Surveillance Tools Emerging From Search Engines and Social Media Data
By Peggy Denny and edited by Neil M. Bressler, MD
Journal Highlights
JAMA Ophthalmology, September 2016
Download PDF
To explore whether social media use and Internet searches can provide a valuable source of information on epidemiologic factors in eye diseases, Diener et al. studied the correlation of diagnoses of conjunctivitis with data from keyword searches and Tweets. They found a strong association between these data and clinically diagnosed conjunctivitis in electronic medical records (EMRs).
The authors analyzed data from 5,816 clinical encounters of 4,143 patients diagnosed with conjunctivitis from June 3, 2012, to April 26, 2014, at the University of California San Francisco (UCSF) Medical Center and used Spearman rank correlation of each weekly observation to compare demographics and seasonality of nonallergic conjunctivitis with allergic conjunctivitis.
The authors found that the seasonality of clinical diagnoses of nonallergic conjunctivitis correlated strongly with results of Google searches in the United States for the term pink eye (ρ, 0.68) and correlated moderately with Twitter results about pink eye (ρ, 0.38) and with clinical diagnosis of influenza (ρ, 0.33) but did not correlate significantly with seasonality of clinical diagnoses of allergic conjunctivitis at UCSF or with results of Google searches in the United States for the term eye allergy. Seasonality of clinical diagnoses of allergic conjunctivitis at UCSF correlated strongly with results of Google searches in the United States for the term eye allergy (ρ, 0.44) and eye drops (ρ, 0.47).
The authors noted that the clinical data, Google search results, and Twitter posts showed a common pattern: Clinical diagnoses of conjunctivitis detected through EMRs appear to be seasonal and are highly correlated with results of Google searches and correlated with relevant Tweets. In keeping with prior studies of allergic rhinitis, they found that searches related to allergic conjunctivitis peaked in the spring. In addition, EMR data on influenza were also correlated with both EMR data on nonallergic conjunctivitis data and search results for pink eye (probably owing to similar seasonality of the underlying infections).
In conclusion, the authors stated that the information that people post and search for online—and the timing of those activities—can be leveraged to promote better understanding of the epidemiologic factors related to conjunctivitis.
The original article can be found here.