Content uploaded by Gianni Pes
Author content
All content in this area was uploaded by Gianni Pes on Feb 06, 2015
Content may be subject to copyright.
Vienna Yearbook of Population Research 2013 (Vol. 11), pp. 87–108
The Blue Zones: areas of exceptional longevity
around the world
Michel Poulain, Anne Herm and Gianni Pes∗
Abstract
The aim of this study was to compare the level of population longevity and the
characteristics of four geographic areas where unusually high proportions of long-
lived individuals have been observed. For these areas (Ogliastra in Sardinia, Okinawa
in Japan, the Nicoya peninsula in Costa Rica and the island of Ikaria in Greece). The
term of ‘blue zone’ (BZ) given to these areas is defined as a limited region where
the population shares a common lifestyle and environment and whose exceptional
longevity has been accurately verified. This paper discusses the use of different
indexes to measure the longevity of a population. As a preliminary result of our
investigations we confirm the exceptional level of male longevity in the Sardinian BZ
and both male and female longevity in Okinawa. Considering possible explanations,
we observed that BZ populations are geographically and/or historically isolated
(islands and mountainous regions). These populations succeeded in maintaining a
traditional lifestyle implying an intense physical activity that extends beyond the age
of 80, a reduced level of stress and intensive family and community support for their
oldest olds as well as the consumption of locally produced food. This is likely to have
facilitated the accumulation of ideal conditions capable of limiting the factors that
negatively impact on health in most Western populations. These people experienced
the epidemiological transition—and its implications—in relative recent times, and
their remarkably good health status during ageing could be the result of a delicate
balance between the benefits of the traditional lifestyle and those of modernity
(increased wealth, better medical care). All these factors could have promoted an
ideal milieu for the emergence of long-lived phenotypes at the population level.
∗
Michel Poulain (correspondance author), Estonian Institute for Population Studies, Tallinn University,
Estonia; IACCHOS, Universit´
e catholique de Louvain, Belgium.
Email: michel.poulain@uclouvain.be
Anne Herm, Estonian Institute for Population Studies, Tallinn University, Estonia.
Gianni Pes, Department of Clinical and Experimental Medicine, Universit
`
a degli Studi di Sassari, Italy.
DOI: 10.1553/populationyearbook2013s87
88 The Blue Zones: areas of exceptional longevity around the world
1 Individual longevity versus population longevity
Longevity is a phenomenon related to individuals (individual longevity) as well as to
populations as a whole (population longevity). While an exceptional age at death or
the extreme survival of the oldest old individuals generally allows identifying the
first phenomenon, the second can only be characterised through various indexes that
are mostly associated with life tables computed for that population as a whole. When
searching for the determinants of population longevity, the relevant characteristics or
behaviours are those shared by a large part of the population. By considering these
common characteristics, the chance to find more powerful explanatory variables is
increased as most persons concerned are born and live in the same place and thus
are more likely to share the same genetic
make-up
, early life conditions as well as
traditional behaviours and habits, including the same locally produced food. So, by
identifying areas where people live the longest, the search for longevity determinants
could be improved.
Practically speaking, gerontologists use the term longevity to refer to any
individual who is beyond the age 90 and actively functional. Living to 100 is an
attractive threshold in the lifespan for the media as well as for research although it
does not have any demographic significance in itself. Other age thresholds could
be used to define the gates of longevity
1
but the majority of research dealing with
longevity considers usually centenarians. Based on the proportion of centenarians
enumerated in censuses or mortality rates, the first potential longevity areas were
identified at the beginning of the 20th century. The US Census Bureau compared
the proportion of centenarians in different countries and pointed out the exceptional
cases of Bolivia with 75 centenarians per 100,000 in 1900, Bulgaria with 60 in 1905
and the Philippines with 51 in 1903 (Bowerman 1939).
In the January 1973 issue of National Geographic magazine, the physician
Alexander Leaf gave a detailed account of his journeys to countries of purported
long-living people: the Hunzas from Pakistan, the Abkhazians from the Soviet Union,
and Ecuadorians from Vilcabamba. According to Leaf, there were ten times more
centenarians in these countries than in most Western countries and he pointed out
that each of them was characterised by poor sanitation, infectious diseases, high
infant mortality, illiteracy, and a lack of modern medical care making the inhabitants’
extreme longevity even more extraordinary (Leaf 1973). However, some years later
Mazess and Forman showed that age exaggeration was predominant in Vilcabamba
with a large number of the men and women who tended to increase their age in
order to improve their social status or to promote local tourism (Mazess and Forman
1979). Later Leaf acknowledged this conclusion and made a final statement agreeing
1
As an example, in Okinawa, every year on the third Monday of September, during the ‘Respect
for the Aged Day’, those persons who reached the age of 97 are celebrated by the local municipality
following a special ritual called ‘Kajimaya’ symbolising one’s return to youth (The Secret of Longevity
the Okinawan Way! (http://www.dietriffic.com/2009/02/06/longevity/)
Michel Poulain, Anne Herm and Gianni Pes 89
that no evidence proves the unusual ages in the village of Vilcabamba (Leaf 1981).
More recently, demographers have become increasingly concerned with the accuracy
of longevity claims given the unprecedented rise in very old people in developed
countries (Jeune and Vaupel 1999; Young et al. 2010). As a consequence, more
careful checks have been conducted resulting in a systematic invalidation of all
allegedly long-living populations on earth as most claims of extreme age appeared to
be undocumented or exaggerated. In 1999, following the publication of data showing
extreme male longevity in Sardinia (Deiana et al. 1999), this scepticism pushed
demographers to assess the validity of the alleged ages of the oldest olds in Sardinia
(Koenig 2001). Based on a strict validation method, the ages of Sardinian centenarians
were thoroughly proven to be correct (Poulain et al. 2006). This validation was
based on investigations in the civil register of births from the two last decades of
the XIX century and the register of deaths from the last two decades of the XX
century. Considering the marginal annotation on death found in the birth register, all
centenarians were identified by place of birth. Surprisingly, the spatial distribution
of Sardinian centenarians according to their place of birth was far from random.
Figure 1 helps to clarify the difference in the spatial distribution of centenarians in
terms of low or high prevalence, as well as presence or absence of concentration of
centenarians.
The number of centenarians born in a given place was reported to the number of
newborns in the same place a century earlier. The spatial distribution of centenarians
by place of birth smoothed by a Gaussian method was fully different from the original
one that considered the distribution of centenarians by place of residence without
considering the variation in the size of the population (Figures 1a and 1b).
As a result of the validation process, an area has been identified in the mountainous
part of Sardinia with a significantly higher proportion of centenarians out of
population born in the same place. This area was called the ‘blue zone’
2
(Poulain
et al. 2004) and the concept of longevity blue zones has since been elaborated as areas
where the population is characterised by a significantly higher level of longevity
compared with neighbouring regions, provided that the exceptional longevity of
people in this population has been fully validated.
In practice, a blue zone (BZ)
is defined as a rather limited and homogenous geographical area where the
population shares the same lifestyle and environment and its longevity has been
proved to be exceptionally high
. Other validated BZs have been found so far in
Okinawa (Japan), on the Nicoya peninsula (Costa Rica) and on the island of Ikaria
(Greece).
The advantages of the BZ concept are evident as potential longevity determinants
may be found among common traits of the concerned population and characteristics
of the shared lifestyle and environment. Moreover some of longevity determinants,
identified for one BZ, could apply to populations of other BZs as well. Accordingly,
2The term was chosen simply because at the time the authors used a blue pen on a map to mark the
villages with long-lived population.
90 The Blue Zones: areas of exceptional longevity around the world
Figures 1a, b, c and d:
Spatial distribution of long-lived subjects in a given area
(c) Spatially randomly distributed
(no concentration)
(d) Non randomly distributed
(concentration)
(a) Low density (b) High density
developing research investigations on several BZ populations and comparing their
characteristics and behaviour is a potential step forwards when searching for longevity
determinants. In this perspective, the present contribution proposes a first comparison
of the longevity level as observed in the four BZs identified so far.
2 Data and method
To identify a BZ and to prove the exceptional longevity of its population it is
necessary first to validate the individual longevity of people living in this area and
more precisely to assess the exact age at death or the extreme survival of oldest olds in
the concerned population. Age misreporting and, more specifically, age exaggeration
must be ruled out. Perls explained that “there are several geographical areas that
have claimed inhabitants with extreme longevity”, and would be therefore considered
as longevity areas, “but after closer examination these claims have been found to
be false” (Perls et al. 2002). He concluded that such cases of extreme longevity
Michel Poulain, Anne Herm and Gianni Pes 91
Figures 2a and 2b:
Spatial distribution of centenarians in Sardinia
(a) Crude distribution by place of residence (b) Smoothed distribution by place of birth
after adjusting for birth cohort size.
required detailed scrutiny because they are so incredibly rare. Young et al. (2010)
listed various cases of invalidation of extreme ages while Poulain (2010) applied
validation rules on a selection of supposedly long-living persons. In addition to the
validation rules for individual centenarians, the validation of an extreme population
longevity demands several specific investigation steps that could vary depending on
the current availability of data sources for every specific population. In order to assess
the level of population longevity that is estimated through the average individual
longevity within a given population, data on all births and deaths occurring within
this population must be collected. The key objective is to ensure exhaustive data on
births a century ago, data on currently surviving centenarians and data on deaths of
centenarians during the last decades.
92 The Blue Zones: areas of exceptional longevity around the world
More precisely, the methodology used for identifying the first BZ in central
Sardinia was the following one:
•
a target area was selected covering all municipalities located in the Nuoro and
Sassari provinces—two areas that emerged as potential places for exceptional
longevity according to AKEA research results (Deiana et al. 1999);
•
birth registers of municipalities in this area from 1880 to 1901 were scrutinised
and the numbers of births were counted;
•
based on marginal annotations of the date of death on the birth record, persons
who died at age 100 and over were exhaustively identified;
•
additional centenarians still alive were identified, whether they were surviving
at their place of birth or elsewhere;3
•
the age of each centenarian was validated according to accepted criteria for
individual age validation;
•
an Extreme Longevity Index (ELI) was computed as the ratio between the
number of centenarians born between 1880 and 1901 and the total number
of births recorded during the same time interval. This index, expressed as the
number of centenarians per 10,000 newborns, is equivalent to the probability
for any person born in that municipality to reach 100 years of age, there or
anywhere else;
•
thereafter a Gaussian spatial smoothing method was used to outline the area
where ELI reached the maximum value.
The extreme longevity area identified in Sardinia and named as Blue Zone (BZ)
includes a group of 14 villages in Barbagia and Ogliastra, covering the highest
mountain area of Sardinia.
4
The total population of this BZ is currently about 40,000
inhabitants who are mainly engaged in pastoral and agricultural activities and follow
a relatively traditional lifestyle.
During the period 1880–1900, covered by the centenarians validation project
(AKEA II) 17,965 people were born within the BZ area, 91 persons (47 men and
44 women) among whom reached the age of 100. The resulting ELI values per
100,000 births are 512 and 506, for men and women respectively. For the rest of
Sardinia, ELI values are significantly lower, with 103 for men and 264 for women
(Poulain et al. 2004).
Starting from 2005, in cooperation with D. Buettner, a journalist writing for
National Geographic, the BZ concept was extended to other relevant regions of the
earth (Buettner 2012). Exceptional longevity at population level has been identified
and validated so far in three other settings: the island of Okinawa in Japan, the
peninsula of Nicoya in Costa Rica and the island of Ikaria in Greece (Figures 3).
Compared to the situation in the first BZ in Sardinia, we faced more difficulties to
3
The survival of centenarians born in the municipality and surviving outside was checked by
contacting the family or the administration of the current municipality of residence.
4The highest mountain of Sardinia is Punta La Marmora, 1856 m above sea level.
Michel Poulain, Anne Herm and Gianni Pes 93
validate the population longevity in these three BZs because of limited availability
of appropriate data sources. Indeed the situation is quite different between the three
settings.
In Okinawa, as elsewhere in Japan, the individual validation of age is based on
the koseki, a family register containing records of all members of a family, including
gender, dates and places of birth, names of parents, dates of marriage and divorce if
any, date and place of death (Saito 2010). Unfortunately, for privacy reasons, data
extracted from the koseki such as birth and death records are only accessible by
directly asking the relatives of the concerned persons or for official legal proceedings
(Willcox et al. 2008). Accordingly, individual age validation can only be performed
by a direct interview of centenarians alive and comparing self-reported information
during the interview with data found on their personal koseki documents. In 2005
Willcox et al. (2008) recruited 52 centenarians alive in Okinawa and reached the
conclusion that 49 of them were actually older than 100 years of age while the other
three were 2, 3 and 6 years younger than alleged. Censuses, statistics on age at death
and regional life tables are available for Okinawa ensuring an exhaustive coverage of
centenarians. Still, one must keep in mind that age in these data is often based on
koseki records. Various additional investigations developed by Willcox et al. (2008)
on aggregate statistics confirmed the overall reliability of the data. Even if some doubt
still exists on the large number of centenarians found in Okinawa as demonstrated by
the first author (Poulain 2011), the exceptional longevity in Okinawa could not be
put into question and Okinawa should be considered as a BZ.
In the Nicoya Peninsula, the identification and validation of age of the oldest olds
and the assessment of the population longevity were performed based on public
voting lists (padr´on electoral) disseminated by the Supreme Electoral Tribunal
(Tribunal Supremo de Elecciones) and produced from the birth registry. The later
includes data on births, naturalisations and deaths of all ever-living Costa Ricans
who contacted the civil registration system since its computerisation in 1970 (Rosero-
Bixby 2008). Each of them was given a unique identification number, either at birth
or at naturalisation, appearing on their identification card or c´edula. The age of the
oldest olds appearing on their c´edula was successfully confronted with self-reported
information given during the interview. The data quality and the higher level of
longevity in Nicoya in comparison to the rest of Costa Rica are confirmed by the
results presented by Rosero-Bixby in this yearbook (2013). These support the BZ
status of the Nicoya Peninsula.
In Ikaria, census data and age-at-death statistics obtained from Statistics Greece
were considered and confronted with individual data extracted from the dimotologio,
an administrative registry that includes demographic information on all Greek
citizens of a given municipality. Unfortunately, for the oldest olds on the island
no birth records could be found and therefore the extinct-cohort method based on
an exhaustive identification of those surviving above 90 as well as of persons being
part of the same birth cohorts who died during the two last decades was used to
estimate the level of population longevity. Individual age validation was successfully
achieved during an exhaustive base of interviews with all those aged 90 and above in
94 The Blue Zones: areas of exceptional longevity around the world
the northern part of the island by using a battery of questions on the occurrence of
demographic events and the age of close relatives (Poulain and Pes 2009).
3 First attempt to compare the BZs
Conventionally, the apparent exceptional longevity of a population is inferred from
the existence of an unusually large proportion of centenarians/nonagenarians. The
centenarian prevalence (CP), i.e. the ratio between the number of living centenarians
in a given population and the total resident population, is largely used in the literature
as well by the media. However, the reliability of this indicator for measuring and
comparing longevity between different populations deserves critical evaluation as
it is sensitive to a number of biases owing to migration and changes in fertility. For
example, in case of a population that experienced large-scale immigration flows
in the younger generations or a baby boom the CP will fail to identify remarkable
survival of persons in old ages as the proportion of elders is artificially lowered.
On the contrary, where the younger population has decreased due to emigration the
proportion of elderly may result artificially increased. In such cases the prevalence or
proportion of oldest olds is no longer reliable to measure longevity and to compare
it between populations. Nevertheless, CP is still the most frequently used indicator
by gerontologists as well as by national and regional authorities eager to claim a
longevity status for the area of the concerned population.
Compared to CP, life tables provide a better measurement of longevity and allow
comparing different populations more safely. The cohort life table is computed from
mortality rates of a given birth cohort observed during one century or longer while
the period life table is calculated on a fictive cohort by considering mortality rates
for each age group at a given time. In most populations longevity is rising so fast that
only the same cohorts of different populations should be compared and accordingly
cohort life tables are preferred. Nevertheless computing a cohort life table is not a
straightforward exercise as different methodologies and assumptions should be used
while ad-hoc data are not always available. As a result, very few countries produce
cohort life tables and no cohort life table exist up to now for any of the BZs.
A closely related indicator, the Centenarian Rate (CR), has been proposed by
demographers to compare longevity levels. It has been introduced by Robine and
Caselli (2005) as the ratio of the number of persons aged 100 years who were forty
years earlier 60 years old in the same population or, more precisely, living within
the same territory. This index is easy to calculate by using census data and is said to
“eliminate the impact of the size of the cohort, the role of migration, naturalization,
fertility and infant mortality” (Robine and Caselli 2005). However, there may still be
Michel Poulain, Anne Herm and Gianni Pes 95
Figures 3:
Geographical location of the Ogliastra, Ikaria, Costa Rica and Okinawa BZs
some bias when migration flows between age 60 and 100 are not negligible.
5
The
CR, as it was initially proposed, measures the average longevity of the members
of a specific cohort by counting how many individuals who were 60 years old
in the census forty years before the time of measurement and later reached age
5
This is a situation existing in some areas regarded as highly favourable for retirement, such as
Florida in the US. In countries where retirement age coincides with a higher level of migration, it may
be better to compare the survival between 75 and 100.
96 The Blue Zones: areas of exceptional longevity around the world
Table 1:
Centenarian Rate (CR) for the 1900 birth cohort (number of persons surviving to 100
years for 10,000 persons alive at 60 years)
Females aged Females aged Female Males aged Males aged Male
Country 60 in 1960 100 in 2000 CR 60 in 1960 100 in 2000 CR
Italy 248 571 1 941 78.1 212 849 412 19.4
of which Sardinia 5 889 64 108.7 5 594 25 44.7
Japan 310 764 3 948 127.1 305 703 900 29.5
of which Okinawa 2 925 145 495.7 2 069 17 82.2
France 270 842 2 965 109.5 239 348 450 18.8
The Netherlands 52 213 388 74.4 47 247 83 17.6
Norway 19 652 146 74.3 17 798 27 15.2
Sweden 41 928 315 75.1 39 262 72 18.3
Switzerland 30 877 271 87.8 25 808 59 22.9
Australia 42 812 524 122.5 40 212 84 21.0
Belgium 57 847 300 51.8 50 438 51 10.1
Canada 60 070 1 080 179.8 61 456 244 39.7
Denmark 24 465 160 65.4 22 076 31 14.0
Finland 21 965 90 41.0 16 671 28 16.8
Poland 122 695 336 27.4 97 732 76 7.8
Russia 470 626 1 605 34.1 235 003 201 8.5
UK 318 869 2 438 76.5 270 280 288 10.6
USA 780 115 12 706 162.9 728 821 2 253 30.9
Source: Human Mortality Database (stock data) and national census data available through national statistical
institutes for Sardinia and Okinawa.
100. Even if the specificity of the selected cohort may limit the comparative power
of the CR, its advantage is that the data required for its computation are easily
available from population censuses as well as in the Human Mortality database. Still,
the interpretation of CR values relies on the assumption that the net international
migration between ages 60 and 100 years can be neglected for the considered cohort.
Using the data from the Human Mortality database for observed countries, and
data provided by the national data sources for Sardinia and Okinawa, we calculated
the CR for the cohort born in 1900. For that specific cohort the observed number of
persons aged 100 on 31 December 2000 was compared with the observed number of
persons aged 60 in the same population on 31 December 1960, that is forty years
earlier. The figures presented in Table 1 support the idea that males in Okinawa
and Sardinia have a higher longevity compared to their peers in other countries.
Extremely high values were observed for women in Okinawa, being three times
higher than for women in all of Japan.
Although the CR can be used with some limitations to compare longevity at
national level, it is less appropriate for smaller populations because of the limited
number of persons reaching age 100 within a cohort corresponding to a single year
Michel Poulain, Anne Herm and Gianni Pes 97
of birth (among the BZs, only Okinawa has a sufficient number of centenarians
to compute the CR). Accordingly, it is preferable to consider the between-census
survival of broader population groups by comparing, for example, the number of
persons aged 60–69 years in a 1970 census with those recorded thirty years later in a
2000 census and aged 90–99 or, alternatively, comparing the number of 60–79 years
olds in 1970 and the aged 90+thirty years later. In Table 2 these alternative CR
are presented for the four longevity BZs as well as a few other countries. When
interpreting these figures the possible bias resulting from migration should be kept
in mind. In cases where the net migration cannot be neglected, these figures are
automatically over-estimated if the population experienced a positive net migration
during the 30 years’ period before observation, or under-estimated in case of negative
net migration. The figures in Table 2 display the superiority of Okinawa for women
and of Sardinia for men. As for Ikaria, owing to high rates of emigration of old
persons to the Athens Metropolitan Area, their figures are under-estimated. By
contrast, for Costa Rica and Nicoya, important immigration from abroad and age
exaggeration among nonagenarians in the 2000 census (Roxero-Bixby 2008) cause
an inflation of CR.
The table in Appendix 1 shows summary ecological and socio-economic variables,
selected from the existing literature, for the four BZs known so far, taking into
account their possible role as longevity determinants. Most geographical features,
though somewhat variable across the four sites, point predominantly to a condition
of insularity and remoteness. The site altitude ranges from sea level to mid-mountain,
with the highest value being found in the Sardinia BZ and the lowest in the Okinawa
BZ. Average steepness of land, which correlates with altitude, reaches a maximum in
the two Mediterranean islands and a minimum in Okinawa. The climate varies greatly
across the four sites with a prevalence of relatively warm temperatures. A noteworthy
observation, even when taking into account the large geographic heterogeneity of
the four sites, is that high levels of sunlight, wind speed and humidity seem to be
prevalent. Some socio-economic factors observed in the BZs, especially the lower
degree of industrialisation and income per capita, seem to indicate the presence of
populations emerging from a long-standing state of economic vulnerability although
they may have experienced a constant improvement of their well-being level during
the last few decades. Besides, some lifestyle and health indicators such as the
moderate average calorie intake and the low prevalence of overweight/obesity in the
adult population indicate a strong persistence of traditional habits, along with a low
susceptibility to adopt a more Western lifestyle.
4 Discussion
The concept of BZ aims at characterising an area where a population sharing a
common lifestyle and environment displays exceptionally high longevity. The first
question to discuss is the very existence of such areas and whether the methods
used to demonstrate their high level of longevity and compare longevity between
98 The Blue Zones: areas of exceptional longevity around the world
Table 2:
Probability to survive between 1970 and 2000 for the 1890–1919 birth cohorts
Survival rates (%)
Persons in 1970 Persons in 2000 1970–2000(1)
Age 60–69 Age 60–79 Age 90–99 Age 90+90–99/60–69 90+/60–79
Males
Italy 2 311 202 3 384 724 84 276 85 131 3.1 2.0
Sardinia 58 178 87 965 3369 3 422 5.8 3.9
Villagrande 172 259 23 23 13.4∗∗∗ 8.9∗∗
Japan 3 098 290 4 544 841 159 302 161 297 5.1 3.5
Okinawa 19 685 30 592 1978 2 031 10.0∗∗∗ 6.6∗∗∗
Costa Rica 27 605 39 590 3 050 3 154 11.0 8.0
Nicoya 1700 2 347 211 221 12.4 9.4∗
Greece 379 960 561 124 8 846 9 386 2.3 1.7
Ikaria 604 800 20 20 3.3 2.5
France 2 259 802 3 352 607 90 236 91 159 4.0 2.7
USA 7 117 859 11 011 845 312 843 318 473 4.4 2.9
Females
Italy 2 697 844 4 292 653 257 827 262 360 9.6 6.1
Sardinia 61 377 98 411 6 446 6 580 10.5 6.7
Villagrande 166 269 22 23 13.3 8.6
Japan 3 467 619 4 514 236 464 402 473 879 13.4 10.5
Okinawa 28 258 46 237 7 226 7 544 25.6∗∗∗ 16.3∗∗∗
Costa Rica 27 456 39 930 4 150 4 282 15.1 10.7
Nicoya 1 518 2 179 225 245 14.8 11.2
Greece 417 664 657 288 26 203 27 379 6.3 4.2
Ikaria 532 856 47 49 8.8∗5.7∗
France 2 720 096 4 599 804 318 095 325 260 11.7 7.1
USA 8 407 784 13 815 593 1 025 834 1 060 367 12.2 7.7
Source: Data source: Human Mortality Database (stock data) and national census data available through national
statistical institutes for Sardinia and Okinawa.
(1) The survival rates for Villagrande, Okinawa, Nicoya and Ikaria have been compared with the corresponding
ones for Sardinia, Japan, Costa Rica and Greece, respectively. The statistical Z-test for two proportions shows
significant differences at 5% (∗), 1% (∗∗) or 0.1% (∗∗∗ ).
different populations are adequate. The second question is whether some ecological
characteristics of the four BZs could be considered as potential determinants of
the exceptional longevity of such populations. However, before discussing these
two questions it is worth reminding that the age of a person is measured by the
number of chronological years from birth. Therefore, validating accurately the
age of every centenarian and ensuring the exhaustiveness of collected data is an
absolute prerequisite for assessing and comparing the levels of population longevity.
Inaccuracy in extreme ages has been often reported in the past and is still commonly
Michel Poulain, Anne Herm and Gianni Pes 99
found today in populations lacking civil registration (Kannisto 1988; Wilmoth and
Lundstrom 1996; Poulain 2010). By the available evidence, data sources differ
considerably between the four BZs, with the most favourable situation in Sardinia
and the weakest one in Ikaria. Nevertheless, based on research done by other scientists
as well as on our own investigations we are confident that the high level of longevity
of these populations is real even if more in-depth validation could still be helpful to
confirm our findings.
Among several indicators, which one should be considered ideal to compare
longevity between different populations? We would briefly like to summarise the
pros and cons of two indicators, i.e. CR and ELI.
•
CR estimates the survival of cohort(s) between two different censuses. It can
be computed by using census data or the Human Mortality Database under the
hypothesis of the net international migration being negligible. Unfortunately,
CR cannot be used to compare longevity at local level as cohorts may be too
small and in a large number of cases no centenarians are found at the second
census in many municipalities. Moreover, the assumption of negligible net
migration, while possibly acceptable at national level, is no longer valid at
local level where a significant number of old people may move away to another
municipality, either to a nursing home or the residence of one of their children.
Thus, CR cannot be used to compare the longevity between sizable long-living
populations.
•
ELI was used at local level in Sardinia because it was possible to match every
birth with the corresponding death by using the marginal annotations in the
birth register. This information allows identifying every death at age 100 and
over, regardless of where it has occurred. Moreover, to overcome the problem of
small number of newborns the data of all Sardinian cohorts from 1881 to 1901
included have been pooled. Unfortunately ELI cannot be computed in the three
other BZs as birth registers are either lacking (Ikaria) or inaccessible (Okinawa)
or no marginal annotations on death exist for every newborn (Nicoya).
Having in mind these limitations the data gathered in Table 2 confirm the exceptional
longevity in Okinawa for both men and women compared to all of Japan while similar
levels of longevity are also found in the Sardinian BZ compared to Sardinia as a
whole, but for men only. Also women in Ikaria are shown to live longer than Greek
women in general, which cannot be confirmed for men. Due to migration biases
it has not been possible to prove any exceptional level of longevity in the Nicoya
BZ and more details on the longevity registered in that population are proposed by
Roxero-Bixby (2013) in this volume.
As far as the second point of the discussion is concerned, the comparison of
the four BZs relies on the underlying assumption that, despite the large number of
potential causes involved in the increased survival at individual level, a more limited
subset of ‘shared longevity determinants’ may exist at population level. The utility of
the BZ concept can be easily appreciated as these populations exhibit an unusual level
of homogeneity in both environment and lifestyle. Moreover the comparison of the
100 The Blue Zones: areas of exceptional longevity around the world
four BZs favours replication studies as well as testing of hypotheses across various
ethnic and cultural settings. The comparison proposed here for the first time is rather
limited, qualitative in essence, and takes into consideration only a few ecological
and socio-cultural variables as potential predictors of longevity, selected on the basis
of the available evidence and sparse literature data. Nevertheless some interesting
features could emerge that will be the starting point for additional investigations.
Among the ecological variables, geographical features such as altitude and
terrain steepness are particularly intriguing. Most populations who enjoy high life
expectancy have been reported to live in mountains. Although many of them turned
out to be myths (Leaf 1981) it has been suggested that living at high altitude could
reduce the risk of mortality especially from cardiovascular disease. While this could
be true for truly high elevations (above 1500 meters), it is rather questionable for
lower altitudes. Besides, a recent study across US counties reported that after
adjustment for confounding variables the association between altitude and life
expectancy is no longer significant (Ezzati et al. 2012). Nevertheless, it is possible
that people living even at modest elevation might benefit from positive effects on their
health that could be partly explained by the relative isolation of these areas. As matter
of fact, life in the mountains or on a remote island is associated with a set of tightly
linked variables affecting both individual behaviour (diet, physical activity) and, more
generally, the social context (habitat, economic activity, community support) and the
environment (degree of pollution, quality of drinking water). All these factors are
regarded as having a positive influence on health. Mountain environment is associated
with increased land steepness which entails a constant stimulus for outdoor physical
activity even by subjects in advanced age. Recently we have suggested that the
elevated average slope of the terrain, fairly common in mountainous areas, may
imply greater energy expenditure during active life, thus resulting in improved
cardiorespiratory fitness, and ultimately in better survival (Pes et al. 2011).
In the past, populations living in mountains were likely to suffer from a degree
of isolation that led to delayed economic growth compared to the whole country.
During the early stages of industrialisation in Western countries, mountain areas
were often regarded as backward pockets where traditional agro-pastoral activities
persisted despite the rise of industrialism. However, this had important individual
and social consequences, including a better preservation of the traditional habitat
and a lower probability of coming into contact with health-threatening pollutants,
not to mention a more satisfying and less alienated individual and occupational life.
This, perhaps unintentionally, allowed a relevant proportion of the population to
maintain the traditional way of life centred on tight social relationships, meaningful
emotional exchange between generations and a considerable degree of support for
weaker individuals by the community as a whole that in the long run could have been
beneficial for survival.
The observation that all the BZs identified so far are slightly economically
underdeveloped compared to the rest of their respective country, as reflected by
their lower per capita income, seems to defy the so-called social gradient theory
postulated by Evans (1994) stating that longevity would increase in proportion with
Michel Poulain, Anne Herm and Gianni Pes 101
the economic prosperity owing to its advantages in reducing anxiety and stress. When
looking at the regional level, as highlighted by Cockerham (2000) in the case of
Okinawa, areas ranking higher on levels of health are those where the process of
modernisation came later and more slowly, as in the case of the three other BZs. This
apparent discrepancy linking low income with longevity can be better understood
if the economic development of these areas is evaluated in relative rather than
absolute terms. When around the middle of the last century areas like Sardinia and
Ikaria started their economic development, largely assisted by their governments, this
resulted mainly in shifting from extreme poverty and deprivation to a relative material
well-being, with limited social stratification. As pointed out by Poland (1998), the
absence of a true class gradient and a consequent absence of social competition and
individual stress, hallmarks of the Sardinian and Greek BZs, might have created
significantly more favourable conditions for individual health than those operating in
the more competitive mainland framework, a hypothesis that should be checked also
for Okinawa and Nicoya.
Other life-shortening factors listed in Appendix A.1, such as smoking habits, the
obesity rate and suicide prevalence, display a wide variation across the various BZs
and their value is surprisingly high in some of them. This, however, does not provide
evidence of their lack of influence on survival if we take into account two particular
aspects: (i) the variable values might refer to the current population and do not
necessarily reflect the historical situation of previous generations; (ii) the influence
of potentially negative factors might be counteracted by concomitant factors acting
in the opposite direction. The latter seems to be the cause of the paradox reported
in some long-lived populations whose lifestyle appears to be unhealthy (Rajpathak
2011) yet it is buffered by coexisting protective factors. In this regard some dietary
factors also deserve attention. The average caloric intake of the BZ’s populations,
with the possible exception of Okinawa, is not distinct from the general population
of their mainland countries and does not fit the criteria for calorie restriction with
optimum nutrition considered up to now the only longevity-promoting diet (Fontana
2010). Of course it is theoretically possible that not the quantity itself but rather the
quality of food might have played an important role in maintaining a high standard
of health in these populations, an aspect that will deserve further research.
In conclusion, the BZs, representing the extreme manifestation of population
ageing operating in many contemporary populations as well, should be potentially
regarded as a promising theoretical model, perhaps as important as the model of
centenarians in the field of individual longevity. In fact, these rare populations to
some extent might have retained significant longevity-related cause–effect links that
in other post-industrial populations are masked by the effect of greater heterogeneity
in genetic, socio-economic and lifestyle aspects.
102 The Blue Zones: areas of exceptional longevity around the world
5 Future scientific research and policy implications
The search for common longevity determinants in the different BZs is just beginning.
Quantitative and qualitative surveys are currently under development in a comparative
way for Sardinia and Ikaria, involving biomedical and behavioural aspects. A
comprehensive approach is being considered, favouring the analysis of individual
life as well as an anthropological viewpoint in order to understand how the various
transitions experienced by the local community (fertility, epidemiology, education,
communication, nutrition
...
) could have interacted to result in exceptional longevity.
Additionally, blood samples are being collected to carry out some genetic and
epigenetic investigations. In particular the study of DNA methylation is expected to
be more useful in populations historically or geographically isolated such as the BZs,
which were exposed to high levels of endogamy and have therefore progressively
developed a significant reduction in their genetic diversity.
As far as policy implications are concerned, Appel (2008) expressed the thought
that “Blue Zones, now limited to just a few populations in the world, can become
commonplace”. The question is how the lessons obtained from the BZs can be
transposed for improving the healthy ageing of our post-industrial societies. This
led Dan Buettner to initiate the Blue Zones Community Project, aimed at creating
a programme at community level to help communities harness schools, businesses,
families and governments into improving their residents’ health and well-being.
The Blue Zones Project initiative to become a Blue Zones Community is a systems
approach that allows citizens, schools, employers, restaurants, grocery stores and
community leaders to work together on policies and programmes that will make the
most impact and move the community towards optimal health and well-being. Blue
Zones Project initiatives have been launched in several places in the US: Albert Lea
in Minnesota, three beach cities near Los Angeles in California and, more recently,
ten communities in Iowa. Specific lifestyle and environmental characteristics in each
of the BZs are transposed to help American cities joining the project to optimise their
own communities for better fitness and longevity.
For all Iowa communities, tools will be available to help them continue their path
to improved wellbeing. The Blue Zones Project provides leadership training and
guidance to help leaders transform their community, work and home environments.
Can good-health practices of the world’s longest-lived communities be transplanted
into small US Midwestern cities? This is the real challenge. First results show how
“the Blue Zones impact in how people eat and exercise, how they find new ways
to build community and get to know each other, how they are cutting back areas
for smoking and expanding opportunities for altruism—all key elements in creating
a Blue Zones community” (Wolfe 2012). Nevertheless it is too early to assess the
improvement in healthy ageing that could be associated with the implementation of
the Blue Zone Community Project. Scientists should be attentive in the forthcoming
years to develop accurate indicators that could help assessing the efficiency of such
policy support initiatives.
Michel Poulain, Anne Herm and Gianni Pes 103
Acknowledgements
Support from the Estonian Ministry of Education and Science (SF1300018s11) and
the Estonian Science Foundation grant No. 8325 for Michel Poulain and Anne Herm
is gratefully acknowledged.
Appendix
Table A.1:
Ecological and socio-economic variables in the four BZ sites
Okinawa BZ Sardinia BZ Ikaria BZ Nicoya BZ
Geographic features
Area (km2) 1 201a1 559b255c778d
Population 1 384 762a42 113b8 312c326 953d
Density (inhab./sq km) 1 015 27 31 67
Average latitude 26◦N39
◦N37
◦N10
◦N
Site altitude Sea level Mid-mountain Mid-mountain Hills
Average land steepness Low High High Medium
Proximity to the sea Yes Yes Yes Yes
Climate
Climate Subtropical Mediterranean Mediterranean Tropical
Average annual temp and range (◦C) 22.4 (16–27)e16.9 (16–18)f18.9 (16–22)g26.4 (23–28)h
Average annual rainfalls (mm) ∼2 000e∼800 i631j2 178h
Relative humidity (%) 71e65i66g81h
Land characteristics
Main geologic features Coralline Granite and Granite rocks Sedimentary
limestone, basalt rocks rocks
volcanic rocks (limestone)
Background radioactivity Low High High Low
Drinking water hardness HighkLowlLowmHighn
Forest coverage (%) 46o83.1p>80 j3q
Socio-economic indexes
Local income (LI) per capita ∼$21 000r$19 872s$26 235t$8 700u
Agriculture (% of LI) 1.9 1.4 12.2 18.2
Manufacturing (% of LI) 4.7 16.8 9.7 7.7
Services (% of LI) 89.5 81.8 78.1 64
Vehicles per 1,000 inhabitants 490r360b200t177u
Lifestyle and health indicators
Daily food intake per capita (kcal) <2000k2 600v<1500w2 392x
Smoking rate (%) 4.6y∼20z82 (men)w40.3x
Obesity rate (%, BMI above 30) 10–40y8.7aa 12.5w23.6bb
Suicide rate (cases/100,000) 32.1/33.5 (men)cc 15.3dd 2.35ee 7.3ff
9.1/14.6 (women)
104 The Blue Zones: areas of exceptional longevity around the world
Sources of data presented in Appendix 1
a. http://en.wikipedia.org/wiki/Okinawa Island
b.
Poulain, M., G.M. Pes, C. Grasland, C. Carru, L. Ferrucci, G. Baggio, C.
Franceschi, L. Deiana 2004. Identification of a geographic area characterised
by extreme longevity in the Sardinia island: the AKEA study. Experimental
Gerontology 39 (9): 1423–1429.
c. http://en.wikipedia.org/wiki/Icaria
d. Relative to the entire province of Guanacaste;
e.
Ikema, T., G.R. Bigg, and R.G. Bryant (2010). Increasing rain intensity
over Okinawa, 1982–2005, and the link to changes in characteristics of
northwest Pacific typhoons. Journal of Geophysical Research, 115 (D24),
doi: 10.1029/2010JD014428.
f. ISTAT (http://www3.istat.it/)
g.
Karavitis, C.A. et al. 2012. “Assessing alternative water resources management
scenarios in islands of the Aegean archipelago, Greece”. Global NEST Journal,
14(3), 264–275.
h.
Campos, O., E. Rodriguez and L. Ugalde 1992. Desarrollo Agropecuario
Sostenible en la Region de Hojancha, Guanacaste, Costa Rica. Rep. No. 195,
CATIE/ROCAP, Turrialba, Costa Rica.
i. Chessa, P. and A. Delitala 1997. Il Clima della Sardegna.SAR, Sassari, Italy.
j.
Gouvas, M., and N. Sakellariou 2011. Climate and forest vegetation in Greece.
National Observatory of Athens, Institute for Environmental Research and
Sustainable Development. Library, Study No. 01/2011.
k.
Willcox B.J., D.C. Willcox, H. Todoriki, A. Fujiyoshi, K. Yano, Q. He, J.D.
Curb and M. Suzuki 2007. Caloric restriction, the traditional Okinawan diet,
and healthy ageing. The diet of the world’s longest-lived people and its potential
impact on morbidity and life span. Annals of the New York Academy of Sciences,
1114: 434–455.
l.
Angelillo, B. and U. Alamanni 1964. Sull’andamento nella provincia di Sassari
della mortalit
`
a per malattie cardiocircolatorie in rapporto anche ad alcuni fattori
alimentari. Igiene e Sanit`a Pubblica, 20: 429–435.
m.
Lambrakis, N. and G. Kallergis 2005. Contribution to the study of Greek
thermal springs: hydrogeological and hydrochemical characteristics and origin
of thermal waters. Hydrogeology Journal, 13: 506–521.
n. Mora-Alvarado, D., N. Alfaro-Herrera, C. Felipe-Portuquez and M. Peinador-
Brolatto 2000. C
´
alculos en las v
´
ıas urinarias y su relaci
´
on con el consumo de
calcio en el agua de bebida en Costa Rica. Revista Costarricense de Salud
P´ublica 9(17).
o.
Li, Q., M. Kobayashi and T. Kawada 2008. Relationships Between Percentage
of Forest Coverage and Standardised Mortality Ratios (SMR) of Cancers in all
Prefectures in Japan. The Open Public Health Journal 1: 1–7.
p.
Piano Faunistico Venatorio Provinciale 2011–2015. Provincia dell’Ogliastra,
Assessorato all’ambiente.
Michel Poulain, Anne Herm and Gianni Pes 105
q.
Sanchez-Azofeifa, G.A. 2001. Deforestation in Costa Rica: A Quantitative
Analysis Using Remote Sensing Imagery. Biotropica 33: 378–384
r.
Lopez-Rodriguez, J., and D. Nakamura 2011. Mind the Remoteness! Income
disparities across Japanese Prefectures. Estudios de Economia 38: 393–417.
s. ISTAT, Distribuzione del reddito e condizioni di vita in Italia (2005–2006).
t.
Eurostat 2008. Regional gross domestic product (PPS per inhabitant), by NUTS
2 regions.
u.
Encuesta de Hogares (2001–2008) del Instituto Nacional de Estad
´
ısticas y
Censos (INEC).
v.
Tivaroni, J. 1928. “Nutrition patterns in Sardinia population”. Rivista di Politica
Economica, 2:1–11.
w.
Panagiotakos, D.B., C. Chrysohoou, G. Siasos, K. Zisimos, J. Skoumas, C.
Pitsavos and C. Stefanadis 2010. Sociodemographic and Lifestyle Statistics
of Oldest Old People (
>
80 Years) Living in Ikaria Island: The Ikaria
Study. Cardiology Research and Practice, 2011 ID 679187, 7 pages, doi:
10.4061/2011/679187.
x.
Kabagambe, E.K., A. Baylin, X. Siles and H. Campos 2002. Comparison of
dietary intakes of micro- and macronutrients in rural, suburban and urban
populations in Costa Rica. Public Health Nutrition, 5(6A): 835–42.
y.
Willcox, D.C., B.J. Willcox, S. Yasura, M.A. Ashitomi and M. Suzuki 2012.
Gender gap in healthspan and life expectancy in Okinawa: health behaviours.
Asian journal of gerontology &geriatrics 7: 49–58.
z. Indagine DOXA, Istituto Superiore di Sanit`
a (ISS), 2011.
aa.
Piras, I., M.E. Ghiani, A. Melis, S. Inghes, L. Salaris, A. Gatti, A. Loviselli,
G. Vona and C.M. Cal
`
o 2006. Incidenza del sovrappeso e dell’obesit
`
a tra i
coscritti sardi nati nel 1980. Antropo, 13: 35–43. www.didac.ehu.es/antropo
bb.
Chen-Ku, C. H. and M.G. Sand
´
ı-Bogantes 2004. Actualizaci
´
on en obesidad.
Acta M´edica Costarricense 46, suppl. 1.
cc.
Chen, J., Y. J. Choi, and Y. Sawada 2009. “How is Suicide Different in Japan?”
Japan and the World Economy, 21(2): 140–50.
dd.
Maggian, R. 2011. Guida al welfare italiano: dalla pianificazione sociale alla
gestione dei servizi. Maggioli ed.
ee.
Giotakos, O., G. Tsouvelas and V. Kontaxakis 2012. Suicide rates and mental
health services in Greece. Psychiatrik´e 23(1): 29–38.
ff.
Granados-Bloise, D. 1998. Magnitud y Tendencia del Suicidio en los Cantones
de Costa Rica: 1980–1994. Revista Costarricense de Salud P´ublica 7: 23–35.
References
Appel, L. J. 2008. “Dietary Patterns and Longevity Expanding the Blue Zones”.
Circulation 118 (3): 214–215. doi:10.1161/CIRCULATIONAHA.108.788497.
106 The Blue Zones: areas of exceptional longevity around the world
Bowerman, W. G. 1939. “Centenarians”. Transactions of the Actuarial Society of
America 40: 361–378.
Buettner, D. 2010. The Blue Zones: Lessons for Living Longer From the People
Who’ve Lived the Longest. Washington, D.C.: National Geographic.
Cockerham, W. C., H. Hattori, and Y. Yamori. 2000. “The Social Gradient in Life
Expectancy: The Contrary Case of Okinawa in Japan”. Social Science &Medicine
51 (1): 115–122.
Deiana, L., L. Ferrucci, G. M. Pes, C. Carru, G. Delitala, A. Ganau, S. Mariotti, A.
Nieddu, S. Pettinato, P. Putzu, C. Franceschi, and G. Baggio. 1999. “AKEntAnnos
The Sardinia Study of Extreme Longevity”. Aging, Clinical and Experimental
Research 11 (3): 142–149.
Evans, R. G. 1994. “Why Are Some People Healthy and Others Not? Introduction.”
In The Determinants of the Health of Populations, ed. by R. G. Evans, M. L. Barer,
and T. R. Marmor, 3–26. New York: Aldine de Gruyter.
Ezzati, M., M. E. M. Horwitz, D. S. K. Thomas, A. B. Friedman, R. Roach, T. Clark,
C. J. L. Murray, and B. Honigman. 2012. “Altitude, Life Expectancy and Mortality
From Ischaemic Heart Disease, Stroke, COPD and Cancers: National Population-
Based Analysis of US Counties”. Journal of Epidemiology and Community Health
66 (7): e17. doi:10.1136/jech.2010.112938.
Fontana, L., L. Partridge, and V. D. Longo. 2010. “Extending Healthy Life Span
From Yeast to Humans”. Science 328 (5976): 321–326. doi:
10.1126/science.
1172539.
Jeune, B., and J. W. Vaupel. 1999. “Species of Evidence of Exceptional Longevity”.
In Validation of Exceptional Longevity, ed. by B. Jeune and J. W. Vaupel, 11–22.
Monographs on Population Aging 6. Odense: Odense University Press.
Kannisto, V. 1988. “On the Survival of Centenarians and the Span of Life”.
Population Studies 42 (3): 389–406. doi:10.1080/0032472031000143546.
Koenig, R. 2001. “Sardinia’s Mysterious Male Methuselahs”. Science 291 (5511):
2074–2076. doi:10.1126/science.291.5511.2074.
Leaf, A. 1973. “Every Day Is a Gift When You Are Over 100”. National Geographic
143 (1): 93–118.
Leaf, A. 1981. “Statement Regarding the Purported Longevous Peoples of
Vilcabamba”. In Controversial Issues in Gerontology, ed. by H. Hershow, 25–26.
New York: Springer Publishing Company, Inc.
Mazess, R., and S. Forman. 1979. “Longevity and Age Exaggeration in Vilcabamba,
Ecuador”. Journal of Gerontology 34 (1): 94–98.
Perls, T. T., L. M. Kunkel, and A. A. Puca. 2002. “The Genetics of Exceptional
Human Longevity”. Journal of the American Geriatrics Society 50 (2): 359–368.
Michel Poulain, Anne Herm and Gianni Pes 107
Pes, G. M., F. Tolu, M. Poulain, A. Errigo, S. Masala, A. Pietrobelli, N. C. Battistini,
and M. Maioli. 2013. “Lifestyle and Nutrition Related to Male Longevity in
Sardinia: An Ecological Study”. Nutrition, Metabolism and Cardiovascular
Diseases 23 (3): 212–219. doi:10.1016/j.numecd.2011.05.004.
Poland, B., D. Coburn, A. Robertson, and J. Eakin. 1998. “Wealth, Equity and
Health Care: A Critique of a Population Health Perspective on the Determinants
of Health”. Social Science &Medicine 46 (7): 785–798.
Poulain, M. 2010. “On the Age Validation of Supercentenarians”. In
Supercentenarians, ed. by H. Maier, J. Gampe, B. Jeune, J.
-
M. Robine, and
J. W. Vaupel, 3–30. Demographic Research Monographs 7. Springer Berlin
Heidelberg.
Poulain, M. 2011. “Exceptional Longevity in Okinawa: A Plea for In-Depth
Validation”. Demographic Research 25 (7): 245–284. doi:
10 . 4054 / DemRes .
2011.25.7.
Poulain, M., and G. M. Pes. 2009. Report on the Validation of the Exceptional
Longevity in Ikaria. Unpublished Internal Report. National Geographic.
Poulain, M., G. M. Pes, C. Carru, L. Ferrucci, G. Baggio, C. Franceschi, and
L. Deiana. 2006. “The Validation of Exceptional Male Longevity in Sardinia”. In
Human Longevity, Individual Life Duration, and the Growth of the Oldest-Old
Population, ed. by J.
-
M. Robine, E. M. Crimmins, S. Horiuchi, and Z. Yi, 147–166.
New York: Springer /Kluwer.
Poulain, M., G. M. Pes, C. Grasland, C. Carru, L. Ferrucci, G. Baggio, C. Franceschi,
and L. Deiana. 2004. “Identification of a Geographic Area Characterized by
Extreme Longevity in the Sardinia Island: The AKEA Study”. Experimental
Gerontology 39 (9): 1423–1429. doi:10.1016/j.exger.2004.06.016.
Rajpathak, S. N., Y. Liu, O. Ben-David, S. Reddy, G. Atzmon, J. Crandall, and
N. Barzilai. 2011. “Lifestyle Factors of People With Exceptional Longevity”.
Journal of the American Geriatrics Society 59 (8): 1509–1512. doi:
10.1111/j.
1532-5415.2011.03498.x.
Robine, J.
-
M., and G. Caselli. 2005. “An Unprecedented Increase in the Number of
Centenarians”. Genus 61 (1): 57–82.
Rosero-Bixby, L. 2008. “The Exceptionally High Life Expectancy of Costa Rican
Nonagenarians”. Demography 45 (3): 673–691.
Rosero-Bixby, L., W. H. Dow, and D. H. Rehkopf. 2013. “The Nicoya Region of
Costa Rica: A High Longevity Island for Elderly Males”. This volume, Vienna
Yearbook of Population Research 11: 109–136.
Saito, Y. 2010. “Supercentenarians in Japan”. In Demographic Research Monographs,
ed. by H. Maier, J. Gampe, J. M. Robine, and J. W. Vaupel, 75–99. Rostock: Max
Planck Institute.
108 The Blue Zones: areas of exceptional longevity around the world
Willcox, D. C., B. J. Willcox, Q. He, N.
-
c. Wang, and M. Suzuki. 2008. “They Really
Are That Old: A Validation Study of Centenarian Prevalence in Okinawa”. The
Journals of Gerontology. Series A: Biological Sciences and Medical Sciences 63
(4): 338–349.
Wilmoth, J. R., and H. Lundstr
¨
om. 1996. “Extreme Longevity in Five Countries:
Presentation of Trends With Special Attention to Issues of Data Quality”. European
Journal of Population 12: 63–93. doi:10.1007/BF01797166.
Wolfe, W. 2012. “Blue Zone Offers Lessons in Longevity”. Star Tribune.
Young, R. D., B. Desjardins, K. McLaughlin, M. Poulain, and T. T. Perls. 2010.
“Typologies of Extreme Longevity Myths”. Current Gerontology and Geriatrics
Research 2010 (423087). doi:10.1155/2010/423087.
