What Actually Kills Us

The leading causes of death — and how wildly the news misranks them.

The same number, utterly transformed

People have always died, and roughly a comparable number die each day now as in past centuries. But behind that steady-looking figure is one of the great transformations in human history. For almost all of the past, death came early and often from infection: plague, dysentery, tuberculosis, childbirth, and above all the deaths of small children. Demographers call the shift away from this the epidemiological transition, and it has rewritten what it means to die.

Today, across most of the world, death has been pushed to the far end of life. The leading causes are no longer infections that strike the young but the chronic diseases of old age — heart disease, cancer, dementia. That is not a grim fact; it is arguably the greatest achievement of the modern era. We did not abolish death. We delayed it, by decades, for billions of people.

The clearest proof is in childhood. As recently as 1950, around a quarter of all children died before their fifth birthday across much of the world; globally that figure has now fallen below 4%, and keeps dropping. Hundreds of millions of children who would once have died now live full lives. It is the most important good-news story almost no one talks about.

What we fear is not what kills us

The widget above captures the single most useful thing to understand about mortality: our sense of danger is wildly miscalibrated. Researchers have compared the actual causes of death with how often each is mentioned in the news. The mismatch is staggering. In one analysis of US deaths and US media, terrorism caused a tiny fraction of one percent of deaths but accounted for roughly a third of coverage, while heart disease — the number-one killer — received a small sliver of attention.

This is not a media conspiracy; it is human nature. We pay attention to the rare, sudden and dramatic, and tune out the common, slow and ordinary. The result is that people worry about plane crashes, sharks and terror attacks while the things most likely to end their lives — poor diet, smoking, high blood pressure, inactivity — attract barely a thought. Understanding the real distribution of death is genuinely empowering: it tells you where your attention, and your worry, are actually worth spending.

What 170,000 a day means

The daily toll, made tangible.

The crossover that is coming

Deaths are one half of a daily ledger. Set them against the roughly 385,000 babies born each day and the world ends each day with about 215,000 more people than it started with. That surplus is what keeps the population growing — for now.

But the two lines are slowly converging. Births have plateaued, while deaths are set to rise as the world ages. Sometime around the 2080s, demographers expect the lines to cross: more people dying each day than being born. That is the moment the global population, traced on our page about how many people are alive today, is projected to peak and begin its first sustained decline in modern history.

Test your risk sense

Six questions on what really ends lives — most people guess wrong.

Frequently asked

Sources & method

Daily and per-second figures are derived from published annual estimates, so they are rounded approximations. The news-versus-deaths comparison is illustrative, based on an analysis of US causes of death against US media coverage; the exact figures vary by study and country.

• World Health Organization — Global Health Estimates: leading causes of death worldwide (cardiovascular disease, cancers and others).
• United Nations, World Population Prospects 2024 — annual deaths (~62 million), the projected rise toward ~90 million, and the births/deaths crossover around the 2080s.
• Our World in Data — "Does the news reflect what we die from?" (analysis of US deaths versus US news coverage, after Shen et al.).
• UN Inter-agency Group for Child Mortality Estimation — the fall in under-5 mortality since 1950.
• Italy and United Kingdom populations (~59 and ~67 million) used for the annual scale comparison.

Related questions