In this study we assessed the feasibility of a VR-guided intervention seeking to improve physical fitness in individuals with chronic pain. In a randomised controlled trial with a cross-over design and participants in rehabilitation for chronic pain were asked to perform, in a counter-balanced order, five minutes of aerobic exercise following identical instructions given through either a VR headset or TV screen. The procedures were then repeated with mindfulness exercises. Heart rate (HR) was monitored throughout all four sessions and participants self-reported perceived exercise intensity, benefit, relaxation, and reward. Paired Student's t-test, Wilcoxon signed rank test and McNemar's test were performed to compare the outcome variables across sessions for individuals, as appropriate.

Participants: were recruited through partner patients' organizations, a local exercise group aimed at people living with rheumatological conditions, and an inpatient occupations rehabilitation centre.

The study was conducted according to the Helsinki declaration and approved by the Local committee for medical and health research ethics at the Inland Norway University of Applied Sciences (Ref. 20616405).

Study design The study was designed as a randomised controlled trial with a cross-over design.

Intervntion: Each participant underwent five exercise sessions in total. Firstly, all participants performed a six- minutes warm-up session following instructions projected on a TV screen. Subsequently, the participants performed two aerobic exercise sessions, identical in content and duration (five- minutes), once following the instructions from VR (VR-A) and once following the identical program that was displayed on a TV screen (TV-A). The allocation of the first aerobic exercise session as VR-A or TV-A was randomly assigned for each individual. After completing the first aerobic exercise session, the participants were subsequently asked to complete the second session guided through the opposite mode of delivery. After the exercise sessions the same randomised cross-over design was followed for two five- minute mindfulness sessions, with the participants undergoing, in a counter-balanced and random order, a guided mindfulness program, once delivered through VR (VR-M) and once displayed on a TV screen (TV-M).

Virtual environment and technology In the VR-A and VR-M, the participants wore a stand-alone HMD (Oculus Quest 2 with stock headstrap, Meta Platforms, Menlo Park, California, USA). The virtual environment was developed by Fynd Reality AS (Hamar, Norway), in line with general criteria outlined during the participatory process and consisted of a computer-generated replication of Hamar's town square (Norway). Each participant would enter this town square as a virtual avatar with virtual arms and body. The position of the arms was tracked by the hand-held controllers. A large screen would appear in the virtual town square, showing a video in which an instructor guided the exercises for the various sessions. This was the same video as shown on the TV screen.

Data collection at baseline Participant were assessed according to the Polysymptomatic Distress Scale (PDS). The Widespread Pain Index and Symptom Scale that together compose thePDS were completed by participants. The PDS is composed of variables used in the 2010 American College of Rheumatology (ACR) fibromyalgia criteria which were later modified for use in clinical research and self-evaluation in epidemiological surveys 17-19. The PSD is thus both a diagnostic aid and a measure of fibromyalgia severity. A diagnosis of fibromyalgia may be indicated by a PSD score of 12 or more in an epidemiological study1. The PDS has been translated and validated in Norwegian 20.

Outcome measures HR was registered throughout the sessions using a HR-monitors worn at the wrist (Garmin® Forerunner 55). The mean HRs for the last three minutes of each session were calculated for each individual. For the aerobic exercise sessions, HR was also categorised into HR-zones according to the ACSM recommendations based on the participants' predicted maximal HR (220 minus age) 4, with the proportion of time spent in " moderate-vigorous exercise intensity" (HR 65% of maximum heart rate or more) being calculated and used for further analysis.

Participants also completed questionnaires

Statistics We compared the mean HR values recorded within the last three minutes of the warm-up vs. all other sessions, as well between each aerobic exercise (VR-A vs. TV-A) and mindfulness sessions (VR-M vs. TV-M), using paired Student's t-test.

The time spent in each HR category was compared using the paired Wilcoxon signed rank test and time in the "moderate-vigorous" vs "low" ACSM zone in the VR-A vs. TV-A sessions using McNemar's test for dichotomised variables.

Comparisons between the self-reported measurements collected after the aerobic exercise (VR-A vs. TV-A) and mindfulness sessions (VR-M vs. TV-M) were performed using the paired Wilcoxon signed rank test for each session for ordinal data and McNemars test for dichotomised variables.

Inclusion Criteria:

• The inclusion criteria were adults (>18 years of age) with chronic musculo-skeletal pain.

Exclusion Criteria:

• The exclusion criteria were inability to perform VR-delivered physical exercise

We have not described sharing data from this small trial in the application to the ethical committe. If a relevant party was to ask for individual level data we would consider the application and consult the ethical committee.

No masking was possible