Chernobyl at 40: The True Human Toll

Giving Green, one of the most rigorous independent climate evaluators, recommends supporting organisations that actively advocate for nuclear energy development — Clean Air Task Force is among them. For some donors, this position feels counterintuitive. The question comes up often: "But what about Chernobyl?"

Romain Barbe

Founder · Mieux Donner · Reading time: 12 min

It is a legitimate question. On 26 April 1986, exactly 40 years ago, a reactor exploded in Soviet Ukraine in what remains the most serious nuclear accident in history. Forty years on, what do the data actually say about its human toll?

The answer is more complex than most people think. And for some researchers, considerably more troubling.

What the data actually say about Chernobyl's death toll

The consensus on direct deaths: 31

The figure that the entire scientific community accepts without debate is precise and, for many, surprising: 31 deaths directly attributable to the accident.

Two people died in the immediate explosion. Twenty-eight workers and firefighters on site died within three months from acute radiation syndrome. A third death, from a heart attack during the emergency response, brings the total to 31. Of the 134 workers hospitalised as emergencies with severe radiation syndrome, 28 died in the weeks that followed. The others survived.¹

This figure is established by UNSCEAR, the United Nations Scientific Committee on the Effects of Atomic Radiation, and confirmed by the Chernobyl Forum report in 2006, which brought together the IAEA, WHO, UNDP and UNEP.² These are the only deaths that science can attribute directly to radiation, without modelling or extrapolation.

Cancers: estimates, not counts

Beyond direct deaths, did the accident cause additional cancers in the population? Here the ranges diverge considerably. Understanding why is essential to reading the debate honestly.

The Chernobyl Forum report (WHO/IAEA, 2006) estimates approximately 4,000 additional cancer deaths among the most heavily exposed populations: liquidators and people evacuated from the most contaminated zones.² This is the most conservative estimate, applied only to people who received significant doses.

The Cardis et al. study, published in the International Journal of Cancer in 2006, extends the scope to all of Europe and arrives at approximately 16,000 additional deaths, with substantial uncertainty.⁴ The TORCH report, commissioned by the Greens–EFA group in the European Parliament and written by researchers Ian Fairlie and David Sumner, puts forward a range of 30,000 to 60,000 deaths.⁵

Other estimates diverge even further from the consensus. Greenpeace's 2006 official report, produced by around sixty scientists, concludes with approximately 93,000 fatal cancers and cites 200,000 additional deaths in Russia, Belarus and Ukraine between 1990 and 2004 — a figure the organisation claims as its own, without peer review.^6a Further still, the book by Yablokov, Nesterenko and Nesterenko, published in 2009 in the Annals of the New York Academy of Sciences, puts forward a figure of 985,000 deaths. Alexei Yablokov was a co-founder of Greenpeace Russia, but the NYAS issued an explicit disclaimer stating that it had neither commissioned, validated, nor peer-reviewed the work. Four of the five published academic reviews judged the methodology to be seriously flawed.^6b

The honest and defensible range sits between 4,000 and 60,000 additional cancer deaths, projected over the lifetimes of the exposed population. These are modelled projections spanning decades, not actual counts.

Thyroid cancer: the only causally established cancer

Amid this debate, one point commands consensus: thyroid cancer is the only cancer whose increase is causally and scientifically attributed to the Chernobyl accident.

Around 6,000 cases of thyroid cancer were diagnosed in children and adolescents exposed at the time of the accident, primarily through ingestion of radioactive iodine via the food chain.¹ The paradox of this finding is that thyroid cancer, when diagnosed and treated, has a survival rate above 98%. The number of deaths attributed to these thyroid cases did not exceed 15 at the time the UNSCEAR 2008 report was published. The disease is serious, the diagnosis shattering, but mortality remains exceptionally low compared to the number of cases.

The invisible cost: when fear kills more than radiation

The coal substitution effect: documented deaths

From 1986, and even more sharply after Fukushima in 2011, fear of nuclear energy led several democratic countries to close plants or freeze new projects. That capacity was replaced, largely, by coal and natural gas. Burning these fossil fuels produces fine particles and atmospheric pollutants that kill.

This mechanism is documented with precision for Germany. After Fukushima, the German government closed eight reactors within months. A study by Jarvis, Deschenes and Jha estimates that this shift toward coal caused approximately 1,100 additional deaths per year in Germany through increased air pollution.⁸ A study published in 2025 in Environmental and Resource Economics, using the synthetic control method, produces a more conservative estimate of around 170 deaths per year, but confirms the direction of the effect.⁹

Kharecha and Sato, in a study published in 2019 in Energy Policy, modelled what would have happened if Japan and Germany had reduced coal output rather than nuclear output after Fukushima: approximately 28,000 premature deaths from air pollution could have been avoided between 2011 and 2017.¹⁰

Plants never built: a 40-year reckoning

Chernobyl's effect on global nuclear policy did not stop at the closure of existing plants. The accident also halted or significantly slowed the construction of new reactors for four decades.

A recent working paper by economists at MIT, Northwestern University and the University of Chicago (Makarin, Qian and Wang, 2024, presented at the National Bureau of Economic Research) quantifies this effect. According to their estimates, Chernobyl reduced the number of nuclear plants built worldwide by 389 units over 38 years. Each plant not built meant more fossil fuels burned and more fine particles emitted. The researchers estimate that the decline in nuclear construction caused by Chernobyl resulted in the loss of 318 million life-years worldwide, including 141 million in the United States and 33 million in the United Kingdom.⁷

The study also points to an irony: post-Chernobyl regulatory delays extended the operation of older plants, which are statistically less safe, paradoxically increasing the risk of a future nuclear accident.

The other direct effects of radiophobia

Fear of radiation also produced immediate health effects, independent of any energy policy.

At Fukushima, the example is documented with precision. The decision to urgently evacuate 154,000 people living around the plant was taken under the pressure of radiological panic. That rushed evacuation caused 1,656 deaths, 90% of them among people aged 65 or over, who were frail and poorly adapted to temporary shelter conditions. In the same geographic area, the earthquake and tsunami had killed 1,607 people.¹¹ Fear of radiation killed more people than the radiation itself.

After Chernobyl, similar effects were documented. In several European countries where abortion was legal, researchers recorded an abnormal rise in requests for terminations of pregnancy, including for healthy pregnancies, in countries where the doses received were so low that no health effects were expected. In Denmark, academic work established that fear of nuclear radiation had caused more foetal deaths than the accident itself.¹²

What Chernobyl teaches us about our decisions

Judge by consequences, not intention

Chernobyl has become a symbol. And symbols are rarely good analytical tools. Reactor number 4 exploded because of a combination of design flaws and human error: that fact is beyond dispute. But the political and emotional response that followed, in Western democracies, was often driven by emotion more than data.

This is not a moral criticism. Fear of radiation is an understandable human reaction. But in energy policy decisions, as in giving decisions, intention is not enough. What matters is the effect. And the effect of a policy that substitutes coal for nuclear is measured in deaths from fine particles, in additional tonnes of CO2, and in life-years lost.

Some organisations, by amplifying fear of radiation, spreading mortality estimates far removed from the scientific consensus, and above all by never asking what would replace nuclear in the event of closure, contributed to energy choices whose human consequences proved far heavier than what they were trying to prevent. The intention was to protect. The effect, by directing countries toward more coal, was often the reverse. It is a reminder that the impact of an action is not measured by its intention, but by its consequences. That is true for donations. It is true for advocacy.

What the comparative data show: deaths per TWh

The question "is nuclear dangerous?" cannot be asked in the abstract. It must be compared with the available alternatives. That is precisely the angle taken by Our World in Data, one of the most rigorous and widely cited data sources in academic literature, in its analysis of mortality rates by energy source.¹³

Source: Our World in Data, based on Markandya & Wilkinson (The Lancet, 2007) and Sovacool et al. (2016). These estimates for fossil fuels are likely conservative: they are based on European plants and older models of pollution's health impact.^13,14

These data warrant two qualifications that intellectual honesty requires mentioning. The first concerns radioactive waste: mortality per TWh does not measure the very long-term effects of nuclear waste storage, which remains an open question in terms of global governance. The second concerns context-specificity: with the massive and continuing fall in solar and wind costs over the past decade, nuclear is not necessarily the right answer in every geographic or temporal context. What remains true is that the relevant question, in each concrete situation, is: what replaces coal and gas?

Energy poverty: the issue Chernobyl debates ignore

There is a dimension that the Chernobyl debate almost always leaves aside. In at least 45 countries, daily residential electricity consumption per person is lower than what it takes to run an air conditioner for one hour. This is not a metaphor about comfort: it is a reality documented by Hannah Ritchie on Our World in Data in 2026.¹⁶ No refrigeration for medicines, no lighting after sunset, no protection against heat waves that, in a warming world, are killing more and more people.

According to IEA data from October 2025, 730 million people have no access to electricity in 2024, 85% of them in sub-Saharan Africa.¹⁵ If a broader definition of energy poverty is used, incorporating people who are connected but unable to use their connection effectively, a study published in Joule in 2024 puts this figure at more than one billion.¹⁷

We can debate technologies. What we cannot do is ignore that the consequences of our collective energy choices do not play out only in wealthy countries, and that access to abundant, clean and affordable energy is a condition of human development.

Conclusion

Forty years after Chernobyl, the human toll of the accident is real but widely misunderstood. Thirty-one direct deaths, attested. Thousands of additional cancers, probable but modelled, with ranges from 4,000 to 60,000 depending on the method and scope used. And a third toll, harder to see: the one fear itself produced, by directing energy policies toward sources that are far deadlier per unit of energy produced.

Judge by consequences rather than emotions. Use the available data rather than symbols. Move beyond ideological divides to assess the real impact of our choices. That is what the scientific method does. It is also what we try to do at Mieux Donner, for every recommended charity.

Frequently asked questions