A Systematic Review of Virtual Reality Therapies for Substance Use Disorders: Impact on Secondary Treatment Outcomes

Abstract

BACKGROUND AND OBJECTIVES:

Virtual Reality (VR) therapy may be an effective tool in treating urges and cravings in substance use disorder (SUD). Given the high co-occurrence of difficulties with mood, anxiety and emotional dysregulation with SUD, this review sought to examine the extant literature on the efficacy of VR for SUD in improving these secondary treatment outcomes.

METHODS:

A systematic literature review was conducted following PRISMA guidelines in PubMed, PsychInfo, and Embase. Studies were included if they utilized immersive VR, were conducted with individuals with substance use disorder/dependence/misuse, and included measures of mood, anxiety, depression, emotional dysregulation, or retention in treatment.

RESULTS:

Seven articles met our inclusion criteria. Five studies were conducted in patients using nicotine and utilized a cue-exposure intervention. VR was effective at reducing substance use and cravings in the majority of studies. Results on the efficacy of VR for improving mood, anxiety, and emotional regulation were mixed. Work examining retention in treatment was limited.

DISCUSSION AND CONCLUSIONS:

VR for SUD has the potential to improve mood and anxiety symptoms as well as retention in treatment for patients with SUD, particularly if the therapy targets these psychological symptoms. Future studies examining components of VR for SUD in individuals with drug use disorders, as well as examinations of targeted interventions for associated mood, emotional dysregulation, and retention in treatment, are necessary.

SCIENTIFIC SIGNIFICANCE:

This is the first systematic review of the impact of VR on mood, anxiety, and emotional dysregulation for individuals with substance use.

INTRODUCTION

Substance use disorders (SUDs) are highly prevalent and morbid conditions. In 2019, over 40 million people in the US reported having an SUD.¹ Broadly defined, SUDs are characterized by difficulty with self-control, risky substance use, significant time spent obtaining a substance, cravings and urges for the substance, and/or social impairments as a result of substance use.² Over 50% of individuals with SUD are reported to have a comorbid psychiatric disorder, with the most common being anxiety (41%) and mood (30%) disorders.³ In clinical populations, the prevalence of co-occurring psychopathology appears to be even higher. For instance, rates of comorbid depression are as high as 60% among individuals seeking treatment for SUD.⁴ Individuals with SUD are also likely to experience high levels of co-occurring transdiagnostic symptoms, including affective dysregulation, emotional lability, reactivity, irritability, and lack of emotional self-regulation,^5,6 also referred to as emotional dysregulation.^7,8 This co-occurring psychiatric symptomatology can produce significant impairment and barriers to treatment in individuals with SUD.

Data suggest that co-occurring anxiety, mood, and emotional dysregulation symptoms predict more frequent and severe substance misuse,^9,10 maintenance of a high-level substance misuse,¹¹ and difficulties in retention in treatment.^12–14 The latter is especially detrimental given recent evidence suggesting that engagement in SUD treatment substantially reduces mortality rates and increases patient utilization of vital pharmacological treatment.¹⁵ Not surprisingly, there is converging agreement that treatment of SUD should address associated comorbidities while also targeting improved retention in treatment.^16–18

The literature suggests that treating low mood, anxiety, and emotional dysregulation in SUD may result in better overall outcomes. For instance, Nunes and colleagues (2004) found that individuals with comorbid depressive disorder and SUD who were given antidepressant medications had significantly lower rates of substance use compared to individuals who received a placebo.¹⁹ Similarly, Fatséas and colleagues (2010) found that combining cognitive behavior therapy (CBT) with antidepressants was the most effective treatment for treating comorbid opioid and anxiety disorders.²⁰ Data from a 2013 meta-analysis suggests that adding psychosocial interventions targeting mood to standard smoking cessation interventions improved cessation rates across several trials.²¹ More recently, targeted training in improving emotional regulation has been shown to improve symptoms of craving in patients with substance dependence.²²

Digital technologies have increasingly been acknowledged as important tools for the treatment of SUD. One novel therapy that has emerged in this portfolio is virtual reality (VR), a computer-simulated environment that can create an immersive, interactive, three-dimensional experience. Many VR experiences also include haptic (touch) feedback as well as spatial audio, that adds to the level of immersion the user experiences. VR has been utilized across disciplines including gaming, flight simulation, and surgery, and has been shown to have useful applications in anxiety and pain management with both acute and longer-term effects.^23–25 The technology has also been harnessed in exposure-based interventions, including addressing fear of flying,²⁶ and posttraumatic stress disorder.²⁷ However, less is known about the role of VR in the realm of nicotine, substance use, and SUD.

Initial VR work demonstrates efficacy in successfully inducing substance cravings,²⁸ and more recent evidence suggests VR can be effectively used as a therapeutic tool for SUD.^29–31 In a recent review, Segawa and colleagues described that VR therapy can be effective in inducing and addressing SUD symptoms.³² Specifically, important signals emerged from the literature for the potential efficacy of VR in treating cravings for individuals with nicotine use disorders, although effects on individuals with alcohol use disorder appeared limited. No data were reported for VR and opioid use disorders or stimulant use disorders.³² While helpful, this review did not report on the impact of VR on co-occurring mood, anxiety, or emotional regulation, as well as retention in treatment in individuals with SUD. Given the importance of addressing co-occurring symptoms of psychopathology in SUD treatment, we completed a systematic qualitative review of the literature attending to the impact of VR treatments for SUD on co-occurring mood, emotional regulation, as well as retention in patients with SUDs.

Given that there is significant variability in what is considered true VR in the scientific field,³³ we chose to adhere to the strict definition of VR interventions as ones utilizing a head-mounted device, as such devices are regarded to be the most immersive. ³² This allowed us to maintain consistency with prior work, namely the recent review by Segawa and colleagues.³²

METHODS

This systematic review was conducted and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.³⁴

Eligibility Criteria

Studies were included if they utilized an immersive VR Device (defined per Segawa et al., 2020’s review as a Head-Mounted Display),³² if their study population included adolescents or adults with substance misuse, dependence, or disorder, and if the study outcomes measure included at least one of the following: mood, anxiety, depression, affect, emotional dysregulation, or retention in treatment. Studies were excluded if results reported only on the efficacy of an intervention to induce symptoms such as craving, without any treatment component. Studies using randomized controlled trial (RCT), controlled trial, and open label trial methodologies were included. Editorials, commentaries, chapters, opinion papers, reviews, study protocols, case reports, and other studies that used a non-experimental design or that did not report results were not included. Only English-language studies were included. Though Grey Literature (i.e., conference abstracts, dissertations) were not initially excluded from the search, no Grey Literature included enough information to successfully pass full-text screening criteria.

Search Strategy

Searches across PubMed (including MEDLINE), PsychInfo, and Embase were conducted for studies published on or before October 15^th, 2021. The searches were re-run in April of 2022 to ensure completeness and no additional relevant papers were identified.

Study Selection

Once study duplicates were removed, the titles and abstracts of all papers were screened by two independent reviewers. Discrepancies were resolved through discussion until consensus was reached. Once irrelevant titles were excluded, two investigators reviewed the full texts of all relevant articles. Articles for which full texts could not be obtained were excluded (N = 3). Records of all duplicate and irrelevant articles that were excluded were maintained.

Data Collection

Two reviewers extracted data from the full-text studies and all extracted data was reviewed by the senior reviewer. The following variables were extracted: ascertainment, study design, study duration, VR hardware used, VR intervention characteristics, SUD outcomes, mood/emotional regulation outcomes, and retention in treatment outcomes.

Risk of Bias Assessment

To assess potential sources of bias in the seven included studies, two investigators reviewed full texts of included papers using the Cochrane Collaboration tool for assessing risk of bias for randomized clinical trials³⁵ and the ROBINS-I for assessing risk of bias in non-randomized trials.³⁶ The Cochrane Collaboration tool examines six domains of bias including sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other issues. Similarly, the ROBINS-I examines seven domains of bias including confounding bias, selection bias, classification of intervention, deviations from intervention, missing data, measurement of outcomes, and reported result.

RESULTS

Our initial search yielded 1,220 studies. Once duplicates were removed, titles and abstracts of 797 studies were screened and 747 were excluded for irrelevancy. Of the remaining 50 studies selected for full text review, three were excluded due to an inability to find full text reports of these publications. In total, 40 publications were excluded following full-text review. The most common reason for exclusion was that the VR did not meet criteria for an immersive intervention (N=21). Articles were also excluded if study outcomes did not include at least one variable of interest to this review (mood, affect, emotional regulation, or retention in treatment) (N=13). Additional information on excluded reports can be found in Figure 1. Our final review consisted of seven articles.

Figure 1:

PRIMSA

We initially found a total of 1220 articles: 272 from PubMed, 652 from Embase, and 296 from PsychINFO. The results were combined to exclude 423 duplicate findings. 797 articles were screened according to the exclusion criteria. 747 titles seemed to be clearly irrelevant to our interest for our review. We then reviewed the remaining 50 articles in full text for exclusion. 3 studies were removed because we could not access the full text. We found 21 studies involving non-immersive VR interventions, 13 that did not measure secondary treatment outcomes, 3 that were not our study design of interest, 2 that did not measure SUD outcome, and 1 study that did not examine individuals with substance use. The remaining 7 articles were included in our review.

Population

A detailed description of the studies is shown in Table 2. Five studies (N=408) examined patients with nicotine/tobacco misuse, dependence, or use disorders,^38,40–43 with four studies including individuals who were daily smokers of at least 6⁴⁰ or at least ten cigarettes daily.^38,41,43 Given that daily smoking of as little as four cigarettes daily has been associated with tobacco use disorder⁴⁴ as well as extant literature defining light smoking as smoking fewer than 6 cigarettes daily,⁴⁵ our study utilizes the broad term of nicotine misuse to reference the behaviors of the daily smokers in the samples above. One study examined patients with alcohol use disorder,³⁹ and one study examined patients with opioid use disorder.³⁷ We found no studies meeting our criteria that examined secondary treatment outcomes of VR for individuals whose primary substances of use were cannabis, hallucinogens, methamphetamine, amphetamine, benzodiazepines, or MDMA.

Table 2.

Virtual Reality Therapies for Substance Use Disorders: Impact on Secondary Treatment Outcomes

Study DOI	Sample Size (N)	Participant Age and Sex Ratio	Study defined “Treatment as Usual”	Primary Substance Assessed	VR Intervention	VR Hardware	Study Duration	SUD Outcomes	Mood/Emotion Outcomes	Retention in Treatment	Comments
10.1089/cpb.2009.0118 Girard, 2009	N (total)= 91 N (active VR) = 46 N (sham VR) = 45	18-65 years M= 44 years (SD=11) 57% Female	No concurring treatment	Tobacco/Nicotine Use	Cue-Exposure 30-minute sessions	eMagin Z800 Head-Mounted Display	12 weeks with a 6-month follow-up	↓ Nicotine addiction (FTND) ↓ Smoking urges (QSU-B)		↑ Retention VR treatment	Participants crushed cigarettes (active VR) or balls (sham VR). Advantage of crushing cigarettes vs grasping balls significant from week 4-12. After week 12 abstinence rates significantly higher in VR vs controls. At 6 month follow up, abstinence significantly higher in VR condition
ISBN: 978-1-949-331-25-7 Malbos, 2018	N(total)= 61 N(initial)=61 N(completed)=37	M= 49 years (SD=13) 72% Female	Brief CBT Intervention	Tobacco/Nicotine Use	Cue-Exposure 45-minute sessions	Sensics® zSight Head Mounted Display	8 weeks	↓ Tobacco craving (TCQ-12; VAS-Craving) ↓ Cigarette dependence (CDS-12)	No significant difference in anxiety, depression or other outcomes in CBT or VR (STAI, BDI, SF-12, SES)		Unclear how many participants were in VR cue exposure vs traditional CBT VR = CBT overall
10.1016/j.jsat.2018.10.003 Pericot-Valverde, 2019	N(total)=102 N (CBT only)=52 N(CBT+CET)=50	18 years+ M= 40 years (SD=13) 64% Female in CBT only 72% Female in CBT+CET	Brief CBT Intervention	Tobacco/Nicotine Use	Cue-Exposure 20-minute sessions	Vuzix iWear	12 months	↓ Cigarette craving (VAS -Craving) No significant difference in Smoking Abstinence (point-prevalence + continuous abstinence) ↑ Relapse rates in VR group (% abstinent at end-of-treatment and relapsed within 12-month follow-up)		No significant difference in VR Tx Retention (% attended treatment, each follow-up, mean number of sessions)	Adjunct CET with CBT = CBT alone in retention and smoking cessation Relapse rates higher in CBT+CET group (64.3%) vs CBT group (37%). Craving only measured during the virtual exposure
ISSN: 1554-8716 Figueras-Puigderrajols, 2020	N(total)= 28 N(VR-CET)=12 N(TAU)=16	36-67 years M= 54 years (SD=8) Sex ratio not reported	Hospital Based SUD Treatment	Alcohol Use Disorder	Cue-Exposure 1-hour sessions	Oculus Rift	6 one-hour sessions	No significant difference in alcohol craving (MACS) No significant difference in attentional bias to alcohol (Alcohol Stroop Test)	No significant difference in anxiety (STAI)		Duration of study not specified Inpatient/Outpatient status not specified Small sample size may underestimate effect
10.1016/j.jsat.2021.108317 Zandonai, 2021	N(total)= 46	18-65 years Mean age not reported 59% Female	No concurring treatment	Tobacco/Nicotine Use	Cue-Exposure 30-minute sessions	Oculus Rift	3 days	↓ Cigarette craving daily (QSU-B) No significant difference between groups (QSU-B)	No significant difference in mood (POMS)		Participants were randomized to 3 groups: VR immersion in smoking environment after 15 seconds, immersed in smoking environment immediately, or immersed in neutral scenario for memory retrieval Did not specify how many participants in each group
10.1016/j.brat.2021.103858 Malchuska, 2021	N(total)=108 N(TAU+VR-ABM)=47 N(TAU+VR control)=49 N(excluded)=25	18 years+ M= 50 years (SD=12) 49% Female	Brief behavioral counseling for smoking cessation (psychoeducation and motivational interviewing)	Tobacco/Nicotine Use	Approach-Avoidance 15-minute sessions	HTC Vive	7 weeks	↓ Number of cigarettes smoked daily No significant effect on: Degree of nicotine dependence (FTND) Attitude toward smoking Alcohol use (AUDIT)	No significant difference on: -Positive mental health -Health behavior checklist (Sports Activities and Eating Habits)		VR-ABM did not modify approach, attentional or association biases. VR-ABM reduced number of daily smoked cigarettes but only during administration No effect of training on other smoking- or health-related outcomes (improvements found in both groups) VR-ABM may have at follow-up.
10.1093/pm/pnab162 Faraj, 2021	N(total)=15	18 years+ Mean age not reported 73% Female	Methadone Maintenance Therapy	Opioid Use Disorder	Mindfulness 30-minute sessions	Windows Mixed Reality Headset	12 weeks	↓ Opioid craving ↓ Pain (VAS)	↓ Anxiety ↓ Depression No significant difference on anger (VAS) ↓ Salivary Cortisol		Decreases in craving rebounded mid-treatment fMRI showed decreases in pain task related activation in the pain neuromatrix

KEY: Fagerstrom Test for Nicotine Dependence (FTND); Questionnaire of Smoking Urges-Brief (QSU-Brief); Tobacco Craving Questionnaire (TCQ-12); Visual Analog Scale- Craving (VAS); Cigarette Dependence Scale (CDS-12); State Trait Anxiety Inventory (STAI); Beck’s Depression Inventory (BDI); Quality of Life Scale (SF-12); Rosenberg Self-Esteem Scale (SES); Cognitive Behavioral Therapy (CBT); Cue-Exposure Treatment (CET); Treatment as Usual (TAU); Treatment (Tx); Multidimensional Alcohol Craving Scale (MACS); Profile of Mood States (POMS); Alcohol Use Disorders Identification Test (AUDIT); Virtual Reality- Approach Bias Modification (VR-ABM)

Ascertainment and Design

Four of our seven studies took place in academic research laboratories and recruited individuals from the general population.^38,40,41,43 Two studies recruited patients from SUD treatment clinics,^37,39one study did not report the origin of ascertainment.⁴² No studies took place in an inpatient setting. Five studies took place in Europe,^39–43 one study took place in Canada,³⁸ and one study took place in the United States.³⁷ One study was a single-arm intervention³⁷ and the remaining six studies were RCTs.

Among the six RCTs, two randomized participants to either treatment VR or a control VR,^38,40 one randomized participants to treatment VR or CBT alone,⁴¹ one randomized participants to VR or TAU at an SUD focused hospital-based clinic,³⁹ one randomized participants to VR and CBT or CBT alone,⁴² and one randomized participants to receiving a smoking or neutral VR.⁴³

In the single arm intervention, all patients received outpatient Methadone Maintenance Therapy in addition to treatment VR.³⁷

Hardware

In keeping with the criteria set by Segawa and colleagues,³² all included studies utilized an immersive VR administered through a head mounted device. Two studies used the Oculus Rift headset,^39,43 one study used the eMagin headset,³⁸ one study used HTC Vive,⁴⁰ one study used the Vuzix iWear,⁴¹ one study used the Sensics zSight,⁴² and one study used a Windows Mixed Reality Headset.³⁷

Intervention

Five studies utilized a cue-exposure intervention,^{38,39,41–43} in which participants interacted with environments designed to induce substance craving. In four of these five studies,^39,41–43 participants were immersed in scenes related to tobacco consumption (i.e., at a bar or party) and encouraged to practice smoking abstinence. Notably, the Pericot-Valverde (2019) study paired this cue-exposure therapy with CBT.⁴¹ The remaining study that utilized a cue exposure intervention³⁸ had participants grasp virtual balls and crush virtual cigarettes. One study tested a mindfulness intervention³⁷ and utilized a combination of guided movement and breathing techniques. Finally, one study used an approach-avoidance intervention⁴⁰ where participants were presented with smoking-related and non-smoking-related cues and instructed to grasp all non-smoking related objects and avoid all smoking related objects.

SUD Outcomes

Substance cravings and urges were the most common primary SUD outcome and were diminished with VR in the majority of studies.^37,41–43

Among five studies targeting nicotine misuse, four reported decreases in urges or cravings in patients in VR conditions from baseline^37,41–43 while one reported a decrease in number of cigarettes smoked daily in treatment VR when compared to a control VR.⁴⁰

In addition to reduction in cravings, several studies reported other significant nicotine related outcomes. Girard and colleagues reported that patients who received treatment VR reported significantly higher cigarette abstinence rates at six month follow up compared to those in the control VR group.³⁸ Machulska and colleagues reported that over the course of the two-week VR administration period, individuals receiving treatment VR reported a greater decrease in daily cigarettes smoked than those receiving control VR, but these effects did not last at follow up.⁴⁰

Notably, nicotine outcomes were not uniform across studies. Although Malbos and colleagues reported a significant reduction in craving and nicotine dependence in individuals receiving treatment VR, individuals in this condition did not differ significantly from those receiving CBT alone in these outcomes.⁴² Additionally, while Pericot-Valverde reported a significant decrease in craving in individuals receiving a VR add on to concurrent CBT, the researchers reported no effect of this intervention on smoking abstinence at follow up. Interestingly, this study also reported that subjects in the VR condition experienced significantly higher rates of smoking relapse over 12 month follow up compared to those receiving CBT alone.⁴¹

Only two out of seven studies examined non-nicotine substance misuse.³⁷ One study in patients with alcohol use disorder found no significant difference in craving between individuals who received VR and those who received only outpatient alcohol use disorder treatment (TAU).³⁹ One study in patients with opioid use disorder reported a significant decrease in opioid craving following VR treatment intervention, with post-session craving levels being significantly lower than pre-session cravings.³⁷

Mood and Emotional Regulation Outcomes

Results for impact of VR intervention on mood outcomes were mixed. Only one study³⁷ found that VR intervention significantly decreased post intervention self-reported anxiety and depression scores in patients with opioid use disorder. In contrast, three studies in patients with nicotine misuse^40,42,43 and one study in patients with alcohol use disorder ³⁹ found no significant effect of VR intervention on mood symptoms. More specifically, results demonstrated no significant differences following VR intervention in anxiety,^39,42 depression,⁴² overall wellbeing,⁴⁰ or mood state.⁴³ The two remaining studies examined patients with nicotine misuse and^38,41 did not measure mood outcomes. No studies found that VR intervention increased anxiety or depression. Additionally, no studies specifically reported on emotional regulation, mood reactivity, and/or mood dysregulation outcomes.

Retention in Treatment Outcomes

Study treatment and follow up ranged between 12 weeks³⁸ and 12 months.⁴¹ While several studies included a concurring therapeutic component,^37,39–42 only two studies specifically examined retention as an outcome.^38,41 Whereas Girard (2009) found that the VR intervention increased retention in the 12-week psychosocial minimum-support treatment program by about 2 weeks,³⁸ Pericot-Valverde (2019) reported no differences in retention in study treatment activities between individuals who received the VR intervention combined with CBT and individuals who only received CBT.⁴¹ Additional information on study outcomes and variables of interest can be found in Table 2.

Risk of Bias Assessment Outcomes

Overall, we determined that most randomized controlled trials were at low risk of bias for random sequence generation and attrition bias, unclear risk of bias for allocation concealment and high risk of bias for performance and detection bias (Figure 2). Similarly, we found that the one non-randomized controlled trial included in this review³⁷ had overall moderate risk, particularly regarding classification of intervention and measurement of outcomes (Figure 3).

Figure 2.

Risk of Bias Assessment Results for Randomized Controlled Trials (RCTs)

Figure 3.

Risk of Bias Assessment Results for Non-Randomized Controlled Trials (Non-RCTs)

DISCUSSION

The aim of this qualitative systematic literature review was to examine the extant literature on the efficacy of VR for treating SUD in addition to addressing co-occurring symptoms of mood, anxiety, emotional dysregulation, as well as retention in treatment. Overall, we found that VR had a positive impact on urges, cravings, and substance use. Mixed findings emerged when examining other outcomes including anxiety, mood, and retention in treatment. Furthermore, our review identified significant gaps in the literature, specifically regarding efficacy of immersive VR interventions for SUD in addressing emotional regulation as well as in the use of VR in inpatient settings.

The results of this review suggest an overall positive impact of VR on SUD related outcomes. Of the seven studies that met our search criteria, five reported that VR interventions effectively addressed SUD symptoms in participants. Specifically, our results suggest that VR has the potential to reduce cravings for nicotine and tobacco and reduce the number of cigarettes smoked daily. ^37,40–43Among the five studies targeting nicotine and tobacco misuse, all studies demonstrated efficacy of a VR intervention in addressing some aspect of nicotine craving or smoking.^38,40–43 Importantly, only two ^38,40 of these studies reported a significant difference in these outcomes between VR and control conditions. Additionally, the single study sampling individuals with opioid use disorders also demonstrated efficacy of a mindfulness VR in reducing cravings in participants post intervention.³⁷ Interestingly, in the sole study examining patients with alcohol use disorders, VR was not effective in improving cravings for alcohol, though the study did not measure actual alcohol use.³⁹

Our mixed findings on the efficacy of VR in improving co-occurring mood and anxiety symptoms suggest that simply participating in VR treatment for SUD may not inherently improve these symptoms. For instance, the only study showing improved mood and anxiety, Faraj et al (2021) utilized a mindfulness meditation-based intervention that, unlike other studies in this review, did not require participants to interact with substance-related triggers, and focused instead on regulating one’s sensations and feelings.³⁷ It may be that cue-exposure therapy alone is insufficient in improving both mood and SUD related outcomes longer-term. In support of this notion, Pericot-Valverde and colleagues (2019) noted that relapse rates were significantly higher in individuals who engaged in cue-exposure therapy combined with CBT compared to CBT alone, despite both interventions significantly reducing cravings in participants immediately after completion of trials.⁴¹ The authors speculated that addressing cravings alone may be inadequate to address SUD outcomes in the longer-term, and effective therapies should target secondary treatment outcomes such as mood and stress/stress tolerance. These speculations are consistent with findings from Machulska and colleagues, who reported that although VR treatment significantly decreased daily smoked cigarettes during the two-week intervention period, this effect was no longer present at the seven week follow up.⁴⁰ However, given that Faraj and colleagues study lacked a control group,³⁷ it is difficult to ascertain whether the therapeutic aspect of the VR is truly responsible for the improvements in mood, and in turn, decrease in opioid craving.

It may also be that VR preferentially improves mood symptoms in individuals with opioid use disorder compared to individuals with alcohol or nicotine use disorders. Alternatively, it could be that findings in studies targeting other substances were limited by small sample sizes. Specifically, the one study targeting alcohol use disorders (N=28) may have been underpowered. However, it is noteworthy that two out of three studies targeting nicotine that reported no effect of VR intervention on mood did so despite relatively large sample sizes of 108 and 61, respectively.^40,42

Given that Faraj and colleagues’ mindfulness intervention appeared to improve both craving and mood symptoms in participants,³⁷ it may be that VR with mindfulness interventions may adequately address both mood and anxiety as well as urges and cravings without a VR substance exposure component. It is noteworthy that VR did not increase cravings, anxiety, or depression in any of these studies.

Despite the importance of retention in SUD treatment,⁴⁶ we found a limited and mixed literature with VR. One study³⁸ found that participants receiving therapeutic VR for SUD demonstrated improved retention in a 12-week psychosocial treatment program when compared to those receiving control VR,. However, another study⁴¹ found no difference in retention between patients receiving VR plus CBT and those receiving CBT alone. It is possible that the latter may have been due to a potential ceiling effect, given relatively high retention rates in the CBT only group (73%), compared to the CBT+VR group (78%). While no study reported serious adverse events, one study reported that participants in the treatment VR condition experienced more severe cybersickness than those in the control VR group.³⁸ Interestingly, Girard and colleagues also reported improved retention in the treatment VR group compared to those receiving control VR.³⁸ Clearly, more studies examining the role of VR on facilitating retention in treatment are needed.

There are a number of substantial limitations in the current review. Given that many of these studies were pilot studies with relatively small sample sizes, there is likely an underestimate of the effect (Figueras-Puigderrajols, 2020, N=28; Faraj, 2021, N=15). Additionally, our review excluded studies that did not use a head-mounted device for immersive VR, as defined by Segawa and colleagues (2020).³² Though our inclusion criteria are in line with best practice recommendations for VR clinical trials,⁴⁷ more standardization is needed to ensure further consistency in interpreting the VR literature. Likewise, although two studies^39,42 both utilized the State-Trait Anxiety Inventory (STAI) to measure anxiety, all other studies included in our review used different measures for mood, anxiety, emotion regulation, and retention, which may limit cross-study comparability. In addition, we excluded studies that did not examine secondary treatment outcomes including mood, anxiety, emotional regulation, and retention. These studies may present more information about the efficacy of VR interventions in addressing SUD outcomes across a variety of substances. Relatedly, and partially due to our stringent inclusion criteria, we did not identify studies examining the use of VR on our target outcomes for cannabis, hallucinogens, methamphetamine, amphetamine, or benzodiazepine misuse, dependence, or use disorders. Regarding demographics, only one³⁷ of the seven studies included in our review reported on race and ethnicity. Likewise, while most individuals in treatment for substance use disorders are male,⁴⁸ the majority of participants included in these studies were female. As a result, these findings may not have been generalizable to the broader population. Similarly, our review lacks studies examining the effect of VR on emotional regulation or the administration of VR in inpatient SUD treatment settings.

Despite these limitations, we found evidence for the efficacy of immersive VR to address certain SUD related outcomes beyond urges and cravings. A limited literature also supports that mindfulness VR improves symptoms co-occurring symptoms of mood and anxiety in patients with SUD. Given the potential utility of therapeutic VR, future studies examining the efficacy of targeted VR therapies on symptoms of mood, anxiety, and emotional dysregulation, as well studies sampling patients with a more diverse array of SUD in a range of treatment settings, are necessary.

Table 1.

PRIMSA Search Terms

Embase	(‘Virtual Reality’:ti,ab,kw OR ‘Virtual Reality’/exp OR ‘Virtual Reality Exposure Therapy’/exp) AND (‘Drug Dependence’/exp OR ‘Substance Use Disorder*’:ti,ab,kw OR ‘SUD’:ti,ab,kw OR ‘drug use disorder*’:ti,ab,kw OR ‘addictive disorder*’:ti,ab,kw OR ‘addict*’:ti,ab,kw OR ‘substance use*’:ti,ab,kw OR ‘cocaine’:ti,ab,kw OR ‘cannabis’:ti,ab,kw OR ‘mdma’:ti,ab,kw OR ‘inhalant*’:ti,ab,kw OR ‘opioid*’:ti,ab,kw OR ‘methamphetamine*’:ti,ab,kw OR ‘amphetamine*’:ti,ab,kw OR ‘stimulant*’:ti,ab,kw OR ‘sedative*’:ti,ab,kw OR ‘hypnotic*’:ti,ab,kw OR ‘benzodiazepine*’:ti,ab,kw OR ‘barbiturate*’:ti,ab,kw OR ‘hallucinogen*’:ti,ab,kw OR ‘heroin’:ti,ab,kw OR ‘opiate*’:ti,ab,kw OR ‘fentanyl’:ti,ab,kw OR ‘drug addict*’:ti,ab,kw OR ‘drug dependency’:ti,ab,kw OR ‘marijuana’:ti,ab,kw OR ‘drug abuse*’:ti,ab,kw OR ‘polysubstance*’:ti,ab,kw OR ‘illicit drug*’:ti,ab,kw OR ‘alcohol’:ti,ab,kw OR ‘nicotine’:ti,ab,kw OR ‘tobacco’:ti,ab,kw OR ‘cigarette*’:ti,ab,kw OR ‘smoking’:ti,ab,kw)
Pubmed	(“Virtual Reality”[MeSH] OR virtual reality[tiab] OR “Virtual Reality Exposure Therapy”[Mesh]) AND (“Substance-Related Disorders”[Mesh] OR SUD[tiab] OR drug use disorder*[tiab] OR addictive disorder*[tiab] OR addiction[tiab] substance use[tiab] OR cocaine[tiab] OR cannabis[tiab] OR mdma[tiab] OR inhalant*[tiab] OR opioid*[tiab] OR methamphetamine*[tiab] OR amphetamine*[tiab] OR stimulant*[tiab] OR sedative*[tiab] OR hypnotic*[tiab] OR benzodiazepine*[tiab] OR barbiturate*[tiab] OR hallucinogen*[tiab] OR heroin[tiab] OR opiate*[tiab] OR fentanyl[tiab] OR drug addict*[tiab] OR drug dependency[tiab] OR marijuana[tiab] OR drug abuse*[tiab] OR polysubstance*[tiab] OR illicit drug*[tiab] OR alcohol[tiab] OR nicotine[tiab] OR tobacco [tiab] OR cigarette*[tiab] OR smoking [tiab])
PsychINFO	(exp Virtual Reality/ or virtual reality.ti,ab. or exp Virtual Reality Exposure Therapy/) and (exp “Substance Use Disorder”/ or SUD.ti,ab. or “drug use disorder”.ti,ab. or addictive disorder*.ti,ab. or addiction.ti,ab. or “substance use”.ti,ab. or cocaine.ti,ab. or cannabis.ti,ab. or mdma.ti,ab. or inhalant*.ti,ab. or opioid*.ti,ab. or methamphetamine*.ti,ab. or amphetamine*.ti,ab. or stimulant*.ti,ab. or sedative*.ti,ab. or hypnotic*.ti,ab. or benzodiazepine*.ti,ab. or barbiturate*.ti,ab. or hallucinogen*.ti,ab. or heroin.ti,ab. or opiate*.ti,ab. or fentanyl.ti,ab. or drug addict*.ti,ab. or drug dependency.ti,ab. or marijuana.ti,ab. or drug abuse*.ti,ab. or polysubstance*.ti,ab. or illicit drug*.ti,ab. or alcohol.ti,ab. or nicotine.ti,ab. or tobacco.ti,ab. or cigarette*.ti,ab. or smoking.ti,ab.)