Effectiveness of health warning labels on alcoholic beverages: A systematic review with a narrative analysis for Austria

Introduction: Alcohol is a toxic and psychoactive substance and a leading modifiable risk factor for numerous diseases, including several cancers, causing a substantial societal and economic burden. Health warning labels (HWLs) on alcoholic beverages are discussed internationally as a low-threshold measure to inform consumers about health risks, yet they remain rare worldwide, and evidence on their effectiveness –especially in Austria– is limited.

Research aims: This report aimed (i) to identify the HWL designs described in the literature, (ii) to assess the effectiveness of HWL compared with no HWL and with one another regarding health-related and economic outcomes in the general population, and (iii) to derive an evidence-based contextualisation for Austria.
Methods. A systematic search of the Cochrane Library, Embase, Ovid Medline, HTA (INAHTA) and PsycINFO was conducted for publications from January 2016 to February 2026 in English and German. Literature was selected independently by two researchers and data extraction was double-checked. Risk of bias was assessed using RoB 2 and CAMELOT, and certainty of evidence using GRADE. The Austrian contextualisation was based on the WHO-INTEGRATE Evidence-to-Decision Framework.
Results. Ten primary studies (nine randomised experimental, one qualitative) were included, covering four HWL types: text-only, text-and-pictogram, text-and-image and image-only. HWLs communicating cancer risk were most consistently associated with higher risk perception, a shift in consumption behaviour and intentions to change alcohol consumption. HWL types did not differ in effectiveness for risk perception and intentions. For behaviour, one study found an image-only HWL (showing a diseased bowel) most effective, followed by a text-and-image and text-only cancer warning. No eligible evidence was identified for alcohol-related disease incidence, attitudes and beliefs towards alcohol consumption, or economic outcomes. Risk of bias was high for risk perception and intentions, mainly due to self-reported and unblinded measurement, and low to moderate for behaviour. Certainty of evidence was moderate for risk perception and behaviour. Contextualisation suggests HWLs may be a favourable, feasible and low-intrusiveness intervention for Austria, although evidence is limited by mostly single-exposure experimental designs.

Conclusion: HWLs, particularly those addressing cancer risk, may strengthen health risk perception, shift consumption behaviour and increase intentions to change alcohol consumption. As robust real-world and long-term evidence is still lacking, HWLs in Austria should be considered only as one component of a broader alcohol policy, not as a stand-alone intervention.