Home-Based Trials in Adolescent Migraine

Randomized trials are needed to identify safe and effective migraine preventive treatments for children.Conventional trials typically require frequent in-person study visits. 1 Many families decline study participation, citing time and distance.2 Home-based trials are a novel, participant-centered design innovation wherein most or all study procedures are completed remotely using technology.Melatonin is safe and effective for migraine prevention in adults.3 In this pilot study, we assessed the feasibility of a homebased trial of melatonin for adolescent migraine prevention.Methods | We conducted a randomized, double-blind, placebocontrolled pilot study of melatonin, 3 mg (Rugby Laboratories), vs placebo for migraine prevention in children aged 12 to 17 years.Participants were recruited via our clinic, flyers, social media, print advertisements, and parent letters (Table 1).Screening occurred on our study website; eligible participants were invited for a 1-time study center enrollment visit.Diagnosis of migraine by International Classification of Headache Disorders, 3rd edition (beta version) criteria 4 was confirmed by a pediatric headache neurologist.Remaining study procedures were conducted from home.The institutional review board of the University of California, San Francisco approved this study (clinicaltrials.govidentifier: NCT02344316).Parents provided written informed consent and adolescents provided assent.Participants received a nightly text message on their smartphone that linked to a secure web-based electronic headache diary.Study staff monitored diary compliance and provided reminders.After a 28-day baseline, those with 80% or greater diary compliance and the requisite number of headaches were randomized to melatonin or placebo for 12 weeks.Study medication was shipped to homes.Participants recorded sleep using Fitbits (Fitbit Inc).Sleep data synced to participants' smartphones or computers and automatically transferred to the study database via application program interface calls.Adverse events were assessed via telephone twice monthly.The enrollment goal was 30 participants.The aims of this study were to (1) determine the success vs cost of various recruitment strategies, (2) demonstrate enrollment feasibility, (3) estimate study completion rate, (4) estimate variance in headache outcomes using home-trial methodology, and (5) assess adverse events.Mann-Whitney, χ 2 , and Fisher exact tests as well as multivariable linear regression were used.