A Virtual Reality Game Suite for Graded Rehabilitation in Patients With Low Back Pain and a High Fear of Movement: Within-Subject Comparative Study

Abstract

Background: Complex movement pathologies that are biopsychosocial in nature (eg, back pain) require a multidimensional approach for effective treatment. Virtual reality is a promising tool for rehabilitation, where therapeutic interventions can be gamified to promote and train specific movement behaviors while increasing enjoyment, engagement, and retention. We have previously created virtual reality-based tools to assess and promote lumbar excursion during reaching and functional gameplay tasks by manipulating the position of static and dynamic contact targets. Based on the framework of graded exposure rehabilitation, we have created a new virtual reality therapy aimed to alter movement speed while retaining the movement-promoting features of our other developments.

Objective: This study aims to compare lumbar flexion excursion and velocity across our previous and newly developed virtual reality tools in a healthy control cohort.

Methods: A total of 31 healthy participants (16 males, 15 females) took part in 3 gamified virtual reality therapies (ie, Reachality, Fishality, and Dodgeality), while whole-body 3D kinematics were collected at 100 Hz using a 14-camera motion capture system. Lumbar excursion, lumbar flexion velocity, and actual target impact location in the anterior and vertical direction were compared across each virtual reality task and between the 4 anthropometrically defined intended target impact locations using separate 2-way repeated measures analysis of variance models.

Results: There was an interaction between game and impact height for each outcome (all P<.001). Post-hoc simple effects models revealed that lumbar excursion was reduced during Reachality and Fishality relative to that during Dodgeality for the 2 higher impact heights but was greater during Reachality than during Fishality and Dodgeality for the lowest impact height. Peak lumbar flexion velocity was greater during Dodgeality than during Fishality and Reachality across heights. Actual target impact locations during Dodgeality and Fishality were lower relative to those during Reachality at higher intended impact locations but higher at lower intended impact locations. Finally, actual target impact location was further in the anterior direction for Reachality compared to that for Fishality and for Fishality relative to that for Dodgeality.

Conclusions: Lumbar flexion velocity was reduced during Fishality relative to that during Dodgeality and resembled velocity demands more similar to those for a self-paced reaching task (ie, Reachality). Additionally, lumbar motion and target impact location during Fishality were more similar to those during Reachality than to those during Dodgeality, which suggests that this new virtual reality game is an effective tool for shaping movement. These findings are encouraging for future research aimed at developing an individualized and graded virtual reality intervention for patients with low back pain and a high fear of movement.

Keywords: biomechanics; exergaming; gamification; intervention; lumbar; movement; physiotherapy; reaching; rehabilitation; serious games; virtual reality.

Figure 1

Diagram of the physics equations used in the different VR games. Four target contact locations (shown in green) are computed for each subject based on anthropometrics and a trunk flexion angle (θ) of 15°, 30°, 45°, and 60° and presented as a static target during Reachality. During Dodgeality, targets are launched with a constant initial velocity (vo), and the launch angle (α) is modified to ensure that the launch trajectory intercepts an intended target contact location. During Fishality, the launch velocity and angle are manipulated to ensure that the launch trajectory reaches a target height (H) and intercepts the intended target contact locations. VR: virtual reality.

Figure 2

Visual depiction of Reachality (A), Fishality (B), and Dodgeality (C) gameplay, the avatar that participants controlled (D), and the motion analysis data collected during the experiment (E).

Figure 3

Study outcomes compared across intended impact heights (IH1–IH4) and virtual reality games (Reachality, Fishality, and Dodgeality). Error bars represent 1 standard deviation. a: significant difference between Dodgeality and Fishality; b: significant difference between Dodgeality and Reachality; c: significant difference between Fishality and Reachality.