Global impact of World Sepsis Day on digital awareness of sepsis: an evaluation using Google Trends

World Sepsis Day (WSD) was established by the Global Sepsis Alliance in 2012 and is held every 13 of September. One of the objectives is to raise global awareness of sepsis. Despite its high mortality rate [1], an international survey reported that 80–90% of the public in western countries are unfamiliar with sepsis [2]. Anno 2018, public knowledge is no longer solely obtained via television and newspapers, but is largely acquired via the Internet and social media. These resources therefore contribute to digital awareness, and can be used to share knowledge. We aimed to investigate whether WSD is indeed associated with a global increase in digital informationseeking behaviour. By using Google TrendsTM data, which are presented as the relative search volume (RSV) [3], we investigated global digital information-seeking on the terms “sepsis”, “septicemia” and “blood poisoning”. The methods were similar to previous work that investigated the effect of World Thrombosis Day on digital information-seeking [4]. The years 2012–2016, in which WSD was held, were considered as exposure years, with the preceding 5 years (i.e. 2007–2011) serving as control years. The period of interest was defined as the 4 weeks surrounding WSD and compared with the control period, defined as the remaining weeks of the year. Global RSV data were downloaded on the 29 of September 2017 using the “health” category. Data were downloaded for each year separately. Mean differences in RSV, both absolute and as percentages, between the period of interest and the control period were estimated for each year separately. In the years that WSD was held, with the exception of the year 2012 when WSD was first introduced, we found a significant increase in digital informationseeking for the weeks surrounding WSD on terms related to sepsis compared with the remaining weeks of the year (Table 1 and Fig. 1). This was not the case for the years in which WSD was not yet held. The strengths of our approach are the focus on allencompassing terms and the ability of comparing exposure years to control years. However, we assumed