Setting, eligibility criteria and recruitment
Participants will be recruited from two Spanish universities in the province of Zaragoza (Spain). The participants will be students of the University of Zaragoza or the National Distance Education University (UNED), who will be recruited according to the following criteria: (a) ≥ 18 years of age; (b) enrolled in undergraduate or master’s degree courses at these universities; (c) enrolled in the faculties of social sciences or health sciences at the Zaragoza campus (in the case of University of Zaragoza students) or at the Calatayud campus (in the case of UNED students); (d) speaking and writing Spanish language; and (e) providing written informed consent before knowing the assignment group. No exclusion criteria will be considered to frame the work as a universal intervention for the promotion of health and stress management in university students.
Students will be recruited through the following three main procedures: (1) informative posters will be put up in university buildings containing a brief explanation of the study and the contact details; (2) several members of the university teaching staff will contact their students and send them an informative e-mail with a brief explanation of the study and the contact details; (3) different tuition centres, student organizations and support services for university students will also send their users an informative e-mail with a brief explanation of the study and the corresponding contact details.
Sample size
The sample size has been estimated taking as a reference a moderate difference (d = 0.40) between the total MBI group and the active control group in the main variable of perceived stress (Perceived Stress Questionnaire - PSS), as it has been observed in previous similar studies (29), accepting an alpha of 0.05 and a beta risk of 0.20 in a bilateral contrast, with a 2:1 allocation rate (so that two participants will be assigned to the mindfulness total group ‒ one in the condition of VR support and another in the condition that does not have such support ‒ for each control subject). This unequal randomization ratio design was established to gain additional exploratory information on the use of VR when added to the mindfulness treatment proposed (30, 31). Therefore, and under these circumstances, 225 participants will be required: 75 participants in the active control group of relaxation therapy, and 150 participants in the total MBI group. Taking into account all the participants, and assuming an attrition rate of around 25% (21), the total sample has been established at 280 university students.
Randomization, allocation and masking of study groups
The assignation of the subjects will be carried out after the baseline evaluation by a member of the research group, who has no knowledge about the study aims, through a simple random sequence generated by computer. Participants will be randomized into three groups ('mindfulness', 'mindfulness + VR', and 'relaxation'), considering the secondary objective of evaluating the effectiveness of each mindfulness subgroup separately. Nevertheless, taking into account the primary aim of isolating the common effects of both mindfulness subgroups, 'mindfulness' and 'mindfulness + VR' will work as an only one group of treatment. Posttreatment and 6-month follow-up measurements will be recorded by a different and independent assessor who will be blind to the study condition of participants. However, due to the characteristics of the intervention, providers and participants will be able to know what kind of intervention they are offering and receiving respectively, so this will be a single-blind study although all the participants will be treated through suitable active conditions. The interventions will be conducted in two cohorts: cohort 1 will be started in November 2018 and cohort 2 in April 2019 (Figure 2).
Interventions
Within the three conditions described below, motivational techniques will be used to increase adherence of participants to sessions and daily tasks. The participants will not receive any amount of economic compensation. In order to improve the adherence of participants to sessions, they will be sent reminders via WhatsApp of the start time of the session, a few hours prior to each session. Daily tasks aids, such as mindfulness or relaxation exercise audio recordings (depending on the condition) will also be sent through a weekly WhatsApp message in each group. All participants will be encouraged to do a daily task at home and will be asked to keep a daily written record of the number and duration of practice sessions.
-Mindfulness condition: In this group, n=93 students will be participating in a programme structured around two central elements, mindfulness and compassion. The programme consists of 90-minute group sessions, held once a week over a space of 6 weeks, and is offered as an extra-curricular activity. Each sub-group of mindfulness condition will be formed by 15 or 16 participants, so there will be six sub-groups of this condition. The sessions combine components of theory and practice, using a methodology which prioritizes inquiry, reflection and debate among the group of students. The theory component will comprise mindfulness and compassion concepts set out over the length of the programme, in addition to others related to time management, stress, anxiety and the balance between personal life and academic/work life. The practical component of the program will consist of formal and informal mindfulness and self-compassion exercises under the guidance of a psychologist, who will be specifically trained in the application of the theoretical and practical components of the mindfulness programme to university students. Table 1 provides a summary of the structure and contents of the mindfulness programme by session.
-Mindfulness condition complemented by VR environments: This condition (n=93) is equivalent to the previous one (group sessions formed by 15 or 16 participants, held once a week over a space of 6 weeks and offered as an extra-curricular activity), except the duration of each session, which is reduced from 90 to 75 minutes. Consequently, the content of each mindfulness session is slightly shortened, but this time difference is replaced by an individual VR session held before or after the group mindfulness session. This short VR session is designed on an individual basis and consists of a mindfulness or self-compassion exercise through VR. The VR kit comprises a set of Samsung GEAR VR goggles, a Samsung Galaxy S6 phone and optional headphones (https://www.psious.com/). The content and objectives of each VR scenario are specified in Table 2. In this condition, the mindfulness instructor will remain the same but the implementation of VR will be carried out by another psychologist who is specifically trained in the application of VR scenarios. Prior to the use of VR, the psychologist will check the participants’ health. In general terms, the use of VR is not recommended for pregnant women; people suffering from hypertension, ear infections, epilepsy or vertigo; patients who have recently undergone surgery; or people who suffer from cardiovascular disease, psychosis or serious mental illness. For this reason, the psychologist will ensure that none of these conditions are present. The participant will be seated on a chair in a quiet and secure place.
-Relaxation condition: This group (n=94) is based on the adapted version proposed by Bernstein and Borkovec of progressive muscle relaxation therapy (32). This programme also consists of 90-minute group sessions formed by 15 or 16 participants, held once a week over a space of 6 weeks, and offered as an extra-curricular activity. The relaxation programme involves the training of 16 muscle groups during the initial sessions, 7 muscle groups during the intermediate sessions, 4 muscle groups in the later stage, and only relaxation through recall in the final session. This programme is also complemented with visualizations, as it was originally proposed by Jacobson (33). Table 3 provides a summary of the programme structure and contents. All participants in this condition will be instructed by a psychologist specifically trained in the application of progressive muscle relaxation therapy.
Outcomes and data collection
Several outcomes will be measured and compared between the mindfulness (with and without VR support) and control groups. We will collect data on socio-demographic variables and experience in the use of new technologies (baseline), as well as on the primary and secondary outcomes (baseline, posttreatment and six month follow-up). The study primary outcome will be perceived stress. Secondary outcomes will be psychological wellbeing (e.g., anxiety, affectivity and emotional regulation), academic functioning (e.g., academic engagement, academic burnout and burnout clinical subtypes) and trait mindfulness and self-compassion. Other secondary VR measurements will be emotional intensity, state of mindfulness and sense of presence. All the measurements will be collected using paper-and-pencil surveys. In the last session, participants of all the groups will be called for a meeting to complete the posttreatment questionnaires. Participants who did not attend this meeting will be sent a message via WhatsApp, in which they will be informed about two possible dates to complete the questionnaires. Six months later, a similar strategy to complete the follow-up-intervention questionnaires will be used (all the students will be called for a meeting via WhatsApp). Participants who did not attend this meeting will be sent a new message with two possible dates. No outcome data will be collected for participants out of the referred meetings.
-Socio-demographic, use of new technologies and acceptability
Demographic Information will be gathered by means of a questionnaire developed by our team, including: 1) date of birth; 2) sex; 3) marital status; 4) educational level; 5) employment status; 6) type of employment contract; 7) previous training in: a) relaxation, b) mindfulness and/or compassion, and c) VR (with two answer options for each one: yes, no); and 8) degree course currently being taken.
Experience on the use of new technologies will be measured with a brief version of the Independent Television Company SOP Inventory (ITC-SOPI) (34, 35). Preliminary analyses have demonstrated adequate psychometrics for the ITC-SOPI (34).
Acceptability of the intervention will be measured using an adapted version of the Credibility/Expectancy Questionnaire (36, 37). Participants will complete this scale before and after the intervention. The adapted questionnaire comprises 6 items rated from 0 (‘not at all’) to 10 (‘very much’), with appropriate psychometric properties (34, 35).
-Primary Outcome
Perceived stress, measured by the 10-item self-report PSS (38, 39), will be the primary outcome. Through the PSS, participants will be asked to rate how unpredictable, uncontrollable and overloaded they have found their life over the past month on a 5-point Likert-type scale (from 0 = ‘never’ to 4 = ‘very often’). Higher scores indicate higher levels of stress. The Spanish PSS (40) provides a reliable and valid measure of perceived stress, with adequate psychometric properties (α = 0.82, test-retest, r = 0.77).
-Secondary Outcomes
Anxiety will be measured by the State-Trait Anxiety Inventory (STAI) (41-43). This inventory is a widely used, validated measure of anxiety and consists of 20 statements that evaluate how the participants feel at the present moment (state anxiety), and 20 statements that evaluate how the participants feel in general (trait anxiety). It uses a 0–3 Likert-type rating scale ‒ the higher scores indicating greater anxiety levels ‒ with good psychometric properties (α = 0.93) in university students (44).
Positive and negative affect will be measured by the Positive and Negative Affect Schedule (PANAS) (45). This self-report questionnaire consists of two 10-item scales to measure both positive and negative affect. Each item is rated on a 5-point Likert-type scale from 1 (‘not at all’) to 5 (‘very much’). Scores for each subscale are totalled, with higher scores indicating more positive or negative affect. The internal consistencies of the Spanish version of the positive and negative PANAS scales are 0.87 and 0.91, respectively (46).
Emotional regulation will be measured by the Emotional Regulation Questionnaire (ERQ) (47). This 10-item scale has been designed to measure respondents’ tendency to regulate their emotions in two ways: cognitive reappraisal (6 items) and expressive suppression (4 items). Participants respond using a 7-point Likert scale (from 1 = ‘strongly disagree’, to 7= ‘strongly agree’) where higher scores indicate increased use of regulatory strategies. The Spanish version of the ERQ shows adequate internal consistency, test-retest reliability, and convergent and discriminant validity, with reliability values of α = 0.79 for reappraisal and α = 0.75 for suppression (48).
Mindfulness will be measured by the Five Facet Mindfulness Questionnaire (FFMQ) (49). This 39-item self-report consists of five facets: observing (8 items), describing (8 items), acting with awareness (8 items), non-judging of inner experience (8-items) and non-reactivity to inner experience (7 items). Respondents indicate on a 5-point Likert-type scale the degree to which each item is generally true for them, ranging from 1 (‘never or very rarely true’) to 5 (‘very often or always true’). Higher scores indicate higher levels of mindfulness. The facet scores can be combined to produce a total mindfulness score. The Spanish version of this scale (50) shows good internal consistency (Cronbach’s alpha between 0.80 and 0.91).
Self-compassion will be measured by the Self-Compassion Scale (SCS) (51) Spanish version (52). This 26-item questionnaire assesses the components of self-compassion across three facets: self-kindness, common humanity and mindfulness. Higher scores indicate greater levels of self-compassion. The facet scores can be combined to produce a total self-compassion score (53). The SCS uses a 5-point Likert scale, ranging from 1 (‘almost never’) to 5 (‘almost always’). The Spanish version has demonstrated to be a valid and reliable instrument (α = 0.87; test-retest = 0.92).
Academic engagement will be measured by the Utrecht Work Engagement Survey Scale-Students (UWES-S) (54). This 17-item questionnaire includes three subscales: vigour (6 items), dedication (5 items) and absorption (6 items). The UWES is scored on a seven-point frequency Likert-type rating scale varying from 0 (‘never’) to 6 (‘always’). Higher scores in each subscale indicate greater levels of engagement. The UWES Spanish version has demonstrated to have good psychometric properties (54).
Academic burnout will be measured by the Maslach Burnout Inventory Student Survey (MBI-SS) (54), where the references to work have been changed for references to study. This 15-item questionnaire includes three subscales: exhaustion (5 items), cynicism (4 items) and efficacy (6 items). High scores in exhaustion and cynicism, and low scores in efficacy are indicative of burnout. Participants have to respond on a Likert-type scale with 7 options ranged from 0 (‘never’) to 6 (‘always’). The psychometric properties of the MBI-SS Spanish validation have been observed to be adequate (54).
Burnout clinical subtypes will be measured by the 36-item Burnout Clinical Subtype Questionnaire (BCSQ-36) (55), which assesses the three following burnout subtypes: frenetic, under-challenged and worn-out. References to work will be changed to references to study. Each subscale comprises 4 items. Participants indicate the point to which they agree with each item using a Likert-type scale with 7 options ranging from 1 (‘never’) to 7 (‘always’). Higher scores mean greater levels of burnout subtypes. The BCSQ-36 has demonstrated good psychometrics in the Spanish population (55).
-Virtual Reality measurements
A visual analogue scale (VAS) (56) will assess the intensity of different emotions before and after the VR intervention. A briefer version of the original measure (16 emotion items) will be used and it will be composed by 7 emotion items (happiness, sadness, anger, surprise, anxiety, relaxation/calm, vigour/energy). Participants can choose responses ranging from 1 (‘not feeling the emotion at all’) to 7 (‘feeling the emotion extremely’). This 7-item scale has been successfully used in previous studies (19, 57).
State of mindfulness will be measured just before and after each VR session using an adaptation of the Mindful Attention Awareness Scale (MAAS-State) (58-60). Item numbers 3, 8, 10, 13 and 14, has been designated to assess state mindfulness (58), and will be included. Each item will be rated on a 7-point Likert-type scale, between 0 (‘not at all’) and 6 (‘very much’). Higher scores reflect greater levels of state mindfulness. This shorter 5-item scale has been successfully used in previous studies (19).
Sense of presence in the VR scenarios will be measured just before and after each VR session, using 4 items with a 7-point Likert scale ranging from 1 to 7. These 4 items were adapted from the Slater, Usoh & Steed Questionnaire (SUS) (61). The Spanish version has been used to measure sense of presence provided by VR, with α = 0.92 (19).
Data monitoring and harms
The Data Monitoring Committee (DMC) of this trial is a group composed of five experts: the person in charge of the project, the head of the group, an independent clinical psychologist, an independent psychiatrist and the psychologist responsible of data handling. The DMC will meet three times throughout the trial (just before starting, in the midst of the intervention and at the end of the trial), in order to evaluate the possible cases of participants presenting adverse effects and to ensure that the trial is carrying out correctly. The practice of relaxation may have associated paradoxical increases in tension, relaxation-induced anxiety and panic and decreases in sympathetic tone which can lead to nausea and vomiting (62). Adverse effects of meditation can include psychological effects such as emotional stress, confusion, disorientation and dependence on practice; psychopathological effects such as anxiety, deliriums and hallucinations; and/or physiological effects such as pain, sensorial dysfunction, exacerbation of neuromuscular/joint diseases, reduction in appetite and insomnia (63). In a recent multi-cultural study on the effects of meditation, 25.4% of the participants reported unwanted effects, but most of them were transitory with no need for medical assistance (64). On the other hand, VR is associated with uncomfortable physical responses such as nausea (burping, increased salivation, etc.), oculomotor symptoms (eyestrain, difficulty concentrating, etc.) and disorientation (dizziness, vertigo, etc.) (65-67). Any adverse effect which is referred from the participants during the sessions, along the trial or observed by the psychologists in charge of the groups will be communicated to the DMC. The DMC will have the responsibility of evaluating and dealing with such possible side effects, bringing it to the attention of the corresponding health professional, in case it is required. The DMC will recommend changing the group or terminating the programme if the unintended effects observed or reported are a consequence of the participation in the intervention and acquire clinical relevance from the point of view of the corresponding health professional. The DMC will function independently of the trial funder.
Data analysis
To report this randomized controlled trial (RCT) we will make use of the recommended CONSORT guidelines (68, 69). Paper-based data entry will be double-checked revising possible out-of-range values. Socio-demographic data will be described at baseline by means of frequencies (percentages), medians (inter-quartile range) and means (SD), depending on the distribution of each variable. The treatment conditions will be compared to evaluate the success of randomization by means of chi-square (or Fisher when necessary), Kruskal-Wallis and one-way ANOVA tests, respectively. This strategy will also be used to analyse possible differences between groups in terms of preference for treatment conditions and credibility as a result of the treatment received.
-Main analysis
The efficacy of the general mindfulness programme ‒ integrating both groups with and without the presence of VR support ‒ compared to the relaxation control condition will be analysed according to the main PSS variable at posttreatment, which will be taken as a continuous outcome. It will be developed by means of a repeated measures design and an intention to treat (ITT) basis, by using multilevel mixed-effects regression models including time as an independent variable and subjects and presence/absence of VR support as random effect variables. For this, the restricted maximum likelihood (REML) estimation will be used, which produces unbiased estimates in case of small or unbalanced sample sizes (70). Non-standardized slopes and 95% confidence intervals (95% CI) will be calculated by adjusting the time of the year in which the programme was undertaken (temporality), and those socio-demographic variables that show significant differences between groups at baseline. The ‘group x time’ interaction will be considered in order to study the specific trajectories of each group throughout the intervention, and to determine whether the possible differences between the general mindfulness group and the relaxation group remain consistent over time. Effect size, as the differences between groups at post-test and at 6-month follow-up, will be assessed using Cohen's d statistic, calculated by the combined standard deviation in the pre-test, thus weighing the differences between the corresponding marginal means (71). Effect sizes are defined as small, when d ≤0.2; medium, when d = 0.5; and large, when d ≥ 0.8 (72).
-Secondary analyses
The efficacy of the general mindfulness programme ‒ integrating both groups with and without the presence of VR support ‒ compared to the relaxation control group regarding to the psychological well-being, academic functioning, trait mindfulness and self-compassion outcomes, will be calculated following the same analytical strategy used for the main analyses. Moreover, the efficacy of each mindfulness subgroup (with or without VR support) vs the control group as regards the main and secondary variables will be estimated, although the mixed regression model in this case will only include subjects as a random effect variable. Per protocol analysis will be carried out, considering only those participants who attend at least ≥50% of the sessions (73, 74). Sensitivity analysis will be developed to assess the effects of missing data, which will be replaced by multiple imputations based on chained equations as long as there are <40% missing data in the corresponding outcomes and whether missing values are distributed at random (MAR) to ensure validity (75). Additionally, possible differences in emotional state, immersion in the VR environment and state of mindfulness in the mindfulness + VR subgroup will be explored through each of the VR environments, using mixed models.
Baseline socio-demographic and psychological characteristics and the fact of belonging to subgroups (with or without VR support) will be evaluated dependent on compliance with the programme. Participants will be considered to have completed the programme when they have attended at least half or more of the sessions (73, 74). The characteristics of the subgroups will be compared by means of the corresponding chi-square (or Fisher) tests, and the Student’s t-test for unpaired groups.
No interim analyses before ending the study will be developed, unless DMC decided to prematurely terminate the study. An alpha level of 0.05 will be set, using a two-tailed test. The probability values relative to the main analysis will be adjusted according to Benjamini-Hochberg’s correction for multiple comparisons. The secondary analyses will be considered as exploratory and thus they will not be corrected (76).