METHODS: Population-based surveys included 30,721 Malay, 10,865 Indian and 25,296 Chinese adults from The Malaysian Cohort, and 413,737 White adults from UK Biobank. Sex-specific linear regression models estimated associations of anthropometry and body composition (body mass index [BMI], waist circumference [WC], fat mass, appendicular lean mass) with systolic blood pressure (SBP), low-density lipoprotein cholesterol (LDL-C), triglycerides and HbA1c.
RESULTS: Compared to Malay and Indian participants, Chinese adults had lower BMI and fat mass while White participants were taller with more appendicular lean mass. For BMI and fat mass, positive associations with SBP and HbA1c were strongest among the Chinese and Malay and weaker in White participants. Associations with triglycerides were considerably weaker in those of Indian ethnicity (eg 0.09 [0.02] mmol/L per 5 kg/m2 BMI in men, vs 0.38 [0.02] in Chinese). For appendicular lean mass, there were weak associations among men; but stronger positive associations with SBP, triglycerides, and HbA1c, and inverse associations with LDL-C, among Malay and Indian women. Associations between WC and risk factors were generally strongest in Chinese and weakest in Indian ethnicities, although this pattern was reversed for HbA1c.
CONCLUSION: There were distinct patterns of adiposity and body composition and cardiovascular risk factors across ethnic groups. We need to better understand the mechanisms relating body composition with cardiovascular risk to attenuate the increasing global burden of obesity-related disease.
METHODOLOGY: A community-based cross-sectional study was conducted among 640 indigenous Temiar OA participants from a remote settlement in Gua Musang, Kelantan, Malaysia. Data was collected using face-to-face interviews with a standardised pretested questionnaire and through blood samples collected for haemoglobin (Hb) testing. Anaemia status was determined using the Hb level cut-off established by the World Health Organization (WHO). Descriptive analysis was used to determine the prevalence of anaemia, while multiple logistic regression was used to determine factors associated with anaemia.
RESULTS: The overall anaemia prevalence was 44.7% (286/640), and the prevalence rates of mild, moderate and severe anaemia were 42.7, 50.7 and 6.6%, respectively. Anaemia-specific prevalence varied significantly by age group (p 40 (aOR 0.25, p