Objectives To examine the influence of socio-economic position on type 2 diabetes risk markers in different ethnic groups and determine whether differences in socio-economic position can explain ethnic differences in type 2 diabetes risk.
Design Cross-sectional survey of children in 200 primary schools in London, Birmingham and Leicester (Child Heart and Health Study England, or CHASE) in which standardised anthropometric and fasting blood measurements were made. Ethnic origin was defined by parental self-report. Parent's socio-economic position (based on occupation) was measured using the National Statistics Socioeconomic Classification (NS-SEC). Statistical analyses were adjusted for age and sex and included a random effect for school.
Participants 4796 children (1153 white European, 1306 South Asian, 1215 black African/Caribbean) aged 9–10 years.
Main outcome measures Height, adiposity (ponderal index, skinfold thickness, fat mass index, waist circumference), glycated haemoglobin (HbA1c), glucose, insulin resistance, triglyceride, HDL-cholesterol, C reactive protein.
Results In the whole study population, NS-SEC showed weak and inconsistent associations with diabetes risk markers. However, there were marked differences between ethnic groups. Low socio-economic position was related to higher adiposity, insulin resistance and triglyceride levels in white Europeans and to a lesser extent South Asians; opposite patterns were observed in black African-Caribbeans (likelihood-ratio tests for interactions between NS-SEC and ethnicity, all p<0.05). There were marked ethnic differences in diabetes risk markers. Compared to white Europeans, South Asian children had higher fat mass index (% difference 7.3; 95% CI 2.8 to 12.0), sum of skinfolds (5.1; 1.1, 9.4), HbA1c (2.1; 1.6, 2.7), glucose (0.8; 0.2, 1.5), insulin resistance (29.6; 23.1, 36.4), triglycerides (12.9; 9.4, 16.5) and C reactive protein (43.3; 28.6, 59.7) and lower HDL-cholesterol (−2.9; −1.3, −4.5). In contrast, black African/Caribbean children had less marked increases in HbA1c, insulin resistance and C reactive protein but conversely, had lower triglycerides and higher HDL-cholesterol; adiposity levels were not consistently increased. However, adjustment for socio-economic position had no material effect on the ethnic differences in metabolic markers observed.
Conclusions Although socio-economic position showed little overall association with diabetes risk markers in this multi-ethnic study population, there were appreciable associations within individual ethnic groups. Ethnic differences in socio-economic position did not explain marked ethnic differences in emerging risks of type 2 diabetes between South Asians, black African-Caribbeans and white Europeans; other explanations for these ethnic differences should be sought.