Web search trends on fibromyalgia: development of a machine learning model

Area of Science:

• * Digital Health
• * Health Informatics
• * Chronic Pain Research

Background:

• * Fibromyalgia (FM) is a complex chronic pain condition impacting millions worldwide.
• * The internet is a primary source for health information, influencing patient understanding and engagement.
• * Analyzing online search behavior provides insights into public interest and information-seeking patterns for FM.

Purpose of the Study:

• * To analyze international online search trends for fibromyalgia (FM) from 2020-2024.
• * To identify temporal patterns and forecast future search interest using statistical and machine learning models.
• * To assess the key drivers of online search interest in FM.

Main Methods:

• * Utilized Google Trends data from 16 countries (2020-2024).
• * Employed time-series analysis, linear regression, Mann-Kendall trend test, and seasonal decomposition.
• * Applied Auto-Regressive Integrated Moving Average (ARIMA), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) models, with SHAP values for feature importance.

Main Results:

• * Significant variation in FM search interest across countries; China, UK, USA, and Canada showed highest engagement.
• * Rising trends observed in Brazil, Italy, and UK; declining trends in Argentina, Canada, Greece, and USA.
• * Short-term search history (previous day, week, month) was the most influential predictor in machine learning models.

Conclusions:

• * Online search behavior analysis offers valuable insights for improving fibromyalgia (FM) patient education.
• * Targeted digital health literacy initiatives and awareness campaigns can enhance engagement and knowledge.
• * Optimizing online health resources and decision aids is crucial for future patient support.