Migraine, Migraine in Adults, Sleep Disorder (Disorder), Stress
Conditions
Keywords
migraine, behavioral therapy, sleep hygiene, stress management, migraine triggers, non-medical prophylaxis
Brief summary
The MiSleepS study investigates how sleep disturbances and stress are linked to migraine attacks. Participants wear a device called a WHOOP band, which tracks sleep and body signals, and answer brief daily questions via a smartphone app about their sleep, stress levels, and migraine symptoms. The goal is to identify personal patterns that may contribute to migraine. Based on these insights, participants receive individualized recommendations to improve their sleep and daily routines - aiming to reduce migraine attacks in the long term without medication. The study is conducted at the University Hospital Zurich and is aimed at adults with episodic migraine.
Detailed description
Migraine is a widespread neurological disorder affecting more than one billion people worldwide, and is among the leading causes of disability, particularly in women. It is characterized by episodic or chronic headaches and often accompanied by nausea, photophobia, and cognitive impairment. Despite advances in pharmacological therapies - such as the advent of CGRP antagonists - a large proportion of patients remain undertreated or refractory to standard interventions. Critically, migraine is influenced by multiple behavioral and environmental triggers, among which sleep disturbances and stress are consistently among the most frequently reported and most modifiable. However, their complex and often bidirectional interactions with migraine are still not fully understood, and most available research is limited by methodological constraints, including short observation periods, retrospective data, and insufficient attention to sex and gender variables. The Migraine Sleep Study (MiSleepS) is a prospective, two-phase clinical study aiming to investigate the role of sleep, circadian rhythm, and stress as dynamic triggers of migraine and to evaluate the effectiveness of individualized, non-pharmacological behavioral interventions. Conducted at the University Hospital Zurich, this monocentric study will combine high-resolution physiological data captured via the WHOOP 5.0 wrist-worn wearable device with real-time, ecological momentary assessments (EMA) collected through the SEMA3 smartphone app. These dual digital tools enable continuous monitoring of key variables such as sleep duration, sleep architecture, heart rate variability, perceived stress, and migraine occurrence and severity. Participants will undergo a five-week observational phase (phase A), during which their natural sleep-stress-migraine interactions will be captured without interference. An interim analysis will be conducted to identify individual behavioral and circadian profiles, including insomnia-like patterns, sleep deprivation, social jetlag, and chronotype mismatch. Based on these results, participants will be stratified into clusters and assigned a tailored behavioral plan to address their specific profile. In the subsequent six-week intervention phase (phase B), participants will implement these behavioral strategies, supported by remote follow-ups and daily app-based tracking. The primary endpoint will be the change in monthly migraine days, while secondary endpoints include migraine severity, sleep quality, stress levels, and adherence to recommendations. To control for observation-related confounding - such as the Hawthorne effect - a run-in cohort of the first ten participants will follow a modified protocol. While they undergo the same assessment and tracking procedures, they will not receive any behavioral recommendations in phase B. This approach allows for differentiation between improvements due to heightened self-awareness and those attributable to the targeted intervention itself. The study further aims to examine sex- and gender-related differences in migraine pathophysiology and response to behavioral interventions, using validated tools such as the Stanford Gender-Related Variables for Health Research (GVHR) score (Nielsen et al., 2021). By integrating physiological, psychological, and gender-related dimensions, MiSleepS aspires to develop a more individualized understanding of migraine and to explore scalable, low-risk, non-pharmacological treatment strategies that can be implemented in clinical practice.
Interventions
BEHAVIORALProfile-based behavioral sleep and stress management
After completion of the 5-week observational phase (phase A), the study team will conduct an interim analysis integrating WHOOP biometric data and SEMA3 self-reports on migraine, sleep, and stress. The aim is to identify individual sleep-stress patterns linked to migraine activity. Based on predefined criteria, participants will be assigned to one of four behavioral profiles: insomnia-like, sleep deprivation, social jetlag, or circadian misalignment. Mixed or unclassified cases will be grouped separately. All participants receive general behavioral recommendations on sleep hygiene, scheduling, and stress management. WHOOP-based personalized tips (e.g., optimal sleep windows, recovery days) will be encouraged. Profile-based participants also receive targeted prioritization of interventions most relevant to their sleep-migraine pattern, including techniques such as rhythm stabilization, relaxation training, or strategic light exposure.
Sponsors
Susanne Wegener
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
OTHER
Masking
NONE
Intervention model description
MiSleepS is a monocentric, prospective clinical trial with a single-group, within-subject design. All participants follow the same protocol: a 5-week observational phase (Phase A), followed by a 6-week interventional phase (Phase B). After interim analysis, participants receive personalized behavioral recommendations based on their individual sleep-stress-migraine profile. The first 10 participants (run-in cohort) undergo the same assessments but do not receive an intervention in Phase B. This subgroup serves to control for observational bias (Hawthorne effect) but is not considered a separate study arm. No randomization or parallel group assignment is applied. The study evaluates within-subject changes in migraine burden before and after the tailored intervention, consistent with a single-group interventional model.
Eligibility
Sex/Gender
ALL
Age
18 Years to 65 Years
Healthy volunteers
No
Inclusion criteria
* Adults aged between 18 and 65 years * Diagnosis of episodic migraine according to The International Classification of Headache Disorders (ICHD-3) criteria confirmed by our headache specialists * 4 to 14 headache days per month (mean value based on the 3 months prior to study enrollment) * Ability to give informed consent and to adhere to the study protocol * Sufficient German language comprehension to follow the study procedures and answer all questions related to the study outcomes * Stable migraine medication regimen for the past 3 months and throughout the study period
Exclusion criteria
* Diagnosis of sleep disorders that could interfere with the sleep intervention, such as obstructive sleep apnea with an apnea-hypopnea index (AHI) \> 15, Restless Legs Syndrome, frequent (i.e. weekly) Non-rapid eye movement (NREM) sleep parasomnia, REM Behavior Disorder (RBD) * Current diagnosis of a psychiatric disorder that is inadequately treated or therapy-resistant and may interfere with study participation or adherence to study procedures (this includes, but is not limited to: schizophrenia, schizoaffective disorder, bipolar disorder (type I), post-traumatic stress disorder with active symptoms, or major depressive disorder with ongoing functional impairment despite treatment). Diagnosis must be confirmed by clinical history or treating physician. * Regular use of benzodiazepines and other central nervous system (CNS)-depressant substances (self-reported) * Concomitant steroid medication (self-reported) * Known or suspected alcohol, drug or medication abuse (i.e. \> 0.5 l wine or 1 l beer per day) * Inability to follow the procedures of the study (e.g., due to language problems, cognitive deficits, instable home situation) * Concurrent participation in another study involving drug and behavioral interventions within 3 months prior to and during the present study, as well as participation in an ongoing study with data collection through SEMA3 * Planned medical intervention of substantial relevance requiring hospitalization for more than 24 hours (e.g. surgery) during intervention (routine assessments, e.g. check-ups will be allowed) * Shift work with working during the night * Travelling more than 2 time zones in the last month before the observation or intervention periods or during the study * Persons who are pregnant or breastfeeding
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Decrease in the number of migraine days (6 weeks) | From phase A (baseline) to the end of phase B (6 weeks of intervention). | The primary endpoint is the change in the number of migraine days per month from phase A (baseline) to the end of phase B, following 6 weeks of intervention. A reduction of 30% in self-reported monthly migraine days is defined as the primary outcome measure. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Association between total sleep time and migraine onset | Phase A (5 weeks). | Correlation between objectively measured total sleep time (hours, derived from wearable device data) and the occurrence of migraine onset (yes/no), assessed using daily entries. |
| Association between sleep latency and migraine onset | Phase A (5 weeks). | Correlation between sleep latency (minutes, derived from wearable device data) and the occurrence of migraine onset (yes/no), assessed using daily entries. |
| Association between sleep-wake time variability and migraine onset | Phase A (5 weeks). | Correlation between variability in sleep-wake timing (derived descriptively from wearable device data) and the occurrence of migraine onset (yes/no), assessed using daily entries. |
| Association between recovery score and migraine onset | Phase A (5 weeks). | Correlation between recovery scores (as calculated by the wearable device application) and the occurrence of migraine onset (yes/no), assessed using daily entries. |
| Association between perceived stress and migraine onset | Phase A (5 weeks). | Correlation between perceived stress measured daily using the Stress Numeric Rating Scale-11 (Stress NRS-11) and the occurrence of migraine onset (yes/no). |
| Difference in total sleep duration between nights with and without migraine attacks | Phase A (5 weeks) | Mean difference in total sleep duration (hours), derived from wearable device data and daily entries, comparing nights with reported migraine attacks (yes) versus nights without migraine attacks (no), as recorded in the SEMA3 app. |
| Difference in sleep quality between nights with and without migraine attacks | Phase A (5 weeks) | Mean difference in subjective sleep quality, assessed via daily entries, comparing nights with reported migraine attacks (yes) versus nights without migraine attacks (no). |
| Difference in light, deep, and REM-sleep proportion between nights with and without migraine attacks | Phase A (5 weeks) | Mean difference in the proportion of time spent in light sleep, deep sleep, and REM-Sleep (% per night), derived from wearable device data, comparing nights with reported migraine attacks (yes) versus nights without migraine attacks (no). |
| Change in subjective sleep quality from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention, 6 weeks). | Mean change in subjective sleep quality measured using a validated questionnaire (Pittsburgh Sleep Quality Index, PSQI) and daily entries in the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). |
| Change in total sleep duration from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention, 6 weeks). | Mean change in total sleep duration (hours per night), derived from wearable device data, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). |
| Change in sleep consistency score from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention, 6 weeks). | Mean change in sleep consistency score, derived from wearable device data, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). |
| Change in sleep onset variability from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention, 6 weeks). | Mean change in variability of sleep onset timing, derived from wearable device data, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). |
| Change in perceived stress from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention, 6 weeks). | Mean change in perceived stress measured daily using the Stress Numeric Rating Scale-11 (0-10) via the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). |
| Decrease in the number of migraine days (3 months) | From phase A (baseline) to the end of the study (3 months) | Change in the number of migraine days per month from phase A (baseline) to the end of the study (3 months) (outcome measures: 30% reduction in the number of self-reported monthly migraine days before vs. after the intervention) |
| Change in migraine severity from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention phase) | Mean change in migraine severity, measured as average daily Numeric Rating Scale (NRS; 0-10) scores recorded in the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). Analysis includes only participants reporting migraine episodes in both phases. |
| Change in migraine attack duration from baseline to end of intervention | From phase A (baseline) to the end of phase B (intervention phase) | Mean change in migraine attack duration, measured as mean hours per episode based on participant self-report in the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the intervention phase (Phase B). Analysis includes only participants reporting migraine episodes in both phases. |
| Change in migraine severity from baseline to 3-month follow-up | From phase A (baseline) to 3-month follow-up | Mean change in migraine severity, measured as average daily Numeric Rating Scale (NRS; 0-10) scores recorded in the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the study (3-month follow-up). Analysis includes only participants reporting migraine episodes in both periods. |
| Change in migraine attack duration from baseline to 3-month follow-up | From phase A (baseline) to 3-month follow-up | Mean change in migraine attack duration, measured as mean hours per episode based on participant self-report in the SEMA3 app, comparing the observational phase (Phase A, baseline) to the end of the study (3-month follow-up). Analysis includes only participants reporting migraine episodes in both periods. |
Countries
Switzerland
Contacts
CONTACTSusanne Wegener
CONTACTMarie Therese Kleinsorge
PRINCIPAL_INVESTIGATORSusanne Wegener
University Hospital Zurich, Department of Neurology
Outcome results
None listed