Evaluating the use of virtual reality for weight management: A systematic review and meta‐analysis

Abstract

Aims

This study aims to clarify the effect of VR as an additional method to common treatments for weight management in adults using a systematic review and meta‐analytic design.

Materials and Methods

A comprehensive literature search was performed through Medline, Embase, CENTRAL, PsycINFO, CINAHL, and Web of Science Core Collection to identify the eligible studies that investigated the effect of VR‐enhanced treatment (i.e., VR to deliver conventional treatment) on weight management among adults seeking weight loss compared to conventional treatments alone. The included studies were assessed by the INSPECT‐SR tool for trustworthiness and authenticity. Body weight and BMI data at baseline, post‐intervention, and at follow‐up (maintenance phase) were collected to estimate the effect size.

Results

Eleven studies met the criteria for inclusion in the systematic review, of which 10 studies were included in the meta‐analysis. The pooled effect size indicated that VR‐enhanced treatment was equally effective as conventional treatment in managing body weight and BMI during the intervention. More interestingly, the results revealed that VR‐enhanced treatment was more effective on body weight management during the maintenance phase compared to conventional treatments (MD: −5.16 kg; 95% CI: −8.46, −1.85; I ²  = 0%).

Conclusion

The findings suggest that VR‐enhanced treatment may offer a more effective approach for maintaining weight loss than conventional methods; however, these results should be interpreted with caution, as they are derived from a small number of studies, some of which raise serious concerns regarding conduct, governance, and transparency, highlighting the need for rigorous clinical trials to reach a definitive conclusion.

1. INTRODUCTION

Obesity has become one of the most important and pressing challenges of the modern world. ¹ A large number of people worldwide struggle with excess weight, and the prevalence of obesity is increasing. ² Excess body fat is directly associated with many chronic diseases, such as diabetes, ³ cardiovascular disease, ⁴ high blood pressure, ⁵ and non‐alcoholic fatty liver disease. ⁶ The severity of consequences associated with obesity makes it evident that it should be treated as a disease, and its harm to human health is much more significant than historically anticipated. ⁷ Although obesity is considered an accumulation of body fat resulting from excessive caloric intake, it is well‐recognised that the aetiology is multi‐factorial. ⁸ For example, psychological and behavioural factors (including disorders) have been suggested as undeniable key factors in the development and progression of obesity. ⁹ Unfortunately, approaches such as diet, enhancing physical activity, and using medical/pharmaceutical intervention alone have not been successful in many cases; in part related to challenges faced by patients in following treatment regimens, resulting in only short‐term success. ¹⁰ , ¹¹ As a result, finding a way to encourage people to adhere to treatment with multiple therapeutic approaches could be crucial in improving success in maintenance.

With the recent advancements in immersive technology, multiple new methods have been suggested with the promise to be applied in the clinical approach. ¹² , ¹³ The beneficial effects of applying immersive technology in post‐stroke rehabilitation, ¹⁴ improving phantom limb pain, ¹⁵ and teaching social skills to individuals with autism spectrum disorder ¹⁶ are among the several well‐documented methods. Moreover, scientists have investigated combining immersive technology with other methods, such as gamification and personal avatars, to improve eating habits by creating highly interactive environments. ¹⁷ , ¹⁸ , ¹⁹ Among various types of immersive technology, virtual reality (VR)—a highly interactive, computer‐based environment that immerses the user in a digital setting resembling the real world ²⁰ —provides a high level of immersion and might be a suitable approach for improving eating behaviours. ²¹ By creating a customizable environment tailored to the treatment setting, VR offers a feasible method for delivering evidence‐based weight management approaches, as supported by previous studies indicating its potential role in the future treatment of obesity. ²² , ²³ This feature has led clinical trials to leverage VR as a medium for delivering cognitive behavioural therapy, ²⁴ relaxation therapy, ²⁵ cue exposure therapy, ²⁶ and standard weight management strategies, ²⁷ , ²⁸ including promoting physical activity and healthy dietary habits.

Despite the considerable potential for VR in weight management and early promising findings, results are mixed, and therefore, the research base is largely inconclusive. Previous systematic reviews were performed to address this and reported some positive findings ²⁹ , ³⁰ ; yet, not all aspects have been explored, including the long‐term effects of VR‐enhanced treatments and a critical appraisal of study quality to determine the reliability and strength of the evidence. To clarify the current state of the literature, we conducted a systematic review and meta‐analysis to formally summarise the existing data and specifically assess the effect of VR‐enhanced treatment, as a way to deliver evidence‐based approaches, compared to conventional therapies, on weight management in the adult population.

2. METHODS

This meta‐analysis was performed in accordance with the Cochrane Handbook for Systematic Reviews of Interventions and is reported based on the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) guidelines. ³¹ The protocol was registered at PROSPERO (ID: CRD42023482744).

2.1. Search strategy

A comprehensive literature search was carried out across databases, including Medline, Embase, Cochrane Central Register of Controlled Trials (Central), PsycINFO, CINAHL, and Web of Science Core Collection from inception to 21 November 2024. Keywords including virtual reality and head‐mounted display, as well as obesity, weight loss, and body mass index, were used in combination with MeSH terms and specific modifiers. The number of articles retrieved from each database in the initial search and the keywords used are presented in Supplemental Tables 1 and 2, respectively. Furthermore, the references of relevant reviews and selected studies were hand‐searched to explore potential studies that might have been missed in the electronic search. All clinical trials that investigated the effect of a VR‐enhanced approach on weight management compared to conventional treatments among adults seeking weight loss were included. The studies were excluded if they reported the results as qualitative data; if the purpose of using VR was only to persuade participants to increase physical activity; if VR was only used to manage body image disturbances and increase generalised self‐efficacy; or if the use of VR was exclusively for relaxation. The PICOS criteria table (Population, Intervention, Comparison, Outcomes, and Study Design) is available in Supplemental Table 3.

2.2. Study selection and data extraction

Two authors (S.P. and M.P.) performed the study selection independently by using Covidence software. After excluding duplicate citations, articles were checked based on their titles/abstracts, and irrelevant ones were excluded. The remaining studies were selected for further assessment via full‐text in case their eligibility could not be determined in the first step of screening. After completion of the selection process, the same authors collected primary information from included references, and the following set of data was extracted: articles' information (first author's last name, years of publication, and country), participants' information (age, BMI, sex, and health condition), method of studies (sample size, design, and duration of study), and summary of intervention and control group. Any discrepancy regarding study selection and data abstraction was resolved by discussion.

2.3. Investigating problematic clinical trials in systematic reviews

The Investigating Problematic Clinical Trials in Systematic Reviews (INSPECT‐SR) tool is a new approach to evaluate the trustworthiness and authenticity of included trials. ³² This tool consists of a series of questions across 4 domains: ‘Inspecting post‐publication notices’, ‘Inspecting conduct, governance and transparency’, ‘Inspecting text and figures’, and ‘Inspecting results in the study’. The questions aim to identify any concerns and are answered with ‘Yes’, ‘No’, ‘Unclear’, and ‘Not applicable’, based on the information provided in the included studies. The decision for each domain is based on the individual response to the corresponding question. While the overall assessment will be classified as ‘some concern’ if only one domain is rated as ‘some concern’, more than one domain with ‘some concern’ or any domain classified as ‘serious concern’ leads to ‘serious concern’.

2.4. Risk of bias assessment

Risk of bias assessment was conducted by two independent investigators (S.P. and M.P.) using the Cochrane Collaboration Risk of Bias assessment tool, ³³ which evaluates the trials according to the existence of selection bias, performance bias, attrition bias, reporting bias, as well as other possible sources of bias. The results are presented as ‘high’, ‘low’, or ‘unclear’ risk of bias. The potential disagreements were resolved by discussion.

2.5. Statistical analysis

All analyses were conducted with RevMan (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration) and Stata software (STATA Corp, College Station, TX, USA). The mean difference and standard deviation (SD) of change were calculated to obtain the net effect size using suggested formulas, where they were not directly provided by selected studies. ³³ Since some of the included trials did not provide sufficient information to calculate r correlation, we assumed it to be equal to 0.6 and performed a sensitivity analysis for r = 0.8 and 0.2 to explore whether the pooled effect size can be significantly influenced by a change in the r‐value. In addition, the follow‐up assessment of the mentioned outcomes of interest was also collected to examine the long‐term effect of VR‐enhanced treatment. To do so, the baseline and the post‐follow‐up assessments were collected. Only the last measurements were taken into account in the studies that conducted multiple time assessments during follow‐up.

Regarding the assessment of overall effect size, a fixed‐model or random‐effect test was applied based on the between‐study heterogeneity, which was categorised as low, moderate, and high when estimated at 0%–30%, 30%–60%, and >60%, according to Cochrane systematic review and meta‐analysis guidelines. ³⁴ In addition to r correlation, sensitivity analysis using a leave‐one‐out‐at‐a‐time approach was also applied to assess whether the overall result was influenced by a single effect size. Subgroup analyses were performed to explore whether certain study or participant characteristics can modify the effect of the intervention. A meta‐regression test was also conducted to determine the possible relationship between changes in outcomes and the duration of intervention, as well as the baseline BMI. Publication bias was evaluated by visual inspection of the funnel plot as well as two formal tests, including the Begg adjusted rank correlation ³⁵ and Egger's regression asymmetry test. ³⁶ p‐value <0.05 was considered the level of significance.

3. RESULTS

The study selection process and the explanation regarding what basis the studies were excluded in each step are illustrated in Figure 1. In brief, 2906 citations remained after removing duplicated references (n = 1638), among which 2858 publications were irrelevant and excluded by title/abstract screening. Forty‐eight studies retrieved from the first step assessment were checked by full text, through which 37 studies were removed based on inclusion/exclusion criteria: case study (n = 1), not reporting the outcomes of interest (n = 23), duplicate data (n = 2), only used for relaxation (n = 1), protocol studies (n = 3), and non‐VR enhanced treatment intervention (n = 7). Finally, 11 trials met the inclusion criteria and were selected for the present study. One study ³⁷ did not provide enough information on the outcomes of interest and was only reported descriptively. The remaining 10 clinical trials were included in the meta‐analysis. Among these, the Cesa et al. ²⁴ and Manzoni et al. ³⁸ compared the VR‐enhanced treatment with Cognitive‐Behavioural therapy and an inpatient programme; each study was treated as two separate arms. Riva et al. ³⁹ followed a similar approach by comparing VR‐enhanced treatment with two groups, and additionally reported outcomes separately for two subgroups: patients with obesity and those with eating disorders. As a result, this study was included in the pooled effect size as four distinct arms.

FIGURE 1.

The flow diagram of the process used in the selection of the clinical trial studies included in this systematic review and meta‐analysis.

3.1. Study characteristics

The characteristics of the included studies are outlined in Table 1. Based on the information presented in the table, most of the studies were conducted in Italy, ²⁴ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ and the remainder were performed in the United States, ²⁷ , ²⁸ , ⁴¹ , ⁴² and Mexico. ⁴³ Only one trial was conducted in Spain and Italy. ²⁶ Seven studies recruited patients with overweight and/ or obesity, ²⁷ , ²⁸ , ³⁷ , ³⁸ , ⁴¹ , ⁴² , ⁴³ whilst the rest enrolled participants with excess body weight, along with having an eating disorder, ²⁴ , ²⁶ , ³⁹ , ⁴⁰ The mean age of participants ranged from 28 to 59 years, and the BMI was between approximately 27 and 44 kg/m². The duration of the intervention and follow‐up ranged from 3 to 28 weeks and from 3 to 12 weeks, respectively.

TABLE 1.

Main information extraction of eligible studies.

First authors (years of publication)	Sample size/design	Sex/Mean age	Baseline BMI (kg/m2)/ Duration of treatment/follow‐up	Health condition	Intervention	Comparison
Cesa et al. (2013) Italy	ECT: 27 CBT: 20 IP: 19/RCT	Female/ECT: 32.9 ± 8.8 CBT: 29.9 ± 7.95 IP: 32.2 ± 6.36	ECT: 39.2 ± 5.3 CBT: 41.1 ± 3.3 IP: 41.8 ± 6.3/6 weeks/12 months follow‐up	Adults with obesity and binge eating disorder	15 VR‐ECT sessions plus basic inpatient treatment	(A) 15 CBT sessions plus basic inpatient treatment. CBT was based on stage 1 treatment for binge eating disorder. (B) IP
Johnston et al. (2012) United States	VR: 33 Face to face: 21/CT	Both genders/VR: 46.3 ± 9.6 Face to face: 37.5 ± 10.6	VR: 33.13 ± 6.13 Face to face: 30.21 ± 5.6/12 weeks/not followed	Adults with an overweight condition	VR‐world (second Life) weight loss programme	Face‐to‐face programme
Manzoni et al. (2016) Italy	VR: 56 CBT: 52 IP: 50/RCT	Female/overall: 35.63 ± 8.04	Overall: 42.24 ± 6.01/6 weeks 12‐month follow‐up	Adults with obesity	15 VR‐ECT sessions plus a standard behavioural programme	(A) CBT plus Standard behavioural programme (B) IP
Phelan et al. (2021) United States	VR: 7 Standard programme: 6/RCT	Both genders/VR: 50.1 ± 11.3 Standard programme: 35.6 ± 14.8	VR: 30.4 ± 4 Control: 36.5 ± 4.9/4 weeks/not followed	Adults with obesity or overweight	Weight‐related VR app plus standard behavioural weight loss	Standard behavioural weight loss plus a non‐weight‐related VR app
Riva et al. (2004) Italy	VR: 30 CBT: 30 IP: 30/RCT	Female/VR: 33.43 ± 8.29 CBT: 32.2 ± 7.95 IP: 33.5 ± 8.17	VR: 39.56 ± 6.20 CBT: 41.14 ± 5.70 IP: 39.95 ± 6.40/28 weeks/not followed	Adults with obesity and eating disorders	10 VR‐ECT sessions plus nutritional and physical treatment	(A) 15 CBT plus Standard behavioural programme (B) IP
Riva et al. (2001) Italy	VR: 14 CBT: 14/RCT	Female/VR: 32.14 ± 8.73 IP: 30.29 ± 12.62	VR: 43.50 ± 5.97 IP: 44.44 ± 10.07/6.5 weeks/not followed	Adults with obesity	7 VR‐ seven sessions of VREDIM plus nutritional and physical treatment	IP
Sullivan et al. (2013) United States	Second life: 10 Face‐to‐Face: 10/RCT	Both genders/second life: 30.3 ± 3.1 Face‐to‐Face: 31.9 ± 4.1	Second life: 34.2 ± 5.7 Face‐to‐Face: 31.5 ± 4.1/12 weeks/9 months follow‐up	Adults with obesity or overweight	VR‐based treatment (Second Life) for both weight loss and weight maintenance	Face‐to‐face, including weight loss and physical activity programmes, followed by the Second Life for weight maintenance
Thomas et al. (2020) United States	Web‐based VR: 72 Weight Watchers: 74/RCT	Both genders/web‐based VR: 57.0 ± 11.2 Weight Watchers: 59.6 ± 9.2	Web‐based VR: 33.0 ± 4.6 Weight Watchers: 33.2 ± 5.1/24 weeks/not followed	Adults with obesity or overweight	Web‐based VR sessions for training in behavioural weight loss, plus Weight Watchers' online platform	Weight Watchers' online platform of a weight management programme containing basic diet and physical activity education
Sansoni et al. (2024) Italy	VR: 12 IP: 12/RCT	Female/VR: 29.25 ± 7.75 IP: 28.83 ± 6.48	VR: 28.52 ± 7.69 IP: 33.68 ± 7.38/5 weeks 12 months	Adults with obesity and bulimia nervosa	10 biweekly VR‐ECT sessions plus 5 weekly CBT sessions	IP
Ferrer‐Garcia et al. (2019) (Spain and Italy)	VR: 31 CBT: 24/Multicenter‐RCT	Both genders/VR: 34.64 ± 10.16 CBT: 34.37 ± 9.55	VR: 27.27 ± 4.88 CBT: 28.69 ± 7.15/3 weeks/6 months follow‐up	Adults with bulimia nervosa and binge eating disorder	6 VR‐based cue exposure therapy sessions	CBT
Cárdenas‐López et al. (2015) Mexico	Overall participants: VR:8 CBT:8/RCT	Both gender/not reported	VR: 33.33 ± 3.86 CBT: 32.88 ± 4.75/12 weeks/not followed	Adults with obesity	12 VR‐ECT sessions	CBT

Abbreviations: CBT, cognitive‐behavioural therapy; CT, controlled trial; ECT, experiential cognitive therapy; IP, Integrated multimodal medically managed inpatient programme including medical, nutritional, physical, and psychological care; VR, virtual reality; RCT, randomized controlled trial.

The treatments delivered via VR varied across studies. Riva et al. ³⁷ applied VREDIM's 3D Healing Experiences™, which delivered cognitive‐behavioural techniques such as assessing stimuli that could elicit abnormal eating behaviours, evaluating anxiety symptoms related to food exposure, and addressing participants' body image concerns. Despite some differences, a generally similar VR‐enhanced cognitive‐behavioural therapy protocol was used in studies by Cesa et al., ²⁴ Riva et al., ³⁹ Manzoni et al., ³⁸ Sansoni et al., ⁴⁰ and Cárdenas‐López et al. ⁴³ These interventions provided immersive situations where participants practiced behavioural eating management, decision‐making, and problem‐solving skills, combined with inpatient programmes offering medical, nutritional, and physical treatments. A behaviour‐based approach was also adopted by Ferrer‐García et al., ²⁶ where VR was used to deliver cue exposure therapy through food‐related scenarios to reduce cravings and anxiety. Phelan et al. ⁴² implemented behaviour‐focused VR simulations, including a restaurant scenario that encouraged healthy food choices and a kitchen ‘hide and seek’ game to identify healthy eating cues and reinforce healthy decisions through feedback and guidance. Johnston et al. ⁴¹ and Sullivan et al. ²⁸ used the Second Life platform to deliver behaviourally grounded VR interventions promoting diet and physical activity. Participants in the Second Life group accessed virtual spaces such as gyms, restaurants, and grocery stores to practice behavioural strategies and receive immediate feedback. Thomas et al. ²⁷ used a standalone ‘Experience Success’ platform, offering PC‐based VR scenarios designed to reflect real‐life challenges, enabling participants to practice weight‐loss strategies in a low‐risk, interactive format alongside participation in an online weight management programme (Weight Watchers). Except for Thomas et al., ²⁷ Johnston et al., ⁴¹ and Sullivan et al., ²⁸ all studies applied therapist‐guided approaches.

3.2. Investigating problematic clinical trials in systematic reviews

The summary of the INSPECT‐SR tool assessment is highlighted in Supplemental Table 4. Although none of the studies had any issues related to the first and second domains, including ‘post‐Publication Notices’ and ‘Text and Figures’, all trials exhibited some to serious concerns in the ‘Conduct, Governance and Transparency’ domain. In addition, four studies were classified as having some concern in the ‘Results in the Study’ domain due to the lack of reported SD and/or insufficient baseline information. ³⁷ , ³⁸ , ³⁹ , ⁴³ The overall judgement for all studies was ‘Serious concern’, except for the studies by Ferrer‐Garcia et al., ²⁶ Thomas et al., ²⁷ and Phelan et al. ⁴²

3.3. Risk of bias assessment

Table 2 presents a summary of the risk of bias assessment. All studies, except Johnston et al., ⁴¹ mentioned that the participants were randomly assigned to the groups; however, only seven studies provided adequate information on the allocation concealment. There was a serious concern regarding the masking of the investigators as well as the participants since none of the trials applied a blinding design. Except for one trial, ³⁷ the attrition of participants was sufficiently addressed in all studies. Most of the studies acknowledged reporting bias, which included financial support (public, commercial, or industry) and authors' relationships with any external agency that could influence the reporting of outcomes. ²⁴ , ²⁶ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² The other bias was considered as accounts regarding usability assessments of VR treatment and possible side effects of using VR among participants, which was sufficiently addressed only in one trial. ⁴²

TABLE 2.

The summary of review authors' judgements about each risk of bias item for included studies.

Study	Random sequence generation	Allocation concealment	Blinding	Incomplete outcome data	Selective reporting	Other bias
Cesa et al. (2013)	L	L	H	L	L	U
Johnston et al. (2012)	U	H	H	L	L	U
Manzoni et al. (2016)	L	L	H	L	L	U
Phelan et al. (2021)	L	L	H	L	L	L
Riva et al. (2004)	L	U	H	L	L	U
Riva et al. (2001)	L	U	H	U	U	U
Sullivan et al. (2013)	L	U	H	L	U	U
Thomas et al. (2020)	L	L	H	L	H	U
Sansoni et al. (2024)	L	L	H	L	L	U
Ferrer‐Garcia et al. (2019)	L	L	H	L	L	U
Cárdenas‐López et al. (2015)	L	U	H	L	U	U

Note: H, high risk of bias; L, low risk of bias; U, unclear or unrevealed risk of bias. Criteria defined for risk of bias assessment are according to the Cochrane guidelines. The other bias is considered as accounts regarding usability assessments of VR treatment and possible side effects of using VR among participants.

3.4. Main findings

3.4.1. The effect of VR‐enhanced treatment on body weight

Eight studies reported on body weight, among which the study by Riva et al., ³⁷ was only presented as the systematic review due to the lack of providing SD. This study suggested that VR‐enhanced treatment (behavioural therapy delivered in a VR environment alongside a low‐calorie diet and physical training) resulted in greater body weight reduction than a non‐VR low‐calorie diet plus physical training (mean body weight reduction: 11.33 kg vs. 7.58 kg). ³⁷

The forest plot illustrates the changes in body weight post‐intervention and follow‐up, as shown in Figure 2A,B, respectively. Seven articles (12 arms) provided information on weight measures at baseline and the end of the treatment, while 3 studies (5 arms) performed further assessment to explore body weight changes in a long‐term follow‐up period. The pooled effect size indicated that VR‐enhanced treatment was not superior to conventional approaches at the end of a weight loss intervention (MD: 0.01 kg; 95% CI: −1.68, 1.70; I ² = 0%). However, the VR‐enhanced treatment group experienced a greater weight improvement than the conventional approaches during the maintenance phase (MD: −5.16 kg; 95% CI: −8.46, −1.85; I ² = 0%). Subgroup analyses based on intervention type (VR‐Experiential Cognitive Therapy vs. other VR treatments) and health conditions (overweight or obesity alone vs. eating disorder and excess body weight) did not reveal significant differences between subgroups for body weight at the end of the intervention. However, a greater effect on body weight at the end of follow‐up was observed in patients with overweight or obesity alone and in those who received VR‐Experiential Cognitive Therapy (Supplemental Table 5). No significant correlation was detected between the change in body weight regarding the VR‐enhanced treatment and the duration of the study, as well as baseline BMI, either after intervention or at the end of follow‐up (Supplemental Table 6).

FIGURE 2.

Forest plot showing the effect of the VR‐enhanced approach on change in body weight and BMI at post‐intervention and follow‐up. Body weight change after (a) intervention (b) follow‐up. BMI change after (c) intervention (d) follow‐up.

3.4.2. The effect of VR‐enhanced treatment on BMI

The summary of the overall effect size of changes in BMI at the post‐intervention and during the maintenance phase is outlined in Figure 2C,D, respectively. Five trials (9 arms) reported on the BMI data for participants at the pre‐ and post‐intervention. The result of the meta‐analysis did not yield any significant difference in BMI changes between VR‐enhanced treatment and conventional therapy (MD: 0.08 kg/m²; 95% CI: −0.84, 1.01; I ² = 0%). Similarly, there was no significant difference in BMI changes during follow‐up between VR‐enhanced treatment and conventional therapy (MD: −0.28 kg/m²; 95% CI: −1.39, 0.82; I ² = 0%). Stratified analyses based on treatment type and health conditions did not reveal any differences between groups for BMI, either at the end of the intervention or at the end of follow‐up (Supplemental Table 5). Additionally, we did not find any significant relationship between the change in BMI resulting from VR‐enhanced treatment and duration of study, as well as baseline BMI, either after intervention or at the end of follow‐up (Supplemental Table 6).

3.5. Sensitivity analysis and publication bias

The result of a sensitivity analysis performed by removing each individual effect size one at a time indicated that the overall result was not influenced by any single estimate. Regarding the r correlation, sensitivity analysis did not show any change in the overall effect size when the r value was considered between 0.2 and 0.8 (Supplemental Table 7).

Visual inspection of funnel plots did not suggest any asymmetry in estimating the effect of VR‐enhanced treatment on all outcomes of interest (Supplemental Figure 1). No potential evidence of publication bias was found in Egger's regression asymmetry test and Begg's rank‐correlation methods test for body weight after treatment (Begg's test p = 0.41; Egger's test p = 0.44). Due to the limited number of studies, publication bias tests could not be applied to the remaining outcomes.

4. DISCUSSION

This systematic review and meta‐analysis critically evaluated the existing evidence from clinical trials investigating the delivery of evidence‐based behavioural treatments through virtual reality for weight management during both intervention and follow‐up periods. The results suggest that VR‐enhanced treatments could be more effective than conventional therapies in weight management over a longer term (during the maintenance phase). This effect was more pronounced among patients who received VR‐experiential cognitive therapy and those with overweight or obesity without eating disorders. The findings also showed that VR‐enhanced treatments were as effective as evidence‐based conventional therapies in weight reduction after treatment. The results were independent of baseline BMI measures and treatment duration.

Excess body weight is regarded as a great challenge to the healthcare system due to its high prevalence and the serious consequences it poses for society. ⁴⁴ Evidence indicates a strong association between excess body weight and the incidence and progression of diabetes, ⁴⁵ hypertension, ⁴⁶ and stroke. ⁴⁷ Obesity is also linked with an increased risk of death among individuals aged 50–71 years, across both sexes and all racial or ethnic groups. ⁴⁸ Reduction in excess body weight is directly associated with a reduction in cardiovascular risk factors, ⁴⁹ , ⁵⁰ highlighting the important role of weight management in human health. Given the fact that the effects of many existing weight loss programmes are often short‐lived, with many patients regaining weight, finding a more consistent method of weight management seems necessary. ⁵¹ , ⁵² This study showed that both conventional weight management methods and VR‐enhanced approaches are equally efficacious; however, the addition of VR to conventional treatment yielded a notable advantage in the longevity of weight‐loss results. This aligns with a previous systematic review, which found that the use of VR for self‐monitoring diet, physical activity, and weight may facilitate weight loss, improve body image satisfaction, and enhance health self‐efficacy. ³⁰ Likewise, a systematic review and meta‐analysis by Yaw et al., ²⁹ reported no significant difference between VR‐based intervention and control groups in either weight or BMI at post‐treatment. While their result at the maintenance phase also suggested a similar effect of VR on BMI, consistent with our findings, they did not observe a significant impact on weight during this phase. This discrepancy may be due to the differences in the analytical approaches and the eligibility criteria; based on these criteria, we exclude the studies by Horne et al. ⁵³ and Seo et al., ⁵⁴ which used a personalised avatar and an indoor cycling programme within a VR environment, respectively, solely to enhance individual motivation. In addition, we did not consider the waitlist group as the control group, since they did not receive any treatment, and it cannot be determined whether the effect is due to the VR or the treatment alone. We also excluded Riva et al. ⁵⁵ and Riva et al., ⁵⁶ as they presented the same data on weight as Riva et al. ³⁹ and Manzoni et al., ³⁸ respectively.

We found that the effect of VR‐enhanced treatment on body weight can last longer during the maintenance phase compared to conventional care. This positive outcome may, in part, be attributed to VR's ability to enhance user engagement. For instance, prior research has demonstrated that VR fosters immersion, which facilitates more effective learning and greater engagement among students. ⁵⁷ In fact, VR provides additional advantages through motivational reinforcement, practice‐oriented instruction, and active learning strategies. These elements together foster an interactive environment that may help reduce resistance to treatments. ⁴¹ , ⁵⁵

Unlike the results for body weight, findings indicated a non‐positive effect for VR‐enhanced treatment in the maintenance phase for BMI. This could be related to differences in the included studies in the pooled effect size of these outcomes. Given that all of the included studies, with the exception of the study by Cesa et al., ²⁴ reported on weight loss or BMI and did not include both outcomes. We also did not detect a significant correlation between the change in body weight and the duration of treatment, as well as the duration of follow‐up. This might be confusing, as we found that the effect of VR‐enhanced treatment could last longer; it could be due to the small number of trials with an approximately similar duration of follow‐up. No significant association was also detected between the baseline BMI and changes in weight or BMI in response to the treatment. Future studies with longer durations of both treatment and maintenance phases are necessary to explore a possible relationship between the change in weight over time resulting from the VR‐enhanced treatment and baseline BMI values.

Despite promising findings, the results remain inconclusive due to a lack of trustworthiness based on the INSPECT‐SR assessment. The INSPECT‐SR assessment indicated some serious concerns across all included trials, with most issues related to the ‘Conduct, Governance, and Transparency’ domain. A common concern among the trials was the absence of an ethical code number, which prevented us from verifying the details of ethical approval. Additionally, the lack of study registration or insufficient information in the registrar panel was frequently observed. Although the INSPECT‐SR tool guidelines recommend excluding problematic studies, we decided to include all studies and report the assessment results since trial registration was not a required practice before 2010. Furthermore, four studies reported the same trial registration ID, ²⁴ , ³⁸ , ³⁹ , ⁴⁰ which was classified as a serious concern. However, these studies appear to be driven by a larger overall project, and the possibility of multiple publications from the same study was low, as their baseline information differed. The limited number of available clinical trials, along with concerns regarding their trustworthiness assessed by the INSPECT‐SR tool, highlights the need for further high‐quality clinical trials to reach a definitive conclusion for this novel approach.

Apart from the direct effect of VR‐enhanced treatment on weight outcome, two additional factors may play a critical role in the decision to adopt this cutting‐edge technology. First, the highly interactive and immersive environments offered by the new generation of VR. The use of VR in clinical studies dates back to the early 20th century; however, due to technological constraints and the historical costs of conducting VR research, VR‐based clinical studies only saw an increase in numbers in recent years. ⁵⁸ As VR technology has developed, it now includes a set of highly intuitive interactions, particularly when used with hand or body gesture controls, ⁵⁹ , ⁶⁰ which significantly broadens the inclusivity spectrum among users with various levels of digital literacy. It can also strengthen the cognitive effects of treatments and consequently decrease the costs of remedies. ⁶¹ In addition, VR has become a great tool for clinical investigations. It is also worth mentioning that rapid improvements in stand‐alone VR capabilities, combined with the rise of affordable head‐mounted (HMD) models, could consolidate VR's position as an effective and viable platform for remote research and clinical assessment. ⁶² , ⁶³ Second, the less chance of cybersickness due to the rapid technological advancements of VR. In the present systematic review and meta‐analysis, none of the included studies mentioned serious symptoms of cybersickness, and they reported that the rate of satisfaction from treatment was the same in experimental and control groups. However, that being said, there are inherent risks associated with the use of VR that should be noted. First, cybersickness is reported as a more prevalent adverse event. ⁶⁴ Associated with several different sources under various circumstances, symptoms of cybersickness can include eye fatigue, headache, nausea, and dizziness. ⁶⁵ , ⁶⁶ Another problem concerning VR is the risk it poses to individuals with a history of epilepsy. However, there have been attempts to alleviate VR‐related issues. New headsets, like HMD Quest 3, offer a considerably higher frame per second as well as lower latency than their predecessors, and the resolution has significantly improved, too. ⁶⁷ , ⁶⁸ These technological advancements have a positive impact on the VR experience and have made VR a better candidate for research and treatment approaches. Substantial improvements in these key factors may indicate an increasing potential for VR‐enhanced treatments to positively impact health outcomes such as weight management in the future.

Similar to other systematic reviews and meta‐analyses, the present study's limitations stem from the included studies; namely, the small number of available trials, most of which were conducted with small sample sizes. Conducting trials with limited participants may influence the power of the study (increasing the risk of type II error) and may mistakenly lead to a non‐significant result. Although only one study did not report whether randomisation was performed, there is insufficient information in five studies regarding allocation concealment, which raises concerns about decreased internal validity, potential overestimation of results, and reduced generalizability as a consequence of selection bias. The lack of blinding in outcome assessment in all trials raised a concern regarding the risk of bias, primarily performance bias, which may lead to an overestimation of the treatment effect. However, masking is almost impossible in these studies, as there is no condition to keep participants and investigators unaware of the treatment's nature. A limited number of included trials provided information on the follow‐up phase after the active intervention ended. Nevertheless, even with these few studies, we were able to demonstrate the potential long‐lasting effects of VR‐enhanced treatment. Furthermore, except for the study by Sansoni et al., ⁴⁰ which is a more contemporary study on younger adults (2024, with participants having a mean age of 28 years), the remaining studies recruited older populations and/ or were carried out in earlier years (2001 onwards). It is noteworthy that the younger generation, thanks to their higher level of digital literacy, is more prepared to learn and adapt to these types of treatments; however, given that virtual reality provides a more natural user interaction compared to preceding digital technologies, VR‐based studies have also proven to be acceptable for older users. ⁶⁹ , ⁷⁰ Additionally, considering that VR has improved significantly in the past decade, some of the included VR trials in this study lack the technical qualities of similar VR hardware and software available today. Therefore, it is reasonable to postulate that with the new generation of VR devices, we might achieve better results with longer‐lasting effects.

5. CONCLUSION

The current study's findings suggest that VR‐enhanced treatments may contribute to longer‐term weight maintenance, as patients in VR‐based trials were less likely to regain weight after completing the treatment. However, the results should be interpreted cautiously, as the small number of available studies, along with their limited quality and reliability, prevents us from drawing a definitive conclusion at this time.

It is noteworthy that recent advancements in mixed and extended reality technology could lead to even more substantial improvements in investigations' outcomes. The development of high‐fidelity head‐mounted displays has made a striking leap forward with realistic graphics and low latency. Creating high‐precision virtual test environments now seems well within reach. In addition, adjunct technologies such as haptics and real‐time body tracking open new avenues towards a more natural interaction, which could remarkably increase the quality of VR test environments in the near future. Taking into account the current state of VR technology, its predicted trajectory, and its potential integration into health care approaches, future high‐quality randomised clinical trials are needed to evaluate the use of adjunct VR in delivering evidence‐based weight management treatments.

AUTHOR CONTRIBUTIONS

MP and SP contributed to the design of the review, study selection, data extraction, and risk of bias assessment. NA developed and executed the search strategy through the databases. MP conducted the data analysis and interpreted the results. MP and SP wrote the initial draft. DM, RM, REG, and BH critically revised the manuscript. All authors read and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests.

Supporting information

Data S1: Supporting Information.

Pourmasoumi M, Pourmasoumi S, El‐Gabalawy R, et al. Evaluating the use of virtual reality for weight management: A systematic review and meta‐analysis. Diabetes Obes Metab. 2026;28(2):1315‐1326. doi: 10.1111/dom.70318

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.