The devastation of hurricanes, earthquakes, tsunamis or tornadoes is often conveyed by how many people have been injured or killed. And based on this, we assess “how bad was it really?” For example, the recent hurricane season in the Atlantic has cost nearly 300 lives in the US and the Caribbean, with Helene killing at least 228 people. These deaths are usually due to flooding of houses and resultant drownings, injuries caused by the destruction of buildings or loss of emergency medical care.
But new research challenges us to see these disasters as broader events that have lasting effects for decades after they hit – whether from stress, financial hardship, pollution or long-term disease. A new paper in Nature magazine develops a methodology to estimate the overall effect of individual tropical cyclones (ie hurricanes and tropical storms) on all causes of mortality across all populations within the US. The authors analysed how mortality rates within a state changed for 20 years after the state was hit by a natural disaster, and took mortality data from 1950 to 2015 to form a longer-term picture.
Their main finding is that a large number of premature deaths in the US could be traced back to tropical cyclone events: people who died earlier than would have been expected in the absence of a natural disaster. For example, looking at the direct toll of tropical cyclones shows that each one killed 24 individuals on average. But expanding this to indirect deaths takes the toll to 7,170-11,430 for each event on average. These indirect early deaths are relatively higher in those under the age of 44 and in Black populations.
The Nature paper is notable because it’s the first systematic attempt to look across hundreds of what we might call “natural disaster” events and attempt to capture the full impact in terms of deaths over decades. While the study focused on hurricanes and similar storms in the US, that type of methodology could be used to study the long-term health impacts of other disasters, from climate floods to heatwaves, and could even be broadened to understand the indirect health impacts of any kind of societal shock – whether a natural disaster or conflict.
For those of us working in global public health, the results aren’t surprising. Individual case studies of natural disasters show the longer-term impact on illness and death. For example, in 2010 a category 7 earthquake hit Haiti. The immediate death toll was estimated to range from 100,000 to 160,000 people. An international response was mounted, but with this came cholera. Genetic sequencing has shown that UN peacekeepers probably brought the disease, and it spread rapidly in conditions of failing water and sanitation systems.
The outbreak resulted in nearly 800,000 Haitians being infected with cholera, and more than 9,000 deaths, and it took until February 2022 – nearly 12 years later – for the country to be declared cholera-free. Haiti is facing numerous other long-term consequences from the earthquake, including poor housing and infrastructure, and weak to nonexistent medical systems, all within multiple failed governments. While we know that such dire conditions will affect people’s health, it’s been hard to draw a direct line back to a single event and estimate its impact.
Another example is the 2004 Boxing Day tsunami, which was triggered by an earthquake in Indonesia. Massive waves up to 100ft high hit nearby countries like Sri Lanka. The direct toll is estimated at roughly 230,000 people, making it the deadliest natural disaster of the 21st century (so far). Reflecting 14 years later on the impact, Mathilda Shannon, studying disaster management at Manchester University, noted the gendered impact of the tsunami, including an increase in girls dropping out of school to take over the roles of their mothers who died in the tsunami, an increase in child marriages as families recover economically, pressure being put on young girls to reproduce to replace children who had died in the tsunami, and increased discrimination against girls.
By doing a systematic analysis across hundreds of events, we can complement these case-by-case examples. We’ve referred to these as the “uncounted” dead in outbreaks: those who die not directly because of a disease such as Ebola, but because of the shutdown of health services and public health outreach. Perhaps that concept can also be used to count the true toll of natural disasters.
We are used to thinking about natural disasters as events confined in time and space: the direct impact in a certain location of an earthquake happens over minutes, a hurricane over hours. While they might be confined in geography, longitudinal studies can help us understand the full range of effects and what extra efforts might be needed to rebuild.
And with the climate crisis increasing the number and severity of disasters like hurricanes and heatwaves, we need to know the true extent of the devastation. Both to alert the public that the danger and harm go on long after the sudden shock of a disaster, and so that governments can plan for the longer-term impacts and develop policies that can save lives in the aftermath.
