Addressing delayed case reporting in infectious disease forecast modeling

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Infectious disease forecasting is crucial for public health and policy.
• Reporting delays in case data can lead to underestimation of disease burden.
• Accurate forecasting is essential for timely and effective public health interventions.

Purpose of the Study:

• To develop a general framework for addressing reporting delays in infectious disease forecasting.
• To improve the accuracy and reliability of disease forecasts.
• To provide guidance on handling reporting delays in practical forecasting efforts.

Main Methods:

• Proposed a framework leveraging historical case data or external internet data to estimate reporting errors.
• Developed methods to adapt forecasting pipelines for under- or over-reporting.
• Applied methods to dengue fever and influenza-like illness (ILI) data.
• Conducted simulation studies to compare method performance and robustness.

Main Results:

• Implementing correction methods for reporting delays consistently increased forecasting accuracy and prediction coverage.
• Method performance varied, with some requiring specific data quality or external sources.
• Alternative strategies like data exclusion and sensitivity analysis were evaluated.

Conclusions:

• Accounting for reporting delays is vital for accurate infectious disease forecasting.
• The proposed framework offers practical strategies and guidance for public health practitioners.
• The study highlights the importance of addressing data limitations in epidemiological modeling.