The abstract is read at the publisher; the summary is JClub's.
Local influenza forecasts for Emergency Department influenza visit percentages across 173 U.S. HSAs beat state-based forecasts in 98.8% (1-week), 90.8% (2-week), and 78.6% (3-week) of areas.
- Why it matters: State-level forecasting can hide within-state differences that matter for local outbreak planning, reducing usefulness for timely public health response.
- What they did: They built and tested gradient boosting quantile regression forecasts for influenza-attributable ED visit percentages in HSAs, then compared accuracy against forecasts derived only from state-level data.
- The result: Across 1–3 week horizons, local models achieved mean weighted interval scores averaging 39.2%, 19.6%, and 11.4% lower than state-based forecasts, supporting finer-scale forecasting for HSA populations over 250,000.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.