Dual diagnosis refers to patients with both severe mental illness and substance abuse. Dual diagnosis is therefore a challenging condition to treat, and the group typically represents the most vulnerable individuals in society. Historically, research on dual diagnosis has been underprioritized, and thus, we still do not know enough about how to best assist this vulnerable group.

However, new studies indicate that virtual reality programs can reduce anxiety in patients with psychotic disorders. They achieve this by providing access to a virtual therapist and lifelike environments where patients can challenge their thoughts about the dangers of navigating the world. For both psychotic disorders and substance abuse, we know that anxiety often plays a role in the clinical picture. Therefore, anxiety almost always has an impact on dual diagnosis patients, where it is crucial in maintaining substance abuse and functional impairment. Despite this, anxiety is rarely a focus in existing treatment options, as it is too resource-intensive in addition to an already intensive treatment process.

This study investigates whether the resource barrier can be overcome and whether hospitalized dual diagnosis patients can experience reduced anxiety, fewer relapses, and better outcomes after discharge when their anxiety is treated through partially automated virtual reality therapy.

Introduction - Dual diagnosis (i.e., comorbid psychosis and substance use disorder) is characterized by more severe psychopathology, higher morbidity, less treatment adherence, and smaller treatment gains compared to psychosis alone. Further, there is a clear association between educational status and both prevalence and mortality of psychosis and substance use. Thus, the debilitating symptoms of these disorders propagate social inequality as well as being a substantial societal burden in purely economic terms. Importantly, this burden seems to increase when patients cannot access evidence-based treatment and recent policy changes mean that the amount of dual diagnosis patients in the Danish mental health services will soon more than triple. Improving the accessibility and efficacy of treatment for this vulnerable population is therefore more than ever, a critical unmet need. For this reason, this study aims to investigate the feasibility, efficacy, and acceptability of a partially auto-mated virtual reality (VR) based psychotherapeutic intervention targeting anxiety in dual diagnosis patients. Social avoidance has been theorized to be a highly relevant treatment target in dual diagno-ses populations since it is thought to play a critical role in maintaining symptoms, worsening mood and functional impairment, and deteriorating treatment adherence. Recent studies have found that VR-based cognitive behavioral therapy for psychosis (CBTp) resulted in significant reductions in anxiety and psychotic symptoms, while being safe and acceptable. Studies also show that this treatment can be automated, thus greatly increasing accessibility of treatment. Further, automatiza-tion allows patients to continue treatment even after hospital discharge, potentially maintaining treatment gains for longer. In short, there are several findings which indicate that automated VR-based CBTp is a promising treatment for dual diagnosis and therefore this present trial will be the first in the world to investigate this.

Background - Though CBTp has been shown to be an effective and safe treatment that patients prefer to medication, it is still poorly implemented in many countries. Only 0.07 % of the 5.5 billion DKK that schizophrenia costs Danish society yearly, are related to contacts with a psychologist or psychiatrist, implying that CBTp is also poorly implemented in Denmark. For dual diagnosis patients, prospects of receiving CBTp are further complicated because hospitals offering CBTp often forward patients with comorbid SUD to municipal services. In addition, the evidence supporting psychotherapy for dual diagnosis is still inadequate and the long-term effect of treat-ment remains unclear. This is likely because psychotherapy can be difficult to access for dual diag-nosis patients, being a long-term and complex intervention, which requires a high level of competence from the administering clinician. The lack of high quality clinical research, along with the complexity of the treatment, represent major barriers for implementation. Incorporating the use of VR technology in high quality clinical research may be one way to develop CBTp towards be-coming more accessible, increase long term efficacy and address anxiety symptoms, and easier to implement.

Inclusion criteria

• Adult patients (+18 years) admitted to the standard dual diagnosis treatment program in the Mental Health Services of the Capital Region of Denmark.
• Fulfilling diagnostic criteria for a F20-29 disorder and dependence syndrome of either alcohol or drugs in the ICD-10 diagnostic manual.
• Scoring 6 or above on the avoidance subscale and 66 or above on the distress subscale scale of the O-AS. This corresponds to severe levels of avoidance and distress36.

Exclusion criteria

• Incapable of providing informed consent.
• Insufficient knowledge of the Danish and/or English language.

When a participant is randomized to TAU+VR, the senior researcher reveals the allocation to the assigned clinician via secure e-mail. The assigned clinician then schedules the eight therapy sessions with the participant who is thus unblinded to their treatment allocation. If the participant is randomized to TAU, the assigned clinician will inform the participant as soon as possible. The researchers collecting data from the patient and the researchers performing data analysis remain blinded. There are no circumstances wherein unblinding of outcome assessors or data analysists is permissible.