Mechanisms of mindfulness in patients with migraine: Results of a qualitative study
Abstract
Objective
To understand the mechanisms of mindfulness' impact on migraine.
Background
Promising mindfulness research demonstrates potential benefit in migraine, but no data-driven model exists from the lived experiences of patients that explains the mechanisms of mindfulness in migraine.
Methods
Semi-structured qualitative interviews were conducted with adults with migraine who participated in two mindfulness-based stress reduction (MBSR) clinical trials (n = 43). Interviews were audio-recorded, transcribed, and summarized into a framework matrix with development of a master codebook. Constructivist grounded theory approach was used to identify themes/subthemes.
Results
Participants who learned mindfulness techniques through MBSR experienced altered pain perception, altered response to migraine attacks and disease, increased awareness of external and internal experiences, improved overall well-being, and group benefits. Mindfulness resulted in earlier stress-body awareness and increased interoceptive awareness resulting in earlier attack recognition, leading to earlier and more effective management. Interictal factors of self-blame, guilt, and stigma decreased while migraine acceptance, hope, empowerment, self-efficacy, and self-compassion increased. Improved emotion regulation resulted in decreased fear of migraine, pain catastrophizing, anticipatory anxiety, and pain reactivity. Although taught as prevention, mindfulness was used both acutely and prophylactically. We created a conceptual model hypothesizing that MBSR skills led to an infusion of mindfulness in daily life, resulting in altered pain perception and experience, ultimately leading to improvement in overall well-being, which may positively feed back to the infusion of mindfulness in daily life. The therapeutic benefit of learning mindfulness in a group setting may moderate these effects.
Conclusions
This study identified several new potential mechanisms of mindfulness' effect on migraine. After learning MBSR skills, participants reported altered pain and migraine perception and experiences. Increased stress-body and interoceptive awareness resulted in earlier migraine awareness and treatment. Mindfulness may target important interictal factors that affect disease burden such as fear of migraine, pain catastrophizing, and anticipatory anxiety. This is the first data-driven study to help elucidate the mechanisms of mindfulness on migraine from patient voices and can help direct future research endeavors.
Abbreviations
-
- ICHD
-
- International Classification of Headache Disorders
-
- MBSR
-
- mindfulness-based stress reduction
-
- MRI
-
- magnetic resonance imaging
-
- QST
-
- quantitative sensory testing
-
- SD
-
- standard deviation
-
- UCNS
-
- United Council for Neurologic Subspecialties
INTRODUCTION
Mindfulness has emerged as a potential treatment option for migraine.1, 2 Pharmacological treatments are often limited by side effects, lack of efficacy, and/or costs.3-5 Prior studies have demonstrated benefits of mindfulness for chronic pain conditions.6 Research assessing mindfulness specifically for migraine is critical as migraine is different from chronic pain, involving intermittent unpredictable attacks of acute pain and sensory amplification (e.g., photophobia, phonophobia, osmophobia, etc.). Recent clinical trials of mindfulness interventions in patients with migraine are promising.2, 7, 8
We conducted two randomized clinical trials evaluating Mindfulness-Based Stress Reduction (MBSR) in adults with migraine.9, 10 MBSR was taught with the standardized curriculum of eight weekly 2-h sessions plus a “retreat” day. During the intervention, certified MBSR instructors led participants through experiential mindfulness practices including body scan, sitting meditation, walking meditation, and sitting and standing gentle mindful yoga. During these practices, participants engaged in both focused attention (sustaining attention on sensations of the breath, while disengaging from distractors such as mind-wandering) and open monitoring (nonreactive awareness of the flow of cognition, emotions, and sensations).11 Dialogue and inquiry provided an opportunity for participants to discuss their experiences with mindfulness practices and how to incorporate them into daily life. Typical mindfulness skills learned included use of the breath as an anchor; noticing automatic “pilot-mode” and stress reactivity; attention to body sensations; nonreactive awareness of thoughts, emotions, and sensations; awareness of pleasant, unpleasant, and neutral sensations; and nonjudgment and equanimity. For these studies, MBSR was not modified for the migraine patient population. Participants were given MP3 files or compact discs and instructed to use them for daily practice at home, with logs demonstrating average participant practice of 34 (standard deviation [SD] = 11) minutes per day in the pilot study9 and 33 (15) min per day in the larger study, with an average practice of 4.2 (2.5) days per week.10
The neurobiological mechanisms of mindfulness are actively being explored by researchers across the world.12, 13 We previously published a narrative review discussing potential mechanisms of mindfulness in migraine, which included the following: (1) alterations in pain perception via decoupling the sensory and affective components of pain; (2) enhanced interoceptive body awareness; (3) regulation of autonomic dysfunction seen in patients with migraine; and (4) improved cognitive modulation of pain.2 Quantitative results from our clinical trials showed meaningful clinical improvements in headache-related disability, pain catastrophizing, headache self-efficacy, and depressive symptoms, with benefits seen out to 36 weeks.9, 10 After MBSR, participants had a 36% reduction in response to experimental heat pain intensity and a 30% reduction in heat pain unpleasantness, suggesting a shift in pain perception that may alter migraine experience.10 To further understand these quantitative findings and to investigate the mechanisms of how mindfulness impacts migraine, we went to the lived experiences of the participants with migraine who took the MBSR course. To ensure that we fully captured their experiences, and since there is currently no data-driven model that explains the mechanisms of mindfulness in migraine, we used a hypothesis-free, constructivist grounded theory approach to analyze the qualitative interviews.9, 10 The objective of our study was to better understand potential mechanisms of mindfulness' benefit on migraine by listening to and understanding the patient perspective.
METHODS
We conducted semi-structured in-person qualitative interviews with participants with migraine from two randomized clinical trials on MBSR9, 10 (n = 43) registered at clinicaltrials.gov (NCT01545466 and NCT02695498), with institutional review board approvals from Brigham and Women's Hospital and Wake Forest University School of Medicine, respectively. All participants gave written informed consent prior to participation. Full methods of both clinical trials have been described previously.9, 10, 14
Study design, setting, and participants
Briefly, a pilot study was conducted in Boston, Massachusetts, of MBSR (n = 10) versus a wait-list control group (n = 9), with enrollment between January and March 2012.9 One hundred percent of those randomized to MBSR completed qualitative interviews. A larger study (n = 89) was conducted in Winston-Salem, North Carolina, of MBSR (n = 45) versus an active comparator Headache Education control group (n = 44), enrolled between August 2016 and October 2018.10 Of those randomized to MBSR in the larger study, 73.3% (33/45) completed qualitative interviews, making the total response rate of participants completing qualitative interviews in both studies 78.2% (43/55). Of the 12 participants not interviewed in the larger study, 4 were lost to follow up, 4 withdrew from the study, and 4 were not interviewed during their 12-week study visit.10 Inclusion criteria for both studies included participants keeping prospective diaries to confirm migraine frequency (inclusion of 4–14 days/month in the pilot study and 4–20 days/month in the larger study). United Council for Neurologic Subspecialties (UCNS)-certified headache physicians conducted in-person evaluations to confirm International Classification of Headache Disorders (ICHD) migraine diagnoses. Participants remained on their migraine medications for the study duration and were blinded to group assignment.
Quantitative sensory testing
Quantitative sensory testing (QST, e.g., “heat pain testing”) was conducted at baseline and at three follow-ups (12, 24, and 36 weeks post-baseline) on 33 participants in the larger study, using a 16 × 16 mm thermal probe with the MEDOC TSA-II. Both intensity and unpleasantness were measured using a 15 cm sliding visual analogue scale, ranging from “no pain sensation” to “most intense imaginable” and from “not at all unpleasant” to “most unpleasant imaginable,” respectively.15, 16
Intervention
The participants in this study who completed qualitative interviews had undergone the standardized MBSR intervention that was delivered with 8 weekly classes of 2 h duration.
Interviews
All interviews were conducted after participants completed MBSR intervention at the 12-week study visit. Expert qualitative interview consultants from the Wake Forest University School of Medicine Qualitative and Patient-Reported Outcomes Shared Resource provided educational training, consultation, and resources to all team members on how to appropriately conduct effective unbiased qualitative interviews. Qualitative interview techniques that were taught included appropriate probing techniques to clarify details without leading participants. A semi-structured interview guide was utilized during qualitative interview sessions that contained mostly open-ended questions (see Supporting Information). The main intent of the interviews was to capture participants' experiences with the clinical trial and with the interventions. All interviews were recorded and transcribed verbatim. The interviews from the entire larger study (n = 71) lasted on average 47 min (SD 13.9) with a range of 18–77 min and a total of nearly 3300 min of recorded interviews.14
Sample size
For the pilot study, the goal sample size of 34 was not reached (time constraints unrelated to the feasibility of the study [e.g., R.E.W.'s relocation] limited recruitment to 3 months and decreased the ability to reach the target sample size).9 For the larger randomized clinical trial, we estimated a final sample size of 44 per group to provide at least 90% power with an alpha of 0.05 (using PASS statistical software).10 For this qualitative study, all participants from both studies who completed MBSR and were interviewed were included (n = 43) to ensure we fully captured all potential mechanisms of mindfulness in migraine.
Qualitative data analyses
Themes and subthemes were identified using Charmaz's hypothesis-free, constructivist grounded theory approach.17 Under the leadership of a qualitative research specialist (from Wake Forest Qualitative and Patient-Reported Outcomes Developing Shared Resource), team members who were not mindfulness experts (undergraduate, masters, and MD/PhD students; n = 7) received weekly training in qualitative data analysis over 6 months. The qualitative data expert provided educational training, consultation, and participation in team meetings. All students were educated on qualitative interview techniques and analyses using educational materials such as the textbook The Coding Manual for Qualitative Research,18 with regularly assigned readings reviewed (for further details, see Estave et al.14). The research team, including the PI, analyzed the data during weekly meetings (August 2019–January 2022), beginning with summarizing and analyzing transcripts into a framework matrix of 12 key domains extrapolated from the interview guides (see Supporting Information).19-22 Investigator triangulation23 was achieved for 50% of the transcripts until a consistent technique was established. All final summaries were cross-checked against the original transcripts to ensure accuracy. Emerging ideas, potential codes, and any challenges were then discussed at team meetings. A master codebook was created until meaning saturation was achieved.24 Dedoose software was used for data analysis of transcripts. The codebook was applied to each transcript by six independent coders. One-third of the interviews were assessed by two independent raters, with interrater reliability tests conducted to confirm consistency. An iterative process of coding, categorization, discussion, and disagreement resolution continued until all team members agreed on a final set of themes, subthemes, and table organization. Experts in qualitative interviewing, mindfulness, and/or behavioral interventions (E.S. and P.G.) were then included to help interpret results. For further details on the research team, please see the “Research Team” section. Additionally, themes, subthemes, and the conceptual model were continually discussed to increase awareness of theoretical sensitivity and facilitate reflexivity of the researchers. An audit trail was maintained throughout the entirety of data analysis with notes from each research team meeting, within Dedoose software, and through use of an online SharePoint team site (maintained with high security behind a firewall compliant with the Health Insurance Portability and Accountability Act) to organize and maintain study files (e.g., framework matrices, etc.).
Statistical analyses
Descriptive statistics were used throughout to describe continuous and categorical data. Percentages and standard deviations were used to describe the baseline characteristics of participants. Percentages and frequency were also provided for the number of participants who completed qualitative interviews. All statistical analyses were performed using R Software.
Research team
At the time of data analysis, the research team included the PI, a paid consultant in qualitative interviewing, and seven students of various training levels, including undergraduate (n = 4), masters (n = 2), and MD/PhD students (n = 1). These students were thoroughly trained in qualitative data analysis as described above and were not mindfulness, migraine, or qualitative data experts. Given the PI's expertise in mindfulness, the PI was able to identify and recognize meaningful data relevant to the themes and conceptual model during data analysis and interpretation, thus employing theoretical sensitivity throughout the process. For data interpretation, content expertise included those with mindfulness clinical trial expertise (R.E.W. and P.G.), behavioral intervention expertise (E.S.), and qualitative interviewing expertise (P.G. and E.S.).
RESULTS
Baseline characteristics
Participants in both the pilot and larger studies were an average of 44 (SD = 13.0) years old, mostly women (92%), White (89%), and had private health insurance (87%) (Table 1). Most participants had a 26-year (SD = 14.0) history of migraine, with 9.8 (3.3) headache days/month, 6.6 (2.7) migraine days/month, and high disability at baseline (HIT-6 score of 63.0, SD = 7.2).
Baseline characteristic | Pilot RCT with MBSR n = 10 n (%) | Larger RCT with MBSR n = 37a n (%) | Combined pilot and larger RCTs (n = 47) N (%) |
---|---|---|---|
Sociodemographics | |||
Age (years); mean (SD) | 46 (17) | 44 (12) | 44.43 (13) |
Sex | |||
Female | 9 (90) | 34 (92) | 43 (92) |
Male | 1 (10) | 3 (8) | 4 (9) |
Race | |||
White | 9 (90) | 33 (89) | 42 (89) |
Black or African American | 1 (10) | 4 (11) | 5 (10) |
Primary health insurance | |||
Private | 8 (80) | 33 (89) | 41 (87) |
Medicare/Medicaid/other public | 2 (20) | 4 (11) | 6 (13) |
Marital status | |||
Married/living with partner | 9 (90) | 27 (74) | 36 (77) |
Divorced/separated/widowed | 0 | 5 (13) | 5 (11) |
Single-never married | 1 (10) | 5 (13) | 6 (13) |
Current employment status | |||
Employed/self-employed full time (>30 h/week) | 4 (40) | 25 (68) | 29 (62) |
Employed part-time | 1 (10) | 3 (8) | 4 (9) |
Student, homemaker, volunteer | 2 (20) | 6 (16) | 8 (17) |
Unemployed, retired | 3 (30) | 3 (8) | 6 (13) |
Education | |||
≤High school | 1 (10) | 3 (8) | 4 (9) |
College | 4 (40) | 20 (54) | 24 (51) |
Graduate degree | 5 (50) | 14 (38) | 19 (40) |
Recruitment source | |||
Academic medical center/provider referral | 8 (80) | 18 (49) | 26 (55) |
Community | 2 (20) | 19 (51) | 21 (45) |
Headache features | |||
Disease characteristics during baseline HA log | |||
Headache days during 28-day baseline, mean (SD) | 10.4 (3.1) | 9.6 (3.4) | 9.8 (3.3) |
Migraine days during 28-day baseline, mean (SD) | 4.2 (2.9) | 7.3 (2.6) | 6.6 (2.7) |
Disease history | |||
Years with migraine, mean (SD) | 26 (19) | 26 (13) | 26 (14.4) |
Migraine with aura | 4 (40) | 14 (38) | 18 (38.3) |
Family history of headache | 7 (70) | 25 (68) | 32 (68.1) |
Use of treatments | |||
Current use of prophylactic treatment | 8 (80) | 12 (32) | 20 (43) |
Current use of acute medicationb | 10 (100) | 33 (89) | 43 (92) |
Experienced headache medication side effect | 5 (50) | 23 (62) | 28 (60) |
Triggers, comorbidities, and disability | |||
Stress or letdown stress as a trigger | 8 (80) | 28 (76) | 36 (77) |
Current or past diagnosis of depression | 2 (20) | 15 (41) | 17 (36) |
Current or past diagnosis of anxiety | 2 (20) | 12 (32) | 14 (30) |
MIDAS-one month at baselinec, mean (SD) | 12.5 (9.8) | 13.7 (11.5) | 13.4 (11.2) |
HIT-6 at baseline, mean (SD)d | 63.0 (8.0) | 63.0 (7.0) | 63 (7.2) |
- Abbreviations: HA, headache; MBSR, mindfulness-based stress reduction; MIDAS, Migraine Disability Assessment; RCT, randomized clinical trial; SD, standard deviation.
- a Baseline characteristics from the larger RCT represent participants who completed visit 2 (4 additional participants than the 33 who participated in the qualitative interviews).
- b Missing data from the larger RCT for acute medication use (n = 3 or n = 4 MBSR group).
- c Migraine Disability Assessment (MIDAS)-one month was used to decrease recall bias and improve the accuracy of the results (compared to the typical 3-month recall period).27, 31, 32 Scores range 0–115, higher scores reflect greater disability, MIDAS is typically used as an average over 3 months; to facilitate interpretation of the MIDAS-one month data presented, the mean estimate results (but not the confidence intervals) can be multiplied by 3 for conversion to the typical 3 month assessment27; score range for 3 month MIDAS: 0–5: little or no disability, 6–10 mild disability, 11–20 moderate disability, 21+ severe disability.
- d Headache Impact Test-6 (36–78), higher scores reflect greater headache impact, score range: <49: little to no impact, 50–55: some/moderate impact, 56–59: substantial impact, 60+ severe impact.
Main findings
The themes and subthemes from our main findings are summarized in Table 2, with representative quotations in Tables 3–7. Full tables can be found online (Tables S1). We found that through learning mindfulness with an MBSR course, participants with migraine had altered pain experience demonstrated through heat pain testing (Table 3); altered response to migraine attacks and relationship to migraine disease (Table 4); infusion of mindfulness skills in daily life with increased awareness of external and internal experiences (Table 5); improvement in overall well-being (Table 6); and therapeutic benefits of the group experience (Table 7).
Main finding (Table) | Themes and subthemes |
---|---|
Altered pain experience seen through experimental heat pain testing (Table 3) | Changed pain experience during heat pain testing
|
Increased migraine awareness from heat pain testing
|
|
Altered response to migraine attack (Table 4A) | Increased interoceptive awareness of migraine
|
Improved emotion regulation during migraine attacks
|
|
Altered relationship to migraine disease (Table 4B) | Increased emotion regulation of migraine disease
|
Increased understanding of migraine and stress
|
|
Infusion of mindfulness skills in daily life with increased awareness of external experiences (Table 5A) | Increased awareness of self and relationship to world
|
Increased awareness of sensory input
|
|
Infusion of mindfulness skills in daily life with increased awareness of internal experiences (Table 5B) | Increased present moment experience
|
Increased interoceptive awareness: Awareness of the body, thoughts
|
|
Increased emotion regulation
|
|
Increased nonjudgmental awareness
|
|
Improvement in overall well-being (Table 6) | Improvement in emotional well-being
|
Increased cognitive well-being
|
|
Improved social well-being: Enhanced relationships
|
|
Meaningful behavior changes
|
|
Therapeutic benefit of group experience of learning mindfulness in patients with migraine (Table 7) | Therapeutic group benefits
|
- Abbreviation: MBSR, mindfulness-based stress reduction.
Theme | Subtheme | Quotations |
---|---|---|
Changed pain experience during heat pain testing | Changed perception of experimental heat pain testing |
|
Decreased reactivity to pain during heat pain testing |
|
|
Increased migraine awareness from heat pain testing | Better understanding of migraine pain through heat pain testing |
|
Differentiating migraine pain intensity from unpleasantness with mindfulness |
|
- Abbreviation: MBSR, mindfulness-based stress reduction.
Theme | Subtheme | Quotations |
---|---|---|
A: Altered response to migraine attack after learning mindfulness | ||
Increased interoceptive awareness of migraine | Increased sensitivity to interoceptive signals resulting in earlier awareness of migraine attack |
|
Earlier migraine attack awareness leads to altered use of medications |
Taking medications earlier
Avoidance of medication
|
|
Use of mindfulness as an acute treatment |
|
|
Use of breath during migraine attacks and with life challenges |
|
|
Acknowledging and accepting pain during an attack |
|
|
Interoceptive stress awareness prevents migraine trigger |
|
|
Improved emotion regulation during migraine attacks | Awareness of, and less: Fear of migraine |
|
Awareness of, and less: Pain catastrophizing |
|
|
Awareness of, and less: Anticipatory anxiety |
|
|
Decreased reactivity to migraine pain |
|
|
Increased pain acceptance with decreased self-blame, guilt, and stigma |
|
|
Self-compassion during migraine attacks |
|
|
B: Altered relationship to migraine disease after learning mindfulness | ||
Improved emotion regulation of migraine disease | Migraine acceptance |
|
Hope and empowerment |
|
|
Improved self-efficacy |
|
|
Increased understanding of migraine and stress | Heightened awareness of migraine and stress relationship |
|
Improved ability to manage stress and migraine |
|
Theme | Subtheme | Quotations |
---|---|---|
A: Increased awareness of external experiences after learning mindfulness | ||
Increased awareness of self and relationship to world | Meta-awareness: Observation of external world separate from self (“self as an observer”) |
|
Broader life perspective |
|
|
Nonreactive awareness through decentering and re-perceiving |
|
|
Increased awareness of sensory input | Heightened awareness and appreciation of nature |
|
Heightened exteroceptive (e.g., audio, visual, olfactory) sensory awareness |
|
|
B: Increased awareness of internal experiences after learning mindfulness | ||
Increased present moment experience | Decreased autopilot mode |
|
Breath as an anchor leads to nonreactivity |
|
|
Increased interoceptive awareness: Awareness of the body, thoughts | Increased sensitivity to interoceptive signals with embodied awareness |
|
Metacognition: awareness of thoughts separate from self |
|
|
Increased emotion regulation | Improved emotional insight |
|
Downregulation of emotional reactivity |
|
|
Earlier stress-body awareness with sensitivity to stress interoceptive signals |
|
|
Improving stress response through awareness |
|
|
Increased nonjudgmental attitudes | Self-acceptance through nonjudgment |
|
Acceptance of life stressors |
|
|
Increased self-compassion |
|
Improvement in emotional well-being | Enjoyment and happiness |
|
Equanimity (mental calmness, ability to relax) |
|
|
Improved depressive symptoms |
|
|
Improved anxiety |
|
|
Prioritizing and accepting time for self-care |
|
|
Increased cognitive well-being | Awareness of mind–body integration |
|
More focused, able to think more clearly |
|
|
Improved social well-being: Enhanced relationships | Increased empathetic understanding and emotional literacy |
|
More patience with self and others |
|
|
Improved listening skills |
|
|
Improved relationships through decreased reactivity |
|
|
Meaningful behavior changes | Improved eating and drinking habits |
|
Improved activity level |
|
|
Improved sleep hygiene |
|
Theme | Subtheme | Quotations |
---|---|---|
Therapeutic group benefits | Decreased feelings of migraine isolation |
|
Creation of a “safe space” |
|
|
Enhanced mindfulness motivation and accountability |
|
|
Empowerment through shared stories |
|
|
Migraine reappraisal through understanding others' experiences |
|
We created a conceptual model (Figure 1) of proposed relationships between the observed effects of learning mindfulness through MBSR in patients with migraine. We hypothesized that the newly learned MBSR skills led to an infusion of mindfulness into daily life, resulting in both an altered pain experience and migraine experience, ultimately resulting in improved overall well-being. The therapeutic benefits of the group experience may be moderating these effects. Additionally, the improvement in well-being may have resulted directly from, and created a positive feedback loop to, the infusion of mindfulness into daily life.

Themes and subthemes
After learning mindfulness, participants' experiences during heat pain testing changed, with alterations in pain perception and decreased reactivity to pain (Table 3). In addition, participants had a better understanding of their migraine pain and the differentiation between pain intensity and unpleasantness through heat pain testing (Table 3).
Participants' response to migraine attacks changed through increased interoceptive awareness (conscious perception of internal body sensations)25 and improved emotion regulation during migraine attacks (Table 4A). Specifically, increased sensitivity to interoceptive signals led to earlier awareness of migraine attacks, which altered medication use. Participants used mindfulness as both an acute and prophylactic migraine treatment, allowing them to take medications earlier or avoid them altogether. Learning mindfulness led to using the breath as an anchor during migraine attacks and life challenges. Mindfulness helped participants acknowledge and accept the pain during an attack. Increased interoceptive awareness of stress sensations in the body may have allowed for mindfulness to be used to alter the progression of migraine attacks. Improved emotion regulation during migraine attacks led to increased awareness of and less fear of migraine, pain catastrophizing, anticipatory anxiety, and reactivity to pain. Participants experienced self-compassion during attacks and increased pain acceptance with less self-blame, guilt, and stigma.
Learning mindfulness also altered the relationship to migraine disease through increased emotion regulation with migraine acceptance, hope, empowerment, and self-efficacy (Table 4B). An increased understanding and awareness of the relationship between migraine and stress developed, possibly resulting in an improved ability to manage both stress and migraine (Table 4B).
Learning mindfulness increased awareness of external experiences (Table 5A). An increased awareness of self and relationship to the world allowed for development of meta-awareness, with observation of the external world as separate from oneself (e.g., “self as an observer”). Participants gained broader life perspectives and nonreactive awareness through decentering and re-perceiving (separating oneself from one's thoughts with better clarity of experiences in the moment).26 Participants experienced increased awareness of sensory input seen through heightened nature awareness and appreciation. Heightened exteroceptive (e.g., audio, visual, olfactory) sensory awareness also occurred.
Mindfulness also increased awareness of internal experiences (Table 5B) in the present moment, decreased autopilot mode, and increased use of breath as an anchor leading to nonreactivity. Interoceptive awareness of the body and thoughts developed with increased sensitivity to interoceptive signals with embodied awareness and metacognition (awareness of thoughts as separate from self). Emotion regulation improved, with enhanced emotional insight, downregulation of emotional reactivity, earlier stress-body awareness (with sensitivity to stress interoceptive signals), and improved stress response. Nonjudgmental attitudes, such as self-acceptance, acceptance of life stressors, and self-compassion, also increased.
Mindfulness improved overall sense of well-being (Table 6). Improvement in emotional well-being was seen through enjoyment and happiness, equanimity (mental calmness and ability to relax), improved depressive symptoms and anxiety, and prioritizing and accepting time for self-care. Increased cognitive well-being was seen through awareness of mind–body integration and being able to think more clearly. Social well-being improved with enhanced relationships, as empathetic understanding and emotional literacy improved. Participants reported being more patient with themselves and others as well as having improved listening skills. Relationships also improved with decreased reactivity. Meaningful behavior changes were seen through improved eating and drinking habits, activity level, and sleep hygiene.
Finally, therapeutic benefits of the group experience of learning mindfulness in participants with migraine was evident (Table 7). Participants experienced feelings of decreased migraine isolation and creation of a “safe space.” The group experience enhanced mindful motivation and accountability. Shared stories led to empowerment, and the connection with others who had similar experiences fostered migraine reappraisal through understanding others' experiences.
DISCUSSION
Participants with migraine who learned mindfulness techniques through MBSR developed increased mindful awareness of internal and external experiences, with shifts in pain perception and migraine experiences resulting in improvement in overall well-being. Learning mindfulness skills changed participants' perspectives, altering their experiences with the world and with migraine attacks. Interoceptive and stress-body awareness increased, with overall improved migraine management. Improved emotion regulation around the migraine experience resulted in decreased impact of interictal factors such as fear of migraine, pain catastrophizing, anticipatory anxiety, and pain reactivity. Receiving the intervention in a group context provided an opportunity for participants with migraine to connect with others experiencing similar challenges, resulting in decreased feelings of migraine isolation and increased mindfulness motivation and accountability. This study's findings support and extend the existing literature regarding the impact and mechanisms of mindfulness on health, emotional wellness, relationships, and healthy behaviors.25-30 Our findings of improved cognitive well-being are consistent with recent magnetic resonance imaging (MRI) evidence suggesting mindfulness improves cognitive efficiency in adults with migraine.8 These qualitative findings expand our understanding of the mechanisms through which mindfulness may positively impact adults with migraine and help explain and validate our quantitative findings previously reported.9, 10
In our quantitative results, we found that mindfulness resulted in a shift in pain perception, with a 30% and 36% reduction in experimental pain unpleasantness and intensity, respectively, in participants with migraine.10 While mindfulness meditation has been shown to attenuate the subjective pain experience in healthy controls,31, 32 our research is unique in demonstrating a reduction in the perception of experimentally-induced pain across time in a clinical trial of a patient population with pain (i.e., migraine). Our qualitative results help interpret these quantitative findings, as participants explained how they experienced this shift in pain perception, noticing decreased reactivity to pain with use of the breath. Mindfulness impacted participants' migraine pain experience by changing their awareness, perspective, and reactivity, leading to an improved ability to differentiate pain intensity from unpleasantness. The neural mechanisms of mindfulness-based pain relief have demonstrated that reductions of pain intensity are associated with heightened activity of the anterior cingulate cortex, the anterior insula, and the orbitofrontal cortex, while reductions in pain unpleasantness are associated with attenuated activity of the thalamus.32 These results demonstrate that mindfulness is engaged in the cognitive modulation of pain through cognitive and affective pain control and reappraisal of sensory nociceptive input.33 Further, our prior results showing that mindfulness-based pain relief is not mediated through endogenous opioids34 highlights the importance of finding the non-opioidergic pathways that could explain the full benefit of mindfulness' impact on migraine. Ongoing research at Massachusetts General Hospital is being conducted to further evaluate the mechanisms of mindfulness meditation in patients with migraine utilizing multimodal positron emission tomography/MRI technology (NIH P01AT009965) to further understand how mindfulness reduces pain perception. This shift in pain perception and altered response to migraine attacks may be key factors contributing to the reductions in migraine-related disability, pain catastrophizing, and depressive symptoms seen quantitatively in our clinical trials.9, 10
This study identified several new potential mechanisms of mindfulness' effects on migraine. The increased interoceptive awareness experienced by participants led to earlier awareness of migraine attacks (e.g., during prodromal state) and triggers, resulting in earlier use of medications. Although many patients and providers may consider mindfulness as an alternative to medication, this finding demonstrates that learning mindfulness can be used as a pharmacological adjunct and may result in better medication management. For example, mindful patients may adhere better to instructions to use acute medications early during a migraine attack because of this enhanced ability to tune in and recognize symptoms of an attack earlier during the prodrome. In our study, identifying a potential migraine earlier even allowed participants to prepare for, or avoid, the attack altogether, often using breathing and relaxation techniques learned through MBSR.
Historically, increased body awareness was thought to lead to somatosensory amplification and hypervigilance towards pain, worsening anxiety sensitivity and creating a maladaptive response.25 On the contrary, our findings demonstrate that the enhanced interoceptive awareness of migraine sensory experiences, including both during the attacks and the prodrome, may be advantageous in patients with migraine. However, such increased interoceptive awareness may only be beneficial when coupled with the reduction of maladaptive responses that occurs with mindfulness, such as attenuated fear of migraine, pain catastrophizing, anticipatory anxiety, and pain reactivity. Mindfulness specifically teaches “turning towards” pain rather than away from it, encouraging awareness of different sensations with curiosity and nonjudgment, ultimately allowing for cognitive pain reappraisal.27, 35, 36 This increased embodied interoceptive awareness may decrease the experience commonly seen in those with chronic pain that the body has “betrayed them,”37 allowing for positive benefits of interoception.
Although this study has identified a potential impact of improved interoceptive awareness, it is unclear which mindfulness practice drives this effect in migraine. For example, while the open monitoring of mindfulness may be key, the experiential practices of focused attention on the breath versus the body scan may also be instrumental in increasing migraine interoceptive awareness.38 Interoceptive awareness localizes to the insula,38 an area of the brain considered a “hub of activity” in migraine, playing a key role in integrating many of the dynamic processes of migraine (sensory, autonomic, cognitive, and emotional).39 Further research is needed to better understand the neurophysiology underlying improved interoceptive awareness from mindfulness in migraine.
Participants described mindfulness skills as allowing them to become aware of their stress response sooner and using MBSR techniques to decrease their stress reactivity, a commonly identified migraine trigger.40 Participants highlighted how mindfulness decreased interictal factors of self-blame, guilt, and stigma, while also increasing migraine acceptance, hope, empowerment, self-efficacy, and self-kindness during attacks. These factors may play a more meaningful role in individuals' day-to-day experiences of living with migraine than previously recognized.14 While the focus in clinical trials and clinical practice is often placed on the ictal burden of migraine pain, with a specific focus on 1–10 pain ratings, mindfulness may target these other factors that contribute to significant interictal disease burden. As those with migraine experience recurrent, intermittent, unpredictable attacks of pain and sensory amplification with resulting significant disability, targeting this interictal disease burden is challenging, especially with pharmacological options alone. Mindfulness may thus be a valuable adjunct to help address these interictal factors that contribute to significant disease disability and poor quality of life.
This clinical trial taught mindfulness as a preventive migraine strategy, as it has been most often studied.2, 8 Interestingly, many participants reported using mindfulness as an acute treatment. To date, only a few small studies have assessed and found the practice of mindfulness beneficial for acute treatment of migraine.41, 42 Psychologically focusing on the breath may physiologically activate the vagus nerve, which has been demonstrated to be an effective acute migraine treatment with the novel noninvasive vagus nerve stimulator.43 This is potentially one of the mechanisms that explains the unique benefits of mind–body modalities in migraine (e.g., yoga,44, 45 tai chi,46 etc.). Use of the breath as an anchor was reported throughout many interviews and can be found in various themes and subthemes, including improving emotion regulation, relationships, and nonreactive awareness. Breath awareness played a key role in understanding migraine and stress, helping to decrease stressors by leading to earlier stress-body awareness, nonreactivity, and avoidance of medications. We recently found that slow breathing combined with attention to breath reduces pain independent of endogenous opioids,47 further highlighting the potential importance of the breath in pain reduction. Although not quantitatively assessed, our data support beneficial effects of mindfulness as an acute treatment, demonstrating a need for further research in this area.
The group setting of MBSR allows for dialogue, inquiry, and didactic discussions, providing an opportunity to learn how to use mindfulness skills in the context of daily life experiences. Group mindfulness sessions have been found to more effectively induce a state of mindfulness and strengthen social connectivity compared to solitary sessions,48 while also providing valuable social support and perspective insight, especially when practiced among others with a shared medical condition.49 Our study identified the group experience as beneficial for connecting people with migraine to each other, decreasing migraine isolation, developing a sense of group support and empowerment, improving migraine perspective, and creating a safe space. The shared group experience also enhanced motivation and accountability for mindfulness practice and engagement. The increased interoceptive awareness that participants experienced from mindfulness may have also played a role in this enhanced social connection.50 Pain can be isolating,37 especially for those with sensory hypersensitivities that preclude full engagement in life activities, such as in migraine.14 The group-based, in-person instruction differentiates this intervention from many app-based programs that exclusively offer asynchronous learning, experiential practices, or self-management strategies. As meditation interest and online options have increased,51 especially during the COVID-19 pandemic,52, 53 understanding the value of group-based and in-person sessions is important to knowing the critical ingredients of learning, engaging with, and benefiting from mindfulness. Given that group environments are known to provide specific therapeutic benefits,54 we designed our study with an active comparator Headache Education group to account for this effect.10 Although MBSR is taught in a group setting and may be a moderator of mindfulness' benefits seen in this study (Figure 1), our quantitative results demonstrated the impact of the mindfulness group beyond our comparator Headache Education group, suggesting the group effect was not the sole mechanism explaining clinical benefits.10 The role of the group setting in learning mindfulness with others who similarly experience migraine requires further investigation.
A multitude of prior models have been developed to understand the mechanisms of mindfulness, which are well-summarized in this review.24 Our conceptual model framework (Figure 1) organizes our main findings to better understand the major factors potentially contributing to altered pain perception and experiences with migraine attacks. These results also highlight the importance of the biopsychosocial model, with biological (awareness), psychological (well-being), and social (group) factors all playing a role in the altered pain and migraine experiences. While we demonstrated the key hypothesized relationships driving the effects seen in this research, further research will be needed to compare our model with previously developed models38 and to validate the hypothesized relationships with effect sizes of the interactions.
Strengths and limitations
This is the first study to use patient voices of those with migraine from their lived experiences taking a mindfulness course to develop a model for the mechanisms of mindfulness in migraine. This study has several important strengths, including a large sample of 43 qualitative interviews, obtained across two clinical trials that were conducted in two different cities across time. Participants were diagnosed with migraine by a UCNS-certified headache specialist using ICHD-diagnostic criteria. Rigorous qualitative methodology was used to effectively capture, organize, and analyze data. While previous research has hypothesized mechanisms of mindfulness through survey responses, this study captured mechanisms through the voices and lived experiences of the participants who had experientially learned mindfulness during a standardized MBSR course. The use of open-ended questions allowed for participants' perspectives to be depicted authentically.
Limitations of this study include a non-diverse patient population, as most participants were White educated females, decreasing study generalizability. Participants in the two trials had either 4–14 or 4–20 migraine days/month, were treatment-seeking, and had baseline severe headache-related disability; therefore, the results reflect the impact of MBSR in patients with this disease burden. In the two clinical trials, 78% randomized to MBSR completed the qualitative interviews, so results only reflect responses of those who completed the interview. MBSR involves weekly group sessions with a guided teacher, and participants practiced mindfulness at home over 30 min per day on the days practiced; results may not generalize to other mindfulness interventions or lower dosages. Although access has increased during the COVID-19 pandemic with online class availability,53 MBSR is limited to those with the time, energy, and financial resources to pay for the sessions. Though iterative data collection would have been a helpful approach, interviews were conducted immediately after the intervention when the perspective on mindfulness' impact may have been greatest, and effects seen may wane over time. Magnitude coding was not conducted to assess the proportion of participants endorsing the various themes as our overarching goal with these analyses was to understand all potential mechanisms that could play a role of mindfulness in migraine. Future research may help identify which mechanisms may have the largest effect. Additionally, inferences about potential mechanisms are based on patient perceptions. Interindividual differences may exist, as the role each factor plays for each person may be more, less, or not at all. Our conceptual model broadly hypothesizes the effects seen across all participants, and each person may not experience each one. Furthermore, gender, race, ethnicity, socioeconomic status, and other individual differences may influence the impact each factor has on mindfulness' mechanism on migraine. We did not capture effect sizes of the different variables in the model.
In summary, we discovered new findings of the potential mechanisms of mindfulness in adults with migraine. Future research should endeavor to replicate these findings in other mindfulness-based interventions. Extending this research to diverse populations is important, with a potential to extend to pediatric populations. Empirical validation of our newly introduced conceptual model is needed, comparing it to prior models.27 In the midst of the COVID-19 pandemic with elevated stress and uncertainty, mindfulness may be an especially beneficial treatment approach for those with migraine.53
AUTHOR CONTRIBUTIONS
Study concept and design: Paige M. Estave, Rebecca Erwin Wells. Acquisition of data: Mariam Shakir, Rebecca Erwin Wells. Analysis and interpretation of data: Paige M. Estave, Caitlyn Margol, Summerlyn Beeghly, Reid Anderson, Mariam Shakir, Alexandra Coffield, Julia Byrnes, Nathaniel O'Connell, Elizabeth Seng, Paula Gardiner, Rebecca Erwin Wells. Drafting of the manuscript: Paige M. Estave, Caitlyn Margol, Summerlyn Beeghly, Reid Anderson, Mariam Shakir, Alexandra Coffield, Julia Byrnes, Rebecca Erwin Wells. Revising it for intellectual content: Elizabeth Seng, Paula Gardiner, Rebecca Erwin Wells. Final approval of the completed manuscript: Paige M. Estave, Caitlyn Margol, Summerlyn Beeghly, Reid Anderson, Mariam Shakir, Alexandra Coffield, Julia Byrnes, Nathaniel O'Connell, Paula Gardiner, Rebecca Erwin Wells, Elizabeth Seng.
ACKNOWLEDGMENTS
We are grateful to all the participants who volunteered for this study and provided unique insights through qualitative interviews. We are thankful for the tremendous support of Charles R. Pierce, Kate Furgurson, Geena George, and Anissa Berger, in addition to the tremendous support of a multitude of other volunteer students, including Nicole Rojas, Hudaisa Fatima, Obiageli Nwamu, MA, Vinish Kumar, Rosalia Arnolda, Paige Brabant, Danika Berman, Nicholas Contillo, Flora Chang, Easton Howard, Camden Nelson, and Carson DeLong. We are thankful for Steven Eugene Albertson's talent for graphical design and help with our Conceptual Model Figure. We appreciate the support from the Qualitative and Patient-Reported Outcomes Developing Shared Resource of the Wake Forest Baptist Comprehensive Cancer Center, the Wake Forest Clinical Translational Science Institute (CTSI), the Clinical Research Unit staff and support, and the Research Coordinator Pool, funded by the National Center for Advancing Translational Sciences (NCATS), National Institutes of Health, through Grant Award Number UL1TR001420.
FUNDING INFORMATION
American Pain Society Grant, Sharon S. Keller Chronic Pain Research Program, (PI-Wells); NCCIH K23AT008406 (Wells) and NINDS K23NS096107 (Seng), NCCIH R01AT011005 (Seng and Wells), NCCIH R01AT011502 (Wells), NCI R01CA266995 (Wells), American Headache Society Fellowship (Wells) and the Headache Research Fund of the John Graham Headache Center, Brigham and Women's Faulkner Hospital. Research supported in part by the Qualitative and Patient-Reported Outcomes Developing Shared Resource of the Wake Forest Baptist Comprehensive Cancer Center's NCI Cancer Center Support Grant P30CA012197 and the Wake Forest Clinical and Translational Science Institute's NCATS Grant UL1TR001420. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
CONFLICT OF INTEREST STATEMENT
Paige M. Estave, Caitlyn Margol, Summerlyn Beeghly, Reid Anderson, Mariam Shakir, Alexandra Coffield, Julia Byrnes, Nathaniel O'Connell, Paula Gardiner, and Rebecca Erwin Wells have no conflicts of interest to report. Elizabeth Seng has consulted for GlaxoSmithKline and Click Therapeutics.
CLINICAL TRIALS REGISTRATION NUMBERS
clinicaltrials.gov Identifiers: NCT01545466 and NCT02695498.