Predicting the cumulative number of disaster deaths during the early stage of earthquakes

BACKGROUND: The quantitative measurement of the anticipated number of disaster deaths is very important shortly after the mainshock because the forecasted fatalities could help determine the size of the health and medical services team to be deployed. This study aimed to devise a simple method to predict the cumulative number of deaths during the immediate or early stage of a large earthquake. METHODS: We analyzed six earthquakes in Japan that involved at least 20 deaths, 1990–2018. Analyzing statistical patterns in the cumulative number of deaths, we used three models—the exponential model, the Weibull model, and the percentile-based model—to predict the likely number of deaths during the early stage of earthquakes. RESULTS: The median time required to reach the median number of deaths was 2.2 (interquartile range: 1.5, 3.8) days from the mainshock. By only multiplying the cumulative number of deaths as on day 2 by a factor of two, the likely number of deaths was calculated using the percentile-based method. The validity of this simple method was better than the results from day 4 using the parametric models. The Great East Japan earthquake was exceptionally large and difficult to predict in real time, and it involved a large number of fatalities following a tsunami. CONCLUSIONS: For all other earthquakes, the median number of deaths was reached on day 2. Even in a setting with poor technical resources, the predicted number of deaths can be obtained by multiplying the reported cumulative number on day 2 by a factor of two.