Extended reality (XR) technologies are increasingly being used to reduce health and procedural anxieties. The global effectiveness of these interventions is uncertain, and there is a lack of understanding of how patient outcomes might vary between different contexts and modalities. This research used panoramic meta-analysis to synthesize evidence across the diverse clinical contexts in which XR is used to address common outcomes of health and procedural anxiety. Review-level evidence was obtained from 4 databases (MEDLINE, Embase, APA PsycINFO, and Epistemonikos) from the beginning of 2013 until May 30, 2023. Reviews that performed meta-analysis of randomized controlled trials relating to patient-directed XR interventions for health and procedural anxiety were included. Studies that analyzed physiological measures, or focused on technologies that did not include meaningful immersive components, were excluded. Furthermore, data were only included from studies that compared intervention outcomes against no-treatment or treatment-as-usual controls. Analyses followed a preregistered, publicly available protocol. Trial effect sizes were extracted from reviews and expressed as standardized mean differences, which were entered into a 3-level generalized linear model. Here, outcomes were estimated for patients (level 1), studies (level 2), and anxiety indications (level 3), while meta-regressions explored possible influences of age, immersion, and different mechanisms of action. Where relevant, the quality of reviews was appraised using the AMSTAR-2 (A Measurement Tool to Assess Systematic Reviews, Revised Instrument) tool. Data from 83 individual trials were extracted from 18 eligible meta-analyses. Most studies involved pediatric patient groups and focused on procedural, as opposed to general, health anxieties (eg, relating to needle insertion, dental operations, and acute surgery contexts). Interventions targeted distraction-, education-, and exposure-based mechanisms, and were provided via a range of immersive and nonimmersive systems. These interventions proved broadly effective in reducing patient anxiety, with models revealing significant but heterogeneous effects for both procedural (d=-0.75, 95% CI -0.95 to -0.54) and general health (d=-0.82, 95% CI -1.20 to -0.45) indications (when compared with nontreatment or usual-care control conditions). For procedural anxieties, effects may be influenced by publication bias and appear more pronounced for children (vs adults) and nonimmersive (vs immersive) technology interventions, but they were not different by indication. Results demonstrate that XR interventions have successfully reduced patient anxiety across diverse clinical contexts. However, significant uncertainty remains about the generalizability of effects within various unexplored indications, and existing evidence is limited in methodological quality. Although current research is broadly positive in this area, it is premature to assert that XR interventions are effective for any given health or procedural anxiety indication.
Read full abstract