Approved Research

Causal effects of maternal vegetarian diets during pregnancy on offspring health

Lay summary

Vegetarian diets are defined as diets without the intake of meat, fish, or seafood. Vegetarianism is socially and culturally patterned, and nutrient intakes also differ between vegetarians and non-vegetarians. While previous studies have found that vegetarian diets are associated with a range of health outcomes in the general population, evidence on the relationship between maternal vegetarian diets during pregnancy and offspring health remains scarce. This three-year project is part of a broader project that aims to investigate the influences of maternal vegetarian diets during pregnancy on a range of offspring health outcomes from birth to childhood. Using the UK Biobank data, we aim to identify the genetic determinants of vegetarianism and use them as a "proxy" of maternal vegetarianism to better examine its causal effects on offspring health. Given the increasing popularity of vegetarianism and the importance of optimal nutrition during pregnancy for maternal and child health, our findings will help comprehensively evaluate the pros and cons of the increasingly popular vegetarianism and have implications for dietary counselling and guidelines for pregnant women.

Scope extension: Research question: Does following a vegetarian diet during pregnancy causally affect the health of offspring?

The overarching aim is to examine the causal effect of maternal vegetarian diets during pregnancy on offspring health using Mendelian randomisation (MR).

Specifically, by using the UK Biobank data, we aim to:

1) conduct a genome-wide association study (GWAS) for vegetarianism;

2) select genetic instruments for vegetarianism based on the GWAS full summary statistics

3) use the genetic instruments to proxy maternal vegetarianism and perform MR analysis (in our own birth cohort samples) to examine its causal effects on offspring health outcomes from birth to childhood.

In addition to the above, we hope to add a continuous variable for our exposure and extend the scope of our outcomes of interest: 1) using dietary data from food frequency questionnaires (FFQs), we will develop a continuous plant-based diet index (PDI) based on pre-defined scoring criteria (Satija et al., PLoS Med, 2016) to reflect the adherence to plant-based/vegetarian diets and include it as a phenotype for our GWAS; 2) in addition to offspring outcomes, we are also interested in other maternal (e.g., hypertensive disorders of pregnancy, gestational diabetes, maternal anaemia, perinatal depression, etc.) and pregnancy (miscarriage, stillbirth, prelabour rupture of membranes, induction of labour, Caesarean section, etc.) outcomes that are highly relevant to them.