Effektivitet og mekanismer for Virtual Reality-behandling av Phantom Leg Pain

Subjects assigned to the Active VR treatment began by selecting an avatar, with features such as gender and skin color that could be chosen according to preference. During treatment, they participated in a variety of games and activities developed and used by our teams including Kick, Dog Food, Quest for Fire, Chess, Checkers, Sudoku, and surfing the internet. Subjects had substantial flexibility to select games according to their interests but were required to spend at least 30 minutes in each session in games that required forceful, large-amplitude movements of the amputated lower limb (e.g., Quest for Fire, Kick).

8 twice-weekly interventions of virtual reality treatment focused on reducing phantom limb pain.

Subjects assigned to the Distractor VR treatment participated in the REAL i-Series immersive VR experience (REAL system), which had been demonstrated to reduce pain in several studies but lacked the hypothesized "active ingredients" of our Active VR treatment (visual and auditory feedback of movement of an extrapolated amputated limb). Subjects navigated through pleasant and relaxing VR environments; they did not see any rendering of their body and made no movements with their legs.

8 twice-weekly interventions of virtual reality treatment focused on reducing phantom limb pain.

Visual Analogue Scale (VAS) of McGill Short Form Questionnaire (Melzack, 1987); 0 minimum score - 10 maximum score; higher scores indicate higher level of pain (worse outcome)

McGill Short Form Questionnaire (Melzack, 1987): a series of 4-point scales (0 none - 3 severe) assessing the intensity of 15 qualitative characteristics of pain (e.g., throbbing, shooting, cramping, etc.), where higher scores are worse (more severe). For every participant we computed the total pain score (maximum 45 points indicating severe pain). Then, we computed the difference in the total scores between time points (e.g. post treatment vs baseline). Therefore, negative scores indicate improvement in pain qualitative characteristics.

Difference between 8th and 1st treatment session on scaled score measuring "average pain intensity since the previous intervention" on an 11-point numerical rating pain scale; (Scale = 0 - minimum score/no pain to 10 maximum score/pain as bad as participants can imagine); higher scores indicate higher level of pain (worse outcome).

The Frenchay Activities Index (FAI) (Holbrook et al., 1983):1983): a scale measuring the physical function and daily activity associated with quality of life in people with amputation. Scale = 0 minimum score to 45 maximum score; higher scores indicate more daily activities (better outcome). We report the average difference between post-intervention (after 4 weeks) and pre-intervention.

The 12-Item Short Form Health Survey (SF-12) (Ware et al., 1996): a measure of quality of life and functional capacity that has been validated in patients with amputation and phantom limb pain. Scale = 0 minimum score to 100 maximum score. Higher scores indicate poorer physical and mental health (worse outcome).

The Pain Interference Scale from the Brief Pain Inventory (BPI) (Cleeland & Ryan, 1994): A scale measuring the degree to which pain interferes with daily activities using a 0-10 numeric rating. Scale = 0 minimum score to 10 maximum score. Higher scores indicate greater pain interference (worse outcome). We report the average difference between post-intervention (after 4 weeks) and pre-intervention.

The Hospital Anxiety and Depression Scale (HADS) (Zigmond & Snaith, 1983): This is a 14- item measure with subscales assessing depressive and anxiety symptoms. For each subscale. range = 0 minimum score to 42 maximum score. Higher scores indicate more depression and/or anxiety (worse outcome). We report average differences on the anxiety scale and the depression scales between post-intervention (after 4 weeks) and pre-intervention.

The 13-item Pain Catastrophizing Scale (Sullivan et al., 1995): this scale investigates pain catastrophizing, which has been associated with pain severity and disability after amputation. Score range = 0-52; Higher scores indicate higher tendency to catastrophize (worse outcome). We report the average difference between post-intervention (after 4 weeks) and pre-intervention.

Insomnia Severity Index (Bastien et al., 2001): a 7-point scale that measures insomnia, a symptom associated with phantom limb pain. Score range = 0-28; higher scores indicate more insomnia (worse outcome). We report the average difference between post-intervention (after 4 weeks) and pre-intervention.

Technology Acceptance Scale (Morris et al., 1997): this is a 7-point scale that measures level of acceptance of the VR system. 0 minimum score - 133 maximum score; higher scores indicate low level of acceptance (worse score).

System Usability Scale (SUS) (Brooke, 1996); Scale measuring the usability of the each game; minimum score 10 - maximum score 50. Higher scores indicate less usability (worse outcome). For the distractor treatment, participants rated the overall experience; for the active treatment, the average of the two games was computed for each participants.

Simulator Sickness Questionnaire - SSQ (Kennedy et al., 1993): 4 points (0 - none; 3 - severe) scale assessing the severity of 16 cybersickness motion sickness symptoms, like nausea or dizziness. We computed the average score across items.

Brief Slater-Usoh-Steed Presence Questionnaire (Usoh et al., 2000; Slater et al., 1998; 1994). 0 minimum score - 42 maximum score. Higher scores indicate more presence in the VR (better outcome

The Modified Limb Deficiency and Phantom Limb Questionnaire (Goller et al., 2013): a questionnaire and visual analog scale assessing prosthesis usage, phantom limb experiences, and pain severity. The outcome measure used was the score on the visual analogue scale assessing pain severity at its worst (Scale = 0 no pain to 10 unbearable pain). We report the average score difference between post-intervention (after 4 weeks) and pre-intervention.

Visual analogue scale that evaluates treatment satisfaction (Robinson et al., 2004; Smith et al., 2005); 0 minimum score - 10 maximum score; Higher scores indicate more treatment satisfaction (better outcome)