aripiprazole (Abilify)

the boxed warning, part one: deaths among older people with dementia-related psychosis

the first part of the boxed warning says that elderly patients with dementia-related psychosis treated with antipsychotic drugs are at an increased risk of death, and that aripiprazole is not approved for them [label, boxed warning and 5.1]. the number behind it is a pooled analysis of 17 placebo-controlled trials, mostly of atypical antipsychotics and mostly about 10 weeks long: the death rate was about 4.5% on drug against about 2.6% on placebo, 1.6 to 1.7 times higher, with most deaths cardiovascular, such as heart failure or sudden death, or infectious, such as pneumonia [fda-dementia].

unlike some labels in the class, this one also reports aripiprazole’s own trials in that population. in three 10-week placebo-controlled studies of 938 elderly people with psychosis associated with alzheimer’s disease, mean age 82, the reactions at least twice as common on aripiprazole were lethargy (5% against 2%), somnolence (8% against 3%), incontinence (5% against 1%), excessive salivation (4% against 0%) and lightheadedness (4% against 1%), and the label says safety and efficacy in that group "have not been established" [label 5.1]. in the same studies there were more strokes and transient ischaemic attacks, including fatal ones, on aripiprazole, with a statistically significant dose-response in the fixed-dose study, mean age 84 [label 5.2]. the geriatric section adds that the schizophrenia, mania and depression trials did not include enough people over 65 to say whether they respond differently [label 8.5].

the boxed warning, part two: suicidal thoughts and behaviors, by age

the second part is the antidepressant class warning, which aripiprazole carries because it is approved as an add-on in depression. it says antidepressants increased the risk of suicidal thoughts and behavior in children, adolescents and young adults in short-term studies, that the studies "did not show an increase" above age 24 and showed "a reduction in risk" at 65 and older, and that everyone starting one should be watched closely [label, boxed warning and 5.3]. the label states the difference against placebo per 1,000 people treated: under 18, 14 additional cases; 18 to 24, 5 additional; 25 to 64, 1 fewer; 65 and older, 6 fewer [label 5.3, table 3]. the pooled trials were 24 short-term trials of 9 antidepressants in more than 4,400 children and 295 trials of 11 antidepressants in more than 77,000 adults, median two months; none was of aripiprazole itself. no suicides occurred in the pediatric trials; the adult trials had too few to conclude anything about suicide [label 5.3].

so the warning has an age boundary that is printed in the label and rarely repeated [age-ceiling]. the label says the risk beyond several months is unknown, that prescribers should consider changing the regimen, "including possibly discontinuing the medication", in anyone whose depression is persistently worse, and that "aripiprazole is not approved for use in treating depression in the pediatric population" [label 5.3]. that is an instruction to prescribers about watching, not an instruction to patients about stopping. a separate warning says the possibility of suicide "is inherent" in psychosis, bipolar disorder and depression, and that prescriptions should be written for the smallest quantity consistent with good care [label 5.14].

the gambling and impulse-control warning

warning 5.7 is the one specific to this drug. "postmarketing case reports suggest that patients can experience intense urges, particularly for gambling, and the inability to control these urges while taking aripiprazole." the less frequent urges it lists are sexual urges, shopping, eating or binge eating, "and other impulsive or compulsive behaviors". because people "may not recognize these behaviors as abnormal", the label tells prescribers to ask patients or their caregivers specifically about new or intense urges of each kind. it concedes that impulse-control symptoms "can be associated with the underlying disorder", reports that "in some cases, although not all, urges were reported to have stopped when the dose was reduced, or the medication was discontinued", and ends with an instruction to the prescriber: "consider dose reduction or stopping the medication if a patient develops such urges" [label 5.7]. pathological gambling also appears in the post-marketing list, which the label says comes from voluntary reports from a population of uncertain size, so no rate can be estimated [label 6.2]. the warning was added to the label after an fda safety communication in may 2016; that document is cited on this site’s legal and safety record page, which is listed as a source here and not repeated [legal-record].

the one population study in the library is a swedish register study of every person in the country diagnosed with pathological gambling between 2005 and 2019, 3,689 people, each matched to two controls, 11,067 in all. within the subgroup with schizophrenia or a related psychotic disorder, an aripiprazole prescription was associated with a gambling diagnosis at an odds ratio of 3.4; dopamine agonists used in parkinson’s disease, which carry the same kind of warning, showed 3.2. the study has no time dimension, so it cannot show which came first, the subgroup is small and heavily selected, and one author holds a chair funded by the state gambling operator, which the paper discloses [wolfschlag-2021]. the finding is consistent with the case reports and with the drug’s pharmacology; it is not proof, and the library record says so.

the metabolic warning, with the label’s tables

warning 5.6 is the class warning that atypical antipsychotics cause hyperglycemia and diabetes, undesirable lipid changes and weight gain, that each drug has its own profile, and that fasting glucose should be tested at the start and periodically in anyone with diabetes risk factors, with regular weight monitoring [label 5.6]. the label adds that because aripiprazole "was not marketed at the time" the epidemiological studies of the class were done, "it is not known if aripiprazole is associated with this increased risk" [label 5.6]. its own tables are among the smallest in the class, and the controlled data mostly stop at three to six weeks.

• blood sugar, adults [5.6, table 4]: across 13 placebo-controlled monotherapy trials, median exposure 25 days, mean fasting glucose rose 4.4 mg/dl on aripiprazole (1,057 people) against 2.5 on placebo (799), not significantly different. the share going from normal to high (126 mg/dl or more) was 3.8% against 3.6%; from borderline to high, 17.6% against 9.2%. at 24 weeks the mean change was +2.2 mg/dl on aripiprazole against +9.6 on placebo, in 42 and 28 people. as an add-on in depression, +0.7 against +0.8 mg/dl over a median 42 days [5.6, table 5].
• lipids, adults [5.6, table 7]: across 17 trials, the share going from normal to high total cholesterol was 2.5% against 2.8%; fasting triglycerides 7.4% against 7.0%; ldl 0.6% against 0.7%; hdl falling to low 11.4% against 12.5%. as an add-on in depression, triglycerides ran the other way, 9.7% against 4.1% [5.6, table 8]. the label says analyses at 12 and 24 weeks "were limited by small numbers of patients".
• weight, adults [5.6, table 12]: +0.3 kg on aripiprazole (1,673 people) against -0.1 kg on placebo (1,100) over a median 21 to 25 days; at 24 weeks, -1.5 kg against -0.2 kg in 73 and 46 people. the share gaining 7% or more of body weight was 8.1% against 3.2% in schizophrenia (4 to 6 weeks), 2.2% against 2.7% in mania (3 weeks), and 5.2% against 0.6% as an add-on in depression (6 weeks), where the mean change was +1.7 kg against +0.4 kg.
• weight, children and adolescents [5.6, table 13; 8.4]: in schizophrenia and bipolar trials of 4 to 6 weeks, +1.6 kg against +0.3 kg, with 5.2% against 1.6% gaining 7% or more; at 24 weeks, +5.8 kg against +1.4 kg, in 62 and 13 children. in the autism trials, 26.3% against 7.1% gained 7% or more over 8 weeks; in tourette’s, 20.0% against 7.6% over 8 to 10 weeks. in the open-label follow-on, 32.8% of the schizophrenia and bipolar children had gained 7% or more by 26 weeks, not adjusted for growth; adjusted for growth, the mean weight z-score change was +0.09 standard deviations, which the label counts as not clinically significant. in the autism follow-on, children treated for more than nine months had a z-score change of +0.26.
• the comparison across the class [pillinger-2020]: in a network meta-analysis of 100 randomised trials in 25,952 people with schizophrenia, aripiprazole was among the antipsychotics with the most favourable metabolic profile, the least weight gain and the smallest blood-sugar and cholesterol changes, while olanzapine and clozapine caused the most.

akathisia and movement side effects: where aripiprazole’s numbers run high

what the metabolic tables give back, the movement tables take, and the pattern depends on the use. akathisia, an inner restlessness with an urge to move that people often find the hardest side effect to live with, was the only reaction at 5% or more and twice placebo in the adult schizophrenia trials: 8% against 4% [label 6.1]. in adult mania it was 13% against 4%, with sedation 8% against 3%, restlessness 6% against 3%, tremor 6% against 3% and extrapyramidal disorder 5% against 2% [label 6.1, table 14]; added to lithium or valproate, 19% against 5%, and akathisia was the main reason people stopped (5% against 1%) [label 6.1, table 16]. as an add-on in depression it was 25% against 4%, with restlessness 12% against 2% and insomnia 8% against 2% [label 6.1, table 21]. the independent trial in older adults found the same thing: akathisia in 26% against 12% [lenze-2015].

in children the movement figures are higher and rise with dose. in the adolescent schizophrenia trial, extrapyramidal events other than akathisia were 25% against 7%, and by dose 5.0% on placebo, 13.0% at 10 mg and 21.6% at 30 mg [label 6.1]. in the pediatric bipolar trial, 26% against 5%, and by dose 3.1%, 12.2% and 27.3%, with somnolence 23% against 3% and salivary hypersecretion 6% against 0% [label 6.1, table 17]. in the autism trials, 18% against 2%, with tremor 10% against 0% and drooling 9% against 0% [label 6.1, table 18]. the label notes that the rating scales used to measure these effects objectively often did not show a difference from placebo in adults, while the counts of reported events did [label 6.1].

tardive dyskinesia [5.5] is the separate, potentially irreversible movement disorder that can follow long antipsychotic treatment. the label says it is impossible to predict who will develop it, that whether antipsychotics differ in their potential to cause it is unknown, that the risk rises with duration and cumulative dose, and that chronic treatment should be reserved for chronic illness with no equally effective, less harmful alternative, at the smallest dose and shortest duration that works, with the need reassessed periodically [label 5.5]. neuroleptic malignant syndrome [5.4], a rare and potentially fatal reaction with high fever, rigid muscles, confusion and unstable pulse or blood pressure, has occurred with aripiprazole; the label says stop at once and treat in hospital.

the other warnings, in plain words

the warnings section runs from 5.1 to 5.15. beyond those above:

• low blood pressure and falls [5.8, 5.9]: orthostatic hypotension 1% against 0.3% in 2,467 adults, syncope 0.5% against 0.4%; caution in heart or cerebrovascular disease, dehydration and with blood-pressure drugs. antipsychotics "may cause somnolence, postural hypotension, motor and sensory instability, which may lead to falls", so the label asks for fall-risk assessment at the start and periodically on long-term treatment.
• drowsiness and driving [5.12]: somnolence including sedation in 11% of adults against 6%, and in 24% of children against 6%; it led 0.3% of adults and 3% of children to stop. in the fixed-dose schizophrenia trials it was the only reaction with a dose relationship, 12.6% at 30 mg against 7.1% on placebo [6.1].
• white cells [5.10]: leukopenia, neutropenia and agranulocytosis have been reported with antipsychotics including aripiprazole; anyone with a history of low white cells should have blood counts checked frequently in the first months.
• seizures [5.11]: 0.1% of adults (3 of 2,467) and 0.1% of children (1 of 732) in the short-term trials, with people with a seizure history excluded.
• swallowing and temperature [5.13, 5.15]: antipsychotics can impair the body’s ability to cool itself and can cause oesophageal dysmotility and aspiration, "a common cause of morbidity and mortality in elderly patients, in particular those with advanced alzheimer’s dementia".
• the enzymes that change the dose [2.6, 7.1, table 1]: aripiprazole is cleared by two liver enzymes, cyp2d6 and cyp3a4, and the label sets dose multipliers rather than cautions. with a strong cyp2d6 inhibitor (it names quinidine, fluoxetine and paroxetine) or a strong cyp3a4 inhibitor (itraconazole, clarithromycin), half the usual dose; with both, a quarter; with a strong cyp3a4 inducer (carbamazepine, rifampin), double the dose over one to two weeks; in a known cyp2d6 poor metabolizer, half, and a quarter with a strong cyp3a4 inhibitor. the elimination half-life is about 75 hours, about 146 hours in poor metabolizers, and steady state takes about 14 days [12.3]. as an add-on in depression, the label says to use the depression dosing without these adjustments [2.6]. no adjustment is needed with lithium, valproate, lamotrigine, lorazepam or sertraline [7.2].
• benzodiazepines [7.1]: with lorazepam, "the intensity of sedation was greater with the combination" and orthostatic hypotension was greater than with lorazepam alone; the label says to monitor both. blood-pressure drugs: aripiprazole may enhance their effect.
• pregnancy and breastfeeding [8.1, 8.2]: babies exposed in the third trimester are at risk of movement and withdrawal symptoms after birth (agitation, abnormal muscle tone, tremor, somnolence, breathing and feeding problems), some recovering within hours and others needing prolonged hospital care. published studies "have not established a drug-associated risk of major birth defects, miscarriage, or adverse maternal or fetal outcomes"; a medicaid study of 9,258 women exposed to antipsychotics in pregnancy "did not indicate an overall increased risk for major birth defects". animal studies at 10 and 19 times the maximum human dose showed fetal harm. exposure in pregnancy "can have variable effects on milk supply".
• not a controlled substance [9.1 to 9.3]: aripiprazole has not been systematically studied for abuse or dependence; monkeys showed withdrawal signs on abrupt stopping, and the clinical trials showed no drug-seeking.
• overdose [10]: no fatality has been reported with aripiprazole alone; the largest known ingestion with a known outcome was 1,260 mg, 42 times the maximum daily dose, with full recovery. the label asks for an electrocardiogram.

what the approval trials measured, and for how long

schizophrenia: five placebo-controlled trials of four or six weeks in acutely relapsed inpatients, 414, 404, 420 and 367 people in the four that worked; "one study, the smallest, did not" distinguish the drug from placebo [label 14.1]. the fourth trial tested 2, 5 and 10 mg, and only 10 mg beat placebo; across the trials "there was no evidence that the higher dose groups offered any advantage over the lowest dose group" [label 14.1], which is why the dosing section says doses above 10 or 15 mg "were not more effective" [label 2.1]. three trials carried a risperidone or haloperidol arm "not designed to allow for a comparison" [label 14.1]. one 26-week trial randomised 310 people who had been stable on other antipsychotics to 15 mg or placebo and measured time to relapse, which was longer on the drug [label 14.1]. in adolescents, one 6-week trial of 302 outpatients, where 30 mg was not better than 10 mg [label 14.1, table 23].

mania: four 3-week placebo-controlled trials in hospitalised adults, 268, 248, 480 and 485 people, and one 6-week trial of 384 people adding it to lithium or valproate [label 14.2]. in the maintenance trials, 161 adults stable on the drug were randomised to continue or switch to placebo: 19 mood events against 36, with the difference in manic episodes (6 against 19) and not in depressive ones (9 against 11); in the 52-week add-on trial of 337 people, 25 events against 43, again manic (7 against 19) rather than depressive (14 against 18) [label 14.2]. the label states that pattern without comment. in children, one 4-week trial of 296 [label 14.2].

depression: two 6-week trials, 381 and 362 adults, whose depression had not responded to one to three prior antidepressants and then to eight weeks of a prospective one. added to that antidepressant at a mean final dose of about 11 mg, aripiprazole beat placebo on the montgomery-asberg scale by 2.84 and 3.01 points [label 14.3, table 25]. "a smaller mean reduction" was seen in men than women [label 14.3]. that is the whole approved record for the use for which most of the library rows exist.

autism and tourette’s: two 8-week trials in 98 and 218 children for irritability, with more than 75% under 13; a maintenance trial in which 85 stabilised children were randomised to continue or switch to placebo for 16 weeks did not establish maintenance efficacy [label 14.4; 8.4]. two trials of 8 and 10 weeks in 133 and 61 children for tics [label 14.5]. maintenance efficacy in children for every use is either unestablished or "extrapolated from adult data" [label 8.4]. the pediatric section also reports juvenile animal studies in which rats and dogs showed harm at every dose tested, with "no safety margin" at the lowest dose relative to the exposure of an adolescent on 15 mg; all effects reversed after two months off the drug [label 8.4].

side effects, with the label’s numbers

across 1,843 adults on aripiprazole against 1,166 on placebo in the short-term schizophrenia and mania trials: headache 27% against 23%, agitation 19% against 17%, insomnia 18% against 13%, anxiety 17% against 13%, nausea 15% against 11%, constipation 11% against 7%, vomiting 11% against 6%, dizziness 10% against 7%, akathisia 10% against 4%, dyspepsia 9% against 7%, sedation 7% against 4%, fatigue 6% against 4%, extrapyramidal disorder, tremor, somnolence and restlessness each 5% against 3%, blurred vision 3% against 1% [label 6.1, table 15]. as an add-on in depression, 371 against 366: akathisia 25% against 4%, restlessness 12% against 2%, fatigue 8% against 4%, insomnia 8% against 2%, blurred vision 6% against 1%, somnolence 6% against 4%, constipation 5% against 2% [label 6.1, table 21]. among children, pooled: somnolence 16% against 4%, headache 12% against 10%, fatigue 10% against 2%, sedation 9% against 2%, tremor 9% against 1%, nausea and vomiting 8% each, increased appetite 7% against 3% [label 6.1, table 20].

dropout for side effects: 6% against 2% as an add-on in depression, 12% against 6% added to lithium or valproate, 5% against 2% in adolescent schizophrenia, 7% against 2% in pediatric bipolar, 10% against 8% in autism, 7% against 1% in tourette’s; the label gives no dropout figure for the adult schizophrenia or mania monotherapy trials [label 6.1]. in the 26-week schizophrenia trial, tremor was 8% against 2%, mostly mild, early and short [label 6.1]. prolactin runs the other way from most antipsychotics: "blood prolactin decreased" is listed as frequent and "blood prolactin increased" as rare [label 6.1]. the aripiprazole label has no qt warning; qt prolongation appears only as a rare laboratory finding and among overdose reports [label 6.1, 10].

what the research says, in both directions

in favour. in the network meta-analysis of 402 randomised trials in 53,463 adults with schizophrenia, all 32 antipsychotics including aripiprazole reduced symptoms more than placebo, the drugs differed more in side effects than in effectiveness, and aripiprazole was one of only five that significantly improved quality of life against placebo (standardised mean difference -0.49) [huhn-2019]. on metabolic effects it sat among the most favourable of 18 drugs [pillinger-2020]. in 181 adults aged 60 or over whose depression had not remitted on an antidepressant, adding aripiprazole produced remission in 44% against 29% on placebo [lenze-2015]. in 619 adults aged 60 or over with treatment-resistant depression, adding aripiprazole improved well-being more than switching to bupropion (a 4.83-point against 2.04-point gain over 10 weeks) and 28.9% remitted [lenze-2023]. a meta-analysis of 16 placebo-controlled trials in 3,480 people found that adding an atypical antipsychotic to an antidepressant roughly doubled the odds of remission (odds ratio 2.00) [nelson-2009].

against, or smaller than it sounds. in the same meta-analysis, dropout because of side effects was nearly four times as likely as on placebo (odds ratio 3.91), and the funding of the analysis and of several included trials could not be read from open sources [nelson-2009]. the late-life trial’s remission gain came with akathisia in 26% against 12% [lenze-2015]. the label’s own depression trials show a difference of about three points on a clinician scale over six weeks, with one in four people restless [label 14.3; 6.1, table 21]. the schizophrenia record is four to six weeks of placebo-controlled data plus one 26-week relapse trial, and the maintenance trials in bipolar disorder prevented manic episodes without a measurable effect on depressive ones [label 14.1, 14.2]. the children’s weight and movement figures are larger than the adults’, maintenance efficacy in children is unestablished or extrapolated, and the juvenile animal studies had no safety margin [label 5.6; 6.1; 8.4]. the impulse-control warning rests on case reports and one register study that cannot show which came first [label 5.7; wolfschlag-2021].

the warnings, measured. the dementia warning’s size is the class pooled analysis, 4.5% against 2.6% over about 10 weeks, plus aripiprazole’s own stroke signal with a dose-response [fda-dementia; label 5.2]. the suicidality warning’s size is the label’s own table, 14 extra cases per 1,000 under 18 and none above 24 [label 5.3]. the metabolic warning’s size, for this drug, is the smallest in the class on the independent comparison [pillinger-2020]. the akathisia figure, 25% against 4% in the depression trials, is the label’s own [label 6.1, table 21].

questions people ask

what does the abilify (aripiprazole) black box warning mean?

it has two parts. first, older people with dementia-related psychosis who are treated with antipsychotics die at a higher rate than on placebo: about 4.5% against 2.6% over about 10 weeks across 17 pooled trials, mostly from heart or infectious causes, and aripiprazole is not approved for them; aripiprazole’s own trials in that group also found more strokes, with a dose-response. second, the antidepressant class warning: in pooled trials, suicidal thoughts and behaviors increased in people under 25 (14 extra cases per 1,000 under 18, 5 extra at 18 to 24) and not above 24, with 6 fewer per 1,000 at 65 and older. aripiprazole carries it because it is approved as an add-on in depression; it is not approved for depression in children. the warning says to watch everyone starting it, especially in the first months and after dose changes.

does abilify cause gambling or compulsive behaviour?

the label says post-marketing case reports suggest people can experience intense urges, particularly for gambling, and an inability to control them while taking aripiprazole, with sexual urges, shopping and binge eating reported less often. it tells prescribers to ask patients and caregivers specifically about new urges, because people may not recognise them as abnormal, notes that in some cases the urges stopped when the dose was reduced or the drug stopped, and tells prescribers to consider dose reduction or stopping if urges develop. the label gives no rate; voluntary reports cannot produce one. the one population study in the library, a swedish register study of 3,689 people with gambling disorder, found an aripiprazole prescription associated with the diagnosis at an odds ratio of 3.4 within people with psychotic disorders, but could not show which came first. the question for a prescriber is whether anything about spending, gambling, eating or sexual behaviour has changed since starting.

does abilify cause weight gain?

less than most antipsychotics in adults, more in children. the label’s adult tables show +0.3 kg against -0.1 kg on placebo over about three weeks and -1.5 kg against -0.2 kg at 24 weeks; 8.1% against 3.2% gained 7% or more of body weight in the schizophrenia trials, 2.2% against 2.7% in mania, and 5.2% against 0.6% as an add-on in depression, where the mean change was +1.7 kg against +0.4 kg. in children, +1.6 kg against +0.3 kg over 4 to 6 weeks, +5.8 kg against +1.4 kg at 24 weeks, and 26.3% against 7.1% of children with autism gained 7% or more over 8 weeks. the independent comparison of 18 antipsychotics placed aripiprazole among the most favourable for weight, blood sugar and cholesterol, with olanzapine and clozapine the least.

what are the most common abilify side effects?

in the label’s adult schizophrenia and mania trials, 1,843 people against 1,166 on placebo: headache 27% vs 23%, agitation 19% vs 17%, insomnia 18% vs 13%, anxiety 17% vs 13%, nausea 15% vs 11%, constipation 11% vs 7%, vomiting 11% vs 6%, dizziness 10% vs 7%, akathisia 10% vs 4%, sedation 7% vs 4%. as an add-on in depression: akathisia 25% vs 4%, restlessness 12% vs 2%, insomnia 8% vs 2%, fatigue 8% vs 4%, blurred vision 6% vs 1%. in children: somnolence 16% vs 4%, fatigue 10% vs 2%, sedation 9% vs 2%, tremor 9% vs 1%, with extrapyramidal effects rising with dose. the label’s one-line summary for adults with schizophrenia is a single reaction at twice placebo, akathisia.

what is akathisia, and how common is it on abilify?

akathisia is an inner restlessness with a compelling urge to move, often felt in the legs, which people frequently describe as the hardest antipsychotic side effect to live with. in the label’s trials it was 8% against 4% in adult schizophrenia, 13% against 4% in mania, 19% against 5% when added to lithium or valproate, and 25% against 4% when added to an antidepressant, where restlessness was a further 12% against 2%. the independent trial in adults over 60 found 26% against 12%. the label counts it among the symptoms that can precede worsening depression or suicidality in people on antidepressants, which is one reason it tells prescribers to watch for it in the first months and after dose changes.

does abilify work for depression?

as an add-on, by a modest margin. the label’s two 6-week trials, 381 and 362 adults whose depression had not responded to prior antidepressants, found aripiprazole added to the antidepressant beat placebo by 2.84 and 3.01 points on the montgomery-asberg scale, with akathisia in 25% against 4% and dropout for side effects 6% against 2%. independent trials in older adults found remission in 44% against 29% on placebo at a cost of akathisia in 26% against 12%, and in a 619-person trial adding aripiprazole beat switching to bupropion on well-being with 28.9% remitting. a meta-analysis of 16 trials found adding an atypical antipsychotic roughly doubled the odds of remission and nearly quadrupled the odds of stopping for side effects. aripiprazole is not approved for depression on its own, or for depression in children.

what happens if you stop abilify suddenly?

the label has no discontinuation warning for aripiprazole and gives no taper. what it does say: tardive dyskinesia may remit, partially or completely, if antipsychotic treatment is withdrawn, while treatment itself can mask it; monkeys showed withdrawal signs on abrupt stopping, and the human trials were not designed to look; when switching antipsychotics, immediate stopping may be acceptable for some people with schizophrenia and gradual stopping better for others; and the need for continued treatment should be reassessed periodically. the elimination half-life is about 75 hours, so the drug leaves the body over a couple of weeks rather than days. none of that is a reason to stop on one’s own; the condition being treated is the larger risk, and a plan belongs with a prescriber.

how is abilify different from seroquel (quetiapine) or risperdal (risperidone)?

all three carry the dementia boxed warning and the metabolic and tardive dyskinesia class warnings. the label facts that differ: aripiprazole’s weight, glucose and lipid tables are among the smallest in the class and prolactin tends to fall rather than rise, while akathisia and restlessness run high, especially as an add-on in depression; it has the impulse-control warning the others do not; and its label carries no qt warning. the quetiapine and risperidone labels each have their own explainer on this site [quetiapine-explainer; risperidone-explainer], and the drugs are compared side by side on the comparison pages [compare-quetiapine; compare-risperidone]. nothing in any of the labels is a reason to prefer one drug; that is a prescriber’s decision.

sources for this section

the library rows link to our one-page summary of each study, with the original paper, its funding and our evidence grade on that page.

1. aripiprazole tablets prescribing information, fda label on dailymed (set id 1d593ada-256d-4dec-9a26-2a6a67d97205, version 13, effective july 31, 2026; remedyrepack, anda 207105): boxed warning; sections 1, 2.1 to 2.6 with table 1, 4, 5.1 to 5.15 with tables 3 to 13, 6.1 with tables 14 to 21, 6.2, 7.1 table 22, 7.2, 8.1, 8.2, 8.4, 8.5, 9.1 to 9.3, 10, 12.3, 14.1 to 14.5 with tables 23 to 27
2. drugs@fda, nda 021436 (abilify, otsuka), approved november 15, 2002
3. ahrq medical expenditure panel survey, prescribed medicines public-use files 2014 (hc-168a) and 2024 (hc-254a): people with at least one aripiprazole purchase and prescriptions, weighted; the 2014 to 2024 prescription change is within the survey’s margin of error; computed by veisund, method on the usage section of this page
4. the antipsychotic dementia boxed warning, read for the library from the risperidone label: 17 pooled trials, about 4.5% against 2.6% deaths, mostly cardiovascular or infectious (library row)
5. the antidepressant suicidality warning’s age ceiling, from the labels themselves, as a library row (library row)
6. wolfschlag m, håkansson a. increased risk for developing gambling disorder under the treatment with pramipexole, ropinirole, and aripiprazole: a nationwide register study in sweden. plos one 2021 (3,689 cases, 11,067 people; the library record explains why it is weak evidence and discloses the funding of one author’s chair) (library row)
7. huhn m, et al. comparative efficacy and tolerability of 32 oral antipsychotics for the acute treatment of adults with multi-episode schizophrenia: a systematic review and network meta-analysis. lancet 2019 (402 trials, 53,463 people) (library row)
8. pillinger t, et al. comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia: a systematic review and network meta-analysis. lancet psychiatry 2020 (100 trials, 25,952 people) (library row)
9. lenze ej, et al. efficacy, safety, and tolerability of augmentation pharmacotherapy with aripiprazole for treatment-resistant depression in late life: a randomised, double-blind, placebo-controlled trial. lancet 2015 (181 adults aged 60 and over) (library row)
10. lenze ej, et al. antidepressant augmentation versus switch in treatment-resistant geriatric depression (optimum). new england journal of medicine 2023 (619 adults aged 60 and over) (library row)
11. nelson jc, papakostas gi. atypical antipsychotic augmentation in major depressive disorder: a meta-analysis of placebo-controlled randomized trials. american journal of psychiatry 2009 (16 trials, 3,480 people; funding not readable from open sources) (library row)
12. aripiprazole: the legal and safety record page on this site, which carries the fda’s may 2016 safety communication on impulse-control problems among its sources (listed here as a source only; nothing from it is repeated on this page)
13. what the quetiapine (seroquel) warnings mean, the companion explainer on this site
14. what the risperidone (risperdal) warnings mean, the companion explainer on this site
15. aripiprazole vs quetiapine, the side-by-side comparison page on this site
16. aripiprazole vs risperidone, the side-by-side comparison page on this site

Akathisia on the aripiprazole label

The aripiprazole FDA label names akathisia in its Adverse Reactions section, in the list below.

Most-reported reactions

• Off label use8,566
• Drug ineffective8,544
• Weight increased8,027
• Product use in unapproved indication5,351
• Anxiety5,268
• Fatigue4,437
• Suicidal ideation4,154
• Depression4,147
• Nausea4,127
• Insomnia4,009