olanzapine (Zyprexa)

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

the boxed warning says that elderly patients with dementia-related psychosis treated with antipsychotic drugs are at an increased risk of death, and that olanzapine is not approved for them [label, boxed warning and 5.1]. the class number is a pooled analysis of 17 placebo-controlled trials, mostly of atypical antipsychotics and mostly about 10 weeks long: about 4.5% died on drug against about 2.6% on placebo, 1.6 to 1.7 times the rate, mostly from cardiovascular causes such as heart failure or sudden death, or infectious ones such as pneumonia [label, boxed warning; fda-dementia]. the label adds that observational studies suggest the older, conventional antipsychotics may carry the same risk, and that how much of that is the drug and how much the patients is not clear [label, boxed warning].

unlike most labels in the class, this one also prints an olanzapine-specific figure: in the placebo-controlled trials of olanzapine in elderly people with dementia-related psychosis, 3.5% died against 1.5% on placebo [label 5.1]. those trials also found significantly more strokes and transient ischaemic attacks, including fatal ones, on olanzapine than on placebo [label 5.1], and in 5 of them (1,184 people) falls, somnolence, swelling of the legs, abnormal gait, urinary incontinence, lethargy, weight gain, weakness, fever, pneumonia, dry mouth and visual hallucinations were all more common on olanzapine, with 13% stopping for side effects against 7% [label 8.5]. the label says nothing about the drug being less safe in older people with schizophrenia, where it found no difference in tolerability [label 8.5].

olanzapine does not carry the antidepressant suicidality boxed warning on its own, because it is not approved as an antidepressant on its own; the fixed combination with fluoxetine does, and the label sends the reader to that product’s boxed warning [label, boxed warning]. its own warning 5.2 says the possibility of a suicide attempt is inherent in schizophrenia and bipolar disorder, that high-risk patients need close supervision, and that prescriptions should be for the smallest quantity of tablets consistent with good management, to reduce the risk of overdose [label 5.2].

the metabolic warning, with the label’s tables

warning 5.5 opens by saying atypical antipsychotics as a class cause hyperglycemia, lipid changes and weight gain, and then does something few labels do: it says where olanzapine sits in the class. on blood sugar, "the association between atypical antipsychotics and increases in glucose levels appears to fall on a continuum and olanzapine appears to have a greater association than some other atypical antipsychotics" [label 5.5]. the label asks for fasting glucose and a fasting lipid profile at the start and periodically, and regular weight monitoring [label 5.5, 5.17]. the tables below are the label’s; where it gives a long-term column, so does this page.

• weight, adults [5.5]: in 13 placebo-controlled trials with a median 6 weeks of exposure, people on olanzapine gained a mean 2.6 kg while people on placebo lost 0.3 kg; 22.2% gained 7% or more of their body weight against 3%, and 4.2% gained 15% or more against 0.3%. in long-term studies of at least 48 weeks (2,021 people, median 573 days), the mean gain was 5.6 kg, and 64% had gained 7% or more, 32% had gained 15% or more, and 12% had gained a quarter of their body weight or more. gain was seen across every starting body-mass index, and it rose with dose: 1.9 kg at 10 mg, 2.3 kg at 20 mg and 3.0 kg at 40 mg over 8 weeks.
• weight, the distribution the label prints [5.5, table 6]: among adults who completed 6 months, 10.9% had gained 10 to 15 kg and 4.3% more than 15 kg; among those who completed 12 months, 14.9% had gained 10 to 15 kg and 14.1% more than 15 kg; among the 147 people who completed 36 months, 17.0% had gained 10 to 15 kg, 11.6% 15 to 20 kg and 10.9% more than 20 kg. these are completers, so the people who stopped because of weight gain are not in the columns; the label says 0.2% stopped for that reason in the short trials and 0.4% in the long ones.
• weight, adolescents aged 13 to 17 [5.5, tables 7 and 8]: a mean 4.6 kg gained in a median 3 weeks against 0.3 kg on placebo, with 40.6% gaining 7% or more against 9.8%. in studies of at least 24 weeks (179 adolescents, median 201 days) the mean gain was 11.2 kg, and 89%, 55% and 29% had gained at least 7%, 15% and 25% of their starting weight. the label states that adolescents gain more than adults and tells prescribers that this "may lead them to consider prescribing other drugs first in adolescents" [label 1.1, 1.2, 8.4].
• blood sugar, adults [5.5, table 2]: in trials up to 12 weeks, 2.2% of people with a normal fasting glucose crossed the diabetes threshold of 126 mg/dl, against 3.4% on placebo; in studies of at least 48 weeks, 12.8% did. of people who started borderline (100 to 125 mg/dl), 17.4% crossed 126 within 12 weeks against 11.5%, and 26.0% in the long-term studies. mean fasting glucose rose 2.76 mg/dl against 0.17 over about 3 weeks; in the catie trial, over a median 9.2 months, the mean rise in glucose was 15.0 mg/dl. the label reports hyperglycemia "in some cases extreme and associated with ketoacidosis or hyperosmolar coma or death" with atypical antipsychotics including olanzapine, and diabetic coma and ketoacidosis among post-marketing reports [label 5.5; 6.2].
• lipids, adults [5.5, table 4]: over up to 12 weeks, mean fasting triglycerides rose 20.8 mg/dl on olanzapine while falling 10.7 on placebo, total cholesterol rose 5.3 while falling 6.1, and ldl rose 3.0 while falling 4.3. triglycerides rose by 50 mg/dl or more in 39.6% against 26.1% in the short trials and in 61.4% in the long ones; total cholesterol crossed 240 mg/dl from normal in 2.8% against 2.4% short-term and 14.8% long-term, and from borderline in 23.0% against 12.5% short-term and 55.2% long-term. in catie, triglycerides rose a mean 40.5 mg/dl over 9.2 months. the label notes that very high triglycerides, over 500 mg/dl, have been observed, and random readings over 1,000 mg/dl have been reported since marketing.
• blood sugar and lipids, adolescents [5.5, tables 3 and 5]: total cholesterol rose a mean 12.9 mg/dl against 1.3, ldl 6.5 against 1.0, and triglycerides 28.4 against a fall of 1.1 over 3 to 6 weeks; 37.0% had a triglyceride rise of 50 mg/dl or more against 15.2%. fasting glucose shifts to the diabetes threshold were rare at 6 weeks (0% against 1.9%) and 0.9% at 24 weeks.

prolactin, movement, blood pressure: the other warnings in plain words

the warnings section runs from 5.1 to 5.17. beyond the boxed warning and the metabolic one:

• prolactin [5.15]: olanzapine raises prolactin and the rise "persists during chronic administration". in trials up to 12 weeks, 30% of adults shifted from a normal to a high level against 10.5% on placebo; in adolescents, 47% against 7%. across 8,136 adults the clinical signs the label could attribute were menstrual changes in 2% of women, sexual-function events in 2% of women and men, and breast events (discharge, enlargement, milk production, gynecomastia) in 0.7% of women and 0.2% of men; in 454 adolescents, breast events in 2% of girls and 2% of boys. the share with a high prolactin rose with dose: 31.2% at 10 mg, 42.7% at 20 mg, 61.1% at 40 mg over 8 weeks. high prolactin can suppress reproductive hormones, and the label says it may reversibly reduce fertility in women [8.3].
• tardive dyskinesia [5.6]: involuntary, potentially irreversible movements, most often after long treatment and high cumulative doses. the label says whether antipsychotics differ in their potential to cause it is unknown, 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.
• acute movement side effects [6.1, tables 16 and 17]: in the fixed-dose schizophrenia trial, any extrapyramidal event was reported by 15%, 25% and 32% at about 5, 10 and 15 mg against 16% on placebo; akathisia by 5%, 11% and 10% against 1%; parkinsonism by 8%, 14% and 20% against 10%. on the rating scales the picture was flatter: parkinsonism 12% to 14% against 15%, akathisia 16% to 27% against 23%. the label states that acute dystonia, including neck spasm and tongue protrusion, is a class effect that is more common in young men and at higher doses of the older drugs.
• low blood pressure, fainting and falls [5.7, 5.8]: orthostatic hypotension was recorded in 20% or more of 6,030 adults across 41 trials when vital signs were measured, and fainting in 0.6%. the label says starting at 5 mg lowers the risk, and that alcohol and diazepam make it worse [7.1]. with the short-acting injection, about a third of people given three 10 mg doses four hours apart had a significant fall in standing blood pressure, and the label says an injected benzodiazepine should not be given alongside it [5.7]. a fall-risk assessment is asked for in anyone whose other conditions or medicines add to drowsiness or unsteadiness.
• neuroleptic malignant syndrome [5.3] and dress [5.4]: nms is a rare, potentially fatal reaction with very high temperature, rigid muscles, confusion and unstable pulse or blood pressure; the label says stop at once and treat in hospital. dress (drug reaction with eosinophilia and systemic symptoms) is a rash with fever, swollen glands and inflammation of the liver, kidneys, lungs or heart, "sometimes fatal"; the label says stop if it is suspected.
• low white cells [5.9], seizures [5.11], swallowing [5.10], heat [5.13]: leukopenia and neutropenia have been reported with olanzapine, with frequent blood counts asked for in anyone with a low count or a history of drug-induced neutropenia. seizures occurred in 0.9% of 2,500 people in the premarketing trials, many with confounding factors. aspiration pneumonia and impaired cooling are class cautions.
• drowsiness [5.12]: somnolence in 26% against 15%, dose-related, and it led 0.4% to stop. the label says not to drive or operate machinery until the effect is known.
• anticholinergic effects [5.14]: constipation, dry mouth and fast heart rate in the trials, and since marketing "the risk for severe adverse reactions (including fatalities) was increased with concomitant use of anticholinergic medications". the label asks for caution in anyone with urinary retention, an enlarged prostate, constipation or a history of paralytic ileus, and with other anticholinergic drugs [7.1].
• liver enzymes [6.1]: alt rose to three times the upper limit of normal in 5% of adults against 1% on placebo, and to five times in 2% against 0.3%, usually transient and without jaundice or liver failure; in adolescents, 12% against 2%. it was the one reaction the label calls a drug-related reason for stopping in the schizophrenia trials (2% against 0%). cholestatic and mixed liver injury, hepatitis and jaundice are among the post-marketing reports [6.2].
• heart rhythm [6.1]: no significant difference from placebo in qt, qtc or pr intervals in pooled adult or adolescent studies, and a mean heart-rate rise of 2.4 beats per minute in adults and 6.3 in adolescents.
• interactions [7.1]: carbamazepine raises olanzapine clearance by about 50%, and rifampin and omeprazole may do the same; fluvoxamine raises olanzapine levels by half to double and the label says to consider a lower dose. tobacco smoke induces the same liver enzyme: the label names non-smoking as a reason for a 5 mg start [2.1], and two studies linked at the end measured the effect at about double the blood level per milligram in non-smokers, with the level climbing when a person stops smoking [haslemo-2006; zullino-2002].
• pregnancy and breastfeeding [8.1, 8.2]: babies exposed to any antipsychotic in the third trimester are at risk of movement and withdrawal symptoms after birth. the label says published studies of olanzapine in pregnancy "have not established a drug-associated risk of major birth defects, miscarriage, or adverse maternal or fetal outcomes", that a medicaid cohort of 9,258 women exposed to antipsychotics did not show an overall increase in major birth defects, and that untreated schizophrenia or bipolar disorder carries its own risks to the mother. olanzapine passes into breast milk, with reports of excess sedation, irritability, poor feeding and tremor in breastfed infants.

the long-acting injection: a second boxed warning and a restricted program

zyprexa relprevv, the olanzapine pamoate injection given every two to four weeks, carries a boxed warning the tablets do not: post-injection delirium/sedation syndrome. after an injection, some people develop signs of olanzapine overdose, from dizziness, confusion, slurred speech and unsteadiness to seizures and coma, apparently because the drug entered the bloodstream faster than intended; the label says the exact mechanism is not known [relprevv, boxed warning and 5.1]. the label’s numbers: events occurred in fewer than 0.1% of injections and in approximately 2% of patients who received injections for up to 46 months; most were hospitalised, several needed intubation, and all had largely recovered by 72 hours. the risk is the same at each injection, so it accumulates with the number of injections [relprevv 5.1].

because of this, the injection is available only through a restricted distribution program (a rems) in which the prescriber, the healthcare setting, the patient and the pharmacy must all be enrolled, it must be given in a certified setting with ready access to emergency services, and the patient must be observed by a healthcare professional for at least 3 hours after every injection, confirmed alert and oriented before leaving, accompanied home, and told not to drive or operate machinery for the rest of that day [relprevv 2.1, 5.2]. tolerability must be established with oral olanzapine first, and the injection has not been studied in anyone under 18 [relprevv 2.1]. its efficacy rests on one 8-week trial and one 24-week maintenance trial in adults [relprevv 1.1, 2.1]. the metabolic, prolactin and dementia warnings are the same as the tablets’; in its own 24-week fixed-dose trial, weight gain was 0.67 kg at the lowest dose and 1.70 kg at the highest [relprevv 5.7].

what the approval trials measured, and for how long

schizophrenia in adults rests on two 6-week trials of inpatients. the first (149 people) compared 1 mg and 10 mg with placebo: 10 mg beat placebo and 1 mg did not. the second (253 people) compared three dose ranges of about 5, 10 and 15 mg: the two higher groups beat placebo, with "no clear advantage for the high-dose group over the medium-dose group" [label 14.1]. a third trial took 326 outpatients who had been stable on olanzapine for at least 8 weeks and randomised them to continue or switch to placebo; it was stopped early because of an excess of relapses on placebo, so the observation ran up to 8 months rather than the planned 12 [label 14.1]. the label says doses above 10 mg were not shown to be more effective than 10 mg, and that olanzapine is not indicated above 20 mg a day [label 2.1]. one 6-week trial in 107 adolescents established the adolescent indication, at a mean dose of about 11 mg [label 14.1].

mania rests on one 3-week trial of 67 people and one 4-week trial of 115; a second 3-week trial run at the same time with the same design "was not shown to be superior to placebo" [label 14.2]. maintenance rests on a randomised-withdrawal trial of 361 people who had responded for about two weeks on average and were then kept on olanzapine or switched to placebo: half of the placebo group had left by day 23 and half of the olanzapine group by day 59 [label 14.2]. the add-on indication rests on two 6-week trials of 175 and 169 people already on lithium or valproate, and the adolescent mania indication on one 3-week trial of 161 [label 14.2]. the agitation indication for the short-acting injection rests on three 24-hour trials measured at two hours [label 14.3].

the arithmetic that matters: the efficacy trials were 3 to 6 weeks, the longest controlled measurement was about 8 months, and the label’s own long-term weight and lipid columns run to 36 months in people who stayed on it that long. the label asks the prescriber who continues it to "periodically reevaluate the long-term usefulness of the drug for the individual patient" [label 2.1, 2.2].

side effects, with the label’s numbers

in the 6-week schizophrenia trials (248 on olanzapine against 118 on placebo), the reactions at 5% or more and at least twice placebo were dizziness 11% against 4%, constipation 9% against 3%, weight gain 6% against 1%, postural hypotension 5% against 2%, akathisia 5% against 1%, and what the label’s old coding dictionary calls personality disorder, meaning non-aggressive objectionable behaviour, 8% against 4% [label 6.1, table 9]. in the 3- and 4-week mania trials (125 against 129): somnolence 35% against 13%, dry mouth 22% against 7%, dizziness 18% against 6%, weakness 15% against 6%, constipation 11% against 5%, dyspepsia 11% against 5%, increased appetite 6% against 3%, tremor 6% against 3% [label 6.1, table 10]. across all the short placebo-controlled trials (532 against 294): somnolence 29% against 13%, accidental injury 12% against 8%, dizziness 11% against 4%, abnormal gait 6% against 1%, fever 6% against 2% [label 6.1, table 11]. dropouts for side effects were 5% against 6% in schizophrenia and 2% against 2% in mania [label 6.1].

added to lithium or valproate (229 against 115), the numbers are the highest on the page: somnolence 52% against 27%, dry mouth 32% against 9%, weight gain 26% against 7%, increased appetite 24% against 8%, tremor 23% against 13%, dizziness 14% against 7%, speech disorder 7% against 1%, and 11% stopped for side effects against 2% [label 6.1, tables 13 and 14]. in adolescents (179 against 89): sedation 44% against 9%, weight increased 30% against 6%, increased appetite 24% against 6%, liver enzymes increased 8% against 1% [label 6.1, table 22].

in overdose, the label reports 67 cases in the premarketing trials, the largest 300 mg with drowsiness and slurred speech only, and since marketing one death possibly from as little as 450 mg taken alone against another person surviving about 2 grams [label 10.1]. olanzapine is not a controlled substance and "has not been systematically studied in humans for its potential for abuse, tolerance, or physical dependence" [label 9.3].

what the research says, in both directions

in favour. in a network meta-analysis of 402 randomised trials covering 53,463 adults with schizophrenia, all 32 antipsychotics reduced symptoms more than placebo and the drugs differed more in side effects than in effectiveness [huhn-2019]. in the lurasidone label’s trial that used olanzapine 15 mg as an active comparator, olanzapine’s placebo-subtracted change on the positive and negative syndrome scale was the larger (12.6 points against 9.7 and 7.5 for the two lurasidone doses), although that trial was not designed to compare the drugs [lurasidone-explainer]. olanzapine’s side-effect profile runs the other way from lurasidone’s: no movement-disorder penalty on the rating scales in the label’s fixed-dose trial [label 6.1, table 16] and no heart-rhythm signal [label 6.1]. the label’s pregnancy section says published studies have not established a drug-associated risk of major birth defects, in its own words [label 8.1].

against, or larger than it sounds. the same metabolic network meta-analysis ranked olanzapine, with clozapine, as having the worst metabolic profile of 18 antipsychotics: the largest rise in body-mass index against placebo (1.07 kg/m2) and the largest rise in ldl cholesterol (0.20 mmol/l), over a median of only 6 weeks [pillinger-2020; pillinger-2020, abstract]. the label’s long-term columns say what 6 weeks cannot: a mean 5.6 kg at a year, 64% gaining 7% or more, and 12.8% of people with normal blood sugar crossing the diabetes threshold [label 5.5]. in 6,671 uk patients on maintenance treatment for bipolar disorder, self-harm rates were roughly 40% to 50% higher on olanzapine, quetiapine or valproate than on lithium, with the usual caution that people prescribed different drugs differ [hayes-2016]. an english case-control study of 58,769 people with dementia found the highest cumulative exposure to strongly anticholinergic drugs, a group that includes olanzapine, associated with about 49% higher odds of dementia, 70% for the antipsychotic class; early dementia leading to prescriptions is the standard alternative explanation [coupland-2019]. tardive dyskinesia remains at about one in five on the newer antipsychotics as a class [carbon-2017].

smoking and cannabis. two studies in the library measure what the label only names. in 73 nursing-home patients on clozapine or olanzapine, non-smokers had about twice the blood level per milligram, and seven to twelve cigarettes a day was enough for the full effect; the authors suggested a 50% lower starting dose in non-smokers [haslemo-2006]. two case reports describe the interaction running backwards, with a patient developing marked movement side effects on olanzapine after cutting down tobacco [zullino-2002]. the mechanism is the smoke, not the nicotine, so patches do not hold the level down. separately, in 256 people after a first episode of psychosis in south london, continued daily use of high-potency cannabis roughly tripled the odds of relapse, and people who stopped did best; this is the substance acting on the illness, not on the drug [schoeler-2016].

how long to stay on an antipsychotic, in both directions. none of this is olanzapine-specific. a finnish register study of 8,719 people after a first schizophrenia admission found the lowest risk of rehospitalisation or death in those who stayed on an antipsychotic continuously, and death 174% to 214% higher among non-users and early stoppers; people who stop are not a random sample [tiihonen-2018]. a dutch randomised trial of 128 people in remission after a first episode found, seven years on, twice the recovery rate (40.4% against 17.6%) in those assigned to guided dose reduction, with more early relapses along the way; it was small and open-label [wunderink-2013]. the label asks for the need for continued treatment to be reassessed periodically [label 5.6].

questions people ask

what does the zyprexa (olanzapine) black box warning mean?

it has one part: in 17 placebo-controlled trials of antipsychotics in elderly people with dementia-related psychosis, about 4.5% died over roughly ten weeks against about 2.6% on placebo, 1.6 to 1.7 times the rate, mostly from heart or infectious causes. the olanzapine label adds its own figure, 3.5% against 1.5% in the olanzapine trials in that group, and more strokes. olanzapine is not approved for dementia-related psychosis. it does not carry the antidepressant suicidality boxed warning on its own because it is not approved as an antidepressant on its own; the olanzapine-fluoxetine combination does. the long-acting injection carries a second boxed warning of its own, on post-injection delirium and sedation.

how much weight do people gain on olanzapine?

on the label’s numbers, a mean 2.6 kg in a median 6 weeks against a 0.3 kg loss on placebo, with 22.2% gaining 7% or more of their body weight against 3%. in studies of at least 48 weeks the mean gain was 5.6 kg, and 64% had gained 7% or more, 32% 15% or more, and 12% a quarter of their weight or more. among people who completed three years, about one in nine had gained more than 20 kg. adolescents gained more: 4.6 kg in 3 weeks and 11.2 kg over about 200 days. a network meta-analysis of 18 antipsychotics ranked olanzapine and clozapine as having the worst metabolic profiles. the label asks for regular weight checks and fasting glucose and lipids at the start and periodically.

does olanzapine cause diabetes?

the label says the link between atypical antipsychotics and blood sugar "appears to fall on a continuum and olanzapine appears to have a greater association than some other atypical antipsychotics". its numbers: 2.2% of people with normal fasting glucose crossed the diabetes threshold within 12 weeks (against 3.4% on placebo) and 12.8% did in studies of at least 48 weeks; of people who started borderline, 26.0% crossed it long-term. in the catie trial, glucose rose a mean 15 mg/dl over nine months. the label reports cases of extreme hyperglycemia with ketoacidosis, coma or death across the class, including olanzapine, and asks for fasting glucose at the start and periodically.

what are the most common olanzapine side effects?

in the schizophrenia trials: dizziness 11% vs 4% on placebo, constipation 9% vs 3%, weight gain 6% vs 1%, akathisia 5% vs 1%. in the mania trials: somnolence 35% vs 13%, dry mouth 22% vs 7%, dizziness 18% vs 6%. added to lithium or valproate: somnolence 52% vs 27%, dry mouth 32% vs 9%, weight gain 26% vs 7%, increased appetite 24% vs 8%. in adolescents: sedation 44% vs 9%, weight increase 30% vs 6%. orthostatic hypotension was recorded in 20% or more when measured, and prolactin rose from normal to high in 30% of adults and 47% of adolescents. movement side effects on the rating scales were no higher than placebo, and there was no qt signal.

what is the zyprexa relprevv boxed warning, and why the 3-hour wait?

the long-acting olanzapine injection carries a boxed warning for post-injection delirium/sedation syndrome: after an injection some people develop signs of olanzapine overdose, from confusion and slurred speech to seizures and coma, apparently because the drug reached the blood faster than intended. the label puts it at fewer than 0.1% of injections and about 2% of patients over up to 46 months, with most hospitalised and all largely recovered within 72 hours; the risk is the same at each injection, so it accumulates. the fda restricts the product to a rems program: certified settings, enrolled prescribers, patients and pharmacies, at least 3 hours of observation after every dose, someone to take the patient home, and no driving that day.

does smoking change olanzapine levels?

yes. the hydrocarbons in tobacco smoke induce the liver enzyme cyp1a2 that clears olanzapine. the label names non-smoking status as a reason for a 5 mg starting dose and says carbamazepine, which induces the same enzyme, raises clearance by about half. a study of 73 patients found non-smokers had about twice the blood level per milligram, with seven to twelve cigarettes a day enough for the full effect, and case reports describe side effects appearing when a smoker cuts down. nicotine patches do not reproduce the effect, so a level climbs when smoking stops. that is a question for a prescriber before any change in smoking, not a reason to change the dose alone.

is olanzapine a controlled substance or addictive?

it is not a controlled substance. the label says olanzapine "has not been systematically studied in humans for its potential for abuse, tolerance, or physical dependence", that the trials showed no drug-seeking behaviour, and that this is not enough to predict misuse once marketed. a discontinuation reaction (sweating, nausea, vomiting) is among the reactions reported since marketing, which is a separate matter from addiction; the label’s maintenance trials also show the illness returning faster after a switch to placebo. a plan for reducing or stopping belongs with a prescriber.

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. zyprexa (olanzapine) tablets, zydis and intramuscular prescribing information, fda label on dailymed (set id b418946a-1ab4-4d89-a012-d94fc361a3c4, version 68, effective february 11, 2026; h2-pharma under cheplapharm licence): boxed warning; sections 1, 2.1, 2.2, 5.1 to 5.17 with tables 2 to 8, 6.1 tables 9 to 22, 6.2, 7.1, 8.1 to 8.5, 9.3, 10.1, 14.1 to 14.3
2. zyprexa relprevv (olanzapine pamoate) extended-release injectable suspension prescribing information, fda label on dailymed (set id 11544cba-64a9-49cb-8d74-da228053b252, version 68, effective may 19, 2026): boxed warning; sections 1.1, 2.1, 5.1, 5.2, 5.3, 5.7
3. drugs@fda, nda 020592 (zyprexa), approved september 30, 1996
4. ahrq medical expenditure panel survey, prescribed medicines public-use files 2014 (hc-168a) to 2024 (hc-254a): olanzapine has no reportable year (fewer than 60 survey respondents in every year, under ahrq’s reliability rule); method on the usage section of this page
5. the antipsychotic dementia-mortality warning as a library row: the 17-trial pooled analysis, the 2005 and 2008 fda actions, and the questions it raises (library row)
6. 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 (library row written up for aripiprazole) (library row)
7. pillinger t, et al. comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: a systematic review and network meta-analysis. lancet psychiatry 2020 (library row written up for aripiprazole; the olanzapine bmi and ldl figures are from the abstract, pubmed 31860457; uk medical research council and wellcome trust funding) (library row)
8. hayes jf, et al. self-harm, unintentional injury, and suicide in bipolar disorder during maintenance mood stabilizer treatment: a uk population-based electronic health records study. jama psychiatry 2016 (library row)
9. coupland cac, et al. anticholinergic drug exposure and the risk of dementia: a nested case-control study. jama internal medicine 2019 (library row)
10. carbon m, hsieh ch, kane jm, correll cu. tardive dyskinesia prevalence in the period of second-generation antipsychotic use: a meta-analysis. journal of clinical psychiatry 2017 (library row)
11. haslemo t, eikeseth ph, tanum l, molden e, refsum h. the effect of variable cigarette consumption on the interaction with clozapine and olanzapine. european journal of clinical pharmacology 2006 (73 patients) (library row)
12. zullino df, delessert d, eap cb, preisig m, baumann p. tobacco and cannabis smoking cessation can lead to intoxication with clozapine or olanzapine. international clinical psychopharmacology 2002 (two case reports) (library row)
13. schoeler t, et al. effects of continuation, frequency, and type of cannabis use on relapse in the first 2 years after onset of psychosis: an observational study. lancet psychiatry 2016 (256 patients) (library row)
14. tiihonen j, tanskanen a, taipale h. 20-year nationwide follow-up study on discontinuation of antipsychotic treatment in first-episode schizophrenia. american journal of psychiatry 2018 (register study; the row records the selection caution) (library row)
15. wunderink l, et al. recovery in remitted first-episode psychosis at 7 years of follow-up of an early dose reduction/discontinuation or maintenance treatment strategy: long-term follow-up of a 2-year randomized clinical trial. jama psychiatry 2013 (randomised, open-label, 128 people) (library row)
16. what the lurasidone (latuda) warnings mean, the companion explainer on this site, whose label trials used olanzapine as an active comparator
17. the olanzapine legal and safety record, a separate page on this site: court and regulator records, kept apart from the label and trial facts here

Weight gain and akathisia on the olanzapine label

The olanzapine FDA label names weight gain and akathisia in its Adverse Reactions section, in the list below.

Most-reported reactions

• Drug ineffective7,452
• Weight increased6,931
• Off label use6,256
• Diabetes mellitus5,413
• Drug interaction4,208
• Toxicity to various agents4,207
• Nausea4,139
• Somnolence4,133
• Fatigue3,686
• Anxiety3,517