Virtual Reality Training for Chronic Low Back Pain: Comparative Review

Key Takeaways

• Shooting games and VR-based equestrian training were associated with greater pain reduction than all control types, while multicomponent mixed games were superior to placebo controls only; shooting games ranked highest for pain.
• No intervention demonstrated statistically significant superiority for disability, even though disability rankings were reported across modalities.
• Shooting games were associated with lower kinesiophobia than traditional exercise controls, while overall evidence certainty ranged from very low to moderate.

Shooting games ranked highest for pain relief in chronic low back pain, with SUCRA indicating a 98% probability of being the most effective intervention in a JMIR systematic review and network meta-analysis comparing VR-based training modalities. The analysis compared whether distinct VR approaches differed across pain intensity, disability, and kinesiophobia rather than functioning as a single intervention class.

The synthesis included 25 randomized controlled trials with 2610 participants. It used a Bayesian network meta-analysis with standardized mean differences, SUCRA rankings, CINeMA, and adapted GRADE. Interventions were grouped as VR-based cognitive behavioral therapy, walking training, equestrian training, shooting games, multicomponent mixed games, pelvic flexibility games, and yoga training. Main outcomes were pain intensity, disability by the Oswestry Disability Index, and kinesiophobia by the Tampa Scale of Kinesiophobia. The network spanned fully immersive, semi-immersive, and nonimmersive approaches delivered 2 to 5 times weekly for 15 to 60 minutes, totaling 30 minutes to 36 hours over one session to 12 weeks.

For pain, 21 studies with 1017 patients contributed to the network. Shooting games showed an SMD of -4.40 (95% CrI -6.80 to -2.20), and VR-based equestrian training showed -2.00 (95% CrI -3.70 to -0.57), with both outperforming all controls. Disability results drew on 10 studies with 1485 participants; multicomponent mixed games ranked first at 84%, followed by yoga training at 75% and equestrian training at 70%, without significant superiority. Kinesiophobia analysis included 7 studies with 250 patients. In that network, shooting games outperformed traditional exercise controls, with an SMD of -3.40 (95% CrI -5.60 to -1.20), and held the top SUCRA ranking at 94%.

Evidence certainty ranged from very low to moderate overall, with disability comparisons rated very low and many pain comparisons rated low or very low. Downgrading reflected within-study bias, imprecision, and heterogeneity or incoherence. Heterogeneity was substantial for pain and disability, with I2 values of 87.6% and 96.6%, while kinesiophobia showed little heterogeneity at I2=0.00%.

The authors also noted VR-related adverse effects and participant dropouts in some studies, and they cautioned that unblinded shooting game assessments could inflate estimated effects. The rankings remained tempered by sparse data in some comparisons, broad dosing variation, and the stated need for additional randomized evidence.