From surveillance data to public health action: the usability gap in WHO digital influenza surveillance

Background Interpretable surveillance evidence connects reported influenza activity with a defined analytical task. Under what conditions can influenza surveillance records shared through the World Health Organization (WHO) FluNet and FluID provide interpretable evidence for local forecasting and alert assessment? Methods We assessed valid-rate availability and continuity in WHO FluNet and FluID for 2015–2025 and compared seasonal-naive, domestic, and target-excluded international models in rolling held-out evaluations for 2022–2025, with alert trade-offs expressed as scenario costs. Sensitivity analyses examined zero coding, model specification, and the COVID-19 period, whereas official-document mapping and exploratory analyses assessed recording context and background conditions. Results Valid-rate construction was possible for 162 FluNet, 61 ILI (influenza-like illness), and 97 severe acute respiratory infection (SARI) reporting codes. Median FluNet coverage increased from 50.0% in the primary snapshot to 76.9% in the zero-restored later snapshot, making zero coding consequential for interpreting coverage. Domestic models outperformed seasonal repetition in the primary comparison, while the tested international aggregate yielded negative mean increments across all robustness specifications. SARI sensitivity gains entailed more false alerts, and the ILI alert pattern was not robust. Restoring zero-positive partner weeks improved the 1-week SARI contrast by 13.1 per 10,000 admissions (95% interval, 0.8–33.4). Conclusion The findings identify conditions for constructing and evaluating interpretable shared evidence: valid local baselines, explicit recording context, and assessment for a specified outcome, horizon, threshold, and loss preference. Continuous domestic surveillance provides the comparator for judging candidate external inputs. Metadata and feedback reforms offer an agenda for testing these conditions in local practice.

Authors

Institutions

Publication Details

Journal
Frontiers in Public Health
Published
2026-11-19
DOI
https://doi.org/10.3389/fpubh.2026.1960979
Primary Topic
Data-Driven Disease Surveillance
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

From surveillance data to public health action: the usability gap in WHO digital influenza surveillance

Mingfei Ma, Yu Teng
Frontiers in Public Health
Data-Driven Disease Surveillance
article

From surveillance data to public health action: the usability gap in WHO digital influenza surveillance

Mingfei Ma, Yu Teng
article en

Abstract

Background Interpretable surveillance evidence connects reported influenza activity with a defined analytical task. Under what conditions can influenza surveillance records shared through the World Health Organization (WHO) FluNet and FluID provide interpretable evidence for local forecasting and alert assessment? Methods We assessed valid-rate availability and continuity in WHO FluNet and FluID for 2015–2025 and compared seasonal-naive, domestic, and target-excluded international models in rolling held-out evaluations for 2022–2025, with alert trade-offs expressed as scenario costs. Sensitivity analyses examined zero coding, model specification, and the COVID-19 period, whereas official-document mapping and exploratory analyses assessed recording context and background conditions. Results Valid-rate construction was possible for 162 FluNet, 61 ILI (influenza-like illness), and 97 severe acute respiratory infection (SARI) reporting codes. Median FluNet coverage increased from 50.0% in the primary snapshot to 76.9% in the zero-restored later snapshot, making zero coding consequential for interpreting coverage. Domestic models outperformed seasonal repetition in the primary comparison, while the tested international aggregate yielded negative mean increments across all robustness specifications. SARI sensitivity gains entailed more false alerts, and the ILI alert pattern was not robust. Restoring zero-positive partner weeks improved the 1-week SARI contrast by 13.1 per 10,000 admissions (95% interval, 0.8–33.4). Conclusion The findings identify conditions for constructing and evaluating interpretable shared evidence: valid local baselines, explicit recording context, and assessment for a specified outcome, horizon, threshold, and loss preference. Continuous domestic surveillance provides the comparator for judging candidate external inputs. Metadata and feedback reforms offer an agenda for testing these conditions in local practice.

Frontiers in Public HealthVol. 14
Dalian Maritime University (CN)
Data-Driven Disease Surveillance
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon