Identification of Unplanned Closure of K-12 Schools via Twitter and Online Systematic Search: A Pilot Study of Public Schools in Michigan, September 2015 – June 2016
Background: Unplanned school closures (USCs) can be used to mitigate the transmission of infectious diseases such as influenza or rota-virus. Currently, CDC researchers use online systematic searches to evaluate interpandemic patterns of publically announced USCs lasting ≥1 day in the United States. We determined whether Twitter provides complementary data.
Methods: Twitter handles of Michigan public schools and school districts were identified. All tweets associated with these handles were downloaded. USC-related tweets were identified using five keywords. Descriptive statistics and multivariable logistic regression were performed in R for each data collection method, with individual schools as the unit of analysis. Sensitivity of each data collection method was calculated using all the schools with USCs identified by either method as the denominator (assumed as true positive).
Results: Among 3,469 Michigan public schools, 2,003 maintained their own active Twitter accounts or belonged to school districts with active Twitter accounts. Of these 2,003 schools, in the 2015–2016 school year, at least one USC announcement was identified for 349 schools via the current method only, 678 schools via Twitter only, and 562 schools via both methods. No USC announcements were identified for 414 schools. Assuming true positives for either method, the sensitivity of the current method is 57% (911/1,589) and that of the Twitter method is 78% (1,240/1,589). Rural schools were less likely than city schools to have active Twitter coverage (adjOR = 0.2171, 95% CI, 0.1734–0.2708) and to announce USCs on Twitter (adjOR = 0.3341, 95% CI, 0.2477–0.4491), but more likely to have USCs identified by the current method (adjOR = 2.3420, 95% CI, 1.9230–2.8564).
Conclusions: Each method identified USCs that were missed by the other. Our results suggest that identifying USCs on Twitter complements the current method.
International Conference on Emerging Infectious Diseases (ICEID)
Jackson, Ashley M., Lindsay Mullican, Zion Tsz Ho Tse, Jingjing Yin, Dharamendra Kumar, Isaac Chun-Hai Fung.
"Identification of Unplanned Closure of K-12 Schools via Twitter and Online Systematic Search: A Pilot Study of Public Schools in Michigan, September 2015 – June 2016."
Biostatistics, Epidemiology, and Environmental Health Sciences Faculty Presentations.