Evaluating the changes in household purchases of foods and drinks high in fat, salt and sugar following Bristol's outdoor advertisement restrictions policy: a quasi-experimental study.

INTRODUCTION

In 2021, Bristol (UK) became the first city outside London to introduce a policy restricting advertisements of high fat, salt and sugar (HFSS) products and other unhealthy commodities (alcohol, gambling and payday loans) on council-owned advertisement sites, predominantly bus stops. This research evaluates the impact of this policy on household purchasing of HFSS products specifically.

METHODS

Take-home purchasing data for food and drink items recorded from October 2020 to May 2023 (74 weeks before and 59 weeks after policy implementation) by 1012 households from Kantar's Worldpanel division were included in the analysis: 217 households from Bristol (intervention) and 795 from Cardiff, Sheffield, Gloucestershire and South Gloucestershire (controls). Controlled interrupted time-series analyses adjusted for confounders were used to estimate the change in average weekly household purchases of energy and nutrients from HFSS products in the post-intervention period (from 1 April 2022) compared with the counterfactual (data from controls and pre-intervention in Bristol).

RESULTS

There was no indication of changes in purchasing of HFSS products in Bristol following the policy; mean change in weekly household purchases of energy and nutrients from take-home HFSS products was 897.6 kcal (95% CI -57.7 to 1.853) or 6.1% (95% CI -0.7 to 12.9) for energy, 4.6% (95% CI -2.6 to 11.9) for fat, 6.1% (95% CI -2.9 to 15.0) for sugar and -1.9% (95% CI -17.4 to 13.6) for salt, respectively, compared with the counterfactual. Stratified analyses by household socio-demographics also showed no evidence of changes in energy purchased from HFSS products in Bristol.

CONCLUSIONS

This study did not observe evidence of changes in purchasing of HFSS products following Bristol's advertising restriction policy. Bristol City Council owns a relatively small part (~30%) of the city's outdoor advertisement estate, so changes in exposure may have been too small to lead to measurable effects. Extending similar policies to cover more outdoor spaces and advertising platforms is probably required to impact purchasing behaviour.