In July 2009, a Distribute Working Group wrote a letter to Nature in response to an earlier article by Ginsberg et al. on using search engine query data for detecting epidmics.
Through retrospectively analyzing billions of internet search queries, Ginsberg et al. identified a collection of specific searches that track the course of influenza-like illness (ILI) reported by the US Centers for Disease Control and Prevention (CDC). Prospective monitoring during 2007-2008 found high correlation between Google estimates and CDC-reported ILI, with next-day timeliness compared to the 1-2 week delay reported in traditional CDC ILI surveillance. The assertion by Ginsberg et al., however, that internet search term estimates enable public health officials to respond better to seasonal and pandemic influenza does not take into account the current practice of public health, or the state of the art in electronic disease surveillance.
Searching for better flu surveillance? A brief communication arising from Ginsberg et al. Nature 457, 1012-1014 (2009). Nature Precedings, 28 July 2009. Donald R. Olson, Atar Baer, Michael A. Coletta, Lana Deyneka, Ryan Gentry, Amy Ising, Erin L. Murray, Marc Paladini, Justin Pendarvis, Karl Soetebier, Kevin J. Konty, Jill Schulmann, Jeffrey Engel, Julia Gunn, Robert T. Rolfs & Farzad Mostashari.
Manuscript available online
Tags: Nature, article, flu
- npre20093493-1.pdf, 1 MB
▶ Reply to This