Observational Study Using Machine Learning to Predict the Effectiveness of Migraine Preventive Treatments in Adults 18 to 99 Years Old

What this Trial Is About

Researchers are evaluating how well machine learning ML models can predict the effects of migraine preventive treatments for adults with episodic or chronic migraine. The study aims to identify patient features such as sociodemographic information, headache patterns, and related health conditions to help predict which treatments may work best. This observational trial compares the predicted treatment effects from ML models to the actual observed effects after treatment. Participants will start by providing information through a phone consultation and questionnaire. Headache days are tracked for 4 weeks before beginning a migraine preventive prescribed by their physician. The study then monitors headache days during the first 12 weeks of treatment, divided into 28-day periods, to assess if the preventive reduces headache frequency by 50% or more. Participants may undergo up to two treatment periods, each followed by a phone call to assess outcomes. The total participation time can be up to 28 weeks. During the study, researchers collect sociodemographic, headache, and comorbidity data before treatment starts. Participants will have follow-up phone calls after each treatment period to evaluate outcomes. The main measure is the accuracy of ML models in predicting treatment response based on a 50% reduction in headache days. Secondary measures include time to treatment response and success rates of first-line therapies. The study observes standard care without interfering with treatment decisions and does not include blinding or control groups.

A Pragmatic Trial of Machine Prescription for Migraine