Study of the relationship between ethnic ancestry and birth weight in babies born to South Asian and White parents in England and Wales

Georgina Ronalds and Dave Leon, London School of Hygiene & Tropical Medicine

[Project number 30012]

Babies born in the UK to women of Indian ethnic origin are lighter on average than the offspring of women of European origin. Studies disagree on whether this ethnic gap has lessened in second-generation births. However, the largest studies have found no improvement in birth weight among Asians born in the UK, or even that the gap between Asian and white European babies has widened. Both genetic and phenotypic adaptation to undernutrition in utero, reflected in low birth weight, have been proposed to explain - in the context of rapidly changing diet and lifestyle - the high rates of cardiovascular disease, type II diabetes and obesity that are observed in Indian populations. I know of no studies to date which have investigated birth weight in babies with mixed Indian-European ethnic descent. I hope to compare the birth weights of babies born to inter-ethnic unions and same-ethnicity unions, a method which provides a previously unexplored opportunity to study the roles of genetic adaptation, phenotypic adaptation as a result of maternal constraint of growth in utero, and other environmental factors in determining inter-ethnic differences in foetal growth.

The LS provides an opportunity to employ an existing data set which contains information on birth weight, ethnic group and a range of potential confounders that is sufficient for the needs of the study. In particular, the capacity of the LS to provide inter-generational data is important. Firstly, it helps tackle the issue of reported ethnic group not necessarily indicating ethnic ancestry. By allowing a mother to be linked to her child and - through information on household membership - to the likely father of the child also, the LS can provide a classification of ethnicity based on parental ethnic group. This classification is likely to have improved validity in the context of the particular research question, which addresses genetic and environmental contributions to variation in birth weight. Secondly, the potential to identify the country of birth of the child's parents would allow children in the second generation of a migrant family to be distinguished from children in later generations. This is an important factor in the light of previous studies into changes in birth weight over successive generations in migrant populations. The aim of the study is to compare birth weight of babies in four groups defined by parental ethnicity as reported in the 1991 census: where both parents are South Asian (Indian, Bangladeshi or Pakistani); where both parents are white; where the mother is South Asian and father is white; where the mother is white and father is South Asian. The comparisons would also be made using the narrower criterion of Indian ethnic group instead of South Asian. As such births are relatively rare, the study would benefit from the large size of the LS.

Top of page