Escolar Documentos
Profissional Documentos
Cultura Documentos
Atherosclerosis
journal homepage: www.elsevier.com/locate/atherosclerosis
Review
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
320
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
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
Japanese
Abbott, Honolulu
Heart Study (US,
2003) [25]
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]
De Munter, Sunset
study
(Netherlands,
2013) [43]
South Asian
Surimanese, African
Surimanese and
White Dutch
Evenson, The
Atherosclerosis
Risk in
Communities
Black Americans
and White
Americans
Up to 7 years
370 South Asian
Surimanese, 689
African Surimanese,
and 567 Dutch men and
women, aged 35e60
years at baseline
7544 men aged 45e68 12 years
years at baseline
Main ndings
321
Japanese
Abbott, Honolulu
Heart Study (US,
1994) [26]
Primary outcome
measure
322
Table 1 (continued )
Ethnic group
studieda
Participants
Follow-up
Gillum, NHANES 1
(US, 1996) [39]
Non-Hispanic
whites, nonHispanic blacks,
Hispanic and other
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]
Hsia, Womens
Health Initiative
Observational
Study (US, 2005)
[45]
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
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]
Primary outcome
measure
Taiwanese
5 years
Chinese
Matthews,
Shanghai
Womens Health
Study (China,
2007) [19]
Nakanishi, (Japan
2004) [31]
Japanese
Japanese
Noda, The Japan
Collaborative
Cohort Study for
Evaluation for
Cancer Risk
(Japan, 2005)
[32]
Okada, Osaka
Health Survey
(Japan, 2000)
[30]
Japanese
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).
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
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
3 years
Incident CHD
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).
Physical activity index
calculated from self-report of
24-h habitual physical activity
levels
Japanese
4 years
Incident T2DM
Native Hawaiians,
Japanese
Americans,
Caucasians
8 years
Incident T2DM
Incident T2DM
Main ndings
Wang, Shanghai
Mens Health
Study (China,
2013) [21]
Chinese
5.4 years
Williams, Health
Survey for
England (UK,
2011) [44]
Indian, Pakistani/
Bangladeshi and
White British
4 or 9 years
CHD mortality
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.
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
327
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
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
South Asians
65
Europeans
328
55
45
35
25
15
0
20
40
60
80
100
120
European
South Asian
6
4
2
p(ethnicity) = 0.001
0
45
50
55
60
65
Exercise intensity (%VO2max)
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)
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].
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
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
[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]
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.
[114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
[128]
[129]
[130]
[131]
[132]
[133]
[134]
[135]
[136]
[137]
[138]
[139]
[140]
[141]
333