Abstract

Background & Aim: Needle-related procedures cause pain and distress in preschool children, and conventional virtual reality distraction cannot adapt to a child's changing state. This study aimed to evaluate an artificial intelligence-assisted adaptive virtual reality distraction tool for reducing procedural pain and behavioral distress in preschool children undergoing invasive procedures.

Methods & Materials: This simulation study used a two-arm randomized framework with synthetic profiles of 120 children aged 3-6 years undergoing venipuncture, intravenous cannulation, intramuscular injection or blood sampling. The study was carried out in the Sv hospital, Sangareddy, Sangareddy (Dist), Telangana, India. Profiles were allocated by block randomization to adaptive virtual reality distraction (n = 60) or customary care with simple verbal distraction (n = 60). Pain was assessed with the Face, Legs, Activity, Cry and Consolability scale and behavioral distress with the Observational Scale of Behavioral Distress-Revised. The present study were ethically approved. Secondary outcomes were heart rate, procedure duration, first-attempt success, cooperation and caregiver satisfaction. Random forest, logistic regression and support vector machine classifiers were compared for distress classification.

Results: The intervention group had lower procedural pain (mean 3.21 versus 4.16) and behavioral distress (2.87 versus 5.52) than the control group (both p < 0.001). Heart rate during the procedure was lower, procedures were shorter, and cooperation and caregiver satisfaction were higher with the intervention. First-attempt success was higher after adjustment (adjusted odds ratio 3.18). Classifier accuracy ranged from 80.0% to 88.3%.

Conclusion: In this simulation, adaptive virtual reality distraction was associated with lower procedural pain and distress in preschool children. These hypothesis-generating findings require confirmation in a registered clinical trial with real participants.