- Design
- assessor-blind, parallel-group randomised controlled trial at one academic centre
- Population
- 136 adults with schizophrenia or schizoaffective disorder and persistent auditory hallucinations; 56.6% on clozapine
- Primary outcome
- change in hallucination severity on the Psychotic Symptoms Rating Scale to 3 months after therapy
- Effect
- time-by-treatment interaction favoured virtual reality, P = 0.013, Cohen d = 0.614 at 3 months
An assessor-blind trial in Montreal randomised 136 adults with schizophrenia or schizoaffective disorder and persistent auditory verbal hallucinations to nine sessions of virtual reality-assisted therapy or nine sessions of CBT targeted at the voices, stratified by sex and clozapine use. More than half, 56.6%, were on clozapine - this was a resistant group. Both interventions worked on the same idea: changing the patient's beliefs about, and relationship with, the voices.
The time-by-treatment interaction on the Psychotic Symptoms Rating Scale hallucination subscale favoured virtual reality (P = 0.013), with a moderate between-group effect at three months after therapy (Cohen d = 0.614). Both arms improved within themselves - large effects for virtual reality (d = 0.81 post-treatment, d = 1.17 at three months), moderate for CBT (d = 0.58 then d = 0.39). Secondary outcomes showed no between-group difference, though both arms improved on general psychotic symptoms, emotional regulation and voice acceptance.
The finding worth carrying is not that a headset is better than a therapist. It is that a short, focused, relational intervention produced large improvements in voices that had persisted despite clozapine, and that the gains were still growing three months later. Nine sessions is deliverable. Where virtual reality is not available - which is almost everywhere - the CBT arm is the deliverable version of the same finding, and it also worked.
- Ask specifically about persisting voices in clozapine-treated patients rather than assuming the ceiling has been reached
- Offer a short, focused course aimed at beliefs about and relationship with the voices
- Measure before and after with a hallucination scale so benefit is visible to the patient
- Continue antipsychotic treatment unchanged - this was adjunctive, not a substitute
- Review at three months: in this trial the virtual reality effect kept growing after therapy ended
Why it matters
Persistent voices on clozapine are usually treated as the end of the road; two short psychological courses moved them.
Don't overread it
Single-centre, patients could not be blinded, and the comparator was a deliberately short nine-session CBT course rather than a full one.
The statistics, in plain English
Cohen d describes effect size in standard deviations: 0.6 is moderate and clinically noticeable, 1.17 is large. The important distinction here is between-group and within-group - both therapies improved patients substantially, and the extra benefit of virtual reality over CBT was the moderate 0.614. The trial was assessor-blind but patients obviously knew which they received, which inflates subjective improvement in the newer, more engaging arm.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for psychiatry, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free