Assessing community response to wildfire smoke: A multimethod study using social media.

Abstract

The purpose of this study was to assess health-related responses to wildfire smoke on social media. We examined whether seasonal wildfire smoke is an active topic on Twitter, the correlation between fine particulate matter (PM2.5 ) and Twitter search terms, and dimensions of community-level expression to wildfire smoke through tweets. Search terms were identified using a conceptual model developed and refined by healthcare providers and public health experts. Wildfire-related tweets were downloaded from Twitter users in Spokane, Washington during the 2017 and 2018 wildfire seasons. PM2.5 data were correlated with the search terms. A subset of tweets was deductively and then inductively coded to identify perceptions and behavioral responses to wildfire smoke. Seasonal wildfire smoke is an active topic on Twitter. The term "smoke" was strongly correlated with poor air quality and "unhealthy" was moderately correlated. Deductive analyses revealed a multidimensional response to wildfire smoke. Inductive analysis identified new areas of concern, such as pet and animal health. Social media is a lens through which public health professionals can assess and respond to local community needs. Findings will be used to broaden the conceptual model, enhance ongoing surveillance of community-identified health risks, and communicate protective actions.