Pregnancy as a driver of household food purchasing choices

Title

Pregnancy as a driver of household food purchasing choices

Purpose

The aim of this work is to help inform public health efforts to improve the food environment before and during pregnancy, potentially sustainably so. We plan to:

1. Describe the nutritional composition of UK households’ food and drink grocery purchases before, during, and after pregnancy.
2. Estimate whether pregnancy is a likely causal factor in driving households to change nutritional composition of their food basket, and assess to what extent these effects are sustained post birth.
3. Identify socio-economic inequalities in observed effects.

Assess whether the most plausible explanation for observed changes in purchasing is changes in consumption.

Background

Early life, starting from conception, have been established as critical periods in a child’s development. Understanding the food environments UK infants are born into and grow up in is therefore key. In parallel, pregnancy is a significant life event flanked by intense public health measures to promote dietary changes towards healthy eating while trying to conceive, during pregnancy and after child birth. Evidence suggests that the nutritional composition and healthiness of the home food environment is amenable during this time, but this could be driven by other factors, such as pregnancy coming with changes in living arrangements. Causal evidence is therefore important to inform policy that aims to change food choices pre-conception and during pregnancy.

Aims and methodology

We will start by using the Kantar Worldpanel data (2015-2020). We will start by providing a descriptive analysis of the nutritional composition of households’ shopping baskets during pregnancy and after childbirth and highlight characteristics of households that show particularly low scores in a Healthy Eating Index and/or Nutrient Profile Score.

To understand the likely causality of pregnancy on households’ nutritional composition will follow Fadlon and Nielsen (2019)’s strategy to construct a counterfactual for pregnant households using households that go through the same experience but some years into the future, also referred to as a ‘dynamic difference-in-difference’ approach. In addition to constructing this control group, we will account for year, month and household fixed effects.

Timing

October 2021 – April 2023