The study is a proof-of-concept designed to assess whether most CP patients are capable of 12-minute sessions, and if one session has measurable results that are comparable to prior studies with healthy people (i.e., are patients capable of focusing inward, and can they learn to change their GSR?).
A study was designed to test the minimal effectiveness of the Virtual Meditative Walk system. In the long term, the system is designed to be used over six sessions, which is an introduction parallel to learning MBSR in more traditional ways, such as FTF group lessons and recorded training sessions. For this VR approach, we wanted to determine if an immersive Virtual Environment, combined with MBSR training and biofeedback, helps pain patients better manage and self-modulate their pain by reducing perceived pain levels among chronic pain patients in the short term, using the minimal possible time period of one session.
4.2 Participants and Procedures
Thirteen patients ranging from 35 to 55 years of age (mean = 49, SD = 8.2) participated in the study in an established pain clinic. Each patient had a diagnosis of chronic pain. Six participants (3 male, 3 female) were randomly assigned to the control group, and the other seven (3 male, 4 female) were assigned to the VR group. Although 20 participants were originally recruited, 7 of them either chose not to complete the study because of pain, or failed to finish our questionnaires.
As for the patients’ type of pain, it is crucial to understand that CP is considered to be a dysfunction of the pain response system. Therefore, categorization is often deemed to be counterproductive, since it draws researchers back into habitual ways of confusing the pain system dysfunction with acute pain; acute pain is the common understanding of pain that results from injury or infection, and functions in the short term as an alert to danger, injury or threat. While subcategories of pain type are used by a subset of health professionals and researchers who focus on CP, they are often categories of neuropathic, nociceptive and idiopathic pain. Also, while patients may cite the source of pain (such as in their lower back, legs, hips, neck, and shoulders), pain can be “referred”. Moreover, chronic pain, as a systemic dysfunction, often leads to complex and distributed pain. For our readership, therefore, this complexity is beyond the scope of the paper (or page limits). More importantly, based on our prior publications in ACM fields, stating “which” category almost always derails the point of the research findings because readers return to habitual ways of thinking that CP is “just” acute pain, that it behaves in known ways, and that it can be cured.
During meditation sessions, we monitored GSR levels of patients and used this data to drive the dynamics of the VE in real-time. This real-time biofeedback system allowed patients to become aware of their progress as they performed mindfulness and encouraged them to pursue the practice. During the session, the patient saw a foggy forest, with the fog representing the patient’s GSR level. As patients intentionally reduced their stress level inferred from GSR data, the fog faded, and indicating that the patient was approaching or in a meditative state. The fog indicated the cause-and-effect mechanism of biofeedback in the VE. Based on early design tests and on one of our former studies , the fog animation was designed with abstraction in mind. The fog aimed to distribute the attention of the user while displaying the changes of GSR in real-time.
The participant was first informed of the whole study and procedures, and then the GSR sensors were attached to two fingers. In the control group, participants listened to the MBSR training audio. In the VR group, participants listened to the same MBSR training audio while immersed in the VMW environment. Both groups were given 12 min for the MBSR training (shown in Fig. 3).
The construction of the physical setup for the VMW required the use of a stereoscopic VR display. The display is mounted on a movable arm to ensure flexibility and to maximize patient comfort. The GSR sensors, which are small clips, were gently put onto two of the patient’s fingertips; none of the participants reported discomfort from their use. GSR data was used to control the biofeedback system in VR, but it was not adopted to compare pain levels. The assumption is: to immerse patients in the VE so that their GSR data were not displayed as a 2D graph, but rather had a major influence on the VE that is consistent with VR’s immersive capabilities. Consequently, the raw GSR data was not used for comparing two groups.
A pain questionnaire was given to the patients both before and after the study session in order to compare perceived reported pain levels. The 11-point Numerical Rating Scale (NRS) was used as the instrument for patients to self-report their pain levels between the numerical values 0 and 10 (10 means the worst pain possible and 0 means no pain at all). The NRS instrument is standard in pain research in the health domain. For patients, NRS is simple in terms of understanding and ease of use; therefore, it was used to avoid distressing the pain patients with complex and lengthy questionnaires, which are commonly and repeatedly used in pain clinics.
4.5 Results and Analysis
A two-way mixed ANOVA was run to analyze the collected data. Time and condition were two independent variables. Time was the within-subjects factor, and the subjects design was between-subjects design – a participant either belonged to the VR group or to the control group.
We found a significant main effect of Time, F(1, 11) = 10.44, p < .01. The main effect of Condition was not significant, F(1, 11) = 1.53, p > .05. This indicated that when the time at which the NRS was measured is ignored, the initial pain level in the VR group was not significantly different than that in the control group. There was a significant Time x Condition interaction, F(1, 11) = 8.16, p < .05, indicating that the changes in the pain level in the VR group were significantly different compared to the change in the control group. Specifically, there was a significant drop in NRS ratings in the VR group, t(6) = 2.86, p < .05, but a very weak drop in the control group, t(5) = 1.24, p > .05. These findings indicate that the VMW (VR paired with biofeedback for MBSR training) was significantly more effective than MBSR alone at reducing reported pain levels among participants.
In the clinical pain clinic settings, we taught participants to learn a basic level of MSBR in the study with our VR system. We also imported VMW to mobile terminals as a software application for reinforcement for regular practice outside of the clinic, and to capture and track adherence to regular MBSR practice. Therefore, in our next phase, we plan to strengthen patients’ self-care and management skills: (1) through VR therapy by providing six VR sessions in the doctors’ clinic; and (2) providing mobile APP training so they can keep learning and practicing pain self-management in the same context.
Although the single trial outlined does not speak to the effectiveness of potential long-term capabilities for VR chronic pain self-management, the VMW enables chronic pain patients to consider that their pain experiences could be further managed through MBSR practiced over the long-term. By multiple training sessions and regularized practice, patients can learn to more easily situate the psychophysical mediation of their internal experiences into everyday life. The pain reduction reported by the NRS data is an early step in proving that VR and biofeedback systems may be an effective first step in promoting this behavioural change, and potentially to afford patients with a greater sense that they are able to self-manage their pain to some degree. This is an important factor since many CP patients report a sense of hopelessness .
For the past two decades, research on VR for pain has been focusing on interventions for acute pain. However, our focus is on how to utilize VR as a technology that could also benefit patients who live with long term chronic pain. By implementing an effective non-pharmacological analgesic approach—MBSR meditation—we believe the VR is more suitable for chronic pain patients and that such pain self-modulation may provide a more appropriate approach useful over the long term than short term pain distraction. Our immersive VR project, the Virtual Meditative Walk, and the study results demonstrate its potential positive effect.
Future studies with longer immersion times and a focus on how long the analgesic effect may linger after the meditative session is the natural next step in continuing this line of inquiry. The introduction of more detailed reporting methods of perceived pain, such as the use of the McGill Pain Questionnaire, could also yield new insights into the details surrounding perceived reported pain after the VR intervention. This will require greater effort put towards the understanding of chronic pain patient experience, studied within the context of the clinic to ensure that patients’ comfort and stamina are not negatively impacted by the time and effort that would be required if patients needed to travel elsewhere, such as to a research lab.