The patients in the intervention group will be offered an 8-week MBCT followed by a 10-month follow-up program. The MBCT consists of 8 weekly sessions in a group format. Each session lasts for 2 hours. In the program, participants practice mindfulness meditation as well as cognitive exercise. Minimum modifications have been made to the original version of MBCT [20] because the study targets patients with anxiety disorders rather than those with depression. Specifically, we have replaced psychoeducation of depression with that of anxiety. Table 1 describes the themes and contents of each session. The participants will be requested to practice mindfulness meditation for approximately 30 minutes daily and to record the duration of time they meditated and the meditation type.

After the completion of the 8-week MBCT, the participants in the intervention group will be offered to continue the 10-month follow-up program. The program is designed to consist of 2 elements. The first is the 10 monthly follow-up sessions. The participants will be encouraged to attend 1.5-hour–long monthly follow-up sessions, where they will meditate together and share their experiences of mindfulness in daily life. The second element is the use of mindfulness apps in daily life. Participants in the group will be provided with mindfulness apps developed by the research team to be used during the follow-up period. The participants can access the apps either from a smartphone or PC and stream/download the meditation instructions easily. In addition, they are encouraged to send their mindfulness experiences in daily life to the research team every month. The research team will post and share them with other participants on the apps. The research team will also post relevant articles to support participants in continuing the practice. No regular homework will be offered during the follow-up period. Participants will be encouraged to meditate depending on their needs.

The first (MS), second (A Ninomiya), and third (MN) authors led the sessions. The first author is a qualified MBSR teacher at the University of Massachusetts, with 10 years of experience in mindfulness practice. The other 2 authors have been practicing mindfulness for more than 5 years and have experience in offering MBCT 5 times under the supervision of the first author.

Participants in the control group will also be offered the 8-week MBCT. During the follow-up period, they will be encouraged to continue practicing by themselves. However, no additional intervention is intended to be provided after the 8-week MBCT.

MBCT and follow-up sessions were initially planned to be offered in person. However, to ensure participant safety during the COVID-19 pandemic, classes will be offered online.

In this feasibility RCT, the primary outcomes are the (1) study inclusion rate, (2) dropout rate, (3) attendance rate, and (4) mean and standard deviation of the below clinical measures at 8 weeks and 5, 8, and 12 months.

The primary clinical outcome is the mean and standard deviation of the STAI score in both groups at 8 weeks and 5, 8, and 12 months after the start of the intervention. The mean difference and standard deviation of the STAI score between the groups is also assessed.

The secondary clinical outcomes are PAS, LSAS, EQ, SF-36, SPANE, RSES, FFMQ, CDRISC, SCS, QIDS, GAD7, PSS, WHO-HPQ, SWLS, FS, MAIA, EQ-5D, HAM-A, and interoception scores (baseline, 8 weeks, and 12 months only for HAM-A and interoception); health service use; engagement in meditation practice; and satisfaction with and expectation of the classes.

Cost-effectiveness is assessed by the incremental cost-effectiveness ratio, which represents the incremental cost divided by the incremental effectiveness between the groups. Incremental effectiveness is evaluated using quality-adjusted life years calculated from the results of EQ-5D. The analyses are conducted from a health care system perspective.

The STAI is a commonly used measure of state and trait anxiety. It can be used in clinical settings to diagnose anxiety and distinguish it from depressive syndromes. It has 20 items for assessing trait anxiety and 20 for assessing state anxiety. Possible scores range from 20 to 80. Higher scores indicate higher anxiety [34].

The PAS is a measure of illness severity in patients with panic disorder (with or without agoraphobia). It has 13 items with a 5-point scale, which covers the following 5 subscales: panic attacks, agoraphobic avoidance, anticipatory anxiety, disability, and functional avoidance (health concerns). Higher scores indicate more severity [35].

This instrument is used to assess patients’ fear in a range of social interactions and performance situations. The scale consists of 24 items, which are categorized into the following 2 elements: performance anxiety (13 items) and social situations (11 items). Scores are between 0 and 144, with higher scores indicating higher social anxiety [36].

The EQ is a 20-item self-report measure using a 5-point Likert scale ranging from 1 (never) to 5 (always). The total score is between 20 and 100. The scale focuses on decentering, defined as the ability to view the self as separate and different from own thoughts, the capacity for not reacting to negative experiences, and the ability to be self-compassionate. The EQ has been found to be reliable, and convergent and discriminant validities are established for both general and clinical samples. The EQ is also internally consistent, with temporal stability over a 1-month period and good convergent validity [37,38].

The SF-36 is a 36-item multipurpose health survey to evaluate 8 health domains of functional health and the level of well-being, as well as physical and mental health summary measures and a health utility index. Possible scores for each domain range from 0 to 100, with higher scores indicating a better health status [39].

This measure is a 12-item scale to assess positive experiences (6 items) and negative experiences (6 items). Because of the generality of items included in the scale, it can not only assess pleasant and unpleasant emotional feelings that are the focus of most scales, but also reflect other conditions, such as interest, flow, positive engagement, and physical pleasure. The positive (SPANE-P) and negative (SPANE-N) scale scores range between 6 and 30. Higher scores indicate a higher positive or negative affective status. The score obtained on subtracting the negative score from the positive score is called the SPANE-B score, which is between −24 and 24 [40].

This is a brief self-rated assessment tool to evaluate self-esteem, self-worth, acceptability, and confidence. It is the most recognized and widely used measure for these metrics. It has 10 items with a Likert scale (1 = strongly disagree, 4 = strongly agree). The total score ranges from 10 to 40, with higher scores indicating better self-esteem [41].

This tool is a self-report questionnaire that assesses mindfulness. It includes 5 factors, which are extracted on the basis of a factor analysis of 5 mindfulness questionnaires developed independently. The 5 facets are observing, describing, acting with awareness, not judging one’s inner experience, and not reacting to one’s inner experience. The total score ranges from 39 to 195, with higher scores representing a better mindfulness status [42].

The CDRISC is a brief self-rated assessment to measure resilience. The scale contains 25 items, all of which feature a 5-point Likert scale (4 = true nearly all of the time, 0 = not true at all). The scale is rated based on how the subjects felt over the past month. The total score ranges from 0 to 100, with higher scores reflecting greater resilience [43].

This scale assesses a person’s ability to be kind and understanding toward themself, as opposed to harsh and self-critical in instances of pain or failure. It includes 29 items and produces scores on 5 subscales (self-kindness, self-judgment, common humanity, isolation, mindfulness, and overidentification). The subscale scores represent the mean of each subscale’s item scores. Participants are asked to answer how often they had certain thoughts and feelings (1 = rarely to 5 = very often or always). Therefore, each subscale score is between 1 and 5. Higher scores indicate more self-compassion [44].

The QIDS is a self-rated questionnaire to assess depressive symptoms, which is widely used. The responses to 16 separate items on the QIDS are converted into 9 DSM-IV symptom criterion domains. The total score is between 0 and 27. Higher scores indicate higher levels of depressive symptoms [45].

GAD7 was developed to ask patients how often they experienced a set of symptoms in the past 2 weeks. Respondents respond using 4 response options on a Likert scale (0 = not at all to 3 = nearly every day). In addition, an item assessing the duration of anxiety symptoms is included. Therefore, GAD7 scores are between 0 and 21. Scores of 5, 10, and 15 represent mild, moderate, and severe anxiety symptoms, respectively [46].

The PSS was designed to measure the degree to which situations in one’s life are appraised as stressful. The scale has the following 2 versions: the 14-item version (PSS-14) and the 10-item version (PSS-10), with 4 items removed from the 14-item version. We use the PSS-10 in this study. This scale assesses perceived stressful experiences or stress responses in the previous month. Each item is rated on a 5-point Likert scale (4 = never, 0 = very often) to identify positive experiences or responses. The total score ranges from 0 to 40, with higher scores representing higher stress levels [47].

The WHO-HPQ is a self-report instrument designed to estimate the workplace costs of health problems in terms of self-reported sickness absence and reduced job performance (presenteeism). Presenteeism is measured using the following two questions: “On a scale from 0 to 10, where 0 is the worst job performance anyone could have at your job and 10 is the performance of a top worker, how would you rate the usual performance of most workers in a job similar to yours?” and “Using the same 0-10 scale, how would you rate your overall job performance on the days you worked during the past four weeks?” A low presenteeism score indicates poor performance [48].

This scale is a 5-item self-reported questionnaire to evaluate the cognitive aspect of subjective well-being. Scores for each subscale range from 1 (strongly disagree) to 7 (strongly agree). The total score ranges from 5 to 35, with higher scores indicating higher satisfaction [49].

This scale includes 8 items relevant to significant aspects of human functioning, ranging from positive relationships to feelings of competence, meaning, and purpose in life. Each item is answered on a 7-point scale that ranges from 1 (strong disagreement) to 7 (strong agreement). Possible scores range between 8 (strong disagreement with all items) and 56 (strong agreement with all items). Higher scores indicate that respondents viewed themselves positively in important areas of functioning [40].

The MAIA is a self-report scale for experimental interoception research and for the assessment of mind-body therapies [50]. It is a 32-item self-report instrument to assess interoceptive awareness on the following 8 subscales: noticing, not distracting, not worrying, attention regulation, emotional awareness, self-regulation, body listening, and trusting. Each subscale has 3 to 7 items, each assessed on a 6-point Likert scale (0 = never, 5 = always). Scores for each subscale range from 0 to 5. Higher scores indicate better interoceptive awareness [51].

EQ-5D is a standardized measure for assessing health-related quality of life. Applicable to a wide range of health conditions and treatments, it provides a simple descriptive profile and a single index value for health status. The score ranges between 0 (death) and 1 (perfect health) [52].

HAM-A is a rating scale used to measure the severity of anxiety symptoms. It includes 14 items that measure both psychiatric and somatic anxiety. Each item is scored from 0 (not present) to 4 (severe), with a total score between 0 and 56. HAM-A has a structured interview guide (Structured Interview Guide for Hamilton Anxiety Scale: SIGH-A) [53,54].

To assess interoception objectively, we use a heartbeat detection task that has been used and validated worldwide. The participants are asked to wear a pulse oximeter on their finger, which is connected to a PC to evaluate their actual pulse. They are also asked to count the heartbeat felt during various measurement periods. Interoceptive accuracy is measured based on the discrepancy between the number of actual and reported heartbeats [55,56]. The validity and reliability of all measures of the original and Japanese versions, except the Japanese version of the PAS, have been confirmed [34-54,57-74]. With respect to the Japanese version of the PAS, although it shows sufficient reliability, the authors recommend using it as a secondary outcome, because the criterion-related validity indicates “relatively strong correlation” (ie, the correlation coefficient ranges between 0.48 and 0.68). We judge it to be sufficient for use as a secondary outcome.