The blue zones are about half real. The habits hold up, but the head counts do not. US states began issuing birth certificates one by one. In the states that did, records of people living past 110 fell by 69 to 82 percent (Preprint). That analysis has never appeared in a peer-reviewed journal, which is the first thing worth knowing about it.
Two claims got tangled together here: how many people in these places reached 100, and how those people lived. Only one of the two is in trouble.
What the Blue Zones Are
In 2004 a team led by Michel Poulain and Gianni Pes ran the AKEA study, a survey of villages across Sardinia (Study). They went to the registry offices of the 40 towns with the most people aged 100 or more. There they checked birth, marriage, death and population records by hand. The name comes from the blue ink they used to ring the area on a map.
Of 18,000 people born there between 1880 and 1900, 91 reached 100 (Study). That was 47 men and 44 women. It works out at about three times the Sardinian average. The near-even split made demographers sit up. Earlier counts of people aged 100 found more than five women for every man.
Then the idea travelled. Five regions became famous: Sardinia, Okinawa, Ikaria, Nicoya and Loma Linda in California. An observation about odd records turned into a lifestyle plan, sold as the explanation. The original paper was more careful. It warned that other claimed longevity hot spots had never been fully checked (Study).
The Birth Certificate Problem
Saul Justin Newman is a demographer, and he asked where the world’s supercentenarians live. A supercentenarian is someone aged 110 or older. His answer was uncomfortable. The records cluster in places with no birth registration and short average lifespans (Preprint).
Three numbers carry his argument. Where US states brought in birth certificates, records of people past 110 fell by 69 to 82 percent. Only 18 percent of the people who passed the strictest age checks hold a birth certificate. In the United States that figure is zero. Their birthdays also bunch on dates you can divide by five (Preprint). Real birthdays do not behave like that, but guessed ones do. The work won an Ig Nobel Prize in 2024, the first ever given in demography.
The analysis itself is a preprint on bioRxiv. That is a site where work goes up before other scientists review it, and no journal has published it. Newman posted it in 2019 and has revised it since, most recently in November 2025 (Preprint). He says one journal put it through nine rounds of review, then neither accepted nor rejected it (Report). The journal has not given its side. Read it as a strong argument that nobody has signed off.
His peer-reviewed record is separate, and it is real. In 2018 he argued in PLoS Biology that death rates only seem to slow in extreme old age (Study). The cause is bad records, not biology. Errors in fewer than 1 in 10,000 records were enough to produce it. A companion paper picked apart the model behind the slowdown claim (Study).
It is not the last word, and the same journal ran the reply. Kenneth Wachter put Newman’s own error rates back through the maths. They implied that almost every recorded 110-year-old was really 100 (Response). Birth certificates had been collected for every Italian over 110 in that data. Italy has registered births since the 1860s, so the mix-up Newman assumes cannot happen there. Wachter called the implication not credible.
What the Pension Audits Found
The strongest part of the case does not rest on the preprint at all. In late summer 2010, Japan’s Ministry of Justice checked its family register, the Koseki, against reality. Over eleven days it examined 47,439,848 registers, about 90 percent of the total. It could not confirm 234,354 people listed as aged 100 or more (Study). Of those, 77,118 would have been 120 or older.
The likeliest cause is dull paperwork. Deaths were often never passed between the family register and the resident registry. A record cannot be deleted without a death certificate. Some town offices had not gone digital, so clerks matched the files by hand (Study).
Japan does not work out life expectancy from the family register. So those 234,354 records changed the national figures by nothing at all (Study). The authors say the resulting doubts about Japanese longevity are misplaced. A second check knocked on doors instead. It found 584 missing people aged 100 or more out of 44,449, or about 1 in every 80 (Study). In Okinawa, the Japanese blue zone, it found none (Response).
Greece is the blunter story. From late 2011 the Ministry of Labour ran a census of pensioners. It stopped the payments that had no paperwork behind them. The projected result was 48,787 pensions that should never have been paid (Report). Against 2.4 million pensioners, that is about 2 in every 100.
The part about people aged 100 is thinner than it sounds. Reuters reported in 2012 that Greece had halted 200,000 benefit payments. They went to people who were not entitled to them, or who were dead. About 9,000 registered 100-year-olds drawing old-age benefits had prompted the review (Report). The source was one unnamed ministry official. That single news story is what the preprint cites for its Greek fraud claim.
A pension gives a family a monthly reason not to report a death. Old-age records rarely get checked until money runs short.
None of that proves anything about a particular village. Ikaria’s registers were checked on their own, back to town birth records from 1890. In 2009 researchers confirmed 124 islanders aged 90 or over still living (Response). That is close to five times the rate on the Greek mainland.
What the Defenders Checked
The blue-zone researchers answered in peer review. Steven Austad and Gianni Pes published their reply in The Gerontologist in November 2025 (Response). Pes helped identify the Sardinian zone, so this is a team defending its own work. Their first point is about what the term means. Blue zones were never about the number of people aged 110, they write. They were about the share of a population that reaches 90. On that measure, the two sides are partly arguing about different things.
Their second point is the archive work. The team matched ages against civil records, marriage and military records, voter rolls, church books and family archives (Response). In Sardinia that meant civil records back to 1866, parish books from the 1600s, and full village family trees. In Okinawa, a 2005 check looked at 52 people said to be 100 or older. Their stated ages matched exactly in 49 cases. In Nicoya, researchers visited 35 claimed 100-year-olds in 2007 and rejected one. He said he was 107 and had not yet turned 100.
In April 2026 a group led by S. Jay Olshansky proposed a formal definition, backed in the announcement by Austad and by Dan Buettner, who made the idea popular (Announcement). A place would have to pass two tests. The first is unusually strong survival after 70. The second is unusually high odds of reaching 100 once you have reached 70. And no place could qualify unless outside researchers were allowed to inspect the records.
The same publishing test applies to their side too. The definition is a proposal, not a settled standard, and the analysis behind it is itself a preprint (Preprint). It confirmed that Sardinia, Okinawa and Nicoya stood out in the year each was named. It also found their advantage narrowing since about 2010. Ikaria, Martinique and Loma Linda could not be tested, for want of usable data.
The biggest concession comes from the defenders, not the critics. Blue zones are not necessarily permanent, they write. Okinawa no longer qualifies as one (Response).
Okinawa Lost Its Lead
For about thirty years the prefecture had the longest life expectancy in Japan. Then it lost that lead. In 2002 the national life tables put Okinawan men 26th of 47 prefectures, down from 4th in 1995 (Study). Locally the drop was called “shock 26”. Writing in 2008, researchers dated the loss of the male advantage to 2000 (Study). Younger Okinawans were already dying at higher rates than the rest of the country.
The decline runs by generation. In 1975, men aged 50 to 54 died at about the national rate. By 2015 they were dying at roughly 40 percent above it (Study). Men aged 80 to 84 were still dying slightly below it. Every generation born after the 1950s did worse than Japan as a whole, and the gap widened with each one.
The official 2020 life tables show the crossover. Okinawan men now rank 43rd of 47 for life expectancy at birth, and 43rd again at age 40. Yet at 75 they rank 2nd, and Okinawan women rank 1st at both 65 and 75 (Life tables). So the oldest Okinawans still live longest in Japan, while the middle-aged sit near the bottom. Okinawa also has Japan’s highest obesity rate today (Study).
The diet story needs the same care. The famous figure is that sweet potato supplied 69 percent of Okinawan calories. It comes from a 1949 survey of 2,279 people (Study). They ate about 1,785 calories a day. That was roughly 200 less than researchers estimated they needed, and even that gap is uncertain. The American occupation government ran the survey, four years after the Battle of Okinawa. A separate check that year examined 1,826 islanders. About 1 in 5 showed a sign of poor nutrition (Study).
Whether that was helpful calorie restriction or plain postwar hunger is still argued over in the journals (Comment). Either way, a hungry occupied island in 1949 is a weak basis for a menu plan.
A place that loses its advantage one generation after changing how it eats tells you something. The advantage was probably never in the soil or the genes. The researchers tracking the decline add a warning of their own. Part of the remaining old-age lead may rest on ages nobody has checked (Study).
The Habits Still Hold Up
People often throw this part away by mistake. Every habit the blue zones made famous has its own evidence. That evidence comes from following ordinary people for years. None of it depended on counting anyone’s 100th birthday.
Take food first: a 2025 analysis pooled 25 long-term studies (Meta-analysis). People eating the most whole plant foods had about a seventh lower risk of dying than those eating the fewest. People eating the most refined and sugary plant foods had about a fifth higher risk. That means white bread, soft drinks, sweets and fries. Both patterns count as plant-based, and only one of them helped. A separate 2025 analysis of 977,763 people landed on nearly the same two numbers (Meta-analysis).
A model of 467,354 UK adults estimated what a lasting change might buy. A 40-year-old moving from a poor diet to a healthier one gained roughly 9 extra years (Study). Starting at 70, the same switch bought about 4 to 5 years. A halfway shift buys roughly half that. It is a projection, not a measured outcome, and it assumes you keep it up. Our look at the Mediterranean diet heart trials covers what happens when this gets tested directly.
Movement works the same way. The blue-zone version is walking to get places, not exercise as a project. The useful dose is lower than most people assume, which we covered in why 7,000 steps beats 10,000.
Connection has a deep evidence base. One pooled analysis followed 308,849 people across 148 studies (Meta-analysis). People with stronger social relationships were roughly 50 percent more likely to be alive at the end. Living alone, on its own, predicted little. What tracks with survival is the depth of the ties, the thread running through loneliness and health risk.
Purpose holds up too, with one honest asterisk. A 2026 analysis pooled 488,765 people and 48,928 deaths, following some of them for 32 years (Meta-analysis). People reporting a stronger sense of purpose had about 30 percent lower risk of dying. Much of that link shrank once health habits and medical conditions were counted. It shrank further once depression was counted too. Purpose seems to work partly through what it makes you do.
One honest limit is left. The branded Blue Zones Project has been through peer review once, across three California beach cities from 2010 to 2017 (Study). Researchers compared them with 15 matched towns. The main measure was a score in which people rated their own well-being. The towns did improve. But nobody counted deaths or disease, and nobody was assigned by coin flip. The study team also included staff who ran the project. Living like a blue zone has never really been tested.
Frequently Asked Questions
Are the blue zones fake?
No, but the popular version oversells them. Counts of very old people in some regions look unreliable. Records of extreme age cluster where birth registration is missing (Preprint). The archive checks the blue-zone researchers ran are real, and their measure is the share of people reaching 90 (Response). The habits stand on firmer ground than the map does.
Has Saul Newman’s blue zones paper been peer reviewed?
No. It is a preprint on bioRxiv, first posted in 2019 and revised since. The latest version is from November 2025 (Preprint). Newman says one journal ran it through nine rounds of review without accepting it (Report). His related work on registration errors did pass peer review, in PLoS Biology in 2018 (Study).
Did people in Okinawa really live longer?
The generations born before World War Two did, and they still lead Japan. Okinawan men rank 2nd of 47 prefectures for remaining years at age 75, and Okinawan women rank 1st (Life tables). But Okinawan men now rank 43rd at birth. Every generation born after the 1950s has done worse than Japan as a whole (Study).
Is the blue zones diet worth following?
The food pattern is well supported, with one condition. Eating mostly whole plant foods is linked to a lower risk of dying in the years that follow. A plant-based diet built on refined and sugary food is linked to the opposite (Meta-analysis). What has never been tested is the branded program. Its one published evaluation measured how people rated their own well-being (Study).
Key Takeaways
- Extreme-age records track missing paperwork. Where US states started issuing birth certificates, records of people past 110 dropped by 69 to 82 percent (Preprint).
- The headline analysis is not peer reviewed. It has sat on bioRxiv since 2019, and its author says one journal has neither accepted nor rejected it (Report).
- National audits found the ghosts on their own. Japan could not confirm 234,354 people registered as aged 100 or more, though its life expectancy figures never used that register (Study).
- The archive checks are real. Ages were matched against civil, church, military and voter records, and false claims were thrown out (Response).
- Okinawa’s lost lead points to habits. Its men fell from 4th to 26th of 47 prefectures around 2000, and the island now has Japan’s highest obesity rate (Study).
- The habits never needed the map. Whole-plant eating (Meta-analysis), strong social ties (Meta-analysis) and a sense of purpose (Meta-analysis) each track with a lower risk of dying.
Copy the Habits, Not the Map
The blue zones are best read as an idea that got popular before anyone audited its data. That is not a scandal, and a lot of science goes this way.
The practical advice barely moves. Eat mostly whole plants, and go easy on the refined stuff (Meta-analysis). Walk most days. Keep people close (Meta-analysis). Keep a reason to get up in the morning. You do not need an island, a diet brand, or a birth certificate from 1912.
Pick one and start this week. If you want the choice with the least guesswork behind it, make it the food.
The evidence on the habits is solid, and the head counts are still being argued over. Both can be true at once.
This article is for educational purposes and is not medical advice. Talk to a qualified clinician before changing your health regimen.
Leave a comment