Off-Label Use of Atypical Antipsychotics

Off-Label Use of Atypical Antipsychotics: An Update

Summary of the Agency for Healthcare Research and Quality “AHRQ” Report

David L. Fogelson, M.D.

Clinical Professor of Psychiatry

David Geffen School of Medicine at UCLA

And The Semel Institute for Neuroscience and Human Behavior at UCLA

Suggested Citation for most material in this presentation

  • Maglione M, Ruelaz Maher A, Hu J, Wang Z, Shanman R, Shekelle PG, Roth B, Hilton L, Suttorp MJ, Ewing BA, Motala A, Perry T.
  • Off-Label Use of Atypical Antipsychotics: An Update. Comparative Effectiveness Review No. 43. (Prepared by the Southern California Evidence-based Practice Center under Contract No. HHSA290-2007-10062-1.) Rockville, MD: Agency for Healthcare Research and Quality. September 2011.
  • Available at:

What is the AHRQ Report?

  • A review of the expanding use of atypical antipsychotics for off label indications
  • Examined the efficacy, comparative effectiveness, benefits, and adverse effects
  • 170 studies of efficacy or comparative effectiveness
  • 180 studies of  adverse effects
  • The review includes studies through May 2011
    • I have reviewed additional articles for this presentation
  • Intended to inform discussion of options and decision making
  • Not meant to be construed as a clinical guideline

What was Reviewed?

  • All extant atypicals were used as search terms
  • Searched all controlled trials comparing an atypical to placebo, to another atypical, or other pharmacotherapy, for off-label conditions
  • Conditions
    • Anxiety disorders
    • ADHD
    • Dementia & Agitation associated with Aging
    • Major Depression
    • Eating Disorders
    • Insomnia
    • OCD
    • PTSD
    • Personality Disorders
    • Substance Abuse
    • Tourette’s Syndrome
  • Observational Studies included if > 1,000 subjects

What are FDA approved Atypicals?

  • Aripiprazole (Abilify)
  • Asenapine (Saphris)
  • Clozapine (Clozaril)
  • Iloperidone (Fanapt)
  • Lurasidone (Latuda)
  • Olanzapine (Zyprexa)
  • Paliperidone (Invega)
  • Quetiapine (Seroquel)
  • Risperidone (Risperdal)
  • Ziprasidone (Geodon)


  • By 2001 96% of prescribed antipsychotics were atypicals
  • 90% prescribed to children are atypicals
  • A class effect of atypicals cannot be assumed for any mental disorder
  • For most atypicals evidence lacking for effectiveness and comparative effectiveness for many indications
  • Risk benefit ratio must be considered
  • Adverse Effects may outweigh benefit

Conclusions about Efficacy, from most effective to least

  • Improve Behavioral Symptoms of Dementia
  • Several have approved indications in Depression and others show efficacy
  • Growing evidence for efficacy in:
    • Obsessive Compulsive Disorder
    • Combat-related PTSD
    • Generalized Anxiety Disorder
  • Evidence is too limited to estimate benefit in Borderline Personality Disorder
  • Evidence is insufficient for treatment of Tourette’s

Conclusions about Adverse Reactions

  • Risk of death in patients 65 and older increased by atypical and typical antipsychotics
    • Recent data contradicts this assertion
  • Weight Gain, most severe with olanzapine
  • Risks for endocrine/metabolic abnormalities & diabetes are less certain, but are present
  • Sedative Effects are common in all age groups
  • Urinary retention is found in the elderly

Conclusions about Inefficacy

  • Evidence is “stronger” that there is no increase in body weight in Anorexia Nervosa (even though weight gain is a common adverse side effect)
  • Evidence is “stronger” that there is inefficacy in Substance Abuse Disorders
  • Most Commonly Prescribed Off-Label
    • Risperidone, Quetiapine, & Olanzapine
  • Most Commonly Studied for Off-Label use
    • Risperidone, Olanzapine, & Quetiapine
  • The Least Studied
    • Aripiprazole, Ziprasidone
  • Not studied for Off-Label use
    • Asenapine, Iloperidone, Lurasidone, & Paliperidone
  • Due to severe blood dyscrasia risk with Clozapine it was not included in this review
  • Since 2005 FDA regulatory warning about severe risks in elderly, use has declined in the elderly
  • Even so, atypical use is higher in long-term care settings than in the community

Does Effectiveness/Harm vary by race, ethnicity, gender, or age?

  • Few studies adequately address this question
  • One study examined effect by gender
  • No studies stratified results by race, ethnicity
  • A couple studies did stratify by age

Strength of Evidence for Efficacy or Inefficacy of Atypicals for Off-Label Indications

  • AHRQ Report uses a 4 point “Strength of Evidence Scale” to indicate confidence in effect
  • Effect is defined as either “medication can improve symptoms or medication does not improve symptoms”
    • 0: evidence unavailable or does not permit a conclusion
    • 1: Low confidence that the evidence reflects the true effect; further research is likely to change confidence level and estimate of effect
    • 2: Moderate confidence, evidence reflects true effect; further research may change confidence level and may change estimate of effect
    • 3: High confidence evidence reflects true effect; further research very unlikely to change confidence level in the estimate of effect

Evidence Supports use of Five Atypicals

  • Aripiprazole
  • Olanzapine
  • Quetiapine
  • Risperidone
  • Ziprasidone
  • The next section will review the Evidence of their Efficacy in specific disorders

Dementia, confidence levels

  • Rated Strength of Evidence in Overall (global scores) symptoms, Agitation, & Psychosis
  • Overall
    • Aripiprazole  2; Olanzapine 1; Quetiapine 1; Risperidone 3
  • Agitation
    • Aripiprazole 1; Olanzapine 2; Quetiapine no trials (a more recent meta-analysis suggests 1); Risperidone 3
  • Psychosis
    • Aripiprazole 1; Olanzapine 1; Quetiapine no trials; Risperidone 3

Dementia, comments on findings

  • General improvement in behavioral symptoms
  • Effect sizes are small
    • Less than .2 standard deviations of difference between treatment and control groups
  • No class effect
    • Risperidone>Aripiprazole=Olanzapine>Quetiapine

Major Depressive Disorder (MDD), confidence levels

  • Rated Strength of Evidence for Monotherapy and as Augmentation for MDD only; 36 trials
  • Monotherapy, MDD
    • Aripiprazole  no trials; Olanzapine 2 does not improve symptoms; Quetiapine 2, NNT =13, remission & 6 for response; Risperidone no trials
  • Augmentation of SSRI/SNRI, MDD
    • Aripiprazole  approved indication; Olanzapine Approved Indication for Rx resistant; Quetiapine Approved Indication; Risperidone 2, NNT = 8, remission & 7 for response; Ziprasidone 1

Bipolar Depression (BPD) Not part of AHRQ Review

  • Monotherapy, BPD
    • Aripiprazole not reviewed; Olanzapine not reviewed; Quetiapine Approved Indication; Risperidone not reviewed; Lurasidone Approved Indication
  • Augmentation, BPD
    • Aripiprazole not reviewed; Olanzapine Approved Indication as Symbyax; Quetiapine not reviewed; Risperidone not reviewed

Obsessive Compulsive Disorder, Augmentation of SSRIs

  • Augmentation of SSRIs, response is a 30% improvement on YBOCS
    • Aripiprazole  no trials
    • Olanzapine, confidence level = 1
    • Quetiapine, confidence level = 1, weak effect
    • Risperidone, level = 2, NNT = 5, OR = 3.9
    • Ziprasidone, level = 1, weak effect

Obsessive Compulsive Disorder, Comments about Findings

  • Risperidone is most effective
  • Olanzapine and Ziprasidone have a weak effect
    • May have similar effect, confidence level = 1
  • Other meds have not been shown to be effective
  • Clozapine may exacerbate OCD Symptoms

• Clozapine may exacerbate OCD Symptoms •Psychopharmacology (Berl). 2013 May 10. [Epub ahead of print] Influence of polymorphisms in genes SLC1A1, GRIN2B, and GRIK2 on clozapine-induced obsessive-compulsive symptoms. Cai JZhang WYi ZLu WWu ZChen JYu SFang Y,

Zhang C.Schizophrenia Program, Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, 200030, People’s Republic of China.

Post Traumatic Stress Disorder, Adjunctive

  • Ten trials
  • Aripiprazole, no trials
  • Olanzapine, evidence unavailable or does not permit a conclusion
  • Quetiapine, evidence unavailable or does not permit a conclusion
  • Risperidone, confidence level = 2, combat related (DOD guidelines disagree based on 2010 VA study in 247 veterans, found ineffective)

Comments about PTSD

  • Evidence insufficient to determine effect of Risperidone in abused women with PTSD
  • Evidence for Olanzapine and Quetiapine insufficient for analysis

Generalized Anxiety Disorder

  • Aripiprazole, no trials
  • Olanzapine, no trials
  • Quetiapine, confidence level = 2
    • Three large trials
    • 26% greater likelihood of responding (> 50% improvement on HAMA) than placebo
  • Risperidone, no trials

Borderline Personality Disorder, confidence levels

  • 12 trials in Personality Disorders
  • Aripiprazole, 1
  • Olanzapine, mixed results in 7 trials, does not permit a conclusion
  • Quetiapine, 1
  • Risperidone, no trials
  • Ziprasidone, 1, ineffective

Personality Disorders: Schizotypal

  • Two small trials with Risperidone
  • One superior to placebo
  • One no difference from placebo
  • Unable to reach a conclusion, data insufficient

Anorexia Nervosa, body weight restitution, confidence levels represent inefficacy

  • Six Trials
  • Aripiprazole, no trials
  • Olanzapine, 2, ineffective
  • Quetiapine, 1, ineffective
  • Risperidone, no trials

Substance Abuse; confidence levels represent inefficacy

  • 33 trials
  • Aripiprazole, Alcohol-2, Methamphetamine-1
  • Olanzapine, Alcohol-1, Cocaine-1
  • Quetiapine, Alcohol-1
  • Risperidone, Cocaine-1, Methadone-1

Other Possible Indications

  • Tourette’s syndrome, data insufficient to permit conclusions
    • One small trial indicated Risperidone is at least as effective as pimozide or clonidine
  • Insomnia, very limited data, n = 13, of inefficacy for Quetiapine

Attention Deficit Hyperactivity Disorder

  • ADHD, no co-occurring disorders
    • One trial showed Risperidone superior to placebo for aggression related to ADHD
    • Risperidone may be efficacious
  • ADHD in the context of mental retardation
    • One trial showed Risperidone more effective than methylphenidate
    • Risperidone may be more effective than methylphenidate

Autism Spectrum Disorders

  • Risperidone, FDA approved
    • Treatment of irritability
    • Efficacy was established in 3 short-term trials in children and adolescents (ages 5 to 17 years)
    • symptoms of aggression towards others, deliberate self-injuriousness, temper tantrums, and quickly changing moods
  • Aripiprazole, FDA approved
    • Treatment of irritability
    • Efficacy established in two 8-week trials in pediatric patients (aged 6 to 17 years)
    • symptoms of aggression towards others, deliberate self-injuriousness, temper tantrums, and quickly changing moods

Adverse Effects, Elderly Patients: Dementia

  • Typical and Atypical Antipsychotics increase risk of death in elderly with dementia
  • Number Needed to Harm (NNH) = 100 for Atypicals (Aripiprazole, OLZ, QTP, Risperidone)
    • 1/100 patients died due to treatment
    • Confidence Level 3
  • Typicals elevated mortality rate
    • NNH not calculated
    • Confidence Level 2

Adverse Effects, Elderly Patients: Dementia

Adverse Effects, Elderly Patients: Dementia, Relative Risk

  • Risperidone has increased risk of cerebrovascular accidents, NNH = 34
  • Risperidone and Olanzapine have increased risk for cardiovascular events
    • Risperidone, NNH = 53
    • Olanzapine, NNH = 48
  • Confidence level, 2

Adverse Events, Elderly: EPS, Sedation, & Urinary Infections and Incontinence

  • EPS is common in the elderly, confidence level = 2
    • Risperidone, NNH = 20
    • Olanzapine, NNH = 10
  • Sedative Effects, NNH = 8-16, confidence level =2
  • Fatigue, NNH = 18-21, confidence level = 2
  • Urinary Adverse Effects, confidence level = 1
    • Infections
    • incontinence

Weight and Metabolic Adverse Events, Adults Aged > 64

  • Weight gain is more common in patients taking olanzapine or risperidone than placebo
  • One study shows no metabolic risks with Risperidone
  • One study showed a trend toward increased risk for diabetes with olanzapine
  • Weight Gain in Children and Adolescents; increased risk in patients taking Risperidone in two placebo controlled studies

Weight and Metabolic Adverse Events, Adults Aged 18-64

  • Most promote weight gain
    • Olanzapine has greatest risk
    • OLZ, NNH = 3 versus NNH = 16-35 for other atypicals, confidence level = 3
    • Could not estimate risk for Ziprasidone
  • Increased Risk for Endocrine
    • Diabetes: Quetiapine, Risperidone, Ziprasidone, Olanzapine
    • Elevated Prolactin with Risperidone & Lurasidone
  • Increased Risk for Metabolic Abnormalities
    • Olanzapine in particular
    • Magnitude of risk unknown
    • Confidence level = 1

EPS, Sedative, & Fatigue Adverse Events, Adults Aged 18-64

  • Elevated risk for EPS, confidence level 1
    • Aripiprazole, NNH = 11, akathisia = 7
    • Quetiapine, NNH = 36
    • Ziprasidone, NNH = 11
  • Sedative Side Effects & Fatigue
    • Applies to Aripiprazole, Olanzapine, Quetiapine, Risperidone, & Ziprasidone
    • Sedation NNH = 3-11, confidence level 2
    • Quetiapine, fatigue, NNH = 14-19

What is known about the risk for tardive dyskinesia?

  • Woods SW et al in J Clin Psychiatry 2010; 71(4):463-474
  • in a naturalistic study found that patients exposed to atypicals have 55% of the risk of conventionals for TD.
  • Woods literature review in the same article, found a lower risk, estimated to be 25%.

What is the risk for TD when combining atypicals with conventional antipsychotic medication?

  • D, Woods et al in J Clin Psychiatry 2010; 71(4):463-474
    • naturalistic study found patients treated with a combination of atypical plus conventional antipsychotic have more than twice the risk for developing TD compared to patients exposed to conventionals alone
    • While it is common practice to employ this combination when monotherapy fails, one must consider the risk benefit of this therapy in light of the markedly increased risk for tardive dyskinesia
  • Note that there is little or no evidence that combination therapy is effective.

Gaps in Knowledge

  • Evidence is insufficient to estimate effect of demographic related factors on outcomes of treatment
    • Race
    • Ethnicity
    • Age (with exception of ADRs in patients with dementia)
  • Evidence is insufficient to permit conclusions about optimal dosage and duration of treatment
  • Few head to head comparisons of typical and atypical antipsychotics
    • Either within or between classes
    • For any indication
    • To permit estimating comparative efficacy
  • Adverse Event Reporting not Standardized
    • Unable to perform “global analysis”
    • Unable to understand Risks

What Shall we tell our Patients?

  • Explain limited evidence for off-label indications
  • Explain risk for adverse effects including Tardive Dyskinesia
  • Trade-offs for benefits and risks in Elderly with Dementia
    • Insufficient data to estimate risks based on 2013 data
    • Refers to death and stroke
    • Always consider non-pharmaceutical interventions first
      • Behavioral
      • environmental
  • High Risk for weight gain
    • Discuss nutritional considerations
    • Discuss benefits of exercise
    • Pharmacotherapy to prevent weight gain should be considered
  • Individualize treatment plans
    • No two patients are exactly alike
    • Consider patients’ and significant others’ preferences