• Center on Health Equity and Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Smartphone App Unable to Accurately Predict Episodic Migraine


In an attempt to develop a multivariable predictive model for days with new-onset migraine headaches, researchers found episodic migraine attacks were not predictable based on self-prediction or on self-reported exposure to common trigger factors.

In an attempt to develop a multivariable predictive model for days with new-onset migraine headaches, researchers found that episodic migraine (EM) attacks were not predictable based on self-prediction or on self-reported exposure to common trigger factors. Study results were published in Headache.

For individuals living with EM, the unpredictability of migraine attacks poses significant challenges. “In theory, accurate real-time forecasts of oncoming migraine attacks could help patients to target medication use and other preventive strategies for susceptible time periods and help decrease the disease burden,” the researchers wrote. Forecasts predicting risks could include data on trigger factors, premonitory symptoms, self-prediction, and physiological signals.

To build upon efforts made in migraine prediction, researchers developed a custom research application for the iPhone and tested the app’s validity through internal assessments.

Between October 2018 and March 2019, investigators recruited patients with EM for a 90-day iPhone app–based, prospective, daily-diary cohort study. All participants were at least 18 years old, lived in the United States, and had suffered from severe headaches for at least 1 year, had migraine headaches within the past 3 months, and had 2 to 10 headache days per month with fewer than 15 headache days per month. Migraine headaches were determined via the 3-question ID migraine diagnostic tool.

After completing an eligibility survey, participants were invited to download the app onto their phones, complete informed consent, enroll in the study, and submit all study-related data.

“At baseline, the application presented one-time surveys on health history and headache history,” the authors wrote. “For the next 90 days, the application administered nightly surveys about headache timing and symptoms, sleep timing, perceived stress, caffeine and alcohol consumption, medication changes, menstruation, premonitory symptoms, and headache self-prediction.”

Daily weather data for each participant were gleaned from the Google Maps Geocoding application programming interface (API) and the Dark Sky weather API.

A total of 178 participants were included in the final analysis. The majority of participants were women (n = 166), and the average patient age was 37 years. The average Migraine Disability Assessment test (MIDAS) score was 32.6, indicating severe headache-related disability, while participants had an average of 7 headache days per month and 16 years since migraine diagnosis. In addition, 43% of the cohort had never taken daily migraine prevention medications.

Overall, 1870 migraine events were included in the analysis. Researchers found:

  • A decrease in caffeine consumption, higher self-predicted probability of headache, a higher level of stress, and times within 2 days of the onset of menstruation were positively associated with next-day migraine risk.
  • The penalized model yielded a within-person C-statistic of 0.56 (95% CI, 0.54-0.58).
  • The multivariable model predicted migraine risk only slightly better than chance (within-person C-statistic: 0.56 [95% CI, 0.54-0.58]).
  • The weak calibration slope of 0.73 (95% CI, 0.57-0.90) was less than 1, suggesting an overfitted model.

“The model performed slightly better than chance in terms of within-person discrimination, but the performance was inadequate for practical use,” the researchers wrote. “Many of the candidate predictors showed a weak and/or noisy correlation with migraine risk, which contributed to the poor discrimination.”

The app-based surveys may have also been susceptible to measurement error due to the brief, simple questions in the end-of-day survey, marking a limitation to the study.

In addition, “the population of episodic migraine patients is heterogeneous,” the authors explained. “The relationships between trigger factors and migraine risk may be more similar among a subset of patients with similar symptoms, perceived triggers, and comorbidities, and thus prediction may be superior in such a subgroup.”

However, despite the findings, the authors noted that the study helped to evaluate the predictive relevance of some common trigger factors of migraine, compare various constructs of the factors in terms of predictive ability, and measure room for improvement in scientific understanding of migraine predictors.

“Incorporating additional predictors, improving the accuracy and frequency of measurement, and studying a more homogeneous migraine population may enable future predictive models to perform better, enable targeted medication use, and reduce the unpredictability of episodic migraine attacks,” the researchers concluded.


Holsteen KK, Hittle M, Barad M, and Nelson LM. Development and internal validation of a multivariable prediction model for individual episodic migraine attacks based on daily trigger exposures. Headache. Published online October 6, 2020. doi:10.1111/head.13960

Related Videos
Will Shapiro, vice president of data science, Flatiron Health
Will Shapiro, vice president of data science, Flatiron Health
Kathy Oubre, MS, Pontchartrain Cancer Center
Emily Touloukian, DO, Coastal Cancer Center
dr krystyn van vliet
dr mitzi joi williams
Stephen Speicher, MD, MS
dr marisa mcginley
Mike Brown, Vice President of Managed Care, Cardinal Health
Related Content
© 2024 MJH Life Sciences
All rights reserved.