Same Cancer, Different Odds: How Cancer Survival Varies Between Countries

The chance of being alive five years after a cancer diagnosis depends heavily on where a person lives. The largest published global study, CONCORD-3, found five-year survival for children with the most common childhood leukaemia ranging from under 50% to over 95% between countries, and even among wealthy countries with universal health care, survival from the same cancers differs by several percentage points.^[1][2]

For a person diagnosed with cancer, these gaps are not abstractions. Two people with the same cancer, at the same age, can face very different prospects because one lives where cancer tends to be found earlier, where treatment starts sooner, or where specialist care is in reach. International comparisons are how researchers find out which of those differences are real and which are artefacts of how data are collected. This piece, part of our cancer research library, sets out what the two main international programmes, CONCORD and the International Cancer Benchmarking Partnership (ICBP), have found, why the gaps exist, and how carefully the numbers need to be read.

The same cancer has very different survival depending on the country

CONCORD-3, published in The Lancet in 2018, is the largest comparison of cancer survival yet published. It analysed individual records for more than 37.5 million people diagnosed with one of 18 cancers between 2000 and 2014, from 322 population-based cancer registries in 71 countries and territories, and estimated age-standardised five-year net survival.^[1] Net survival is the share of people who would be alive five years on if cancer were the only possible cause of death, which lets countries with different life expectancies be compared fairly.

For most cancers, survival was among the highest in the world in the USA and Canada, Australia and New Zealand, and Finland, Iceland, Norway and Sweden. For women diagnosed with breast cancer in 2010 to 2014, five-year survival was 90.2% in the USA and 89.5% in Australia, but as low as 66.1% in India. Children's brain tumours showed a very wide range: from 28.9% in Brazil to nearly 80% in Sweden and Denmark.^[1] Our overview of cancer worldwide sets these survival gaps beside incidence and mortality.

The ranking is not simply rich versus poor. South Korea had the highest five-year survival in the world for stomach (68.9%), colon (71.8%) and rectal cancer (71.1%), Japan for oesophageal cancer (36.0%) and Taiwan for liver cancer (27.9%). In the same region, survival from melanoma and from lymphoid and myeloid blood cancers was generally lower than elsewhere.^[1] A country's position can differ sharply from one cancer to the next, so a single national ranking hides as much as it shows.

Even among wealthy countries with universal health care, survival gaps persist

The International Cancer Benchmarking Partnership was set up to ask a narrower question: why do countries with similar wealth and universal health systems get different results? Its SURVMARK-2 study, published in Lancet Oncology in 2019, used data on 3.9 million people from cancer registries in Australia, Canada, Denmark, Ireland, New Zealand, Norway and the UK, diagnosed with cancer of the oesophagus, stomach, colon, rectum, pancreas, lung or ovary between 1995 and 2014.^[2]

The good news is real: one-year and five-year net survival rose in every country, for almost all of the cancers studied, over those two decades, and five-year survival from rectal cancer increased by more than 13 percentage points in Denmark, Ireland and the UK. Gains were larger for people diagnosed before 75 than for those aged 75 and over. But for people diagnosed in 2010 to 2014, survival was still generally higher in Australia, Canada and Norway than in New Zealand, Denmark, Ireland and the UK.^[2]

Later ICBP papers looked cancer by cancer. For people diagnosed in 2012 to 2014, one-year net survival from stomach cancer was 55.2% in Australia and 44.8% in the UK; for oesophageal cancer it was 50.3% in Ireland and 41.3% in Canada.^[4] In a separate ICBP study of pancreatic cancer, also for people diagnosed in 2012 to 2014, one-year net survival ranged from 21.1% in New Zealand to 30.9% in Australia.^[5]

Sources: Arnold et al., Gut 2022^[4]; Cabasag et al., British Journal of Cancer 2022^[5]. Seven countries: Australia, Canada, Denmark, Ireland, New Zealand, Norway and the UK.

Europe shows a similar pattern on a wider canvas. The EUROCARE-6 study found five-year relative survival from skin melanoma of 87.9% across Europe for adults diagnosed in 2008 to 2013, but 75.0% in Eastern Europe against 91.2% in Central Europe. Within Eastern Europe, the gap between countries reached 24.7 percentage points.^[6]

Stage at diagnosis explains much of the gap, but not all of it

The single biggest influence on survival from most solid cancers is how far the cancer has spread when it is found. England's official statistics show the scale: for men diagnosed with lung cancer in 2018 to 2022, five-year net survival was 59.6% when the cancer was found at stage 1 and 4.2% at stage 4. Of people diagnosed with non-small cell lung cancer, 42.7% were diagnosed at stage 4.^[7] As the publication notes, survival for all stages combined depends on the share of people diagnosed at each stage, so a country that finds more cancers early will have higher overall survival even if treatment at each stage were identical.^[7] Our explainer on stage at diagnosis covers this in more depth.

Stage distributions do differ between countries. Among the seven ICBP countries, the share of pancreatic cancers diagnosed at distant (metastatic) stage ranged from 53.9% in Ireland to 83.3% in New Zealand.^[5] For oesophageal and stomach cancer, between 56% and 90% of people had regional or distant disease, depending on the country, and the authors concluded that the disparities "can partly be explained by earlier diagnosis resulting in more favourable stage distributions".^[4]

The word "partly" matters. Survival also differed between countries within the same stage group. For oesophageal and stomach cancer, variation between countries was greatest for localised disease, where one-year survival from stomach cancer ranged from 75.5% in Australia to 94.3% in New Zealand.^[4] For pancreatic cancer, survival differences were seen between countries in every stage category at one year, narrowing over time, particularly for distant-stage disease.^[5] If stage were the whole story, people with the same stage would fare the same everywhere. They do not, which points to differences in treatment and care after diagnosis, as well as to differences in how stage is recorded.

Access to treatment, and the money behind it, shapes survival too

ICBP has begun to measure treatment directly. A study published in BMJ Oncology in 2025 compared people diagnosed with lung cancer in 2012 to 2017 across 16 jurisdictions in Australia, Canada, the UK and Norway. The share receiving chemotherapy ranged from 23% in Northern Ireland to 45% in Norway, and radiotherapy from 32% in England to 48% in New South Wales and 50% in Newfoundland and Labrador. The variation held within groups of the same age and stage. Time to starting radiotherapy was longer in the UK jurisdictions than elsewhere. The authors write that the reasons "are unclear", but that differences in non-surgical treatment "are plausibly associated with international variation in lung cancer survival".^[8]

Spending appears in the European data as well. A EUROCARE-6 analysis of people diagnosed with myeloid blood cancers in 2001 to 2013 found five- and ten-year relative survival from acute myeloid leukaemia, myeloproliferative neoplasms and myelodysplastic syndrome was higher in the European countries with the highest total national health expenditure than in those with the lowest.^[9] This is an association between countries, not proof that spending alone explains the difference.

The widest gaps are between high-income countries and countries where diagnosis and treatment are hardest to reach. The ABC-DO study, published in The Lancet Global Health in 2025, followed 2,153 women with breast cancer at eight hospitals in Namibia, Nigeria, South Africa, Uganda and Zambia for up to seven years. Crude (observed) survival was 51% at three years and 40% at five years. Five-year age-standardised net survival was 35% to 42% in Zambia and Nigeria, 52% to 58% for Black women in Uganda, South Africa and Namibia, and over 83% for non-Black women in Namibia.^[3] The gap within Namibia is a reminder that survival tracks inequality inside countries as well as between them. In the fifth year after diagnosis, the annual risk of death was still 8% to 21% for Black women in Namibia, Uganda and Nigeria.^[3] These figures come from a hospital cohort rather than a national registry, so they are not directly comparable with CONCORD estimates. Our study of cancer in low- and middle-income countries looks at these settings in more detail.

Some of the gap is in the data, so comparisons must be read with care

Survival figures depend on how registries work. Researchers behind SURVMARK-2 identified five areas where differences in method can shift a country's estimate: how complete the registry is and what sources it draws on; whether death certificates are used to find cases; how the date of diagnosis is defined; how people with more than one primary cancer are handled; and how well cases are linked to death records. ICBP's own assessment is that these registry differences are unlikely to explain the survival gaps it found, but they can shift individual estimates.^[10] The oesophageal and stomach cancer study similarly warned that differences in registration practice and staging systems "may have impacted the comparisons".^[4] None of this makes the gaps disappear, but it is why a difference of a percentage point or two between similar countries should not be read as a verdict on their health systems.

Survival data also look backwards. Five-year survival can only be measured for people diagnosed at least five years earlier, so CONCORD-3 describes people diagnosed up to 2014, and the ICBP stage studies people diagnosed in 2012 to 2014. Treatment and services have changed since. Rising survival can also partly reflect earlier detection: if a cancer is found sooner, the clock starts earlier even when death is not delayed (lead-time bias), and screening can find cancers that would never have caused symptoms (overdiagnosis). A US analysis of the 20 most common solid cancers found that changes in five-year survival between 1950 and 1995 bore little relationship to changes in mortality, and tracked changing patterns of diagnosis instead.^[11] Our guide to cancer survival statistics explains these measures.

The COVID-19 pandemic adds a new complication. An ICBP study in Lancet Oncology in 2026 estimated that across 18 jurisdictions in the same seven countries, 55,713 (16%) of 347,666 expected cancer diagnoses were missing between April and December 2020, with the largest shortfalls for prostate (24%) and female breast cancer (18%). The impact was greatest in the first months of lockdown, "when barriers in access to health care were greatest", and the authors say more research is needed on whether those people were diagnosed later and at a later stage.^[12]

The next global round, CONCORD-4, extends the series to more recent diagnoses and, for the first time, covers all cancers in children. Its first paper, on childhood cancers, was published in The Lancet in March 2026, using records for 613,021 children diagnosed in 1990 to 2019 from 307 registries in 68 countries and territories. For children diagnosed in 2015 to 2019, its index of five-year net survival for all childhood cancers combined was above 80% in most participating high-income countries, 60% to 80% in most upper-middle-income countries, and 50% to 60% in the five lower-middle-income countries that took part.^[13] At the time of writing (November 2026) the adult results had been presented but not published in a peer-reviewed journal, so for adults CONCORD-3 remains the most recent global comparison.

What the evidence asks of us

The international evidence points the same way from several directions. Where survival is lower, cancers tend to be found later, and people with the same stage of cancer receive different treatment. Both are things health systems can change.

• Find cancer earlier, and measure it. WHO's Global Breast Cancer Initiative rests on three pillars: "health promotion and early detection; timely diagnosis; and comprehensive breast cancer management". Its aim is to reduce breast cancer mortality by 2.5% a year, which over 20 years would save 2.5 million lives.^[14] The ABC-DO researchers modelled that reaching the initiative's target of 60% of breast cancers diagnosed at stage I or II, together with access to treatment, would cut deaths by roughly a third among Black women in the five countries studied.^[3] That is a projection, not an observed result, but it shows the scale of what is at stake.
• Close treatment gaps, not only diagnosis gaps. Because survival differs within stage groups, commissioners need data on who receives surgery, chemotherapy and radiotherapy, how quickly, and at what ages. The ICBP lung cancer findings, with treatment use varying by age even at the same stage, show why.^[8]
• Put children first where the gap is widest. WHO's Global Initiative for Childhood Cancer aims to raise survival for children with cancer worldwide to at least 60% by 2030, citing "substantial differences in survival across countries and health systems".^[15] CONCORD-4's childhood analysis found most participating countries at or near that level already, and its authors suggest the target "may not be ambitious enough".^[13]
• Fund population-based cancer registries. Every comparison in this article depends on them. Better registration, consistent staging and reliable linkage to death records, the improvements ICBP recommends, are what allow countries to see where they are falling behind.^[10]
• Watch for the after-effects of the pandemic. Missed diagnoses in 2020 may surface as later-stage disease. Services should track stage and survival for people diagnosed since then.^[12]

Language: this piece says "people with cancer" and "people living with and after cancer", and describes the illness and its treatment plainly, without combat metaphors. Clinical terms such as "stage", "incidence" and "net survival" are used in the sense the cited source defines them.

Why we're publishing this

Charities, support groups and hospices often hear the question "would I have done better somewhere else?" The honest answer, at population level, is that survival does differ by country, and that much of the difference comes from when cancer is found and what treatment follows, not from anything people with cancer did or did not do. We publish this so that organisations campaigning for faster diagnosis, fairer access to treatment and better cancer data can cite the evidence accurately, with its limits stated, and so that people living with and after cancer can see that the gaps are a matter for health systems to close.

How to read this data

• Net, relative and observed survival differ. CONCORD-3, SURVMARK-2 and England's statistics report net survival, which estimates survival from cancer alone. EUROCARE-6 reports relative survival, a closely related measure. ABC-DO reports both crude (observed) survival, which counts deaths from any cause, and net survival. Observed survival is always lower. Do not compare one measure with another.
• Check the period and the time point. Figures here describe people diagnosed between 1995 and 2022, at one, three or five years after diagnosis. A one-year figure cannot be set beside a five-year figure.
• Age-standardisation matters. The registry figures are age-standardised so that countries with older or younger populations can be compared.
• Coverage is uneven. Wealthy countries have long-running national registries; many low-income countries have none, or registries covering one city. CONCORD-3 includes 71 countries and territories, and the African figures here come from a hospital cohort, not a national register.^[1][3]
• Survival is not mortality. Survival follows people who have been diagnosed; mortality counts deaths in the whole population. Rising survival with flat mortality can reflect earlier detection or overdiagnosis rather than people living longer.^[11]
• Population figures say nothing about an individual. They describe groups of people diagnosed years ago, not anyone's own outlook.

Use this data

Free to cite with attribution to Better Health For Everyone. Charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's health, please speak to a qualified health professional.

Sources

1. Allemani C, Matsuda T, Di Carlo V, et al. Global surveillance of trends in cancer survival 2000–14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. The Lancet, 2018. europepmc.org/article/MED/29395269
2. Arnold M, et al. Progress in cancer survival, mortality, and incidence in seven high-income countries 1995–2014 (ICBP SURVMARK-2): a population-based study. Lancet Oncology, 2019. europepmc.org/article/MED/31521509
3. Mo T, Joffe M, Cubasch H, et al. Breast cancer overall survival, annual risks of death, and survival gap apportionment in sub-Saharan Africa (ABC-DO): 7-year follow-up of a prospective cohort study. The Lancet Global Health, 2025. europepmc.org/article/MED/40975077
4. Arnold M, Morgan E, Bardot A, et al. International variation in oesophageal and gastric cancer survival 2012–2014: differences by histological subtype and stage at diagnosis (an ICBP SURVMARK-2 population-based study). Gut, 2022. europepmc.org/article/MED/34824149
5. Cabasag CJ, Arnold M, Rutherford M, et al. Pancreatic cancer survival by stage and age in seven high-income countries (ICBP SURVMARK-2): a population-based study. British Journal of Cancer, 2022. europepmc.org/article/MED/35236937
6. Crocetti E, Bennett D, Jooste V, et al. Skin melanoma survival is improving in Europe but regional differences persist: results of the EUROCARE-6 study. European Journal of Cancer, 2025. europepmc.org/article/MED/41172568
7. NHS England. Cancer Survival in England, cancers diagnosed 2018 to 2022, followed up to 2023: cancer survival by stage. NHS England Digital, 2026. digital.nhs.uk/…/cancers-diagnosed-2018-to-2022-followed-up-to-2023/cancer-survival-by-stage
8. Barclay ME, McPhail S, Johnson SA, et al. Chemotherapy and radiotherapy use in patients with lung cancer in Australia, Canada, the UK and Norway 2012–2017: an ICBP population-based study. BMJ Oncology, 2025. europepmc.org/article/MED/40662158
9. Vener C, Lillini R, De Angelis R, et al. Long-term survival for myeloid neoplasms and national health expenditure: a EUROCARE-6 retrospective, population-based study. European Journal of Cancer, 2025. europepmc.org/article/MED/40056559
10. Andersson TM, Myklebust TÅ, Rutherford MJ, et al. Five ways to improve international comparisons of cancer survival: lessons learned from ICBP SURVMARK-2. British Journal of Cancer, 2022. europepmc.org/article/MED/35058590
11. Welch HG, Schwartz LM, Woloshin S. Are increasing 5-year survival rates evidence of success against cancer? JAMA, 2000. europepmc.org/article/MED/10865276
12. Morgan E, Bardot A, Langselius O, et al. International disruptions to cancer diagnosis and stage at presentation during the COVID-19 pandemic in 2020: an International Cancer Benchmarking Partnership (ICBP) population-based study. Lancet Oncology, 2026. europepmc.org/article/MED/42225103
13. Allemani C, Di Carlo V, Ssenyonga N, et al. Progress towards the WHO Global Initiative for Childhood Cancer target of 60% 5-year survival for all childhood cancers combined, 1990–2019 (CONCORD-4): a Cancer Survival Index derived for 68 countries by analysis of individual records for 613 021 children from 307 population-based cancer registries. The Lancet, 2026. europepmc.org/article/MED/41905383
14. World Health Organization. The Global Breast Cancer Initiative. WHO. who.int/initiatives/global-breast-cancer-initiative
15. World Health Organization. The Global Initiative for Childhood Cancer. WHO. who.int/initiatives/the-global-initiative-for-childhood-cancer