Intervention Review

You have free access to this content

Selective serotonin reuptake inhibitors (SSRIs) for autism spectrum disorders (ASD)

  1. Katrina Williams1,2,3,*,
  2. Amanda Brignell1,
  3. Melinda Randall1,
  4. Natalie Silove4,
  5. Philip Hazell5

Editorial Group: Cochrane Developmental, Psychosocial and Learning Problems Group

Published Online: 20 AUG 2013

Assessed as up-to-date: 7 AUG 2013

DOI: 10.1002/14651858.CD004677.pub3


How to Cite

Williams K, Brignell A, Randall M, Silove N, Hazell P. Selective serotonin reuptake inhibitors (SSRIs) for autism spectrum disorders (ASD). Cochrane Database of Systematic Reviews 2013, Issue 8. Art. No.: CD004677. DOI: 10.1002/14651858.CD004677.pub3.

Author Information

  1. 1

    University of Melbourne, The Royal Children's Hospital, Department of Paediatrics, Parkville, Victoria, Australia

  2. 2

    The Royal Children's Hospital, 3W Clinical Offices, Parkville, Victoria, Australia

  3. 3

    Murdoch Childrens Research Institute, Parkville, Victoria, Australia

  4. 4

    The Children's Hospital at Westmead, Child Development Unit, Westmead, New South Wales, Australia

  5. 5

    Sydney Medical School, Discipline of Psychiatry, Concord West, New South Wales, Australia

*Katrina Williams, Department of Paediatrics, University of Melbourne, The Royal Children's Hospital, 50 Flemington Rd, Parkville, Victoria, 3052, Australia. katrina.williams@rch.org.au.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 20 AUG 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Barthelemy 1989

MethodsCross-over
No loss to follow-up


ParticipantsN = 13; Children only
8 boys, 5 girls
Age range 3 to 10 yrs, mean age 6 yrs, 4 months.
Diagnosis DSM-III autism

IQ range 30 - 75. Obsessive-compulsive behaviours not required


InterventionsTreatment: Fenfluramine twice daily divided dose at total 1.5 mg/kg
Reduced to 0.8 mg/kg in 2 children due to adverse effects
Duration: 3 months
Placebo: identical, placebo phase duration 1 month


OutcomesWeight
Behavior Summarised Evaluation
Urinary dopamine metabolites


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and treating physicians blinded

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear if there was blinding of outcome assessors

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo loss to follow-up

Selective reporting (reporting bias)Unclear riskNo information about what as planned

Buchsbaum 2001

MethodsCross-over
No loss to follow-up


ParticipantsN = 6; Adults only
5 men, 1 woman
Mean age 30.5 ∓ 8.6 yrs.
Diagnosis DSM-IV, ADI

5 autism, 1 Asperger disorder

IQ scores ranged from 53 to 119 and all participants were verbal. Obsessive-compulsive behaviours were not a requirement


InterventionsTreatment: fluoxetine starting dose 10 mg/day up to maximum dose 40 mg/day for 8 weeks
Placebo not described. Duration of placebo phase = 8 weeks


OutcomesYale-Brown Obsessive Compulsive Scale
Hamilton Rating Scale for Anxiety
Clinical Global Impression Scale - Improvement

Also mention of adaptation for baseline assessment for autism and scores for this measure are reported after intervention
Positron Emission Tomography


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and treating physicians blinded

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear if there was blinding of outcome assessors

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo loss to follow-up

Selective reporting (reporting bias)Unclear riskNo information about what was planned

Hollander 2005

MethodsCross-over

44 children randomised of 62 consented. 5 lost to follow-up


ParticipantsN = 44, 39 completed. Children only
30 boys, 9 girls
mean age 8.18 ∓ 3.0, range 5 to 16.
Diagnosis: DSM-IV-TR of Autism, PDD-NOS or Asperger Syndrome

IQ range 30 - 132. No required threshold for obsessive-compulsive behaviours


InterventionsTreatment: fluoxetine 8 weeks treatment, 4 weeks wash-out, 8 weeks cross-over
2.5 mg/day up to 0.8 mg/kg/day maximum


OutcomesYale-Brown Obsessive-Compulsion Scale
Clinical Global Improvement Scale Adapted to Global Autism
Suicidality Subscale of Overt Aggression Scale
Fluoxetine side effects checklist


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Double blind"

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"All CY-BOCS and CGI-AD outcome assessments were completed by an independent evaluator (IE) who did not have access to side effect data and who was blind to treatment condition"

Incomplete outcome data (attrition bias)
All outcomes
High riskLoss to follow-up of 1 non-responder and 3 who were non-compliant and no ITT analysis possible

Selective reporting (reporting bias)Unclear riskNo information about what was planned

Hollander 2012

MethodsPlacebo-controlled trial
Endpoint Classification: Efficacy Study
Intervention Model: Parallel Assignment


ParticipantsN = 37; Adults only
Men (26) and women (11) aged 18 to 60 years (mean age 38.1 yrs (SD 14.26)) who meet DSM-IV and ADI criteria for autistic disorder and have a Clinical Global Impression-Severity Scale for Autistic Disorder score of 4 and are medication-free.

Excluded if: history of hypersensitivity or side effects while receiving fluoxetine; abnormal ECG, lab test or physical exam findings or have co-morbid conditions such as schizophrenia, schizoaffective disorder, active seizures, cardiovascular disease, bipolar disorder, haematopoietic


InterventionsParticipants will receive either placebo or fluoxetine. The drug dosage will be administered by a fixed schedule for first week: week 1 = 1 x 10 mg capsule per day after breakfast and increased each week by 20 mg per day as tolerated by the participant to max of 80 mg. Serum levels of fluoxetine and norfluoxetine will be documented at Week 12. Minimum dose required to continue study was 20 mg.


OutcomesNot listed on the trials register site

In paper the primary outcome was reported as: repetitive behaviours as measures on the compulsion subscale of the Yale-Brown Obsessive Compulsive Scale (only compulsion subscale used, not obsession subscale as authors reason that obsessions difficult to measure in the population)

Secondary outcomes reported are:

- Clinical Global Impresion improvement rating for repetitive behaviours (encompassed both measures from Yale-Brown and obsessive thought patterns observed in high-functioning adults with ASD). Was also based on all available information including rating scales, clinical observations, participant report.

- Global rating: CGI rating overall symptoms

- Aberrant Behaviour Checklist (for irritability)


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskParticipants were grouped into blocks chronologically based on timing of their recruitment to the studies. 'Blocks’ of participants were then randomly allocated to intervention or placebo.

Allocation concealment (selection bias)Unclear riskNo information provided

Blinding of participants and personnel (performance bias)
All outcomes
Low riskBoth participants and assessors were blinded as to participants' allocation.

Study medications were administered in an identical double-blind fashion

Blinding of outcome assessment (detection bias)
All outcomes
Low riskAs reported side-effects were minimal it would have been possible to maintain clinician blinding

Incomplete outcome data (attrition bias)
All outcomes
Low riskNumber of participants lost to pre-trial were balanced in numbers, and reasons given are similar, across groups. No loss to follow-up occurred once trial began

Selective reporting (reporting bias)Low riskResults reported for primary and secondary outcome measures but no outcomes measures reported in protocol.

King 2009

MethodsMulti-centre trial (six centres)

Parallel trial


Participants149 children randomised, 76 to placebo and 73 to treatment group
13 withdrew from each group
Aged 5 to 17.
Autistic Disorder, Asperger Disorder or PDD-NOS, severity of at least moderate on CGI severity of illness scale
At least moderate compulsive behaviours
61% > 70 non-verbal IQ


InterventionsLiquid citalopram obtained commercially.
Placebo matched for smell, taste and viscosity


OutcomesCGI improvement scale
CYBOCS-PDD (clinician-rated)
Composite measure of the CGI improvement scale and CYBOCS-PDD
6 subscales of the Repetitive Behaviour Scale (parent-rated)
Aberrant Behavior Checklist-Community version


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomisation using permuted blocks with randomly varying block sizes stratified by site and age

Allocation concealment (selection bias)Low riskAllocation distal to investigators

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNo mention of blinding of participants and personnel

Blinding of outcome assessment (detection bias)
All outcomes
Low riskTwo "masked" clinicians met with participants during each scheduled evaluation.
The evaluating clinician monitored efficacy and was blinded to adverse events

Incomplete outcome data (attrition bias)
All outcomes
Low riskBalanced "drop outs" for each group. ITT analyses used and suitable substitution of missing data used - as described: "For subjects to complete all post-randomisation assessments, the last observation was carried forward"

Selective reporting (reporting bias)Low riskResults for all outcomes and scales prespecified in the trial registry were available

Leventhal 1993

MethodsTwo phase: placebo-treatment-placebo followed by randomised cross-over


ParticipantsN = 15; children only
3 to 12½ yrs (mean age 7.6 ∓ 2.6yrs).
Diagnosis infantile autism DSM-III
No loss to follow-up. Incomplete data for some outcomes
IQ range 16 - 63. Obsessive-compulsive behaviours not required.


InterventionsFenfluramine


OutcomesRitvo-Freeman Real Life Rating Scale
Connors Abbreviated Parent and Teacher Questionnaires


NotesPrevious use of fenfluramine


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Double blind", no details given

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUncertain if outcome assessors blind to treatment group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo loss to follow-up but incomplete data for some outcomes

Selective reporting (reporting bias)Unclear riskNo information about plans available

McDougle 1996

MethodsParallel trial
No loss to follow-up


ParticipantsN = 30; adults only
27 men, 3 women
Mean age 30.1 ∓ 7.7 yrs, age range 18 to 53 yrs. Adults only
Diagnosis of autism using DSM-III-R and ICD-10 at least "moderate" in severity using the CGI global severity of illness rating
Obsessive-compulsive behaviours not required


InterventionsFluvoxamine to max. 300 mg/day for 9 to 12 weeks
Identical placebo, 9 to 12 weeks
Equality of treatment between groups
Compliance measure unclear


OutcomesRitvo-Freeman Real Life Rating Scale
Clinical Global Impression Scale, global improvement
Brown Aggression Scale
Vineland Maladaptive Behavior
Yale-Brown Obsessive Compulsive Scale


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskNot stated

Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Double blind"

Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors blind to treatment group

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo loss to follow-up

Selective reporting (reporting bias)Unclear riskNo information about plans available

NCT00609531

MethodsParallel trial


ParticipantsIn adults (Citalopram average age 22.1 (SD = 9.0), Placebo average age 23.9 (SD = 10.8)) with diagnoses of autistic disorder or Asperger’s Disorder and high levels of repetitive behaviours as defined by baseline CYBOCS-PDD scores ≥ 8.

Diagnosis was made using DSM-IV-TR for Autistic Disorder and informed by above spectrum cutoff scores on the Autism Diagnostic Observation Schedule-Generic with standard cutoffs.

Exclusion criteria included:
(1) history of intolerable adverse effects with 2 or more SSRIs that were not attributable to inappropriate dosing, dose escalation, concomitant treatments or developmental stage;
(2) history of an exposure to citalopram/escitalopram of sufficient dose or duration to determine response status;
(3) history of adequate clinical trials of two SSRIs;
(4) anticonvulsant medication use;
(5) history of gestational age < 34 weeks, birth weight < 2000 grams, or intraventricular haemorrhage;
(6) history of known medical condition associated with autism including Fragile X syndrome, tuberous sclerosis, neurofibromatosis, phenylketonuria, epilepsy and gross brain injury;
(7) MRI contraindication;
(8) concomitant psychotropic use;
(9) intelligence scores < 70 as assessed by the Weschler Abbreviated Scale of Intelligence


InterventionsCitalopram or placebo

Citalopram doses were initiated at 5 mg/day and the maximal dosage was 30 mg/day. Doses were flexibly adjusted after considering whether the clinical response was sufficient as defined by a CGI-I score of  “1 - very much improved” or “2 - much improved” and whether there were significant adverse effects.


OutcomesAs per trial register -
Primary: (1) Functional Magnetic Resonance Imaging (not relevant to this review); (2) Clinical Global Improvement Scale.
Secondary - Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS), as supplied CY-BOCS and RBSR-total.


NotesParticipants were paid USD 50 for completing the imaging portion of the study and USD 10 per hour for all other study activities


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskStated randomly assigned but method not stated

Allocation concealment (selection bias)Low riskAllocation by the UNC Chapel Hill’s Department of Pharmacy’s Investigational Drug Service, matched for appearance

Blinding of participants and personnel (performance bias)
All outcomes
Unclear risk"The blind was broken after the post-treatment scan, after which appropriate health care provider referrals were given, if desired." Blinding not reported for participants and personnel

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"The blind was broken after the post-treatment scan, after which appropriate health care provider referrals were given, if desired." Blinding was reported as for Investigator on the trials register.

Incomplete outcome data (attrition bias)
All outcomes
Low riskData provided for 6 participants in each group. Appropriate methods to adjust for missing data used for the few data points with missing data. No other participants were enrolled. Planned enrolment was 40 but recruitment stopped at 12

Selective reporting (reporting bias)Low riskData provided for tools listed on trial registry and some additional tools and data not yet published

Sugie 2005

MethodsCross-over


ParticipantsN = 19, 18 completed; children only
15 boys, 4 girls
Mean age 5.3 yrs, range 3 to 8.4 yrs.
Diagnosis: DSM-IV Autism
1 lost to follow-up


InterventionsPlacebo or fluvoxamine 1 mg/kg/day for 2 weeks, 2 mg/kg/day for 3 weeks, 3 mg/kg/day for 6 weeks, 1.5 mg/kg/day for 2 weeks, 2 week wash-out, cross-over


OutcomesBehavioural Assessment Scale
Clinical Global Impression Scale


NotesHaematological and molecular genetic analysis


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated randomisation sequence

Allocation concealment (selection bias)Low riskAt a distance from investigators

Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Double blind"

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear if outcome assessors blind to treatment group

Incomplete outcome data (attrition bias)
All outcomes
Low riskLittle attrition, excluded one participant due to non-compliance

Selective reporting (reporting bias)High riskOnly reported effectiveness for genetic subgroups

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Doyle 2001Not a trial of SSRIs. Cost analysis
Not RCT, no placebo

Gordon 1993Clomipramine: not SSRI

Humble 2001Participants not ASD

McDougle 1998Open-label, non-randomised, no placebo control

Peral 1999Open-label, no randomisation, no placebo

Remington 2001Clomipramine: not SSRI

Sanchez 1996Open-label, not RCT

Scahill 2011Rationale, design and sample characteristics of a citalopram trial. Not RCT

 
Characteristics of studies awaiting assessment [ordered by study ID]
EUCTR2008-003712-36-FR

MethodsNo information available

ParticipantsChildren with autism

InterventionsFluoxetine

OutcomesNo information available

NotesAuthors attempted to contact investigators but no details provided on European register and no response from register to two emails sent requesting investigator details

NCT00183339

MethodsAllocation: Randomised
Endpoint Classification: safety/efficacy Study
Intervention Model: parallel assignment
Masking: Quadruple-blind (participant, caregiver, investigator, outcomes assessor)

ParticipantsChildren aged 30 to 58 months with a diagnosis of autism

InterventionsPlacebo: Between 2 mg per day and 20 mg per day of liquid placebo will be given in the morning using a flexible dosing strategy, following a 36-week dose titration schedule.

Intervention: Between 2 mg per day and 20 mg per day of liquid fluoxetine will be given in the morning using a flexible dosing strategy, following a 36-week dose titration schedule.

OutcomesPrimary outcome: Feasibility and safety of conducting placebo control trial of fluoxetine Time Frame: Measured over 12 months

Secondary outcome: Side effect and drop-out evaluation (Time Frame: Measured at Month 12)

NotesAuthors contacted investigators of this study. Linmarie Sikich responded to query and reported study completed and currently being prepared for publication (N = 18).

NCT00515320

MethodsAllocation: randomised
Endpoint Classification: safety/efficacy study
Intervention Model: parallel assignment
Masking: quadruple-blind (participant, caregiver, investigator, outcomes Assessor)

ParticipantsChildren aged 5 - 17 years who meet DSM-IV criteria for autistic disorder and have a CYBOCS-PDD score of at least 10 at screening.

InterventionsFluoxetine: Once daily oral dispersible tablet 2 mg, 9 mg or 18 mg

Placebo: oral dispersible tablet

OutcomesPrimary outcome: The percentage change from baseline to the endpoint visit for the CYBOCS-PDD score.

Secondary outcomes:

  • The time- and dose-related course of therapeutic effects
  • The inter-relationship between these effects in the context of global clinical changes
  • The indirect effects on participant caregivers of the dose regimen in these subjects compared to placebo during treatment.
  • Safety measures will be physical examination, vital signs, EKG/ECG and clinical laboratory tests.

NotesAuthors contacted investigators of this study. Duke Meriman responded and reported study completed and currently being prepared for publication

NCT00655174

MethodsAllocation: Randomised

Intervention model: parallel assignment

Masking: double-blind

ParticipantsChildren with autism aged 3 - 10 years

InterventionsFluvoxamine and sertraline

OutcomesFrequency and severity of aggressive behaviour, obsessive symptoms or anxiety

NotesAuthors sent two emails to both investigators listed on trial site. No response received to date.

 
Characteristics of ongoing studies [ordered by study ID]
ACTRN12608000173392

Trial name or titleFluoxetine for the treatment of repetitive behaviours in children and adolescents with autism: A randomised double-blind placebo-controlled trial.

MethodsParallel randomised controlled trial

ParticipantsBoys and girls aged 8 to 17 years who

1. meet criteria for an ASD based on the DSM-IV or ICD-10.

2. have a score of ≥ 15 on the total score of the Repetitive Behavior Scale - Revised (RBS-R) at the time of screening.

Participants with an intellectual disability must have previously documented psychometric testing.

InterventionsFluoxetine or placebo (sugar syrup) will be administered as an oral syrup (2 mg/ml), once daily. Trial medication (fluoxetine or placebo) will begin at 2 mg/day, and increased in weekly increments of 2 mg (provided that side effects do not emerge), until an effective dose is reached. The maximum dose will be 12 mg/day by week 6 of the trial. The final effective dose will be maintained from week 7 to 12 of the trial. Placebo (sugar syrup with raspberry flavouring). The placebo syrup will be of similar appearance and taste to the fluoxetine syrup. The packaging for both will be identical.

OutcomesPrimary outcomes: Total score on the RBS-R.

Secondary outcome:Score for each subscale of the RBS-R.

Starting date2010

Contact information

NotesOngoing. Recruitment slower than anticipated.

 
Comparison 1. Proportion improved for Clinical Global Impression Improvement (CGI-I)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Clinical Global Impression - Improvement (CGI-I)3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 Adults
262Risk Ratio (M-H, Random, 95% CI)12.58 [1.77, 89.33]

    1.2 Children
1149Risk Ratio (M-H, Random, 95% CI)0.96 [0.61, 1.51]

 
Table 1. Methods reported in protocol but not used in this review

IssueMethod

Measures of treatment effectContinuous data

Where standardised assessment tools generate a score as the outcome measure, we plan to compare the means of these scores and calculate a mean difference for inclusion in meta-analysis, from data available from trial authors or calculated using methods outlined in Chapters 7 and 9 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Where studies do not use sufficiently similar instruments to measure an outcome, but the construct measured is similar, we plan to conduct meta-analysis using standardised mean difference.

Unit of analysis issuesWe will assess all included trials to determine the unit of randomisation and whether or not this unit of randomisation was consistent with the unit of analysis. Where cross-over trials are used, we will extract mean and standard error of paired t-tests and assess the clinical (pharmacodynamic) suitability of 'wash-out' period and risk of spillover of drug effect for those who received drug treatment first.

Subgroup analysisSubgroup analysis will be undertaken if clinically different intervention are identified or there are clinically relevant differences between subject groups:
• age of participants (adult vs paediatric, preschool vs school age)
• diagnostic classification
• medication dose

Sensitivity analysisSensitivity analysis will be conducted to assess the impact of study quality on the results of meta-analyses. For example, we will test to see if studies with high rates of loss to follow-up or inadequate blinding are more likely to show positive outcomes.

 
Table 2. Outcome measures used in included trials

Outcome measureNCT00609531BarthelemyKingBuchsbaumHollander

2005
LeventhalMcDougleSugieHollander 2012

Core features of autism

1Behavioural Assessment Scalex

2Behaviour Summarized Evaluation Scalex

3Repetitive Behavior Scale–Revisedxx

Obsessive-compulsive behaviour

4Yale-Brown Obsessive Compulsive Scale (Y-BOCS)x (combined obsession and compulsion score)x (combined obsession and compulsion score)x (obsession and compulsion subscales reported separately)x (compulsion subscale only)x Modified Y-BOCS used (combined and separate subscales reported)x (compulsion subscale only)

5CGI-I for obsessive-compulsive symptomsxx

Anxiety

6Hamilton Rating Scale for Anxiety (HAM-D)x

Depression

7Hamilton Rating Scale for Depressionxx

Behaviour

8Ritvo-Freeman Real Life Rating Scalexx

9Vineland Adaptive Behaviour Scalesx

10Clinical Global Impression Scale (CGI)x CGI-Ix

CGI-I
x

CGI-I
x

CGI-AD
x

CGI-I
x

Genotype specific only
x

CGI-I

11Aberrant Behavior Checklist (ABC)xx (irritability subscale only)

12Connors' Abbreviated Parent and Abbreviated Teacher Questionnairesx

Aggression

13Brown Aggression Scalex

Other standardised outcomes

14Merrill-Palmer Scale of Mental Testsx

15Wechsler Intelligence Scale for Childrenx

16Alpern-Boll Developmental Profilex

Adverse events

17Fluoxetine Side Effects Checklistx

18Suicidality Subscale, Overt Aggression Scale - Modifiedx

 Clinical Global Improvement Scale Adapted to Global Autism (CGI-AD)
Clinical Global Impression - Improvement scale (CGI-I)
Clinical Global Impression - Severity scale (CGI-S)