Gill (2014) Physical Activity, Ethnicity and Cardio-Metabolic Health. Does One Size Fit All.

Atherosclerosis 232 (2014) 319e333
Contents lists available at ScienceDirect
Atherosclerosis
journal homepage: www.elsevier.com/locate/atherosclerosis
Review
Physical activity, ethnicity and cardio-metabolic health: Does one size

t all?
Jason M.R. Gill a, *, Carlos A. Celis-Morales b, Nazim Ghouri a
a
b
Institute of Cardiovascular and Medical Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK
Human Nutrition Research Centre, Institute for Ageing and Health, Newcastle University, Newcastle on Tyne, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 August 2013
Received in revised form
7 November 2013
Accepted 8 November 2013
Available online 23 November 2013
A large and consistent body of epidemiological evidence indicates that low levels of physical activity, low
levels of cardiorespiratory tness and high levels of sedentary behaviour are associated with increased
risk of cardio-metabolic diseases. However, most such studies have been undertaken in populations of
White European descent. The available data from non-White populations suggests that physical activity
is also protective in these groups, but the threshold level of activity needed to confer low risk, particularly for type 2 diabetes, may not be the same across all ethnic groups. In patients with impaired glucose
regulation, lifestyle interventions, including physical activity as a component (often in combination with
weight loss), are effective at reducing risk of incident diabetes across a range of ethnic groups. However,
the optimal levels of physical activity for prevention of diabetes and cardiovascular disease amongst the
general populations of different ethnic groups have not been rmly established. Emerging data suggest
that innate differences in cardiorespiratory tness levels and capacity for fat oxidation potentially
contribute to ethnic differences in the cardio-metabolic risk prole and that ethnicityespecic physical
activity guidelines may be conceptually warranted. More study is needed to understand how and why
the doseeresponse relationship between physical activity and cardio-metabolic risk differs according to
ethnicity and to determine the best approaches to promote physical activity in non-White ethnic groups.
2013 Elsevier Ireland Ltd. All rights reserved.
Keywords:
Cardiovascular disease
Diabetes
Ethnicity
Race
Physical activity
Fitness
Exercise
Contents
1.
2.
3.
4.
5.
6.
7.
8.
9.
Physical activity, cardiorespiratory fitness and sedentary behaviour: three related factors influencing cardio-metabolic disease risk . . . . . . . . . . . . . . . .1
Physical activity guidelines and population attributable risk associated with inactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
Physical activity and cardio-metabolic disease risk in non-White populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9
Lifestyle intervention trials for cardiovascular disease prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
Lifestyle intervention trials for diabetes prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Ethnicityeenvironment interactions and diabetes risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
Can physiological differences between ethnic groups explain the ethnic differences in diabetes risk? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13
Can ethnicespecific physical activity recommendations help to overcome innate differences in diabetes risk between ethnic groups? . . . . . . . . . . . .14
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1. Physical activity, cardiorespiratory tness and sedentary

behaviour: three related factors inuencing cardio-metabolic
disease risk
* Corresponding author. BHF Glasgow Cardiovascular Research Centre, Institute
of Cardiovascular and Medical Sciences, College of Medical, Veterinary and Life
Sciences, University of Glasgow, Glasgow G12 8TA, UK. Tel.: 44 (0) 141 3302916;
fax: 44 (0) 141 3305481.
E-mail address: jason.gill@glasgow.ac.uk (J.M.R. Gill).
0021-9150/$ e see front matter 2013 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.atherosclerosis.2013.11.039
A large and consistent body of prospective epidemiological data

(>1.5 million participants) has rmly established that a low level of
physical activity is an important modiable risk factor for all-cause
mortality, cardiovascular disease and type 2 diabetes [1e5]. The
320
J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333
most active individuals experience about a 30% reduction in cardiovascular disease, diabetes all-cause mortality risk compared to
the least active [1,3e5], although the magnitude of the diabetes risk
reduction reduces by about half after adjustment for BMI [2,4].
These risk reduction values are likely to be underestimates because
the studies generally assessed physical activity by self-report
questionnaire which can attenuate the true relationship between
activity and health outcome measures, due to regression dilution
bias errors due to mis-reporting of activity levels [6,7] and
(particularly for all-cause mortality and cardiovascular disease)
because many studies adjusted for factors that are likely to mediate
some of the benecial effects of physical activity (e.g. blood pressure, diabetes) [1,3,5]. Studies of the association of cardiorespiratory tness (i.e. a physiological trait dened as the ability of the
cardiovascular and respiratory systems to supply oxygen to working muscles during sustained physical activity, which is typically
assessed by measuring maximal oxygen uptake (VO2max) during an
incremental exercise test) and health outcomes generally report
larger effects than studies of physical activity (i.e. a behaviour
dened as bodily movement produced by skeletal muscles that
results in energy expenditure, which can be assessed by self-report
questionnaire or objective (e.g. accelerometers) methods). A high
level of tness is typically associated with a reductions in all-cause
mortality risk of w40e45% [1,3], cardiovascular disease risk by
w50e60% [1,3] and type 2 diabetes risk of w50e70% [1], compared
to a low level of tness, in prospective cohort studies. Limited data
suggest that there may be an interaction between physical activity
level, cardiorespiratory tness and cardiovascular disease risk, with
the benets of a high level of physical activity in reducing risk
appearing to be most pronounced in individuals with low innate
cardiorespiratory tness levels and those with innately high tness
levels appear to be at relatively low risk cardiovascular disease risk
irrespective of their physical activity level [8]. In recent years, it has
emerged that time spent engaged in sedentary behaviours e
dened as non-sleeping activities in a sitting or reclining posture
with energy expenditure 1.5 METS (where 1 MET is resting energy
expenditure) [9] e is associated with increased risk of cardiovascular disease, diabetes, metabolic syndrome, obesity and death
from any cause, with this effect often being independent of time
spent engaged in moderate-to-vigorous physical activity (i.e. activities at an intensity >3 METS) [10e12].
2. Physical activity guidelines and population attributable
risk associated with inactivity
Underpinned by this substantial and consistent evidence base,
many national and international bodies have made guideline recommendations that all adults should engage in at least 150 min of
moderate intensity (3e6 METS), or 75 min of vigorous intensity
(>6 METS), physical activity per week [13e15]. This is equivalent to
expending w1000 kcal/week (w4.2 MJ/week) in physical activity of
at least moderate intensity, a level of activity shown to substantially
(although not maximally e higher levels of activity reduce risk
further) reduce risk of mortality and cardio-metabolic disease [1].
Recent UK physical activity guidelines have also recommended that
individuals should limit extended periods of sedentary time [13],
although there are currently insufcient data available to provide a
specic target to aim for.
The number of people worldwide not achieving these physical
activity levels, and the consequent public health implications, are
substantial. It has recently been estimated that about 35% of the
adults worldwide do not meet these physical activity targets and,
accordingly, that inactivity is responsible for 9% of premature
mortality worldwide, equating to 5.3 million deaths annually e
more than is attributable to smoking [16]. In a study of American
adults, it was estimated 16% of mortality could be attributed to low

cardiorespiratory tness; more than smoking, obesity and hypercholesterolaemia combined [17].
3. Physical activity and cardio-metabolic disease risk in nonWhite populations
However, it is important to note that the majority of studies
prospectively investigating the effect of physical activity dose on
cardio-metabolic disease and mortality risk have been performed
in people of white European ethnicity, with a relatively small
number of studies including groups of non-white origin. This is key
limitation to the evidence base that has been highlighted in systematic reviews of the literature [1] and physical activity guideline
statements [14]. Prospective cohort studies which have evaluated
the effects of physical activity on cardio-metabolic disease or
mortality risk in non-White populations are summarised in Table 1.
There is evidence of a protective effect of physical activity in Chinese and Japanese populations. A study from the National Health
Interview Survey (NHIS) in Taiwan reported a dose-dependent inverse relationship between level of physical activity and mortality
risk in Taiwanese adults aged over 65 years [18]. Data from the
Shanghai womens health study indicated that high levels of
physical activity were protective against cardiovascular and allcause mortality [19], and incident type 2 diabetes [20] in middleaged Chinese women. A recent report from the Shanghai Mens
Health Study showed that regular participation in walking, jogging
and Tai Chi was associated with reduced risk of CVD and all-cause
mortality [21]. A series of publications from the Honolulu Heart
Program have reported inverse associations between physical activity level and risk of coronary heart disease [22e24], stroke
[25,26], diabetes [27], and mortality [28] in middle-aged to elderly
Japanese-American men. The protective effect of physical activity
against incident diabetes has been conrmed in Japanese men [29e
31], and against CHD, stroke and total CVD, and all-cause mortality
in has been conrmed in Japanese men and women [32,33] in reports from a number of other cohorts. There is also evidence that
high physical activity reduces incidence of all-cause mortality in
elderly Mexican-Americans [34]. However, while these studies
demonstrated a protective effect of physical activity in Chinese,
Japanese and Mexican-American populations, none of these reports
provide a head-to-head comparison of the doseeresponse relationship between physical activity and disease risk between
different ethnic groups. Because of differences in the quantication
of physical activity between the various questionnaire-based
physical activity instruments used in different studies, such headto-head comparisons within a single study are needed to determine whether the doses of physical activity required for benet are
similar across ethnic groups. Cohort studies including more than
one ethnic group provide data that suggest that the doseeresponse
relationship between physical activity level and cardio-metabolic
disease or mortality risk may not be identical across all ethnicities. An early report (after 4e7 years follow-up) from the Atherosclerosis Risk in Communities (ARIC) study in the US found that
increasing levels physical activity attenuated risk of incident coronary heart disease (CHD) events in non-Black (who were 99%
White), but not in Black adults [35]. The authors could not provide a
denitive explanation for this differential effect of physical activity
between the Black and non-Black groups, but suggested that the
small number of Black participants engaging in vigorous exercise,
lower reliability of the physical activity questionnaire in Blacks, and
the small number of CHD events in the Black cohort, may have
contributed to this null effect in this early report [35]. This interpretation is borne out by a more recent analysis from ARIC cohort
(after 19e21 years follow-up), which reported similar associations
Table 1
Prospective cohort studies which included non-White European populations investigating the association between physical activity and risk of cardio-metabolic incidence, cardio-metabolic disease mortality, or all-cause
mortality.
First author, study
(country and
publication year)
Ethnic group
studieda
Participants
Follow-up
7530 men aged 45e68 22 years

years at baseline
Japanese
Abbott, Honolulu
Heart Study (US,
2003) [25]

years at baseline
African-Americans
Autenrieth, The
Atherosclerosis and White
Americans
Risk in
Communities
Study (US, 2013)
[41]
Median follow-up
13,069 White and
African-American men 18.8 years
and women aged 45
e64 at baseline
African-Americans
Bell, The
Atherosclerosis and Caucasians
Risk in
Communities
Study (US, 2013)
[36]
Burchel, Honolulu Japanese
Heart Study (US,
1995) [27]
3707 African-American Up to 21 years

and 10,018 Caucasian
men and women aged
45e64 years at baseline
De Munter, Sunset
study
(Netherlands,
2013) [43]
South Asian
Surimanese, African
Surimanese and
White Dutch
Donahue, Honolulu Japanese

Heart Study (US,
1988) [23]
Evenson, The
Atherosclerosis
Risk in
Communities
Black Americans
and White
Americans

years at baseline
Up to 7 years
370 South Asian
Surimanese, 689
African Surimanese,
and 567 Dutch men and
women, aged 35e60
years at baseline
years at baseline
14,575 White and Black Average follow-up

American men and
7.2 years
women aged 45e64 at
baseline
Baseline physical activity

assessment
Main ndings
In younger middle-aged men (45e54 years) there was no signicant effect of

physical activity on stroke incidence. In older middle-aged men (55e68 years),
greater levels of physical activity were associated with lower incidence of stroke
(p < 0.001). This signicant trend in the older men was evidence for
haemorrhagic, intracerebral haemorrhage and subarachnoid haemorrhage
stroke sub-types.
The protective effect of physical activity on the incidence of stroke (rate/1000)
Physical activity index
Incident
calculated from self-report of became statistically signicant in men aged 75 years (Lower physical activity
thromboembolic
24-h habitual physical activity index tertile: 40.5/1000; Middle physical activity index tertile: 17.1/1000;
stroke
Upper physical activity index tertile: 28.3/1000; p-trend 0.032).
levels
Incident ischaemic Self-reported physical activity Signicant inverse associations were found between physical activity categories
stroke
by questionnaire (Baecke)
and ischaemic stroke in adjusted models (age, sex, race-centre, education,
cigarette-years). Compared with poor physical activity, adjusted HR for ideal
physical activity was 0.78 (95% CI 0.62e0.97) for ischaemic stroke. Additional
adjustments for cardio-metabolic risk factors attenuated the HR. Further sexand race-specic analyses revealed that the association was predominantly
observed among males and among African-Americans.
Incident CVD
Self-reported physical activity Physical activity inversely related to CVD, heart failure, and CHD incidence in
by questionnaire converted to both ethnic groups (p values for trend 0.0001), and with stroke in Africanminutes/week of moderate and Americans. Hazard ratios for CVD for each higher physical activity category were
similar for both ethnic groups: 1.0, 0.65 (95% CI 0.56e0.75), and 0.59 (95% CI
vigorous activity
0.49e0.71) for African-Americans and 1.0, 0.74 (95% CI 0.66e0.83), and 0.67
(95% CI 0.59e0.75) for Caucasians (p value for interaction 0.38).
Overall 6-year cumulative incidence of T2DM decreased progressively with
Incident T2DM
calculated from self-report of increasing quintile of physical activity from 73.8 to 34.3 per 1000 (p < 0.0001 for
24-h habitual physical activity trend) in all men and from 53.9 to 21.7 per 1000 (p < 0.0001 for trend) among
men with a non-fasting glucose level < 225 mg.dl1 1 h after a 50 g glucose load,
levels
the latter group being less likely to have unrecognised T2DM at baseline. Ageadjusted ORs for diabetes comparing the upper with the lower four quintiles of
physical activity were 0.55 (95% CI 0.41e0.75) for all men and 0.50 (95% CI 0.33
e0.74) for men with glucose < 225 mg.dl1. After adjustment for age and other
cardio-metabolic risk factors, ORs were still statistically signicant and similar
in magnitude. Restriction of analyses to men who remained free of CVD during
the study period produced similar results, suggesting that inactivity due to
subclinical CVD is unlikely to be responsible for these ndings.
CVD-related
Physical score calculated from Overall, low amount of physical activity (HR 1.88, 95% CI 1.23e2.86) was
hospital discharge self-report of weekly habitual associated with CVD-related hospital admission with no statistically signicant
effect of ethnicity on association. After adjustment for age, gender, lifestyle and
physical activity levels
metabolic syndrome risk factors, HRs for CVD-related hospital discharge for a
low amount of physical activity was 2.77 (95% CI 1.31e5.87) for Africans, 1.53
(95% CI 0.76e3.05) for South Asians, and 1.55 (95% CI 0.73e3.30) for the Dutch
(p-value for ethnicity by lifestyle interaction 0.41).
Among those aged 45e64 years, rate of CHD in men who led active lifestyles was
Incident CHD
calculated from self-report of 30% lower than the rate in those who were less active (RR 0.69; 95% CI 0.53
24-h habitual physical activity e0.88). In those>64 years, rate of CHD in active men was less than half the rate
in those who led more sedentary lifestyles (RR 0.43, 95% CI 0.19e0.99). Results
levels
were similar after controlling for several CVD risk factors and potentially
confounding variables.
Incident ischaemic Self-reported physical activity For the overall cohort, Ischaemic stroke incidence rates were highest in the
stroke
by questionnaire
lowest quartile of sport, leisure, and work scores. HRs ischaemic stroke for the
highest quartile compared with the lowest quartile of activity adjusted for age,
sex, race-centre, education, and smoking, were sport 0.83 (95% CI 0.52e1.32),
Incident stroke

calculated from self-report of
24-h habitual physical activity
levels
321
(continued on next page)
Japanese
Abbott, Honolulu
Heart Study (US,
1994) [26]
Primary outcome
measure
First author, study

(country and
publication year)
322
Table 1 (continued )
Ethnic group
studieda
Participants
Follow-up
Study (US, 1999)

[40]
Fretts, The Strong American Indians

Heart Study (US,
2009) [47]
Gillum, NHANES 1
(US, 1996) [39]
Black and nonBlack Americans
14,040 Black and non- Up to 4e7 years

Black men and women
(non-Black participants
were 99% White), aged
45e64 at baseline
Non-Hispanic
whites, nonHispanic blacks,
Hispanic and other
2896 men and women 8 years

with diabetes (73.8%
Non-Hispanic whites,
16.6% non-Hispanic
blacks, 7.9% Hispanic
and 2.7% other) aged
18e95 years at baseline
Black Americans
and White
Americans
Mean follow-up
771 Black and 5058
white men and women 11.6 years
aged 45e74 at baseline
12 years
Japanese
Hakim, Honolulu
Heart Study (US,
1998) [28]
707 non-smoking
retired men aged 81
e61 years at baseline
Japanese
Hakim, Honolulu
Heart Study (US,
1999) [24]
2678 men aged 71e93 2e4 years

years at baseline
Hsia, Womens
Health Initiative
Observational
Study (US, 2005)
[45]
Caucasian, AfricanAmerican, Hispanic,

American Indians
and Asian/Pacic
Islander
5.1 years
74,240 Caucasian
women, 6465 AfricanAmerican women, 3231
Hispanic women, 327
American Indian
women, and 2445
Asians/Pacic Islander
women, aged 50e79
years at baseline

assessment
Main ndings
leisure 0.89 (95% CI 0.57e1.37), and work 0.69 (95% CI 0.47e1.00). Further
adjustment for factors that likely were intermediate variables (hypertension,
diabetes, brinogen, and BMI) between physical activity and stroke attenuated
the associations. No ethnicity interaction given. Ischaemic stroke incidence
declined with increasing physical activity score for Black Americans, while
among White Americans the pattern was less consistent (numerical data not
reported).
Incident T2DM
Self-reported physical activity Compared with participants who reported no physical activity, those who
reported any physical activity had a lower risk of diabetes: ORs were 0.67 (95%
questionnaire over the past
CI 0.46e0.99), 0.67 (95% CI 0.45e0.99), and 0.67 (95% CI 0.45e0.99) for
year
increasing tertile of physical activity, after adjustment for age, sex, study site,
education, smoking, alcohol use, and family history of diabetes. Further
adjustment for body mass index and other potential mediators attenuated the
risk estimates.
Incident CHD
Self-reported physical activity Age-, race-, and eld centre-adjusted RR of CHD was 0.73 in women and 0.82 in
by questionnaire converted to men per each SD increment in the sports index (P < 0.05). For the leisure index,
sport, leisure and work indices these relative risks were 0.78 for both sexes (p < 0.05). The work index was not
associated with CHD. Associations held for non-blacks, but there was no
association between the sport or leisure indices and CHD among blacks
(numerical data not reported).
All-cause mortality Self-reported physical activity Compared with inactive individuals, those who walked at least 2 h per week had
a 39% lower all-cause mortality rate (HR 0.61, 95% CI 0.48e0.78) and a 34% lower
and CVD mortality and time spent walking in
CVD mortality rate (HR 0.66, 95% CI 0.45e0.96), controlling for sex, age, race,
preceding 2 weeks by
body mass index smoking, and comorbid conditions. Mortality rates were
questionnaire
lowest for persons who reported that their walking involved moderate
increases in heart and breathing rates (all-cause mortality HR 0.57, 95% CI 0.41
e0.80; CVD mortality HR 0.69, 95% CI 0.43e1.09). Ethnicity interaction was not
signicant.
In White women aged 65e74 years, low non-recreational activity was
Incident stroke
Recreational and habitual
associated with an increased risk of stroke (RR 1.82, 95% CI 1.10e3.02) after
physical activity
adjusting for the baseline risk factors of age, smoking, history of diabetes,
characterisation by
history of heart disease, education, systolic blood pressure, serum total
questionnaire
cholesterol, body mass index, and haemoglobin concentration. Similar
associations were seen for men and for Blacks (numerical data not reported).
All-cause mortality Self-reported distance walked Mortality rate among men who walked less than 1 mile per day was nearly
per day
twice that among those who walked more than 2 miles per day (40.5% vs. 23.8%,
p 0.001). The cumulative incidence of death after 12 years for the most active
walkers was reached in less than 7 years among the men who were least active.
Incident CHD
Self-reported distance walked Men who walked 0.25 miles per day had a 2-fold increased risk of coronary
per day
heart disease versus those who walked 1.5 mile/d (5.1% versus 2.5%; p <0.01).
Men who walked 0.25 to 1.5 mile per day were also at a signicantly higher risk
of coronary heart disease than men who walked longer distances (4.5% versus
2.5%; p < 0.05). Adjustment for age and other risk factors failed to alter ndings.
Incident T2DM
Self-reported physical activity Among Caucasian women, walking (multivariate-adjusted HRs 1.00, 0.85, 0.87,
by questionnaire
0.75, 0.74; p < 0.001 for trend across exercise quintiles) and total physical
activity score (hazard ratios 1.00, 0.88, 0.74, 0.80, 0.67; p 0.002) demonstrated
a strong inverse relationship with diabetes risk. In BMI-adjusted models,
African-American women in higher physical activity categories were less likely
to develop diabetes than women in the lowest physical activity category. After
adjusting for age and multiple risk factors, however, no signicant association
between physical activity and diabetes risk was apparent for African-American,
Hispanic, or Asian women.
Folsom, The
Atherosclerosis
Risk in
Communities
Study (US, 1997)
[35]
Gregg, National
Health Interview
Survey (US,
2003) [42]
1651 men and women, 10 years

aged 45e74 years at
baseline
Primary outcome
measure
Inoue, Japan Public Japanese

Health Centrebased
Prospective
Study (Japan,
2008) [33]
Lan, The National

Health Interview
Survey in Taiwan
(Taiwan, 2006)
[18]
Manson, Womens
Health Initiative
Observational
Study (US, 2002)
[37]
Taiwanese
83,034 men and

women aged between
45 and 74 years at
baseline
5 years
2113 men and women 2 years

aged 65 years at
baseline
Chinese
Matthews,
Shanghai
Womens Health
Study (China,
2007) [19]
74,942 women aged 40 5.7 years

Nakanishi, (Japan
2004) [31]

years at baseline
Japanese
Japanese
Noda, The Japan
Collaborative
Cohort Study for
Evaluation for
Cancer Risk
(Japan, 2005)
[32]
Okada, Osaka
Health Survey
(Japan, 2000)
[30]
Japanese
32,953 men and 44,723 9.7 years

women aged 40e79
years at baseline
6013 Japanese men

aged 35e60 years at
baseline
16 years
Compared with subjects in the lowest quartile, increased daily total physical
activity was associated with a signicantly decreased risk of all-cause mortality
in both sexes, with HRs for the second, third, and highest quartiles of 0.79 (95%
CI 0.71e0.87), 0.82 (95% CI 0.74e0.91), 0.73 (95% CI 0.66e0.81), respectively, for
men and 0.75 (95% CI 0.66e0.85), 0.64 (95% CI 0.56e0.74), 0.61 (0.52e0.73),
respectively, for women, in fully-adjusted models. Signicantly decreasing risk
of CHD and cerebovascular mortality was also observed with increasing physical
activity quartile in both men and women.
All-cause mortality Self-reported physical activity Reduced risk of mortality in active individuals compared to sedentary
individuals for all cause of mortality (HR 0.65, 95% CI 0.47e0.91). Similarly,
questionnaire (number of
leisure activities and exercise individuals that engage in leisure physical activity once and two or more times a
condition active vs. inactive) week had an HR of 0.70 (95% CI 0.50e0.98) and 0.35 (95% CI 0.15e0.82),
respectively compared to those that did no leisure physical activity.
Incident CHD and Self-reported physical activity Increasing physical activity score had a strong, inverse association with the risk
of CHD and CVD events (similar effects in White and Black women, but no
CVD events (Results questionnaire
physical activity*ethnicity interaction was reported). White women in
reported for all
increasing quintiles of energy expenditure (METs) had RRs of CV events of 1.00,
women, White
0.82, 0.73, 0.63, and 0.55 (p < 0.001) respectively. Similar associations were
women and Black
reported for Black women (1.00, 0.81, 0.58, 0.68 and 0.48; p 0.02,
women only)
respectively).
For the combined multi-ethnic cohort, overall physical activity participation

was associated with reduced mortality risk compared to lack of physical activity
participation history at baseline visit (HR 0.54, 95% CI 0.39e0.76). In
multivariable models physical activity participation remained associated with
lower mortality risk (HR 0.66, 95% CI 0.46e0.93) and CVD mortality (HR 0.56,
95% CI 0.32e0.97). Similar results were seen when restricting to AfricanAmericans for CVD death in fully-adjusted models (HR 0.57 95% CI 0.31e1.05).
Due to the low number of deaths amongst Whites and Hispanics, ethnic-specic
analyses for these subgroups were not performed.
All-cause mortality Self-reported physical activity Increasing overall activity, adult exercise, non-exercise related activities and
and CVD mortality questionnaire (housing, leisure cycling were all related to a signicant reduction in all-causes of mortality
compared with inactive individuals (HR 0.6, 95% CI 0.51e0.73; HR 0.64, 95% CI
and transportation related
0.36e1.14; HR 0.66 95% CI 0.55e0.79; and HR 0.66, 95% CI 0.40e1.07,
activities).
respectively). Increasing overall physical and adult exercise signicantly
reduced CVD-related mortality (HR 0.66, 95% CI 0.46e0.95 and 0.23, 95% CI 0.03
e1.64, respectively).
Incident IGT and
Physical activity was measured RR of developing IGT and T2DM according to increasing quartiles of overall daily
energy expenditure was 1.00, 0.71, 0.67, and 0.40 (p < 0.001) and 1.00, 0.76, 0.70
T2DM
using a 1-day activity record.
This data was used to estimate and 0.41 (p < 0.001), respectively.
energy expenditure.
Incident CHD and Self-reported physical activity Men and women who reported having physical activity in the highest category
(i.e., walking 1 h/day or doing sports 5 h/week) had a 20%e60% lower ageStroke
questionnaire was used to
collect information related to
adjusted risk of mortality from CVD, compared with those in the second lowest
time spent walking and time
physical activity category (i.e., walking 0.5 h/day, or sports participation for 1
spent doing sport.
e2 h/week). The multivariate-adjusted HRs for the highest versus the second
lowest categories of walking or sports participation were 0.71 (95% CI 0.54
e0.94) and 0.80 (95% CI 0.48e1.31), respectively, for ischaemic stroke; 0.84 (95%
CI 0.64e1.09) and 0.51 (95% CI 0.32e0.82), respectively, for CHD; and 0.84 (95%
CI 0.75e0.95) and 0.73 (95% CI 0.60e0.90), respectively, for CVD.
Incident T2DM
Self-reported physical activity Men who engaged in regular physical exercise at least once a week had an RR of
questionnaire. Physical activity T2DM of 0.75 (95% CI 0.61e0.93) compared with men engaging in exercise less
data related to leisure and work often. Even vigorous activity only once a week at weekends was associated with
physical activity at week and
a reduced risk of T2DM. Men who engaged in vigorous activity at least once a
weekend days were collected. week at weekends had a multiple-adjusted RR of T2DM of 0.55 (95% CI 0.35
e0.88) compared with sedentary men.
All-cause mortality Self-reported physical activity
participation and health beliefs
as part of a detailed
questionnaire.
5.1 years
61,574 Non-Hispanic
White women, 5661
Black women, 2880
Hispanic women, 1340
American Indian
women, 2288 Asian/
Pacic Islander women,
aged 50e79 years at
baseline
Mathieu, The Dallas African-American, 3018 men and women 7 years
Heart Study (US, Hispanic and White (White 31%, AfricanAmerican 50%, Hispanic
2012) [38]
17%) aged between 18
and 65 years at baseline
Non-Hispanic
White, Black,
Hispanic, American
Indians and Asian/
Pacic Islander
All-cause mortality, Self-reported physical activity

CHD mortality and by questionnaire
cerbrovascular
disease mortality
(continued on next page)

323
First author, study

(country and
publication year)
324
Table 1 (continued )
Ethnic group
studieda
MexicanOttenbacher,
Americans
Hispanic
Established
Population for
the
Epidemiologic
Study of the
Elderly (US,
2012) [34]
Japanese
Rodriguez,
Honolulu Heart
Study (US, 1994)
[22]
Follow-up
Primary outcome
measure
948 men and women

aged 75 years at
baseline
3 years
All-cause mortality Self-reported physical activity

by questionnaire (Physical
Activity Scale for the Elderly).

years at baseline
Incident CHD

assessment
Compared to the lowest quartile of physical activity participants in quartiles, 2, 3

and 4 had mortality HRs of 0.35 (95% CI 0.24e0.51), 0.25 (95% CI 0.17e0.38) and
0.23 (95% CI 0.15e0.35), respectively. These values were attenuated slightly, but
remained signicant following adjustment for confounders.
Using the lowest physical activity index tertile for reference, RR for incident CHD
for the highest tertile of physical activity was 0.83 (95% CI 0.70e0.99). After
adjusting for age, hypertension, smoking, alcohol intake, diabetes, cholesterol,
and BMI, RR was 0.95 (95% CI 0.80e1.14). For CHD mortality, the age-adjusted
RR was 0.74 (95% CI 0.56e0.97) and 0.85 (95% CI 0.65e1.13) after risk factor
adjustment.
Self-reported questionnaire of Risk of developing T2DM 27% lower among those individuals that walk to work
minutes spent walking to work more than 21 min compare to those that walk less than 10 min per day
(p 0.007).
calculated from self-report of
24-h habitual physical activity
levels
Japanese
11,073 Japanese men

aged 40e55 years at
baseline
4 years
Incident T2DM
Native Hawaiians,
Japanese
Americans,
Caucasians
74,913 men and

women years (14%
Native Hawaiians, 47%
Japanese Americans
and 39% Caucasians)
aged 45e75 years at
baseline
8 years
Incident T2DM
Self-reported physical activity

questionnaire. Hours spent on
strenuous sport, vigorous work,
moderate activity during the
previous year were collected.
70,658 women aged 40 4.6 years

Incident T2DM

by questionnaire evaluating
regular physical activity and
sport participation over the
previous 5 years.
Villegas, Shanghai Chinese

Womens Health
Study (China,
2006) [20]
Main ndings
Wang, Shanghai
Mens Health
Study (China,
2013) [21]
Chinese
61,477 men aged 40

5.4 years
CVD and All-cause

mortality

by questionnaire evaluating
regular physical activity and
sport participation over the
previous 5 years.
Williams, Health
Survey for
England (UK,
2011) [44]
Indian, Pakistani/
Bangladeshi and
White British
13,293 White, 1244

Indian and 876
Pakistani/Bangladeshi
men and women,
aged35 years at
baseline
4 or 9 years
CHD mortality

over the previous 4 weeks by
questionnaire
HR for T2DM incidence for the highest quartiles of strenuous sport compared to
the lowest quartile on men was 0.80 (95% CI 0.72e0.88), 0.72 (95% CI 0.58
e0.89), 0.83 (95% CI 0.66e1.04) and 0.85 (95% CI 0.74e0.99) for combined ethnic
groups, Caucasian, Native Hawaiian and Japanese-American, respectively. A
similar trend was found for increasing quartiles of vigorous work (h/day) and
moderate activity (h/week), except for in Japanese-Americans in vigorous work
and for Hawaiians and Japanese-Americans in moderate activity. Women
showed a similar association for strenuous sport compared to men, the highest
compared to lowest quartile was associated with an HR of 0.67 (95% CI 0.57
e0.79), 0.54 (95% CI 0.38e0.77), 0.66 (95% CI 0.47e0.91) and 0.76 (95% CI 0.61
e0.95)) for combined ethnic groups, Caucasian, Native Hawaiian and JapaneseAmerican, respectively. No signicant associations with diabetes risk were
observed for either vigorous or moderate activity in women (except for
moderate activity in Caucasian women (HR 0.69 (95% CI 0.53e0.90)).
Leisure physical activity (LPA) and daily living physical activity (DPA) were
associated with a moderately reduced risk of T2DM. RR for T2DM by categories
of increasing LPA and DPA categories were 1.00, 0.89, 1.05, and 0.83, (p
trend 0.12) and 1.00, 0.98, 0.95, and 0.88, (p trend 0.06) respectively.
Occupational physical activity was not associated with T2DM risk in this
population, but commuting to work was associated with a reduction in risk.
After adjustment for potential confounders, men who exercised regularly had
an HR for total mortality of 0.80 95% (CI 0.74e0.87) compared with men who did
not exercise. The corresponding HRs were 0.80 (95% CI 0.72e0.89) for practising
Tai Chi, 0.77 (95% CI 0.69e0.86) for walking, and 0.73 (95% CI 0.59e0.90) for
jogging. Similar inverse associations were also found for cancer and
cardiovascular mortality.
Undertaking 30 min per week of moderate intensity physical activity
compared to none was associated with reduced risk of CHD mortality in White
(HR 0.48; (95% CI 0.34e0.66)) and South Asian (HR 0.29; (95% CI 0.11e0.78)). No
signicant interaction with ethnicity (p 0.79).
a
Ethnic group description uses terminology adopted by the authors of each paper. Abbreviations: CHD e coronary heart disease, CI e condence interval, CV e cardiovascular, CVD e cardiovascular disease, HR - hazard ratio,
OR e odds ratio, RR e relative risk, SD e standard deviation, T2DM e type 2 diabetes mellitus.
Sato, The Kansai

Healthcare Study
(Japan, 2007)
[29]
Steinbrecher, The
Multi-ethnic
Cohort Study
(US, 2012) [46]
Participants
between physical activity category (poor, intermediate or recommended level) and incidence of CVD, CHD and heart failure between Black and White groups [36]. In both White and Black
groups, intermediate levels of physical activity (i.e. more than zero,
but less than recommended levels) were associated with signicant
reductions in risk compared to no physical activity. Interestingly,
however, when physical activity level was categorised into quartiles, signicant physical activity by ethnicity interactions were
observed for CHD and CVD in women, and for CHD in both sexes
combined, with a greater protective effect observed in the Black
group [36]. Furthermore, a report from the Womens Health
Initiative Observational Study (WHIOS) indicated that the dosee
response relationship for reduction in cardiovascular events with
increasing physical activity was similar amongst White and Black
American postmenopausal women [37] and data from the Dallas
Heart Study indicated that risk of CVD mortality decreased with
increasing physical activity in African-American adults [38]. Analysis of data from different cohorts has revealed contrasting effects
of ethnicity on the relationship between physical activity and
stroke. In the NHANES I Epidemiologic Follow-up Study, the signicant reduction in stroke risk with increasing levels of physical
activity observed in White adults was not seen in Black adults,
which may be a consequence of insufcient statistical power to
detect the effect due to smaller number of Black participants within
the cohort [39]. However in the ARIC cohort, in analyses undertaken at multiple follow-up time-points, the protective effect of
physical activity against stroke incidence was more consistent in
Blacks than Whites [36,40,41]. Interestingly, the greater protective
effect of physical activity on ischaemic stroke in Blacks in the ARIC
cohort persisted in models adjusted for smoking, blood pressure,
anti-hypertensive medication use, left ventricular hypertrophy,
diabetes, lipids, brinogen, van Willebrand factor, white blood cell
count, waist-to-hip ratio, as well as demographic factors, suggesting that the greater benet of physical activity for ischaemic stroke
protection in Blacks was largely independent of effects on these
conventional risk factors [41]. Data from the US National Health
Interview Survey, found that the mortality risk reduction associated
with walking at least 2 h per week in adults with type 2 diabetes
was similar for White and non-White subgroups [42]. Prospective
data on physical activity and cardio-metabolic disease or mortality
risk in populations of South Asian origin are particularly limited. A
report from the Dutch SUNSET study investigated the effect of
baseline physical activity level on risk of subsequent for CVDrelated hospital discharge in middle-aged South Asian Surimamese, African Surimansese, and Dutch European adults over a
5.5-year follow-up period [43]. While an overall benet of physical
activity was observed in the group as a whole, the relatively small
sample size (n 1626 in total) and low event rate (178 total cases),
meant that when ethnicity-stratied analyses were undertaken,
the protective effect of physical activity against CVD-related hospital discharge was not statistically signicant for either the Dutch
European or South Asian Surimamese groups (but was for the African Surimansese group), and no signicant ethnicityephysical
activity interaction was observed [43]. A longitudinal analysis of the
1999 and 2004 Health Survey for England datasets (with 9 or 4
years of follow-up) found that undertaking at least 30 min.week1
of moderate intensity activity was associated with reduced risk of
CHD mortality in both White and South Asian adults, with no signicant interaction with ethnicity [44]. However, the dichotomous
nature of physical activity classication in this analysis means that
potential for interpretation of any ethnic differences in dosee
response relationship between physical activity and CHD in this
study is limited [44].
The most consistent evidence for differential effects of physical
activity on health outcomes between ethnic groups is for type 2
325
diabetes. Data from the WHIOS revealed that whilst higher levels of
physical activity were signicantly associated with reduced risk of
incident diabetes in White post-menopausal women, these ndings were not replicated in other ethnic groups, with no signicant
association being observed between physical activity level and
diabetes risk in African-American, Hispanic or Asian/Pacic
Islander women in multivariate-adjusted analysis [45]. This may
reect lack of statistical power to detect a relationship in the nonWhite groups, in which the cohort sizes were more than 10-fold
smaller than the White group; ethnic differences in the selfreporting of actual physical activity levels and/or incident diabetes; and/or real ethnic differences in the doseeresponse relationship between physical activity and diabetes risk. Data from The
Multiethnic Cohort was also suggestive of ethnic differences in the
doseeresponse relationship between physical activity and type 2
diabetes risk. In both men and women, there was a signicant inverse association between increasing levels of moderate activity
and diabetes risk in Caucasians (Whites), but not in Native Hawaiians or Japanese-Americans, with signicant ethnic interaction
effects [46]. Whilst engaging in strenuous sports was inversely
associated with diabetes risk in men and women from all three
ethnic groups, there was a tendency for the protective effect of
strenuous sports to be greater in Caucasian men, than in men from
the other two ethnicities (p 0.07 for ethnicity interaction) [46]. In
addition, data from the Strong Heart Study show higher levels of
physical activity are associated with reduced risk of incident diabetes in American Indians [47], but there was no direct comparison
with other ethnic groups in this cohort. Thus, there is emerging
evidence from the limited data available, that the doseeresponse
relationship between physical activity and risk of type 2 diabetes
may not be the same across all ethnic groups, with the level of
physical activity needed to gain benet potentially being higher in
non-white populations. One clear limitation to the available evidence, is the lack of data comparing the doseeresponse relationship for physical activity and diabetes risk between Europeans and
South Asians. Interestingly, diabetes risk is elevated compared to
white European groups, in most non-White ethnic groups, particularly when they adopt Westernised or urbanised lifestyles (see
section 6 below), and it may be that a higher threshold for physical
activity level is needed in these groups to obtain a low level of
absolute diabetes risk. Thus, currently recommended levels of
physical activity, based on the available epidemiological evidence
base of largely White European ethnicity cohorts, may not necessarily be appropriate for all ethnic groups.
4. Lifestyle intervention trials for cardiovascular disease
prevention
The randomised controlled trial (RCT) represents the gold
standard for evaluating the efcacy and effectiveness of an intervention on a health outcome. Given the low CVD event rate in the
general population, it is not feasible to perform RCTs to evaluate the
effectiveness of physical activity (either alone or in the context of a
wider lifestyle intervention trial) for the primary prevention of
CVD events in the general population (in order to have sufcient
power such a trial would need to be too large and too long to be
logistically viable). Patients with type 2 diabetes have increased
risk of CVD compared to the general population [48], so form a
group in which a lifestyle intervention RCT could conceivably
demonstrate benet in terms of reduced cardiovascular event rate.
However, the recent Look AHEAD trial in which 5145 overweight or
obese patient with type 2 diabetes were randomised to intensive
lifestyle intervention (physical activity, dietary modication and
weight loss) or diabetes support education (control), was stopped
early (after 9.6 years of the planned 13.5-year follow-up period)
326
after an interim analysis revealed that it was unlikely for a significant difference between groups in the primary endpoint (composite of death from CVD, non-fatal myocardial infarction, non-fatal
stroke and hospitalisation for angina) to be seen by the planned
end-date [49]. Potential reasons for this include a lower than expected CVD event rate reducing power to detect any differences
between groups; the relatively intensive intervention in the control
group (which was comparable in intensity the intervention group
in many lifestyle intervention trials) [50] that would act to attenuate any potential differences between groups; diminishing
compliance to the intervention over time in the intervention group
e differences in tness and weight between the intervention and
control groups decreased steadily from a peak difference at 1-year
as the trial progressed; and the greater use of diabetes, antihypertensive and lipid-lowering drugs over the course of the trial
in the control group [49]. At the time the trial was stopped, the
hazard ratio for a CVD event in the intervention compared to
control group was 0.95 (95%CI 0.83e1.09) overall, 0.94 (0.80e1.11)
in Whites, 1.34 (0.91e1.96) in Blacks, 0.71 (0.06e8.28) in Asians or
Pacic Islanders, 0.66 (0.41e1.05) in Hispanics and 0.74 (0.21e1.76)
in Native Americans. There was no signicant interaction with
ethnicity (p 0.17), but power to detect any ethnicityeintervention
interaction effect, would have been particularly limited given the
absence of a signicant effect of the intervention overall. Although
the Look AHEAD intervention did not signicantly impact on CVD
events, rates of full (reversion to normoglycaemia without diabetes
medication) and partial (reversion to pre-diabetes glycaemia levels
without diabetes medication) remission of type 2 diabetes were 3e
6 fold greater over the course of the trial in the intervention
compared with the control group [51]. In addition, the intervention
reduced HbA1c concentrations and improved cardiorespiratory
tness levels [51], with the change in tness level at 4-years being
associated with change in HbA1c, independent of a range of confounders including change in weight [52]. A study 20-year followup data (i.e. continuing for 14 years after completion of the 6-year
trial) from the Da Qing Study, a diabetes-prevention lifestyle
intervention trial in Chinese adults with impaired glucose tolerance, reported 20-year hazard rate ratio for CVD death of 0.83 in the
intervention compared to the control group, although the 95% CI
included 1 (0.48e1.40) [53]. However, it is important to note that
the Da Qing Study was a diabetes prevention trial, rather than a
CVD mortality prevention trial, and thus was not designed to have
sufcient power to detect an intervention effect for this outcome.
Interestingly, separation of CVD death rates between control and
intervention groups did not start to become apparent until 10-years
of follow-up, (indeed reduction in hazard rate ratio for CVD death
numerically larger at 0.73 (95% CI 0.42e1.26) when only the 14-year
post-trial period was considered), which raises the possibility that
the Look AHEAD trial, which was stopped after 9.6 years of followup, may have been too short for a clear effect of lifestyle intervention on CVD event rates to have become evident. Event rates are
much higher in secondary CVD prevention settings, making RCTs
evaluating the effects of physical activity on hard CVD endpoints
more feasible. Meta-analyses of RCTs of exercise-based cardiac
rehabilitation programmes in patients post-myocardial infarction
indicate signicant reductions in all-cause, CHD and CVD mortality
in patients in the exercise intervention groups [54,55]. Thus, in
secondary prevention settings, where CVD event rates are high
enough for clear intervention effects to be seen, RCTs reveal clear
benets of exercise in reducing CVD mortality risk.
It is, however, important to note that despite the equivocal nature of the end-point trial evidence to date, there is clear and
consistent experimental evidence to demonstrate that increasing
physical activity positively inuences a number of mechanistic
pathways and bio-markers known to play a causal role in the
atherosclerotic and metabolic disease process, including favourable

alterations to the lipid and lipoprotein prole [56e58], improvements in insulin sensitivity [59,60], improvements in endothelial
function [61e63], reductions in blood pressure [64e66], and reductions in systemic inammation [67,68], which supports the
large body of observational epidemiological evidence indicating
that high levels of physical activity are associated with reduced
levels of cardio-metabolic disease risk. Nevertheless, given the
experiences with the Look AHEAD trial, which had estimated costs
of over a quarter billion US Dollars [69], it seems unlikely that a
denitive RCT to determine the effects increasing physical activity
on prevention of hard CVD outcomes in a primary prevention
setting will ever be successfully completed.
5. Lifestyle intervention trials for diabetes prevention
Unlike CVD, where progression rate to events is relatively low in
primary prevention settings, even in groups regarded as having
elevated risk (for example, a 20% 10-year risk of a CVD event, puts
someone in a high-risk category [70]), it is possible to easily identify
individuals with the pre-diabetic conditions of impaired glucose
tolerance (IGT) and impaired fasting glucose (IFG), where progression rates to diabetes are high (w3e18% per year [2]). This
higher event rate and faster disease progression means that undertaking RCTs with sufcient power to evaluate the effects of
lifestyle intervention on diabetes incidence are much more feasible
than undertaking lifestyle intervention RCTs evaluating CVD outcomes. Consequentially, over the last two decades, there have been
a substantial number of major RCTs which have demonstrated that
lifestyle intervention (increased physical activity and dietary
change, usually coupled with weight loss) is effective at reducing
incidence of diabetes in individuals with IFG/IGT (reviewed in Ref.
[2]). Recently published reports of long-term follow-up data from
the Chinese Da Qing study [53], the Finnish Diabetes Prevention
Study (DPS) [71] and the US Diabetes Prevention Program Outcomes Study [72], have shown that the protective effect of lifestyle
intervention in reducing incident diabetes persists for at least 10e
20 years, indicating clear long-term benet. Importantly, there is
evidence that lifestyle intervention is effective at preventing
microvascular complications of diabetes as well as onset of the
disease. Data published from the Da Qing study showed a 47%
reduction in incidence of diabetic retinopathy in the lifestyle
intervention compared to control group over 20 years of follow-up
[73].
The protective effect of lifestyle intervention against incident
diabetes has been demonstrated in a number of different ethnic
groups including Chinese [53,74], Japanese [75] and Indians [76] in
trials containing single ethnic groups; in White Europeans, AfricanAmericans, Hispanics, American Indians, and Asians/Pacic Islanders within the multi-ethnic US Diabetes Prevention Program
(DPP) [77]; and in White Europeans in a number of trials [2,78].
Although differences in the protocols for lifestyle intervention and
compliance of participants to the interventions makes direct
comparison of intervention effect sizes between studies difcult,
the magnitude of diabetes risk reduction, at w28e67% has been
broadly comparable between trials. Risk reduction in the Indian
Diabetes Prevention Programme (IDPP), at 28% [76], was at the
lower end of this range. Potential explanations for this smaller
relative reduction in risk include the fact that this was one of the
few trials in which participants did not lose weight on the intervention and the fact that the absolute rate of progression to diabetes, at 18.3% per year in the Control group, was approximately
twice as high in this study than the other major diabetes prevention
trials (e.g. 11.0% per year in the DPP [77] and 7.8% per year in the DPS
[79]). The latter point is an important, but subtle, one: the
reduction in absolute rates of progression to diabetes with lifestyle

intervention in the IDPP (reduction of 52 cases per 1000 personyears) [76] was comparable with reductions seen in the DPS
(reduction of 46 cases per 1000 person-years [79]) and DPP
(reduction of 62 cases per 1000 person-years [77]). In the DPP,
which was the one trial which had large enough numbers of participants from different ethnic groups to allow the effects of lifestyle intervention to be separately evaluated, the reduction in
diabetes incidence rates were comparable between White (reduction of 51%; 51 cases per 1000 person-years), African-American
(reduction of 61%; 73 cases per 1000 person-years), Hispanic
(reduction of 66%; 75 cases per 1000 person-years), American Indian (reduction of 65%, 82 cases per 1000 person-years) and Asians/
Pacic Islander (reduction of 71%, 83 cases per 1000 person-years)
ethnic groups [77].
Thus, there is good evidence that lifestyle intervention is
comparably effective at preventing diabetes in individuals with IFG/
IGT across a range of ethnic groups. However, it is important to
recognise that individuals with IFG/IGT are at high risk of diabetes
irrespective of ethnic group. They progress to frank diabetes at the
rate of w3e18% per year without intervention, and even with
intensive lifestyle intervention progress to diabetes at the rate of
w2e13% per year [2]. Thus, an argument can be made that it may
be too late to primarily focus on individuals with IFG/IGT in efforts
for diabetes prevention and earlier intervention with lifestyle
intervention for diabetes prevention in groups at increased diabetes risk, but who are still normoglycaemic, is warranted. It is also
important to recognise that, with the exception of the IDPP in
which rates of progression to diabetes were higher [76], rates of
progression to diabetes in the control arms of the various diabetes
prevention trials of patients with IFG/IGT were broadly similar
across the different ethnic groups [74,77]. Thus, when an individual
progresses to IFG/IGT, they are at high risk of diabetes, irrespective
of ethnic group. This, however, is not reective of the pattern of
diabetes risk amongst the general population where there are large
differences in risk between different ethnic groups, for example
diabetes risk for South Asians living in the UK is three to vefold
higher than the background White European population [80,81]
(see Section 6 below). Thus to develop IFG/IGT a European has
progressed further along the diabetes risk continuum compared to
the average European in the population, than a South Asian with
IFG/IGT has progressed compared to the average South Asian in the
population. Accordingly, ndings from studies of patients with IFG/
IGT from different ethnic groups cannot necessarily be extrapolated
to guidelines for diabetes prevention for different ethnic groups
within the normoglycaemic general population. The optimal level
of physical activity for diabetes prevention may differ between
ethnic groups with widely varying levels of diabetes risk.
6. Ethnicityeenvironment interactions and diabetes risk
There are clear differences in risk of type 2 diabetes between
ethnic groups and while prevalence of diabetes is increasing
worldwide, the rate of differs widely between regions [82]. This can
be partly attributed to differences in urbanisation and obesity [83],
but differences in environment alone do not appear to tell the
whole story, particularly in elucidating why certain populations
and ethnic groups experience a disproportionately higher prevalence of type 2 diabetes when they adopt urbanised lifestyles with
greater energy-dense food availability and low levels of physical
activity. There is a growing body of evidence that the adverse effects of urbanisation on diabetes risk are substantially larger in
non-White populations than in populations of White European
descent. For example, diabetes prevalence in South Asian adults
living in the United Kingdom and North America is 12e26% [81,84e
327
86] compared to 4e8% amongst the UK and US populations of

white European descent [82,85,87]. However, diabetes prevalence
in rural India is low, at w3% (0.7% in active rural dwellers without
abdominal obesity) [88], rising to 7e18% in the cities [88,89]. Thus,
when lean and active and living in rural environments, South
Asians have low diabetes risk but adoption of an urban or Westernised lifestyle appears to have a disproportionately large inuence on diabetes risk in South Asians, compared to those of White
European descent. A similar pattern is observed when comparing
White populations to populations of Black African descent. Diabetes prevalence amongst African-Americans is about twice as high
as for White Americans for a given BMI [87], and similarly in the UK
diabetes prevalence is more than twice as high amongst the Black
population compared to the background White population [81]. In
contrast diabetes prevalence in African countries e where there is
limited urbanisation e is low [82]. In the Multiethnic Cohort Study
of residents of California and Hawaii in the US (i.e. different ethnic
groups sharing a similar Westernised environment) the prevalence
odds ratio for diabetes, in age, sex and BMI-adjusted models was
greater than two for all non-White ethnic groups (Native Hawaiians, Latinos, African-Americans, Japanese) compared to White
Europeans [90].
This disproportionately large adverse effect of urbanisation on
diabetes risk is also observed in a number of indigenous populations throughout the Americas and Australasia [91]. The classic
example is the Pima Indians, who have relatively low diabetes
prevalence at 6.9% when living rurally in Mexico (which was not
signicantly different from the rate in non-Pima Mexicans), which
rises more than vefold to 38% amongst Pima Indians living in the
US [92,93]. It is of particular note that reported energy intake did
not differ signicantly between the Mexican and US Pimas, but
Mexican Pima men and women reported w2.5-fold and w7-fold
higher physical activity levels than the US Pima counterparts
[92,93]. This large difference in physical activity level was veried
by using doubly-labelled water measurement in a sub-group of the
Mexican and US Pima cohorts. Activity energy expenditure levels in
Mexican Pimas were over 500 kcal/day higher than their US Pima
counterparts [94]. Thus, Pima Indians appear to be able to overcome their clear genetic susceptibility to diabetes [95] by engaging
in very high levels of physical activity e levels of activity much
higher than are needed to confer low diabetes risk in populations of
White European descent.
We have recently reported similar observations for the Mapuches, an indigenous American-Indian group living in Chile who
have low levels of diabetes when they live in traditional rural environments, which rises substantially when they adopt an urbanised lifestyle [96e99]. We made detailed phenotypic
measurements (including a cardiorespiratory tness test, objective
assessment of physical activity by accelerometer, dietary assessment, body composition measurement, and assessment of a range
of cardio-metabolic risk biomarkers) in 123 Mapuche adults living
in traditional rural environments, 124 Mapuche adults living in an
urban environment, 91 Chileans of European descent living rurally
and 134 Chileans of European descent living in an urban setting
[100]. Our principal ndings were twofold. Firstly, there was a
signicant ethnicityeenvironment interaction for insulin resistance (as estimated by HOMAIR [101]) (p 0.0003): the adverse
effect of urbanisation on this biomarker of cardio-metabolic risk
was over fourfold greater in Mapuches than in Chileans of European
origin [100]. Secondly, the adverse effects of low levels of physical
activity and high levels of time spent sedentary were substantially
greater in Mapuches than European Chileans (p 0.0001 for both
for ethnicity interactions), an observation that persisted after
adjustment for several potential confounders (see Fig. 1). Thus, it
appears that Mapuches are protected from insulin resistance when
Europeans
Mapuches
HOMAIR
6
4
p(interaction) = 0.0001
2
0
Lower
Middle
Upper
Tertiles of Sedentary Time
8
HOMAIR
p(interaction) = 0.0001
4
2
0
Lower
Middle
Upper
Tertiles of Physical Activity
Fig. 1. Homoeostasis model assessment-estimated insulin resistance (HOMAIR) in

Chilean adults of Mapuche or European ethnic origin according to level of sedentary
time (top panel) and level of physical activity (bottom panel). Bars show mean SEM.
p(interaction) describes the ethnicity physical activity/sedentary time interaction for
HOMAIR after adjustment for potential confounding covariates (i.e. age, sex, environment (rural or urban), socio-economic level, education level, smoking status, BMI,
waist, body fat, accelerometer wear time, tness, energy intake, and physical activity
(for sedentary time relationship) or sedentary time (for physical activity relationship)).
Modied from Ref. [100].
they are very active, but the consequences of low levels of physical
activity and high levels of sedentary behaviour for them are
particularly large. In other words, they need to undertake greater
levels of physical activity than their counterparts of European
origin to maintain a favourable metabolic risk prole.
Thus, the interaction between innate susceptibility and environmental inuence on diabetes risk and insulin resistance appears
to be much steeper most non-White compared with White European populations. Accordingly, it could be argued that people of
White European descent (w15% of the worlds population) are in
fact the outliers with respect to the role of environmental inuences on diabetes and metabolic risk. Given the fact that the
global burden of diabetes is increasingly falling on non-White
populations, understanding of how environmental factors, such
as physical activity, modulate diabetes risk in non-White ethnic
groups is of pressing global public health concern.
7. Can physiological differences between ethnic groups
explain the ethnic differences in diabetes risk?
Of the major worldwide ethnic groups, South Asians, comprising
w20% of the worlds population, are amongst those with the
highest susceptibility to type 2 diabetes, particularly when they
migrate to Westernised countries [81,84e86]. When living in the
UK, South Asians tend to develop diabetes about a decade earlier in
life at lower levels of BMI than White Europeans [80,81] and nondiabetic South Asians in the general UK or US population have
higher glucose levels and are more insulin resistant than Europeans
[102e105]. A number of hypotheses have been proposed to explain
South Asians increased insulin resistance and diabetes risk
(reviewed in detail here [106]), including greater adiposity for a
given BMI, with more central fat distribution [84,104,107], and
potentially greater liver fat accumulation [108]; early origins/thrifty
phenotype effects [109e111]; genetic factors including potential
differences in the prevalence and effects of diabetes risk alleles in

South Asian compared to other populations [112,113]; differences in
lifestyle characteristics, particularly lower physical activity level
[114e117]; and differences in cardiorespiratory tness and skeletal
muscle metabolism [105,118]. South Asians increased adiposity has
received considerable focus as a potential explanatory factor for
their increased insulin resistance and diabetes risk. However, South
Asians remain more insulin resistant than Europeans after matching or adjustment for a range of adiposity markers [105,119], indicating other factors also contribute to the South Asian insulin
resistance phenotype. A number of studies have reported that
South Asians living in Western countries are less physically active
than Europeans. This has been observed in studies using both
objective [118,120,121] and self-report [114e117] measures of
physical activity and is evident throughout the lifecourse, starting
in childhood [121] and continuing throughout adulthood into
middle-age and older [114e118], and has even been observed a
study that objectively measured physical activity levels during
pregnancy [120]. Furthermore, while these lower levels of physical
activity are likely to contribute to the greater insulin resistance and
diabetes risk in South Asians, this does not explain the whole story,
South Asians remain more insulin resistant than Europeans even
after adjustment for differences in physical activity level [105,118].
One particularly interesting observation is that cardiorespiratory
tness levels are lower in South Asians than comparable groups of
Europeans [105,118,122,123]. This is a consistent nding that is not
explained by South Asians lower levels of physical activity. While
tness increases with increasing levels of physical activity in both
Europeans and South Asians, cardiorespiratory tness levels are
lower in South Asians than Europeans for any given level of
objectively measured physical activity (Fig. 2). This nding of innate
ethnic differences in tness levels is supported by recent data from
the US NHANES study, which showed systematic differences in
tness levels between Mexican Americans (highest), non-Hispanic
whites and non-Hispanic blacks (lowest) groups in the US, independent of the level of (self-reported) physical activity and demographic factors [124]. The heritable contribution to
cardiorespiratory tness has been estimated to be up to 50%
[125,126] and thus, it seems likely that there will be a genetic (and/
or early origins) contribution to the lower tness levels observed in
South Asians
65
Europeans
VO2max (ml kg-1 min-1)
328
55
45
35
25
15
0
20
40
60
80
100
120
Moderate-to-vigorous physical activity

(min.day-1)
Fig. 2. Relationship between cardiorespiratory tness (as assessed by maximal oxygen
uptake; VO2max) and objectively-measured physical activity in South Asian and European men. Dotted regression line describes the relationship in European men; solid
regression line describes the relationship in South Asian men. For all levels of physical
activity, VO2max was signicantly lower in South Asians. Modied from Ref. [118].
8. Can ethnicespecic physical activity recommendations

help to overcome innate differences in diabetes risk between
ethnic groups?
The concept of ethnicity-specic health guidelines is not a new
one. For example, recent consensus statements have recommended
ethnicity-specic cut-points for obesity, suggesting that the conventional threshold of BMI 30 kg.m2 is lowered to BMI 25 kg.m2
in South Asian populations [134,135]. This was based on published
data indicating that South Asians required BMI values of w21e
23 kg.m2 to have equivalent glycaemia, and w22e26 kg.m2 to
have equivalent lipid proles, to those seen in Europeans with BMI
30 kg.m2 [103,136]. However, the concept of ethnicityespecic
physical activity guidelines, which account for differences in the
cardio-metabolic risk prole between different ethnic groups has
not yet been considered in depth. A recent consensus statement on
Fat oxidation (mg.kg-1.min-1)
European
South Asian
6
4
2
p(ethnicity) = 0.001
0
45
50
55
60
65
Exercise intensity (%VO2max)
Fat oxidation @ 55% VO2max

(mg.kg-1.min-1)
12
10
8
6
4
r = 0.423
p = 0.011
2
0
1
2
3
4
5
Square-root insulin sensitivity index
12
Fat oxidation @ 55% VO2max
(mg.kg-1.min-1)
South Asians. This innate low tness in South Asians may

contribute to their increased diabetes risk. We demonstrated that
statistical adjustment for the difference in tness between South
Asian and European men reduced the excess insulin resistance
observed in South Asians by over two-thirds [118]. This is supported by data from other studies that have reported strong associations between cardiorespiratory tness and insulin sensitivity
[127] or between change in cardiorespiratory tness with exercise training and change in insulin sensitivity [128]. There is clear
epidemiological evidence that low tness levels are associated with
increased diabetes risk [129,130], and animal model data shows
articial selection solely for the trait of low tness in rats induces
insulin resistance and a metabolic syndrome-like phenotype [131].
Further evidence of innate physiological differences between South
Asians and Europeans contributing to differences in insulin resistance and diabetes risk, come from our data indicating that rates of
fat oxidation during exercise (which primary reects muscle
metabolism) in South Asian men was about 50% lower than in
European men at any given exercise intensity (Fig. 3), and that the
rate of fat oxidation during exercise was signicantly associated
with insulin sensitivity both at the whole body level and the level of
insulin signalling within skeletal muscle (Fig. 3) [105]. Further
study is needed to replicate these ndings and conrm whether
they extend to women, but as the only way to improve tness is to
increase physical activity and one of the key adaptations to exercise
training is increased ability of muscle to oxidise fat during exercise
[132], these data suggest that South Asians may need to undertake
higher levels of physical activity than Europeans to overcome
innate physiological differences which contribute to South Asians
greater insulin resistance. The concept of a low tness phenotype,
where individuals with low intrinsically low tness and oxidative
capacity might also choose to be more inactive because they nd
physical activity more difcult, thus compounding their innate
metabolic disadvantage has postulated in the literature [133]. The
consistent reports of lower levels of physical activity in South
Asians compared to Europeans [114e118,120,121], together with
data demonstrating that South Asians have lower levels of cardiorespiratory tness [105,118,122,123] and lower capacity for fat
oxidation during exercise [105], suggests that a low tness
phenotype is likely to play a role in explaining the increased cardiometabolic disease risk observed in South Asian populations living in
urbanised and Westernised environments. There is a clear challenge to increase physical activity levels in groups with this
phenotype, who have the double burden of needing to undertake
more physical activity than others to minimise risk of cardiometabolic disease and nding it harder to become more active,
but identifying appropriate physical activity targets for such groups
represents an important rst step in this process.
329
10
r = 0.475
p = 0.025
8
6
4
2
0
0.5
1
1.5
2
2.5
Log PKB Ser473 phosphoryation
Fig. 3. Top panel: Rate of fat oxidation during incremental submaximal exercise in
South Asian and European men. Bars show mean SEM. p(ethnicity) describes maineffect difference in fat oxidation between Europeans and South Asians. Middle and
bottom panels: Relationship between fat oxidation during exercise and insulin
sensitivity index [141] (Middle panel) and skeletal muscle protein kinase B (PKB)
phosphorylation at Ser473 (Bottom panel) in South Asian and European men. r- and pvalues for combined South Asian and European group. All panel modied from
Ref. [105].
Physical Activity Guidelines for Asian Indians recommended that

Indians undertake 30 min of moderate-intensity physical activity,
15 min of work-related activity and 15 min of muscle strengthening
exercises with light weights every day [137], which differs slightly
from other national and international physical activity guidelines
(150 min of moderate intensity or 75 min of vigorous intensity
physical activity per week [13e15]), but the evidence base underpinning this specic recommendation is unclear. A recent report in
Sri Lankan women found that women with newly diagnosed dysglycaemia undertook less self-reported physical activity than normoglycaemic women [138]. Based on this they used Receiver
Operating Characteristics (ROC) analysis to determine physical activity cut-off values for dysglycaemia, reporting that women undertaking less than 1435 MET-minutes per week of walking
330
(w400 min per week) and less than 2640 MET-minutes per week of
moderate and vigorous activity (w400e650 min of moderate intensity physical activity per week) were four times as likely to be
dysglycaemic compared to women exceeding these physical activity thresholds [138]. The self-reported nature of physical activity
measurement in this study is a clear limitation, with evidence that
this can lead to a 4e10-fold overestimation of physical activity
compared with objective measures [6,139] and lead to a misrepresentation of the true dose response relationship between physical activity and risk [6]. Nevertheless, it important to note that the
level of physical activity needed to avoid dysglycaemia in these
South Asian women was substantially higher than level of physical
activity recommended in current guidelines [13e15,137]. In an
attempt to work toward provision of accurate evidence-based
guidelines for physical activity in South Asians, we recently used
an approach adopted by others for the calculation of South Asianspecic obesity cut-points [103,136]. In 75 non-diabetic middleaged South Asian men and 83 age and BMI-matched European men
living in Scotland, we derived individual summary cardiometabolic risk factor scores based on glycaemia, insulin resistance, lipid and blood pressure biomarkers and constructed ageadjusted regression models describing the relationship between
objectively measured moderate-intensity physical activity and
cardio-metabolic risk factor score in the two ethnic groups [140].
We then determined predicted values for the cardio-metabolic risk
factor score in European men undertaking 150 min.week1 of
moderate-intensity physical activity (i.e. the currently recommended level) and calculated the level of physical needed in South
Asians to provide the same cardio-metabolic risk factor score [140].
Our analysis suggested that South Asian men need to undertake
more than 250 min.week1 of moderate-intensity physical activity
to confer equivalent risk to European men achieving current
physical activity guidelines (Fig. 4). Further study is needed to
replicate these ndings, ideally prospectively, in larger cohorts and
to extend these analyses to women and other ethnic groups, and
debate is required at policy level to determine the best approach for
messaging this new evidence. However, this work does provide an
Cardio-metabolic risk factor score

(SD units)
0.6
150 min.week-1
in Europeans
0.4
266 min.week-1
in South Asians
0.2
0.0
-0.2
-0.4
-0.6
-0.8
50
100
150
200
250
300
350
Moderate-intensity physical activity (min.week -1)

Fig. 4. Relationship between the cardio-metabolic risk factor score and objectively
measured moderate-intensity physical activity in South Asians (solid red line) and
Europeans (solid blue line). 266 min week1 of moderate-intensity physical activity in
South Asians gave an equivalent cardio-metabolic risk factor level to that observed in
Europeans undertaking 150 min.week1 of moderate-intensity physical activity (i.e.
the amount of physical activity currently recommended in guidelines). Dotted red lines
represent the 95% condence bands around the regression line for South Asians.
Regression lines are adjusted for age, BMI, daily accelerometer wear time, and number
of days of accelerometer wear. Modied from Ref. [140].(For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of
this article.)
important rst step towards the provision of ethnicity-specic

physical activity guidelines. Such quantication of appropriate
levels of physical activity in non-White ethnic groups will
contribute to the optimisation of lifestyle interventions in these
populations to help overcome the growing challenge of reducing
the growing burden of cardio-metabolic disease in non-White
populations throughout the world.
9. Conclusions
While there is a large body of evidence that physical activity
plays an important protective role against cardio-metabolic disease
risk in populations of White European origin, data on non-White
populations are more limited. The available evidence suggests
that non-White populations do also benet from increased physical
activity levels, but it seems likely that the optimal dose of physical
activity required to achieve low cardio-metabolic disease risk differs according to ethnic group. More study is needed to understand
more about how and why the doseeresponse relationship between
physical activity and disease risk differs by ethnicity. This will
require inclusion of more non-White populations in prospective
cohort studies of the relationship between physical activity (preferably objectively measured) and cardio-metabolic disease outcomes; intervention trials comparing the effects of multiple levels
of physical activity on health outcomes (or biomarkers of health
outcomes) across ethnic groups; and more laboratory studies
investigating the mechanisms underpinning ethnicity-related differences in physiological and metabolic to physical activity. There is
also a clear need for further study to determine the best approaches
to promote physical activity in ethnic groups (such as South Asians)
who e despite physiologically needing to undertake higher levels
of physical activity to minimise cardio-metabolic risk e appear to
undertake less physical activity than White Europeans when they
live in Westernised countries.
References
[1] Warburton DE, Charlesworth S, Ivey A, Nettlefold L, Bredin SS. A systematic
review of the evidence for Canadas physical activity guidelines for adults. Int
J Behav Nutr Phys Act 2010;7:39. http://dx.doi.org/10.1186/1479-5868-7-39.
1479-5868-7-39 [pii].
[2] Gill JM, Cooper AR. Physical activity and prevention of type 2 diabetes
mellitus. Sports Med 2008;38:807e24.
[3] Nocon M, Hiemann T, Muller-Riemenschneider F, Thalau F, Roll S, Willich SN.
Association of physical activity with all-cause and cardiovascular mortality: a
systematic review and meta-analysis. Eur J Cardiovasc Prev Rehabil 2008;15:
239e46.
http://dx.doi.org/10.1097/HJR.0b013e3282f55e09.
00149831200806000-00001 [pii].
[4] Jeon CY, Lokken RP, Hu FB, van Dam RM. Physical activity of moderate intensity and risk of type 2 diabetes: a systematic review. Diabetes Care
2007;30:744e52. http://dx.doi.org/10.2337/dc06-1842. 30/3/744 [pii].
[5] So F, Capalbo A, Cesari F, Abbate R, Gensini GF. Physical activity during
leisure time and primary prevention of coronary heart disease: an updated
meta-analysis of cohort studies. Eur J Cardiovasc Prev Rehabil 2008;15:247e
57. http://dx.doi.org/10.1097/HJR.0b013e3282f232ac.
[6] Celis-Morales CA, Perez-Bravo F, Ibanez L, Salas C, Bailey ME, Gill JM.
Objective vs. self-reported physical activity and sedentary time: effects of
measurement method on relationships with risk biomarkers. PLoS ONE
2012;7:e36345. http://dx.doi.org/10.1371/journal.pone.0036345. PONE-D12e06438 [pii].
[7] Shephard RJ. Limits to the measurement of habitual physical activity by
questionnaires. Br J Sports Med 2003;37:197e206.
[8] Blair SN, Cheng Y, Holder JS. Is physical activity or physical tness more
important in dening health benets? Med Sci Sports Exerc 2001;33:S379e
99.
[9] Sedentary Behaviour RN. Letter to the editor: standardized use of the terms
sedentary and sedentary behaviours. Appl Physiol Nutr Metab 2012;37:
540e2. http://dx.doi.org/10.1139/h2012-024.
[10] Edwardson CL, Gorely T, Davies MJ, et al. Association of sedentary behaviour
with metabolic syndrome: a meta-analysis. PLoS ONE 2012;7:e34916. http://
dx.doi.org/10.1371/journal.pone.0034916. PONE-D-11e16815 [pii].
[11] Wilmot EG, Edwardson CL, Achana FA, et al. Sedentary time in adults and the
association with diabetes, cardiovascular disease and death: systematic
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
review and meta-analysis. Diabetologia 2012;55:2895e905. http://

dx.doi.org/10.1007/s00125-012-2677-z.
Thorp AA, Owen N, Neuhaus M, Dunstan DW. Sedentary behaviors and
subsequent health outcomes in adults a systematic review of longitudinal
studies, 1996e2011. Am J Prev Med 2011;41:207e15. http://dx.doi.org/
10.1016/j.amepre.2011.05.004. S0749e3797(11)00312-6 [pii].
Department of Health. Start active, stay active: a report on physical activity
for health from the four home countries Chief Medical Ofcers; 2011. pp. 1e
59.
U.S.Department of Health and Human Services. 2008 physical activity
guidelines for Americans; 2008.
World Health Organisation. Global recommendations on physical activity for
health; 2010. pp. 1e58.
Lee IM, Shiroma EJ, Lobelo F, Puska P, Blair SN, Katzmarzyk PT. Effect of
physical inactivity on major non-communicable diseases worldwide: an
analysis of burden of disease and life expectancy. Lancet 2012;380:219e29.
http://dx.doi.org/10.1016/S0140-6736(12)61031-9. S0140e6736(12)610319 [pii].
Blair SN. Physical inactivity: the biggest public health problem of the 21st
century. Br J Sports Med 2009;43:1e2. 43/1/1 [pii].
Lan TY, Chang HY, Tai TY. Relationship between components of leisure
physical activity and mortality in Taiwanese older adults. Prev Med 2006;43:
36e41. http://dx.doi.org/10.1016/j.ypmed.2006.03.016. S0091e7435(06)
00137-X [pii].
Matthews CE, Jurj AL, Shu XO, et al. Inuence of exercise, walking, cycling,
and overall nonexercise physical activity on mortality in Chinese women.
Am J Epidemiol 2007;165:1343e50. http://dx.doi.org/10.1093/aje/kwm088.
kwm088 [pii].
Villegas R, Shu XO, Li H, et al. Physical activity and the incidence of type 2
diabetes in the Shanghai womens health study. Int J Epidemiol 2006;35:
1553e62.
Wang N, Zhang X, Xiang YB, et al. Associations of Tai Chi, walking, and
jogging with mortality in Chinese men. Am J Epidemiol 2013. http://
dx.doi.org/10.1093/aje/kwt050. kwt050 [pii].
Rodriguez BL, Curb JD, Burchel CM, et al. Physical activity and 23-year
incidence of coronary heart disease morbidity and mortality among
middle-aged men. The Honolulu Heart Program. Circulation 1994;89:2540e
4.
Donahue RP, Abbott RD, Reed DM, Yano K. Physical activity and coronary
heart disease in middle-aged and elderly men: the Honolulu Heart Program.
Am J Public Health 1988;78:683e5.
Hakim AA, Curb JD, Petrovitch H, et al. Effects of walking on coronary heart
disease in elderly men: the Honolulu Heart Program. Circulation 1999;100:
9e13.
Abbott RD, Curb JD, Rodriguez BL, et al. Age-related changes in risk factor
effects on the incidence of thromboembolic and hemorrhagic stroke. J Clin
Epidemiol 2003;56:479e86. S089543560200611X [pii].
Abbott RD, Rodriguez BL, Burchel CM, Curb JD. Physical activity in older
middle-aged men and reduced risk of stroke: the Honolulu Heart Program.
Am J Epidemiol 1994;139:881e93.
Burchel CM, Sharp DS, Curb JD, et al. Physical activity and incidence of
diabetes: the Honolulu Heart Program. Am J Epidemiol 1995;141:360e8.
Hakim AA, Petrovitch H, Burchel CM, et al. Effects of walking on mortality
among nonsmoking retired men. N Engl J Med 1998;338:94e9. http://
dx.doi.org/10.1056/NEJM199801083380204.
Sato KK, Hayashi T, Kambe H, et al. Walking to work is an independent
predictor of incidence of type 2 diabetes in Japanese men: the Kansai
Healthcare Study. Diabetes Care 2007;30:2296e8. http://dx.doi.org/10.2337/
dc07-0090. dc07-0090 [pii].
Okada K, Hayashi T, Tsumura K, Suematsu C, Endo G, Fujii S. Leisure-time
physical activity at weekends and the risk of type 2 diabetes mellitus in
Japanese men: the Osaka Health Survey. Diabet Med 2000;17:53e8.
Nakanishi N, Takatorige T, Suzuki K. Daily life activity and risk of developing
impaired fasting glucose or type 2 diabetes in middle-aged Japanese men.
Diabetologia 2004;47:1768e75.
Noda H, Iso H, Toyoshima H, et al. Walking and sports participation and
mortality from coronary heart disease and stroke. J Am Coll Cardiol 2005;46:
1761e7.
http://dx.doi.org/10.1016/j.jacc.2005.07.038.
S0735e1097(05)
01782-1 [pii].
Inoue M, Iso H, Yamamoto S, et al. Daily total physical activity level and
premature death in men and women: results from a large-scale populationbased cohort study in Japan (JPHC study). Ann Epidemiol 2008;18:522e30.
http://dx.doi.org/10.1016/j.annepidem.2008.03.008. S1047e2797(08)00063X [pii].
Ottenbacher AJ, Snih SA, Karmarkar A, et al. Routine physical activity and
mortality in Mexican Americans aged 75 and older. J Am Geriatr Soc
2012;60:1085e91. http://dx.doi.org/10.1111/j.1532-5415.2012.03995.x.
Folsom AR, Arnett DK, Hutchinson RG, Liao F, Clegg LX, Cooper LS. Physical
activity and incidence of coronary heart disease in middle-aged women and
men. Med Sci Sports Exerc 1997;29:901e9.
Bell EJ, Lutsey PL, Windham BG, Folsom AR. Physical activity and cardiovascular disease in African Americans in atherosclerosis risk in communities.
Med Sci Sports Exerc 2013;45:901e7. http://dx.doi.org/10.1249/
MSS.0b013e31827d87ec.
331
[37] Manson JE, Greenland P, LaCroix AZ, et al. Walking compared with vigorous
exercise for the prevention of cardiovascular events in women. N Engl J Med
2002;347:716e25.
[38] Mathieu RA, Powell-Wiley TM, Ayers CR, et al. Physical activity participation,
health perceptions, and cardiovascular disease mortality in a multiethnic
population: the Dallas Heart Study. Am Heart J 2012;163:1037e40. http://
dx.doi.org/10.1016/j.ahj.2012.03.005. S0002e8703(12)00170-6 [pii].
[39] Gillum RF, Mussolino ME, Ingram DD. Physical activity and stroke incidence
in women and men. The NHANES I Epidemiologic Follow-up Study. Am J
Epidemiol 1996;143:860e9.
[40] Evenson KR, Rosamond WD, Cai J, et al. Physical activity and ischemic stroke
risk. The atherosclerosis risk in communities study. Stroke 1999;30:1333e9.
[41] Autenrieth CS, Evenson KR, Yatsuya H, Shahar E, Baggett C, Rosamond WD.
Association between physical activity and risk of stroke subtypes: the
atherosclerosis risk in communities study. Neuroepidemiology 2013;40:
109e16. http://dx.doi.org/10.1159/000342151. 000342151 [pii].
[42] Gregg EW, Gerzoff RB, Caspersen CJ, Williamson DF, Narayan KM. Relationship of walking to mortality among US adults with diabetes. Arch Intern Med
2003;163:1440e7. http://dx.doi.org/10.1001/archinte.163.12.1440. 163/12/
1440 [pii].
[43] de Munter JS, Agyemang C, Stronks K, van Valkengoed IG. Association of
physical activity, smoking, and alcohol intake with CVD-related hospital
discharge in people of European, South Asian, or African descent. Eur J Prev
Cardiol
2013;20:80e8.
http://dx.doi.org/10.1177/2047487311434232.
2047487311434232 [pii].
[44] Williams ED, Stamatakis E, Chandola T, Hamer M. Physical activity behaviour
and coronary heart disease mortality among South Asian people in the UK:
an observational longitudinal study. Heart 2011;97:655e9. http://dx.doi.org/
10.1136/hrt.2010.201012. hrt.2010.201012 [pii].
[45] Hsia J, Wu L, Allen C, et al. Physical activity and diabetes risk in postmenopausal women. Am J Prev Med 2005;28:19e25.
[46] Steinbrecher A, Erber E, Grandinetti A, Nigg C, Kolonel LN, Maskarinec G.
Physical activity and risk of type 2 diabetes among Native Hawaiians, Japanese Americans, and Caucasians: the Multiethnic Cohort. J Phys Act Health
2012;9:634e41. 2010e0209 [pii].
[47] Fretts AM, Howard BV, Kriska AM, et al. Physical activity and incident diabetes in American Indians: the Strong Heart Study. Am J Epidemiol
2009;170:632e9. http://dx.doi.org/10.1093/aje/kwp181. kwp181 [pii].
[48] Sattar N. Revisiting the links between glycaemia, diabetes and cardiovascular
disease. Diabetologia 2013;56:686e95. http://dx.doi.org/10.1007/s00125012-2817-5.
[49] Wing RR, Bolin P, Brancati FL, et al. Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes. N Engl J Med 2013;369:145e54. http://
dx.doi.org/10.1056/NEJMoa1212914.
[50] Wesche-Thobaben JA. The development and description of the comparison
group in the Look AHEAD trial. Clin Trials 2011;8:320e9. http://dx.doi.org/
10.1177/1740774511405858. 8/3/320 [pii].
[51] Gregg EW, Chen H, Wagenknecht LE, et al. Association of an intensive lifestyle intervention with remission of type 2 diabetes. J Am Med Assoc
2012;308:2489e96. http://dx.doi.org/10.1001/jama.2012.67929. 1486829
[pii].
[52] Jakicic JM, Egan CM, Fabricatore AN, et al. Four-year change in cardiorespiratory tness and inuence on glycemic control in adults with type 2 diabetes in a randomized trial: the Look AHEAD Trial. Diabetes Care 2013;36:
1297e303. http://dx.doi.org/10.2337/dc12-0712. dc12-0712 [pii].
[53] Li G, Zhang P, Wang J, et al. The long-term effect of lifestyle interventions to
prevent diabetes in the China Da Qing Diabetes Prevention Study: a 20-year
follow-up study. Lancet 2008;371:1783e9. http://dx.doi.org/10.1016/S01406736(08)60766-7. S0140e6736(08)60766-7 [pii].
[54] Heran BS, Chen JM, Ebrahim S, et al. Exercise-based cardiac rehabilitation for
coronary heart disease. Cochrane Database Syst Rev 2011:CD001800. http://
dx.doi.org/10.1002/14651858.CD001800.pub2.
[55] Lawler PR, Filion KB, Eisenberg MJ. Efcacy of exercise-based cardiac rehabilitation post-myocardial infarction: a systematic review and meta-analysis
of randomized controlled trials. Am Heart J 2011;162:571e84. http://
dx.doi.org/10.1016/j.ahj.2011.07.017. S0002e8703(11)00559-X [pii].
[56] Durstine JL, Grandjean PW, Davis PG, Ferguson MA, Alderson NL, DuBose KD.
Blood lipid and lipoprotein adaptations to exercise: a quantitative analysis.
Sports Med 2001;31:1033e62.
[57] Malkova D, Gill JMR. Effects of exercise on postprandial lipoprotein metabolism. Future Lipidol 2006;1:743e55.
[58] Al-Shayji IA, Caslake MJ, Gill JM. Effects of moderate exercise on VLDL and
intralipid kinetics in overweight/obese middle-aged men. Am J Physiol
Endocrinol
Metab
2012;302:E349e55.
http://dx.doi.org/10.1152/
ajpendo.00498.2011. ajpendo.00498.2011 [pii].
[59] Maarbjerg SJ, Sylow L, Richter EA. Current understanding of increased insulin
sensitivity after exercise - emerging candidates. Acta Physiol (Oxf) 2011;202:
323e35. http://dx.doi.org/10.1111/j.1748-1716.2011.02267.x.
[60] Turcotte LP, Fisher JS. Skeletal muscle insulin resistance: roles of fatty acid
metabolism and exercise. Phys Ther 2008;88:1279e96. http://dx.doi.org/
10.2522/ptj.20080018. ptj.20080018 [pii].
[61] Green DJ, Maiorana A, ODriscoll G, Taylor R. Effect of exercise training on
endothelium-derived nitric oxide function in humans. J Physiol 2004;561:1e25.
http://dx.doi.org/10.1113/jphysiol.2004.068197. jphysiol.2004.068197 [pii].
332
[62] Szostak J, Laurant P. The forgotten face of regular physical exercise: a natural anti-atherogenic activity. Clin Sci (Lond) 2011;121:91e106. http://
dx.doi.org/10.1042/CS20100520. CS20100520 [pii].
[63] Padilla J, Simmons GH, Bender SB, Arce-Esquivel AA, Whyte JJ, Laughlin MH.
Vascular effects of exercise: endothelial adaptations beyond active muscle
beds. Physiology (Bethesda) 2011;26:132e45. http://dx.doi.org/10.1152/
physiol.00052.2010. 26/3/132 [pii].
[64] Whelton SP, Chin A, Xin X, He J. Effect of aerobic exercise on blood pressure:
a meta-analysis of randomized, controlled trials. Ann Intern Med 2002;136:
493e503.
[65] Cornelissen VA, Buys R, Smart NA. Endurance exercise benecially affects
ambulatory blood pressure: a systematic review and meta-analysis. J Hypertens
2013;31:639e48. http://dx.doi.org/10.1097/HJH.0b013e32835ca964.
[66] Cornelissen VA, Smart NA. Exercise training for blood pressure: a systematic
review and meta-analysis. J Am Heart Assoc 2013;2:e004473. http://
dx.doi.org/10.1161/JAHA.112.004473. 2/1/e004473 [pii].
[67] Mathur N, Pedersen BK. Exercise as a mean to control low-grade systemic
inammation. Mediators Inamm 2008;2008:109502. http://dx.doi.org/
10.1155/2008/109502.
[68] You T, Arsenis NC, Disanzo BL, LaMonte MJ. Effects of exercise training on
chronic inammation in obesity: current evidence and potential mechanisms. Sports Med 2013;43:243e56. http://dx.doi.org/10.1007/s40279-0130023-3.
[69] Ryan D. Risks and benets of weight loss: challenges to obesity research. Eur
Heart J Suppl 2005;7:L27e31.
[70] Collins GS, Altman DG. Predicting the 10 year risk of cardiovascular disease
in the United Kingdom: independent and external validation of an updated
version of QRISK2. BMJ 2012;344:e4181.
[71] Lindstrom J, Peltonen M, Eriksson JG, et al. Improved lifestyle and decreased
diabetes risk over 13 years: long-term follow-up of the randomised Finnish
Diabetes Prevention Study (DPS). Diabetologia 2013;56:284e93. http://
dx.doi.org/10.1007/s00125-012-2752-5.
[72] Knowler WC, Fowler SE, Hamman RF, et al. 10-year follow-up of diabetes
incidence and weight loss in the Diabetes Prevention Program Outcomes
Study. Lancet 2009;374:1677e86. http://dx.doi.org/10.1016/S0140-6736(09)
61457-4. S0140e6736(09)61457-4 [pii].
[73] Gong Q, Gregg EW, Wang J, et al. Long-term effects of a randomised trial of a
6-year lifestyle intervention in impaired glucose tolerance on diabetesrelated microvascular complications: the China Da Qing Diabetes Prevention Outcome Study. Diabetologia 2011;54:300e7. http://dx.doi.org/
10.1007/s00125-010-1948-9.
[74] Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM
in people with impaired glucose tolerance. The Da Qing IGT and Diabetes
Study. Diabetes Care 1997;20:537e44.
[75] Kosaka K, Noda M, Kuzuya T. Prevention of type 2 diabetes by lifestyle
intervention: a Japanese trial in IGT males. Diabetes Res Clin Pract 2005;67:
152e62. http://dx.doi.org/10.1016/j.diabres.2004.06.010. S0168e8227(04)
00178-0 [pii].
[76] Ramachandran A, Snehalatha C, Mary S, Mukesh B, Bhaskar AD, Vijay V. The
Indian Diabetes Prevention Programme shows that lifestyle modication and
metformin prevent type 2 diabetes in Asian Indian subjects with impaired
glucose tolerance (IDPP-1). Diabetologia 2006;49:289e97.
[77] Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of
type 2 diabetes with lifestyle intervention or metformin. N Engl J Med
2002;346:393e403.
[78] Penn L, White M, Lindstrom J, et al. Importance of weight loss maintenance
and risk prediction in the prevention of type 2 diabetes: analysis of European
Diabetes Prevention Study RCT. PLoS ONE 2013;8:e57143. http://dx.doi.org/
10.1371/journal.pone.0057143. PONE-D-12e30924 [pii].
[79] Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes
mellitus by changes in lifestyle among subjects with impaired glucose
tolerance. N Engl J Med 2001;344:1343e50.
[80] Mukhopadhyay B, Forouhi NG, Fisher BM, Kesson CM, Sattar N. A comparison
of glycaemic and metabolic control over time among South Asian and European patients with type 2 diabetes: results from follow-up in a routine
diabetes clinic. Diabet Med 2006;23:94e8.
[81] Sproston K, Mindell J. The health of minority ethnic groups, vol. 1; 2006.
[82] International Diabetes Federation. IDF diabetes atlas. Brussels: International
Diabetes Federation; 2009.
[83] Zimmet P. Globalization, coca-colonization and the chronic disease
epidemic: can the Doomsday scenario be averted? J Intern Med 2000;247:
301e10. jim625 [pii].
[84] McKeigue PM, Shah B, Marmot MG. Relation of central obesity and insulin
resistance with high diabetes prevalence and cardiovascular risk in South
Asians. Lancet 1991;337:382e6.
[85] Venkataraman R, Nanda NC, Baweja G, Parikh N, Bhatia V. Prevalence of
diabetes mellitus and related conditions in Asian Indians living in the United
States. Am J Cardiol 2004;94:977e80. http://dx.doi.org/10.1016/j.amjcard.2004.06.048. S0002e9149(04)00976-2 [pii].
[86] Kanaya AM, Wassel CL, Mathur D, et al. Prevalence and correlates of diabetes
in South Asian Indians in the United States: ndings from the metabolic
syndrome and atherosclerosis in South Asians living in America study and
the multi-ethnic study of atherosclerosis. Metab Syndr Relat Disord 2010;8:
157e64. http://dx.doi.org/10.1089/met.2009.0062.
[87] Zhang Q, Wang Y, Huang ES. Changes in racial/ethnic disparities in the

prevalence of type 2 diabetes by obesity level among US adults. Ethn Health
2009;14:439e57.
http://dx.doi.org/10.1080/13557850802699155.
910365654 [pii].
[88] Mohan V, Mathur P, Deepa R, et al. Urban rural differences in prevalence of
self-reported diabetes in Indiaethe WHO-ICMR Indian NCD risk factor surveillance. Diabetes Res Clin Pract 2008;80:159e68.
[89] Ramachandran A, Mary S, Yamuna A, Murugesan N, Snehalatha C. High
prevalence of diabetes and cardiovascular risk factors associated with urbanization in India. Diabetes Care 2008;31:893e8. http://dx.doi.org/10.2337/
dc07-1207. dc07-1207 [pii].
[90] Maskarinec G, Grandinetti A, Matsuura G, et al. Diabetes prevalence and
body mass index differ by ethnicity: the Multiethnic Cohort. Ethn Dis
2009;19:49e55.
[91] Yu CH, Zinman B. Type 2 diabetes and impaired glucose tolerance in
aboriginal populations: a global perspective. Diabetes Res Clin Pract
2007;78:159e70. http://dx.doi.org/10.1016/j.diabres.2007.03.022. S0168e
8227(07)00249-5 [pii].
[92] Valencia ME, Bennett PH, Ravussin E, Esparza J, Fox C, Schulz LO. The Pima
Indians in Sonora, Mexico. Nutr Rev 1999;57:S55e7.
[93] Schulz LO, Bennett PH, Ravussin E, et al. Effects of traditional and western
environments on prevalence of type 2 diabetes in Pima Indians in Mexico
and the U.S. Diabetes Care 2006;29:1866e71. http://dx.doi.org/10.2337/
dc06-0138. 29/8/1866 [pii].
[94] Esparza J, Fox C, Harper IT, et al. Daily energy expenditure in Mexican and
USA Pima indians: low physical activity as a possible cause of obesity. Int J
Obes Relat Metab Disord 2000;24:55e9.
[95] Baier LJ, Hanson RL. Genetic studies of the etiology of type 2 diabetes in Pima
Indians: hunting for pieces to a complicated puzzle. Diabetes 2004;53:1181e
6.
[96] Perez-Bravo F, Carrasco E, Santos JL, Calvillan M, Larenas G, Albala C. Prevalence of type 2 diabetes and obesity in rural Mapuche population from
Chile. Nutrition 2001;17:236e8. S0899900700005505 [pii].
[97] Perez-Bravo F, Fuentes M, Angel B, et al. Lack of association between the fatty
acid binding protein 2 (FABP2) polymorphism with obesity and insulin
resistance in two aboriginal populations from Chile. Acta Diabetol 2006;43:
93e8. http://dx.doi.org/10.1007/s00592-006-0221-7.
[98] Larenas G, Arias G, Espinoza O, et al. Prevalence of diabetes mellitus in a
Mapuche community of Region IX, Chile. Rev Med Chil 1985;113:1121e5.
[99] Carrasco EP, Perez FB, Angel BB, et al. Prevalence of type 2 diabetes and
obesity in two Chilean aboriginal populations living in urban zones. Rev Med
Chil 2004;132:1189e97.
[100] Celis-Morales CA, Perez-Bravo F, Ibanes L, et al. Insulin resistance in Chileans
of European and Indigenous descent: evidence for an ethnicity x environment interaction. PLoS ONE 2011;6:e24690.
[101] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC.
Homeostasis model assessment: insulin resistance and beta-cell function
from fasting plasma glucose and insulin concentrations in man. Diabetologia
1985;28:412e9.
[102] Forouhi NG, Sattar N, Tillin T, McKeigue PM, Chaturvedi N. Do known risk
factors explain the higher coronary heart disease mortality in South Asian
compared with European men? Prospective follow-up of the Southall and
Brent studies, UK. Diabetologia 2006;49:2580e8.
[103] Gray LJ, Yates T, Davies MJ, et al. Dening obesity cut-off points for migrant
South Asians. PLoS ONE 2011;6:e26464. http://dx.doi.org/10.1371/journal.pone.0026464. PONE-D-11e06667 [pii].
[104] Chandalia M, Lin P, Seenivasan T, et al. Insulin resistance and body fat distribution in South Asian men compared to Caucasian men. PLoS ONE 2007;2:
e812.
[105] Hall LM, Moran CN, Milne GR, et al. Fat oxidation, tness and skeletal muscle
expression of oxidative/lipid metabolism genes in South Asians: implications
for insulin resistance? PLoS ONE 2010;5:e14197. http://dx.doi.org/10.1371/
journal.pone.0014197.
[106] Hall LML, Sattar N, Gill JMR. Risk of metabolic and vascular disease in South
Asians: potential mechanisms for increased insulin resistance. Future Lipidology 2008;3:411e24.
[107] Misra A, Vikram NK. Insulin resistance syndrome (metabolic syndrome) and
obesity in Asian Indians: evidence and implications. Nutrition 2004;20:482e
91.
[108] Petersen KF, Dufour S, Feng J, et al. Increased prevalence of insulin resistance
and nonalcoholic fatty liver disease in Asian-Indian men. Proc Natl Acad Sci
U S A 2006;103:18273e7.
[109] Yajnik CS. Early life origins of insulin resistance and type 2 diabetes in India
and other Asian countries. J Nutr 2004;134:205e10.
[110] Yajnik CS, Lubree HG, Rege SS, et al. Adiposity and hyperinsulinemia in Indians are present at birth. J Clin Endocrinol Metab 2002;87:5575e80.
[111] Yajnik CS, Fall CH, Coyaji KJ, et al. Neonatal anthropometry: the thin-fat
Indian baby. The Pune Maternal Nutrition Study. Int J Obes Relat Metab
Disord 2003;27:173e80.
[112] Ramachandran A, Ma RC, Snehalatha C. Diabetes in Asia. Lancet 2010;375:
408e18.
http://dx.doi.org/10.1016/S0140-6736(09)60937-5.
S0140e
6736(09)60937-5 [pii].
[113] Gujral UP, Pradeepa R, Weber MB, Narayan KM, Mohan V. Type 2 diabetes in
South Asians: similarities and differences with white Caucasian and other
[114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
populations. Ann N Y Acad Sci 2013;1281:51e63. http://dx.doi.org/10.1111/

j.1749-6632.2012.06838.x.
Fischbacher CM, Hunt S, Alexander L. How physically active are South Asians
in the United Kingdom? A literature review. J Public Health (Oxf) 2004;26:
250e8. http://dx.doi.org/10.1093/pubmed/fdh158. 26/3/250 [pii].
Hayes L, White M, Unwin N, et al. Patterns of physical activity and relationship with risk markers for cardiovascular disease and diabetes in Indian,
Pakistani, Bangladeshi and European adults in a UK population. J Public
Health Med 2002;24:170e8.
Yates T, Davies MJ, Gray LJ, et al. Levels of physical activity and relationship
with markers of diabetes and cardiovascular disease risk in 5474 white
European and South Asian adults screened for type 2 diabetes. Prev Med
2010.
http://dx.doi.org/10.1016/j.ypmed.2010.06.011.
S0091e7435(10)
00238-0 [pii].
Williams ED, Stamatakis E, Chandola T, Hamer M. Assessment of physical
activity levels in South Asians in the UK: ndings from the Health Survey for
England. J Epidemiol Community Health 2011;65:517e21. http://dx.doi.org/
10.1136/jech.2009.102509. jech.2009.102509 [pii].
Ghouri N, Purves D, McConnachie A, Wilson J, Gill JMR, Sattar N. Lower
cardiorespiratory tness contributes to increased insulin resistance and
fasting glycaemia in middle-aged South Asian compared to European men
living in the UK. Diabetologia 2013. http://dx.doi.org/10.1007/s00125-0132969-y.
Forouhi NG, Jenkinson G, Thomas EL, et al. Relation of triglyceride stores in
skeletal muscle cells to central obesity and insulin sensitivity in European
and South Asian men. Diabetologia 1999;42:932e5.
Berntsen S, Richardsen KR, Morkrid K, Sletner L, Birkeland KI, Jenum AK.
Objectively recorded physical activity in early pregnancy: a multiethnic
population-based study. Scand J Med Sci Sports 2012. http://dx.doi.org/
10.1111/sms.12034.
Duncan MJ, Birch S, Al-Nakeeb Y, Nevill AM. Ambulatory physical activity
levels of white and South Asian children in Central England. Acta Paediatr
2012;101:e156e62. http://dx.doi.org/10.1111/j.1651-2227.2011.02566.x.
Davey GJG, Roberts JD, Patel S, et al. Effects of exercise on insulin resistance
in South Asians and Europeans. Journal of Exercise Physiology 2000;3:6e11.
Hardy CP, Eston RG. Aerobic tness of Anglo-Saxon and Indian students. Br J
Sports Med 1985;19:217e8.
Ceaser TG, Fitzhugh EC, Thompson DL, Bassett Jr DR. Association of physical
activity, tness, and race: NHANES 1999e2004. Med Sci Sports Exerc
2013;45:286e93. http://dx.doi.org/10.1249/MSS.0b013e318271689e.
Bouchard C, Daw EW, Rice T, et al. Familial resemblance for VO2max in the
sedentary state: the HERITAGE family study. Med Sci Sports Exerc 1998;30:
252e8.
Bouchard C, Lesage R, Lortie G, et al. Aerobic performance in brothers,
dizygotic and monozygotic twins. Med Sci Sports Exerc 1986;18:639e46.
Larsen FJ, Anderson M, Ekblom B, Nystrom T. Cardiorespiratory tness predicts insulin action and secretion in healthy individuals. Metabolism
[128]
[129]
[130]
[131]
[132]
[133]
[134]
[135]
[136]
[137]
[138]
[139]
[140]
[141]
333
2012;61:12e6. http://dx.doi.org/10.1016/j.metabol.2011.05.010. S0026e

0495(11)00139-9 [pii].
Gan SK, Kriketos AD, Ellis BA, Thompson CH, Kraegen EW, Chisholm DJ.
Changes in aerobic capacity and visceral fat but not myocyte lipid levels
predict increased insulin action after exercise in overweight and obese men.
Diabetes Care 2003;26:1706e13.
Lynch J, Helmrich SP, Lakka TA, et al. Moderately intense physical activities
and high levels of cardiorespiratory tness reduce the risk of non-insulindependent diabetes mellitus in middle-aged men. Arch Intern Med
1996;156:1307e14.
Wei M, Gibbons LW, Mitchell TL, Kampert JB, Lee CD, Blair SN. The association between cardiorespiratory tness and impaired fasting glucose and
type 2 diabetes mellitus in men. Ann Intern Med 1999;130:89e96.
Wisloff U, Najjar SM, Ellingsen O, et al. Cardiovascular risk factors emerge
after articial selection for low aerobic capacity. Science 2005;307:418e20.
Holloszy JO, Coyle EF. Adaptations of skeletal muscle to endurance exercise
and their metabolic consequences. J Appl Physiol 1984;56:831e8.
Church T. The low-tness phenotype as a risk factor: more than just being
sedentary? Obesity (Silver Spring) 2009;17(Suppl. 3):S39e42. http://
dx.doi.org/10.1038/oby.2009.387. oby2009387 [pii].
Misra A, Chowbey P, Makkar BM, et al. Consensus statement for diagnosis of
obesity, abdominal obesity and the metabolic syndrome for Asian Indians
and recommendations for physical activity, medical and surgical management. J Assoc Physicians India 2009;57:163e70.
Kumar S, Hanif W, Zaman MJ, Sattar N, Patel K, Khunti K. Lower thresholds
for diagnosis and management of obesity in British South Asians. Int J Clin
Pract 2011;65:375e85.
Razak F, Anand SS, Shannon H, et al. Dening obesity cut points in a
multiethnic population. Circulation 2007;115:2111e8.
Misra A, Nigam P, Hills AP, et al. Consensus physical activity guidelines for
Asian Indians. Diabetes Technol Ther 2012;14:83e98. http://dx.doi.org/
10.1089/dia.2011.0111.
Waidyatilaka I, Lanerolle P, Wickremasinghe R, Atukorala S,
Somasundaram N, de SA. Sedentary behaviour and physical activity in South
Asian women: time to review current recommendations? PLoS ONE 2013;8:
e58328.
http://dx.doi.org/10.1371/journal.pone.0058328.
PONE-D-12e
33340 [pii].
Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical
activity in the United States measured by accelerometer. Med Sci Sports
Exerc 2008;40:181e8. http://dx.doi.org/10.1249/mss.0b013e31815a51b3.
Celis-Morales CA, Ghouri N, Bailey MES, Sattar N, Gill JMR. Should physical
activity recommendations be ethnicity-specic? Evidence from a crosssectional study of South Asian and European men. PLoS ONE 2013;8:
e82568. http://dx.doi.org/10.1371/journal.pone.0082568.
Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral
glucose tolerance testing: comparison with the euglycemic insulin clamp.
Diabetes Care 1999;22:1462e70.

Gill (2014) Physical Activity, Ethnicity and Cardio-Metabolic Health. Does One Size Fit All.

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Gill (2014) Physical Activity, Ethnicity and Cardio-Metabolic Health. Does One Size Fit All.

Enviado por

Direitos autorais:

Formatos disponíveis

Atherosclerosis 232 (2014) 319e333

Contents lists available at ScienceDirect

Physical activity, ethnicity and cardio-metabolic health: Does one size

1. Physical activity, cardiorespiratory tness and sedentary

A large and consistent body of prospective epidemiological data

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

adults, it was estimated 16% of mortality could be attributed to low

7530 men aged 45e68 22 years

7589 men aged 45e93 6 years

3707 African-American Up to 21 years

Donahue, Honolulu Japanese

6815 men aged 45e68 6 years

14,575 White and Black Average follow-up

Baseline physical activity

In younger middle-aged men (45e54 years) there was no signicant effect of

Physical activity index

(continued on next page)

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

First author, study

Study (US, 1999)

Fretts, The Strong American Indians

Black and nonBlack Americans

14,040 Black and non- Up to 4e7 years

2896 men and women 8 years

2678 men aged 71e93 2e4 years

Caucasian, AfricanAmerican, Hispanic,

Baseline physical activity

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

1651 men and women, 10 years

Inoue, Japan Public Japanese

Lan, The National

83,034 men and

2113 men and women 2 years

74,942 women aged 40 5.7 years

2924 men aged 35e59 7 years

32,953 men and 44,723 9.7 years

6013 Japanese men

For the combined multi-ethnic cohort, overall physical activity participation

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

All-cause mortality, Self-reported physical activity

(continued on next page)

First author, study

948 men and women

All-cause mortality Self-reported physical activity

8006 men aged 45e68 23 years

Baseline physical activity

Compared to the lowest quartile of physical activity participants in quartiles, 2, 3

11,073 Japanese men

74,913 men and

Self-reported physical activity

70,658 women aged 40 4.6 years

Self-reported physical activity

Villegas, Shanghai Chinese

61,477 men aged 40

CVD and All-cause

Self-reported physical activity

13,293 White, 1244

Self-reported physical activity

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

Sato, The Kansai

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

J.M.R. Gill et al. / Atherosclerosis 232 (2014) 319e333

atherosclerotic and metabolic disease process, including favourable