Epidemics and pandemics pose a sporadic and sometimes severe threat to human health. How should policymakers prioritize preventing and preparing for such events, relative to other needs? To answer this question, we used computational epidemiology and extreme events modeling simulations to estimate the risk of future mortality from low-frequency, high severity epidemics and pandemics in two important categories—respiratory diseases (in particular those caused by pandemic influenza viruses and novel coronaviruses) and viral hemorrhagic fevers (VHFs) such as Ebola and Marburg virus diseases. We estimate a global annual average of 2.5 million deaths, attributed to respiratory pandemics. We estimate an annual average of 26,000 VHF deaths globally, 72 percent of which would be in Africa. Annual averages conceal vast year by year variation, and the reported analyses convey that variation—as well as variation across regions and by age. Our estimates suggest that both the frequency and severity of such events is higher than previously believed—and this is likely to be a lower bound estimate given the focus of this chapter on deaths caused by a subset of pathogens. Our simulations suggest that an event having the mortality level of COVID-19 should not be considered a “once in a century” risk, but rather occurring with an annual probability of 2-3 percent (that is, a one in 33-50-year event). Despite the substantial uncertainty in heavy-tail distributions, policymakers can use these estimates to develop risk-informed financing, prevention, preparedness, and response plans.
