Methodological guidelines to estimate population-based health indicators using linked data and/or machine learning...

Area of Science:

• Health Informatics
• Data Science
• Public Health Research

Background:

• Estimating health indicators using linked data and AI varies across Europe.
• Challenges include data interoperability, diverse data sources, and limited big data analytical skills.
• Existing methods for health indicator estimation from linked data are inconsistent.

Purpose of the Study:

• To develop methodological guidelines for calculating population-based health indicators.
• To guide European countries in utilizing linked data and machine learning (ML) techniques.
• To establish standardized, high-quality research methods for population health.

Main Methods:

• A systematic, step-wise approach was employed.
• Included a comprehensive scientific literature review.
• Identified best practices from European countries and developed a guideline checklist.

Main Results:

• Methodological guidelines for population health indicator studies were developed.
• Guidelines offer a systematic approach for linked data and ML-based research.
• A detailed checklist covers study design, data preparation, analysis, and limitations.

Conclusions:

• This is the first set of guidelines for population health studies using linked data and ML.
• Guidelines will support researchers in adopting systematic, high-quality research methods.
• Emphasizes the need for advanced methodologies to improve population health information and outcomes.