Evaluating value mediation in patients with chronic low-back pain using virtual reality: contributions for empirical research in Value Sensitive Design

This study is part of the PijnVRij study (ClinicalTrials.gov Identifier: NCT04042090). The PijnVRij study is a mixed-methods study to assess the effects of therapeutic VR on improving the quality of life in patients with non-specific chronic low-back pain. The quantitative results of the study will be reported separately. Here, we address the qualitative outcomes of the study. The study was conducted from January 2020 till January 2021. Participants were randomized at a pain clinic of a large teaching hospital (Rijnstate, Arnhem, the Netherlands). Approval was obtained under Dutch law by the medical ethical committee (CMO Arnhem-Nijmegen, NL70042.091.19) in compliance with the Declaration of Helsinki.

Study participants were adult patients suffering from chronic non-specific low-back pain not attributable to a recognizable, known specific pathology (e.g. infection, tumor, osteoporosis, lumbar spine fracture, or cauda equina syndrome). All participants provided informed consent before participation. Participants were included when they: (1) reported a pain score ≥ 4, (2) did not receive any invasive treatment for non-specific low-back pain in the last year, (3) are willing and able to comply to the study protocol. Participants meeting the following criteria were excluded from the study: (1) participant is included in an alternative trial to evaluate ways of treating pain, (2) participant suffers from severe anxiety, (3) participant has difficulties handling VR (participant has a delirium, dementia, seizure, epilepsy, severe visual/hearing impairment, or head or face is not intact), (4) participant has a high risk of Methicillin-Resistant Staphylococcus Aureus (MRSA).

We studied the VR application Reducept (Reducept, Leeuwarden, The Netherlands). Reducept is a novel psychological VR intervention for treating chronic pain. Reducept aims to teach patients that pain can be influenced and managed by changing how they think about their pain. In the game, the user travels with a spaceship through the body. The journey starts in the spinal cord and ends in the brain. Along the way, the user is educated about the mechanisms of pain. Reducept hereby makes use of the ‘explain pain theory’, which describes that understanding pain and influencing the cognitive, emotional, and behavioral processes related to this pain enables patients to lower their pain sensations [25]. In addition to the pain education, the user plays several serious games stimulating the visual, auditory, and proprioceptive senses to manage chronic pain through distraction and relaxation. The games incorporate the principles of cognitive behavioral therapy, mindfulness, and acceptance and commitment therapy. For example, in one game, the user has to shoot away little ‘insects’ in the body that represent his pain. According to the cognitive behavioral therapy, patients hereby gain control over their pain which can lower the pain experience. In another game, the user is asked to focus on the images visualized in front of him whilst hearing relaxing music. This results in relaxation and mindfulness, which are both beneficial for reduction of a pain sensation. Also, following the acceptance and commitment therapy, users are taught by a relaxing voice that they should accept their pain in their daily lives, which can again reduce the experience of pain [25, 26].

At the time of the study, the application was still being optimized but a fully working version was available for testing in an experimental controlled setting. Figure 1 portrays screenshots of the application. Reducept was played on the VR headset Oculus Go (Facebook, Inc.; Menlo Park, USA). Participants of our study were inquired to use the application each day at home during a four weeks intervention period. The use of VR and reported pain scores could be monitored through an online dashboard that provided insight into the usage data from a distance.

To comprehend what values patients consider important in the context of their pain, and how VR mediates the values of patients with chronic low-back pain, we interviewed each participant twice, either personally or by phone, using a value-oriented semi-structured interview [27]. The first interview intended to understand the current value framework of patients in the context of chronic pain: what values mattered, the definition of these values, and the effect of pain on each value in terms of norms and experiences. After this interview, participants received education on VR and used the VR application for four weeks. Education was provided to participants in their own homes by a researcher. The researcher explained all buttons and technical features of the VR headset and provided support to connect the headset to the internet. When the participant showed enough knowledge to use the headset autonomously, the researcher left. Participants also received a paper manual with instructions on using VR that repeated the instructions of the researcher. On days 7 and 14, participants received a call from the research team for support when needed. Also, a telephone helpdesk was constantly available when participants needed technical support. The second interview was held directly after four weeks and investigated technological value mediation. We aimed to construct a novel value framework including what values were important in the use of VR, the definition of each value, and patients’ experiences of each value via norms. Both interview guides consisted of open-ended questions and probes derived from literature on important values. Interview guidelines can be found in Appendix A.

Interviews were recorded and transcribed verbatim. An independent researcher with a background in industrial design and ethics (MS) analyzed all results of the interviews held before using VR separately from al interviews held after using VR. Data were categorized in patient experiences, these into norms, and norms into values. The analysis resulted in two value frameworks. The first value framework consisted of all values patients considered important in the context of their chronic pain before using VR. The second framework refers to the experience of patients after using VR. Value frameworks were then compared to understand how VR technology and this particular application mediates values. Based on this empirical work, the authors then assigned a normative claim to the value mediations. This normative claim is needed to make design decisions. Value mediations were either marked as ‘positive’ (or ‘neutral’), or ‘negative’. Negative value mediations or opportunities for better value mediations were consequently translated into recommendations for the design and implementation of this VR technology. The guidelines for thematic content analysis were followed to ensure study trustworthiness [28].

Twenty patients with an average age of 51 (27–71 years old) used VR at home. Seventeen were interviewed before using VR, twenty after using VR. Education levels varied between participants. Three were male, seventeen were female. Only three participants had used a VR headset before. Below, first, the interviews held before using VR are reported. These interviews resulted in six values important in the context of chronic pain: health, self-perception, safety, hope, autonomy, and social comfort. After reporting how these values are defined and expressed in norms and experiences through patient quotes, we report what values matter when VR is used. VR affected the six aforementioned values, and four additional ones were introduced: privacy, accessibility, sensory comfort, and spatial comfort. VR did not affect how people defined the values. Yet, norms related to the values and value experiences were greatly mediated by VR. Recommendations are provided per value to better consider value mediation in this VR design and implementation. Figure 2 visualizes a summary of the results. A more comprehensive overview of the results can be found in Appendix B.

Interviews held after participants made use of VR were translated into a novel value framework. Below, we present all values affected by the use of VR in terms of norms and experiences. The same values as discussed in the previous section were found, and four additional ones were introduced: privacy, accessibility, spatial comfort, and sensory comfort. A comparison of the empirical findings of the two value frameworks resulted in VR design and implementation recommendations at the end of each value.

In contrast to the other values, ‘health’ has been studied empirically often. Multiple studies have been conducted in the medical domain on the effectiveness of VR for health. Our study results are in line with those of Snelgrove, Edwards [29], showing chronic pain to be constantly affecting patients’ well-being. Interviews held after using VR suggest that VR can improve health. These findings comply with findings from studies on improving pain management skills by VR [22, 30‐32]. To even better deal with pain in real life, we recommended the provision of support to patients to translate virtual lessons to real life. Games that support changing real-life user behavior have been described as persuasive games [33]. Therapeutic VR could consider principles of persuasive game design to optimize healthy patient behavior in real life.

Several authors have speculated on the ability of VR to alter the self-perception of individuals [34, 35]. They stated that because VR can create ‘virtually real experiences’ [36], users of VR quickly identify themselves with the virtual arms or virtual heartbeat that they can hear [37‐39]. This illusion of embodiment could result in loss of self-identity by alienating users from their own body, and a loss of connection to the real world, which could negatively affect behavior in real life [40]. In the VR application we tested, this ability to change the way people phenomenologically experience their bodies is intentionally used to lower experiences of chronic pain. Before using VR, most participants expressed that chronic pain has altered the way they perceive themselves, which complies with prior research [41]. After using VR, several participants positively experienced a change in self-perception; most participants indicated to accept pain better as part of daily life and to comprehend the concept of pain better. Our findings did not correspond with the speculative concerns. It was in this context that Tack [42] posed the concern of VR providing a temporary illusion of a ‘stronger’ back. He questioned whether this would cause patients to act beyond their physical capacity and if the return to self-perception would lead to disappointment. Again, our findings showed differently. In leaving VR, our participants did not experience disappointment. Even better, some were able to apply lessons learned from the VR application in real life. By, for example, ‘shooting away’ their pain, patients could embody that ‘stronger’ back. This might result from cognitive-behavioral therapeutic principles, including acceptance and commitment therapy and mindfulness, underlying the VR game [25, 43] which is a promising tool for any therapeutic VR to increase patient empowerment. This value particularly shows that empirical research into values provides more positive results than speculation.

Participants commented that their experiences of chronic pain often negatively affected their feelings of physical safety. Previously, Horsfield [40] illustrated that the virtual world could enable people to experience greater safety by escaping from pain and mortality as these are part of our embodiment. We found that VR indeed allows for pain distraction and thereby temporarily increased feelings of safety in a virtual world. Horsfield continued that temporary feelings of safety in the virtual world could lead to alienation and disappointment in the real world. We have not found this to be true. Actually, virtual feelings of safety were embodied by three participants, that consequently felt safer in the real world. In contrast, virtual environments could also harm safety by causing adverse effects like motion sickness [44]. In our study, only a few patients experienced minor dizziness after using VR, as the VR application was designed to prevent this. We did find another negative effect of VR on safety though. Especially the first time, some participants felt insecure in the virtual environment due to its novelty and being disclosed from the real world. These empirical findings generate other insights than speculative claims and should be considered by any therapeutic VR team during design and implementation.

Horsfield [40] speculated that the virtual world is a place of hope. It can provide hope that daily life could be lived without pain. The author continued that this could lead to disappointment and disillusion. Our research provided nuance to this claim. In the study, the experience of VR itself did not directly mediate hope. It was the medical condition of patients that initially created a boundless hope that pain one day will disappear. VR became the object of hope. Bolstered by advertisements, media attention, and enthusiasm of the medical staff, this hope might have generated false beliefs and affected the actual impact of VR. Our findings are in line with a concern earlier on posed by Madary and Metzinger [35] called ‘therapeutic misconception’. They argued that VR, as a novel technology, generates high expectations when applied in healthcare. As these could result in desperation and disappointment, a shared decision-making process between patient and caregiver in which realistic expectations are discussed was recommended before prescribing therapeutic VR.

In literature, speculative concerns have been raised on the dependence of patients on VR [34]. We did not find a dependence on VR but only found that VR offers potential for self-management and increased feelings of autonomy. Nevertheless, we recommended that support is still necessary to allow any patient to make use of the technology. This recommendation complies with previous recommendations on support needed in ehealth use to foster autonomy [45].

Concerns in the literature addressing social relationships, including escapism from the real world [46], were not encountered in our study. Yet, we found that VR is not contributing to social relations either. We also identified an opportunity for VR to become more social. Previously, it has been concluded that VR can contribute to social relations and empathy [47, 48]. Within healthcare, VR can be used in this way by improving empathy for people with dementia in caregivers [49, 50] and by using the virtual environment as a ‘playground’ for hospitalized children [51]. These cases illustrate that social comfort is a value that could be improved by VR and might become incorporated in any therapeutic VR design.

Our findings contrast common literature regarding privacy as one of the most pressing ethical concerns of VR. Madary and Metzinger [35], O’Brolcháin, Jacquemard [52], and Spiegel [53], for example, speculated on the value of privacy. The authors feared for people’s privacy when eye movement is traced, data is being stolen or misused, and third parties obtain access to VR’s camera to record faces. Privacy was not seen as an issue by patients. Even though our patients did not consider these concerns, designers should make sure these items never become patient concerns. Therefore, we recommended VR developers to consider, amongst others, general data protection regulation guidelines, and the use of VR headsets designed for medical purposes. This way, privacy will not constrain data sharing between patient and caregiver for better support. This value shows that empirical research might not provide a full insight into all values of importance, as users might not always know what is best for them.

Access to VR is repeatedly speculated as an ethical concern [35, 40]. Our results correspond with this concern. We made VR available to patients by offering them VR headsets for free and providing education and technical support. Still, these aspects remained a barrier for accessibility as some desired continuing use of VR and encountered financial problems, and others pointed to technology as a barrier for future use. To achieve distributive justice in virtual care technology, there is a need for insight into the cost-effectiveness of VR therapy compared to traditional therapy so that medical insurance agencies are willing to reimburse. Furthermore, we recommended education and support services to make therapeutic VR accessible for all.

Technology development is oftentimes steered by the possibilities that technology offers. Yet, these possibilities do not always meet the needs of users. Therefore, Kellmeyer [34] argued that instead of opting for a technology-centered approach, VR development should follow a user-centered approach. Our study demonstrated the importance of this statement through the values of sensory and spatial comfort. We labeled these values design values because they do not directly relate to moral considerations. We encouraged both ethicists and designers to consider these values together with moral values to optimize the mediation of VR.

This study has a few limitations. First, as the VR experience was new to all but three participants, this novelty might have influenced, for example, positive outcomes related to health, negative experiences of safety, and high expectations and hope. Repetition of this study within some years, when VR is a well-known technology, will clarify whether the novelty of this technology has affected the study outcomes or whether outcomes are intrinsic to VR. Second, during the study, the Reducept application was updated three times with minor changes. The updates went together with several technological failures. These have affected values of sensory and spatial comfort. Ideally, studies into (VR) software should be conducted via only testing one version. Third, we studied participants with non-specific chronic low-back pain awaiting medical treatment. Our study results reflect their opinions and might differ from people suffering from different types of chronic pain, from patients not following a trajectory for another treatment, or from patients with a shorter history of pain, especially regarding the values of health and hope. Only with a longitudinal study in a larger patient population, we would be able to generalize our value results to chronic pain and VR. Fourth, as we applied semi-structured value-oriented interviews, we had identified through literature a set of values of which we believed pain and VR might affect. Although we provided the interviewees the opportunity to come up with topics themselves and to only continue with the topics they deemed most important, this technique of interviewing might have influenced the outcomes of the value identification process. Also, the normative process of translating empirical research into design and implementation recommendations remains a subjective endeavor. Nonetheless, to empirically identify values, and appreciate value experiences, no better methods exist at present. Fifth, we did not study the relative importance of one value over another even though this is part of a value framework [16, 20]. This relative importance is particularly important in the case of value tensions. When these occur, insight into users’ value prioritization could aid in solving the tensions [54]. Finally, this study has been conducted during the COVID-19 pandemic. This pandemic required people to remain at home and decreased their mobility. Several participants questioned whether their experiences were also affected by changes in social contact and stress. This observation stresses the importance of taking into account both micro and macro contextual factors when studying the value experiences of people.