In this large population-based survey of the multi-ethnic Pan-Malaysian population, the overall prevalence of the metabolic syndrome was 27.5%, with higher prevalence in women (30.1%) compared to men (24.8%). Compared to ethnic Malays, the prevalence of metabolic syndrome was 16% lower among Chinese, 12% higher among Indigenous Sarawakians, and 31% higher among Indians. In addition, our data suggest that the association between ethnicity and metabolic syndrome was modified by age. Among younger participants, the distribution of metabolic syndrome was more tightly clustered among Malays, Chinese, and Indians, while Indigenous Sarawakians had a higher prevalence than the other ethnic groups. This difference among younger age groups may result, in the coming years, in a new ethnic distribution in the prevalence of metabolic syndrome in Malaysia.
The overall prevalence of metabolic syndrome observed in our study was lower than that reported in India and in Western Countries, similar to other reports from neighboring South East Asia countries, and is higher than reported rates in China. Using the same harmonized IDF definition of metabolic syndrome, the prevalence of metabolic syndrome was 31.4% in Kolkata, India 
, 23.3% in Thailand 
and 7.3% in a large study from Guangdong, China 
. Using the National Cholesterol Education Program, Adult Treatment Panel III (NCEP ATP III) diagnostic criteria, Bhardwaj et al identified a 45.3% prevalence of metabolic syndrome among 459 subjects in New Delhi, India 
, whilst a Chinese study of 15,540 subjects identified a 7.8% and 17.8% prevalence among men and women, respectively 
In our study, Indians had the highest overall prevalence of metabolic syndrome. It appears that Asians from the Indian sub-continent (i.e. South Asia) have a markedly higher prevalence of metabolic syndrome compared to the Chinese (i.e. East Asia). A higher prevalence of the metabolic syndrome among ethnic Indians residing in Western countries 
and in smaller communities in Asia 
have been reported before. Ethnic Indians may be predisposed to several features of the metabolic syndrome. Compared to age-matched Caucasians of a similar body mass index and waist circumference, Indian adults, particularly females, have a higher percentage body fat, abdominal/central adiposity and subcutaneous adiposity 
, all of which are known determinants of insulin resistance. Furthermore, type 2 diabetes and non-alcoholic fatty liver disease (NAFLD), also related to the metabolic syndrome, occurs more frequently in Indians compared to other ethnic groups in Asia or in Western Countries 
. Lastly, cultural and lifestyle practices of ethnic Indians may further contribute to their higher prevalence of the metabolic syndrome. Compared to other ethnic groups, Indians have been shown to be less physically active on a daily basis 
, and the Indian diet is traditionally high in carbohydrate and fat, with a lower fiber component, all of which may contribute towards insulin resistance and the metabolic syndrome 
The prevalence of metabolic syndrome among younger age groups in our study was higher among Indigenous Sarawakians compared to the other ethnic groups. For individual metabolic syndrome criteria, young Indigenous Sarawakians had higher prevalence of central obesity, raised blood pressure, raised triglycerides, and lower HDL-C compared to young participants of other ethnic groups. There is only limited information on the prevalence of cardiovascular risk factors in the state of Sarawak 
. The Malaysia National Health Morbidity Survey III (NHMS III), conducted among subjects aged ≥18 years, reported that Indigenous Sarawakians had a higher prevalence of hypertension compared to other ethnic groups, but comparisons across ethnicities were not available for other metabolic syndrome components. Further research is needed to better explain the variation in the prevalence of metabolic syndrome among Indigenous Sarawakians as well as the evolution of metabolic syndromes at young ages.
Among older age groups, the prevalence of metabolic syndrome was higher in Malays compared to Chinese and Indigenous Sarawakians, but lower than in Indians. The prevalence of hypertension, among those aged >30 years in NHMS III was higher among Indians and Malays compared to the Chinese and Indigenous Sarawakian 
. In the same survey, the prevalence of obesity was highest in Indians, followed by Malays and Chinese.
Apart from ethnic differences, our study also demonstrated age and gender differences in the prevalence of the metabolic syndrome, consistent with other Asian community-based studies 
. The higher prevalence of central obesity and low HDL-C, that explains some of the gender variation of metabolic syndrome, may be due to the higher prevalence of obesity and lower levels of physical activity among females. A higher prevalence of the metabolic syndrome among females compared to males is well recognized 
, partly due to an increase in the prevalence of the components of the metabolic syndrome at the peri-menopause and early menopause stage in females 
. The cardiovascular risk profile of women can worsen at this time, possibly due to weight gain and changes in lipid profile. Regardless of gender, all components of the metabolic syndrome increase with age, likely due to time-related life style factors and diminishing physical activity with age 
Some limitations, of our cross sectional survey of the population of Malaysia in 2004, need to be considered. First, our cross-sectional design does not allow establishing causal statements. Second, it was difficult to attribute ethnic differences to more specific pathways. The association between ethnicity and disease observed in this study is a likely combination of genetic and environmental factors and their interactions along with temporal changes of society. Third, we excluded an ethnic sample from the state of Sabah (see Materials and Methods
) and had a higher proportion of urban dwellers compared to the general population of Malaysia. However, there was appropriate representation of the major ethnic groups including Malays, Indigenous Sarawakians, Indians and Chinese, data on whom may be of relevance to other regions in Asia. The rigorous methodology, large sample size and representative educational levels of the study sample with respect to the general population are major strengths of our study.
We conclude that the prevalence of the metabolic syndrome in Malaysia, a multi-racial South East Asian population was 27.5%, a rate which is considerably higher than that of adults in East Asia but lower than that from South Asia (Indian sub-continent). Whilst rapid socio-economic development in Malaysia may be partly responsible for this pattern, we observed ethnic differences between ethnic groups, with ethnic Indians having a greater prevalence of metabolic syndrome compared to other ethnic groups. In addition, we found that the association between ethnicity and metabolic syndrome was modified by age. The distribution of metabolic syndrome among the different ethnic groups was more similar in the younger age groups compared to the older age groups, although young Indigenous Sarawakians had much higher prevalence of metabolic syndrome compared to other ethnic groups. It is thus critical to assess age-specific trends in metabolic syndrome among young and middle age groups in Malaysia, especially among indigenous populations. Further studies are also warranted to elucidate specific ethno-cultural related factors that may be intervened on in future prevention programs to reduce the burden of metabolic risk factors.