This research evaluates how effectively Large Language Models can support public health professionals in analyzing and classifying health-related text data.

• Combines 13 datasets (6 external, 7 new) to test LLM performance across public health tasks
• Assesses capabilities for identifying health burdens, epidemiological risk factors, and public health interventions
• Provides systematic evaluation framework for determining LLM suitability in health applications

Practical applications include enhanced disease surveillance, more efficient health data processing, and supporting evidence-based public health decision-making—potentially reducing expert workload while maintaining accuracy.

Evaluating Large Language Models for Public Health Classification and Extraction Tasks