See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/250005141
A growth study of Japanese descendants from
São Paulo, Brazil
Article in Anthropological Science · August 2005
DOI: 10.1537/ase.040702
CITATION
READS
1
25
2 authors:
Lucio Vinicius
Marta Mirazon Lahr
40 PUBLICATIONS 484 CITATIONS
125 PUBLICATIONS 5,086 CITATIONS
University College London
SEE PROFILE
University of Cambridge
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
IN-AFRICA View project
All content following this page was uploaded by Marta Mirazon Lahr on 16 November 2014.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
ANTHROPOLOGICAL SCIENCE
Vol. 113, 163–167, 2005
A growth study of Japanese descendants from São Paulo, Brazil
LUCIO VINICIUS1*, MARTA M. LAHR1
1
Leverhulme Centre for Human Evolutionary Studies, Department of Biological Anthropology, University of Cambridge,
Downing Street, Cambridge CB2 3DZ, UK
Received 3 July 2004; accepted 2 October 2004
Abstract We describe the growth pattern of a sample of Japanese descendants born in São Paulo,
Brazil. Cross-sectional data from 1297 subjects were obtained in eight middle- and upper middle-class
schools. Weight and height of children of Japanese ancestry were compared both to populations of similar socio-economic status, namely the well-off European descendants born in São Paulo and the
United States, and to populations of similar genetic background from Japan and urban China. Adult
size is smaller in Brazilians of Japanese ancestry than in the samples of European ancestry. We show
that the Brazilian-born Japanese descendants do not display any growth deficits during the preadolescent period but fall short of the US growth standard afterwards, suggesting that the adolescent growth
spurt is responsible for their smaller body size at adulthood. The observed height deficit is more pronounced than the weight deficit in both boys and girls. On the other hand, when comparisons include
only populations of Asian origin, the Japanese descendants from São Paulo are shown to be similar in
height and weight to the better-off children living in Japan, and significantly larger than the urban Chinese children of relatively lower socio-economic status.
Key words: growth, Japanese descendants, anthropometric references, Brazil
immigration from Europe, mainly driven by the expansion
of coffee plantations during the second half of the 19th century. Between 1873 and 1903, approximately 2,000,000
Europeans migrated to Brazil, especially from Italy (almost
60% of the total), Portugal, Spain, and Germany (Levy,
1974). During the first half of the 20th century, Japanese
migrants (mainly peasants with very little means) added to
the huge foreign contingent totaling 250,000 by 1950 and
giving rise to the one of the largest (if not the largest) Japanese communities outside Japan (Yamashita, 2001).
As the process of industrialization and urbanization intensified in the first decades of the 20th century, many Japanese
immigrants and descendants moved from rural environments to the city of São Paulo. The Japanese community of
São Paulo had remained relatively closed but today it is very
much integrated with the rest of the population. Japanese
descendants in São Paulo make up an important proportion
of the wealthier private school sector, as well as of the city’s
university graduates. Therefore, Japanese descendants differ
in terms of geographic origin from European descendants of
similar socio-economic status growing up in the same city
neighborhoods and attending the same wealthy schools, and
also contrast with Brazilians in general due to their successful socio-economic history.
This study describes growth in weight and height of Japanese descendants born in São Paulo, and compares them to
children and adolescents of European background from the
same city, school, and socio-economic status. Growth of the
Japanese sample is also interpreted relative to better-off populations from the USA and to other anthropometric studies
undertaken in Japan and China.
Introduction
Since the rebirth of the debate on human origins in the late
1980s (Stringer, 1990; Lahr, 1996), evolutionary anthropology has again drawn attention to population dynamics and
geographical differentiation. The sustained effort of paleoanthropology revealed that metric variation between
human groups is partially associated with the process of dispersal of modern humans from an ancestral African source
and subsequent diversification driven by different adaptive
and demographic histories (Cavalli-Sforza et al., 1994; Fix,
1999). On the other hand, differences in human growth also
result from environmental and socio-economic factors, such
as nutrition, health, and social status (Eveleth, 1986). The
phenomenon of a secular trend in body size (van Wierigen,
1986), or increase in weight and height of a population in the
timescale of a few generations or within a single generation,
observed in North America, Europe, Japan, and developing
countries, results from a general improvement of living standards observed in those populations.
This context grounds the present report on the growth of
children and adolescents of Japanese descent from the city
of São Paulo, Brazil. The biometrical identity of this population derives from both genetic and environmental factors,
since the demographic and economic history of the Japanese
descendants distinguish them from the other inhabitants of
São Paulo. The city was characterized in the past by intense
* Corresponding author. e-mail: lvc22@hermes.cam.ac.uk
phone: +44-1223-335-461; fax: +44-1223-335-460
Published online 10 December 2004
in J-STAGE (www.jstage.jst.go.jp) DOI: 10.1537/ase.040702
© 2004 The Anthropological Society of Nippon
163
164
L. VINICIUS AND M.M. LAHR
ANTHROPOLOGICAL SCIENCE
Subjects and Methods
The study is based on a cross-sectional sample of 1297
children and adolescents aged 7–17 years (647 boys and 650
girls) of Japanese ancestry, from eight private schools in the
city of São Paulo in 1999. These schools are attended by
middle- and upper middle-class children, which are therefore wealthier on average than the total Japanese population
either in São Paulo or in Brazil. We included in the sample
only children whose four grandparents were born in Japan
(identified through the assessment of school records), which
corresponded to approximately 10% of the student population of these schools. Ages were assessed directly from
school registers. In tables and graphs, age cohorts are represented in the ‘±’ scale (e.g. cohort 7 designates children aged
6.5–7.49 years). A correction was made whenever we used
comparative data based on the alternative ‘+’ scale. For
example, in the growth survey carried out by the Japanese
government (MEXT, 2004), cohort 7 represents children
aged 7–7.99 years, and this was converted to age 7.5 in our
‘±’ scale. Since our data were collected in 1999, we used
Japanese data for the year 2000 in the comparisons, instead
of the more recent 2003 dataset (MEXT, 2004).
In some cases, it was not possible to measure both weight
and height of the same subject, so that the total number of
weight measurements is 647 for boys and 650 for girls, and
the total number of height measurements is 646 for boys and
648 for girls. Data collection was mostly performed by one
of us (L.V.) and complemented with data collected by graduate schoolteachers. Inter-observer error was not significant
(data not shown). Anthropometrical measurements were
taken during physical exercise classes, so that weights and
heights were taken of subjects barefoot with light clothing.
Stature was recorded to the nearest centimeter by the use of
a stadiometer, and no upward pressure was applied beneath
the mastoid process. Weight was recorded to the nearest
0.1 kg, except where a digital scale displaying exact values
in intervals of 0.05 kg was used. Differences between values
obtained from each scale are not statistically significant.
Comparative data from children and adolescents of European origin (3984 boys and 3894 girls) were collected at the
same schools and are described elsewhere (Castilho and
Lahr, 2001). The Japanese sample was also compared to the
2000 CDC (Centers for Disease Control and Prevention,
Department of Health and Human Services) growth reference from the USA (Kuczmarski et al., 2000). We used the
software ‘EPI Info’ (CDC) to convert weight for age and
height for age of our samples into CDC centiles. We
assessed differences in weight and height between Japanese
descendants born in São Paulo and populations from Japan
(MEXT, 2004) and urban China (Lin et al., 1992) by means
of independent samples t-tests.
Results
Comparison between children of Japanese and European
ancestry born in São Paulo reveals lower mean values of
weight and height throughout the growth period in the Japanese subjects (Figure 1, Figure 2, Table 1), but those differences are statistically significant only after the beginning of
Figure 1. Weight vs. age of boys (circles) and girls (squares) of
Japanese (black symbols) and European (white symbols) ancestry
from São Paulo, Brazil.
Figure 2. Height vs. age of boys (circles) and girls (squares) of
Japanese (black symbols) and European (white symbols) ancestry
from São Paulo, Brazil.
adolescence. Boys of European ancestry are significantly
taller only from age 15 and heavier only from age 17 (independent samples t-test, P < 0.001), while European girls are
significantly taller and heavier (P < 0.001), respectively,
from the ages 14 and 15 on. At 17 years, the mean weight
and height of Japanese boys are 64 kg and 170.4 cm, respectively, while European boys show mean values of 68.8 kg
and 176.2 cm (Castilho and Lahr, 2001). For Japanese girls,
the average weight and height at age 17 are 52 kg and
160.4 cm, while in European girls the mean values are
56.1 kg and 164 cm. In spite of the cross-sectional nature of
our data, the results point to the adolescent growth spurt as
responsible for the Japanese shortfall, in both sexes, in adult
body size relative to the population of European ancestry.
Japanese descendants were also compared to a wealthy
US sample of European ancestry, by converting their mean
values into centiles of the reference CDC distribution (Figure 3, Figure 4). The mean weight of Japanese boys and girls
is significantly above the US reference during the pre-adolescent period. At 8 years, for example, the weight and
Vol. 113, 2005
GROWTH OF JAPANESE DESCENDANTS IN BRAZIL
165
Table 1. Mean weight (kg) and height (cm) of boys and girls of Japanese descent born in São Paulo, Brazil
Age
7±
8±
9±
10 ±
11 ±
12 ±
13 ±
14 ±
15 ±
16 ±
17 ±
Boys
Girls
Weight (SD, n)
Height (SD, n)
Weight (SD, n)
Height (SD, n)
24.94 (4.8, 18)
28.36 (6.0, 28)
33.74 (8.2, 28)
34.43 (5.0, 28)
40.91 (11.8, 44)
43.61 (11.0, 36)
50.99 (11.8, 50)
56.39 (14.3, 61)
60.74 (12.5, 135)
64.54 (12.3, 110)
64.05 (13.0, 109)
120.58 (5.78, 18)
128.32 (5.2, 28)
134.21 (5.6, 28)
137.15 (4.5, 28)
144.86 (8.1, 44)
148.93 (6.1, 36)
158.06 (7.7, 50)
163.31 (6.5, 61)
168.07 (6.8, 135)
169.67 (6.7, 110)
170.42 (5.6, 108)
23.59 (3.4, 16)
26.59 (4.4, 15)
32.02 (7.9, 26)
33.21 (7.3, 17)
40.1 (7.9, 39)
43.31 (8.8, 34)
48.97 (10.7, 41)
49.96 (8.0, 72)
51.1 (9.0, 150)
52.64 (7.0, 155)
52.01 (6.6, 85)
122.38 (5.2, 16)
128.13 (3.5, 15)
132.35 (7.2, 26)
137 (6.6, 17)
146.84 (6.9, 39)
151.51 (7.0, 34)
156.6 (5.5, 41)
158.39 (5.0, 72)
159.38 (6.1, 150)
159.84 (5.6, 155)
160.4 (5.2, 85)
Standard deviation and sample size are given as (SD, n).
Figure 3. Weight (white symbols) and height (black symbols) of
boys (circles) and girls (squares) of Japanese ancestry from São Paulo,
expressed as centiles of the CDC growth reference.
height of Japanese boys are above the 81st and 69th centiles,
respectively. The height of boys and girls also oscillates
around the 50th centile, indicating no shortfall in growth.
However, moving towards the oldest cohorts, a decrease of
the Japanese values for both weight and height in both sexes
is evident. Differentiation from the US reference is faster in
height, whose average in girls falls below the 50th centile at
age 13, and in boys at age 14. The weight of girls is below
the 50th centile at age 15. By 17 years, height in both sexes
is well below the reference values, corresponding respectively to the 25th (boys) and 35th (girls) centiles, while
female weight corresponds to the 33rd centile. Only weight
in boys remains close to the US average (48th centile),
despite a similar fall in relative values during adolescence.
The results confirm the role of the adolescent growth spurt
as responsible for the shortfall in growth of Japanese descendants born in São Paulo. Our sample population is smaller in
size than the US sample despite similar socio-economic status.
Comparisons between the Brazilian Japanese sample and
two populations living in Japan and China were also carried
out. While Japanese descendants from Brazil seem to grow
slightly faster than children born and living in Japan
(MEXT, 2004), catch up growth in the latter implies that at
age 17 only female height is still significantly larger
(P < 0.01, independent samples t-test) in the children from
São Paulo.
On the other hand, comparison with Chinese children and
adolescents living in urban areas of China (Lin et al., 1992)
revealed significant differences. With the exception of
female height, Japanese descendants born in São Paulo
exhibited significantly larger sizes compared to Chinese
boys and girls, not only throughout the growth period but
also as young adults (age 17). The differences emerge
despite the fact that the Chinese sample is exclusively urban.
If children living in rural China (also sampled by Lin et al.,
1992) had been included in the comparisons, their growth
deficit relative to the Japanese descendants born in São
Paulo would be more evident. Unfortunately, more recent
data for the Chinese are not available.
Discussion and Conclusions
We identified significant shortfalls in the growth of welloff Japanese descendants from São Paulo, Brazil, both in
relation to a population of European descendants living in
the same city and to an US sample of equally high economic
status. The Brazilian Japanese sample begins to display
growth deficits after the onset of adolescence, suggesting
that a different timing and intensity of the growth spurt may
be responsible for their smaller adult weight and height, but
the cross-sectional nature of our data does not allow a deeper
assessment of this question, which would require longitudinal data in order to be fully addressed. We showed previously that European descendants from the same schools do
not exhibit any growth deficits in relation to the WHO/
NCHS reference (Castilho and Lahr, 2001).
A different picture emerges when the comparisons comprise only populations of Asian origin. Adult Japanese subjects from São Paulo and Japan show comparable weight
and height, with the single exception of female height. The
Japanese national survey in the year 2000 (MEXT, 2004)
portrayed the growth of an Asian population of high socio-
166
L. VINICIUS AND M.M. LAHR
ANTHROPOLOGICAL SCIENCE
Figure 4. Weight (A and C) and height (B and D) of boys and girls of Asian populations from São Paulo (white circles), Japan (black circles),
and China (white squares).
economic status, and therefore both the common genetic
background and equivalent social status explain the similarity in growth patterns with the Brazilian sample. On the
other hand, the Chinese anthropometric study used in our
analyses was undertaken in the 1980s and thus predated
most of the recent economic boom in China, and therefore
represents a population of lower socio-economic status than
that of Japanese descendants from São Paulo. It seems evident that, in growth studies of children of Asian background,
the national Japanese data (MEXT, 2004) should be used as
the standard for healthy growth instead of the 2000 CDC reference.
Acknowledgments
We thank the schools Rainha da Paz, Etapa, Humboldt,
Augusto Laranja, Pueri Domus, Rio Branco, Santa Cruz and
Sion. This research was funded by FAPESP (grants 96/88831 to L.V. and 95/9386-5 to M.M.L.).
References
Castilho L.V. and Lahr M.M. (2001) Secular trends in growth
among urban Brazilian children of European descent. Annals
of Human Biology, 28: 564–574.
Cavalli-Sforza L.L., Menozzi P., and Piazza A. (1994) The History
and Geography of Human Genes. Princeton University Press,
Princeton.
Eveleth P.B. (1986) Population differences in growth: environmental and genetic factors. In: Falkner F. and Tanner J.M. (eds.),
Human Growth: A Comprehensive Treatise, Volume 2. Plenum Press, New York, pp. 221–240.
Fix A.G. (1999) Migration and Colonization in Human Microevolution. Cambridge University Press, Cambridge.
Kuczmarski R.J., Ogden C.L., Grummer-Strawn L.M., Flegal
K.M., Guo S.S., Wei R., Mei Z., Curtin L.R., Roche A.F., and
Johnson C.L. (2000) CDC Growth Charts: United States.
Advance data from vital and health statistics, Number 314,
National Center for Health Statistics, Hyattsville, Maryland.
Lahr M.M. (1996) The Evolution of Modern Human Diversity.
Cambridge University Press, Cambridge.
Levy M.S. (1974) O papel da migracao internacional na evolução
da população brasileira (1872 a 1972). Revista de Saude
Pública, 8: 49–90.
Lin W.S., Zhu F.C, Chen A.C.N., Xin W.H., Su Z., Li J.Y., and Ye
G.S. (1992) Physical growth of Chinese school children 7–
18 years, in 1985. Annals of Human Biology, 19: 41–55.
MEXT (2004) Statistical Tables. Ministry of Education, Culture,
Vol. 113, 2005
Sports, Science and Technology of Japan. http://
www.mext.go.jp/english/statist/index.htm
Stringer C.B. (1990) The emergence of modern humans. Scientific
American, 263: 98–104.
van Wierigen J.C. (1986) Secular growth changes. In: Falkner F.
GROWTH OF JAPANESE DESCENDANTS IN BRAZIL
167
and Tanner J.M. (eds.), Human Growth: A Comprehensive
Treatise, Volume 2. Plenum Press, New York, pp. 307–332.
Yamashita K.T. (2001) Circle K Cycles. Coffee House Press, Minneapolis.