The Effectiveness of a Neurofeedback-assisted Mindfulness Training using Mobile Application on Stress Reduction in Employees: A Randomized Controlled Trial

Background: Mindfulness-based training programs are known to be effective in reducing stress levels. Objective: This study examines the effectiveness of a 4-week neurofeedback-assisted mindfulness-based training program delivered via a mobile application for employees. Methods: Participants (n = 92) with a full-time job were assigned to one of three groups: Group 1 (n = 29) receiving mobile neurofeedback-assisted mindfulness training; group 2 (n = 32) receiving mobile mindfulness training without neurofeedback function; and group 3 (n = 31) receiving self-learning with paper materials about stress management. The primary outcome measures were perceived stress scale and brief resilience scale. Heart rate variability and two-channel electroencephalography were utilized as physiological measurements. These measurements were collected at three different times; at baseline, posttraining, and one-month follow-up. Generalized estimating equation was used for the analysis. Results: In regard to the resilience scale, a significant interaction (time × group) was observed. In the post-hoc analysis, a significant difference between groups 1 and 3 was found at the post-training assessment, and the significant change remained at 1-month follow-up assessment in group 1. As for physiological measurements, relaxation index showed a significant interaction (time × group difference), and group 1 showed the highest improvement compared to group 2 and group 3 at 1-month follow-up assessment. Conclusions: During the 4-week mindfulness-based training program delivered via mobile application, group 1 with neurofeedback-assistance showed better outcomes on resilience and relaxation compared to the other groups. Further research with a larger sample and a longer follow-up period is needed for generalizing our results. Clinical Trial: NCT 03787407 (JMIR Preprints 21/09/2022:42851) DOI: https://doi.org/10.2196/preprints.42851 Preprint Settings https://preprints.jmir.org/preprint/42851 [unpublished, non-peer-reviewed preprint] JMIR Preprints Min et al 1) Would you like to publish your submitted manuscript as preprint? Please make my preprint PDF available to anyone at any time (recommended). Please make my preprint PDF available only to logged-in users; I understand that my title and abstract will remain visible to all users. Only make the preprint title and abstract visible. No, I do not wish to publish my submitted manuscript as a preprint. 2) If accepted for publication in a JMIR journal, would you like the PDF to be visible to the public? Yes, please make my accepted manuscript PDF available to anyone at any time (Recommended). Yes, but please make my accepted manuscript PDF available only to logged-in users; I understand that the title and abstract will remain visible to all users (see Important note, above). I also understand that if I later pay to participate in <a href="https://jmir.zendesk.com/hc/en-us/articles/360008899632-What-is-the-PubMed-Now-ahead-of-print-option-when-I-pay-the-APF-" target="_blank">JMIR’s PubMed Now! service</a> service, my accepted manuscript PDF will automatically be made openly available. Yes, but only make the title and abstract visible (see Important note, above). I understand that if I later pay to participate in <a href="https://jmir.zendesk.com/hc/en-us/articles/360008899632-What-is-the-PubMed-Now-ahead-of-print-option-when-I-pay-the-APF-" target="_blank">JMIR’s PubMed Now! service</a> service, my accepted manuscript PDF will automatically be made openly available. https://preprints.jmir.org/preprint/42851 [unpublished, non-peer-reviewed preprint] JMIR Preprints Min et al


Table of Contents
1) Would you like to publish your submitted manuscript as preprint? Please make my preprint PDF available to anyone at any time (recommended). Please make my preprint PDF available only to logged-in users; I understand that my title and abstract will remain visible to all users. Only make the preprint title and abstract visible. No, I do not wish to publish my submitted manuscript as a preprint. 2) If accepted for publication in a JMIR journal, would you like the PDF to be visible to the public?
Yes, please make my accepted manuscript PDF available to anyone at any time (Recommended).
Yes, but please make my accepted manuscript PDF available only to logged-in users; I understand that the title and abstract will remain v Yes, but only make the title and abstract visible (see Important note, above). I understand that if I later pay to participate in <a href="http

Original Manuscript
Sohee Oh 0000-0002-3010-448X Je-Yeon Yun 0000-0002-5531-2410 Mindfulness-based training programs are known to be effective in reducing stress levels. Objective: This study examines the effectiveness of a 4-week neurofeedback-assisted mindfulness-based training program delivered via a mobile application for employees.
Methods: Participants (n = 92) with a full-time job were assigned to one of three groups: Group 1 (n = 29) receiving mobile neurofeedback-assisted mindfulness training; group 2 (n = 32) receiving mobile mindfulness training without neurofeedback function; and group 3 (n = 31) receiving selflearning with paper materials about stress management. The primary outcome measures were perceived stress scale and brief resilience scale. Heart rate variability and two-channel electroencephalography were utilized as physiological measurements. These measurements were collected at three different times; at baseline, post-training, and one-month follow-up. Generalized estimating equation was used for the analysis.
Results: In regard to the resilience scale, a significant interaction (time × group) was observed. In the post-hoc analysis, a significant difference between groups 1 and 3 was found at the post-training assessment, and the significant change remained at 1-month follow-up assessment in group 1. As for physiological measurements, relaxation index showed a significant interaction (time × group difference), and group 1 showed the highest improvement compared to group 2 and group 3 at 1month follow-up assessment.
Conclusion: During the 4-week mindfulness-based training program delivered via mobile application, group 1 with neurofeedback-assistance showed better outcomes on resilience and relaxation compared to the other groups. Further research with a larger sample and a longer followup period is needed for generalizing our results.

Introduction
Mental stress is a major public health concern, with high prevalence. Stress is suggested to negatively impact both the physical and mental health of individuals [1,2], ranging from cardiovascular mortality [3], musculoskeletal pain [4], type 2 diabetes [5], depression, and anxiety [6][7][8]. Work-related stress is associated with presenteeism, absenteeism, diminished productivity, and high employee turnovers [9][10][11]. According to an annual U.K. survey, stress, depression, or anxiety accounts for 44% of all work-related health issues, and 54% of all working days were lost due to a recent increase in poor health of employees [12]. Thus, the importance of stress management in employees can be seen at both the industrial and individual levels.
Recently, there has been an increase in the interest, and thereby increased research effort, surrounding the mindfulness-based training programs to reduce stress and improve mental wellness for employees [13]. Many organizations and corporations have adopted mindfulness-based programs to enhance their employees' well-being and performance [14]. Several meta-analyses focusing on healthy individuals suggested that mindfulness-based training programs were able to reduce perceived stress, anxiety, depression, and enhance quality of life, spirituality values, and resilience [15,16]. Mindfulness-based training programs also showed to have benefits for clinicians, nursing staffs, and health profession students with high stress levels and burnout rates [17][18][19]. Health benefits on medical professionals could lead also indirectly benefit patients by improving patient and clinical care [20]. However, despite these possible benefits, several issues, including training duration and delivery methods -and their varying effectiveness -have been addressed [17,21].
According to one qualitative study that reviewed 67 published studies [13], the most common method of delivery was in-person lectures (64, 95.5%), followed by audio recordings (39, 58.2%) and group discussions (34, 50.7%). Among these 67 studies, only four (6%) used online modules as a means of content delivery. However, with technological advancements and greater immersion of smartphones in daily life, there has recently been an increase in the interest of incorporating such technology into the field of mental healthcare [22]. With the advent of smartphone technology and mobile applications, health professionals are realizing their advantages and ways to enhance patient care. Advantages of mobile application-based approaches include increased accessibility, interactivity, and usability of various functions, such as treatment monitoring, appointment reminders, and recordkeeping [23,24]. These advantages could enhance patient's or employees' adherence to the mental health intervention program. A recent study suggested that mindfulness training using smartphone applications may provide immediate positive effects on mood and stress, as well as long-term benefits for attentional control [25].
In addition, in terms of training duration, a traditional 8-week mindfulness training program, like mindfulness-based stress reduction (MBSR), may be a barrier for a widespread organizational adoption and raises the question of whether a similar beneficial effect can be achieved with a shorter program. [13]. A previous study reported that even a 4-day mindfulness training program was effective in reducing anxiety and fatigue, as well as increasing mindfulness [26]. In a systematic review, the correlation between the mean effect size and number of in-class hours for MBSR training was not significant in clinical and nonclinical samples. This finding suggests that reduced -or shorter -class time may be worthwhile for those with time constraints or those with little to no motivation to participate [27].
Meanwhile, a recent meta-analysis concluded that the most consistent electroencephalography (EEG) findings associated with mindfulness training may be increased theta and alpha power [28]. Several previous studies evaluated the relationship between mindfulness-based training and neurofeedback. Neurofeedback is a training method to control brain waves consciously, and an EEG is used to record these waves. During training, subjects are instructed to focus or relax, and EEG is presented as a type of visual stimuli in real-time so that subjects can recognize their present state of brain activity. In a previous study, participants undergoing mindfulness meditation with alpha-neurofeedback demonstrated a higher alpha amplitude compared with their SHAM neurofeedback counterparts [29]. Other research suggest that neurofeedback could mediate the effect of mindfulness meditation [30,31].
The purpose of this study is to verify the effects of neurofeedback-assisted mindfulness training program delivered via a mobile application. We hypothesize that the neurofeedback-assisted mindfulness training might have augmenting effects on stress and related indices compared to mindfulness training only group. neurological disorders, such as epileptic disorders, stroke, brain tumors, or others; 4) current or previous history of psychosis, such as schizophrenia or bipolar I disorders; 5) current report of suicidal ideation; 6) other conditions that might influence the measurement of heart rate variability (HRV), such as cardiac or pulmonary disease; and 7) non-pharmacological psychiatric treatment, counseling or meditation training within the past 6 months. As mental stress at the workplace is commonly associated with depression, anxiety, and insomnia [6][7][8], participants with these conditions, but in stable status, were not excluded, as long as the type and dosage of medication remained the same for the past six months. In the screening process, the Mini International Neuropsychiatric Interview (MINI) [32], which is a short, structured psychiatric interview designed to find a wide range of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, and

Participants
International Classification of Diseases, Tenth Revision of psychiatric disorders, was used to diagnose psychiatric disorders. The MINI was translated into Korean and has good validity and reliability [33]. The interviews were conducted by two psychologists with a master's degree who were familiar with the tool. The estimated total sample size using G-Power was 90 (30 participants per group) with a predicted effect size of Cohen's d = 0.3, alpha level of .05, and a desired power of 0.7. Considering a drop-out rate of 10 %, we targeted to recruit 100 participants.

Assessments
Demographic information, including age, sex, marital status (unmarried, or others), educational status (more than college education or less), and length of career (more than 3 years or less), was obtained using self-reported questionnaires.
As In the original form as developed by Cohen et al. [34], the PSS is a 14-item scale; however, a modified version with a 10-item scale is commonly used [35]. The questionnaire is rated on a 5-point Likert scale; 0 (never), 1 (almost never), 2 (sometimes), 3 (fairly often), and 4 (very often). Higher scores reflect higher levels of perceived stress.
The BRS, composed of 6 items and measured on a 5-point Likert Scale (1: strongly disagree, to 5: strongly agree), was used to evaluate resilience of each individual and assessed the ability to bounce back or recover from stress [36]. Higher scores indicated better resilience.
The K-MAAS was originally developed by Brown & Ryan [37] and validated by Kwon et al. [38] for its application to Koreans. This 15-item scale focuses on the attention and awareness of mindful states, ranging from 1 (almost always) to 6 (almost never). Higher scores indicated higher levels of dispositional mindfulness.
The KELS was developed to measure emotional labor of employees and validated by Lee et al. [39] for its application to Koreans. Emotional labor was defined as the process by which employees have to control their feelings in accordance with the organizational demand and occupational role [40,41]. The KELS has 5 subscales: effort to control emotion (5 items), organizational monitoring system (4 items), demands of emotional labor (3 items), emotional damage (6 items), and organizational support system (7 items). Each item was rated on a 4-point Likert scale (1: not at all, to 4: very much) with higher scores reflecting higher levels of emotional labor.
The KOSS-Short Form consists of 24 items, and each item is rated on a 4-point Likert scale; (1: never to 4: always). It was validated by Chang et al. [42], and most commonly used and studied for the evaluation of job stress in South Korea. It is comprised of 7 subscales, including job demand (4 items), job control (4 items), interpersonal conflict (3 items), job insecurity (2 items), organizational system (4 items), lack of reward (3 items), and workplace environment (4 items). In this study, the sum of each subscale was calculated and then converted to 100 points. Higher scores reflected higher levels of job stress.
Insomnia was assessed by the AIS, which consisted of 8 items with a 4-point Likert scale, with higher scores reflecting greater severity of insomnia [43].
The PHQ-9 was used to evaluate depressive symptoms, and higher scores indicate higher level of depressive symptoms [44].
In addition, we measured the physiological variables, including HRV and EEG, which are widely used as stress-related physiological bio-signals [45,46]. These physiological variables were acquired with headsets named neuroNicle FX2 manufactured by Laxtha (Inc., Daejeon, South Korea). It has two EEG monopolar electrodes located in the prefrontal area (FP1, FP2) with the ground and reference electrodes on a clamp placed on the right earlobe, and one pulse wave sensor (PPG, plethysmograph) attached to the clamp. In a previous work using FX2 device, the authors introduced a reliability of prefrontal EEG marker from the device, and they reported that significant differences in the mean values of the tested variables were not observed between the prefrontal and occipital regions [47]. In our study, EEG and HRV were recorded for five minutes for all subjects in a resting state with eyes closed. Ocular, muscular, and other type of artifacts were detected by automatic removal algorithm, and contaminated period longer than 10 seconds were removed. The Hz) bands were calculated. All these variables were normalized by sex and age.
Throughout the whole study period, self-reported psychological and physiological indices were measured three times: at enrollment (baseline assessment), immediately after training (posttraining assessment), and 4 weeks after training (one-month follow-up assessment).

Randomization and Intervention Conditions
Participants were assigned to three groups by 1:1:1 block randomization (stratified by organization and age) with randomly selected block size (3, 6,  They also discussed any issues with the training and path forward. As for the control, or self-care group (group 3), self-learning paper materials with stress management were provided at the first visit, with no additional weekly meetings. The material included definition of stress, signs, and symptoms of stress, as well as how to manage stress.

Mobile application
The mindfulness-based training program consisted of awareness training (7 minutes), abdominal breathing (4 minutes), and body scan meditation (8 minutes) sessions with audio guide (Figure 1).
Each session was followed by a self-exercise time, for approximately 10 minutes. The order and frequency of each session were structured for the controlled research protocol, and participants were instructed to follow the schedule ( Table 1). Duration of total training in one session was 20 minutes, including self-exercise.

Neurofeedback function
For group 1, the neurofeedback function was supported with the Omnifit Brain (Omni C&S, Inc., Seoul, South Korea) during the self-exercise. The neurofeedback training included alpha protocol to enhance the power of alpha frequency compared to other frequency bands. The alpha wave is known to be associated with relaxation state [48], and alpha protocols are widely used neurofeedback methods reducing anxiety and stress, improving mental performance [49]. Participants of group 1 were instructed that a positive feedback sound (ringing of bell for one second) would be given when they are in focused and relaxed state whereas a negative feedback sound (chirping sound of cricket for one second) would be given when they are in distracted and unrelaxed state.
In the self-exercise session, the ratio of alpha power (8)(9)(10)(11)(12) to high beta (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) power was computed at 2-second intervals. The auditory positive neurofeedback sound was given through the earphone when the ratio of alpha power to high beta power increased to a level greater than or equal to 2.775. The criterion was tested in a previous clinical study conducted with the same EEG system in 1,500 Korean healthy volunteers, and the value was determined by the distribution of T-score of the population [50].

Statistical analysis
Demographic variables and baseline psychological measurements were analyzed using an analysis of variance (ANOVA) or Kruskal-Wallis test for continuous variables, and chi-square tests or Fisher's exact tests were used to compare the three groups. Marital status, educational levels, and length of career were divided into two entities for sake of simplicity. For all outcome measures, including psychological and physiological variables, generalized estimating equation (GEE) was applied to examine the effects of interventions. GEE is commonly used when the outcome variables are measured repeatedly from the same subject, and probable dependencies were assumed between those variables [51]. In this study, group, time, and their interaction term were tested using GEE because of assessing the differential group effect at each time. In the absence of an interaction, the effect of group was tested. As a significance level, a two-tailed P < .05 was selected. All statistical analyses were performed using IBM SPSS statistics version 25 software (IBM Corp., NY, USA).

Ethics statement
The study protocol was approved by the Institutional Review Board of Seoul National University Hospital in Seoul and Seoul National University Bundang Hospital in Seongnam, South Korea. All participants were provided with information of the study and signed informed consent forms prior to participation. This study is registered in the ClinicalTrials.gov (NCT 03787407).

Results
Of the 100 participants screened for this study, 6 participants did not meet the inclusion criteria. As a result, a total of 94 participants were initially included and randomly assigned to one of the three groups: 30 in group 1, 33 in group 2, and 31 in group 3. After the beginning of intervention, one participant in the group 2 dropped out (reason: compatibility problem in between mobile app and personal cellular phone). Thus, a total 93 participants completed the training. One participant in group 1 dropped out due to unemployment before the completion of the 1-month follow-up assessment. Thus, 29 participants in group 1, 32 in group 2, and 31 in group 3 completed all assessments. Figure 2 presents a flow diagram of the study process. The drop-out rates after training engagement were 3.3% (1/30), 3% (1/33), and 0% (0/31) for group 1, group2, and group3, respectively; this distribution was not statistically significant (P = 0.76) by Fisher's exact test. Demographic and clinical characteristics of the participants at baseline who completed all assessments are presented in Table 2. There were no significant differences in age, sex, marital status, educational levels, and length of career between the groups.  Post hoc test using Tukey's method was performed; neurofeedback-assisted mindfulness training group showed higher mindfulness attention awareness scale than self-care group b Post hoc test using Tukey's method was performed; neurofeedback-assisted mindfulness training group showed higher occupational stress scale than mindfulness training group, and mindfulness training group showed higher score compared to self-care group  Table 3.   The EEG and HRV data for 25 participants were excluded from the final analysis; one of the three measurements was not available due to either inappropriate length or continuity of data after removal of artifacts for analysis. Thus, the physiologic data for 67 participants ( (LSM ± SE = 3.87 ± 0.11) showed significant difference compared to group 2 (LSM ± SE = 3.47 ± 0.10) and 3 (LSM ± SE = 3.42 ± 0.12) (P = .009 and P = .007, respectively). In terms of time, the significant difference was found between at the baseline (LSM ± SE = 3.48 ± 0.08) and post-training (LSM ± SE = 3.66 ± 0.08) (P = .01). In additional analysis, we examined differences between groups with relative changes compared to baseline.  Figure   9). In the high beta frequency band, the power of group 1 and 2 remained similarly but that of group 3 decreased ( Figure 10). In the next period, between the post-training and 1-month follow up, the alpha frequency band of group 1 and 2 moved downward together ( Figure 9), but in the beta frequency band, they moved separately in opposite directions ( Figure 10). LF of HRV parameters showed a significant effect by time (Wald Chi-square = 14.073, P = .001) in the model without interaction term ( Figure 11). However, no other significant results were revealed via the HRV parameters.

Discussion
In this study, we found that a 4-week mindfulness-based, self-training program delivered via mobile application was effective in stress reduction and also improved other psychological indices for employees. Additionally, we hypothesized that neurofeedbackassisted training might enhance the effects of mindfulness training. A significant interaction (times × group) was found in the BRS between the neurofeedback-assisted mindfulness training group and the control group; the former showed a significant improvement at the post-training assessment, with continued improvements through the 1-month follow-up assessment period. Group 2 (mindfulness training only) showed significant improvement at the post-training assessment, but at the 1-month follow-up the improvement was not significant. In the relaxation index, neurofeedback-assisted mindfulness training group showed continued improvements compared to the baseline, and the improvement showed significant differences compared to other groups at the 1-month follow-up assessment.
Originally, high fidelity MBSR [52] requires 8 weeks of training with 7.5-hour silent retreat during week six. Practically, this type of program is difficult for general employees to participate. To mitigate this barrier, various mindfulness-based training programs have been developed, with shorter duration and/or incorporation of other kinds of programs. Such programs have been proven to be effective [13]. In our study, the mobile application delivered 3 training modules (awareness training, abdominal breathing, and body scan) over a 4-week period. One of the primary outcome measures was resilience using BRS. Group 1 showed the largest improvement compared with other groups (Figure 4) at the second assessment; and group 1 also showed a significant training effect compared with group 3, which was also the case at the 1-month follow-up assessment.
Previous studies have suggested that mindfulness training could enhance resilience [53][54][55]. One study has shown that increased mindfulness may be related to increased resilience and decreased burnout [53]. Another study has demonstrated that high resilience and higher mindfulness may be significant predictors of lower levels of burnout and psychological distress [56]. Moreover, resilience has been known to be correlated with less stress, better mental health, and more mindfulness [57].
The other outcome measures of our study were evaluated as the effect of mindfulness intervention programs from previously published studies. Several studies reported that PSS were improved after a 4-week mindfulness training program [58,59]. Regarding emotional labor, mindfulness-based intervention appeared to cause less emotional exhaustion [60,61].
One study reported the mediating role of emotional labor to negative correlation between mindfulness and burn out [62]. Regarding the effect of mindfulness on sleep, despite some evidence of improved sleep, a systematic review concluded that controlled studies have not clearly demonstrated positive effects of MBSR on sleep quality and duration [63]. There were no statistically significant differences between groups 2 and 3 in these psychological measures. This could be explained by a self-learning effect of the control group to whom we 26 provided self-learning materials. According to a recent qualitative review with 67 published studies, five studies had an active control group (12.2 %) [13]. There was only one study with a three-arm design that included inactive control and active control groups. It had similar design and total number of training hours with our study. However, significant effects of mindfulness-based program were not found, even in the experimental group [64]. There were two two-armed studies utilizing experimental group and active control group (education only) in the span of 8 weeks. In one of these studies, no significant differences were found between the groups on PSS [65]; however, in the other study, significant differences were found between the two groups on work-related stress scales [66].
On the other hand, for the relaxation index of the right prefrontal channel, neurofeedback-assisted mindfulness training group showed significant differences compared with the other groups at 1-month follow-up assessment. We depicted the changes of alpha and high-beta frequency power (Figure 9 and 10). The alpha power of groups 1 and 2 moved in the same direction during measurements, but the high-beta power of groups 1 and 2 moved separately after the post-training assessment. The alpha and the high-beta power of control group moved in completely different way compared to groups 1 and 2. This transition made the interaction significant, increasing the differences between the groups. Group 1 showed the highest relaxation index at the last assessment. Taken together, we could suggest that a 4- week mindfulness-based training program may increase the power of alpha frequency in groups 1 and 2. Several studies has shown that mindfulness training may be associated with enhanced alpha and theta power, but in beta, delta, and gamma power, the change patterns may be inconsistent [28].
For the power of LF in HRV, all 3 groups showed a decreased pattern without significant differences between the groups or interactions ( Figure 10). This implies that neurofeedback-assisted mindfulness training did not demonstrate additional effects on HRV 27 compared with the other groups. The power of low frequency band of HRV is known to be influenced by parasympathetic, sympathetic nervous activity, and baroreflex [67].
Interpretation of HRV should be careful [68] because of many contextual factors, such as artifacts [69], measurement [70] and others. Several studies reported decreased LF band after the mindfulness-based training program [71,72]. The effects of mindfulness training on LF from these studies were similar with our results. However, another study reported no significant changes in LF, but an increase in the LF/HF ratio [73], which is not consistent with our result.
In this study, we developed a mindfulness-training program embedded in mobile application and verified its effectiveness over a 4-week period. The program delivered by mobile application was effective on PSS, BRS, KELS, AIS, and PHQ-9. The neurofeedbackassisted mindfulness training group showed a significant difference in BRS compared with the control group. In addition, the effect of neurofeedback-assisted mindfulness training on relaxation remained significant at the 1-month follow-up assessment. Therefore, we could suggest that a 4-week program via mobile application may contribute to stress reduction of employees and could improve resilience and relaxation when the mindfulness training is supported with the neurofeedback function.

Figures
Flowchart of the study process.