First, we want to thank Verma & Verma for their review and also comment that the use of only self-report is not ‘acknowledged by the authors’. Patient's performance in delivering an impromptu speech was recorded and rated by independent assessors. Results on the behavioural assessment task support the results from the self-reports.

To clarify the randomisation procedure that was used, as stated in the article, we used a table of random numbers. Following the procedures and table from Kirk,Reference Kirk¹ before starting recruitment we created a list of which condition participants would be randomly assigned to. When a participant met the selection criteria, he or she was assigned to the next available slot on the list, with the numbers 1, 2 or 3 corresponding to the experimental conditions.

As for the potential effect of self-exposure at home, it is indeed a potential limitation of the study. Care was taken to limit self-exposure with anti-in-vivo exposure instructions given to participants in the in virtuo condition and no homework given to all participants. Self-exposure, intended or unintended by participants, remains a possibility. But it is doubtful the success of in virtuo exposure, especially when it was superior to in vivo, could be attributed to this limitation.

Verma & Verma mentioned relevant limitations associated with the use of a waiting list compared with a placebo control condition. But when designing a clinical study, these limitations must be carefully weighed against ethical and practical considerations, especially when a gold-standard control condition is also included in the research design. Using Solomon's 4-group design is indeed a very effective solution to control for the effect of assessment.

In conclusion, it is interesting to highlight one of the findings of the study that might have not caught the attention of Verma & Verma. When it comes to cost-effectiveness, our results showed that using virtual reality to conduct exposure was at least as effective as in vivo, and more effective on some measures such as the Liebowitz Social Anxiety Scale, but also less costly in terms of efforts and financial burden based on the Specific Work for Exposure Applied in Therapy scale.