Are Supercentenarian Claims Based on Age Exaggeration?

Updated June 2024

A pre-print was posted to a pre-print server with a provocative headline questioning research in 3 of the 5 blue zones regions: “Supercentenarians and the oldest-old are concentrated into regions with no birth certificates and short lifespans.”

It has been latched onto by bloggers and even some reputable outlets as newly published peer-reviewed research. Since we’ve received so many questions about it, we felt the need to respond thoughtfully to the most misleading statements and assumptions in the pre-print here.

READ DEMOGRAPHERS’ STATEMENT IN RESPONSE TO PRE-PRINT

THE PRE-PRINT

The pre-print is not new data, research, or a study. It is a theory – an opinion paper that the author has been unable to get published in any scientific journal or peer-reviewed publication.
The original version of the paper was released as a pre-print in 2019 and still has not been properly published, meaning the theories have never passed any scientific peer review. The author has never published a peer-reviewed paper on this subject, despite releasing multiple versions of his paper to non-academic news media over the past five years. His failure to be accepted by any credible, peer-reviewed journal for so long indicates that his theories are not taken seriously by actual experts in the field.
Even some established media outlets do not understand the difference and many have reposted the idea as new data – or worse, as fact – without fact-checking any of the claims within.
The fact that even some reputable outlets gave so much attention to this pre-print without consulting any actual experts or checking sources is disconcerting. The pressure for clicks and views has incentivized media outlets to publish fringe “science” with negative effects. This is an unfortunate example of focusing on controversy (Myth-busting! Fraud!) while neglecting the overwhelming consensus of actual experts in the field.

THE HYPOTHESIS AND CLAIMS OF BIAS

The opinion paper is based on a hypothesis that the author sets out to prove. It is not new. It is the same one Buettner set out to investigate – are claims of exceptional longevity true? Some were, and those were named blue zones. Others were found not to be. There was no bias to claim some locations over others. (Also, the blue zones regions are not defined as places with supercentenarians, as the pre-print author seems to think.)
Buettner relied on some of the world’s best demographers and researchers to identify validated regions where people live the longest, including:

1 Age Verification in Blue Zones Research

Blue zones researchers did not rely on data crunching from afar; the research project’s goal was to actually go to these supposed longevity hotspots and validate or invalidate claims of exceptional longevity. It wasn’t based on datasets from afar. The author of the pre-print didn’t do his research, unfortunately. A team of researchers, demographers, and scientists visited each region to validate age. We checked birth certificates and cross-referenced with church baptism records or other available local records in all the international blue zones regions.

“The identification and certification of a blue zones area or group is based on demographic criteria that are country-specific and depending on available documentation and its reliability.”

The evidence supporting the existence and characteristics of blue zones regions includes demographic studies, epidemiological research, and ethnographic fieldwork.

The FIRST CRITERION for a blue zones region is:

“The reliability of individual longevity should be thoroughly proved and therefore the extreme ages of the oldest olds, dead or still alive should be proved based on ad hoc documentation. Without documentation like birth and death records and continuous population registration system, that reliability cannot be ensured.”

Blue Zones Criteria

An example of how vigorously data collection was in blue zones regions:

“The database developed for the present study includes all individuals born in Villagrande from 1876 to 1912. For each individual we traced the exact date at death or the proof that he/she was still alive at the date of investigation. The data was gathered from civil registers (which record all births, marriages, and deaths), parish registers and the population register (anagrafe). All information has been collected in the municipality population registration office and was cross-checked with information reported in the military register and orally reported information from any relatives of these persons. With regard to those who died outside the village, the information was recovered by using annotations on date and place of death reported in the margin of the birth certificate or transcription of the date of death in the anagrafe. For those who emigrated and for whom no death has been reported, the survival status has been verified with the municipality of current residence.”

Source: A Population Where Men Live As Long As Women: Villagrande Strisaili, Sardinia

The age validation in blue zones regions has been well-documented in both academic and popular literature:

The world’s demographers and researchers do understand the need for age validation:

2 Blue Zones are NOT Regions with Supercentenarians

Blue zones areas are not those with the most supercentenarians as the pre-print describes. The blue zones areas are places with the highest healthy life expectancy, where people reach their 90s with low rates of chronic disease, and where there is a high probability to reach 100. The blue zones criteria are clearly laid out.

3 Relative Poverty in Blue Zones Regions

The preprint makes this conclusion: “Relative poverty and short lifespan constitute unexpected predictors of centenarian and supercentenarian status, and support a primary role of fraud and error in generating remarkable human age records.”

Yes, certain blue zones regions like the Ogliastra cluster of villages in Sardinia are among the more remote and poorer regions of the country. They are all also places where modernization and the Western diet have been slower to take hold. People there walk rather than drive, eat more beans and plant foods than processed and fast foods, and they connect with their neighborhoods more than watch TV and stare at their phones. Most people in America will die of diseases of affluence, like heart disease, cancer and diabetes and in this sense, poverty in the blue zones asserted a benefit to the people living there. All blue zones do, however, enjoy very good or excellent public health services.

SOME EXPERT COMMENTS ON ASSUMPTIONS AND ERRORS IN THE PRE-PRINT:

OKINAWA ERRORS:

Response from Dr. Bradley Wilcox, Director of the Kuakini Center of Biomedical and Translational Research on Aging, the Principal Investigator of the Hawaii Lifespan Study and Hawaii Healthspan Study, and Professor and Director of Research in the Department of Geriatric Medicine at the John A. Burns School of Medicine, University of Hawaii:

“[The paper] has one poorly done “analysis” of U.S. demographic data on exceptional longevity…and then generalizes his “data” to make wild, unsubstantiated claims about Okinawa and other areas of exceptional longevity.

[The author] talks about the current state of some of these longevity hotspots, with no understanding whatsoever of these populations. For example, he states that the Okinawans are currently very unhealthy. Who is he talking about? The young Okinawans eat and drink too much. They are not physically active. They have rampant diabetes.

Yet, the older Okinawans are the healthiest of the Japanese prefectures. Life expectancy at birth has dropped. Life expectancy at ages 65 and 85 years is still the longest in Japan. What can explain this? It’s pretty obvious for any epidemiologist. They led healthy lifestyles. It’s called a cohort effect.

He is correct that the koseki had to be reconstructed in some areas. But in many other areas (which were not bombed to smithereens) it was intact and there were no significant differences in age errors between them.

The sad part of all of this is the gullible media who have not done their homework and only wish for a sensationalist story. Even some credible media outlets have fallen for this tripe.

But there is an actual story here – how a post-doc can fool the world that he is an expert on aging and longevity with no expertise in the field, no publication record, and a total lack of understanding of the populations that he is pontificating about.”

SARDINIA ERRORS:

Response from Gianni Pes, MD, PhD, Senior Researcher in the Department of Clinical and Experimental Medicine at the University of Sassari, Italy. Dr. Pes graduated in Medicine and received a PhD in Medical Statistics from the University of Pavia, Italy. He started to study Sardinian longevity in 1996 and was the first to report in 1999 on the exceptional longevity of the population living in the mountains of Sardinia; Dr. Pes has published more than 100 scientific articles in peer-reviewed journals.

In the article entitled “Supercentenarians and the oldest-old are concentrated into regions with no birth certificates and short lifespans” Dr. Saul Newman casts doubts on the authenticity of the areas with a high density of centenarians, semi-centenarians, and supercentenarians, especially found in those areas known as “Blue Zones”. It suggests that the alleged concentration of long-lived people in areas known for relative poverty and poor-quality documentation may be due to fraud and errors in age validation. He also notes that resorting to unclear “lifestyle factors” is not a sufficient explanation for the striking gap between longevity and socioeconomic indicators of poor health. Finally, the author cites various cases of supercentenarians in the US in whom documentation is absent or poor. It can be replicated that the fact that the US population as a whole is not a Blue Zone.

Page 2, line 35. The majority if not all remarkable age records may be errors.

Reply: Regarding the birth registration system in the Sardinia BZ, it must be noted that since 1866 both civil and ecclesiastical registrations have been available. As Wachter et al. (reference no. 21) noted, birth certificates in Italy, including Sardinia, were very accurate “filled out by Civil Status Officers, well educated for their time, who entered births chronologically into bound volumes, a separate volume for each year in each municipality”. In addition, ecclesiastical certification (Quinque Libri of the Catholic Church) is available. For none of the Sardinian supercentenarians identified up to now, a discrepancy between civil and ecclesiastical data has been found, with only one exception, duly published (Poulain, Pes et al., 2006). This double certification ensures that the probability of errors in the date of birth is virtually zero. At least for Sardinia, the implication of an “absence of birth certificates” (page 3, par. 45) is ridiculous.

Page 3, line 50. Supercentenarian birthdates also exhibit ‘age heaping’ distributional patterns.

Reply: The distribution of birth dates of the 14 Sardinian supercentenarians identified up to now is reported below, and appears to be quite homogeneous with no signs of ‘age heaping’. Although three of them were born in the month of November this is because, in general, it is more likely for a centenarian to be born in winter rather than in another season.

22 1 1889

23 1 1908

25 1 1900

14 2 1901

30 5 1902

19 6 1894

12 8 1901

12 9 1883

23 10 1904

12 11 1893

13 11 1905

18 11 1883

10 12 1901

29 12 1890

Page 3, line 50. Fewer than 15% of exhaustively validated supercentenarians are associated with birth or death certificates:

Reply. In the Sardinia BZ, all supercentenarians (100%) underwent age validation. Regarding French supercentenarians, the RNIPP database (Répertoire national d’identification des personnes physiques) identified a total of 231 potential supercentenarians between 1988 and 2016. Birth and death certificates were found in 213 cases (92%). Of the remaining 18 incomplete cases, 12 were cases of disappearance, five had errors in their year of death, and one had the wrong year of birth. In the case of women, the percentage of validated cases was 95% (Ouellette et al. Supercentenarians and Semi-supercentenarians in France. In: Exceptional lifespans, Eds. Heiner Maier, Bernard Jeune, James W. Vaupel, Rostock, Mecklenburg-Vorpommern, Germany).

Page 4, line 60 Regional distribution for the density of remarkable longevity.

Reply: It is not clear where the author obtained the data on Italy’s supercentenarians. It can be seen in Figure 2 that in Sardinia the greatest concentration of supercentenarians corresponds to the northernmost part (Gallura) and in the south-eastern part. This is absurd, and the real distribution of supercentenarians is that of the figure below.

Page 6, line 75. In total, over 81% of the total global supercentenarian population were located to their region of birth.

Reply: perhaps this is because supercentenarians were born in an epoch in which mobility was much more limited than now, especially in rural and economically disadvantaged regions.

Page 6, line 90. The number of supercentenarians falls by 80% per year overall after introduction of state-wide birth certification.

Reply: From figures 2a and 2b it is not clear whether the author reported the number of resident supercentenarians or the number of newborns per year who have become supercentenarians. In any case, the number of supercentenarians after 1900 appears to be between 3 and 5. For the US population, which was 76.3 million in 1900, this corresponds to a level comparable to that of Europe at that time. Besides, the US population is not characterized by exceptional longevity.

Page 7, line 105. High poverty rates were the best predictor of remarkable age records.

Reply. First, it is not clear whether the author refers to poverty recorded in the year of birth of supercentenarians or the year in which these very old people became supercentenarians. Second, care must be taken not to correlate the individual probability of reaching a very advanced age with poverty as measured by social indicators at the aggregate level. Perhaps the probability of becoming centenarians and supercentenarians is related not to poverty in itself, but to the persistence of healthier traditions/habits that are often associated with a relatively delayed economic development.

Page 8, line 115. The level of income deprivation […] predicts around 40% of the variation in supercentenarian abundance…

Reply: the indicator used to corroborate this statement is the Pearson correlation coefficient. This can only be used only when the relationship between two variables is linear and both variables are normally distributed and have no outliers. An unusually high correlation coefficient could be due to various reasons including small sample size: the number of supercentenarians is low in all populations, and this can lead to an artificially high correlation that may not be reflective of the true relationship between the variables. Second, supercentenarians are outliers within the group of long-lived people: the presence of extreme values, not representative of the general trend in the data, can artificially raise the correlation coefficient as well. Third, if the relationship between the variables is non-linear this can result in a misleadingly high correlation. How can we imagine of using linear prediction models if the probability of finding a supercentenarian is one in 1 million in most populations and calculating the ‘density’ of supercentenarians is a completely imaginary concept?

Page 8, line 120. Urban Manchester, Tyneside, and Liverpool, the UK “blue Zones”.

Reply. The areas mentioned are not at all validated Blue Zones. In the article, the author continually confuses the areas with a presumed high concentration of long-lived people with the real Blue Zones.

Page 8, line 125. …the three overseas departments (Guadeloupe, Martinique and La Réunion)…; page 17, lines 272-273. Guadelupe and Martinique rank equal second for total supercentenarians after Paris.

Reply. The age of supercentenarians from Martinique was validated by Jacques Vallin in 2017. The validation of the age of centenarians is still ongoing, and at the moment the age of approximately 50% of centenarians born before 1921 is considered reliable. The status of Martinique as a Blue Zone is therefore pending until the publication of the final validation data. For Martinique, it has been hypothesized (but not proven) that a strong genetic selection for survival occurred during the period of slavery, when those less resistant to disease were eliminated through a Malthusian process.

Page 9, line 135. Across the 66 region of France […] the poverty rate over age 75 predicted half of the variation in supercentenarians abundance…

Reply. Looking at Figure 3b, we note the presence of outliers that may have falsely increased the correlation coefficient. In addition, the number of supercentenarians in a single district is so small (and sometimes zero) that the concept of “the density” of supercentenarians is problematic.

Page 10, lines 152-154. In Japan […] it can be predicted that these estimates of old-age poverty may provide better predictor of centenarian abundance…

Reply. A certain degree of economic underdevelopment that has historically characterized the Blue Zones seems to call into question the “social gradient theory” proposed by Albert Evans in 1994, according to which the longevity of a given population would grow in direct proportion to its economic prosperity thanks to the reduction levels of anxiety and stress (Evans A. Why are Some People Healthy and Others Not?: The Determinants of Health Populations, 1994). In reality, the cases of Okinawa and Sardinia demonstrate that the highest longevity is observed where the modernization process began later and increased more slowly, as Cockerham appropriately underlined in a 2001 article (Cockerham WC et al. Okinawa: an exception to the social gradient of life expectancy in Japan Asia Pac J Clin Nutr. 2001).

Page 13, line 205 Autocorrelation between age-specific survival breaks down and inverts at advanced ages; page 15, line 236-238 It seems unusual that modern economic conditions and poverty rates are predictive of reaching age 110…

Reply. In Figure 5 of Newman’s paper, it can be seen how the direction of the line depends on several outliers. Any inference about such unevenly distributed numbers is unwise. Further, survival at age 55 and survival after age 95 refers to different generations, which may have been exposed to different positive (or negative) health factors.

Page 16, lines 246-247. Tower Hamlets has the highest poverty rate, the shortest disability-free life expectancy, the highest income inequality and the worst index of multiple deprivation.

Reply. On the premise that Tower Hamlets has never been considered as a Blue Zone, the observation that it seems to have the highest number of semi-supercentenarians must be proven through a serious age validation. If data were reliable, the evidence must be accepted. Incidentally, the influence of genetic factors on becoming semi-supercentenarians and supercentenarians is stronger than on becoming “simple” centenarians. The existence of genetic factors acting locally could, in theory, generate a cluster of related supercentenarians (it was observed in Sardinia) and therefore explain the apparent increase in their density.

Page 18, par. 305. The Italian province of Olbia-Tempio ranked as the best province for survival to ages 100, 105 and 110

Reply. It is not clear from what source these data came, given that the province of Olbia Tempio (Gallura) in Sardinia is the one with the lowest longevity index (extreme longevity index, ELI).

Page 19. Birthdays generally varying by less than two percent across different days of the month

Reply. The examination of the birth dates of approximately 3,500 Sardinian centenarians shows that the percentage of months of birth is between 5% and 11% (this variation is explained by the fact that winter months are slightly overrepresented). The distribution of the days of the month is between 2.5% and 4.5%, compatible with a randomly chosen birthday.

Page 19, line 331. Age heaping.

Reply. The percentages of the months of birth of 100 Italian supercentenarians are as follows:

JAN 7%, FEB 7%, MAR 11%, APR 11%, MAY 9%, JUN 3%

JUL 4%, AUG 6%, SEP 7%, OCT 13%, NOV 13%, DEC 9%

No sign of deviation from randomness. The first day of the month was not overrepresented.

Page 21, par. 350 Social and economic indicators normally linked to worse health outcomes are positively associated with a higher probability of reaching and extreme age.

Reply. Living in poverty does not necessarily mean poor health (Glass TA, McAtee MJ. Behavioral science at the crossroads in public health: extending horizons, envisioning the future. Soc Sci Med. 2006;62:1650-1671). Sardinia is a region with lower income (€ 23,700) than the Italian national average (€33,000) but this does not mean it is ‘poorer’ if by poverty we mean the lack of economic resources indispensable to meet the needs of life and a lower access to social services and healthcare.

Page 22, lines 367-368. Italy is a historically smaller, poorer, less well-educated, and shorter-lived country.

Reply. Historically there was no longevity in Italy. The phenomenon of centenarians in Italy only exploded between 1980 and 2000 in conjunction with economic development. Forty years ago the country was by no means less well-educated and shorter-lived. However, after 2000 Italy suffered long-term economic and financial problems and consequently, the growth of centenarians levelled off.

Page 22, lines 382-383. The existence of widespread errors and pension fraud is often assumed to be unlikely or impossible in these countries.

Reply. The potential existence of pension fraud was carefully examined during the validation of the centenarians of the Sardinian Blue Zone and was definitively excluded, and is at least unlikely for the other three official Blue Zones, although they have not been investigated as systematically. The author makes insinuations without any material evidence.

Page 23, lines 394-395. In 1997 Italy discovered it was paying 30,000 pensions to dead people.

Reply. Here there is some confusion between the Italian social security database and the demographic database maintained by ISTAT. The latter, unlike the former, depends on the availability of a death certificate. As proof of this, in Sardinia in 1997 the number of centenarians registered in the social security database was almost double the number in the demographic database (provided by municipalities). Consequently, many centenarians were formally in receipt of a pension even if they were dead, but were rightly absent in the databases of the demographic services of the municipalities.

Page 24, lines 430-434. Despite prior claims of dietary benefits based on vegetable and sweet potato consumption […] Okinawa has the single lowest consumption per capita of fruit…

Reply. I am afraid that again there is confusion between different generations. The generations born in Okinawa after 1945 have partly abandoned traditional eating habits, and consequently, the level of longevity is no longer at the top among Japanese prefectures. But what matters is that, in previous generations, exceptional longevity was real and was most likely linked to a diet rich in vegetables etc. These criticisms lack historical perspective.

Page 26, lines 456-457. The large scale US-bombing and invasion of Okinawa involved the destruction of entire cities and town, obliterating around 90% of the birth and death records.

Reply. As far as I know, a copy of the Okinawan birth records has been preserved and microfilmed in Hawaii and has allowed the documents lost during the war to be reconstructed without necessarily resorting to personal recollections of the families, which may introduce errors.

Page 26, line 469-470. […] the Blue Zone of Sardinia which during a survey had the highest murder rate in Italy.

Reply. The homicide rate in Sardinia is 0.8 per 100,000 inhabitants, slightly lower than the European rate which is 1 homicide per 100,000 inhabitants, and 30% lower than that of the author’s country, Australia.

Page 28, lines 502-503. Smoking rates of e.g. 17-50% and illiteracy rates of 50-80% are often observed in samples of the oldest old.

Reply. In a paper on nonagenarians from the Sardinian Blue Zone, the percentage of current smokers was 1% in males and zero percent in females (Pes et al., Rejuvenation Res. 2020;23:341-348) and practically no one was illiterate. If how long the smoking habit lasted is not specified, it makes no sense to make correlations with longevity. In Ikaria, the percentage of male smokers was higher (Poulain et al., Mech Ageing Dev. 2021;198:111543) but below that mentioned by the author.

Page 30, lines 545-546. The observation of 60% smoking rate in centenarians who are older and representative of the same cohort presents some difficulty.

Reply. Centenarians are what remains of generations where smoking rates were higher than now. It may be interesting to ask why they didn’t die sooner since they smoked. It is possible that some of them were protected by genetic factors, the importance of which for longevity increases with age. It is also possible that the amount or the method of smoking made the habit less harmful.

Page 30, lines 559-560. It is unclear why clinically excessive drinkers or daily smokers should survive at equal or higher rates; page 35, line 666 Anomalous health habits of supercentenarians.

Reply. In none of the Blue Zones long-lived people drink alcohol or smoke excessively. In Sardinia and Ikaria, the hypothesis has been put forward that moderate drinkers are protected from cardiovascular diseases due to the polyphenol content in red wine, although there could be other positive effects due to the adoption of the Mediterranean diet. The assertion that these behaviors are linked to the perpetration of fraud is gratuitous.

Page 31, lines 577-579. Overall only 6.6% of supercentenarians have an original birth certificate, and 74% of cases have no reported birth documents of any kind.

Reply. Not all alleged supercentenarians have birth and death documents, but the validated ones do. Here the ambiguity is due to avoiding the distinction between supercentenarians included in non-scientific databases such as GRG and those maintained by scientific organizations, like IDL, that are required to publish validation documents.

Page 32, lines 598-599. Like birth certificates, death certificates are completely absent or severely under-represented.

Reply. If a supercentenarian is still alive, obviously there is no death certificate, and most supercentenarians are entered into the databases while they are still alive. In the Blue Zones, all supercentenarians have been validated and their birth and death records are available. Rarely, in Sardinia, false supercentenarians have been identified and mentioned in specialized publications.

Page 36, lines 694-695. …a conclusion that questionably re-shapes past smoking status as a positive indicator of survival.

Reply. As long as is not specified how long one has smoked, any inference regarding the relationship between smoking and longevity is invalid. In Sardinia and Ikaria some centenarians, registered in questionnaires as “smokers”, had smoked only during military service and then stopped. The number of daily cigarettes may have been small even for prolonged periods. Finally, even if the author derides the use of “genetic” factors to explain the lack of danger of smoking, it cannot be ruled out a priori.

Page 37, lines 715-715. Perhaps, supercentenarian databases remain riddled with errors.

Reply. Correct. Provisional databases of supercentenarians can be full of false cases. After careful age validation their genuineness can be accepted.

Page 38, line 744. Physical possession of valid document is not an age guarantee.

Reply. Correct. Only after cross-referencing various documents relating to an alleged supercentenarian, and having found them consistent, we can be certain of his/her age. In Sardinia, this was done systematically during the complete reconstruction of the village genealogy. In this way, the exchange of an alleged supercentenarian with his/her older sister who had already deceased was discovered. If the birth intervals between siblings do not present any gaps, such a substitution is unlikely.

Page 39, lines 769-770 …it requires the measurement of biological ages from fundamental physical properties, such as amino acid chirality or isotopic decay.

Reply. In addition to not being feasible when biological material is lacking, the request to use chemical methods to validate the chronological age of a supercentenarian is not a guarantee of certain results. The method that is currently preferred is the use of epigenetic clocks, which however show the existence of a significant gap between chronological age and biological age. It is not clear why simple demographic evidence should be rejected provided the documents consulted are mutually consistent.

Comments from Robert Young, a gerontologist and researcher who validates supercentenarian cases and debunks longevity claims. (Source: Columbia University Statistics website, which posted the pre-print for discussion):

“As the Senior Consultant for Gerontology for Guinness World Records and the Director of the Gerontology Research Group-Supercentenarian Research and Database Division, I am surprised to see this paper improperly using GRG data without talking to us first about it. This paper, intentional or not, is loaded with so many errors and misrepresentations that I feel that the author should contact the GRG (and the IDL), and myself personally, to discuss a major revision, if not outright retraction.

…The author appears to selectively misuse data to make a point, which is more reminiscent of a political than scientific approach. For example, many of the “missing” Japanese centenarians were actually persons that disappeared in WWII and is not relevant. Further, the island of Ikaria, Greece has never produced a validated supercentenarian so is not a part of the IDL or GRG datasets. In addition to that, the Sardinia data has been properly validated with actual evidence and the data from Sardinia fits well with other validated data. For example, no one in Sardinia’s data has been older than 112. Compare that to the typical longevity claim to ages 115-130 and it’s a totally different situation.

Strangely enough, Newman seems unaware that demographers already have methods to detect ID frauds…and have already done so, in Italy.

It seems strange, and uninformed, to set up a “thought experiment” about a hypothetical question that was already answered decades before the question was asked. Perhaps Newman needs to study the topic area further before trying to shoehorn in controversial, unproven assertions.”