People receiving AVATAR-Extended therapy reported major improvements in distress and intensity of troubling voices after 16 weeks, with lasting benefits for mood and well-being.
Study: Digital AVATAR therapy for distressing voices in psychosis: the phase 2/3 AVATAR2 trial. Image Credit: Jason Alden Photography Ltd. Commissioned by Wellcome.
In a recent study published in Nature Medicine, researchers evaluated a new therapy to reduce distressing voices in people with psychosis.
Their findings indicate that after 16 weeks, the therapy showed some benefits in reducing the frequency of voices and associated distress, with no adverse impacts associated with the treatment.
Background
Digital tools are transforming mental health care by improving access, engagement, and effectiveness. Auditory verbal hallucinations or voices are common in schizophrenia and cause distress, but responses to medication and therapy are often insufficient, highlighting the need for new treatment methods.
An emerging digital treatment creates an avatar of the distressing voice, and the person experiencing the sound can customize its sound and appearance. Therapy involves guided dialogues between the patient and the avatar, aiming to reduce distress and promote empowerment.
Previous studies have established the safety of the intervention and showed positive impacts on the severity of the voices. An early trial found a significant reduction in voice severity compared to supportive counseling, while additional pilot studies supported effectiveness and feasibility and highlighted the potential to personalize the therapy further to address past trauma and other individual experiences.
About the study
From January 2021 to November 2022, 642 individuals were assessed, resulting in 345 participants being randomized into three groups: a brief avatar therapy (AV-BRF) including 116 people, an extended version of the avatar therapy (AV-EXT) including 114, and a treatment-as-usual (TAU) control including 115.
After 16 weeks, data were available for 300 participants (87%), with 298 (86%) remaining at 28 weeks. Loss to follow-up included 12 for TAU, 17 for AV-BRF, and 16 for AV-EXT after 16 weeks, and 11 for TAU, 15 for AV-BRF, and 21 for AV-EXT after 28 weeks.
Baseline characteristics were not significantly different between groups; however, the sample was predominantly male (61.4%), mostly diagnosed with schizophrenia (43.8%), with 40% from minoritized ethnic backgrounds and a mean of 13 years in mental health care.
Completion criteria required 4 of 6 sessions for AV-BRF and 10 of 12 for AV-EXT, with an 82% completion rate in AV-BRF and 58% in AV-EXT. Participants attended a mean of 5.1 (AV-BRF) and 8.2 (AV-EXT) sessions overall, with completed treatments averaging 6.2 sessions for AV-BRF and 11.5 for AV-EXT, and active treatment sessions lasting around 65 minutes for both treatments.
Findings
Significant reductions in voice-related distress were observed at 16 weeks for both AV-BRF and AV-EXT. AV-BRF showed a mean reduction of 1.05 points, and AV-EXT showed a reduction of 1.6 points. By 28 weeks, no significant distress reduction remained in either group.
Voice severity improved significantly at 16 weeks in both AV-BRF (a 2.04-point reduction) and AV-EXT (a 2.32-point reduction), but effects were not maintained at 28 weeks. Voice frequency was significantly reduced in AV-EXT at both 16 weeks (−0.62 points) and 28 weeks (−0.89 points); however, AV-BRF showed no significant effect on the frequency at either time point.
Both AV-BRF and AV-EXT led to improvements in voice acceptance at both time points. AV-EXT alone showed an improvement in perceived voice omnipotence at 28 weeks. For well-being and recovery outcomes, AV-EXT showed improvements at both time points, while AV-BRF showed effects only at 16 weeks. Reductions in anxiety and depression were significant for AV-BRF at both follow-ups but only at 16 weeks for AV-EXT.
Researchers recorded 58 serious adverse events (SAEs), with 51% occurring in the AV-EXT group. Two mortalities in the AV-EXT group were deemed unrelated to the treatment. No significant moderating factors were identified. Compliance analyses suggested larger treatment effects at 16 weeks, with diminishing differences by 28 weeks.
Conclusions
This multisite trial compared brief (AV-BRF) and extended (AV-EXT) forms of avatar therapy for auditory hallucinations conducted by various therapists. Both AV-BRF and AV-EXT improved voice-related distress and severity at 16 weeks, though these effects lessened by 28 weeks.
AV-EXT also significantly reduced voice frequency at both points, addressing a key patient priority. AV-EXT showed larger effects on distress than AV-BRF, with AV-EXT surpassing the threshold for clinical significance. Secondary outcomes differed; AV-BRF led to sustained improvements in mood, while AV-EXT showed greater benefits in well-being and empowerment.
Avatar therapy was safe, though AV-EXT reported more adverse events, possibly due to longer, trauma-focused sessions. Limitations included a TAU comparison instead of active therapy, with AV-EXT having lower completion rates, highlighting a need for better engagement strategies. Future directions include the integration of artificial intelligence and further studies to assess cost-effectiveness and explore broader adoption in healthcare settings.
Yale study links paranoia to visual perception deficits