World Allergy Organization (WAO) Consensus on the use and practice of precautionary allergen labelling: The international ACT-UP! e-Delphi

Background: The Food and Agriculture Organization (FAO) of the United Nations and the World Health Organization (WHO) recently convened an expert consultation to address the increasing but inconsistent use of Precautionary Allergen ("may contain") Labelling (PAL). Objectives: We conducted a Delphi study to explore the perspectives of clinicians, patient representatives, and other interest-holders on PAL practice. Methods: We followed previous guidance on Delphi study development and reporting. A narrative review exploring the FAO/WHO Expert Consultation informed the Delphi process. The working group generated 35 items based on the review of FAO/WHO consultation, categorized into 7 domains: (i) Importance of the issue, (ii) Informing use of PAL, (iii) Action level cut-offs, (iv) Communication, (v) Liability considerations, (vi) Free From statements, and (vii) Areas for further research. Items underwent 3 iterative Delphi rounds involving a panel of 35 experts who rated their agreement narratively and on a 5-point Likert scale. Consensus was defined as ≥70% consistency in agreement or disagreement (excluding neutral responses); stability was defined as <10% variation in average scores across rounds. A public consultation was held to collect different viewpoints arising from the consensus activity. Results: The Delphi was concluded after 3 rounds with consensus achieved for 33/35 items. Response rate was 100% across all 3 rounds. The panel strongly agreed that PAL should only be applied when an unintended allergen may be present above a cut-off (action level) and that it should always be based on a formal risk assessment, preferably quantitative, and mandated through legislation. Furthermore, action levels should be based on "reference doses" (amount of total allergenic protein), rather than on a concentration (eg, 10 ppm), and more specifically on population-adjusted eliciting doses ED05 (amount of allergen expected to elicit an objective reaction in no more than 5% of the allergic population). To achieve a balance between protection and the risk of PAL overuse, ED05 was deemed preferable to ED01. There was 70% agreement that consumers should be informed of residual risk (ie, unintended allergen presence below the reference dose), but no consensus as to how this might be done. Finally, PAL should not be used for an allergen declared to be absent with a "free from" statement, nor should a PAL appear for an allergen listed as an ingredient. Conclusions: This Delphi explored a comprehensive set of statements across multiple aspects of PAL development, application, and communication. On the one hand, it provides systematically derived considerations, aligned with the FAO/WHO consultation recommendations, to support future harmonization of practices. On the other, it highlights several areas of uncertainty to prioritize for future research.