effectifs and depenses · updated 6 July 2026 · ODbL licenceThe sex ratio of people under care fell from 2.00 in 2015 to 1.37 in 2024. This change is usually attributed to a generation effect: the generations of women whose smoking caught up with men's are now reaching the ages at which diagnosis is most frequent. If the 2015-2024 trends continue, parity would be reached around 2034.
The dataset publishes no survival data. Two indicators nonetheless say something about severity: two in three people are in active treatment, whereas follow-up would predominate for cancers with a better prognosis; the annual spending on a person in treatment is 8.8 times that on a person in follow-up.
Between 2015 and 2024, reimbursed spending was multiplied by 2.5, and the number of people under care by 1.4. The gap is mainly due to expensive drugs billed on top of hospital stays: since immunotherapies arrived in this indication in 2017, the share of the liste en sus in total spending has risen from 10% to 34%.
France's statutory health insurance, the Assurance Maladie, reimburses care for almost the entire population. Every reimbursement is recorded in a national database, the SNDS (Système national des données de santé). The Cnam, the national body that runs the Assurance Maladie, applies an algorithm to it, the Cartographie des pathologies, which assigns each insured person to one or more diseases from three kinds of evidence: long-term condition status (ALD, affection de longue durée), hospital stays with a diagnosis, and dispensing of specific drugs. data.ameli.fr publishes yearly totals of this classification, never individual records.
France has 18 regions: 13 in mainland France, Corsica included, and 5 overseas (Guadeloupe, Martinique, French Guiana, Réunion and Mayotte). Amounts are in euros.
Two in three people are in treatment, and they account for 95% of spending. A year of treatment costs 8.8 times a year of follow-up.
The whole gap comes from hospital care. It absorbs 69% of spending during treatment, including 53% for liste en sus drugs alone. In follow-up, it falls to 12%.
One item ignores this boundary, cash benefits: €892 per person in treatment, €916 in follow-up. The end of treatment stops the care, not the loss of income.
| IN ACTIVE TREATMENT | IN FOLLOW-UP | |
| Hospital care | €19,458 | €379 |
| of which liste en sus | €10,245 | < €1 |
| Outpatient care | €7,776 | €1,917 |
| of which pharmacy | €5,000 | €1,175 |
| Cash benefits | €892 | €916 |
| Total | €28,125 | €3,212 |
Each amount is rounded to the euro separately: a total may differ by one euro from the sum of its items.
Nearly three quarters of the rise comes from the coral band. In ten years, spending on a person in treatment rose from €16,300 to €28,100 a year; of this extra €11,800, €8,500 comes from the liste en sus alone: hospital drugs costly enough to be billed on top of the hospital stay tariff.
Behind this item: immunotherapies and targeted therapies. The take-off dates from 2017, the year they arrived in this indication: the liste en sus rose from 10% to 34% of lung cancer spending.
The curve has levelled off. Multiplied by six per person, it peaked in 2022 (€10,973) and has eased since (€10,245 in 2024). The other items did not move in the same way: ×1.5 for outpatient care, stable for cash benefits.
These prevalences are crude, not age-standardised. An area with a young population automatically looks spared: this accounts for much of the gap between overseas France and mainland France.
Differences in smoking add to this, and so does access to diagnosis. The “age-matched” button applies the correction to ages 65-79 (the 65-69, 70-74 and 75-79 bands grouped), around the peak of the disease seen in question 03: the gap between mainland regions narrows, Corsica stays ahead, and overseas France stays far behind, a sign that its young population does not explain everything.
Each year, this file takes a snapshot of a total. It follows no one from one snapshot to the next, so by design it cannot answer three obvious questions:
It counts people treated, not diagnoses. Someone treated for five years appears in five snapshots, and a new case cannot be told from an old one.
A person who dies simply disappears from the next snapshot, exactly like a person who recovered or had no care that year. The three exits cannot be told apart.
No diagnosis means no reimbursement, and so no presence in the file. Nothing here allows their number to be estimated.
Before 60, the bars for both sexes are almost the same height. After 75, the men's bars are twice as high. The reason lies not in age but in era: a woman of 80 grew up when women hardly smoked; a woman of 55, when they smoked almost as much as men. Each generation ages with its tobacco. This is why the women's bars will catch up with the men's, one age group after another.
A few dozen cases per age group across all of France. Any cell counting fewer than 11 patients is masked in the file: this is the statistical confidentiality rule, which protects anonymity. These age groups therefore often appear empty.
This chart counts people treated at each age, never deaths. Two cautions follow:
Impossible to know: no death is recorded in this file, at any age. The question remains open.
Not because the disease spares the oldest. Deaths, generations that smoked less and less frequent diagnosis are mixed together here, with no way of separating them.
All the per-person costs on this page are averages: total spending divided by a number of people. The file never publishes the median, nor the distribution of amounts. If a minority of people on very costly treatments pulls the average up, most people then cost far less than the figure shown, and nothing here makes it possible to know.
Not directly: they are in current euros, not adjusted for inflation. Part of the rise since 2015 reflects the general rise in prices, not more care.