The paper by Bouchard et al Reference Bouchard, Dumoulin, Robillard, Guitard, Klinger and Forget¹ builds upon many facets of methodological deficiencies in previous literature on the issue of the utility of virtual reality therapies in managing social anxiety disorder (SAD).

Positive critique: the study used multiple scales for assessing the various facets of SAD and therapy that usually are lacking in studies using psychotherapies (in particular virtual reality/computer-stimulated therapies). These included measures for assessing burden, challenges and the costs of conducting the exposure; measures to assess how feasible it was for the therapist to perform the therapy; and scales to assess unwanted negative side-effects induced by immersions in virtual reality (cybersickness). The study made the sample more comparable with a real-world setting by including patients with other comorbidities, such as depression, generalised anxiety disorder, panic attacks and substance use, which increases the generalisability of their findings. It also describes the cognitive–behavioural therapy (CBT) used in detail, which usually is lacking in psychotherapy studies.

The study also reports on follow-up at 6 months, with gains maintained. Moreover, it used both performance and non-performance social situations in the treatment of SAD with virtual reality thereby increasing the importance of findings.

Negative critique: the findings are based only on self-reports (as acknowledged by the authors too) restricting its firm conclusions. Information regarding the randomisation process is not mentioned, although the authors do mention that both the therapists and patients did not know which group they were in until the first therapy session.

A possible bias could have arisen because, in individuals receiving in virtuo CBT, after the first seven sessions of cognitive therapy they would have gone out into the real world and may have experienced real in vivo exposures that might create bias in outcomes by rendering the differences attributable not just solely on the basis of in virtuo exposure.

The researchers used individuals on a waiting list as controls, which has been reported in the psychotherapy literature to act not as a placebo but rather as a nocebo.Reference Hart, Fann and Novack² The effect size using a waiting list has been reported to be greater compared with that when using a placebo. Psychological placebo is associated with a significantly greater reduction in symptoms than placement on a waiting list, possibly because of the effects of anticipation, hope or faith. Also, there was no independent, masked investigator to assess the outcome.

Another facet of consideration in such psychotherapeutic studies is the recognition of the Solomon 4-group design effect as assessments may interact with interventions to either strengthen or weaken the observed effects producing biased estimates of effects.Reference McCambridge, Butor-Bhavsar, Witton and Elbourne³ Overall, the study paves the way towards a strengthening of the evidence towards utility of virtual reality applications in psychiatric conditions for a positive outcome.