Xứ
Việt Nam 
XUVN.com
|
|
|
|
|
|
|
Ethnicity and body fatness in New Zealanders
Elizabeth Duncan, Grant Schofield, Scott Duncan, Gregory Kolt, Elaine Rush
http://www.nzma.org.nz/journal/117-1195/913/ The obesity epidemic
The prevalence of obesity has rapidly reached epidemic
proportions in both developed and developing countries around the world. The
World Health Organization (WHO) estimates that there are now more than 300
million obese people worldwide—an increase of 100 million since 1995.1
This rising level of obesity has increased the incidence of obesity-related
morbidities, such as cardiovascular disease, Type 2 diabetes, and
hypertension, thus imposing a major burden on healthcare systems and lowering
the quality of life for those affected. Furthermore, it is predicted that
rates of obesity in the next 20 years could be as high as 45–50% in the USA,
and 30–40% in Australia and England.2
New Zealand is no exception to these trends. National survey results show an
88% increase in the number of obese adults since 1989,3
to a stage where more than half of all adults are overweight and
obesity-related illnesses cost the health care system an estimated $303
million each year.4
Maori and Pacific Island (PI) adults appear particularly susceptible, with
obesity rates 1.9 and 2.5 times (respectively) higher than that of New Zealand
Europeans.5 Similar patterns have been
observed in younger New Zealanders. Results from the 2002 National
Children’s Nutrition Survey indicated that Maori and PI children were 3.0
and 5.3 times (respectively) more likely to be obese than children from other
ethnicities.7 Overall, 21.3% of New Zealand
children aged 5-14 years were classified as overweight, with a further 9.8%
obese.
Body Mass Index as a measure of obesity
Obesity is defined as a condition of excessive fat
accumulation to the extent that health and wellbeing may be impaired.8
In population research, body fatness (BF) is most commonly estimated using
body mass index (BMI), a simple anthropometric measurement of weight (kg)
divided by squared height (m2), which tends
to correlate well with both percent body fatness (%BF) and health-risk.9–13
Although more accurate techniques are available; such as four compartment
models that measure bone mineral content, body water, and body density
independently; BMI remains the most cost-effective and practical tool in
studies of this type.
In 1998, the WHO provided international BMI standards for classifying
overweight and obesity in adults based on the risk of obesity-related disease
for Europeans at each BMI category.8
Overweight was defined as BMI ≥25 kg.m-2
and obesity as BMI ≥30kg.m-2, with the
latter corresponding to approximately 25% and 35% BF in young European men and
women, respectively.14 An obvious limitation
of this measure is its inability to distinguish between fat and fat-free mass.
As such, standard BMI cut-offs for overweight and obesity may not represent
the same levels of BF in populations that differ significantly from the
typical European phenotype.
For young people, different BMI standards are required. The US Centers for
Disease Control and Prevention issued age- and sex-specific BMI charts for
defining overweight and obesity in those aged 2 to 20 years based on the 85th
and 95th percentile of an American reference population.15
Alternative thresholds have been provided by the International Obesity Task
Force (IOTF) using the mean of the BMI-age curves from six major countries.16
At a given age, individuals are classified as overweight or obese if they have
a BMI greater than the mean BMI-age curve that passes through 25 kg.m-2
or 30 kg.m-2 (respectively) at age 18 years.
The intention of these IOTF cut-offs was to establish a higher degree of
international applicability, although the averaging of the six diverse
datasets could be considered arbitrary.
Obesity and ethnicity
The WHO BMI thresholds for overweight and obesity are
widely used in field research; however, their relevance to all populations is
questionable. It is generally accepted that associations between BMI and BF
are dependent on age and gender. More recently, these associations have been
shown to vary with ethnicity. For example, Pacific Islanders tend to have
lower levels of BF than Europeans at a given BMI.17–19
Conversely, many Asian ethnic groups have higher levels of BF than Europeans
at specific BMI, thus putting them at greater risk of obesity-related disease
at relatively low BMI scores.20–22
Even at the same level of BF, risk profiles may differ between ethnic groups.23
This may be explained by ethnic-specific variation in the patterns of fat
distribution. Indeed, central fat accumulation (ie, an android fat pattern)
appears to be a greater predictor of obesity-related health risks than overall
fatness.24,25
Research indicates that, in general, Asian adults are more prone to visceral
and central obesity than Europeans.26,27 In
particular, Hughes et al28 found that Asian
Indians had a greater predisposition for central obesity than Malay and
Chinese Asians. Likewise, there is evidence that Asian children29
and adolescents30 have a greater central fat
mass when compared with Europeans and other ethnic groups. In accordance with
a higher %BF at a given body size, and a more centralised pattern of fat
distribution, elevated disease risks have been observed in Asian populations
at BMI scores well below the WHO thresholds defining overweight and obesity.
In response, the WHO released provisional recommendations that overweight and
obese BMI cut-off points for Asian populations in the Asia-Pacific region be
reduced to ≥23, and ≥25 kg.m-2,
respectively.31 Although a good starting
point, these guidelines do not take into consideration variance among
different Asian populations.
More recently, a WHO expert consultation on BMI in Asian populations concluded
that there is no single cut-off point appropriate for defining overweight or
obesity in all Asian groups.22
Recommendations from the consultation include: (1) retaining current WHO BMI
cut-off points for international classification; (2) adding ‘action
points’ of ≥23 and ≥27.5 kg.m-2
(representing ‘increased’ and ‘high’ risk) as a trigger for public
health action; (3) developing ethnic- and country-specific BMI action points;
and (4) refining BMI action points with waist circumference in populations
predisposed to central obesity.
Obesity in young people
Compared with adults, less is known about the body
composition of children and adolescents. There is, however, evidence that BMI
is not an equivalent measure of BF among young people from different ethnic
groups. At a given BMI, Chinese32 and
Hispanic33,34 youth have a higher level of BF
than Europeans, who in turn have more BF than African-Americans,35
and Maori and Pacific Islanders.36 These
disparities may evolve, or at least increase, during puberty. For example,
Ellis et al37 observed that ethnic
differences in body composition between Hispanics, African-Americans, and
Europeans were much less pronounced in children younger than 8 years of age
(pre-puberty).
Sexual maturation processes (that occur during puberty) affect body
composition, and can alter the associations between BMI and fat mass.35,38–40
Thus, differences in body composition observed during childhood and
adolescence may, in part, reflect ethnic-specific growth and development
patterns.41 In a 6-year follow-up study of
Chinese children, Wang et al42 noted that
overweight prevalence (defined according to IOTF age- and sex-specific BMI
cut-off points) decreased as children became adolescents. This apparent
reduction in overweight may be due to different BMI-age relationships between
the study and the IOTF reference populations.
Although the authors did not determine pubertal stage, they suggested that
Chinese adolescents tend to mature later than the IOTF reference populations,
thereby causing them to be misclassified. Consequently, the IOTF cut-offs may
not be appropriate for these populations. For future studies, consideration of
sexual maturation may be beneficial.
Explaining ethnic-specific relations between Body Mass Index and percentage of body fatness
Several factors have been proposed to help explain the
dependency of the BMI/%BF relation on ethnicity. First, body build/frame size
(as measured by wrist and knee girths) tends to vary among different ethnic
groups. A number of studies have noted that ethnic populations with relatively
high levels of BF at a given BMI also have a more slender build.43,44
Furthermore, Deurenberg et al45 found that
correcting for body build eliminated most of the ethnic-specific differences
associated with %BF prediction equations for bioelectrical impedance analysis
(BIA) in Chinese, Malay, and Indian Singaporeans.
In contrast, earlier research concluded that the prediction of %BF from BMI
was only slightly improved by the inclusion of body build parameters.46,47
It is possible that the effects of body build were not observed in these
studies due to low inter-group and/or high intra-group variability.44
A second factor that may contribute to ethnic-specific relationships between
BMI and %BF is variation in sitting height relative to total height.
Individuals with long legs (low sitting height) generally have a lower BMI
and, as such, %BF may be underestimated from BMI.48
Relative sitting height tends to be higher in Asian ethnic groups, although
the effects on BMI are inconclusive—most likely due to the large intra-group
variation in this parameter.43,44
Given that differences in body build may explain a large proportion of the
ethnic variation in relationships between BMI and %BF, frame size represents
an alternative criterion to ethnicity on which to base BMI cut-offs. As
ethnicity is self-identified, individuals may affiliate with an ethnic group
with which they have no genetic relation. As such, classification according to
body build (rather than ethnic group) may help control for inaccuracies when
defining ethnicity. However, it is unlikely that collecting data on frame size
will be practical in population studies.
Finally, there may be differences in physical activity level among ethnic
groups. More active individuals are likely to have a higher proportion of
muscle mass, and therefore the potential for overestimation of %BF from BMI.44
Such a tendency may only be observable in athletes performing high levels of
activity. Nevertheless, future studies should include anthropometric measures
of body build and physical activity levels in order to increase our
understanding of differences in the BMI/%BF relationship among ethnic groups.
New Zealand’s issues regarding body fatness and ethnicity
New Zealand has an ethnically diverse population
comprising mainly New Zealand Europeans (80.0%), Maori (14.7%), Asians (6.6%),
and Pacific Islanders (6.5%).49 Despite this
diversity, ethnic variation in BF and other body composition variables has yet
to be investigated in all major ethnic groups.
However, researchers have compared ethnic differences in BMI and %BF among
Maori, PI, and European populations. Several studies have found that Maori and
PI adults tend to be leaner (ie, have a lower %BF, and higher fat-free mass)
than New Zealand Europeans of the same body size.17-19
Similar results have been observed in children. Rush et al36
noted that Maori and PI girls have (on average) 3.7% less BF than New Zealand
European girls of the same body size. Furthermore, a related study by Tyrell
et al6 found a small, but statistically
significant, difference in the relationship between BMI and %BF in New Zealand
European, Maori, and PI schoolchildren aged 5–10 years; although they
suggested that the effects of ethnicity were not clinically relevant.
Even though Maori and Pacific Islanders tend to have a higher proportion of
lean mass to fat mass than New Zealand Europeans at a given BMI, as a
population they maintain a greater absolute fat mass. Indeed, when higher BMI
thresholds are applied to Maori and PI peoples to counteract the high
lean-to-fat mass ratio (26 kg.m-2 and 32 kg.m-2
for ‘overweight’ and ‘obesity’, respectively), these two groups remain
twice as likely to be obese than the ‘European and Other’ group.5
Not surprisingly, Maori and PI populations also have a much higher prevalence
of type 2 diabetes when compared to Europeans.50
However, it is noteworthy that the prevalence of type 2 diabetes among New
Zealand Indians exceeds that seen in Maori and Pacific Islanders.
The high prevalence of diabetes among Indians is in line with the elevated
levels of BF at a given body size seen among Asian populations overseas. This
is an issue of increasing importance to New Zealand given that Asian people
make up the fastest growing ethnic group, more than doubling in number between
1991 and 2001.49 Furthermore, Asians are
projected to account for 13% of New Zealand’s population by 2021.
In spite of their population growth, Asian ethnic groups have been largely
neglected by New Zealand health and research policies. For example, only Maori
and PI children were over-sampled in the 2002 National Children’s Nutrition
Survey. In addition, Maori and PI children were analysed separately, whereas
children of Asian decent were grouped with New Zealand Europeans. This is a
common theme in national surveys by government organisations; such as the
Ministry of Health, and Sport and Recreation New Zealand. In order to
understand the public health needs of Asian populations in New Zealand, and to
tailor preventative health strategies, it is vital that future surveys
distinguish between these ethnic groups.
At present, standard BMI thresholds for ‘overweight’ and ‘obesity’ are
applied to Asian populations as there are no robust New Zealand data available
on the relationship between BMI and body composition variables in this ethnic
group. Consequently, Asian groups at risk for health complications
(accompanying their overweight and obesity conditions) may not be targeted in
interventions to prevent/treat obesity. The only evidence available is from a
study by Tyrell et al6 that included two
small groups of Asian children in their investigation of the relationship
between BMI and body composition in Maori, Pacific Islanders, and New Zealand
Europeans. Although results were not presented, the authors commented that
Asian Indian children tended to have a higher %BF at a given BMI compared with
New Zealand Europeans. However, caution must be taken when interpreting this
statement given the small sample size and the fact that %BF was estimated from
bioelectrical impedance analysis using a prediction equation that was not
specifically developed for Asian Indian children.
The recommendations put forward by the recent WHO expert consultation22
offer promise for the classification of overweight and obesity in New
Zealand’s multiethnic society. For clinicians assessing the health status of
individuals, BMI thresholds should be refined by consideration of ethnicity
and other risk factors such as waist circumference. At a population level,
implementation of additional BMI action points will better reflect the
continuum of BF and associated health risk. However, valid comparisons with
overseas statistics will only be possible if the criteria used to define
overweight and obesity are consistent. In these instances, the retention of
the standard WHO cut-offs (25 and 30 kg.m-2)
is advisable. Ultimately, national BMI action points should be developed for
all major ethnic groups in New Zealand based on large-scale population studies
of BMI, BF, and health risk.
Conclusions
Accurate assessment of overweight and obesity is vital
to assist public health organisations in identifying at-risk groups and to
facilitate development of appropriate preventative strategies. At a population
level, BF is most commonly assessed using BMI. Although WHO established
universal BMI standards for defining overweight and obesity, studies have
shown that these BMI thresholds do not provide an equivalent measure of BF and
associated health risk across different ethnic groups. Consequently, WHO
recently recommended the use of additional BMI cut-offs as public health
action points, such as ethnic- and country-specific BMI cut-offs for
overweight and obesity.
In New Zealand, knowledge of the ethnic variation in BF and other body
composition variables is restricted to New Zealand European, Maori, and PI
ethnic groups. New Zealand Asians are of particular interest because of their
rapid population growth, and the lack of published data on their BMI/%BF
relationships. Furthermore, compared with Europeans, Asians from other
countries show elevated levels of BF and greater morbidity and mortality at a
given BMI.
In conclusion, large-scale studies are needed to determine the relations
between BMI, %BF, BF distribution, and health risk across all major ethnic
groups in New Zealand. For young people, these studies should also consider
maturational stage. Resulting data will enable development of ethnic-specific
BMI thresholds for overweight and obesity—however this is only a starting
point.
There is also a clear lack of knowledge concerning ethnic variation in other
areas, such as physical activity and diet. An understanding of these issues is
imperative for tailoring preventative interventions that will counteract the
burgeoning epidemic of obesity in New Zealand.
Author information: Elizabeth K.
Duncan, Doctoral Student, Division of Sport and Recreation, Faculty of Health,
Auckland University of Technology, Auckland; Grant Schofield, Lecturer,
Division of Sport and Recreation, Faculty of Health, Auckland University of
Technology, Auckland; Scott Duncan, Doctoral Student, Division of Sport and
Recreation, Faculty of Health, Auckland University of Technology, Auckland;
Gregory Kolt, Professor, Faculty of Health, Auckland University of Technology,
Auckland; Elaine Rush, Associate Professor, Division of Sport and Recreation,
Faculty of Health, Auckland University of Technology, Auckland.
Acknowledgement: Elizabeth K Duncan
acknowledges the support of the Foundation for Research, Science and
Technology through the Top Achiever Doctoral Scholarship.
Correspondence: Elizabeth K Duncan,
Division of Sport and Recreation, Faculty of Health, Auckland University of
Technology, Private Bag 92006, Auckland. Fax: (09) 917 9746; email: elizabeth.duncan@aut.ac.nz
References:
|
