Skip to main content
. 2019 Jun 21;2019(6):CD008223. doi: 10.1002/14651858.CD008223.pub3

MTA 1999.

Methods Design: RCT
Participants Country: USA
Setting: six multisite outpatient clinics
Sample size calculation: 576 participants required
Sample size: 576 children
Sex: 465 (81%) = boys, 111 (19%) = girls
Age: range 7.0 to 9.9 years
Ethnicity: white = 61%, African‐American = 20%, Hispanic = 8%
Socioeconomic status: not reported
IQ: below 80 excluded
ADHD diagnosis:DSM‐IV, ADHD combined type
ADHD medication: 97% received methylphenidate
Comorbidity: anxiety disorder = 33.5%, conduct disorder = 14.3%, ODD = 39.9%, affective disorder = 3.8%, tic disorder = 10.9%, other = 2.2% such as bulimia, enuresis
Medications for comorbid disorders: balanced between groups.
Inclusion criteria:

  1. boys and girls

  2. 7‐9.9 years of age (1st‐4th grades)

  3. residing with primary caretakers for at least six months

  4. meet dimensional criteria for hyperactivity on the basis of parent and teacher rating scales and full diagnostic criteria for ADHD combined type


Exclusion criteria:

  1. currently in hospital (inability to obtain school assessments)

  2. currently in another treatment study (confounding of assessments and treatments)

  3. below 80 on WISC‐III Verbal IQ, Performance IQ or Full Scale IQ scores, and on Scales of Independent Behavior (insufficient ability to participate in psychosocial interventions)

  4. bipolar disorder, psychosis, pervasive developmental disorder, severe obsessive‐compulsive disorder (treatment may be incompatible with the study intervention)

  5. chronic, serious tics or Tourette's Disorder (possible contraindication for stimulant treatment)

  6. neuroleptic treatment in previous six months (may need resumption, which is incompatible with the study intervention)

  7. major neurological or medical illness that would interfere with study participation or require medications incompatible with the medications used in the study (inability to participate in the study intervention)

  8. history of intolerance to MTA medications (dangerous if participants assigned to arm involving medications)

  9. suicidal or homicidal (needs more intensive treatment than the study intervention provides)

  10. ongoing or previously undisclosed child abuse (risk of removal from home precludes parent intervention and consistent parent data)

  11. missed more than 25% of school days in previous two months (interference with teacher assessments and school intervention)

  12. another child in household already participating in the study intervention (cross‐arm contamination if two children in same household randomised to different arms)

  13. same classroom as child already participating in the study intervention (cross‐arm contamination if two pupils in same classroom are randomised to different arms)

  14. parental stimulant/cocaine abuse in past two years (possible co‐opting of child's medications)

  15. inability of parent to speak English (inability to participate in parent training)

  16. no telephone (inability to participate in telephone calls with therapists)


Baseline characteristics: no significant differences among study groups
Interventions 576 children allocated to one of four groups

  1. Group one (n = 144): medication. In total, 14 months of medical intervention implemented as follows: one month of blind titration with methylphenidate for best dose, if unsatisfactory, then open titration with d‐amphetamine, pemoline, imipramine, and others. Supplementary general advice and selected readings without systematic behavioural intervention. Monthly visits after the titration period, doses adjusted as indicated by monthly monitors

  2. Group two (n = 144): psychosocial. Intensive behavioural treatment consisting of three major components: 1) 27 group and eight individual sessions of parent training; 2) school intervention comprising teacher consultations (six to 20 sessions) and 12 weeks with classroom behavioural specialist for half the time in the classroom; and 3) a child treatment component anchored in an intensive 8‐week, full‐time, summer treatment programme. No medication

  3. Group 3 (n = 145): combined treatment. Integration of all treatment components in medication + psychosocial treatment groups and standard community care

  4. Group 4 (n =143): standard community care. Treatment of own choosing in the community; no treatment provided by the study group


Attendance: in group two and three, the families attended an average of 77.8% of the parent training sessions and 36.2 of 40 possible summer treatment programme days. In the school component, there was an average of 10.7 teacher consultation visits and 47.6 out of 60 possible days of work with a classroom aid
Outcomes Primary outcomes

  1. Social skills: SSRS, parent‐, child‐ and teacher‐rated

  2. General behaviour: CBCL, parent‐ and teacher‐rated


Secondary outcomes

  1. Core ADHD symptoms: Swanson, Nolan and Pelham Rating Scale, teacher‐ and parent‐rated

  2. Performances and grades in school: Wechsler Individual Achievement Test (WIAT)


Outcome assessment: post‐intervention, end of treatment and follow‐up data
Notes Study ID:NCT00000388
Sponsorship source: The study was supported by grants from the National Institute of Mental Health (UO1 MH50461, U01 MH50447, U01 MH5044, Uo1 MH50453, U=1 MH 50454 and U01 MH50467).
Year conducted: not stated
Duration of the study: 38 months
Comments: The Multimodal Treatment Study (MTA study) is a cooperative study, performed by six independent research teams in collaboration with the National Institute of Mental Health, Rockville, MD, and the Office of Special Education Programs, US Department of Eduation, Washington, DC.
Lead author: MTA Cooperative Group. Corresponding author: Peter S Jensen
Institution: Department of Child Psychiatry, Unit 78, Center for the Advancement of Children's Mental Health, New York State Psychiatric Institute/Columbia University
Email: jensenp@child.cpmc.columbia.edu
Address: 1051 RiversideDr, New York, NY 10032
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Low risk Comment: adequate method used
Allocation concealment (selection bias) Low risk Comment: adequate method used
Blinding of participants and personnel (performance bias) 
 All outcomes High risk Comment: no blinding
Blinding of outcome assessment (detection bias) 
 All outcomes High risk Comment: blinded and unblinded raters
Incomplete outcome data (attrition bias) 
 All outcomes Unclear risk Comment: used imputation
Selective reporting (reporting bias) Unclear risk Comment: where is the consumer satisfaction and the CBCL data reported? We requested clarification from one of the study investigators, but had received no response when this review was finished.
Vested interest bias Low risk Comment: no vested interest
Other sources of bias? Low risk Comment: no other apparent sources of bias