The impact of virtual reality (VR) on pain management in the emergency department: a systematic review and meta-analysis

Abstract

Background

Although pharmacological and non-pharmacological therapy are available, pain management in the emergency department (ED) may be difficult. In recent years, virtual reality (VR) has emerged as a practical distraction approach for pain relief, especially in children and adolescents. However, little is known about the efficacy of VR in the ED. Therefore, the current study investigated the effect of VR on pain management in adult and pediatric patients in the ED.

Objectives

Primary objective: To assess the effect of VR on pain intensity. Secondary objectives: To investigate patient satisfaction with VR as a pain management method and to assess the incidence of cybersickness after VR intervention.

Search methods

We comprehensively searched CENTRAL, Google Scholar, PubMed, and MEDLINE databases for all studies published until May 2024. The search was limited to records authored in English and it did not include grey literature, such as theses and dissertations.

Selection criteria

We included randomized and non-randomized studies reporting the use of VR to manage pain in patients presenting to the ED.

Results

The pooled analysis demonstrated a significant reduction in pain scores with the use of VR (SMD: -0.67; p = 0.001). Furthermore, subgroup analyses showed consistent pain reduction with the use of VR across adult and pediatric patients (SMD: -1.08; p = 0.01 and SMD: -0.39; p = 0.009, respectively). Significant pain reduction was also observed in patients undergoing minor medical procedures and those with acute pain unrelated to medical procedures (SMD: -1.55; p = 0.03 and SMD: -0.32; p = 0.002, respectively).

Conclusion

Overall, VR offers effective pain management in adults and pediatric patients with non-procedural acute pain and those undergoing painful procedures in the ED.

Systemic review protocol registration

PROSPERO: CRD42024609121.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12873-025-01285-x.

Introduction

Pain is one of the most frequent symptoms among patients brought to the emergency department (ED). However, prompt and effective pain management in the ED might be problematic due to various factors [1, 2]. Currently, pain relief options in the ED include pharmacological and non-pharmacological techniques. While pharmacological methods such as opioids are known to substantially relieve pain, they often lead to inconsistent and inadequate pain management. Therefore, individuals with pain are likely to demand more dosages or other pharmacological medications for pain management, which might result in side effects such as respiratory and/or central nervous system depression [3, 4].The prescription of opioid medications in emergency departments has been associated with contributing to the opioid dependence epidemic in several countries [5, 6]. However, there are ways to mitigate this by properly prescribing opioids and providing adequate education to both patients and families.

On the other hand, distraction is the frequently utilized non-pharmacological strategy for pain management, with research proving its usefulness in numerous therapeutic contexts [7, 8]. Traditionally, distraction approaches such as TV, music, or other audiovisual media have been utilized to alleviate pain and discomfort during laceration repair, insertion of intravenous catheters, lumbar puncture, and other uncomfortable bedside operations [9–11]. However, these strategies are generally underutilized in a hectic ED setting. As a result, there is a need for a simple, non-opioid, and effective approach that may enhance pain reduction in the ED.

With technological breakthroughs, virtual reality (VR) has been recognized as an effective psychological treatment and active distraction because it creates a sense of “presence” that conventional distraction techniques cannot deliver. So far, data supports the use of VR to alleviate pain and anxiety during burn treatment or venous access placement, particularly in children and adolescents [12–15]. However, little is known regarding the effectiveness of virtual reality in the ED. As a result, the present research explored the impact of VR on pain management in adult and pediatric patients in the ED.

Methods

Literature search and information sources

CENTRAL, Google Scholar, PubMed, and MEDLINE databases were extensively searched for randomized and non-randomized studies related to our study objective. Furthermore, bibliographies of relevant studies retrieved from these databases were searched for additional studies. The search was restricted to texts authored in English; however, no time limitation was provided as all articles published from inception until May 2024 were included. The initial search in the electronic databases mentioned above was comprehensive and inclusive as it used keywords such as virtual reality, pain, emergency department, etc., combined with the Boolean “AND” and “OR.” Moreover, duplicate articles and grey literature containing unpublished data were overlooked as they were likely to undermine the statistical power of the present meta-analysis. The full search strategy is outlined in Appendix A.

Eligibility criteria

Two independent reviewers screened full-text articles of potential studies and included those that met the following PICOST (Participants, Intervention, Control, Outcome, Study design, and Time frame) criteria. P: Adult or pediatric patients presented to the ED with pain or undergoing painful procedures, I: Virtual Reality, C: pharmacologic treatments, standard of care, or other distraction techniques. O: pain intensity before and after intervention, patient satisfaction to pain management, and motion sickness (cybersickness) S: randomized controlled trials (RCTs) and non-randomized studies, T: no time frame was provided. Conversely, studies that failed to satisfy the PICOST criteria or those designed as case reports, narrative reviews, meta-analyses, conference abstracts, animal studies, or letters to the editor were excluded. Moreover, studies that described the adoption of VR in alleviating pain in hospitalized patients or those undergoing procedures in other medical departments were excluded. Furthermore, studies that reported alternative distraction strategies or pain levels longer than an hour after the medical procedure were omitted. All differences during this phase were addressed by constructive discussions between the reviewers.

Data extraction and data items

Two independent reviewers used a standardized Excel spreadsheet to collect all the data required for a comprehensive review and meta-analysis. The data extracted included the surname of the first author, publication date, country in which the study was performed, characteristics of enrolled participants such as gender, sample size, and age groups (adults or children), VR device used, type of VR environment (interactive or non-interactive), cause of pain, the time point pain scores were and the tool used for measuring pain. Any disagreement throughout the entire process was settled by productive dialogue between the two reviewers or by contacting a third reviewer.

The major goal of this research was to investigate the impact of VR on pain management among patients presented to the ED. Therefore, the primary endpoint was the difference in pain intensity before and after VR intervention. The secondary endpoints were patient satisfaction to pain management and the incidence of cybersickness.

Quality appraisal

This meta-analysis included RCTs and non-randomized studies; therefore, quality appraisal was performed using Cochrane’s Risk of Bias (ROB-2) tool and the Newcastle Ottawa Scale (NOS). The NOS was used to rate studies based on the selection, comparability, and outcome domains. Afterward, the overall methodological quality was categorized as poor (NOS score 0–3), fair (NOS score 4–6), or good (NOS score 7–9).

On the other hand, we chose the ROB-2 tool for the evaluation of RCT as it provides a more detailed assessment of bias compared to the ROB-1. Using the ROB-2 tool, RCTs were assessed according to the randomization process, allocation concealment, completeness of outcome data, selective outcome reporting, and other potential sources of bias. Blinding was not included in the risk of bias assessment due to its inherent impracticality in VR interventions, where participants are aware of the treatment condition.

Data synthesis

All statistical analyses in the present study were performed using the Review Manager software (RevMan version 5.4.1). Since pain intensity and satisfaction levels were measured using different tools, the standard mean difference (SMD) was used to test for the overall effect size. Data regarding motion sickness was computed using the Mantel-Haenszel method for binary data and pooled using odds ratio (OR). Moreover, the extent of interstudy heterogeneity was measured using the I [2] statistics, of which values lower than 25%, between 25% and 50%, and above 50% indicated low, moderate, and high heterogeneity, respectively. A random-effects model was used to pool outcomes with high interstudy heterogeneity, while a fixed-effect model was used for outcomes with low heterogeneity. For studies where pain and satisfaction levels were reported as median and interquartile range, the mean and standard deviation were calculated using the formula provided by Wan and colleagues [16].

Subgroup analyses were carried out to investigate the potential effect of modifiers such as patient age group, type of VR environment, type of pain managed (needle-related pain, pain due to minor procedures, and acute pain), study design, and the tools used for pain measurement. Needle-related pain referred to pain arising from needle-rated procedures, such as intravenous cannulation and venipuncture, while minor procedural pain referred to pain due to procedures, such as wound closure, laceration repair, casting, fracture reduction of joint dislocation, thoracotomy, paracentesis, or arterial blood gas measurement.

Certainty of evidence

The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) tool was used to evaluate the certainty of all the pooled outcomes. Using this tool, the outcomes were rated as having high, moderate, low, and very low certainty of evidence based on five different domains, namely risk of bias, consistency of effect, imprecision, indirectness, and publication bias. The decision to downgrade or upgrade the certainty of evidence under the “risk of bias” domain was made using the results of ROB-2 tools, while funnel plots inform the rating of publication bias.

Results

Study selection

1819 potential studies were identified from the electronic databases mentioned earlier. After thorough screening for duplicates, 815 articles were eliminated. Titles and abstracts of the remaining 1004 articles were screened, of which 774 deemed irrelevant to our objective were excluded. In addition, 96 of the remaining 230 records were eliminated before assessment for eligibility because they were case reports, letters to the editors, review articles, or meta-analyses. Finally, only 13 studies were deemed eligible for inclusion. The other 121 studies were excluded because they included hospitalized patients or those in outpatient clinics, anesthesia departments, or trauma units (n = 108) and did not report the outcomes of interest (n = 13) (Fig. 1).

Fig. 1.

PRISMA flow diagram for study selection

Summary of study characteristics

Thirteen studies with 1120 patients were included for review and analysis [17–27]. The majority of these patients were male (56.1%). Ten of the included studies were RCTs, while the other 3 studies were non-randomized. Moreover, 8 of these studies included pediatric patients, and 5 included adult patients. Pain intensity was measured with tools such as the 10-point visual analog scale (VAS), 100 mm VAS, 10-point numerical rating scale (NRS), 10-point color analog scale, and 10-point Faces Pain Scale-Revised (FPS-R) (Table 1).

Table 1.

Summary of study characteristics

Study ID	Study design	Country	Participant characteristics			VR device used	VR type	Source of pain	Time point for pain assessment	Measure of pain
			Sample (n)	M/F	Age group
Bosso et al. 2023 [18]	RCT	Switzerland	123	81/42	Adults	VR headset (unspecified)	Non-interactive	Minor procedures such as suturing, wound exploration, casting, fracture reduction of joint dislocation, thoracotomy, paracentesis, or arterial blood gas measurement.	Immediately after the procedure.	VAS (0–100)
O’Sullivan et al. 2023 [19]	Prospective cohort study	United States	22	10/12	Children	Samsung Gear VR head-mounted display (HMD) with Samsung 7 S and headphones	Non-interactive	Intravenous cannulation or venesection.	2 min after the procedure	CAS (0–10)
Chan et al. 2019 [20]	RCT	Australia	123	67/56	Children	Google Pixel Xl/Google Dream VR headset	Interactive	Venipuncture or intravenous cannulation	Immediately after the procedure.	FPS-R (0–10)
Dumoulin et al. 2019 [21]	RCT	Canada	59	38/21	Children	eMagin z800 VR HMD	Interactive	Needle-related procedures	Immediately after the procedure.	VAS (0–100)
Sikka et al. 2019 [22]	Prospective cohort study	United States	93	36/57	Adults	Samsung Gear VR head-mounted display (HMD) with Samsung 7 S and headphones	Non-interactive	Abdominal, Back, Chest, musculoskeletal, Flank, genitourinary, or other pain	Immediately after the VR intervention.	NRS (0–10)
Rezal et al. 2023 [23]	RCT	Iran	160	125/35	Adults	Samsung Gear VR R325 and P-NET VR-100 VR headsets	Non-interactive	Laceration repair	Immediately after the VR intervention.	NRS (0–10)
Osmanlliu et al. 2021 [24]	RCT	Canada	62	24/38	Children	Oculus Rift	Interactive	Intravenous insertion or venipuncture	2 min after the procedure	NRS (0–10)
Canares et al. 2021 [25]	RCT	United States	55	20/35	Children	Oculus Go	Interactive	Venipuncture	Immediately after the procedure.	VAS (0–10)
Birrenbach et al. 2022 [26]	Prospective cohort study	Switzerland	52	20/32	Adults	Pico G2 4 K VR headset	Non-interactive	Traumatic and non-traumatic	Immediately after the VR intervention.	NRS (0–10)
Schlechter et al. 2021 [27]	RCT	United States	115	55/60	Children	VR headset (unspecified)	Interactive	Intravenous line placement	Immediately after the procedure.	FPS-R (0–10)
Butt et al. 2022 [28]	Prospective RCT	United States	110	66/44	Children	Oculus Go	Interactive	Acute mild to moderate traumatic/nontraumatic pain	15 min after VR intervention	FPS-R (0–10)
Ko et al. 2024 [29]	RCT	China	80	50/30	Adults	VR goggles (unspecified)	Non-interactive	Wound closure	5 min after wound closure	VAS (0–10)
Goldman et al. 2021 [30]	Prospective RCT	Canada	66	36/30	Children	VOX + Z3 3D Virtual Reality Headset	Interactive	Intravenous catheterization	Immediately after the procedure.	FPS-R (0–10)

Note: RCT: Randomized controlled trial VR: virtual reality NRS: numerical rating scale VAS: visual analog scale FPS-R: Faces Pain Scale-Revised CAS: Color analog scale

Effect of VR intervention in reducing pain intensity

Of the 13 included studies, 12 reported pain intensity before and after VR intervention. A pooled analysis of data from these studies showed a significant reduction in pain after utilizing VR (SMD: -0.67; p = 0.001), but with a substantial heterogeneity across the studies (I² = 91%) (Fig. 2).

Fig. 2.

Forest plot showing the effect of VR on pain intensity

Patient satisfaction

Four studies involving 267 patients reported patient satisfaction to pain management. The pooled analysis of data from these studies showed that satisfaction levels were significantly higher in the VR group than in the standard care group (SMD: 0.37; 95% CI: 0.11 to 0.63; p = 0.005) (Fig. 3), suggesting that patients were more satisfied with VR as a method for pain management.

Fig. 3.

Forest plot comparing satisfaction levels between VR and standard care groups

Cybersickness

While our meta-analysis revealed that the incidence of cybersickness was higher in patients receiving VR intervention compared to patients receiving other methods for pain management (31/153 (20.2%) vs. 33/182 (18.1%)), the difference between these two groups was statistically insignificant (p = 0.39). The pooled results also showed a high interstudy heterogeneity (I² = 61%) (Fig. 4).

Fig. 4.

Forest plot comparing the incidence of cybersickness

Subgroup analyses

The subgroup analyses showed consistent pain reduction with the use of VR across adult and pediatric patients (SMD: -1.08; p = 0.01 and SMD: -0.39; p = 0.009, respectively) (Fig. 5). Significant pain reductions were also observed in interactive and non-interactive VR environments (SMD: -0.46; p = 0.002 and SMD: -0.88; p = 0.03, respectively) (Fig. 6). Insignificant pain reduction with the use of VR was observed in patients with needle-related pain (p = 0.06) (Fig. 7), non-randomized studies (p = 0.11) (Fig. 8), and studies using the 10-point CAS (p = 0.56) and 100 mm VAS for the measurement of pain (p = 0.17) (Fig. 9). Therefore, the observed heterogeneity was likely caused by the study design, the type of pain, and the variation in tools used to measure pain.

Fig. 5.

Forest plot showing the effect of VR on pain intensity stratified according to patient age group

Fig. 6.

Forest plot showing the effect of VR on pain intensity stratified according to the type of VR environment

Fig. 7.

Forest plot showing the effect of VR on pain intensity stratified according to the type of pain managed

Fig. 8.

Forest plot showing the effect of VR on pain intensity stratified according to the study design

Fig. 9.

Forest plot showing the effect of VR on pain intensity stratified according to the pain measurement tools

Quality assessment outcome

Table 2 in Appendix B shows the results of the quality appraisal using the Newcastle Ottawa Scale. The assessment showed that all three observational studies had fair methodological quality. The main reason most studies did not achieve maximum scores under the selection domain was that they were carried out in single centers. Moreover, none of the studies achieved maximum scores under the outcome domain due to a lack of information about how outcomes were assessed.

On the other hand, Supplementary Fig. 1 in Appendix B summarizes the risk of bias. The assessment shows that all but one study had a high risk of bias under the other bias domain. This high risk was because the studies were conducted in single medical institutions.

Certainty of evidence

The GRADE assessment has shown that evidence regarding the impact of VR on pain intensity had a moderate level of certainty due to high interstudy heterogeneity. Very low and high levels of certainty were also observed in outcomes regarding cybersickness and patient satisfaction, respectively. The very low certainty was attributed to high heterogeneity and serious imprecision due to a wide 95% confidence interval (Tables 3 and 4 in Appendix C).

Discussion

Pain is one of the most common complaints among patients in the ED. Therefore, effective pain management is of utmost importance. In the present study, we evaluated the effect of VR on pain relief in the ED and found that VR is an efficient and viable technique for reducing pain associated with minor procedures and non-procedural acute pain. Our findings also indicate that VR is equally beneficial at managing pain in both children and adults.

The results presented in the current research compare favorably with previous systematic reviews and meta-analyses. For instance, a meta-analysis of 31 studies with 1947 patients undergoing various medical procedures in different medical departments reported that VR was an effective tool for alleviating pain related to post-surgery wound dressing, burn wound repair, needle-related procedures, and thermal stimuli [31]. Another meta-analysis investigating the role of VR in pain relief among 7133 participants undergoing different medical procedures found that VR offered an effective alternative for pain control [32]. Similarly, a meta-analysis of 20 articles reported that VR could effectively mitigate pain related to labor, dental procedures, burn wound repair and episiotomy repair [33]. A meta-analysis investigating the efficacy of VR in the alleviation of pain in 2224 pediatric patients having needle-related procedures also found that VR was an efficient pain control tool [34].

While our findings indicate that VR is useful for pain control, the mechanism by which it alleviates pain remains unclear. However, it has been hypothesized that VR can relieve pain through distraction due to its potential ability to propel a patient into an alternate reality, resulting in a delayed reaction to inbound pain signals [35]. Multiple arguments have been offered concerning how VR distraction could hinder or reduce pain perception. Melzack and Wall proposed the Gate control theory in 1965 [36]. This hypothesis contends that central nervous system processes such as attention, emotion, and memory play a key part in how one perceives things. Thus, as pain impulses navigate throughout the body, they have to go through “nerve gates” before the body can evaluate the degree of consciousness. This implies that the degree of attention paid to the pain, the emotion attached to the pain, and previous experience with pain all serve an important part in how pain is perceived individually. This theory was further expanded by McCaul and Mallot in 1984 by describing the limited capacity of attention in humans [37]. According to their argument, if individuals are exposed to signals (e.g., visual and audio stimuli from VR devices) other than painful stimuli, they are more likely to sense less pain.

Contrary to our findings, some included have shown no added benefit on pain management with the use of VR. For instance, Doumolin and colleagues reported that although VR was associated with a significant reduction in fear of pain, it did not offer any significant effect on pain intensity. This partial efficacy of VR in regard to pain management was explained by a number of factors. First, most patients (about 75–87%) had used anesthetic cream before the needle-related treatment, which may have played a role in low pain levels at baseline. Another potential explanation is that the cognitive load may not have been adequate. In other words, the virtual environment established in that study may not have been as interactive as anticipated due to the limited funding. Consequently, it is conceivable that certain younger patients (aged 7–10) encountered difficulties operating a wireless mouse, which may have resulted in a lower level of engagement with the task. In contrast, older children may have found the task less engaging and too simple than the computer games they normally play at home or on their mobile devices. Conversely, a prospective study of 22 pediatric patients revealed a slight rise in pain scores when employing the validated CAS. The precise cause of this increase is unknown; however, it may have been due to the non-uniform use of oral and topical analgesics. Nevertheless, this discovery was not unexpected, as another research study that employed this pain measure also demonstrated a minimum clinically significant difference [38].

Interestingly, the subgroup analysis showed no significant difference in pain intensity among patients with needle-related pain. After conducting a sensitivity analysis, we noticed that this insignificant difference was caused by the study conducted by Goldman and colleagues [30]. According to that study, children randomized to the VR group had less pain before starting the IV procedure, which might have caused the insignificant difference. The authors also reported that children were exposed to the VR system before consenting to participate in the trial, meaning that they were excited about the VR experience and reported less pain.

Our study also showed that VR was associated with significantly higher satisfaction levels compared to standard care. This finding might be attributed to several reasons. First, VR is an effective distractor that diverts the attention of patients from painful stimuli. Therefore, the thrill of escaping from “the painful real world” might have helped with satisfaction from pain management. Second, reports have suggested that VR is enjoyable and interesting [39]which might influence patient satisfaction. Finally, apart from reducing pain intensity, VR can also reduce situational anxiety which might lead to increased patient satisfaction [40].

In addition, the current study investigated the incidence of cybersickness, a symptom that is known to occur during exposure to a virtual environment [41]. The pooled results showed that although cybersickness prevalence is slightly higher in the VR group, the difference compared to other methods of pain management was not statistically significant. This finding should be interpreted cautiously due to high interstudy heterogeneity, and further research is needed to support it.

Compared to other meta-analyses [31–34]which focused on the effects of VR on reducing pain intensity across different medical departments, our meta-analysis only focused on patients in the ED. Focusing on patients in the ED provides greater relevance than broadening the scope to include other departments, due to specific contextual and logistical considerations. For instance, applying VR in the ED may present unique logistical challenges such as device sanitization between uses, equipment storage, and the potential cost of replacing damaged or missing components—factors that warrant evaluation in real-world implementation studies. Unlike the other meta-analyses, the current study has also investigated one of the most common harms (cybersickness) associated with virtual environments.

Implications for clinical practice and future research

Our review has revealed that VR has the potential to ameliorate pain due to minor procedures and non-procedural acute pain in the ED. The review also demonstrated that VR significantly reduces pain in adult and pediatric patients, with no significant difference in the incidence of cybersickness compared to other pain management methods. Therefore, it can serve as an alternative tool for managing pain in ED patients.

The current review has shown that the use of various pain scales across studies contributed to the high heterogeneity observed during data synthesis. Standardized and validated measurement tools could help reduce this heterogeneity and enable more consistent evaluation of the relationship between VR and pain control.

Limitations

While the current study has shown VR has a promising ability to mitigate pain during uncomfortable medical procedures and in non-procedural acute pain, its results should be carefully interpreted due to some fundamental limitations. First, most of the studies included in the data synthesis had small sample sizes, thus reducing the statistical power of the meta-analysis. Second, considerable heterogeneity was observed in the pooled analysis. We explored this heterogeneity and found that the study design and variation in pain measurement tools were the driving factors. Nevertheless, we employed the random effects models to counter the heterogeneity and provide conservative estimates. Third, the current study included studies carried out in the ED only; therefore, these findings cannot be generalized to outpatients, hospitalized patients, or patients in other medical departments. Fourth, due to insufficient data, we were unable to conduct a meta-analysis on other harms associated with VR, such as dissociation and dissatisfaction. Finally, only articles published in English were eligible for analysis. Thus, it is possible that data from studies published in different languages that could have been used to improve the statistical power of our meta-analysis were omitted.

Conclusion

Overall, VR seems to be an effective pain management tool in adults and pediatric patients with non-procedural acute pain and those undergoing painful procedures in the ED. However, as VR technology continues to increase in availability and new virtual environments developed, further research will be required to investigate its usability.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ED

Emergency Department

VR

Virtual Reality

MD

Mean Difference

CI

Confidence Interval

CENTRAL

Cochrane Central Register of Controlled Trials

PubMed

PubMed Database

MEDLINE

Medical Literature Analysis and Retrieval System Online

PROSPERO

International Prospective Register of Systematic Reviews

RCT

Randomized Controlled Trials

NOS

Newcastle Ottawa Scale

RoB

Risk of Bias

RevMan

Review Manager (software)

I²

I-squared (measure of heterogeneity)

PICOST

Participants, Intervention, Control, Outcome, Study design, Time frame

VAS

Visual Analog Scale

NRS

Numerical Rating Scale

FPS-R

Faces Pain Scale-Revised

CAS

Color Analog Scale

fMRI

Functional Magnetic Resonance Imaging

HMD

Head-Mounted Display

Author contributions

H.A.Z: conceptualization and review protocol development. H.E, B.P, M.F.A, AM.S, and AH.S: did the quality assessment and statistical analysis. A.T.T, M.G.F and B.T.A: did the data extraction. M.E and E.E.S: wrote up the result and prepared the draft manuscript. H.A.Z: revised the draft manuscript. All authors have approved the manuscript for publication.

Funding

The author(s) received no financial support for this article.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.