Hypnosis for Functional Neurological Disorder: What the Evidence Shows

Hypnosis and functional neurological symptoms have been entangled since the 1800s, when the conditions were called hysteria and the treatments came with an audience [3]. That history is exactly why the old me filed the whole area under “theater.” The modern trial evidence turns out to be two small randomized trials, from the same Dutch team, which on a first read contradict each other. On a second read, they’re the clearest lesson in why comparison groups matter that I’ve found anywhere.

What the condition is

Functional neurological disorder (FND), formerly called conversion disorder, causes real neurological symptoms (weakness, tremor, abnormal movements, seizure-like episodes) from a problem with how the nervous system functions rather than from structural damage. It’s diagnosed by a neurologist, using positive signs on examination, not just by ruling everything else out. It can be very disabling. Both hypnosis trials studied the motor type: weakness, paralysis or movement symptoms.

Trial one: hypnosis against the lobby

In the outpatient trial, 44 people with motor conversion symptoms were randomized to a hypnosis-based treatment or a waiting list [1]. The hypnosis group improved more, on an observer-rated video scale of their motor symptoms and on an interview measure of disability, and the gains held at six months. There was no difference on a broad measure of psychological symptoms.

That’s a positive trial. It’s also, in the words I use for most “hypnosis works” headlines, a victory over the lobby. People on a waiting list get no attention, no structure, no reason to expect improvement. A treatment beating a waiting list tells you it’s better than nothing, which for a disabling condition is genuinely worth knowing, but not much about whether hypnosis was the active part.

Trial two: hypnosis as an add-on

The inpatient trial asked the sharper question. 45 people already in a comprehensive inpatient treatment program were randomized to that program with hypnosis, or the same program without it [2]. Both groups improved, substantially and clinically meaningfully. Hypnosis added nothing measurable on top. Hypnotizability didn’t predict who did well, either.

So: hypnosis beat waiting, and didn’t beat a good treatment program. Those results don’t contradict each other. They’re what you’d expect if a structured, attentive, hopeful treatment helps people with FND, and hypnosis is one way to deliver that, but not a uniquely powerful one. The old me would call that a point to the skeptics. I’d call it a point to reading the methods section. I’ll admit I enjoyed it anyway.

The wider context

The obvious follow-up question is whether anything does reliably beat a good comparison in FND. The answer, from two much larger trials, is: not on the main outcome, so far.

• CBT for dissociative seizures. In a trial of 368 adults, CBT added to standardized care didn’t significantly reduce seizure frequency at 12 months, though it did better on several secondary outcomes, including how bothersome the seizures were and quality of life [4].
• Specialist physiotherapy for motor FND. In a trial of 355 adults, specialist physiotherapy didn’t significantly improve physical functioning at 12 months compared with usual care, in a trial heavily disrupted by COVID-19 [5].

That doesn’t mean these treatments don’t help; both are widely used, and secondary outcomes and clinical experience matter. It does mean FND is hard to move in a trial against an active comparison, and hypnosis is in good company in having trouble doing it.

Limitations of the evidence

Limitation	Consequence
Two trials, 89 people in total	Far too few to estimate an effect with any precision
Both from one research team	No independent replication
Motor type only	Nothing on functional seizures, sensory symptoms or other presentations
Wait-list control in the positive trial	Can’t separate hypnosis from attention and expectation
Trials from 2002 and 2003, older diagnostic criteria	May not match how FND is diagnosed and treated now
No comparison with physiotherapy or CBT directly	No way to rank hypnosis against established approaches

What remains uncertain

Almost everything that would matter for a decision. Whether hypnosis helps functional seizures at all. Whether it adds anything to modern physiotherapy-led treatment. Whether some people with FND, perhaps the highly hypnotizable, benefit more, which the inpatient trial didn’t find. And, more broadly, a 2016 review noted that the evidence base is still dominated by case reports and case series, which can’t answer any of these [3].

What this means in practice

For functional neurological disorder, hypnosis has one small trial showing it’s better than waiting and one small trial showing it adds nothing to a good treatment program. That’s not enough to recommend it as a treatment in its own right, and anyone selling hypnosis as a cure for FND is selling well past the data. Within a specialist team that already uses it, as one part of a broader plan, it’s a reasonable thing to discuss. On its own, from a practitioner outside neurology, it isn’t a substitute for a neurologist’s diagnosis and an FND-informed treatment plan.

For why the control group changes the answer, see Expectation, Placebo and Hypnosis: Why Comparison Groups Matter. For the add-on question more broadly, see Hypnosis as an Adjunct to Medical Treatment.

When to see a clinician first. New weakness, numbness, loss of speech or vision, or a seizure needs urgent medical assessment, and sudden weakness or facial drooping is an emergency: call 911 in the US or 999 in the UK. Functional neurological disorder should be diagnosed by a neurologist, because its symptoms overlap with stroke, multiple sclerosis, epilepsy and other conditions that need different treatment.