Optimal multi-source forecasting of seasonal influenza.

Zeynep Ertem,Dorrie Raymond,Lauren Ancel Meyers,Mark M. Tanaka

doi:10.1371/journal.pcbi.1006236

Zeynep Ertem, Dorrie Raymond + Show 2 more

Open Access

https://doi.org/10.1371/journal.pcbi.1006236

Copy DOI

Abstract

Forecasting the emergence and spread of influenza viruses is an important public health challenge. Timely and accurate estimates of influenza prevalence, particularly of severe cases requiring hospitalization, can improve control measures to reduce transmission and mortality. Here, we extend a previously published machine learning method for influenza forecasting to integrate multiple diverse data sources, including traditional surveillance data, electronic health records, internet search traffic, and social media activity. Our hierarchical framework uses multi-linear regression to combine forecasts from multiple data sources and greedy optimization with forward selection to sequentially choose the most predictive combinations of data sources. We show that the systematic integration of complementary data sources can substantially improve forecast accuracy over single data sources. When forecasting the Center for Disease Control and Prevention (CDC) influenza-like-illness reports (ILINet) from week 48 through week 20, the optimal combination of predictors includes public health surveillance data and commercially available electronic medical records, but neither search engine nor social media data.

Highlights

IntroductionSeasonal influenza epidemics annually result in significant global morbidity and mortality [1], and influenza pandemics can cause catastrophic levels of death, social disruption, and economic loss [2]
We introduce a framework for identifying optimal combinations of data sources, and show that public health surveillance data and electronic health records collectively forecast seasonal influenza better than any single data source alone and better than influenza-related search engine and social media data
Seasonal influenza epidemics annually result in significant global morbidity and mortality [1], and influenza pandemics can cause catastrophic levels of death, social disruption, and economic loss [2]

Summary

Introduction

Seasonal influenza epidemics annually result in significant global morbidity and mortality [1], and influenza pandemics can cause catastrophic levels of death, social disruption, and economic loss [2]. Detection and forecasting of both emergence and peak epidemic activity can inform an effective allocation of resources, surge planning, and public health messaging [1, 3,4,5]. The Center for Disease Control and Prevention (CDC) relies on data from two primary national influenza surveillance systems: (1) the U.S World Health Organization (WHO) and National Respiratory and Enteric Virus Surveillance System (NREVSS) collaborating laboratories (WHO US) and (2) the US Outpatient Influenza-like Illness Surveillance Network (ILINet).

Methods

Results

Discussion

Conclusion