# Impacts of alcohol health warning labels in a real-world setting: protocol for a randomised controlled trial among supermarket customers in Barcelona

## TL;DR

This study tests how alcohol warning labels affect people's knowledge and behavior in Barcelona through a randomized trial.

## Contribution

The novelty lies in evaluating real-world impacts of different health warning label types and positions on alcohol-related knowledge and behavior.

## Key findings

- Cancer message labels may increase knowledge of alcohol's health risks more than responsibility messages.
- Label placement (front vs. back of containers) could influence participants' responses and perceptions.
- Pre-measurement testing may affect how knowledge outcomes are assessed.

## Abstract

Alcohol is causally related to more than 200 diseases and injuries. Alcohol health warning labels are a promising intervention to address alcohol-related harm with multiple possible roles, but research on its real-world impacts is lacking. This study aims to experimentally evaluate the impact of exposure to two types of content (responsibility and cancer message) and positioning of the message (front or back) on knowledge of alcohol causing cancer as the primary outcome and alcohol consumption behaviour, intentions, risk perception, emotional response, product appeal and policy support as secondary outcomes. The study also aims to assess the potential testing effect of pre-measurement on the primary outcome.

Participants (of the legal drinking age in Spain (18 years or older), purchased at least one alcoholic beverage (with alcohol by volume (ABV) ≥ 1.2% for their own consumption and speaking Catalan or Spanish) will be recruited outside of supermarkets in Barcelona after purchasing alcohol, randomly assigned into one of the eight experimental groups, complete a baseline questionnaire (with half of the sample answering baseline questions measuring knowledge) and receive label stickers displaying either responsibility or cancer message, and applied to either front or back of every alcohol container they have purchased. They will complete follow-up surveys measuring the primary and secondary outcomes 1 week and 1 month after the intervention, either online or via telephone. The key hypotheses are that the label containing a cancer message will have a greater impact on the primary and secondary outcomes compared with the responsibility label. To evaluate the impact of health warning labels on knowledge of alcohol causing cancer, logistic regression will be employed to model the probability of a correct response as a function of the key independent variables, with results reported as ORs. Secondary outcomes will be modelled through linear regressions for continuous variables, and through logistic regressions for dichotomic variables or categorical variables that will be dichotomised a priori. The target sample size is 1300 participants.

The study has been approved by the Ethical Committee for Research with Medicines (CEIm) IDIAP Jordi Gol (24/228-P) and the Ethical Research Committee of the WHO (ERC.0004213). The results will be disseminated in peer-reviewed journals, on social media and policy fora in national, European and global context, and will inform WHO and European Union-level policy recommendations.

European Commission, Directorate General for Health and Food Safety, SANTE/2022/SI2.883729.

NCT06915298. https://clinicaltrials.gov/study/NCT06915298