The association between social risk profile and self-reported severe headache or migraine with all-cause mortality risk: A machine learning-based prediction model and interpretability analysis.

This study investigates how social risk factors affect migraines and mortality. Researchers used data from 11,861 US adults collected between 1999 and 2004. They examined the relationship between social risk profile and self-reported severe headaches, also considering all-cause mortality. The study found that higher social risk scores correlated with lower migraine prevalence. Specifically, those with higher scores had a 56% lower chance of migraines. Additionally, higher scores were linked to a 69% reduction in mortality risk among migraine patients. Machine learning techniques helped identify key predictors like age and social risk. The researchers applied various models to assess migraine and mortality risks. They found the XGBoost model most effective for predicting migraines, while the Random Survival Forest was best for mortality. The findings emphasize social determinants' significant role in migraine risk and long-term survival outcomes for patients. Clinical relevance: This research highlights the importance of social factors in migraine risk and mortality, aiding clinicians in developing targeted interventions.