What a 2025 Systematic Review Found About Immersive Virtual Reality-Based Interventions for Mental Disorders

Introduction

Mental disorders remain among the leading causes of disability worldwide, yet access to effective treatment is still uneven, delayed, and often insufficient. Even where evidence-based therapies exist, many patients do not receive them in a timely way or do not complete them. Barriers include cost, stigma, shortages of trained clinicians, long waiting lists, low treatment adherence, and the practical difficulty of delivering interventions that require repeated exposure, social rehearsal, or highly structured behavioral training. In this context, digital psychiatry has expanded rapidly, but not all digital tools offer the same clinical possibilities. Among the most intriguing developments is immersive virtual reality, which has moved from an experimental technology to a serious therapeutic platform in mental health research.

This article reviews the findings of a 2025 systematic review with meta-analysis on immersive virtual reality-based interventions for mental disorders. Rather than treating VR as a general innovation story, it focuses on what the review actually found about effectiveness across diagnostic categories, where the evidence appears strongest, and where it remains uncertain.

The review screened 2443 studies and ultimately included 55 studies involving 3031 participants across 10 mental disorders. It reported statistically significant post-treatment effects versus active controls for alcohol use disorder (reduced state anxiety) and schizophrenia spectrum disorders (reduced psychotic symptoms). Compared with passive controls, significant effects were observed for panic and agoraphobia, social anxiety disorder, specific phobias, and depressive symptoms in PTSD. At the same time, more than half of included studies were judged to have a high risk of bias, and the overall certainty of evidence was generally low to very low. These two facts need to be held together: the field is promising, but the current evidence is still provisional.

Immersive virtual reality differs from standard screen-based digital interventions because it places the user inside a computer-generated environment that can be experienced as spatially and psychologically real. Through head-mounted displays and interactive environments, patients can engage with simulations that evoke emotions, behaviors, and cognitive responses closer to those triggered by real-life situations. This feature is particularly important in psychiatry, where treatment often depends not only on insight or information, but on lived experience. A patient with social anxiety may intellectually understand that a conversation is safe, yet still react with intense distress when placed in a realistic social setting. VR can reproduce that setting in a controlled, repeatable, and clinically adjustable way.

This capacity for controlled immersion explains why VR has attracted growing interest across multiple psychiatric conditions. In anxiety disorders, it offers a way to deliver exposure-based therapy with a level of precision and flexibility that may be difficult to achieve in vivo. In psychotic disorders, it can be used to simulate social contexts for skills training or to reduce distress associated with paranoid ideation. In substance use disorders, it can present craving-related cues while allowing clinicians to observe and intervene in real time. More broadly, VR offers something that many digital mental health tools do not: the possibility of combining standardization, ecological realism, and therapeutic intensity within a single platform.

The sections that follow examine what immersive VR adds to treatment, where the 2025 review found the clearest benefits, why some VR interventions appear to work better than others, and what methodological and practical barriers still limit adoption.

What Immersive VR Adds to Mental Health Treatment

The therapeutic interest in immersive virtual reality does not rest on novelty alone. Its importance lies in the fact that it can reproduce psychologically meaningful experiences in a controlled environment, allowing clinicians to work not only with what patients think and report, but with how they feel, react, and behave in simulated situations that resemble real life. This makes immersive VR fundamentally different from many conventional digital mental health tools, which are often limited to psychoeducation, self-monitoring, or text-based support.

A central concept here is presence, meaning the subjective sense of being inside the virtual environment rather than merely observing it on a screen. Presence matters clinically because many psychiatric symptoms are activated by context. Fear, craving, suspiciousness, social discomfort, and avoidance are often difficult to access through discussion alone. Patients may describe them accurately, but description is not the same as therapeutic exposure or behavioral engagement. Immersive VR allows patients to encounter situations that trigger relevant responses while remaining in a setting that is safe, adjustable, and supervised. In practical terms, this means that treatment can become more experiential without requiring the unpredictability of real-world exposure.

This feature is especially relevant for exposure-based treatment. In anxiety disorders, effective therapy often depends on repeated confrontation with feared situations until avoidance decreases and emotional learning occurs. Yet in vivo exposure can be difficult to organize, expensive, or unacceptable to patients. Some situations are logistically hard to recreate, while others are too intense to begin with. VR offers a middle path. It can simulate heights, crowds, public speaking, enclosed spaces, or trauma-related environments with a level of control that allows clinicians to titrate intensity step by step. This is one of the clearest reasons why VR has shown promise in conditions such as social anxiety disorder, panic and agoraphobia, and specific phobias.

Immersive VR may also provide advantages beyond classical exposure. In disorders where skills practice is important, virtual environments can be used to rehearse behaviors repeatedly under standardized conditions. A patient with psychosis-related social difficulties may practice interaction in settings that would otherwise be stressful or unavailable for structured rehearsal. A patient with substance use problems may be exposed to craving-related cues while learning and testing coping strategies. In these settings, VR is not only a simulation tool. It becomes a platform for behavioral training, emotional regulation, and therapeutic experimentation.

Another important mechanism is embodiment and environmental controllability. Because VR creates a surrounding sensory world, it can shape attention more powerfully than standard screen-based interventions. This may increase emotional activation when activation is needed for treatment, but it may also improve engagement by making the intervention feel more concrete and less abstract. For some patients, imaginal techniques are difficult because they cannot vividly generate feared or emotionally relevant scenarios. VR reduces reliance on imagination by presenting the scenario directly. This can be especially useful in patients who are cognitively avoidant, less verbally expressive, or less responsive to traditional imaginal work.

The standardization of VR is another practical advantage. Once a therapeutic environment is designed, it can be repeated with high consistency across sessions and patients. This allows interventions to be delivered with a level of procedural stability that is difficult to achieve in many real-world settings. Standardization has value not only for research, but for clinical care, because it supports structured progression and makes it easier to calibrate intensity. At the same time, good VR programs can still be individualized. The therapist can modify pace, duration, environmental complexity, and task demands, preserving flexibility within a reproducible treatment frame.

The role of the clinician remains important here. Although some VR interventions are more automated, the best current use cases often involve therapist-guided delivery, where immersion is embedded within an evidence-based treatment model rather than treated as a standalone device experience. The therapist can frame the session, identify targets, monitor emotional response, and help translate what happens in virtual space into insight and behavior change outside it. This is likely one reason VR tends to appear most convincing when it extends an established therapeutic principle rather than trying to replace therapy wholesale.

Immersive VR may also help with motivation and adherence. Some patients find it easier to engage with active, visually rich, task-based interventions than with purely verbal formats. The novelty of VR is not sufficient in itself to sustain treatment, but increased engagement can matter in disorders where dropout is a major problem. When patients experience the intervention as immediate and relevant, they may be more willing to continue, especially if they perceive clear emotional or behavioral change across sessions.

At a broader level, immersive VR adds something clinically valuable because it sits at the intersection of realism and control. It is more emotionally and behaviorally potent than many digital tools, yet more structured and modifiable than real-world exposure or unstructured rehearsal. That combination explains why VR has become such an important area of study in modern psychiatry. Its potential does not lie in replacing existing treatment logic, but in giving that logic a more precise and immersive delivery format.

Evidence Across Diagnostic Categories: Where the 2025 Review Found the Strongest Signals

The most important question for immersive virtual reality in psychiatry is not whether the technology is engaging, but whether it produces clinically meaningful improvement. The 2025 systematic review and meta-analysis is especially useful here because it examined immersive VR interventions across a wide range of mental disorders and compared outcomes across symptom domains, control conditions, and patient groups. Its overall picture is encouraging but selective.

One of the clearest areas of benefit was anxiety disorders. Compared with passive control conditions, the review found statistically significant effects for panic and agoraphobia, social anxiety disorder, and specific phobias. This makes strong clinical sense because anxiety disorders are often maintained by avoidance and exaggerated threat prediction, and VR is especially well suited to graded, repeatable exposure. In these disorders, immersive simulation aligns closely with a well-established therapeutic mechanism rather than introducing a completely new one.

The review also found a significant effect for depressive symptoms in PTSD versus passive controls. This is noteworthy, but it should be interpreted more cautiously than the clearest anxiety findings. PTSD is clinically more complex than many phobic disorders, and symptom domains do not always move together. A positive finding in one domain does not necessarily imply equally strong evidence across the full PTSD symptom structure.

Compared with active control conditions, the review found significant post-treatment effects for alcohol use disorder, specifically reduced state anxiety, and for schizophrenia spectrum disorders, specifically reduced psychotic symptoms. These are especially interesting findings because active controls are more demanding comparators than waitlist or minimal treatment. They suggest that in some contexts immersive VR may offer clinically specific value rather than benefiting only from novelty or intensity.

At the same time, the review found no significant differences versus active controls for functioning and quality of life in schizophrenia spectrum disorder and panic or agoraphobia. This matters because symptom change is not the same as broader recovery. A treatment may reduce distress without clearly improving day-to-day functioning, autonomy, or overall life quality.

Taken together, the review suggests a relatively clear hierarchy. The strongest and most coherent support currently lies in selected anxiety disorders, where immersive exposure closely matches the underlying treatment logic. Alcohol use disorder and schizophrenia spectrum disorders show meaningful promise in selected symptom domains, but the evidence there is more provisional. PTSD-related findings are clinically interesting, though less uniformly settled across outcomes.

Why Some VR Interventions Work Better Than Others

The uneven results of immersive virtual reality in mental health care are not random. Some interventions produce stronger and more consistent outcomes because they fit the clinical logic of the disorder more closely, are delivered with better structure, and target mechanisms that are already known to matter in treatment. In other words, VR is not effective simply because it is immersive. It tends to work best when immersion is used in a way that reinforces a sound therapeutic model.

One of the most important factors is diagnostic fit. Disorders differ in how well their core symptoms can be translated into a virtual environment. Anxiety disorders are a strong example because fear, avoidance, and threat anticipation can often be activated through realistic simulations. A patient with fear of social scrutiny, enclosed spaces, or crowded public settings can encounter these triggers in VR in a way that is emotionally salient but still controlled. This means that the intervention is not relying on novelty. It is amplifying an established treatment principle, namely exposure and corrective learning.

By contrast, conditions with more diffuse, internally driven, or biologically complex symptom patterns may be harder to treat through immersion alone. This does not mean VR has no place in these disorders, but rather that its effect may be less direct. In schizophrenia spectrum disorders, for example, VR may be helpful for social rehearsal, distress reduction, or work on suspiciousness in interactive contexts, but the route from simulation to symptom improvement is more complex than in a simple phobia.

A second factor is the degree of immersion and realism. Therapeutic VR depends partly on whether the patient experiences the simulation as sufficiently believable to trigger relevant emotional and behavioral responses. If the environment feels artificial, limited, or emotionally flat, it may fail to activate the processes that treatment is meant to change. Yet realism is not just a technical issue. The environment must also be clinically relevant.

Therapist involvement is another major variable. The strongest interventions are often not purely technological. They are clinically guided experiences embedded within a broader treatment structure. A therapist can prepare the patient, identify goals, calibrate the challenge level, monitor distress, and help consolidate learning after the session. This is especially important because raw exposure is not enough on its own. The therapeutic effect depends on what the patient learns during exposure, how avoidance changes, and whether new responses generalize beyond the virtual environment.

Treatment duration and session structure also matter. Brief or poorly sequenced interventions may produce weaker effects simply because they do not provide enough repetition or progression. In many psychiatric treatments, meaningful improvement depends on cumulative learning across multiple sessions. VR is unlikely to be different.

Another reason some VR interventions work better is target specificity. Studies that focus on a clearly defined symptom process tend to be more convincing than those aiming at broad improvement across multiple domains at once. A VR program designed to reduce public speaking fear, cue-triggered alcohol craving, or social avoidance may have a sharper effect because the target is narrow and measurable. When the outcome is more global, such as overall functioning or quality of life, the pathway from virtual session to broad improvement becomes less direct and more difficult to detect.

Overall, immersive VR appears most effective when it translates a well-established therapeutic mechanism into a realistic, structured, and clinically guided environment. The technology itself is not the treatment. Its value depends on how precisely it delivers the treatment logic already known to matter.

Limits of the Evidence: Bias, Heterogeneity, and Low Certainty

The current literature on immersive virtual reality in mental health care is promising, but it is also methodologically fragile. This tension is one of the defining features of the field. On the one hand, many studies report symptom improvement, strong patient engagement, and clinically plausible mechanisms of action. On the other hand, the overall strength of the evidence is weakened by high risk of bias, substantial heterogeneity, and generally low certainty across many outcomes.

A central concern is the risk of bias in the underlying studies. In psychotherapy and digital intervention research, bias can enter at multiple stages, including patient selection, randomization, allocation concealment, blinding, outcome assessment, and reporting practices. VR trials are especially vulnerable in some of these areas because participants usually know whether they are receiving an immersive intervention, and clinicians often know as well.

Small sample size is another major weakness. Many VR studies in psychiatry are still relatively small, which limits statistical power and increases the likelihood of unstable findings. Small trials are more vulnerable to exaggerated effect sizes, especially when early positive studies receive disproportionate attention.

The issue of heterogeneity is equally important. Immersive VR is not a single intervention. It is a delivery platform that can be used in very different ways across disorders, settings, and treatment models. Studies vary in hardware, software design, number of sessions, duration of exposure, degree of therapist guidance, and type of control condition. They also differ in which outcomes they prioritize. When such diverse studies are pooled together, it becomes difficult to know whether a positive result reflects the general value of VR or the success of a particular version of it.

Control groups create another layer of interpretive difficulty. A comparison with waitlist or treatment as usual is not equivalent to a comparison with an active and well-structured alternative intervention. VR may outperform passive controls simply because it is more engaging, more intensive, or more novel. When compared with strong active controls, the effect may narrow or disappear. This does not mean VR is ineffective. It means that the relevant question shifts from whether VR is better than minimal intervention to whether it adds meaningful benefit beyond already established approaches.

Outcome measurement itself is not always consistent. Some studies use clinician-rated scales, others rely on self-report instruments, and some combine several outcome types without clear prioritization. Follow-up periods may also be short, making it hard to determine whether observed gains are durable. A treatment that reduces symptoms immediately after the final session may still have limited clinical value if the effect fades quickly.

Finally, the problem of certainty of evidence must be taken seriously. A field can generate many positive results and still provide low-certainty evidence if those results come from studies with design weaknesses, inconsistent findings, or wide confidence intervals. That is exactly why the current state of VR psychiatry should be described as promising but provisional.

Clinical Implementation: Practical Promise and Real-World Barriers

If immersive virtual reality is to move from promising research into routine psychiatric care, its success will depend not only on efficacy, but on whether it can be implemented in ordinary clinical settings without excessive burden. A treatment may work well under study conditions and still remain marginal in practice if it requires too much equipment, too much staff training, or too much disruption to established workflows.

At present, the most realistic clinical use cases appear to be settings where the intervention logic is already well established and the value of controlled simulation is obvious. Anxiety-focused clinics are a leading example, because exposure-based therapy already has strong evidence and VR can be integrated as a more flexible delivery format. Addiction treatment is another plausible domain, particularly for craving-related cue exposure and coping practice. Psychosis services may also benefit, especially in programs that already use structured psychosocial interventions and can incorporate VR into social skills work or distress-focused cognitive-behavioral approaches.

Even in favorable settings, implementation requires infrastructure. Clinics need head-mounted displays, compatible software, reliable technical support, and physical space where sessions can be delivered safely and privately. Services also need protocols for device maintenance, software updates, cleaning procedures, and troubleshooting during sessions.

Staffing is equally important. A VR intervention is not simply a headset handed to a patient. In most serious clinical applications, it requires a clinician or trained facilitator who understands both the treatment rationale and the technical platform. This means services need time for training, supervision, and adaptation of existing workflows.

Patient tolerability also matters. Some users experience cybersickness, visual fatigue, dizziness, disorientation, or rapid emotional exhaustion, all of which can reduce adherence or make the intervention unsuitable. Careful patient selection and session pacing are therefore essential.

Another barrier is integration into care pathways. For VR to be genuinely useful, clinicians need clarity about when to introduce it, how to monitor progress, and how to connect virtual-session learning with treatment outside the headset. Without that link, VR risks becoming an isolated experience rather than a meaningful therapeutic component. This is why it is best framed as an adjunct to established care, not a wholesale substitute for psychotherapy, rehabilitation, or psychiatric management.

Future Directions for VR Psychiatry Research

The next stage of research on immersive virtual reality in psychiatry needs to focus less on proving that VR is interesting and more on demonstrating when, for whom, and under what conditions it produces durable clinical benefit. Larger randomized trials are an obvious priority, but size alone will not solve the field’s problems. Future studies need stronger active comparators, clearer intervention protocols, and more consistent outcome measures so that results can be meaningfully compared across disorders and treatment formats.

Longer follow-up is especially important. Much of the current literature emphasizes short-term symptom change, yet psychiatry also depends on maintenance of benefit, relapse prevention, and functional recovery. Future work should therefore pay greater attention to durability, quality of life, social functioning, and real-world behavior outside the treatment setting.

Another important direction is better matching of intervention design to disorder profile and patient characteristics. VR is unlikely to function as a one-size-fits-all tool. Its strongest role may lie in more precise applications, where the virtual environment is closely aligned with the patient’s symptom structure, avoidance pattern, or behavioral goals.

Conclusion

Immersive virtual reality has emerged as one of the most clinically interesting developments in contemporary mental health care because it allows established therapeutic principles to be delivered in highly controlled, emotionally engaging, and behaviorally relevant environments. The 2025 systematic review suggests that its strongest current support appears in selected anxiety disorders, where exposure-based mechanisms align especially well with the logic of immersive simulation. At the same time, findings in alcohol use disorder and schizophrenia spectrum disorders suggest that VR may also have meaningful value beyond anxiety, particularly when it is used to target clearly defined symptom processes rather than broad and diffuse outcomes.

However, the present evidence remains promising rather than definitive. Many studies are small, methodologically heterogeneous, and affected by substantial risk of bias. More than half of the included studies in the review were rated as high risk of bias, and the overall certainty of evidence was generally low to very low. This means that enthusiasm for the technology should be balanced by caution about the certainty of current conclusions.

For now, immersive VR is best understood not as a replacement for standard psychiatric care, but as a potentially valuable adjunct within carefully selected clinical contexts. Its long-term place in mental health treatment will depend on stronger trials, better implementation models, and clearer evidence that gains extend beyond short-term symptom improvement into durable functional benefit.

References

1. 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. https://doi.org/10.1111/acps.13777
2. Schröder, D., Wrona, K. J., Müller, F., Heinemann, S., Fischer, F., & Dockweiler, C. (2023). Impact of virtual reality applications in the treatment of anxiety disorders: A systematic review and meta-analysis of randomized-controlled trials. Journal of Behavior Therapy and Experimental Psychiatry, 81, 101893. https://doi.org/10.1016/j.jbtep.2023.101893
3. Thaysen-Petersen, D., Hammerum, S. K., Vissing, A.-C., Oestrich, I. H., Nordentoft, M., Düring, S. W., & Fink-Jensen, A. (2024). Virtual reality-assisted cognitive behavioral therapy for patients with alcohol use disorder: A randomized feasibility study. Frontiers in Psychiatry, 15, 1337898. https://doi.org/10.3389/fpsyt.2024.1337898
4. Jeppesen, U. N., Nordentoft, M., Jaya, E. S., Pile, V., Fagerlund, B., Mikkelsen, S. S., Glenthøj, L. B., Clemensen, L., Albert, N., & Glenthøj, B. Y. (2025). Virtual reality-based versus standard cognitive behavioral therapy for paranoia in schizophrenia spectrum disorders: A randomized controlled trial. Nature Medicine, 31, 3425–3439. https://doi.org/10.1038/s41591-025-03880-8
5. Bazhanov, A., & Meixner, G. (2026). Virtualizing alcohol use disorder therapy: A systematic scoping review. Advances in Therapy. Advance online publication. https://doi.org/10.1007/s12325-025-03449-4