Floods, storms, earthquakes, droughts and heatwaves still kill tens of thousands of people and affect over 100 million every year; the death rate has fallen as early warning and preparedness have spread, but the number of people affected and the cost of disasters keep rising, and the poorest countries carry a disproportionate share.
Updated October 14, 2026. Written from the sources listed below, not independently peer-reviewed.
Countries report disaster deaths to the UN. Compared with 2005–2014, how did the global disaster death rate per 100,000 people change in 2015–2024?
41,000
Average disaster deaths per year
Deaths and missing people reported by countries, yearly average for 2015–2024, excluding COVID-19. The average for 2005–2014 was about 88,000.
Data year 2024. Source: United Nations Statistics Division, 2026
123 million
People affected by disasters each year
Yearly average for 2015–2024, as reported by countries. The rate per 100,000 people more than doubled compared with 2005–2014.
Data year 2024. Source: United Nations Statistics Division, 2026
119
Countries reporting multi-hazard early warning systems
About 60% of all countries, a record high in the 2025 global status report. Only 43% of small island developing states report having such systems.
Data year 2025. Source: UNDRR and WMO, 2025
Unit: countries
Countries that report having adopted and implementing a national disaster risk reduction strategy in line with the Sendai Framework (SDG 1.5.3). The 141 countries in 2025 represent 72% of the world. Having a strategy says nothing about how well it is carried out.
The UN's Early Warnings for All initiative aims to protect every person on Earth with an early warning system by 2027, which is also how the Sendai Framework's Target G is tracked. By 2025, 119 countries, about 60% of the total, reported multi-hazard early warning systems, more than double the number ten years earlier. But only 43% of small island developing states have them, Africa and the least developed countries still score lowest, and fewer than a third of countries report solid capabilities to understand the risks they face. With about a year to go, closing these gaps will require faster investment than so far.
In 2015–2024 the least developed countries accounted for 28.7% of reported disaster deaths while holding only 12% of the population of reporting countries.
Source: United Nations Statistics Division, 2026Landlocked developing countries accounted for 20.8% of reported disaster deaths with only 5.1% of the population, and the least developed countries for 12.91% of reported economic losses.
Source: United Nations Statistics Division, 2026Only 43% of small island developing states report having multi-hazard early warning systems. Africa has improved the most since 2015 but still has the lowest scores of any region.
Source: UNDRR and WMO, 2025Insurance covers less than 1% of disaster risk in countries such as Bangladesh, India, Viet Nam, the Philippines, Indonesia, Egypt and Nigeria, so families and governments there bear most losses themselves.
Source: United Nations Office for Disaster Risk Reduction, 2025Each country is shaded by its latest available value, so you can see where the gap is widest. Hover or tap a country, search for one, or open the full table.
Loading country values…
Values in per 100,000 people.
Average yearly deaths and missing people from droughts, earthquakes, floods, storms, landslides, wildfires, volcanoes and extreme temperatures recorded in EM-DAT, per 100,000 people, for the decade starting 2020 (shown as 2020; the decade is not yet complete). A single large event can dominate a country's value. Heatwave deaths are recorded far more completely in Europe than elsewhere, so many European countries appear high while heat deaths in Africa and South Asia are largely missing.
Source: EM-DAT (CRED / UCLouvain), via Our World in Data. Latest year available per country; years differ.
Values load from Our World in Data. Simplified Natural Earth borders; they imply no position on disputed territories.
Instead of one “percent solved” score, DO GOOD rates three separate dimensions. A problem can have mature solutions that still reach too few people. These ratings are an editorial reading of the sources on this page, not a measured index.
Has the situation improved in the real world?
ModerateThe reported disaster death rate fell from 2.11 to 0.75 per 100,000 people between 2005–2014 and 2015–2024, but the rate of people affected more than doubled, from 1,199 to 2,445 per 100,000, and losses remain high.
How mature and validated are the available solutions?
HighEarly warning, preparedness, risk-informed building and land use are well established. Countries with more comprehensive early warning systems report disaster mortality nearly six times lower than countries with limited systems, and UNDRR estimates an average return of US$15 in avoided recovery costs for every dollar invested in risk reduction.
How widely are effective solutions actually used?
Moderate141 countries have national strategies and 119 report multi-hazard early warning systems, but coverage is lowest in small island states, Africa and the least developed countries, and risk reduction receives only about 2% of global aid.
Loss figures are reported by countries with uneven completeness, and many deaths are counted late or not at all: CRED notes that heatwave deaths for 2025 had not yet been included and that provisional estimates may add thousands of deaths in Europe alone.
Death rates are falling and early warning coverage and national strategies are expanding quickly. At the same time, more people are affected each year and UNDRR finds that direct disaster costs have risen from US$70–80 billion a year in 1970–2000 to over US$200 billion.
Disasters happen when a natural hazard such as a flood, storm, earthquake, drought or heatwave meets people and places that are exposed and unprepared. Countries reporting to the UN recorded an average of about 41,000 disaster deaths a year in 2015–2024 and 123 million people affected each year. Single events can be devastating: in 2025 the Myanmar earthquake killed at least 3,820 people, an earthquake in Afghanistan an estimated 2,200, and Tropical Cyclone Senyar left 1,109 people dead or missing on Sumatra in Indonesia, where it also left 3.2 million people in need of help. Beyond the deaths, disasters destroy homes, schools, clinics, roads and harvests. Reporting countries lose more than US$110 billion a year in direct economic losses, and UNDRR estimates that the full cost, once knock-on effects and damage to ecosystems are counted, exceeds US$2.3 trillion a year.
The burden is unequal. The least developed countries account for 28.7% of reported disaster deaths but only 12% of the population of reporting countries, and landlocked developing countries for 20.8% of deaths with 5.1% of the population. Poorer households tend to live in less safe housing and in more exposed places, have little or no insurance and fewer savings to recover, so one disaster can undo years of progress. The good news is that disaster deaths can be prevented: the reported death rate has fallen by nearly two thirds compared with the previous decade, a change the UN links to better early warning and preparedness. But the number of people affected has doubled over the same period, driven by growing cities, environmental degradation and climate change.
The Sendai Framework for Disaster Risk Reduction 2015–2030, adopted by UN member states in March 2015, sets seven global targets tracked with 38 indicators, several of which are also SDG indicators. Countries report their own losses to UNDRR through the Sendai Framework Monitor, while the independent EM-DAT database records major events worldwide. The main measures are:
Deaths and missing people
The number of people who died or went missing because of disasters, per 100,000 population (SDG 1.5.1, 11.5.1 and 13.1.1). Because one large event can dominate a single year, the UN compares ten-year averages; its latest figures exclude COVID-19 deaths.
People directly affected
People injured, made ill, evacuated, displaced or whose homes or livelihoods were damaged. The global rate rose from 1,199 per 100,000 people in 2005–2014 to 2,445 in 2015–2024.
Economic losses and damage to services
Direct economic losses (SDG 11.5.2) and damage to critical infrastructure and disruption of basic services such as health, education and water (SDG 11.5.3). Reporting countries recorded an average of 91,847 damaged or destroyed facilities and more than 1.5 million service disruptions a year.
Strategies and early warning
Whether countries and local governments have disaster risk reduction strategies in line with the Sendai Framework (SDG 1.5.3 and 1.5.4), and whether people are covered by multi-hazard early warning systems, the Sendai Framework's Target G. These measure preparation, not whether it works in practice.
What works
Systems that monitor hazards, forecast their impact, send clear warnings through many channels and make sure people know how to act give communities time to evacuate and protect property. Countries with more comprehensive systems report disaster mortality nearly six times lower than those with limited systems, and 119 countries now report having one.
Limits
The comparison between countries shows an association, not proof that warnings alone cause the difference. Warnings only help if they reach everyone, including people without mobile phones, and if there are safe places to go; coverage is weakest in small island states, Africa and the least developed countries.
What works
Strategies assign responsibilities across government, set priorities and link risk reduction to budgets. By 2025, 141 countries had a national strategy and 116 reported local strategies in line with it, with an average of 72% of local governments in those countries covered.
Limits
The indicator records whether a strategy exists, not whether it is funded or enforced. Strategies often sit with disaster agencies that have little influence over planning, finance or infrastructure ministries.
What works
Building codes for earthquakes and storms, keeping new homes and services out of floodplains and landslide zones, and strengthening schools and hospitals reduce both deaths and damage. UNDRR expects about 1.2 billion more people to live in cities by 2050 than in 2020, 98% of the growth in the Global South, so building safely now prevents future losses.
Limits
Enforcement is weak where much housing is built informally, and retrofitting existing buildings is expensive. UNDRR warns that unsafe construction in earthquake zones is increasing risk.
What works
Contingency funds, insurance, catastrophe bonds and social protection that can be scaled up after a shock let governments and families recover faster without selling assets or cutting essential spending.
Limits
Insurance covers less than 1% of risk in countries such as Bangladesh, India, Indonesia and Nigeria, premiums are rising where risk grows, and pre-arranged financing is still only a very small fraction of crisis funding.
What works
Evacuation plans and drills, trained local volunteers, stocked shelters and first aid skills save lives in the first hours, before outside help arrives, and help warnings turn into action.
Limits
Preparedness needs regular practice and local leadership, and it fades between rare events. It cannot make up for unsafe buildings or the absence of warnings.
Promising, but not proven at scale. The stage shows how far each approach has come; it is not a recommendation.
Agencies agree in advance which actions to fund, such as cash transfers, moving livestock or reinforcing homes, and release the money when a forecast crosses a set threshold, before the disaster strikes. The 2025 early warning status report expects the number of active frameworks to keep growing; evidence on which actions work best for which hazards is still building.
Alerts sent to every phone in a danger zone, together with impact-based forecasts that say what a hazard will do rather than only how strong it is, are reaching more people. The 2025 status report notes that the digital divide still leaves some groups out.
Many low-income countries lack the weather and climate observations that forecasts depend on. Through the Systematic Observations Financing Facility, 20 national weather services with strong data expertise are advising 62 beneficiary countries as peers.
UNDRR's 2025 report models what would happen if low-income and emerging economies could draw on 10% of their IMF Special Drawing Rights after a major disaster, and finds that fiscal crises would be pushed back by 19 and 12 years respectively. This is a modelled policy proposal, not yet in use.
Implementation gap
Only 43% of small island developing states report multi-hazard early warning systems, and Africa and the least developed countries still have the lowest scores.
Knowledge gap
Warnings and plans depend on knowing which hazards threaten which people and places, yet fewer than a third of countries report having disaster risk knowledge capabilities, despite a 16% improvement since 2022.
Implementation gap
Activities aimed at reducing disaster risk receive about 2% of global aid, and pre-arranged financing is a very small share of crisis funding, even though prevention costs far less than recovery.
Coordination gap
Under Early Warnings for All, national implementation plans had been completed in only 17 countries by the time of the 2025 status report, and work across weather services, disaster agencies, telecoms and local governments is often fragmented.
Technology gap
Warnings sent through mobile networks miss people without phones, signal or electricity, and messages are not always in local languages or formats that people with disabilities can use.
Knowledge gap
Countries report their own losses with uneven completeness, and some deaths, especially from heat, are recorded late or not at all. CRED's 2025 figures did not yet include heatwave deaths, and provisional estimates point to 16,469 preventable heat-related deaths in European cities in summer 2025 alone.
Unresolved questions, not vetted breakthroughs. Answering any of them would change what is possible.
How much of the fall in disaster deaths is due to early warning systems, and how much to better buildings, health care, rising incomes or chance differences in where large events struck?
How can countries in Africa, South Asia and elsewhere estimate deaths from extreme heat reliably and in time, so that the global picture is not dominated by the few regions that measure them?
Why do people who receive a warning sometimes not act on it, and which message formats, channels and community arrangements lead to timely evacuation among the most vulnerable groups?
How can the indirect and cascading costs of disasters, such as lost schooling, health effects, debt and damage to ecosystems, be measured consistently so that investment decisions reflect them?
Find out which hazards affect where you live and work, sign up for official alerts, and agree a household plan with a meeting point, important documents and a small emergency kit.
A first aid course helps you care for injured people in the hours before professional help arrives, when many lives are saved or lost.
After a disaster, cash donations to established organisations working with local partners are usually more useful than collected goods, which are hard to transport and sort and may not match what people need.
Back organisations and local projects that work on early warning, safer schools and community preparedness between disasters, and ask local representatives how your area plans for floods, storms and heat.
Ideas visitors sent in for this challenge, read by the DO GOOD team and pinned with permission. The colour shows the kind of contribution.
Loading the board…
What happens next
The DO GOOD team reads every idea and sorts it by colour: solution, research question, existing project, evidence or local action.
If you agree to publication, it can be pinned to this board. Contact details are never shown.
Pinned ideas are not reviewed by scientists and are not endorsed by DO GOOD.
This page is a research overview. DO GOOD does not run a dedicated programme on this challenge, and donations are not earmarked for it. You can join our volunteer work and outreach.
United Nations Statistics Division (2026), published 2026
Latest UN progress report on SDG targets 11.5 and 11.b, based on country reports to the Sendai Framework Monitor. Source of the average deaths, people affected and rates for 2005–2014 and 2015–2024, the shares of deaths and losses in least developed and landlocked developing countries, direct economic losses, damage to infrastructure and services, and local strategies.
United Nations Statistics Division (2026), published 2026
UN progress report on SDG target 13.1. Source of the fall in the disaster death rate from 2.11 to 0.75 per 100,000 (nearly 65%, excluding COVID-19), the rise in people affected, and the number of countries with national disaster risk reduction strategies in 2015 and 2025.
UNDRR and WMO (12 November 2025), published 2025
Annual report on Sendai Framework Target G and the Early Warnings for All initiative; the newest edition. Source of the 119 countries (60%) with multi-hazard early warning systems, the 43% of small island developing states, regional gaps, the nearly six times lower mortality in countries with comprehensive systems, risk knowledge capabilities, national implementation plans, SOFF and anticipatory action. Figures confirmed in the joint WMO/UNDRR press release of the same date.
United Nations Office for Disaster Risk Reduction (27 May 2025), published 2025
UNDRR's flagship report, published every few years; the latest edition. Source of direct disaster costs (US$70–80 billion a year in 1970–2000, over US$200 billion now), total costs above US$2.3 trillion including cascading and ecosystem effects, the US$15 average return per dollar invested, the 2% share of aid, insurance coverage, urban growth and the Special Drawing Rights modelling.
Centre for Research on the Epidemiology of Disasters (CRED), UCLouvain (April 2026), published 2026
Annual summary of the EM-DAT international disaster database for 2025 (data as of 17 March 2026). Source of the deadliest events of 2025, including the Myanmar and Afghanistan earthquakes and Tropical Cyclone Senyar on Sumatra, Indonesia, and of the note that heatwave deaths were not yet included.
United Nations Office for Disaster Risk Reduction, published 2015
UNDRR's overview of the Sendai Framework for Disaster Risk Reduction 2015–2030, adopted on 18 March 2015. Source of its period, seven global targets, 38 indicators and four priorities for action.