No abstract

Objective: To evaluate factors associated naturalistically with adherence to a mobile headache diary. Background: Self-monitoring (keeping a headache diary) is commonly used in headache to enhance diagnostic accuracy and evaluate the effectiveness of headache therapies. Mobile applications are increasingly used to facilitate keeping a headache diary. Little is known about factors associated with adherence to mobile headache diaries. Methods: In this naturalistic longitudinal cohort study, people with headache (n = 1,561) registered to use Curelator Headache ® (now called N1-Headache ®), an application that includes a mobile headache diary, through their physician (coupon), or directly through the website or app store using either a paid or free version of the application. Participants completed baseline questionnaires and were asked to complete daily recordings of headache symptoms and other factors for at least 90 days. Baseline questionnaires included headache characteristics and migraine disability. Daily recordings included headache symptoms and anxiety ratings. Adherence to keeping the headache dairy was conceptualized as completion (kept the headache diary for 90 days), adherence rate (proportion of diary days completed 90 days after registration), and completion delay (the number of days past 90 days after registration required to complete 90 days of headache diary). Results: The majority of participants reported migraine as the most common headache type (90.0%), and reported an average of 30.8 headache days/90 days (SD = 24.2). One-third of participants completed 90 days of headache diary (32.4%). Endorsing higher daily anxiety scores [8/10 OR = 0.97 (95% CI = 0.96, 0.99); 10/10 OR = 0.96 (95% CI = 0.91, 0.99)] was associated with lower odds of completion, whereas higher age [OR = 1.04 (95% CI = 1.03, 1.05)], and downloading the app paid vs. free [OR = 4.27 (95% CI = 2.62, 7.06)], paid vs. coupon [OR = 2.43, 95% CI = 1.41, 4.26)], or through a physician coupon vs. free [OR = 1.75 (95% CI = 1.27, 2.42) were associated with higher odds of completion. The median adherence rate at 90 days was 0.34 (IQR = 0.10-0.88), indicating that half of participants kept 34 or fewer days 90 diary days after registration. Endorsing high daily anxiety scores [5/10 OR = 0.98 (95% CI = 0.97, 1.00); 8/10 OR = 0.96 (95% CI = 0.94, 0.98); 10/10 OR = 0.96 (9% CI = 0.92, 0.98)] and higher age [OR = 1.05 (95% CI = 1.04, 1.07)] were associated with lower odds of adhering at 90 days, whereas downloading the app paid vs. free [OR = 9.63 (95% CI = 4.61, 25.51)], paid vs. coupon [OR = 2.39, 95% CI = 1.27, 5.10)], or through a physician coupon vs. free [OR = 4.01 (95% CI = 2.54, 7.26) were associated with higher odds of adhering at 90 days. Among completers, the median completion delay was 6.0 days (IQR = 2.0-15.0). Among completers, endorsing high daily anxiety scores [9/10 OR = 1/06 (95% CI = 1.01, 1.12)] and younger age [OR = 0.98 (95% CI = 0.97, 1.00)] was associated with completion delay; downloading the app through physician ...

Objectives To describe patterns of perceived stress across stages of the migraine cycle, within and between individuals and migraine episodes as defined for this study. Methods Individuals with migraine aged ≥18 years, who were registered to use the digital health platform N1‐HeadacheTM, and completed 90 days of daily data entry regarding migraine, headache symptoms, and lifestyle factors were eligible for inclusion. Perceived stress was rated once a day at the participant’s chosen time with a single question, “How stressed have you felt today?” with response options graded on a 0‐10 scale. Days were categorized into phases of the migraine cycle: Ppre = pre‐migraine headache (the 2 days prior to the first day with migraine headache), P0 = migraine headache days, Ppost = post‐migraine headache (the 2 days following the last migraine day with migraine headache), and Pi = interictal days (all other days). Episodes, defined as discrete occurrences of migraine with days in all 4 phases, were eligible if there was at least 1 reported daily perceived stress value in each phase. Individuals with ≥5 valid episodes, and ≥75% compliance (tracking 90 days in 120 calendar days or less) were eligible for inclusion in the analysis. Results Data from 351 participants and 2115 episodes were included in this analysis. Eighty‐six percent of the sample (302/351) were female. The mean number of migraine days per month was 6.1 (range 2‐13, standard deviation = 2.3) and the mean number of episodes was 6.0 (range 5‐10, standard deviation = 1.0) over the 90‐day period. Only 8 (8/351, 2.3%) participants had chronic migraine (defined as 15 or more headache days per month with at least 8 days meeting criteria for migraine). Cluster analysis revealed 3 common patterns of perceived stress variation across the migraine cycle. For cluster 1, the “let down” pattern, perceived stress in the interictal phase (Pi) falls in the pre‐headache phase (Ppre) and then decreases more in the migraine phase (P0) relative to Pi. For cluster 2, the “flat” pattern, perceived stress is relatively unchanging throughout the migraine cycle. For cluster 3, the “stress as a trigger/symptom” pattern, perceived stress in Ppre increases relative to Pi, and increases further in P0 relative to Pi. Episodes were distributed across clusters as follows: cluster 1: 354/2115, 16.7%; cluster 2: 1253/2115, 59.2%, and cluster 3: 508/2115, 24.0%. Twelve participants (12/351, 3.4%) had more than 50% of their episodes fall into cluster 1, 216 participants (216/351, 61.5%) had more than 50% of their episodes fall into cluster 2, and 25 participants (25/351, 7.1%) had more than 50% of their episodes fall into cluster 3. There were 40 participants with ≥90% of their episodes in cluster 2, with no participants having ≥90% of their episodes in cluster 1 or 3. Conclusions On an aggregate level, perceived stress peaks during the pain phase of the migraine cycle. However, on an individual and episode basis, there are 3 dominant patterns of perceived stress variation across the migraine cycl...

The usual Hotelling T 2 control chart is not appropriate for monitoring processes where the quality characteristic is a mixture. The composition of mixtures are vectors of positive elements that represent parts of a whole, to which standard multivariate techniques are not appropriate due to their restricted sample space. There are many applications where a mixture is monitored against time, such as in the chemical industry, product composition, impurity profile, or gas components analysis. In this paper, a multivariate control chart for individual compositional observations based on the T 2 statistic is proposed and compared with the typical one in terms of ARL. We show how results are more consistent with compositional data nature and illustrate implementation in a real-world example.

Objective To investigate the relationship between self‐reported triggers and the occurrence of migraine attacks using a smartphone application. Background One of several issues around the study of migraine attack triggers is that limited available evidence supports whether self‐reported triggers can induce a headache on a particular subject. Methods This is an observational longitudinal cohort study of individuals with migraine registered to track their headaches prospectively using a smartphone application. For 90 days, participants entered daily data about triggers (potential triggers and premonitory symptoms) that may be associated with attack risk, as well as migraine symptoms. The statistical significance of univariate associations between each trigger and migraine recurrent events was determined for each individual. Statistically identified triggers were then compared to self‐reported triggers. Results In 328 individuals (290/328 [88.4%] female; mean [standard deviation] 4.2 [1.5] migraine attacks/month) the mean (standard deviation) number of triggers moderately or highly endorsed per individual was 28.0 (7.7) in individuals presented with up to 38 possible triggers. Of these, an average (standard deviation) of 2.2 (2.1) triggers per individual were statistically associated with increased risk of attacks. Even the most commonly endorsed triggers (sleep quality, stress, tiredness/fatigue, sleep duration, dehydration, neck pain, missed meals, eyestrain, mean barometric pressure, and anxiety) were statistically associated in fewer than one third of individuals suspecting each, with the exception of neck pain (117/302 [38.7%]). Conclusions Individuals with episodic migraine believe that many triggers contribute to their attacks; however, few of these withstand statistical testing at the individual level. Improved personal knowledge of potential triggers and premonitory symptoms may help individuals adopt behavioral changes to mitigate attack risk.