USA: Can Twitter and Google help improve heat wave warning systems?

Heat wave warning systems are often used in countries with frequent heat waves to warn the public and prevent heat-related illnesses. A recent study in Environmental Health has shown that tracking tweets on Twitter and searches on Google with heat-related keywords can help provide early and more efficient warnings for heat waves. In this blog, Jihoon Jung, co-author of the research, tells us about the background of the study and how this type of web data can be used to help reduce heat-related illnesses.

By Priya Mistry

[...]

Relating web data to illnesses

We hypothesized that there are more patients with heat-related illnesses when there are more frequent Twitter tweets or Google search results mentioning heat-related keywords. The study collected Twitter messages that mentioned “air conditioning (AC)” and “heat” and Google search data that included weather (heat wave, hot weather), medical (heat exhaustion, heat stroke), recreational (drink, beer, park, pool, swim, water), and adaptation information (AC repair) from May 7 to November 3, 2014, focusing on the state of Florida, U.S. We then tried to find the association between these web data and five emergency room or hospital admission disease categories (cardiovascular disease, dehydration, heat-related illness, renal disease, and respiratory disease).

Research findings

The results show that the number of heat-related illness and dehydration cases exhibited a significant positive relationship with web data. Specifically, heat-related illness cases showed positive associations with tweets (heat, AC) and web searches (drink, heat stroke, park, swim, and tired). In addition, terms such as park, pool, swim, and water tended to show a consistent positive relationship with dehydration cases. Furthermore, the research suggests that tweets and Google searches related to activity patterns such as swimming or going to the park or pool, are positively associated with heat-related illness and dehydration cases. We found inconsistent relationships for renal illnesses and were not able to find any associations between web data and cardiovascular and respiratory illness.

[...]