou: Fao 12, P= 2) Community Care ‘Community care partcipantsreceived none ofour treatments, but were provided arepor of ther inital study assessments, slong witha st of community mental health resources. They ‘were subsequently reassessed in parallel wth partiipantsin ‘ue 3 treatment arms, At each assessment point, the types of lreatmentsthey obained in the community were documented ‘Mostcommuntycaresubject(n = 07 [6748] received ADHD ‘medications (principally ne ofthe stimulants) romtheirown provider during the 14 months: methyiphenidate(n = 84), pemoline (n= 7), amphetamine (1 =6), tricyclics (n= 6). clonidine/guanfacine (n = 4, and/or buproprion (n = 1) 10. subjects received more than I medication). Inaddltion, 16 of ‘these 7 children were treatedby their physician withanother antidepressant (not counting tieyelics or buproprion). For ‘those treated with methylphenidate, the meantotal daily dose atstudy completion was 22.6 mg. averaging? 3doses per day (530 doses perday for MTA -treated subjects). Information ‘concerning community care psychotherapeutic reatmentshas rot yet been coded and will nat be presented in this article. Fidelity and Compliance The MTA study achieved s high degree of adherence to pro- tocol by crr-arm emphasis on subject rapport, manuals tion ofa tremens regular supervision of pharmacothers sts and peychotherapstsby sled clinician investigators, Eos ate ec werent panel and auton sessions. Good compliance (fled by acceptance anda tendanceat eatment sessions) by patients wih the protocol ‘esata by momthypillcunt, inert saliva ea $rmenfsomonitr akingofmethypenidte and encour agement of amis to makeup mised vss, Only 13(0.0%) of 4 medication management subject and 5 (3%) of 143 Combined ireaimentsubjetfalled iostart medication More ‘emarlably none of 144bshavioral treatment subjecsandonly (0.7%) of 143 combined treatment subject fused beat total reatment, There was nodiferencebetween medication ‘managementorcombined treatment in mediation seston at teedance nor between behavior eatmentatl combined eat tment inthe degree ofbehavioral treatment aceptance oat tendanc. These factors did ot affect the overall ndings,” Ssubjects/families reusing their respective treatments were ‘encouraged toreconsider their decision throughout the sty. aswellasto continue to complete all assessments. Th lading all subjects who continued to participate in ascese- ‘mens (despite refusing pat oalloftheiasigned treatments), theabsolueatrtionateoverthe course ofthestudy was3.5%, withonly 20 completedropouts by 14 months (6in thecom- munity care group, 3 in behavioral treatment, 8 in medica tion management, 3 in combined treatment). STATISTICAL ANALYSES Given the well escribed advantages of random-ffects r= gression (RR) techniques over rational analyses of vari ce fr clinical ial data” we used RR whenever pos- Sle or ur primary intent to-teaanalyses2" Rather than define asingl outcome, we specified muliploutcome var shes, anicpating ferential impacis ofthe reatment mo- dalies on vriou outcome domains" Based om ou data reduction procedures outlined above, the 6 domains were represented by 19 measures (Table 4). Foreach outcome va Sets for site, ime, imeXtreament (he eatment ef fect overtime), and siteXteatmend time were conducted ‘within the intent-to-treat RR analyses. When omnibus RR alyses comparing all groupe were sigan, 3 sets of puree comparisons were performed, each set sddresing Tof the principal study questions (1) medication manage" tment vs behavioral eatment (tailed) (2) combined tres tment vs medication management and combined treatment ‘behavioral eatment (ted, assuming the superioty of combined treatment); and (3) communiy cares me Cation management, community care vs behavioral teat tment, and community carevs combined treatment (1 tle, ‘ssuing the superiority of MPA treatments). onferonnicorrectionswereapplistoallomtubus tess, based onthe number of measures the respective domain. Thus, foreach ofS measures within the ADHD domain stan- darddetintions of signtcance (P= 03) werecoreted yd widinghy 3 requiring P01 Torsigalicance. The pairwise Conirastwerlliewcadsted by dividing omnbue erected Sigllcanclevesfurher by (P<01 = 0017 inthisex- imple). Given our aprior hypotheses abou the superiority oftultimeda ireaiment (combined treatment vsmedication ‘management and behavioral treatment), swell asthe supe- Foriyofall3 of ur treatments over community care (com- ‘munity cares combined eiment, medication management, betavlorreatmen), we sed Ital tstforthesespecl contrasts. This approach wa laken asa means of tring & talance between the dangers of commiting type I erorsts overnterpretingigfcant ndings that occuredsmply by chance (ype ertos).Readersarecncoursged inexerle ca {ion when interpreting Led finding and may choose to double lle P values “ Because ial R analyses revealed bth quadratic nd linear efecto eon trestmentoutcmnes,we computed the logofihe number ofdayssince randomization foreachasssr sent point and used tes log vals nal RR analyses Bespite high compliance, we checked whether com- pllance with asessments (le, missing data) could have hanged our findings Random-electe regression analyses were competed 2 ways: once with inclusion of ll sub- jects and then with only those subjects wh provided data Over multiple time points during the study. No dilfer ences emerged among these 2sets ef analyse, lending con- fidence tothe overl findings (©1909 American Med Association, All rights reserved. ‘Downloaded From: httpi//jamanetwork.com/pdfaccess.ashx?url=/data/journals/psyet +8/5107/ by a University of Wisconsin -Madison User on 04/29/2017 yen or yper oe Principal Collaborators ional Institute of Mental Health, Rockville, Md: Peter S, Jensen, MD (Ollie ofthe Director); L. Eugene Arnold, MD (Department of Psychiatry, Ohio State University); John E. Richters, PhD (Developmental Psychopathology and Prevention Research Branch); Joanne B. Severe, MS (Research Projects and Publications Branch); Donald Vereen, MD (Otfice of Drug, Control Policy);and Benedetto Vitielo, MD (Child and Adolescent Treatment and Preventive Interventions Research Branch). Office of Special Education Programs, US Department of Education, Washington, DC: Ellen Schiller, PAD. Principal Investigators and Coinvestigators University of California, Rerkeley/University of California, San Francisco: Stephen P. Hinshaw, PAD (Dej chology, University of California, Berkeley); Glen R Elliot, MD, PhD (Department of Psychiatry, Uni ‘San Francisco). Duke University, Durham, NC: C. Keith Conners, PhD, Karen C. Wells, PAD, and John March, MD (Department of Psy- chiatry and Behavioral Sciences). University of California, irvine/University of California at Los Angeles: James Swanson, PhD, and Timothy Wigal, PhD (Department of Pediatrics and Cognitive Science, Univesity of California, irvine); Dennis P. Cantwell, MD (deceased) (De- partment of Psychiatry, Neuropsychiatric Institute, University of California at Los Angeles). Long Island Jewish Medical Center, New York, NY/Montreal Children's Hospital, Montreal, Quebec: Howard B. Abiko PRD (Department of Psychiatry, New York University School of Medicine); Lily Hechtman, MD (Department of Psychiatry Mccill University, Montreal) ‘New York State Psychiatric Institute/Columbia University/Mount Sinai Medical Center, New York, NY: Laurence L Green hill, MD (Department of Psychiatry, Columbia University); Jellrey H. Neweorn, MD (Department of Psychiatry, Mount Si- nai School of Medicine) University of Pitsburgh, Pittsburgh, Pa: William E. Pelham, PhD (Department of Psychology, Sate University of New York at Bullalo); Betsy Hoza, PhD (Department of Psychological Sciences, Purdue University, West Lafayette, Ind) Statistical and Design Consultation Helena C. Kraemer, PhD (Department of Psychiatry and Behavioral Science, Stanford University, Palo Alto, Calif) ‘Table 1. Subject Exclusion Criteria Child curantyn Rospal Child currant naothar study alow 8 a a WISI scale and on SIB Bor eso psychos, or persoralty order Chronic tou es or Tutte syndrome (C0 serous enough to requis spaateeatment Neurlpimeieston in prevaus 6 mane Majer neuroogal or media nas History of itlrance to MTA medeations (going or previously uneportd abuse Missed ne fourth of school day in ravius 2 months Sams clazeroom ae hil aad in MTA study Parental stimulant abuse in previous 2 years Nan-Englsh-peaking primary career ‘atari same hovshod in NTA td No tephone Sul or homicidal Tabi to parts in school eorponent Confounding of assessments and proceduat Inability to participa in psychosocial veatment equa treatment, may be incompaibe wth MT study Possible conrndcatio or etimlants “eaten may be incapable wih MTA sty Nay ned reaimpion:ncompatble wit MTA study Ira to paras uly in treatment Irby to parcipata in maiatan canton Riek of rma rom home Inability to partcipata in school component Poesiblcrose-arm contamination by acher Fisk of pret co-oping cis medeations Irby to participa in pret tang groups Passbl crosearm contamination Inability to parcipatainangeing contacts “Treatment requirments beyond aby of MTA study “WISCI ndeates Wechsler neligence Seat for Chlren-Thrd Edo, SB, cas of independent Behavior, OCD, obsess compu asad; and [MIA Mutimoda Tesiment Sta of Caren With Atention Det pert Diseaee the first-ever description of the relative effectiveness of these treatments throuigh 14 months, using an intent- to-treat analytic strategy with random-ellects regres- sion techniques describe the results of the paired comparisons, in order of out original hypotheses, Do medication and behavioral treatments result in com- parable levels of improvement in pertinent outcomesat the ene 9f treatment? Robuist differences were found according to 2 different data sources, indicating the superiority of me (as For 10 of 19 variables, omnibus tests revealed signifi- cant treatment effects over time. For these analyses, we (©1900 American Medical As cation management over behavioral treatment for ADHD symptoms (Fable 8); namely, parents’ and teachers rat- {ngs of inattention and teachers ratings of hyperactivity- lation, All rights reserved. Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 ‘Table 2. Number of Subjects and Reasons for Exclusion my Diagnosis and Telephone Seren Malad Ratings School Agrement__Ba Fearon Tor oxhson ding wach pass, No. (2) Wrong age or grade 01 207" Distanc trom schol an family 600 (4) Medal xeuson 73) 150) 179) 81) Parent tia 410) 335 (14 95,10) 13) Moving, anguage, no tephone, schoo retusa 248) 1136) s18(12) Bit) or school neg, toot, conainain Pant or acter symptom checks cutpits nt mat 5222) Comps pacage nat returned 2308) ‘Getta nt met 17) Diagnost itrviow eter not met 19) Pent ug use 3(1) ‘oa. (of ube excuses During Eaen Phase zza8 (49) +408 (60) 320 38) 3065) “oa aot Subject Entering Eaen Phase set 2337 ve 008 Percentage ter tothe proportion of subjects exuded fo tat ason among the toa number a subjects who entered at erutmenasessment phase impulsivity. Medication management and behavioral teat- ment did not differ significantly on any other outcomes. Do participants assigned (o combined treatment show higher levels of improvement in overall functioning in perti- nent outcome domains than those assigned to either medica- tion management or behavioral treatment atthe end of treat- ‘ment (I-tailed hypotheses)? These analyses indicate that ‘combined treatment and medication management did not differ significantly across any domain, Compared with behavioral treatment, combined treatment was superior in benefitting ADHD symp- toms, according to parents’ and teachers ratings of in- aliention and parent-rated hyperactivity-impulsivity Combined treatment also significantly outperformed be- havioral treatment on parents’ SNAP oppositional/ aggressive behaviors, parentrated internalizing symp- toms (Table 5), and Weschler Individual Achievement Test reading achievement score (Table 4). Do participants assigned to cach of the 3 MTA treat ments (medication management, behavioral treatment, and combined treatment) show greater improvement over 14 months than those assigned to community care (1 tailed)? These analyses reveal that combined treatment and medi cation management were generally superior to commu nity care for parent- and teacher-reported ADHD symp- toms, whereas behavioral treatment was not (Table 5) Innon-ADHD domains, medication management and be- havioral treatment were superior to community care on 1 domain only (teacher-reported social skills and 1 mea- sure of parent-child relations, respectively). In contrast, ‘combined treatment was significantly superior to com= munity care on all 5 non-ADHD domains of functioning -reported oppositional/aggressive behaviors, in- ing symptoms, teacher-reported social skills, par- ent-child relations, and Weschler Individual Achieve- ‘ment Test reading achievement scores) Because ot RR intent-to-treat analyses included all, subjects’ data points through 14 months, itis possible that, some treatment groups (especially behavioral treatmes where 38 crossovers to medication occurred) may have (©1909 American Med fared ditional treatments over the course of the study, To ad- srbecause of the number who had received ad- dress this issue, we conducted additional RR analyses, ce soring any observations obtained alter crossover subjects hhad received the additional treatments. These analyses yielded no differences from the findings noted in Table 5 (analyses available from the authors on request). The Figure shows the # RRanalyses selected to high light findings [rom different domains, as well as from ral- ers who were likely to witness the target behaviors: hyperactivity-impulsivity (a core ADHD symptom) (eacher report), internalizing symptoms (parent re- port), social skills (teacher report), and parent-child ar- guing (power assertion, parent report). The remaining, IRR graphs are available from the authors on request ea} All4 groups showed marked reductions in symptoms over lume, with significant differences among them in de- grces of change. Combined treatment and medication ‘management treatments were clinically and statistically superior to behavioral treatment and community care in reducing children’s ADHD symptoms. Combined be- havioral intervention and stimulant medication— multimodal treatment, the current criterion standard for ADHD interventions—yielded no significantly greater benefits than medication management for core ADHD symptoms; this parallels findings reported by oth- ers." Also consistent with previous reports, our com- bined treatment outcomes were achieved with signili- cantly lower medication doses than used in medication management." For other areas of function (oppositional/aggre- ssive behaviors, internalizing symptoms, social skills, par- ent-child relations, and academic achievement), few dif- ferences among our treatments were noted, and when found, were generally of smaller magnitude. In fact, com- bined treatment, medication management, and behav- oral treatment never differed significantly among Association, All rights reserved. Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 ‘Table 3, Baseline Characteristics of the Sample (N = 878)*4 oe _— = = = = Eten “WISCIndeales Weck neligenee Seat fr Chlren-Thrd Eaton, DIS, agra interview Setedue or Oden and ADR, anton ypracity sore Epes indeat tat the value vas nt sigan ‘Teatment groups aifeed sgnicanty on ony T variable age), wh tes dered sigiteanty on most varies. ‘Hiiues measured asing overly est ‘empty tr ote proportion ote cape whos parents hl futine os; marie reso ose with act 2 parent fans fared or common iw) (©1909 American Medical Ass lation, All rights reserved. Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 Table 4, Basen ‘combined Treatment Meaieaton management Baseline ($0) “me (60) Baseline ($0) “na (60) uteome domain Measure and Rater (Woof subjects] (Mo of sujets)__(No. Sunes) __Wo. of Su {DAD symptoms | Trtenon Teacher 216 (057 137] 1.12(075){t34) 227 (081) [138] 1.14 077 (120) Pant 27st) [140] 102/088) (t33] 203 (080 [14] 1.12070) (12), ypeactlinulie Teacher 189 (077 (137] 075,071) (138) 208071) 138) 0.82069) (120) Patent 4191050) [140] —1.85(083) (133) 189082) [140] ot 045) 12), (Gaeeraom obser 033,022) [122] —ozt(o20)[t14) ost (oat) [tra] 016015) [110), ‘Acgression-000 (00 aggression Teacher 120 (091){137] at (068){134) 139 (00)(092] 065068) (120) Patent 4130(071) [140] 076(088 (133) 145 (080) (139) 0.04074) 12), (Gateraom cbse 001g o.zyiizz] coor coors) 118} 0.04 (0.025)(119) a. (0.011) (10 Intaraliing symptoms SAS intrnaiig symptoms Teacher 073,051) {113] 68 4a) {t08) ——079(047 [N17] 06047 [2 Patent 098,037) (138) ag7(o37) (127) 097 (087 [137] 071030) [120) 25a (o4n (tas) 233(047 [193] 2B 04m) (148) 2.72 (047) [125] Soil tills ee (oz) (sta) 4.19090) {108) ga (oatyftH7] 1.18022) 108023) (138) 122027) {127} 101 (02m [137] 1.17025) (120) 280(001) [79] 284091) (68) Parent-hilraons Pwr asarton Pant 266 (05 (041) 2.31(056)(133] 275 056) (140) 2.46 (057) (122) Pesan closeness Pant 256(052){141] 364(052)(133] _858(040)(140) 955082) [122) ‘came achinvement] Chis scores Beading 965(146) (045) 941519] BACT) [NAA] 879,144) 124) Nsarstes s79(151){045] 100516196) 97 2r26) (144) 907 (130) [124] paling 95:1(14B) (04s) 7.0419] 95221) [NAA] 96.0148) [124] ADHD ndeatesatenbon-ceiivhyperactuity sorter DD, oppostonat dent dover, SERS, Socal Skis Rating System, and WASC, Muidimensioral Ait Scale for Chien Flpses inate tha he arable vas amined an at end pan ‘igher score nates Increased symotos or inpatmeat uss eters acted. $A Nighe core indatsinseased aby SPluned contests: combined teatnent nd medaton management were more tect tan behavioral reatment(P <0, not sigan ate Bonteroni aircon. Ana} of variance compare saciomeri scores across group (M = 261): teatment group, F= 23, P08 (at sigan ate Bonteron| fartction si teament se = 15 nas of covariance compared eating subtest score across group (M = $53): oveal, F= 38. P= 01; weament group, F= 3.75, P00; and site regiment not signfcant. hanover a of 23,528 Parwise comparisons: combined eatment as move fective tan behavioral eaten and fammuny cre in parse contasts. The anaes of math and speling yelled no significant main eet for treament group, so 0 pase agasons were pefamed. "eased osing Wechsler Indl Aclevement Tes: abgher score was beter. themselves, with 3 exceptions (combined treatment > Our finding that MTA treatments (most notably behavioral treatment for parent-reported internalizing _ combined treatment) offered greater benefits than com- problems and oppositionaVaggressive symptoms, and We- munity care for oppositional/aggressive behaviors, in- schler Individual Achievement Test reading achieve- _ternalizing symptoms, peer interactions, parent-child ment score). relations, and reading achievement has not been pre\ With respect tocomparisons of MTA treatments with ously reported in long-term studies."*"" However, the dif community care, combined treatment and medication ferential benelils in these non-ADHD domains are con- management fared substantially better than community sistent with the theoretical aims of multimodal care on most ADHD outcome measures, while behavioral approaches." For example, medication is known to r treatment did not. According to least | informant, com- duce negative peer interactions dramatically, but in- bined treatment also fared significantly better than com- creases in positive social behavior are far less robust." munity careforall Snon-ADHDdomains:parent-reported Such changes might require intensive and long-term ap- ‘oppositional/aggressivesymploms,parent-reportedinter- plication ofthe behavioral components of combined treat- nalizing problems, teacher-reported socialskills, parent- ments, such as those found in our STP and school- childrelations.andreadingachievement-Incontrast,medi- based interventions.” Similarly, parent training, which cation management and behavioral treatmenteach fared includes positive parental attention and rewards for the betterthan community carein !non-ADHD domain only child's appropriate behavior, when combined with medi (teacher-rated social skills and parent-child relations). __cation, might be expected to decrease oppesitionality and (©1909 American Medical Association, All ights reserved. Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 ‘Bevavira eniment ‘essrement and Reeral “4m (80) (0) 40 (60) (No.of Subjects} (No.of Sublets} _ No.of Subjects) 228,064) (196) 1.47 (081) [119] 249 (060)[15] 1.48 0.82) (128) 4190 (053) 139] 1.40(068) (129) 205 065)(142] 149/087) (190) 205 075) {136] 1.410107 [119] 193 (081) [135] 1.25 088) (128) ‘an 0.4) 140] 1.24(0.72) (129) 1.95 (087 [142] 1.95(0.72) (190) 0:37 (0.26) (120) 0.20(026) [107] 0:38(027) (118) 018109) (0.15) 4143 (096) 136] 0.97 (280) [149] 1.25 088)(135] 1.00084) (128) 4.37 (070) (140) 4.05(0.78) (129) 1.4 (0.70)(142) 1.11 (67 (190) 0020 (0.045) [120] 610 (018) (107) 0.019 (0.025) (118) 2006 (0.014) 109) 82 (045) {115] 0.58 (0.40) (100) 0.78 (040 [15] 969 0.48 (105) (093 (048) (133] 0.77 (240) (191] 097 (036){137] 082/043) (125) 246 (055) 143] 2.27 (040) (192) 240 058)(145] 227 (045) (124) 80 026) {815] 4.06(032) (100) 0.87 (020)(15] 1.05031) (105) 1020.22) (833) 1.15(024) [191] 1.03 (028)(137] 1.18(028) (125) 323,108) (69) 3.05 (0.82) [68] 275 (050) (141) 247 047) [191] 271 (057 [142] 252(057) [190] ‘352(040) (141) 3.50(048) (192) 358 0.48)142] 2.69(0.48) (190) 95.1141) (148) 96.2(140) (194) 95:5 (143) 146) 96.4142) (191) 97:7 (192) (04a) 1003 (12:7 (194) 986 (141) 146) 1004152) 131] 928 (125)(t4a] 987 (29) (194) 98.7(181) (146) 9420041) 191] ‘enhance parent-child relations more than medication alone. For internalizing symptoms, the relatively greater improvements for subjects given combined treatment are particularly noteworthy, as none of our wreatments were Aesigned to address this domain specifically. The MTA study extends the findings of previous studies that demonstrated short-term, robust efficacy of medication management, showing thal these benelitsper= sist during treatment up to 14 months. In contrast to frequently expressed concerns, children given com- bined treatment and medication management tolerated medication well, including a third dose given im the al- temoon, The relative improvements attabuted to medi- cation management also parallel findings from other, longer-duration stimulant tials." Given the MTA\ssize and scope, however, we saw elfects across diverse set Lungs, patient groups, provider characteristies, and out- ‘come domains. These findings were further strength- ‘ened by the absence of any site X treatment interactions, (©1909 American Med Although combined treatmentand medication man- agement were generally superior to community eate,com- ‘munity treatments usually included medication: hence, ‘tisunclear which components of the 2 MTA medication treatments may have rendered them more elfecive than community care Further analyses ofthese findings re pre- sented in our companion report,” but reviewing the ap- parent differences is instructive. We used a manualized ‘dication tration procedure and “thrice-daily” dosing, aswellashigher,carcfully monitored daly dosesto maxi- ‘mize positive effects and minimize side elects." We met ‘with parents monthly and obtained systematic Feedback Irom both them and the children's teachers, Parent guid. ance and selected readings were provided as needed: this, is reported to provide benefits over simple pill dispens- ing alone.» These components, particularly the system- atic and regular feedback from teachers, do not seem to be part of routine pediatric ADHD treatment practices,” and may have enhanced the elfectiveness of or medica- tion management. The modest benefits for some non-ADHD domains obtained by multimodal treatments have been reported previously’ alter 3to-4 months of teatment. In con- trast, the study by Hechtman and Abikot® filed to dem onstrate these fleets alter 12 months of active treatment, perhapsbecause of theirsmallersample (103 subjects dis tributed aross’ groups). Inaddition to the MTA’S sample sizeadvantages, ts behavioral component design required that behavioral treatment interventionsbe delivered aeross smuliplesettingsand caretakers (home, school, and STP), augmented with further strategies to facilitate the gence” alization of elects across settings and over time”!”—all enhancements not found in previous studies” Whether there is greater vale for multimodal teat- rents for ADHD depends on which intervention is com sidered as the comparison. If one assumes that a behav foral intervention should always be used as the first-line ADHD treatment (often the preference for many’ parents, and the practice in many European countries), and that the possibly greater benefits of combined treatment should be determined, then combined treatment seems to ollera great deal of benefit over behavioral treatment alone. But sone provides carefully monitored medication treat- ‘ment similar to that used inthis study asthe frst line of tweatment, our results suggest that many treated culdren ‘may not require intensive behavioral interventions." The significantly lower total daily doses of methy phenidate inthe combined teeatment arm are notewor thy but not unforeseen,” The importance ofthis find- dng is unclear, and a rigorous test ofthis question would likely require different design. Nonetheless this issue remains an ongoing source of concern of many patents and clinicians and should not be dismissed, particularly since side ellects are usualy related to dosage. I equiva- lent, sometimes better results can be obiained by acom- bined weatment that uses lower doses, such findings may have public health importance Concerning the relative benefits of our behavioral treatment alone, results must be understood within the context ofthe limitations of our study design. Most imi- portanly, our design did not include a no-treatment oF placebo group (an ethically unacceptable option for an Association, All rights reserved. Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 ‘Table 5, Random-Eitects Regression Analysis* Resi of Fakrwise comparisons Comb eed Mgt utcoms Ooms} Measure an ator Random Regressions Med mote Behav P and com ve Behav ‘OHO synpions Tatton Peot-002" “Teacher ‘Weatmant tine Figs =106;P001 Mad Mgt 01 Combes Mat “reatmen st F =D; P=.86 Comb>Bebav Ste: Fa27 Pea? Parent eaten tine Fiyy=215: F001 Mad Mgt 01 Combes Mat Treatments F=06; P=. 88 Comb>Bebav Ste: F=40 Peo? ypeacte-inulsie Teacher Teatmentctine:Fie=100:P=001 Mad Mgt Behav 0046 Comb = Med Mgt “eaten ate f= 13; P=. £0 Comb = Beh Ste: F=40 Peo? Parent ‘eaten ctine: Fi =215;P001 Mad Mgt (01 Comb Med Mgt ‘eaten ate F=13 P=.28, Comb Behav Ste: Fat Pe 0006 Ciasstoom (Gassoom observer Teatmant tine: fs =26; P= 05 Med Mgt Bahay 02§ Comb = Mad Mgt ‘eatmantcsite F215; P= 11 Comb = Beh Ste:F= 115; P01 ‘Aagression-000 (000 = aggression P02 009 Teacher ‘Teatmentctine: Fin =65;P=0002 Mad Mgt Behav —01§ Comb = Med Mgt “eaten ate f= 12; P= 25, Comb = Beh Ste: Fa42- Pe On Parent ‘eatmant tin: Fuye=7A4;Pe001 Ma Mgt = Behav 007§ Comb = Med Mgt “eaten at F21.%, P=.40 Comb Behav Ste: F=43 P= 0007 (csesoom obeaner Intaraliing symptoms SSAS nterealing Pe0m 008" “Teacher “eaten tine Fy =21; P= 10 “reatment st F=05; P=.02 Parent ‘eatmanb tine Fuyy=02;Pe001 Mad Mgt = Behav 03§ Comb — Mad Mgt “reatmentc at F211; P= 35 Comb Behav Ste: F=22 Pe05 ase hia “eaten tine Fun = 06; P= 65 “eaten ie: Te Ste: F=22 Pe 08 Soil kills sss P02, 008 Teacher “Teatmentctine:Fi=6:1;P=0004 Mad Mgt = Behar Comb = Md Met “eaten at F = 05; P=06 Comb = Beh Ste: Fe42 Pe one Parent ‘eaten tine i =22; P= 08 Med Mgt = Behav Comb = Mad Met “reatment at F= 1.0; P=86 Comb = Beh Ste: Fa38 Pe02 Parenti rations Power assertion Pe, 00 Parent “Teatmentctine: isn =86;P= 0008 Mad Mgt = Behar Comb = Md Met “eaten at f= 1.0; P= a5 Comb = Beh Ste ep=02 Przonl closeness Parent ‘eaten tne Fag = 20; P= 0008 “reatment ate F=1.0; P= 54 Med gnats medion management Bea, Beavaraleanent, Comb, combed ramet, CG community ae ADR, atentor de hypeactily ister 00, opposina-dent dor SEAS, Soci Sil Rang System and MSC, Mulsmensonal Anty Scab or hen Epes nate a the ‘bu andr pase st rest wasnt sigan {Outcome domains were measured by Bonerontcoracted eso sgiteance, with orabus) ard pai) comparisons. {Done any when arabs nahi eaten tine ec as sian. vasa sigan after Bonerancorecton. (©1909 American Medical Association, A rights reserved, Downloaded From: http:/jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/S107/ by a University of Wisconsin -Madison User on 04/29/2017 Att stating) nd non aes gen com fe reiincicrenown een. Substantial im rs ma Sy own provement occured overtime acrossal groups inci erase cei: coms, havnt Termica | ing communiy cate), regardless of rating source oF Prey ea ab re woe es method. Whe some of his change ould be repression combo ct oot -a1a8 0017) “Stantial group differences reported here. Mote than three 008 Med Migt>Co oot -0196 (007) fourths of subjects given behavioral treatment were suc- Soo ce Sion IZ 001 | Cssflly maintained without mediation throughout the conb>0z oo: “atmtan) | study, Consequently, it should not be concluded that ot Meahigece Oo 2161001) | bchavioal weatment interventions didnot work Beta = OF ie th oa ‘One caveat concerns our choice and number of out- scont-ce oy 9B ER% | come ensures, Werelt tha the treatment effects in d= Frees got ROOT | rent domains and from diferent responders might ar Soot 29} “atgmanm) | and that these variations were nccerery in interpreting Bitton | the resuls of the study. Consequently. despite the lows fom Oot 0176 (00013) of power caused by Bonferonni corrections, we chose 19 001 ‘Med MigtOC_ O01 -0.165 (0.013) Pe fher than a singh Soon ct ABI | primary outcome measures rater than » single sum “Gow (omi) | faary outcome score Power calculations” that underlay cons = co “aosrionis) | the scuing of sample size were based on setting as the ig Medes ox 398510519 | ower limit ofelinteal significance an eect size of 04 Soot ce anion | ‘ small to moderate) and om the requirement of 80% “aoer oni) : ore cog SOR | ower (Sx level of confidence) to detect elects of that aie Wehigcoe fot afar) | magnitude. Thus, the chance ts high of declaring effet Soot Soren, aes lower than 0.4 wot statsealy significant, ever “ost (0.00) though some clinicians might regard such effects as clini. cont 0z oe “35001 ugh som ait Ahi ox ooms ago(oois) | cel signin owt ct

    You might also like