Antipsychotic Adverse Events · Chapter 2 · Movement Disorders (EPS)
Akathisia
The Restlessness Everyone Blames on the Illness
A patient gets more restless after you start or raise an antipsychotic, and the reflex is to read it as worsening illness — and reach for more drug. That one misread is how akathisia turns into a spiral. It is common, it is genuinely dangerous, and it is one of the most treatable adverse events in psychiatry — once you name it.
Bottom Line Up Front
- Akathisia is subjective inner restlessness plus objective movement — and the inner distress is usually worse than anything you can see across the room. Onset is typically within four weeks of a start or dose increase.
- The classic trap is misreading it as agitation and raising the antipsychotic — which feeds the akathisia. Timing is the tell: did the restlessness start after a med change?
- Highest risk with high-potency and partial-agonist agents — haloperidol (up to ~57%), risperidone (~20%), cariprazine (~17%), lurasidone (~13%), aripiprazole, brexpiprazole. Lower with quetiapine and olanzapine (~3–4%), and clozapine. It isn't only antipsychotics: SSRIs and antiemetics (metoclopramide, prochlorperazine) do it too.
- Treat the drug first: reduce the dose, stop antipsychotic polypharmacy, or switch to a lower-liability agent — dose is the strongest predictor. Then reach for the top of the network meta-analysis: mirtazapine 15 mg, biperiden, or vitamin B6 — that is the efficacy order, with B6 singled out because its tolerability profile is the best of the three rather than because it outranks biperiden. Propranolol is effective but ranks behind them. Benzodiazepine evidence is conflicting.
- Screen for dysphoria and suicidal thoughts. A causal link isn't firmly proven, but it's plausible and concerning enough that routine screening is the right call — treat akathisia as a safety problem, not a nuisance.
What Akathisia Actually Is
Akathisia is a drug-induced movement disorder with two halves that have to be held together: a subjective sense of inner restlessness the patient can name — "I can't stay in my skin," "I need to move" — and an objective motor picture: pacing, rocking, marching in place, shifting weight foot to foot, crossing and uncrossing the legs, an inability to sit through a visit. The subjective half is the hallmark. A patient can hold themselves relatively still and still be in torment, so a calm-looking exam does not rule it out.
Subjective — what they feel
- "I can't stay in my skin"
- "I need to move"
- Inner restlessness the patient can name
- Usually worse than anything visible across the room
Objective — the motor picture
- Pacing, rocking, marching in place
- Shifting weight foot to foot
- Crossing and uncrossing the legs
- Inability to sit through a visit
Timing sorts it into three forms. Acute akathisia appears typically within four weeks of starting the drug, raising the dose, or switching to a higher-potency agent — this is the common one. Tardive akathisia emerges later, after months, and behaves like the other tardive syndromes. Withdrawal akathisia shows up when a drug is reduced or stopped. In every case, the single most useful piece of history is what changed in the regimen and when.
Acute and tardive akathisia look identical but respond in opposite ways to the same move. Acute akathisia improves when you reduce or stop the antipsychotic; tardive akathisia worsens when you do. So for late-onset restlessness (months into stable treatment), the reflexive dose reduction that helps acute akathisia can make things worse — approach it as a tardive syndrome, not a recurrent acute reaction.
| Acute akathisia | Tardive akathisia | |
|---|---|---|
| Onset | Within about four weeks of a start or increase | Months into stable treatment |
| Appearance | Identical — you cannot tell them apart by looking | |
| Reduce or stop the antipsychotic | Improves | Worsens |
| So | Run the ladder in this chapter | Approach it as a tardive syndrome, not a recurrent acute reaction |
The Great Masquerader
Here is the reason this chapter exists. The restlessness of akathisia looks, from the outside, like the agitation of a worsening psychiatric illness — so the reflex is to attribute it to the disease and increase the antipsychotic. That escalates the exact agent driving the akathisia, the patient gets worse, the dose climbs again, and a treatable side effect becomes a spiral that can end in a distressed, non-adherent, or dangerous patient.
The way out is a single question asked early: did this start after a medication change? Akathisia is driven from the body up and eases, at least briefly, with movement; the patient localizes the problem to a compulsion to move rather than to fear or paranoia. Agitation from the illness is driven from the inside out. When the timeline points at a recent start or dose increase, suspect the drug before you suspect the disease.
Rule Out
Four look-alikes account for almost all the confusion. Separating them is mostly a matter of timeline and where the patient locates the sensation.
| Look-alike | How to tell it apart |
|---|---|
| Psychotic agitation / anxiety | The critical fork. Akathisia eases with movement and the patient localizes it to a need to move; agitation is driven by internal distress or paranoia. The tell is timing after a med change — miss it and you escalate the cause. |
| Restless legs syndrome | Worse at rest and in the evening/night, relieved by walking, with a dysesthesia the patient localizes to the legs. Tied to low ferritin — check iron. |
| Drug-induced parkinsonism | Tremor, rigidity, bradykinesia — a slowed, stiff picture rather than a driven one. Can coexist with akathisia, which muddies the exam. |
| Tardive dyskinesia | Choreoathetoid, often orobuccal movements with later onset. Tardive akathisia is the late-onset akathisia variant and can travel with TD. |
Who's at Highest Risk
Liability tracks with potency and with how hard a drug drives the dopamine system. Partial agonists earn a spot near the top precisely because of how they act at the D2 receptor. Rapid titration and higher doses raise risk across the board — in fact, after adjusting for confounders, antipsychotic dose is often the single strongest predictor, and some agents (lurasidone especially) keep climbing with dose.
| Tier | Agents (approx. incidence) |
|---|---|
| Higher liability | High-potency FGAs — haloperidol (up to ~57% in first-episode data); risperidone (~20%); the partial agonists cariprazine (~17%), aripiprazole (~9–18%); lurasidone (~13%). |
| Intermediate | Brexpiprazole (~10%) — lower than aripiprazole/cariprazine (less intrinsic D2 activity, higher 5-HT1A/2A affinity), but still above the low-liability agents. |
| Lower liability | Asenapine (~7%), iloperidone (~4%), olanzapine (~3.6%), quetiapine (~3.5%), clozapine. |
| Don't forget — not just antipsychotics | SSRIs and SNRIs (phenomenologically identical, and the same "misread as anxiety → escalate the dose" trap — treat by reducing/switching the antidepressant first), TCAs, dopamine agonists, calcium-channel blockers, and the antiemetics metoclopramide and prochlorperazine. On a medical floor the diagnosis is easily missed — in one cancer cohort, antiemetic-induced akathisia took 4–6 months on average to be recognized. |
Score It: the Barnes Scale
Anchor the diagnosis with a number so you can track response instead of relying on impression. The Barnes Akathisia Rating Scale (BARS) takes about two minutes and rates four things: the objective movement you observe, the patient's subjective awareness of restlessness, the subjective distress it causes, and a global clinical judgment. Re-score after each intervention — a falling BARS is how you know the ladder below is working.
Objective movement
What you observe across the room.
Subjective awareness
The patient's own sense of restlessness.
Subjective distress
How much that restlessness costs them.
Global judgment
Your overall clinical rating.
Two minutes to score, and it converts an impression into a number you can re-check. Re-score after each intervention — a falling BARS is your evidence the plan is working.
Managing It
The evidence points to a clear sequence: fix the offending drug first, then reach for the agents at the top of the current network meta-analysis. By efficacy that order is mirtazapine (SMD −1.20), then biperiden (SMD −1.01), then vitamin B6 (SMD −0.92) — all three ahead of propranolol, which the older teaching led with. Vitamin B6 gets singled out below for its tolerability and acceptability, not because it beats biperiden on efficacy.
| Option | How to use it | Notes |
|---|---|---|
| Reduce dose / stop polypharmacy / switch | Lower the offending antipsychotic, stop any antipsychotic polypharmacy, or switch to a lower-liability agent — before adding anything. | The most direct fix and the lever most often skipped. Dose is the strongest single predictor. |
| Mirtazapine (5-HT2A antagonist) | 15 mg at night. | Ranked first for efficacy in the 2024 network meta-analysis (SMD −1.20; P-score 0.826); better tolerated than propranolol in head-to-head data. |
| Biperiden (anticholinergic) | Anticholinergic option with the best evidence in its class. | Ranked second for efficacy in the same analysis (SMD −1.01; P-score 0.703) — ahead of vitamin B6, not behind it. Specifically effective (unlike the class as a whole for isolated akathisia); most useful when parkinsonism coexists. Cognitive/constipation effects limit use. |
| Vitamin B6 | 600–1200 mg/day. | Third for efficacy (SMD −0.92; P-score 0.641) but with an excellent tolerability/acceptability profile — which is why it is often highlighted as the best risk-benefit option, rather than because it outranks biperiden. |
| Propranolol (beta-blocker) | Start 20 mg BID; titrate across ~30–120 mg/day. | Effective but ranked behind all three above. Watch tolerability — ~17% discontinued for hypotension/bradycardia in a head-to-head trial. Avoid in asthma/COPD, bradycardia, decompensated HF. |
| Benzodiazepine | Short-term only, for distress. | Conflicting evidence — clonazepam failed to separate from placebo in the 2024 network meta-analysis, while a 2026 network meta-analysis found a possible benefit (SMD −1.62) at very low confidence. Treat as unsettled, and as a bridge rather than a treatment. |
| Other adjuncts | Trazodone, mianserin, cyproheptadine. | Modest supporting evidence; reserve for when first-line options fail or aren't tolerated. |
A walk-through of the study that sets the ranking above. Gerolymos C, Barazer R, Yon DK, Loundou A, Boyer L, Fond G. Drug efficacy in the treatment of antipsychotic-induced akathisia: a systematic review and network meta-analysis. JAMA Netw Open. 2024;7(3):e241527. Read the original article →
The Action Ladder
- Confirm it's akathisia, not the illnessAsk what changed in the regimen and when. Localize the sensation. Score a BARS. If the timeline points at a recent start or increase, treat the drug as the prime suspect.
- Work the drug before the add-onsIn order: reduce the dose, stop antipsychotic polypharmacy, then switch to a lower-liability agent. The guideline sequence is deliberate — exhaust these before reaching for an adjuvant.
- Add mirtazapine, biperiden, or vitamin B6 firstThese are the top three on current network meta-analysis, in that efficacy order — mirtazapine 15 mg at night, biperiden, or vitamin B6 600–1200 mg/day (B6 favoured on tolerability, biperiden when parkinsonism coexists). Propranolol is an effective alternative ranked behind them; benzodiazepine evidence is conflicting. Re-score in a few days.
- Watch for the tardive formLate-onset restlessness on stable treatment may be tardive akathisia — where reducing the drug can worsen it. Don't apply the acute playbook blindly.
- Screen for the dangerAsk directly about dysphoria and suicidal thoughts. The causal link isn't proven, but screening is the right call — and catching it changes the urgency of everything above.
- Ask the timing question before you raise the dose. "Did this start after a med change?" is the single most useful thing you can say.
- A calm exam doesn't rule it out. The subjective torment can outrun the visible movement.
- Partial agonists sit near the top of the risk list — aripiprazole, cariprazine, lurasidone — not the bottom.
- Look off the psych service too. Metoclopramide and prochlorperazine cause akathisia that gets missed on medical floors.
- Mirtazapine, biperiden, and B6 all lead now — current network meta-analysis puts them ahead of propranolol, in that efficacy order. Update the old reflex to reach for propranolol first, and don't skip biperiden: it ranks second, not last.
- Late-onset restlessness may be tardive akathisia — where cutting the dose backfires. Don't run the acute playbook on it.
- Score it, then re-score it. A falling Barnes number is your proof the plan is working.
- New or worsening restlessness after an antipsychotic start or dose increase being treated as "agitation" with more antipsychotic.
- Acute dysphoria, hopelessness, or suicidal thoughts emerging after a med change — screen for akathisia as the driver.
- New aggression or a patient abruptly wanting to stop treatment soon after starting an antipsychotic.
- Anticholinergics escalated repeatedly without response — reconsider the diagnosis and move to a 5-HT2A antagonist or dose strategy.
- Restlessness on metoclopramide or prochlorperazine mistaken for anxiety and left unaddressed.
"That constant need to move, where you can't get comfortable — that's a side effect of the medicine, not your illness getting worse, and it's very treatable. We're going to adjust things: I may lower or change the medication, and I can add something that settles the restlessness. This isn't something you have to just push through. And tell me right away if the restlessness comes with dark or hopeless thoughts — that can happen with this, and it's a reason to call me, not to wait it out."
This is one event. The course is all of them.
You've just seen how a single adverse event gets worked up and managed — recognized, separated from its look-alikes, and treated in a clear ladder. The full Antipsychotics course does the same for the rest of the movement disorders, plus the metabolic, cardiac, prolactin, and clozapine-specific events behind daily prescribing. If this chapter earned its keep, the rest is waiting.
Unlock the full course →References
- Barnes TRE. A rating scale for drug-induced akathisia. Br J Psychiatry. 1989;154:672–676 — the Barnes Akathisia Rating Scale.
- Gerolymos C, Barazer R, Yon DK, Loundou A, Boyer L, Fond G. Drug efficacy in the treatment of antipsychotic-induced akathisia: a systematic review and network meta-analysis. JAMA Netw Open. 2024;7(3):e241527 — efficacy ranking mirtazapine (SMD −1.20; P-score 0.826) → biperiden (SMD −1.01; P-score 0.703) → vitamin B6 (SMD −0.92; P-score 0.641), all ahead of propranolol; clonazepam did not separate from placebo. Full text. Video summary embedded above.
- Furukawa et al. Network meta-analysis of treatments for antipsychotic-induced akathisia. Schizophr Bull. 2026 — found benzodiazepines might be beneficial (SMD −1.62) at very low confidence, conflicting with the 2024 analysis above; the chapter presents the benzodiazepine evidence as unsettled on this basis.
- Pringsheim T, Gardner D, Addington D, et al. The assessment and treatment of antipsychotic-induced akathisia. Can J Psychiatry. 2018;63(11):719–729 — structured algorithm: dose reduction → stop polypharmacy → switch → adjuvants.
- Laoutidis ZG, Luckhaus C. 5-HT2A receptor antagonists for the treatment of neuroleptic-induced akathisia: systematic review and meta-analysis. Int J Neuropsychopharmacol. 2014;17(5):823–832.
- Poyurovsky M. Acute antipsychotic-induced akathisia revisited. Br J Psychiatry. 2010;196(2):89–91 — head-to-head mirtazapine vs. propranolol tolerability.
- Lohr JB, Eidt CA, Alfaraj AA, Soliman MA. The clinical challenges of akathisia. CNS Spectr. 2015;20(S1):1–14. Note: a 2021 systematic review found the akathisia–suicidality link based on small, low-quality studies and inconclusive — screening is recommended, but causation is not established.
Responses