Immersive VR simulates real-life environments with visual, sensory, and auditory input sent to the user in real time via a head-mounted display (HMD) – otherwise known as VR goggles, which many of us will have seen before. In this way, immersive VR can easily recreate environments that may be difficult to access in real life, which makes it a great tool in the treatment of various mental health disorders (Freeman et al., 2017). For example, in VR exposure treatment (VRET) for anxiety disorders, clients can become immersed in their feared environment regardless of how difficult it may be to encounter in real life (e.g., combat, flying) (Eshuis et al., 2021; Maples-Keller et al., 2017).
In the UK, the excellent gameChange (VR for psychosis) work led by Daniel Freeman has seen impressive results in an RCT and qualitative study, and is now being rolled out in the NHS and studied at a larger scale.
Previous narrative (e.g., Freeman et al., 2017; Maples-Keller et al., 2017; Park et al., 2019) and systematic reviews (e.g., Eshuis et al., 2021) show encouraging findings on the effectiveness of VR treatments. However, they often focus on specific disorders, or include studies with non-immersive VR (e.g., desktop-based treatments). Meta-analyses are sometimes conducted (e.g., Eshuis et al., 2021), but not always, which means that effects are not always clearly quantified.
In their comprehensive review and meta-analysis, Zeka and colleagues (2025) looked exclusively at immersive VR and whether it works as a treatment for various mental health disorders.
Immersive virtual reality is able to simulate real-life environments, making it a potentially great tool in the treatment of various mental health disorders, like anxiety.
Methods
The authors searched four databases and identified 55 randomised controlled trials (RCTs) comparing immersive VR to either i) other active treatments (e.g., CBT) or ii) passive controls (e.g., waitlist) for various mental health disorders. Measures of symptoms, cognition, functioning, and quality of life were the outcomes of interest.
Meta-analyses based on a random effect model were conducted separately for each disorder and each control condition. The studies that could not be included, due to the use of incompatible measures, were discussed separately in the paper.
Risk of bias was assessed with the Cochrane Risk-of-Bias Tool 2 (Higgins et al., 2019), and certainty of evidence with the GRADE tool (Schünemann et al., 2013).
Results
Of the 55 included studies (n = 3,031 participants), 45 compared immersive VR to an active treatment, whereas 27 compared immersive VR to a passive treatment (with 17 comparing both active and passive treatments – hence the odd numbers!). Fifty-one studies measured symptoms, five studies measured cognition, and 10 studies measured functioning or quality of life.
Per disorder, the meta-analysis found that:
- Addiction disorders (2 studies)
- VR outperformed active treatment in measures of state anxiety (Hedge’s g = 0.89, 95% CI [0.24 to 1.55]),
- but not of alcohol craving.
- Schizophrenia spectrum disorders (6 studies)
- VR outperformed active treatment regarding positive symptoms like hallucinations at post-treatment (g = 0.37, 95% CI [0.04 to 0.70]) but not at follow-up.
- There was no difference between treatments in negative symptoms (e.g., apathy), depressive symptoms, beliefs about voices, quality of life, or functioning.
- Panic and agoraphobias (7 studies)
- VR exposure treatment (VRET) was much better than passive controls at reducing agoraphobia symptoms at post-intervention (g = 0.90, 95% CI [0.45 to 1.35]).
- However, it did not differ from active treatments in measures of symptom and functioning, either at post-treatment or at follow-up.
- Social anxiety disorder (7 studies)
- Similarly, VR was better than passive controls at reducing social anxiety symptoms (g = 0.83, 95% CI [0.49 to 1.17]), though not depression symptoms.
- Compared to active treatments, VR did not differ in the initial analysis and appeared less effective (g = -0.24) when studies with a high risk of bias were removed.
- Specific phobias (15 studies)
- In comparison to passive controls, VR greatly reduced phobia symptoms (g = 1.07, 95% CI [0.22 to 1.92]) and phobia-related attitudes and beliefs (g = 0.92, 95% CI [0.46 to 1.38]).
- Although the initial analysis showed no difference between VR and active treatments, removing studies with a high risk of bias changed findings in favour of VR (g = 0.30).
- Generalised anxiety disorder (2 studies)
- VR interventions performed similarly to passive controls.
- Post-traumatic stress disorder (6 studies)
- VRET reduced depression (but not PTSD) symptoms more than passive controls (g = 0.67, 95% CI [0.22 to 1.13]).
- Initially, VRET was equally effective as active treatments at reducing PTSD or depressive symptoms, but removing studies with a high risk of bias changed results in favour of other active treatments (g = -0.20).
- Eating disorders (3 studies)
- VR was not better than other active treatments at reducing state anxiety and improving body satisfaction.
Approximately half of the reviewed studies were found to have a high risk of bias, mainly due to lacking information on how participants were randomised, and on blinding of assessors. For another 25% of studies, there were some concerns of bias, whereas the remaining 25% was judged as having a low risk of bias. The certainty of evidence was generally judged as low to very low.
The largest body of evidence (15 studies) was found for specific phobias, in which virtual reality was shown to reduce phobia-related symptoms and attitudes and beliefs in relation to passive controls, with large effects.
Conclusions
Generally speaking, the findings from this meta-analysis suggest that immersive VR is more effective than passive controls and, generally, at least as effective as other active treatments for a variety of mental health disorders. This is particularly the case for anxiety disorders, for which the larger number of studies has been conducted (n = 37). However, it is important to keep in mind that the literature is characterised by a small number of studies and a high risk of bias, which limits our confidence in these conclusions.
Generally speaking, immersive virtual reality appears to be more effective than passive controls, and at least as effective as other active treatments – but the evidence base is small, with high risk of bias.
Strengths and limitations
This review by Zeka and colleagues (2025) has several strengths. First, it was preregistered and followed a clear protocol in line with the PRISMA guidelines, thus adhering to high quality standards. Second, it focused explicitly on immersive VR, unlike some previous reviews that included non-immersive techniques. Non-immersive techniques employ advanced technology but may lack the element of “presence” in the virtual environment, which is considered important for treatment effects, and which is easier to achieve with immersive technologies (Bell et al., 2024). By focusing on a specific type of technology-assisted therapy, the review helps us understand its unique effects and challenges without obscuring them with these of different technology-assisted therapies.
However, there are also some limitations that need to be kept in mind.
The main limitation regards the quality of the included studies. Most had small sample sizes, which limit their statistical power to reveal true effects. Similarly, the pool of studies for each individual mental health disorder was relatively small. Apart from the meta-analysis on specific phobias, which contained 15 studies, almost all other meta-analyses included seven studies or fewer, again limiting the statistical power of each individual meta-analysis.
Also, many studies had a high risk of bias, with information on randomisation and blinding of assessors missing. Randomising participants to conditions and blinding the people who administer the questionnaires to conditions ensures that results are trustworthy – something which, at the moment, we have insufficient information to claim.
Taking everything together, we can say that the paper of Zeka and colleagues (2025) gives a good overview of the current state of the science as regards immersive VR treatment for mental health disorders, whilst at the same time highlighting the areas where more high-quality research is needed before safe conclusions can be drawn.
The literature in this review is characterised by a small number of studies per disorder, generally small study samples, and lack of information on randomisation and blinding, which limits our confidence in the findings.
Implications for practice
Despite the limitations mentioned above, the review of Zeka and colleagues (2025) has several potential implications.
First, a basic implication for researchers and funding agencies is that more (and more rigorous) research is needed in the area of immersive VR treatments for mental health disorders. At the time of writing this blog, VR goggles for gaming can be bought for a couple of hundred pounds. As VR technology becomes increasingly affordable, it is likely that immersive VR treatments will become more widely available as well. But will these be as effective as they will claim to be? A good understanding of how immersive VR treatment works, and for whom, will give us the necessary information to produce evidence-based immersive VR treatments. This way, we can avoid the concern that is often raised in the field of mental health apps, that is, that many of them are not, or not enough, evidence-based (e.g., Van Daele et al., 2020).
What about implications for clinical practice, though? Zeka and colleagues suggest that immersive VR interventions may cautiously be considered effective, but that:
more information on efficacy, tolerability, and barriers is needed prior to initiating broader implementation (p. 226)
In other words, the data is encouraging, but it seems rather premature to be suggesting the broader use of immersive VR in mental health.
Individual differences need to be taken into account when planning the use of immersive VR. For example, it has been suggested that not everyone has the capacity to feel present in the VR environment (Maples-Keller et al., 2017). It is likely that people who lack this “presence” will not benefit from the treatment, at least not to the same degree as someone feeling present. It may also be the case that people who do not feel present in the VR environment will not adhere to the treatment, for example because it does not seem relevant to them – and of course, a treatment that is not adhered to is a treatment that cannot be beneficial.
Another issue to not neglect is tolerability. Side effects of VR are not negligible, and include a variety of physical symptoms such as motion sickness and dry eyes (Park et al., 2019), which are collectively called “cybersickness” (Lundin et al., 2023). Concerningly, a disconnection from the self and the environment are also reported by users; however, side effects are not always considered in VR research (Lundin et al., 2023).
Before being able to confidently recommend the use of immersive VR with patients, we need a better understanding of both its effectiveness for treating mental health disorders, as well as the impact of its side effects.
Statement of interests
This elf has no conflict of interests to report.
Links
Primary paper
Zeka, F., Clemmensen, L., Valmaggia, L., Veling, W., Hjorthøj, C., & Glenthøj, L. B. (2025). The Effectiveness of Immersive Virtual Reality‐Based Treatment for Mental Disorders: A Systematic Review With Meta‐Analysis. Acta Psychiatrica Scandinavica, 151(3), 210-230.
Other references
Bell, I. H., Pot-Kolder, R., Rizzo, A., Rus-Calafell, M., Cardi, V., Cella, M., Ward, T., Riches, S., Reinoso, M., Thompson, A., Alvarez-Jimenez, M., & Valmaggia, L. (2024). Advances in the use of virtual reality to treat mental health conditions. Nature Reviews Psychology, 3(8), 552–567.
Eshuis, L. V., Van Gelderen, M. J., Van Zuiden, M., Nijdam, M. J., Vermetten, E., Olff, M., & Bakker, A. (2021). Efficacy of immersive PTSD treatments: A systematic review of virtual and augmented reality exposure therapy and a meta-analysis of virtual reality exposure therapy. Journal of Psychiatric Research, 143, 516-527.
Freeman, D., Reeve, S., Robinson, A., Ehlers, A., Clark, D., Spanlang, B., & Slater, M. (2017). Virtual reality in the assessment, understanding, and treatment of mental health disorders. Psychological Medicine, 47(14), 2393–2400.
Higgins, J. P., Savović, J., Page, M. J., & Sterne, J. A. C. (2019). RoB 2 Guidance: Parallel Trial. Cochrane Collaboration, 28, 1–24.
Lundin, R. M., Yeap, Y., & Menkes, D. B. (2023). Adverse effects of virtual and augmented reality interventions in psychiatry: Systematic review. JMIR Mental Health, 10(10), e43240.
Maples-Keller, J. L., Bunnell, B. E., Kim, S. J., & Rothbaum, B. O. (2017). The use of virtual reality technology in the treatment of anxiety and other psychiatric disorders. Harvard Review of Psychiatry, 25(3), 103–113.
Park, M. J., Kim, D. J., Lee, U., Na, E. J., & Jeon, H. J. (2019). A literature overview of virtual reality (VR) in treatment of psychiatric disorders: Recent advances and limitations. Frontiers in Psychiatry, 10, 1–9.
Schünemann, H., Brożek, J., Guyatt, G., & Oxman, A. E. (2013). GRADE Handbook for Grading Quality of Evidence and Strength of Recommendations. Grade Working Group.
Van Daele, T., Karekla, M., Kassianos, A. P., Compare, A., Haddouk, L., Salgado, J., Ebert, D. D., Trebbi, G., Bernaerts, S., Van Assche, E., & De Witte, N. A. J. (2020). Recommendations for policy and practice of telepsychotherapy and e-mental health in Europe and beyond. Journal of Psychotherapy Integration, 30(2), 160–173.
