What actually kills people
The top of the table barely moves. Everything underneath it has been rearranging itself for fifty years.
Ask people what kills the most people worldwide and the answers skew violent and infectious: war, terrorism, murder, epidemics. The actual answer is duller and much larger. It is cardiovascular disease, and it has been for a very long time.
The table
Cardiovascular disease — ischaemic heart disease and stroke, principally — accounts for somewhere near a third of all deaths globally, on the order of 18 to 20 million a year. Ischaemic heart disease alone is the single largest cause, killing roughly 9 million people annually. Stroke is second at around 6 to 7 million.
Cancers, taken together, come to roughly 10 million a year. Chronic respiratory disease — mostly COPD — adds another 3 to 4 million. Then comes a long tail: dementias, diabetes and kidney disease, lower respiratory infections, neonatal conditions, road injuries.
This is why the Ledger's Pathological column dwarfs its Accidental column. Cardiovascular demise ticks at 0.589 per second. Fatal electrocution ticks at 0.001. That six-hundred-fold gap is not a presentational choice; it is the shape of the data.
What has changed
The rearrangement underneath the top line is the real story, and it has a name: the epidemiological transition. As countries grow wealthier, infectious disease recedes — through clean water, sanitation, vaccination, antibiotics — and the deaths that used to happen in childhood happen decades later instead, from chronic disease. People do not stop dying. They start dying of different things, older.
Two movements stand out.
Dementia has climbed steeply. Alzheimer's disease and other dementias have risen sharply in the rankings over recent decades and are now the leading cause of death in several wealthy countries, including the United Kingdom. Very little of this is a real increase in risk. It is mostly that dementia is a disease of advanced age, more people now reach advanced age, and — not trivially — doctors have become far more willing to record dementia on a death certificate rather than the pneumonia that finally arrived.
Infectious disease has receded, unevenly. Lower respiratory infections, diarrhoeal disease, tuberculosis, HIV and malaria have all fallen substantially. But they remain leading causes in low-income countries, which is why the Ledger's country atlas shows such radically different tables for Nigeria and Germany. The transition is not finished; it is just much further along in some places than others.
The certification problem
A death certificate names one underlying cause. Real deaths frequently do not have one.
Consider someone with type 2 diabetes and chronic kidney disease who dies of a heart attack. Diabetes damaged the blood vessels; kidney disease strained the heart; the heart attack was the final event. Depending on the certifying physician's judgement and the local coding convention, that death might be counted under diabetes, kidney disease, or ischaemic heart disease. All three are defensible. Only one gets counted.
This is why cause-of-death comparisons across countries are shakier than they look, and why the Global Burden of Disease study reports risk factors — high blood pressure, tobacco, diet, air pollution — alongside causes. High systolic blood pressure is associated with more deaths than any single named disease, because it feeds into so many of them at once. That framing is often more useful than the cause table, and much less frequently quoted.
What this does not tell you
Counting deaths tells you nothing about when they happened. A cause that kills 500,000 infants and one that kills 500,000 ninety-year-olds appear identically in a mortality table, and are not remotely the same event.
This is what measures like years of life lost exist to capture, and by that measure the table reshuffles considerably: road injuries, neonatal conditions and self-harm all climb sharply, because they take people young. The Ledger's demographic bars gesture at this — half of all deaths occur at 70 and over — but the mortality table on its own will always flatter causes that kill the old.
None of this is health advice. Population statistics describe populations, not people. Knowing that cardiovascular disease leads the global table tells you nothing useful about your own risk, which depends on factors no statistic on this site measures. If you have concerns about your health, speak to a doctor.