SEARCH

SEARCH BY CITATION

Keywords:

  • Disparity;
  • ethnicity;
  • obesity;
  • UK

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

Ethnic minority groups are growing as a proportion of the British population. Although disparate, literature suggests inequalities in obesity risk within and among ethnic minority groups relative to Caucasians in the UK. We summarize and appraise the existing peer-reviewed literature about the prevalence and determinants of obesity among ethnic minority groups relative to Caucasians among children and adults in the UK. There was no consensus about obesity prevalence relative to Caucasians among South Asian or Black children or among South Asian adults relative to Caucasians. Black adults generally had higher risk for obesity than Caucasians. Both Chinese children and adults had lower risk for obesity than Caucasians. Few studies have considered differences in the aetiology of obesity by ethnicity. The lack of consensus regarding obesity risk among large ethnic minority groups relative to Caucasians in the UK, and the paucity of studies concerned with differences in obesity aetiology by ethnicity warrant further research in this area. Certain obesity metrics may bias obesity prevalence among particular ethnic groups relative to Caucasians. We summarize key methodological limitations to the current literature and suggest avenues for future research.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

Obesity is a growing epidemic in the UK among both children(1,2) and adults (3). The Foresight Obesity project, which forecasted obesity prevalence among the general population in the UK from current trends suggests that by 2050, 60% of men and 50% of women will be obese, with growing disparities in risk for obesity by ethnicity projected (3).

Obesity is an important determinant of cardiovascular disease risk, and is associated with risk for hypertension, hypercholesterolemia, angina pectoris and coronary heart disease (4). Obesity is also associated with several other diseases(5,6) including diabetes mellitus (7), cancer(8–10), stroke (11), rheumatoid(11) and osteoarthritis (12), and depression (13), among several others (5). Among children, obesity predicts orthopaedic abnormalities, idiopathic intracranial hypertension, asthma, sleep apnoea, gallstones, insulin resistance and subsequent diabetes mellitus, hyperandrogenemia, and poor mental health (6). Obesity also contributes to excess mortality (14,15).

Disparities in health between ethnic minority and majority groups have been identified since the advent of the quantitative measures of population health metrics (14,16,17), including in the UK (16). A review of the literature about ethnic disparities in health metrics in the UK found important differences in high burden diseases and mortality by ethnic group (17). Such differences have been demonstrated in respiratory diseases (18), hypertension (19), heart disease (20–23), and diabetes mellitus (19,21,23). A study about adult mortality among immigrants (24) found differences by country of origin in ischemic heart disease mortality, stroke mortality, suicide and cancer mortality. Data about the healthcare experience of ethnic minorities in the UK also suggests disparities: ethnic minority groups are consistently less likely to report positive experiences with healthcare providers relative to Caucasians (23,25).

According to data from the most recent UK census (26), ethnic minority groups comprise almost 8% of the total population, at about 4.6 million people. The largest ethnic minority group in 2001 was Indian (1.8%), followed by Pakistani (1.3%), mixed ethnicity (1.2%), and Black Caribbean (1.0%). In 2001, the ethnic minority population had grown by 53% since 1991 (26).

Ethnic minorities are generally younger than the Caucasian British population (23,27), and tend to be of lower socioeconomic status than their Caucasian counterparts; Pakistani and Bangladeshi groups have the lowest proportions in ‘managerial and professional occupations’, and Bangladeshis and Black Africans in the UK have the highest proportions of children eligible for free school meals (23). Ethnic minorities are also more likely to be unemployed (27).

Recent trends have shown increasing hostility towards immigration and multiculturalism among the British public, with the proportion who agree or strongly agree with the statement, ‘there are too many immigrants in Britain’ rising from 56% to 70% between 1999 and 2008, and the proportion of those claiming that immigration is the most important issue facing Britain rising from just under 5% in 1997 to over 45% in May 2006 (28). A recent report highlighted increases in hate crimes against Muslim ethnic minority groups in the UK (29).

In the setting of inequalities in socioeconomic wellbeing by ethnicity, and rising tension between ethnic minority and majority groups, it is crucial that the public health community continue to study systematic differences in health by ethnicity in the UK. Given the importance of obesity in health status, ethnic differences in obesity are of particular importance. Studies have demonstrated ethnic inequalities in obesity risk in the UK (30–32). However, this literature remains disparate and disorganized, as to our knowledge, there has never been a systematic appraisal or synthesis of findings regarding ethnic inequalities in obesity in the UK. So as to organize and appraise the present literature, and to identify directions for future research and intervention, we review the literature about ethnic disparities in obesity risk in the last 30 years, summarizing and evaluating important differences in the prevalence and determinants of metrics of obesity by ethnicity among South Asian, Black, Chinese, and other groups relative to Caucasians among children and adults in the UK.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

We sought to systematically review the literature about ethnic disparities in obesity in the UK so as to understand how ethnicity influenced obesity risk in this context. This review encompassed the peer-reviewed literature published between 1 January 1980 and 8 March 2010. We limited our review to these years so as to reflect current thinking about the relation between ethnicity and health. The literature reviewed was identified through the MEDLINE database through the ‘pubmed.gov’ interface and it included papers that included any empirical assessment of the relation between ethnicity and metrics of obesity. We used MeSH search terms ‘Obesity’ and ‘Great Britain’ for English-language articles published in the peer-reviewed literature. The primary author carried out all queries during the month of March 2010.

Our original search yielded 1189 articles, 54 of which were judged to consider the relation between ethnicity and obesity in the UK after screening by title. Another 11 were discarded because the abstract showed that studies were not set in the UK, or because the studies did not empirically analyse differences in obesity by ethnicity. Of the remaining 43, manuscripts were included in the review if they fulfilled the following criteria:

  • • 
    Considered at least two defined, ethnic subpopulations in the UK, and described attribution of ethnicity among respondents.
  • • 
    Described the method used to define obesity, including metric of interest, and threshold for overweight or obesity utilized in analysis.
  • • 
    Conducted a direct empiric analysis of differences in obesity outcome by ethnic subpopulation.

After reading the manuscripts, 21 were excluded because they did not explicitly state the prevalence of obesity in at least two different ethnic groups. This left 22 articles from the original search considered in this review. Reference lists from these articles were searched, and yielded a further 7 articles which fulfilled the inclusion criteria, yielding a final total of 29 articles reviewed here.

Because of the diverse methods used to measure the prevalence of obesity in different studies, a meta-analysis of the results was not attempted. For each of the papers, the primary author extracted the following information: definition of ethnic groups, definition of obesity, population and setting, sample and methods, and findings and conclusions.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

Throughout this section, we will summarize ethnic differences in obesity prevalence and incidence by ethnicity. We organized our findings by age group, considering first ethnic disparities in obesity among children and adolescents in the UK, and then moving to ethnic disparities in obesity among adults. Each study is reviewed in detail (including definition of ethnic groups, definition of obesity, population and setting, sample and methods, and findings and conclusions) in Table 1.

Table 1.  Empirical, peer-reviewed studies about ethnicity and obesity in the UK, 1980–2010
ReferenceEthnic groups representedObesity-related outcomeLocationSampleMethodsConclusions
  1. BMI, body mass index; DXA, dual energy X-ray absorbtiometry; IOTF, International Obesity Taskforce.

Age group: children & adolescents
Rona and Chinn (40)Caucasian, Afro-Caribbean, Indo-Pakistani (Gujurati, Punjabi, Urdu), OtherWeight-for-height, triceps skin-foldEngland13 073 children 5–11 years old who took part in the National Study of Health and Growth in 1982–1983.A cross-sectional analysis of differences in obesity among children by ethnicity.Median weight for height was above the 50th percentile among Caucasians and Afro-Caribbean children, although there was no significant difference. Gujurati girls were lighter than Punjabi and Urdu girls. ‘Other’ children were above the 50th percentile for weight-for-height. Caucasian children were above the 50th percentile in triceps skin-folds, while Afro-Caribbean children were below the 50th percentile. All Asian girls subgroups were below the 50th percentile. Gujurati boys were below the 50th percentile while all other Asian boys were above the 50th percentile in triceps skin-folds. Gujuratis had the lowest weight-for-height standard deviation score while ‘Other’ groups had the highest. ‘Others’, followed by Caucasians had the highest triceps skin-fold standard deviation scores, while Gujuratis followed by Afro-Caribbeans had the lowest.
Duran-Tauleria et al.(39)Caucasian, Afro-Caribbean, AsianOverweight (triceps fold, subscapular fold, triceps and subscapular fold sum, and weight-for-height >75 percentile)England, ScotlandAn English representative sample of 6463 children in 1990; a Scottish representative sample of 4165 children in 1990–1991; an inner-city sample of 7049 mixed-ethnic children from England in 1991. All students aged 5–11.A cross-sectional analysis of the relation between social and biological factors with metrics of obesity in children.The interaction between family size and ethnic group was associated with weight-for-height, triceps skin-fold, and the sum of triceps and subscapular skin-folds in linear regression models adjusted for potential confounders. Punjabi and Urdu children were at higher risk of obesity according to logistic regression models of sum of triceps and subscapular folds adjusted for potential confounders.
Chinn et al.(41)Caucasian, Urdu or Punjabi speaking, Afro-Caribbean, Gujurati speaking, other IndianWeight-for-height, triceps skin-fold thicknessEngland, ScotlandOver 2000 inner-city Caucasian, 1500 Urdu or Punjabi speaking, 1000 Afro-Caribbean, 300 Gujarati speaking, and 300 other Indian children from the five wards with the highest percentage of Asians and the five with the highest percentage of Afro-Caribbeans, and ten other wards selected for deprivation in England, as well as 5000 English representative Caucasian and 3000 Scottish representative Caucasian children.A mixed longitudinal analysis of growth trends and obesity among ethnic minority and inner city children from 1983–1994.Height significantly increased between 1983 and 1994 among all groups except ‘other’ Indian boys. Weight increased significantly among all groups except for ‘other’ Indian boys. There were significant increases in weight-for-height among Gujurati and inner city Caucasian girls. Triceps skin-fold trends showed the greatest increases among Indian subcontinent girls and boys, and girls in the English representative sample. The greatest increase was among Gujurati girls (11%). Inner-city groups were thinner on average at the end of the study than the representative groups. Afro-Caribbeans were slimmest.
Whincup et al.(38)South Asian, CaucasianMean ponderal index, waist/hip ratio, waist circumferenceEngland, Wales227 South Asian and 3415 Caucasian children aged 8 to 11 who took part in the ‘Ten Towns Heart Health Studies’.A cross-sectional analysis of ethnic differences in cardiovascular disease risk.Upon adjustment for age, sex and town, Caucasians had a significantly higher ponderal index than South Asians, although waist circumference and waist/hip ratio were similar.
Jebb et al.(34)Caucasian; Afro-Caribbean; AsianOverweight and obesity (using IOTF (44) BMI cut-offs)England, Wales and ScotlandNationally-representative sample of 1836 children between 4–18 years old in 1997.A cross-sectional analysis of the prevalence of obesity among young people in the UK.Asian subjects had almost 4 times higher prevalence of obesity compared to Caucasians (13.6% vs. 3.5%, P < 0.001), although there was no significant difference in overweight by ethnicity.
Saxena et al., 2004 (32)Black Caribbean, Indian, Pakistani, Bangladeshi, Chinese, Irish and General PopulationOverweight and obesity (using IOTF (44) BMI cut-offs)England5689 children and young adults aged 2–20 from the 1999 Health Survey for England.A cross-sectional analysis of the burden of obesity among children by ethnic group and socioeconomic status.Indian and Pakistani boys had the highest prevalence of overweight (30 and 26%, respectively) and obesity (7.9 and 9.0%, respectively) compared with boys in the general population (22% and 5.8%, overweight and obesity). Bangladeshi boys had the lowest risk (2.8%). In multivariate models adjusted for social class, Indian and Pakistani boys had significantly higher odds for overweight, but not obesity. By contrast, Bangladeshi boys had significantly lower odds for both overweight and obesity. Among girls, Afro-Caribbean had the highest prevalence of obesity (13.0%), while Chinese girls had the lowest prevalence of obesity (1.2%) compared to the general population (5.8%).In multivariate models adjusted for social class among both genders, only Afro-Caribbean girls had higher odds of overweight (OR = 1.79, 95% CI 1.29–2.33) and both Afro-Caribbean (OR = 2.74, 95% CI 1.74–4.31) and Pakistani (OR = 1.71, 95% CI 1.06–2.76) girls had higher odds of obesity than the general population. Chinese girls had significantly lower odds of overweight (OR = 0.52, 95% CI 0.29–0.91), and Indian (OR = 0.39, 95% CI 0.17–0.86) and Chinese (OR = 0.08, 95% CI 0.01–0.56) girls had significantly lower odds of obesity relative to the general population.
Taylor et al.(45)Caucasian, Black African, Indian, Pakistani, Bangladeshi, Black CaribbeanOverweight and obesity (using IOTF(44) BMI cut-offs, or taken from UK 1990 reference BMI data where overweight was >85 percentile, obesity was >95 percentile, and extreme obesity was >99.86 percentile in BMI)East London2482 children aged 11–14 who participated in the ‘Research in East London Adolescents Community Health Survey’.A cross-sectional analysis of ethnic and socioeconomic disparities in overweight and obesity.Pakistani, Black African, and Bangladeshi boys had lower mean BMI than Caucasian British boys. Ethnicity was significantly associated with mean BMI in both sexes after adjusting for age and height. Prevalence of overweight and obesity was highest among Black African girls and lowest among Bangladeshi and Pakistani girls. Among boys, Indian boys had the highest prevalence of obesity and overweight (26% and 47%, respectively, according to UK 1990 thresholds). In multivariate regression models using IOTF cut-offs adjusted for age and height, among girls, only Pakistani girls had significantly lower risk for overweight (OR = 0.56, 95% CI 0.36–0.87) compared to Caucasian British girls. Pakistani boys (OR = 0.53, 95% CI 0.29–0.99) and Black African (OR = 0.56, 95% CI 0.34–0.88) boys had significantly lower risk of overweight, and Indian boys had significantly higher risk of overweight (OR = 1.96, 95% CI 1.20–3.03) compared to Caucasian British boys.
Wardle et al.(31)Caucasian, Asian/mixed Asian, Black/mixed BlackOverweight and obesity (using IOTF(44) BMI cut-offs); waist circumference and waist circumference standard deviation scores (relative to UK 1990 reference data)London5863 students recruited from 36 schools at age 11–12 in 1999.A longitudinal analysis of ethnic and socioeconomic differences in the development of obesity in adolescence.Black/mixed black boys were heaviest and tallest, and Asian/mixed Asian boys were shortest. Waist circumferences were highest in black/mixed black boys and girls. Black/mixed black girls were significantly heavier and taller, and had a higher BMI, than Caucasian or Asian/mixed Asian girls. In time trends of prevalence of overweight and obesity by ethnicity, adjusted for socioeconomic status, in all 5 years of the study, black/mixed black girls were significantly more likely to be overweight or obese than were Caucasian or Asian/mixed Asian girls. Caucasian girls were slightly more likely to be obese or overweight than were Asian/mixed Asian girls, with significant differences in 3 of the 5 years. Asian/mixed Asian boys had the highest prevalence of overweight and obesity in all years, but there were no significant differences in any year. For waist circumference, there was a significant sex by ethnic group interaction, as there were substantial differences in the waist circumferences of girls, but not boys by ethnicity. Black students had higher waist standard deviation scores, but the annual rate of increase was lower than among Caucasian students.
Shaw et al.(42)Caucasian, South Asian, African– CaribbeanDXA body fat percentage; overweight and obesity BMI according to IOTF(44) cut-offs)Birmingham and Middlesbrough1251 healthy children and adolescents aged 5–18.A cross-sectional analysis of body fat percentage differences in adolescents by ethnic group.African–Caribbean children were significantly taller and heavier than the South Asian and Caucasian children. Both African–Caribbean and Caucasian children were significantly taller than UK 1990 references, but South Asians were significantly shorter. At 5 years old, South Asian girls had the highest percent body fat, which was significantly higher than African–Caribbeans, but at age 15, South Asian girls had significantly higher values than both other groups. Among boys, South Asian boys had significantly higher values than African–Caribbeans by age 7, and significantly higher values than both other groups by age 15. African–Caribbean and Caucasian groups showed a reduction in body fat after 12 years among boys. South Asian boys and girls were significantly over-represented in the group with 25% or higher body fat percentage, while using BMI cut-offs, African–Caribbeans were significantly over-represented in the ‘obese’ category using IOTF cut-offs.
Balakrishnan et al.(33)Caucasian, Afro-Caribbean and South AsianOverweight (>85 percentile) and obesity (>95% percentile) according the UK National BMI percentile classificationEast Berkshire health system16 364 Caucasian and South Asian children (with anthropometric and ethnicity details recorded) between 5–7 years old born between 1991 and 1999 in the East Berkshire Child Health System of 51 565 total.A mixed cross-sectional analysis of prevalence trends in overweight and obesity by year of birth, sex and ethnicity.In logistic regression models, South Asian boys were more likey to be overweight and obese than South Asian girls. No significant differences were found among other groups. South Asian boys also had higher risk for overweight (OR = 1.77, 95% CI 1.56–2.00) and obesity (OR = 1.76, 95% CI 1.50–2.06) than Caucasian boys. South Asian girls (OR = 0.86, 95% CI 0.75–0.99) had lower risk for overweight than Caucasian girls. There were no significant differences in overweight and/or obesity risk by gender among Afro-Caribbean children, nor between Afro-Caribbean children and Caucasian children. There was no significant difference in increases in overweight or obesity between Caucasians and South Asians.
Rutter (37)Caucasian, mixed, Chinese/other, Black or black British, Asian or Asian British, not statedObesity (using IOTF (44) BMI cut-offs)UKOver 876 000 children who were measured at ages 4–5 or 10–11 in 2006–2007.A cross-sectional analysis of obesity among children in the UK.Black ethnic groups had the highest prevalence of obesity, followed by mixed, and then Chinese/other. South Asians also had a higher prevalence of obesity than Caucasians.
Harding et al.(43)Caucasian (UK), Caucasian (other), mixed, Black Caribbean, Black African, Indian, Pakistani/ BangladeshiOverweight and obesity (using IOTF (44) BMI cut-offs)London6407 children aged 11–13 between September 2002 and July 2003 from 51 schools in 11 London Boroughs.A cross-sectional analysis of the relations between overweight, obesity and high blood pressure among adolescents in London.Among boys, prevalence of overweight was 19.7% among Caucasian (UK) boys, 23.0% among Caucasian (other), 15.4% among Black Caribbeans, 17.5% among Black Africans, 19.6% among Pakistani/Bangladeshi, and 21.4% among mixed. Prevalence of obesity was 6.4% among Caucasian (UK), 9.2% among Caucasian (other), 10.3% among Black Caribbeans, 8.2% among Black Africans, 6.8% among Indians, 8.4% among Pakistani/Bangladeshi, and 8.2% among Mixed. Among girls, prevalence of overweight was 20.2% among Caucasian (UK) girls, 20.6% among Caucasian (other), 28.1% among Black Caribbeans, 27.9% among Black Africans, 20.8% among Pakistani/Bangladeshi, and 17.9% among mixed. Prevalence of obesity was 6.6% among Caucasian (UK), 7.2% among Caucasian (other), 12.6% among Black Caribbeans, 9.9% among Black Africans, 4.2% among Indians, 4.1% among Pakistani/Bangladeshi, and 8.9% among Mixed.
Harding et al.(36)Caucasian (UK), Caucasian (other), mixed, Black Caribbean, Black African, Indian, Pakistani/ BangladeshiOverweight and obesity (using IOTF (44) BMI cut-offs)London6599 children aged 11–13 between September 2002 and July 2003 from 51 schools in 11 London Boroughs.A cross-sectional analysis of the relations between ethnicity, obesity-related behaviours, and obesity among adolescents in London.Black Caribbean girls (12.6%) and boys (10.1%) had the highest prevalence of obesity. Caucasian UK boys (6.2%) and Indian (4.1%) and Pakistani/Bandladeshi girls (4.1%) had the lowest prevalence of obesity. In models adjusted for age, height, and pubertal status, Black Caribbean girls had significantly higher odds of overweight (OR = 1.38, 95% CI 1.02–1.87) and obesity (OR = 1.65, 95% CI 1.05–2.58) than Caucasian (UK) girls, and Black African girls had higher risk for overweight than Caucasian (UK) girls (OR = 1.35, 95% CI 1.02–1.79). Among boys, Black Caribbean boys had significantly lower risk for overweight than Caucasian (UK) boys, and Caucasian (other) boys had significantly higher risk for obesity than Caucasian (UK) boys. In models further adjusted for parental influences, generation status, family type and SES, Black Caribbean and Black African girls had higher risk for overweight and obesity than Caucasian (UK) girls, and Caucasian (other) girls. Black Caribbean and Pakistani/Bangladeshi boys had higher risk for obesity than Caucasian (UK) boys.
Hawkins et al.(35)Caucasian, mixed, Indian, Pakistani, Bangladesh, Black, otherOverweight and obesity (using IOTF (44) BMI cut-offs)UK13 188 singleton children aged 3 years in the Millennium Cohort study, born between 2000 and 2002, who had complete weight and height data.A cross-sectional analysis of the relation between individual, family, community, and area-level deprivation and overweight and obesity among children.In unadjusted models, black ethnicity was significantly associated with 40% higher odds of obesity, Indian ethnicity was associated with 65% lower odds of obesity, and Pakistani ethnicity was associated with 25% lower odds of obesity relative Caucasian ethnicity. After fully adjusting for individual, family, community and area-level factors, black ethnicity was significantly associated with (OR = 1.41, 95% CI 1.11–1.80) higher odds of overweight and Indian ethnicity was associated with (OR = 0.63, 95% CI 0.42–0.94) lower odds of overweight relative to Caucasian ethnicity.
Age group: adults
McKeigue et al.(19)European, South Asian (further categorized as Sikh, Punjabi Hindu, Gujarati Hindu, and Muslim), Afro-Caribbean (men only)BMI; subscapular folds; suprailiac folds; waist-hip ratioLondon3193 men and 561 women aged 40–69 recruited from Industrial workforces and general practitioners’ lists.A cross-sectional analysis of ethnic differences in the relation between obesity and insulin resistance, diabetes, and cardiovascular disease risk.Afro-Caribbean men had the highest mean BMI (26.3 kg m−2), and South Asians the lowest. South Asian men had significantly higher waist-hip ratios than both European and Afro-Caribbean men. South Asian women had a significantly higher mean BMI (27.0 kg m−2) and waist-hip ratio than European women. South Asian men and women both had significantly larger suprailiac and subscapular skin-folds than all other groups.
Bose (47)Caucasian, Muslim PakistaniMean BMI, minimum waist circumference, maximum hip circumference, sum of all circumferences, waist/abdomen ratio, chest/waist ratio, chest/abdomen ratio, chest/hip ratio, chest/sum of all ratio, abdomen/sum of all ratioPeterborough262 Caucasian and 100 Muslim Pakistani migrants ≥20 years old from three general practices in Peterborough.A cross-sectional analysis of ethnic differences in obesity and fat distribution among Caucasian and Muslim Pakistani men.9.9% of Caucasians and 11% of Pakistanis had BMI between 30 and 40. More Pakistanis (49%) had appropriate weights-for-height than Caucasians. There was no significant difference in obesity risk between Caucasians and Pakistanis. Pakistanis had a significantly higher mean abdomen/sum of all circumferences ratio. Caucasians had significantly higher upper body, lower body, upper body relative to central body, upper body relative to lower body and total adiposity than Pakistanis, but Pakistanis had significantly more fat in the abdomen relative to total adiposity than Caucasians.
McKeigue et al.(46)European, South Asiansubscapular/triceps; subscapular/anterior thigh; abdominal diameterLondon2936 men and 537 women aged 40–69 of European and South Asian descent.A cross-sectional analysis of ethnic differences in the relation between obesity and glucose intolerance.South Asians men and women had significantly higher abdominal diameter, subscapular/triceps and subscapular/anterior thigh measurements than Europeans.
Knight et al.(54)Asian, non-Asianoverweight (25 < BMI < 30 kg m−2) and obese (BMI ≥ 30 kg m−2)Bradford288 male manual workers in two textile factories aged 20–65 years.A cross-sectional analysis of differences in cardiovascular disease risk factors by ethnicity.A higher proportion of non-Asian men were categorized as obese (8.1%) or overweight (40.0%) compared to Asian men (3.9% and 38.3%, respectively), however, the differences were not statistically significant.
Bose (56)Caucasian, PakistaniBMI (>30 kg m−2) and multiple skin-foldPeterborough262 Caucasian and 100 Muslim Pakistani migrants ≥20 years old from three general practices in Peterborough.A cross-sectional analysis of ethnic differences in anatomic fat mass accumulation between Caucasian and Pakistani men.11.0% of Pakistani and 9.9% of Caucasian men were obese, but there was no statistical difference between these prevalence. Pakistani men had significantly higher truncal skin-folds and subcutaneous adiposity compared to Caucasians, and Caucasians had significantly higher upper extremity subcutaneous adiposity than Pakistani men.
Harland et al.(22)European, ChineseOverweight (BMI ≥ 25 kg m−2 in women and 27 kg m−2 in men), obesity (BMI ≥ 30 kg m−2), mean BMI, waist/hip ratio, waist circumferenceNewcastle-upon-Tyne380 Chinese and 625 European adults aged 25–64 years old.A cross-sectional analysis of differences in cardiovascular disease risk factors among Chinese and European adults.Among men, Chinese significantly lower average BMI at 23.8 kg m−2 average, compared to 26.1 kg m−2 among Europeans. Chinese men also significantly lower waist/hip ratio, and waist circumference. Among women, Chinese women had significantly lower average BMI (23.5 kg m−2 compared to 26.1 kg m−2). Chinese women also had significantly lower waist circumference but significantly higher waist/hip ratio. Both Chinese men and women were significantly less likely to be obese (4.5% and 2.1%, respectively) compared to European men and women (14.5% and 15.8%, respectively). Chinese men and women were also significantly less likely to be overweight (17.1% and 31.3%, respectively) compared to European men and women (35.7% and 51.4%, respectively).
Cappuccio et al.(51)Caucasian, African ancestry (Caribbean and West African), South Asian origin (Hindus and Muslims)Obesity (BMI ≥ 27.0 kg m−2) and severe obesity (BMI ≥ 30.0 kg m−2)Former Wandsworth Health Authority in South LondonPopulation-based survey of 1578 men and women without cancer, psychiatric disorders, severe disability or pregnancy aged 40–59 years (524 Caucasian, 549 African descent, 505 South Asian origin) during 1994–1996.A cross sectional analysis of ethnic differences in prevalence, detection, and management of cardiovascular risk factors.Among men, age-adjusted prevalence of obesity was 30.5% among Caucasians, 38.7% among Africans, and 24.2% among South Asians. Among women, prevalence of obesity was 34.3% among Caucasians, 67.9% among Africans, and 48.0% among South Asians. Prevalence of severe obesity among men was as follows: 14.8% among Caucasians, 14.8% among Africans, and 8.4% among South Asians. Among women prevalence of severe obesity was 18.8% among Caucasians, 39.8% among Africans, and 19.7% among South Asians. Among Africans, West African men had higher risk for obesity (40.5%) than Caribbean men (34.2) and higher risk for severe obesity (13.7%) than Caribbeans (12.2%). Among women, West Africans had higher risk for obesity (72.0%) and severe obesity (40.3) than Caribbean women (65.5% and 39.9%, respectively). Among South Asian men, Muslims had higher risk for obesity (26.2%) and severe obesity (8.9%) than Hindus (21.1% and 6.6%, respectively). Among South Asian women, Muslims also had higher risk for obesity (54.7%) and severe obesity (24.3%) than Hindus (42.1% and 15.6%, respectively). African women had significantly higher risk for obesity and South Asian men had significantly lower age-adjusted prevalence of obesity than Caucasians.
Unwin et al.(55)European, ChineseMean BMI; waist circumference; waist/hip ratioNewcastle-upon-Tyne375 Chinese and 610 European men and women aged 25–64.A cross-sectional analysis of the relation between obesity and glucose intolerance among Chinese and European adults.Age adjusted mean BMI and waist circumferences were significantly lower among Chinese men and women compared to European men and women. However, waist/hip ratios among Chinese men were significantly lower than among European men, while those among Chinese women were significantly higher than among European women.
Bhopal et al.,1999 (48)Indian, Pakistani, Bangladeshi, and EuropeanOverweight (BMI 25.0–29.9 kg m−2 and waist-to-hip ratio ≥ 0.95) and Obese (BMI ≥ 30.0 kg m−2)Newcastle-upon-Tyne249 Indian, 305 Pakistani, 120 Bangladeshi and 825 European men and women aged 25–74 years identified from the family health services authority register for the ‘Newcastle Health and Lifestyle Survey’.A cross-sectional analysis of cardiovascular disease risk factors among Indian, Pakistani, Bangladeshi and Europeans.Among men, 69% of Pakistanis, 66% of Indians, and 47% of Bangladeshis were overweight (by BMI) compared to 56% among Europeans. 57% of Indians, 61% of Pakistanis, 63% of Bangladeshis had Waist-to-hip ratios ≥ 0.95 as compared to 27% of Europeans. Among women, 69% of Indians, 66% of Pakistanis and 58% of Bangladeshis were overweight (by BMI) compared to 52% of Europeans. 38% of Indians, 34% of Pakistanis and 15% of Bangladeshis were obese compared to 16% of Europeans. 15% of Indians, 12% of Pakistanis and 5% of Bangladeshis had waist-to-hip ratio ≥ 0.95 compared to 17% of Europeans. Overall, among women, Indians had the highest risk for obesity among South Asians, but Europeans had significantly higher obesity risk than South Asians. Among men, Indians had the highest risk for obesity among all South Asians, and South Asians had significantly higher risk than Europeans.
Wardle et al.(53)Caucasian, Black, Asian, OtherObesity (BMI ≥ 30 kg m−2)England15 061 individuals sampled in the1996 Health Survey for England.A cross-sectional analysis of the relation between socioeconomic status and obesity.Black women had significantly higher odds of obesity in both bivariate and multivariate models (including age at last education, occupational status, benefits, housing, age, marital status and ethnicity). Among men, Asians and those of ‘other’ ethnicity had lower odds of obesity than Caucasians in bivariate analysis, but were not significantly different in mutlivariate models.
Vyas et al.(49)Pakistani, European, African–CaribbeanMean BMI, waist/hip ratio, mean hip circumference, mean waist circumferenceInner City Manchester86 Europeans, 246 African–Caribbeans, and 84 Pakistanis aged 25–79 years from 7 general practice registers.A cross-sectional analysis of nutrient intakes by ethnic group.Mean BMI was highest among European men (29.7 kg m−2) and Pakistani women (30.2 kg m−2). There were no significant differences in mean BMI by ethnicity among men, while Pakistani women had significantly higher mean BMI than European women. Both European and Pakistani men had significantly higher mean waist/hip ratios than African–Caribbean men, while Pakistani women had significantly higher ratios than European and African–Caribbean women. European men had significantly higher mean hip circumferences than other groups, while there were no significant differences in mean hip circumference among women. African–Caribbean men had significantly lower waist circumferences than other men, while Pakistani women had significantly higher waist circumferences than other women.
Rennie and Jebb (30)Black Caribbean, Indian, Pakistani, Bangladeshi, Chinese, Irish and General PopulationObesity (BMI ≥ 30.0 kg m−2)England3204 men and 3699 women aged 16–55+ years old sampled in the 1999 Health Survey for England.A cross-sectional analysis of obesity in England.Black Caribbean women had the highest prevalence of obesity (31.9%), while Chinese women had the lowest (4.5%). Among men, Irish men had the highest prevalence of obesity (20.4%), and Bangladeshi men had the lowest (5.4%).
Miller and Cappuccio (50)Caucasian, African origin, South AsianBMI; waist/hip ratioSouth London261 Caucasian (120 women), 188 African origin (99 women), and 215 South Asian (99 women) individuals who took part in the ‘Wandsworth Heart and Stroke Study’.A cross-sectional analysis of the relations between metrics of obesity and blood concentration of adhesion molecules in a multiethnic population.There were significant differences in both age and sex-adjusted BMI and waist/hip ratio by ethnic group. Individuals of African origin had the highest BMI (27.3 kg m−2) and South Asians the lowest (25.5 kg m−2). However, South Asians had the highest waist/hip ratios and Caucasians the lowest.
Moon et al.(78)Caucasian, Black, AsianOverweight (BMI 25.0–29.9 kg m−2) and obesity (BMI ≥ 30.0 kg m−2)England18 526 individuals sampled in the 1998–1999 Health Surveys for England.A cross-sectional multilevel analysis of variation in overweight and obesity by locality in England.Estimated obesity prevalence was highest among ethnic Blacks (19.7%) and lowest among ethnic Asians (16.9%) relative to ethnic Caucasians (18.3%). However, ethnic Caucasians had highest prevalence of overweight (42.2%) compared to 38.0% among ethnic Asians, and 40.1% among ethnic Blacks. There were no significant differences in estimated prevalence of either obesity or overweight by ethnic group. Ethnic Asians and Blacks in urban areas had a higher prevalence of obesity than those in other areas, while Caucasians in rural areas had higher prevalence of obesity.
Diaz et al.(52)Caucasian, Black, Bangladeshi, Pakistani, Indian, ChineseObesity (BMI)England2003 and 2004 Health Surveys for England population ≥ 20 years old.A cross-sectional analysis of the utility of metrics of obesity in assessing risk for diabetes across ethnic groups.Blacks had the highest mean BMI (28.5 kg m−2) and was significantly higher than among Caucasians, while Chinese had the lowest (24.0 kg m−2). Chinese, Banglandeshi and Indian populations had significantly lower mean BMI than Caucasians.

The studies included in this review utilized two empirical study designs: 2 studies included in this review were longitudinal analyses, while the remaining 27 were cross-sectional in nature. The majority of the studies (18) reported on parochial datasets from localities throughout the UK (London, Peterborough, etc.); eight of these reported on data from London or one of its boroughs. The remaining studies reported on representative data from at least one British country (England, Scotland, Wales, Northern Ireland or a combination). Only two studies about ethnic disparities in obesity included data from Northern Ireland, and each were reporting on UK-wide data; there were five studies that reported on data from Scotland, and four that reported on data from Wales. All 29 studies reported on data from England.

Ethnic disparities in obesity among children and adolescents

The literature about ethnic disparities in obesity among children and adolescents includes 13 studies that assessed differences in prevalence of metrics of obesity among South Asian subgroups, Black subgroups, Chinese and ‘other’ relative to Caucasians in the UK. One study assessed differences in predictors of metrics of obesity by ethnicity.

Childhood obesity among South Asians in the UK

Among South Asian children, there is no consensus with regard to obesity prevalence relative to Caucasians and other ethnic minority groups (31–42). Several studies have suggested that South Asian children have significantly higher risk for metrics of obesity compared to Caucasians and other ethnic minority groups in the UK (31–34,37,39–42), while others have suggested that South Asian children have significantly lower risk (32,35,36,38,40,43), with some studies showing differences in risk for obesity relative to Caucasians by gender (32,40).

Among those studies that suggest that South Asians have higher risk is a study by Jebb and colleagues(34) which considered ethnic differences in overweight and obesity risk among a representative sample of 1836 children in England, Scotland and Wales aged 4–18 in 1997 using International Obesity Taskforce (IOTF) body mass index (BMI) cut-offs (44). The study found that South Asian respondents had almost four times higher prevalence of obesity compared to Caucasians (13.6% compared with 3.5%, P < 0.001).

Several of these studies have noted higher risk for obesity among South Asian boys, in particular. Saxena and colleagues(32) using data about 2- to 20-year-old young people from the 1999 Health Survey for England found that Indian and Pakistani boys had the highest prevalence of overweight (30% and 26%, respectively) and obesity (using IOTF(44) BMI cut-offs) (7.9% and 9.0%, respectively) among boys. Balakrishnan and colleagues (33), found that among 5- to 7-year-old Caucasian, Afro-Caribbean, and South Asian children born between 1991 and 1999 in East Berkshire using UK National BMI percentile classifications, South Asian boys had higher risk for overweight (odds ratio [OR] = 1.77, 95% confidence interval [CI] 1.56–2.00) and obesity (OR = 1.76, 95% CI 1.50–2.06) than Caucasian boys. These findings have also been supported in other studies (31,32,40,45).

A few studies have noted higher obesity prevalence among both South Asian boys and girls. For example, using dual energy X-ray absorbtiometry (DXA) among 1251 healthy children aged 5–18 in Birmingham and Middlesbrough, Shaw and colleagues(42) found that among 5 year olds, South Asian girls had significantly higher adiposity than Black African girls, and by 15 years, they had significantly higher adiposity than Black African and Caucasian girls. Among boys, by age 7, South Asians had significantly higher adiposity than Black African boys, and by 15 years, they had significantly higher adiposity than Black African and Caucasian boys.

Studies have also suggested that South Asian children may have lower risk for obesity metrics than other groups (32,35,36,38,40,43). For example, data from the Millennium Cohort Study among 13 188 3-year-old children showed that in models adjusted for individual, family, community and area-level factors, Indian children had lower risk for overweight (using IOTF (44) BMI cut-offs) (OR = 0.63, 95% CI 0.42–0.94) compared to Caucasian children (35).

Several studies found that South Asian girls, in particular, had lower obesity prevalence relative to other ethnic groups. For example, Saxena and colleagues (32), in a previously described study, found that Bangladeshi boys had the lowest prevalence (2.8%) of obesity (using IOTF(44) cut-offs), and that Indian (OR = 0.39, 95% CI 0.17–0.86) girls had significantly lower odds of obesity relative to the general population. Two studies by Harding and colleagues(36,43) considered ethnic differences in overweight and obesity risk among over 6400 children aged 11–13 from 51 schools in 11 London Boroughs using IOTF(44) BMI cut-offs. They found that Indian and Bangladeshi girls (4.1%, respectively) had the lowest prevalence of obesity among all girls.

One study considered predictors of triceps skin-fold standard deviation scores and weight-for-height among South Asian (by subgroup) relative to Caucasian children. Predictive factors were similar among Caucasians and South Asian groups for triceps skin-folds. However, there were substantial differences in predictors of weight-for-height among South Asian and Caucasian children (40). Among Caucasian children, mother and father's weights and heights, child birth weight, and number of siblings were predictors of weight-for-height. Predictors among Punjabi children were similar to among Caucasians. Among Urdu children, there were no significant predictors. Among Gujarati children, only father's weight and height and child's birth weight were predictors. Among other South Asians, school meals, mother's weight, and father's weight and height were predictors (40).

Childhood obesity among Black Africans and Black Caribbeans in the UK

Like South Asians, the literature about childhood obesity among Black Africans and Black Caribbeans is contested. Several studies have demonstrated higher risk among Black subgroups (31,32,35–37,42,43), and others have demonstrated lower risk among Black subgroups relative to Caucasians in the UK (40–42,45), with one study showing differences in risk for obesity relative to Caucasians by gender (42). Still others have demonstrated no difference between Black subgroups and Caucasians (34).

Among studies that have shown that Black subgroups have higher risk for obesity is a study by Hawkins using data from the UK-representative Millenium Cohort Study, which found that Black subgroup children had 40% higher unadjusted odds of obesity and 41% higher adjusted odds of overweight (using IOTF(44) cut-offs) than Caucasians (35). Two studies by Harding and colleagues(36,43) that found that the prevalence of obesity (using IOTF(44) cut-offs) was highest among Black Caribbean girls (12.6%) and boys (10.1%) aged 11–13. After adjusting for age, height, and pubertal status, Black African girls had significantly higher overweight (OR = 1.38, 95% CI 1.02–1.87) and obesity (OR = 1.65, 95% CI 1.05–2.58) risk than Caucasian (UK) girls, and Black African girls had higher risk for overweight than Caucasian (UK) girls (OR = 1.35, 95% CI 1.02–1.79). Although Black Caribbean boys had significantly lower risk for overweight than Caucasian (UK) boys in these models, after further adjusting for parental influences, generation status, family type, and socioeconomic status, both Black African and Caribbean boys had higher risk for obesity than Caucasian (UK) boys.

Two studies found that Black subgroup girls, in particular, had higher risk for obesity than Caucasians. Saxena and colleagues(32) found that Black African girls had the highest obesity prevalence (13%) among girls. In multivariate models adjusted for social class, Afro-Caribbean girls had higher odds of overweight (OR = 1.79, 95% CI 1.29–2.33) and obesity (OR = 2.74, 95% CI 1.74–4.31) than the general population. Wardle and colleagues(31) found that Black girls were significantly more likely to be obese (according to IOTF(44) BMI cut-offs) than Caucasian or Asian girls.

Findings from other studies, however, have suggested that Black subgroups have lower risk for obesity. Shaw and colleagues(42) assessed obesity by DXA and found that African–Caribbean boys and girls had lower adiposity than Caucasians at all ages, and that this was significantly different at most ages. Chinn and colleagues(41) assessed trends in weight-for-height and triceps skin-folds and found that Afro-Caribbeans were slimmest overall.

One study found that Black African boys, in particular, had lower risk for obesity: Findings from East London showed that Black African boys had lower mean BMI than Caucasian British boys, and in multivariate models of overweight and obesity by IOTF(44) cut-offs adjusted for potential confounders, Black African boys (OR = 0.56, 95% CI 0.34–0.88) had significantly lower risk of overweight (45).

Two studies found no difference in obesity risk between Black subgroups and Caucasians in the UK. Jebb and colleagues(34) found no difference in obesity (using IOTF(44) BMI cut-offs) risk between Black subgroups and Caucasians among a representative sample of 1836 children in England, Scotland, and Wales aged 4–18 in 1997. Duran-Tauleria and colleagues(39) found similar results.

One study(40) considered predictors of triceps skin-fold standard deviation scores and weight-for-height among Afro-Caribbean relative to Caucasian children. Predictive factors were similar among Caucasians and Afro-Caribbeans for triceps skin-folds. However, for weight-for height measures, father's height and weight and number of siblings were not predictive among Afro-Caribbean children, although they were among Caucasian children.

Childhood obesity among Chinese and ‘other’ ethnic groups in the UK

Relatively few studies have considered childhood obesity among Chinese and/or ‘other’ ethnic groups in the UK. There was one study concerned with childhood obesity among Chinese in the UK. Saxena and colleagues(32) found that Chinese girls had the lowest (1.2%) prevalence of obesity (using IOTF(44) BMI cut-offs) overall. Chinese girls had significantly lower odds of overweight (OR = 0.52, 95% CI 0.29–0.91) and obesity (OR = 0.08, 95% CI 0.01–0.56) than the general population. Several studies considered obesity prevalence among ethnically ‘other’ children in the UK. For example, Hawkins and colleagues(35) found no difference in risk for overweight (using IOTF(44) BMI cut-offs) between ethnic ‘others’ and Caucasians in the UK. However, two studies by Harding and colleagues(36,43) found that ethnic Caucasian ‘other’ boys had significantly higher overweight and obesity risk (using IOTF(44) BMI cut-offs) than Caucasian (UK) boys, and Rona and Chinn(40) found similar results.

One study considered predictors of triceps skin-fold standard deviation scores and weight-for-height among ‘others’ relative to Caucasian children. Predictive factors were similar among Caucasians and others for triceps skin-folds. However, for weight-for-height measures, there were no significant predictors of weight-for-height among ‘others’, although maternal and paternal height and weight, child's birth weight, and number of siblings were all significant predictors of weight-for-height among Caucasians (40).

Ethnic disparities in obesity among adults

The literature about ethnic disparities in obesity among adults includes 15 studies concerned with differences in the prevalence of metrics of obesity among South Asian subgroups, Black subgroups, Chinese and ‘others’ relative to Caucasians in the UK. We found no studies that assessed differences in the predictors of adult obesity by ethnic group.

Adult obesity among South Asians in the UK

At present, the literature about the relation between South Asian ethnicity and obesity risk relative to Caucasians among adults is equivocal. Several studies have shown that South Asian adults have higher risk for metrics of obesity than Caucasians in the UK (19,46–50), while others have demonstrated lower risk compared to Caucasians (30,48–52), and one showed no significant differences (53). Studies have also demonstrated differences in prevalence of obesity by nationality among South Asian subgroups (30,51,52).

Several studies have demonstrated higher risk for metrics of obesity among South Asians relative to Caucasians in the UK. For example, two studies among a cross-sectional sample of 3193 men and 561 women aged 40–69 in London by McKiegue and colleagues(45,46) found that Asian men had significantly higher waist/hip ratios than both Afro-Caribbean and Caucasian men and that Asian women had significantly higher mean BMI (27.0 kg m−2) and waist/hip ratios than Caucasian women. South Asians also had significantly larger subscapular and suprailiac skin-folds than all other groups (19). Asian men and women both had significantly higher abdominal diameter, subscapular/triceps skin-fold ratio and subscapular/anterior thigh skin-fold ratio than Caucasians (46). These findings have been supported by others, as well (30,48,49)

Among studies demonstrating lower risk for metrics of obesity among South Asian adults is a study by Diaz and colleagues(52) who examined data from the 2003–2004 Health Surveys for England and found that Bangladeshi and Indian populations had significantly lower mean BMI than Caucasians. Cappuccio and colleagues(51) assessed ethnic differences in obesity (BMI > 27 kg m−2) and severe obesity (BMI > 30 kg m−2) prevalence among Caucasians, South Asians, and Black subgroups among 1578 men and women aged 40–59 in South London and found that South Asian men had significantly lower age-adjusted prevalence of obesity than Caucasians. Other studies have also shown similar findings (50,53).

One study found no difference in overweight (BMI > 25 kg m−2) or obesity (BMI > 30 kg m−2) risk between Caucasians and South Asians, although prevalence of each metric was higher among Caucasians in a study of 288 male South Asian and Caucasian manual workers aged 20–65 years in Bradford (54).

Several studies have also assessed differences among South Asians by ethnic origin. For example, Diaz and colleagues(52) found that although Indians and Bangladeshis had lower mean BMI than Caucasians, Pakistanis did not. Rennie and Jebb(30) found that among South Asian women, Pakistani women, at 25.6%, had the highest risk for obesity (BMI > 30 kg m−2), while 19.6% of Indian women were obese, and 9.5% of Bangladeshi women were obese. Cappuccio and colleagues(51) showed that among South Asian men, Muslims had significantly higher risk for obesity(BMI > 27 kg m−2) (26.2%) and severe obesity (BMI > 30 kg m−2) (8.9%) than Hindus (21.1% and 6.6%, respectively). Similarly, among South Asian women, Muslims also had significantly higher risk for obesity (54.7%) and severe obesity (24.3%) than Hindus (42.1% and 15.6%, respectively). Bhopal and colleagues(48) also showed that among Indians, Pakistanis, and Bangladeshis, Indian men and women had highest risk for obesity (BMI > 30 kg m−2).

Adult obesity among Black Africans and Black Caribbeans in the UK

Black adults in the UK have higher prevalence of metrics of obesity than Caucasian adults (46,50–52). As several studies have demonstrated this relation (46,50–52), while only one study found significantly lower waist circumference among African–Caribbean men relative to Caucasian men (49).

Among studies that showed higher prevalence of obesity among Black adults is a study using data from the 2003–2004 Health Surveys for England, which showed that Black groups had the highest mean BMI (28.5 kg m−2), which was significantly higher than among Caucasians (52). Another study by Cappuccio and colleagues(51) among adults in South London showed that among men, age-adjusted prevalence of obesity (BMI > 27 kg m−2) was 38.7% among Africans compared with 30.5% among Caucasians, and prevalence of severe obesity (BMI > 30 kg m−2) among men was 14.8% among Africans and Caucasians, respectively. Differences among men were not significant. Among women, prevalence of obesity was 67.9% among Africans, which was significantly higher than among Caucasians (34.3%). Prevalence of severe obesity was 39.8% among African women, which was significantly higher than 18.8% among Caucasian women. Only one study, by Vyas and colleagues (49), found that African–Caribbean men had significantly lower waist circumferences than men from other groups.

One study considered differences in the prevalence of metrics of obesity between Black Africans and Black Caribbeans and found no significant differences: Cappuccio and colleagues(51) assessed obesity (BMI > 27 kg m−2) and severe obesity (BMI > 30 kg m−2) among Black subgroups in South London. West African men had higher risk for obesity (40.5%) and severe obesity (13.7%) than Caribbean men (34.2% and 12.2%, respectively). Among women, West Africans had higher risk for obesity (72.0%) and severe obesity (40.3%) than Caribbean women (65.5% and 39.9%, respectively). Neither finding was significant.

Adult obesity among Chinese and ‘other’ ethnic groups in the UK

Several studies were concerned with metrics of obesity among Chinese adults in the UK relative to Caucasians. For example, Diaz and colleagues found that Chinese adults had the lowest mean BMI among all groups studied (including Blacks, Caucasians, Indians, Pakistanis and Bangladeshis) using data from the 2003 and 2004 Health Surveys for England (52). In more detailed analyses, Harland and colleagues(22) and Unwin and colleagues(55) studied ethnic differences in overweight (BMI > 25 kg m−2 in men and 27 kg m−2 in women) obesity (BMI >  30 kg m−2) waist circumference and waist/hip ratios among Caucasians and ethnic Chinese in Newcastle-upon-Tyne. They found that mean BMI was significantly lower among Chinese men (23.8 kg m−2) compared with Caucasian men (26.1 kg m−2). Waist/hip ratio and waist circumference were also significantly lower among Chinese men compared to Caucasian men. Chinese women also had significantly lower waist circumference but significantly higher waist/hip ratio. Both Chinese men and women were significantly less likely to be obese (4.5% and 2.1%, respectively) compared to European men and women (14.5% and 15.8%, respectively). Chinese men and women were also significantly less likely to be overweight (17.1% and 31.3%, respectively) compared to European men and women (35.7% and 51.4%, respectively). Findings from Rennie and Jebb(30) support these results.

One study considered obesity among ‘other’ ethnic grouping relative to Caucasians. In unadjusted models from 1996 Health Survey for England data, ‘others’ had lower risk for obesity (BMI > 30 kg m−2) compared with Caucasians in unadjusted models. However, after adjusting for socioeconomic factors, marital status and age, the association was attenuated (31).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

We reviewed the peer-reviewed literature about ethnic inequalities in obesity in the UK between 1980 and 2010. Overall, there was little consensus in the literature about obesity risk among South Asian and/or Black children relative to Caucasian children in the UK. There was also little consensus with regard to obesity risk among South Asian adults relative to Caucasians. However, we found that the literature about the prevalence of obesity among Black adults suggests a higher prevalence than Caucasians in the UK. Also, both Chinese children and adults appear to have lower risk for obesity than Caucasians in the UK. ‘Other’ children appear to have higher risk for obesity than their Caucasian counterparts, while there appears to be no differences between ‘other’ adults and Caucasians in obesity prevalence. Our findings, taken together, also allow for several inferences with regard to obesity risk among these groups.

Obesity among South Asians

With regard to obesity risk among South Asians, our findings lead to several inferences. First, the findings suggest a possible sexual dimorphism with regard to obesity risk among South Asian children relative to Caucasians. Five studies found heightened obesity risk among South Asian boys, in particular (31–33,40), and three found lower risk among girls, in particular (32,36,43). This suggests that South Asian boys may have higher risk for obesity relative to Caucasians while South Asian girls may have lower risk for obesity relative to Caucasians in the UK. However, this dimorphism seems to shift among South Asian adults – several studies suggest a sexual dimorphism in risk for obesity among South Asian adults relative to Caucasian adults that is opposite to the dimorphism observed among children, described above (19,30,49).

Second, our observations suggest that the choice of obesity metrics may bias estimates of obesity prevalence among South Asian children and adults. Two studies highlight this tendency among children. Shaw and colleagues(42) who assessed obesity by DXA as well as by IOTF(44) BMI cut-offs showed that South Asian boys and girls had significantly higher adiposity than Black Africans at young ages, and by 15, had significantly higher adiposity than Black Africans and Caucasians of the same age. However, neither group was overrepresented in the ‘obese’ category according to IOTF(44) BMI cut-offs. Another study by Rona and Chinn(41) showed similar findings. These findings suggest that weight-for-height metrics, such as the BMI, may systematically underestimate obesity relative to adiposity metrics, such as DXA or skin-fold measurements among South Asian children in the UK.

We made similar observations with regard to adult obesity among South Asians relative to Caucasians in the UK. Our findings among adults suggest that South Asian adults in the UK may accumulate more weight around the abdomen, and have greater adiposity in general, than Caucasians, as all of the studies concerned with ethnic differences in obesity between South Asian adults and Caucasian adults in the UK which measured waist circumference or waist/hip ratio as metrics of obesity found that South Asian adults had significantly larger waists and higher waist/hip ratios(19,47–50) than their Caucasian counterparts. Moreover, each of the studies reviewed above that included skin-fold measurements of any kind found that South Asian adults had significantly greater adiposity than Caucasian adults (19,47,56). These findings, among both children and adults, are in agreement with the existing literature about obesity among South Asian populations, which led to a revision of World Health Organization recommendations about appropriate overweight and obesity cut-off levels for BMI among South Asians (57).

Obesity among Black Africans and Black Caribbeans

Our findings lead to several inferences about risk for childhood and adulthood obesity among Black subgroups in the UK relative to Caucasians. First, like South Asians, there may be sexual dimorphism with regard to obesity risk among Black subgroups in the UK among both children and adults. Our findings suggest that Black subgroup girls may have higher risk, and boys, lower risk of obesity compared to Caucasian girls and boys, respectively, in the UK.

The available evidence also suggests that this sexual dimorphism in obesity risk relative to Caucasians may remain in adulthood among Blacks. For example, none of the studies reviewed above which stratified by sex found significantly higher risk for obesity metrics among Black subgroup men relative to Caucasians, although all of them found significant differences between Black subgroup and Caucasian women (45,51,53).

Also similar to South Asians, some metrics of obesity may bias prevalence estimates among Black children and adults. For example, among children, Shaw and colleagues(42) assessed obesity by DXA as well as IOTF(44) BMI cut-offs and found that while African–Caribbean children were overrepresented in the ‘obese’ category according to IOTF(44) BMI cut-offs, both boys and girls had significantly lower adiposity by DXA scan than Caucasians at most age categories. It is plausible that among children in Black subgroups, weight-for-height based metrics of obesity, such as the BMI, may systematically overestimate obesity relative to adiposity metrics, such as the DXA scan. Similar observations have been made among a Black sample in the USA (58). These findings hold among Black adults. Moreover, contrary to findings among South Asian adults, Black adults may accumulate less weight around the abdomen than Caucasians (49,50).

Methodological limitations

There are six important methodological limitations to our understanding of obesity among ethnic minority groups compared to Caucasians in the UK: (i) the paucity of studies in the literature that considered predictors of obesity among ethnic minority groups; (ii) the paucity of longitudinal studies concerned with obesity among ethnic minority groups; (iii) the lack of multiple metrics of obesity in most studies; (iv) the overreliance on aggregated ethnic groups, such as ‘South Asians’ or ‘Blacks’; (v) the lack of studies that considered obesity risk among groups other than South Asians, Blacks and Chinese and (vi) the paucity of studies that have analysed national or UK-wide datasets. We will discuss each of these limitations in turn, commenting on how each limitation might influence our interpretations of the available literature.

First, the central limitation of the literature about ethnic disparities in obesity in the UK is that there was only one study that directly considered differences in the determinants of obesity by ethnic group. That study focused on differences in parental anthropometry and socioeconomic status as determinants of metrics of obesity in childhood (40).

Explanations for disparities in health outcomes between ethnic minority and majority groups have centered on differences in socioeconomic status, physiology, cultural behaviour, discrimination and acculturative stress between majority and minority groups (17). With regard to the socioeconomic explanation, data suggests that ethnic minorities in the UK have lower socioeconomic status than Caucasians (59); because low socioeconomic status is associated with metrics of poor health, differences in socioeconomic status between ethnic minorities and Caucasians may explain poor outcomes among minority populations (17). Differences in genetically based physiological tendencies, although controversial (17,60), have also been suggested to explain ethnic differences in population health metrics. Other findings have suggested that ethnic differences in health may rest on culturally mediated differences in high-risk health behaviours. For example, religious restrictions and food and alcohol intake among ethnic groups have been shown to influence hypertension and obesity risk (19). Moreover, differences in degree of social networking between cultures may have a complex influence on health metrics (61). Yet other explanations have suggested that differences in cumulative discrimination-associated stress between ethnic minorities and Caucasians may explain ethnic differences in health (62–64). Smith and colleagues suggest that discrimination can influence the health of ethnic minorities in three ways. First, it can directly harm their wellbeing. Second, it can systematically worsen their socioeconomic status, and therefore, their health. And third, it can exacerbate their recognition of relative disadvantage, and thereby negatively impact their health (65). Finally, acculturative stress is the mental or emotional strain of the process of change that individuals or populations undergo when they come in contact with a different culture (66,67), and has been shown to influence ethnic minority health metrics(67–70) and may therefore explain ethnic differences in health.

Each of these factors has been shown to be associated with increased obesity risk among ethnic minority populations in other contexts (71–75). The only potential mediator of the relation between ethnic status and obesity outlined above that was systematically studied was socioeconomic status (40), in one study among children only. The paucity of literature concerned with mediators of the relation between ethnicity and obesity risk has two effects: First, it limits our understanding of the contributors to disparities in obesity risk in the UK as well as our understanding of the aetiology of obesity among ethnic minority groups. And second, it limits our ability to conceptualize interventions that could potentially address obesity among ethnic minority groups.

The second methodological limitation to the existing literature about ethnic disparities in obesity in the UK is the lack of longitudinal studies that have considered ethnicity and obesity in the UK. Only two studies reviewed above reported on findings from longitudinal analyses – and both studies were among children. This paucity of longitudinal studies limits our understanding of ethnic inequalities in obesity risk in the UK for several reasons. First, without longitudinal studies, estimating population-based incidence of obesity among ethnic minority groups is problematic. Second, longitudinal studies would allow for an understanding of the causal and temporal mechanisms upon which associations between ethnicity and obesity risk are based. Third, with regard to the first methodological limitation, longitudinal studies would allow investigators to study the roles of socioeconomic status, physiology, cultural behaviour, discrimination and acculturative stress as predictors of obesity risk among ethnic minority groups in the UK.

Third, few studies about ethnic disparities in the UK have used multiple metrics of obesity so as to address the potential for anthropometric differences by ethnic group. As discussed above, we found substantial anthropomorphic differences by gender and ethnicity among ethnic minority groups in the UK. Because of these systematic differences, in studies of obesity among ethnic minority groups, multiple metrics of obesity should be used. Metrics including a weight-for-height based metric (such as BMI), an adiposity metric (such as skin-folds and/or DXA scans) and a metric of central adiposity (waist circumference and/or waist/hip ratio) should be included in future studies of obesity among ethnic minority groups.

As a fourth methodological limitation, relatively few studies have stratified between ethnic/national groups within broader, aggregate ethnic categories. Studies about health metrics among ethnic minority groups can often be limited by ethnic prevalence in sampling frames, therefore investigators often aggregate populations from geographically similar groups so as to increase the prevalence among an exposure group (76). For example, an investigator might aggregate data about Bangladeshis, Indians, and Pakistanis as ‘South Asians’. Our findings suggest that there may be important differences by ethnic/national group within broader, traditional ethnic classifications. For example, we found significant differences in obesity risk between Indians, Pakistanis and Bangladeshis; between Muslims and Hindus; and between West Africans and Caribbeans (29,48,51,52). This suggests that aggregating may not be appropriate in studies about obesity among ethnic minority groups in the UK.

The fact that many of the studies reviewed above did not disaggregate their findings further among ‘Black’ and/or ‘South Asian’ ethnic aggregates poses a substantial limitation to the available literature about ethnic disparities in obesity risk in the UK for several reasons. First, if groups within ethnic aggregates have substantially different risk for obesity relative to Caucasians, such as Pakistanis and Bangladeshis (30), conceptually, it may not be appropriate to group them together. And estimates of obesity prevalence among the aggregate group relative to Caucasians may not reflect organic differences in obesity risk between those groups, but the proportion of one ethnic group within the aggregate to others. Second, the distribution of potentially mediating factors between ethnicity and obesity, such as socioeconomic status, may differ by ethnic group within ethnic aggregates, such as has been demonstrated between Indians and other South Asian groups (23). Therefore, assessing health metrics such as obesity among ethnic aggregates relative to Caucasians may be complicated by confounding of potentially mediating variables within those ethnic aggregates. Third, the social history and demographic realities of different groups within ethnic aggregates may differ substantially, and therefore, aggregating ethnic groups may distort our understanding of how demographic factors and social histories influence health risk among ethnic groups in the UK.

The fifth methodological limitation is that aside from South Asians, Blacks and Chinese, the literature is scant about obesity risk among other ethnic minority groups relative to Caucasians. Of the 29 studies reviewed here, only three considered the health of ‘other’ ethnic groups, which in all three studies, were poorly defined and aggregated without geographic similarities between groups comprising them. Therefore, the present literature is weak with regard to obesity risk among Arab, East Asian, or Eastern European groups in the UK relative to British Caucasians.

The sixth methodological limitation to the current literature about ethnicity and obesity in the UK is that there are relatively few studies that have considered ethnic inequalities in obesity using nationally or UK-representative data. The majority of the studies (18 of 29) reviewed here reported on datasets from sub-national localities throughout the UK. Only two studies reported on data that was UK-representative, and both of these studies were among children. Studies regarding obesity in the UK have noted important regional differences in risk for obesity overall by region in the UK (30,77). However, little is known about regional differences in obesity inequalities by ethnicity. Given regional differences in overall obesity risk, and the plausibility that the characteristics of ethnic minority populations may differ substantially by geographic context, there may be important differences in obesity inequalities by region. The lack of UK-representative studies sufficiently powered to ascertain regional differences in ethnic inequalities in obesity therefore presents a substantial limitation to the available literature about ethnic inequalities in obesity in the UK.

Limitations

There are several limitations that should be considered when interpreting the findings of this review. First, we organized our findings by age group and then by ethnicity within each group. It is possible that our organization scheme may have partially shaped the inferences drawn from the existing literature. Second, because our inclusion criteria were restricted to studies published in the peer-reviewed literature, it is possible that there may have been a publication bias with regard to the articles discussed above, although our inclusion criteria were thorough, and included data about many of the largest health surveys in the UK. Therefore, our findings may not accurately reflect current knowledge about ethnicity and obesity in the UK. Third, our findings were limited to studies about ethnic minority populations living in the UK, and therefore, it may not be appropriate to generalize findings here when considering similar ethnic populations in other contexts or globally.

Directions for future research

We suggest that six directions for future research emerge from this review. First, investigators interested in ethnic disparities in obesity risk in the UK might undertake studies to specifically assess anthropometric differences by ethnicity and sex among children and adults relative to Caucasians in the UK. Second, longitudinal studies of obesity risk among ethnic minority groups should be a priority among investigators interested in this area. Third, studies about the roles of socioeconomic status, physiology, cultural behaviours, discrimination and acculturative stress as determinants of metrics of obesity are needed. Fourth, research about ethnic minority health should disaggregate traditional ethnic classifications, such as ‘South Asian’ and ‘Black’ in future studies so as to understand heterogeneity among particulate ethnic minorities within these larger taxonomies. Fifth, research about obesity risk among ethnic minority groups other than South Asians, Blacks, and Chinese is needed, as we found no studies in the available literature about obesity risk specifically among Arabs, East Asians, and/or Eastern Europeans relative to British Caucasians in the UK. Sixth, future research about obesity among ethnic minority groups should include multiple metrics of obesity, one each of the following: a weight-for-height based metric (such as BMI), an adiposity metric (such as skin-folds and/or DXA scans) and a metric of central adiposity (waist circumference and/or waist/hip ratio).

Conflict of Interest Statement

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

The authors declare no conflicts of interest.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References

This study was funded in part by the British Heart Foundation and the Rhodes Trust.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of Interest Statement
  8. Acknowledgements
  9. References
  • 1
    Seidell JC. Obesity: a growing problem. Acta Paediatr Suppl 1999; 88: 4650.
  • 2
    Livingstone MB. Childhood obesity in Europe: a growing concern. Public Health Nutr 2001; 4: 109116.
  • 3
    McPherson K, Marsh T, Brown M. Tackling obesities: future choices-modelling future trends in obesity and the impact on health. Foresight 2007; 2: 171.
  • 4
    Wilson PW, D’Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as determinants of cardiovascular risk: the Framingham experience. Arch Intern Med 2002; 162: 18671872.
  • 5
    Wellman NS, Friedberg B. Causes and consequences of adult obesity: health, social and economic impacts in the United States. Asia Pac J Clin Nutr 2002; 11: 705709.
  • 6
    Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int J Obes 1999; 23: 211.
  • 7
    Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, Bales VS, Marks JS. Prevalence of obesity, diabetes, and obesity-related. JAMA 2003; 289: 7679.
  • 8
    Veer P, Kampman E. Food, nutrition, physical activity and the prevention of cancer: a global perspective. Washington, D.C.: World Cancer Research Fund/American Institute for Cancer Research; 2007. [WWW document]. URL http://www.dietandcancerreport.org/ (accessed May 2010).
  • 9
    Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. N Engl J Med 2003; 348: 16251638.
  • 10
    Møller H, Mellemgaard A, Lindvig K, Olsen JH. Obesity and cancer risk: a Danish record-linkage study. Eur J Cancer 1994; 30: 344350.
  • 11
    Suk SH, Sacco RL, Boden-Albala B, Cheun JF, Pittman JG, Elkind MS, Paik MC. Northern Manhattan stroke study. abdominal obesity and risk of ischemic stroke: the northern Manhattan stroke study. Stroke 2003; 34: 15861592.
  • 12
    Coggon D, Reading I, Croft P, McLaren M, Barrett D, Cooper C. Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord 2001; 25: 622627.
  • 13
    Onyike CU, Crum RM, Lee HB, Lyketsos CG, Eaton WW. Is obesity associated with major depression? Results from the third national health and nutrition examination survey. Am J Epidemiol 2003; 158: 11391147.
  • 14
    Bender R, Jöckel KH, Trautner C, Spraul M, Berger M. Effect of age on excess mortality in obesity. JAMA 1999; 281: 14981504.
  • 15
    Gunnell DJ, Frankel SJ, Nanchahal K, Peters TJ, Davey Smith G. Childhood obesity and adult cardiovascular mortality: a 57-y follow-up study based on the Boyd Orr cohort. Am J Clin Nutr 1998; 67: 11111118.
  • 16
    Engels F, Kelley F. The Condition of the Working-Class in England in 1844. Penguin: Harmondsworth, 1845.
  • 17
    Smith GD, Chaturvedi N, Harding S, Nazroo J, Williams R. Ethnic inequalities in health: a review of UK epidemiological evidence. Crit Public Health 2000; 10: 375408.
  • 18
    Duran-Tauleria E, Rona RJ, Chinn S, Burney P. Influence of ethnic group on asthma treatment in children in 1990–1991: national cross sectional study. BMJ 1996; 313: 148152.
  • 19
    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: 382386.
  • 20
    McKeigue PM, Ferrie JE, Pierpoint T, Marmot MG. Association of early-onset coronary heart disease in south asian men with glucose intolerance and hyperinsulinemia. Circulation 1993; 87: 152161.
  • 21
    Chaturvedi N, McKeigue PM, Marmot MG. Resting and ambulatory blood pressure differences in Afro-Caribbeans and Europeans. Hypertension 1993; 22: 9096.
  • 22
    Harland JO, Unwin N, Bhopal RS, White M, Watson B, Laker M, Alberti KG. Low levels of cardiovascular risk factors and coronary heart disease in a UK Chinese population. J Epidemiol Community Health 1997; 51: 636642.
  • 23
    Fitzpatrick J, Jacobson B, Aspinall P. Ethnicity and Health: Executive Summary. Association of Public Health Observatories: York, 2007.
  • 24
    Harding S, Maxwell R. Differences in mortality of migrants. In: DreverF, WhiteheadM (eds). Health Inequalities. The Stationary Office: London, 1997.
  • 25
    Department of Health, Healthcare Commission. Report on the self reported experience of patients from black and minority ethnic groups. London: Office of National Statistics; 2009. [WWW document]. URL http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsStatistics/DH_100467/ (accessed 19 May 2010).
  • 26
    Office for National Statistics. 7.9% from a minority ethnic group. [WWW document]. URL http://www.statistics.gov.uk/CCI/nugget.asp?ID=273 (accessed May 2010).
  • 27
    Bhattacharyya G, Ison L, Blair M. Minority Ethnic Attainment and Participation in Education and Training: The Evidence. Department for Education and Skills: Nottingham, 2003, RTP01-03.
  • 28
    Ipsos MORI. Trend Briefing #1: doubting Multiculturalism. London: Ipsos MORI; 2009. [WWW document]. URL http://centres.exeter.ac.uk/emrc/publications/Islamophobia_and_Anti-Muslim_Hate_Crime.pdf (accessed 19 May 2010).
  • 29
    Githens-Mazer J, Lambert R. Islamophobia and anti-muslim hate crime: a London case study. Exeter: European Muslim Research Center; University of Exeter; 2010. [WWW document]. URL http://muslimsafetyforum.org/docs/Islamophobia_and_Anti-Muslim_Hate_Crime.pdf (accessed 19 May 2010).
  • 30
    Rennie KL, Jebb SA. Prevalence of obesity in Great Britain. Obes Rev 2005; 6: 1112.
  • 31
    Wardle J, Brodersen NH, Cole TJ, Jarvis MJ, Boniface DR. Development of adiposity in adolescence: five year longitudinal study of an ethnically and socioeconomically diverse sample of young people in Britain. BMJ 2006; 332: 11301135.
  • 32
    Saxena S, Ambler G, Cole TJ, Majeed A. Ethnic group differences in overweight and obese children and young people in England: cross sectional survey. Arch Dis Child 2004; 89: 3036.
  • 33
    Balakrishnan R, Webster P, Sinclair D. Trends in overweight and obesity among 5–7-year-old white and south asian children born between 1991 and 1999. J Public Health 2008; 30: 139144.
  • 34
    Jebb SA, Rennie KL, Cole TJ. Prevalence of overweight and obesity among young people in Great Britain. Public Health Nutr 2004; 7: 461465.
  • 35
    Hawkins SS, Cole TJ, Law C, Millennium Cohort Study Child Health Group. An ecological systems approach to examining risk factors for early childhood overweight: findings from the UK millennium cohort study. J Epidemiol Community Health 2009; 63: 147155.
  • 36
    Harding S, Teyhan A, Maynard MJ, Cruickshank JK. Ethnic differences in overweight and obesity in early adolescence in the MRC DASH study: the role of adolescent and parental lifestyle. Int J Epidemiol 2008; 37: 162172.
  • 37
    Rutter H. Data briefing. Child obesity worst in deprived areas. Health Serv J 2008; 21: 21.
  • 38
    Whincup PH, Gilg JA, Papacosta O, Seymour C, Miller GJ, Alberti KG, Cook DG. Early evidence of ethnic differences in cardiovascular risk: cross sectional comparison of British South Asian and white children. BMJ 2002; 324: 635.
  • 39
    Duran-Tauleria E, Rona RJ, Chinn S. Factors associated with weight for height and skinfold thickness in British children. J Epidemiol Community Health 1995; 49: 466473.
  • 40
    Rona RJ, Chinn S. National study of health and growth: Social and biological factors associated with weight-for-height and triceps skinfold of children from ethnic groups in England. Ann Hum Biol 1987; 14: 231248.
  • 41
    Chinn S, Hughes JM, Rona RJ. Trends in growth and obesity in ethnic groups in Britain. Arch Dis Child 1998; 78: 513517.
  • 42
    Shaw NJ, Crabtree NJ, Kibirige MS, Fordham JN. Ethnic and gender differences in body fat in British schoolchildren as measured by DXA. Arch Dis Child 2007; 92: 872875.
  • 43
    Harding S, Maynard MJ, Cruickshank K, Teyhan A. Overweight, obesity and high blood pressure in an ethnically diverse sample of adolescents in Britain: the medical research council DASH study. Int J Obes (Lond) 2008; 32: 8280.
  • 44
    Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide; international survey. BMJ 2000; 320: 12401243.
  • 45
    Taylor SJ, Viner R, Booy R, Head J, Tate H, Brentnall SL, Haines M, Bhui K, Hillier S, Stansfield S. Ethnicity, socio-economic status, overweight and underweight in East London adolescents. Ethn Health 2005; 10: 113128.
  • 46
    McKeigue PM, Pierpoint T, Ferrie JE, Marmot MG. Relationship of glucose intolerance and hyperinsulinaemia to body fat pattern in South Asians and Europeans. Diabetologia 1992; 35: 785791.
  • 47
    Bose K. Generalised obesity and regional adiposity in adult white and migrant Muslim males from Pakistan in Peterborough. J R Soc Health 1996; 116: 161167.
  • 48
    Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti KG, Harland J, Patel S, Ahmad N, Turner C, Watson B, Kaur D, Kulkarni A, Laker M, Tavridou A. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. BMJ 1999; 319: 215220.
  • 49
    Vyas A, Greenhalgh A, Cade J, Sanghera B, Riste L, Sharma S, Cruickshank K. Nutrient intakes of an adult Pakistani, European and African-Caribbean community in inner city Britain. J Hum Nutr Diet 2003; 16: 327337.
  • 50
    Miller MA, Cappuccio FP. Cellular adhesion molecules and their relationship with measures of obesity and metabolic syndrome in a multiethnic population. Int J Obes 2006; 30: 11761182.
  • 51
    Cappuccio FP, Cook DG, Atkinson RW, Strazzullo P. Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in South London. Heart 1997; 78: 555563.
  • 52
    Diaz VA, Mainous AG III, Baker R, Carnemolla M, Majeed A. How does ethnicity affect the association between obesity and diabetes? Diabet Med 2007; 24: 11991204.
  • 53
    Wardle J, Waller J, Jarvis MJ. Sex differences in the association of socioeconomic status with obesity. Am J Public Health 2002; 92: 12991304.
  • 54
    Knight T, Smith Z, Lockton JA, Sahota P, Bedford A, Toop M, Kernohan E, Baker MR. Ethnic differences in risk markers for heart disease in bradford and implications for preventive strategies. J Epidemiol Community Health 1993; 47: 8995.
  • 55
    Unwin N, Harland J, White M, Bhopal R, Winocour P, Stephenson P, Watson W, Turner C, Alberti KG. Body mass index, waist circumference, waist-hip ratio, and glucose intolerance in Chinese and Europid adults in Newcastle, UK. J Epidemiol Community Health 1997; 51: 160166.
  • 56
    Bose K. A comparative study of generalised obesity and anatomical distribution of subcutaneous fat in adult white and Pakistani migrant males in Peterborough. J R Soc Health 1995; 115: 9095.
  • 57
    WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004; 363: 157163.
  • 58
    Daniels SR, Khoury PR, Morrison JA. The utility of body mass index as a measure of body fatness in children and adolescents: differences by race and gender. Pediatrics 1997; 99: 804807.
  • 59
    Nazroo JY. The health of Britain's ethnic minorities: findings from a national survey. London: Policy Studies Institute; 1997. [WWW document]. URL http://hdl.handle.net/10068/424798 (accessed 10 May, 2010).
  • 60
    Herman AA. Toward a conceptualization of race in epidemiologic research. Ethn Dis 1996; 6: 720.
  • 61
    Valle R, Yamada AM, Barrio C. Ethnic differences in social network help-seeking strategies among Latino and Euro-American dementia caregivers. Aging Ment Health 2004; 8: 535543.
  • 62
    Geronimus AT. The weathering hypothesis and the health of African-American women and infants: evidence and speculations. Ethn Dis 1992; 2: 207221.
  • 63
    Geronimus AT. Black/white differences in the relationship of maternal age to birthweight: a population-based test of the weathering hypothesis. Soc Sci Med 1996; 42: 589597.
  • 64
    Brown TN, Williams DR, Jackson JS, Neighbors HW, Torres M, Sellers SL, Brown KT. ‘Being black and feeling blue’: the mental health consequences of racial discrimination. Race Soc 2000; 2: 117131.
  • 65
    Wilkinson RG. Unhealthy Societies: The Afflictions of Inequality. Routledge: New York, 1996.
  • 66
    Berry JW, Kim U, Minde T, Mok D. Comparative studies of acculturative stress. Int Migr Rev 1987; 21: 491511.
  • 67
    Hovey JD, King CA. Acculturative stress, depression, and suicidal ideation among immigrant and second-generation Latino adolescents. J Am Acad Child Adolesc Psychiatry 1996; 35: 11831192.
  • 68
    Wrobel NH, Farrag MF, Hymes RW. Acculturative stress and depression in an elderly Arabic sample. J Cross-Gult Gerontol 2009; 24: 273290.
  • 69
    Lee J, Koeske G, Sales E. Social support buffering of acculturative stress: a study of mental health symptoms among Korean international students. Int J Intercult Relat 2004; 28: 399414.
  • 70
    Finch BK, Vega WA. Acculturation stress, social support, and self-rated health among Latinos in California. J Immigr Health 2003; 5: 109117.
  • 71
    Paeratakul S, Lovejoy JC, Ryan DH, Bray GA. The relation of gender, race and socioeconomic status to obesity and obesity comorbidities in a sample of US adults. Int J Obes Relat Metab Disord 2002; 26: 12051210.
  • 72
    Gee GC, Ro A, Gavin A, Takeuchi DT. Disentangling the effects of racial and weight discrimination on body mass index and obesity among Asian Americans. Am J Public Health 2008; 98: 493500.
  • 73
    O'Dea JA. Gender, ethnicity, culture and social class influences on childhood obesity among Australian schoolchildren: implications for treatment, prevention and community education. Health Soc Care Community 2008; 16: 282290.
  • 74
    Cossrow N, Faulkner B. Race/ethnic issues in obesity and obesity-related comorbidities. J Clin Endocrinol Metab 2004; 89: 25902594.
  • 75
    Sundquist J, Winkleby M. Country of birth, acculturation status and abdominal obesity in a national sample of Mexican-American women and men. Int J Epidemiol 2000; 29: 470477.
  • 76
    Chaturvedi N, McKeigue PM. Methods for epidemiological surveys of ethnic minority groups. J Epidemiol Community Health 1994; 48: 107111.
  • 77
    Scarborough P, Allender S. The North-South gap in overweight and obesity in England. Br J Nutr 2008; 100: 677684.
  • 78
    Moon G, Quarendon G, Barnard S, Twigg L, Blyth B. Fat nation: deciphering the distinctive geographies of obesity in England. Soc Sci Med 2007; 65: 2031.