lisdexamfetamine (Vyvanse)

the boxed warning: abuse, misuse and addiction

quoted in full so that it is the label’s and not ours: "lisdexamfetamine dimesylate has a high potential for abuse and misuse, which can lead to the development of a substance use disorder, including addiction. misuse and abuse of cns stimulants, including lisdexamfetamine dimesylate, can result in overdose and death, and this risk is increased with higher doses or unapproved methods of administration, such as snorting or injection. before prescribing lisdexamfetamine dimesylate, assess each patient’s risk for abuse, misuse, and addiction. educate patients and their families about these risks, proper storage of the drug, and proper disposal of any unused drug. throughout lisdexamfetamine dimesylate treatment, reassess each patient’s risk of abuse, misuse, and addiction and frequently monitor for signs and symptoms of abuse, misuse, and addiction" [label, boxed warning].

this wording is not specific to lisdexamfetamine. in may 2023 the fda required the boxed warning of every prescription stimulant, the amphetamines and the methylphenidates alike, to be updated so that it reads the same across the class [fda-2023; stimulant-warning]. the agency gave its reason plainly: the old labels did not say clearly enough "that most individuals who misuse prescription stimulants get their drugs from other family members or peers", and "sharing these medicines with others can lead to development of substance use disorder and addiction in those with whom these drugs are shared" [fda-2023]. so the warning is about the drug class and about what happens to pills that leave the person they were prescribed for, as much as about the person taking them. the label turns that into instructions: store it "in a safe place, preferably locked", and do not give it to anyone else [label 5.1]. the fda notice adds that stimulants "can be an important treatment option" for the conditions they are approved for; it is a warning about misuse, not a finding that the drug does not work [fda-2023]. the non-stimulant adhd medicines carry no such box; guanfacine, for one, has no boxed warning at all [guanfacine-explainer].

the label also reports how lisdexamfetamine compares with plain amphetamine for liking, in a study of 38 people with a history of drug abuse: 100 mg produced "significantly less" drug liking than 40 mg of immediate-release d-amphetamine, but 150 mg produced similar liking; injected, 50 mg produced positive responses greater than placebo but less than an equivalent injected dose of d-amphetamine [label 9.2]. read both ways: the prodrug design blunts the rush at ordinary doses, and it does not remove the abuse potential at higher ones, which is why the box applies to it in full.

on stopping, the label says physical dependence and tolerance can develop, and that after "abrupt discontinuation or dose reduction following prolonged use" the withdrawal signs include "dysphoric mood; depression; fatigue; vivid, unpleasant dreams; insomnia or hypersomnia; increased appetite; and psychomotor retardation or agitation" [label 9.3]. it gives no taper schedule.

the other warnings, in plain words

before the first dose the label asks for two checks: a careful history, family history of sudden death or ventricular arrhythmia, and physical exam for heart disease, and a family history and examination for tics or tourette’s syndrome [label 2.1]. it is contraindicated with an maoi antidepressant or within 14 days of stopping one, including linezolid or intravenous methylene blue, because of the risk of a hypertensive crisis [label 4]. the warnings themselves, with the label’s words:

• serious heart disease: "sudden death has been reported in patients with structural cardiac abnormalities or other serious cardiac disease who were treated with cns stimulants at the recommended adhd dosage", so the label says to avoid it in people with known structural heart abnormalities, cardiomyopathy, serious arrhythmia or coronary artery disease [label 5.2].
• blood pressure and heart rate: stimulants raise blood pressure by a mean of about 2 to 4 mmhg and heart rate by about 3 to 6 beats a minute, "some patients may have larger increases", and every patient should be monitored for a fast heart rate and high blood pressure [label 5.3].
• psychosis and mania: stimulants can worsen a pre-existing psychotic disorder and "may induce a manic or mixed episode", so the label asks for screening for bipolar risk before starting (depressive symptoms, or a family history of suicide, bipolar disorder or depression). in people with no such history, "psychotic or manic symptoms (e.g., hallucinations, delusional thinking, or mania)" can still appear at the recommended dose; across pooled short trials of stimulants this happened in about 0.1% of people on a stimulant against 0% on placebo, a class figure, not one for this drug alone [label 5.4].
• growth in children: stimulants "have been associated with weight loss and slowing of growth rate". in the 4-week trial in children aged 6 to 12, weight fell 0.9, 1.9 and 2.5 pounds at 30, 50 and 70 mg against a 1 pound gain on placebo; children followed for 12 months on daily treatment dropped from the 60.9th to the 47.2nd weight percentile for their age and sex. the label asks for height and weight to be watched and says a child not growing as expected "may need to have their treatment interrupted" [label 5.5, 6.1]. the height figure the label prints, about 2 cm less growth over 3 years, comes from studies of methylphenidate, another stimulant, and the label says so [label 6.1].
• fingers and toes: stimulants are associated with peripheral vasculopathy, including raynaud’s phenomenon. symptoms are "usually intermittent and mild", but have led to "digital ulceration and/or soft tissue breakdown", and usually improved after a dose reduction or stopping [label 5.6].
• serotonin syndrome: possible when amphetamines are combined with drugs that raise serotonin, which the label lists as maois, ssris, snris, triptans, tricyclics, fentanyl, lithium, tramadol, tryptophan, buspirone and st john’s wort, and more likely with drugs that block the liver enzyme cyp2d6, which raises dextroamphetamine levels [label 5.7].
• tics: stimulants "have been associated with the onset or exacerbation of motor and verbal tics", and worsening of tourette’s syndrome has been reported [label 5.8].

the shortage, from the regulators’ own pages

the fda drug shortages database lists lisdexamfetamine dimesylate capsules, and the chewable tablets, as "currently in shortage" when read on 2026-10-13; the capsule entry was first posted on july 14, 2023 [shortage]. the same entry lists each manufacturer’s strengths separately and they differ: some list every strength as available, others list "limited availability" or one strength unavailable, and the reason most often given is "shortage of an active ingredient" [shortage]. in practice that means the shortage is patchy: whether a pharmacy can fill a given strength depends on whose product it stocks that week.

the fda and the dea explained the origin of the wider stimulant shortage in a joint letter on august 1, 2023. it "began last fall due to a manufacturing delay experienced by one drug maker", and its effects continued "in combination with record-high prescription rates": overall stimulant dispensing in the us rose 45.5 percent from 2012 to 2021 [fda-dea-letter]. the letter also explains why supply cannot simply be turned up: stimulants are controlled substances, so the dea sets limits, called quotas, on how much can be made [fda-dea-letter]. the letter is about all prescription stimulants; it does not give a figure for lisdexamfetamine alone.

what the trials measured, in whom, and for how long

adhd. the main short trials lasted 4 weeks and used the clinician-rated adhd rating scale (adhd-rs-iv, 0 to 54 points) [label 14.1, table 6]. in 290 children aged 6 to 12 (study 1), starting near 43 points, scores fell 15.6, 17.2 and 20.5 points more than placebo at 30, 50 and 70 mg; placebo itself fell 6.2. in 314 adolescents aged 13 to 17 (study 4), starting near 38, the extra drop was 5.5, 8.3 and 7.9 points, and placebo fell 12.8, so the gap was much smaller in teenagers. in 420 adults aged 18 to 55 (study 7), starting near 40, it was 8.0, 9.2 and 10.4 points, with placebo falling 8.2. an 8-week dose-optimisation trial in 336 children and adolescents (study 5), which also had an active comparator the label does not name, found an 18.6-point extra drop [label table 6]. two classroom studies in children and one simulated workplace study in adults found effects lasting from about 2 hours to 12 or 14 hours after the dose, not significant at 1 hour [label 14.1].

staying on it. in children and adolescents who had taken it open-label for at least 26 weeks and responded, 15.8% of those kept on it met the definition of treatment failure over 6 weeks, against 67.5% of those switched to placebo (study 6). in adults who had taken it for at least 6 months, the figures were 9% against 75% (study 9) [label 14.1]. these are withdrawal designs: they measure what happens when a working drug is taken away for 6 weeks, not how well it works over years.

binge eating disorder. a phase 2 trial found 30 mg no different from placebo, and 50 and 70 mg better (study 10). in the two 12-week trials that support the approval (studies 11 and 12, 374 and 350 adults aged 18 to 55 with at least 3 binge days a week), people started near 4.7 binge days a week; on 50 or 70 mg they ended about 3.9 days lower, against 2.5 and 2.3 days lower on placebo, a gap of 1.35 and 1.66 binge days a week [label 14.2, table 7]. a 26-week withdrawal trial in 267 adults who had responded found a longer time to relapse on the drug than on placebo; the label shows that only as a curve and prints no percentages [label 14.2]. the label says to stop it "if binge eating does not improve" [label 2.4]. safety and effectiveness for binge eating have not been established under 18, and people with heart risk factors other than obesity and smoking were excluded from these trials [label 6.1, 8.4].

side effects, with the label’s numbers

appetite and sleep lead every table. in children aged 6 to 12: decreased appetite 39% against 4% on placebo, insomnia 22% against 3%, upper abdominal pain 12% against 6%, irritability 10% against 0%, vomiting 9% against 4%, weight loss 9% against 1% [label table 1]. in adolescents: decreased appetite 34% against 3%, insomnia 13% against 4%, weight loss 9% against 0% [label table 2]. in adults: decreased appetite 27% against 2%, insomnia 27% against 8%, dry mouth 26% against 3%, diarrhea and nausea 7% each against 0%, anxiety 6% against 0%, raised blood pressure 3% against 0%; erectile dysfunction was 2.6% of men against 0% [label table 3, 6.1]. in binge eating disorder over 12 weeks: dry mouth 36% against 7%, insomnia 20% against 8%, decreased appetite 8% against 2%, raised heart rate 7% against 1%, feeling jittery and constipation 6% each against 1%, anxiety 5% against 1% [label table 4].

stopping because of side effects: 8% of children aged 6 to 12 against 0% on placebo, 3% of adolescents against 1%, 6% of adults against 2%, and 5.1% of adults with binge eating disorder against 2.4% [label 6.1]. adults lost 2.8 to 4.3 pounds over 4 weeks against a half-pound gain on placebo [label 6.1]. after approval, reports added among others cardiomyopathy, seizures, stevens-johnson syndrome, angioedema, eosinophilic hepatitis, tics, depression, aggression, frequent or prolonged erections, rhabdomyolysis and intestinal ischemia; the label says such reports cannot give a rate or prove cause [label 6.2].

practical points from the label. it is taken in the morning, because afternoon doses can cause insomnia; the capsule can be swallowed whole or opened and mixed into yogurt, water or orange juice, and a single dose should not be divided [label 2.2]. the starting dose is 30 mg and the maximum 70 mg a day, lower in severe kidney disease [label 2.3, 2.5]. substances that acidify the urine, such as vitamin c, lower amphetamine levels and alkalinising ones such as sodium bicarbonate raise them [label 2.6]. the comparison page sets this label beside the mixed amphetamine salts label [compare]. in a trial in children aged 4 to 5, exposure was about 44% higher than in children aged 6 to 11 at the same dose, with more weight loss, insomnia, infections and irritability, which is the basis of the under-6 limit [label 8.4].

what the research says, in both directions

a limit to state first: this site’s library holds one row tagged lisdexamfetamine: Amphetamine-type ADHD medicines are reported for psychosis more often than methylphenidate. the one it held when this section was checked is a pharmacovigilance study in which lisdexamfetamine was pooled with dexamfetamine [hamard-2024]. the other research below is about amphetamines or stimulants as a class, and each is labelled that way. none of it is a trial of lisdexamfetamine alone except the label’s own.

in favour. every adhd trial on the label beat placebo on its main measure, and both binge eating trials did [label 14]. a cochrane review of 19 trials in 2,521 adults found that amphetamines reduced clinician-rated adhd symptoms, and in its subgroup for this drug, 7 trials and 896 adults, lisdexamfetamine reduced them with a standardised effect of 1.06 (95% ci 0.85 to 1.26), larger than for mixed amphetamine salts; the authors rated it low-quality evidence [castells-2018]. the largest network meta-analysis of adhd medicines, 133 double-blind trials, pooled lisdexamfetamine with the other amphetamines (class evidence) and found amphetamines the most effective on clinician ratings in children and adolescents (standardised effect 1.02) and in adults (0.79), and better than methylphenidate, atomoxetine and modafinil head to head on that measure; it recommended amphetamines as the first choice in adults [cortese-2018]. two large observational studies of adhd medicines as a class point the same way on harms that worry people most: in 443,198 adults aged 25 to 64, current use was not associated with more heart attacks, sudden cardiac deaths or strokes [habel-2011], and in 148,578 swedes newly diagnosed with adhd, starting medication was associated with 21% lower all-cause mortality over 2 years, mainly fewer deaths from accidents and overdoses [li-2024].

against, or smaller than it sounds. the same network meta-analysis found amphetamines the least well tolerated: children and adolescents were 2.3 times, and adults 3.3 times, as likely as on placebo to drop out because of side effects, and on teacher ratings only methylphenidate and modafinil beat placebo among the comparisons available; it recommended methylphenidate as the first choice for children and adolescents, and found too little data to say anything at 26 or 52 weeks [cortese-2018]. the cochrane review found that amphetamines nearly tripled dropouts for side effects (risk ratio 2.69) without keeping more people in treatment, that 16 of its 19 trials were industry funded, that none had a low risk of bias in every domain (partly because the drugs’ effects are easy to feel, which breaks the blinding), and that trials averaged 5.3 weeks [castells-2018]. on psychosis, in 221,846 insured adolescents and young adults starting a stimulant for adhd, new psychosis occurred in 0.21% of those starting an amphetamine against 0.10% of those starting methylphenidate (class evidence, observational) [moran-2019]; in the who’s adverse-reaction database, amphetamine-type drugs, lisdexamfetamine pooled with dexamfetamine, were reported for psychosis 1.61 times as often as methylphenidate in people aged 13 to 25, a reporting ratio that gives no rate and cannot separate the two drugs [hamard-2024]. on the heart over the long run, a swedish national study of 278,027 people with adhd found each added year of adhd medication associated with 4% higher cardiovascular risk, about 23% to 27% higher after 3 or more years, mostly high blood pressure and arterial disease (class evidence, observational) [zhang-2024]. and the label’s own short trials lasted 4 weeks for adhd and 12 for binge eating [label 14].

questions people ask

does vyvanse (lisdexamfetamine) have a black box warning?

yes. its boxed warning says lisdexamfetamine has a high potential for abuse and misuse that can lead to a substance use disorder, including addiction, and that misuse can result in overdose and death, especially at higher doses or when snorted or injected. it asks prescribers to assess each patient’s risk before prescribing, educate families about storage and disposal, and keep monitoring. the wording comes from a 2023 fda update that made the boxed warning the same for every prescription stimulant.

why did the fda update the stimulant boxed warning in 2023?

the fda said the old labels did not adequately warn that most people who misuse prescription stimulants get them from family members or peers, and that sharing them can cause a substance use disorder in the person they are shared with. it required the boxed warning to describe misuse, abuse, addiction and overdose consistently across all prescription stimulants, amphetamines and methylphenidates alike, and said stimulants remain an important treatment option.

is there still a vyvanse shortage?

the fda drug shortages database listed lisdexamfetamine capsules and chewable tablets as currently in shortage when read on 2026-10-13; the capsule entry was first posted on july 14, 2023. availability differs by manufacturer and strength: some list every strength as available, others limited availability, most often because of a shortage of an active ingredient. a 2023 fda and dea letter traced the wider stimulant shortage to a manufacturing delay and record prescribing, with production capped by dea quotas.

is vyvanse approved for binge eating disorder?

yes, for moderate to severe binge eating disorder in adults, approved in january 2015. in the two 12-week trials, adults starting near 4.7 binge days a week had 1.35 and 1.66 fewer binge days a week than on placebo at 50 or 70 mg; 30 mg was no better than placebo in an earlier trial. the label says it is not for weight loss and its safety and effectiveness for obesity have not been established.

how well does lisdexamfetamine work for adhd?

in the label’s 4-week trials, adhd rating scale scores fell 15.6 to 20.5 points more than placebo in children aged 6 to 12 (from about 43), 5.5 to 8.3 points more in adolescents (from about 38) and 8.0 to 10.4 points more in adults (from about 40). a cochrane review found a large effect in adults on low-quality evidence; a 2018 network meta-analysis found amphetamines as a class the most effective on clinician ratings but the least well tolerated.

what are the most common vyvanse side effects?

decreased appetite and insomnia in every age group (39% and 22% of children aged 6 to 12, against 4% and 3% on placebo), then dry mouth, upper abdominal pain, irritability, weight loss, nausea, diarrhea and anxiety. in adults with binge eating disorder, dry mouth was the commonest at 36% against 7%. between 3% and 8% of people in the adhd trials stopped because of side effects.

is lisdexamfetamine the same as adderall?

not the same product, but the same active drug family. lisdexamfetamine is a prodrug that red blood cells convert into dextroamphetamine; adderall is a mix of amphetamine salts. both are schedule ii stimulants and carry the same 2023 class boxed warning on abuse, misuse and addiction. the comparison page on this site sets the two labels side by side.

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. lisdexamfetamine dimesylate capsules prescribing information, fda label on dailymed (set id c11cb3b6-3dff-477b-a249-b4c3118edc5f, version 25, effective august 11, 2026; a generic of vyvanse carrying its tables): boxed warning, sections 1, 2.1 to 2.6, 4, 5.1 to 5.8, 6.1, 6.2, 8.4, 9.1 to 9.3, 12.1, 12.3, 14.1, 14.2 and tables 1 to 4, 6 and 7
2. drugs@fda, nda 021977 (vyvanse, lisdexamfetamine dimesylate), approved february 23, 2007
3. fda supplement approval letter, nda 21977/s-037, january 30, 2015: safety and effectiveness of vyvanse for moderate to severe binge eating disorder
4. us food and drug administration. fda updating warnings to improve safe use of prescription stimulants used to treat adhd and other conditions (drug safety communication, may 11, 2023); read on the verification date
5. fda drug safety communication, 2023: fda updating warnings to improve safe use of prescription stimulants used to treat adhd and other conditions (the library row) (library row)
6. us food and drug administration, drug shortages database: lisdexamfetamine dimesylate capsule (status currently in shortage, date first posted 07/14/2023) and lisdexamfetamine dimesylate tablet, chewable; read on 2026-10-13
7. us food and drug administration and drug enforcement administration. joint letter to the public on the prescription stimulant shortage, august 1, 2023 (pdf)
8. ahrq medical expenditure panel survey, prescribed medicines public-use files 2014 (hc-168a) to 2024 (hc-254a): people with at least one lisdexamfetamine purchase, weighted; 2024 from 86 respondents; computed by veisund, method on the usage section of this page
9. centers for medicare and medicaid services, medicare part d prescribers by geography and drug, national rows, 2024: 87,449 beneficiaries across the brand and generic rows (an upper bound on distinct people), against 54,375,198 people with part d drug coverage in cms monthly enrollment for 2024
10. castells x, et al. amphetamines for attention deficit hyperactivity disorder (adhd) in adults. cochrane database of systematic reviews 2018 (amphetamine class review; the library row is written up for amphetamine; the lisdexamfetamine subgroup and the funding and bias findings were read from the pubmed abstract, pmid 30091808, on 2026-10-13) (library row)
11. cortese s, et al. comparative efficacy and tolerability of medications for attention-deficit hyperactivity disorder in children, adolescents, and adults: a systematic review and network meta-analysis. lancet psychiatry 2018 (amphetamines including lisdexamfetamine pooled as one class; the library row is written up for amphetamine; read with the pubmed abstract, pmid 30097390, on 2026-10-13) (library row)
12. moran lv, et al. psychosis with methylphenidate or amphetamine in patients with adhd. new england journal of medicine 2019 (class evidence, observational) (library row)
13. hamard j, rousseau v, durrieu g, et al. psychosis with use of amphetamine drugs, methylphenidate and atomoxetine in adolescent and adults. bmj mental health 2024 (who vigibase pharmacovigilance; lisdexamfetamine pooled with dexamfetamine) (library row)
14. habel la, et al. adhd medications and risk of serious cardiovascular events in young and middle-aged adults. jama 2011 (class evidence, observational) (library row)
15. li l, et al. adhd pharmacotherapy and mortality in individuals with adhd. jama 2024 (class evidence, observational) (library row)
16. zhang l, et al. attention-deficit/hyperactivity disorder medications and long-term risk of cardiovascular diseases. jama psychiatry 2024 (class evidence, observational; the library row is written up for methylphenidate) (library row)
17. what the guanfacine (intuniv) warnings mean, the companion explainer on this site, for the non-stimulant adhd medicine with no boxed warning
18. amphetamine vs lisdexamfetamine, the side-by-side comparison page on this site

Most-reported reactions

• Drug ineffective4,950
• Fatigue2,150
• Off label use2,115
• Anxiety1,944
• Headache1,857
• Nausea1,642
• Product availability issue1,587
• Insomnia1,570
• Depression1,434
• Feeling abnormal1,401