An ingredient co-occurrence network gives insight into e-liquid flavor complexity

Jeroen L A Pennings; Ina M Hellmich; Sanne Boesveldt; Reinskje Talhout

doi:10.18332/tid/175955

An ingredient co-occurrence network gives insight into e-liquid flavor complexity

Tob Induc Dis. 2024 Jan 10:22. doi: 10.18332/tid/175955. eCollection 2024.

Authors

Jeroen L A Pennings¹, Ina M Hellmich^{1

2}, Sanne Boesveldt², Reinskje Talhout¹

Affiliations

¹ Centre for Health Protection, National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands.
² Division of Human Nutrition and Health, Wageningen University, Wageningen, The Netherlands.

Abstract

Introduction: Part of the appeal of e-cigarettes lies in their available flavors. To achieve attractive flavors, e-liquids contain many different flavoring agents, which allow many flavoring combinations. To advance our knowledge of e-liquid flavors and compositions and to evaluate the effect of legislation, we determined whether there are ingredient combinations that are frequently used together.

Methods: We used e-cigarette ingredient data from the European Common Entry Gate system (EU-CEG) as available on 31 December 2022.

Results: In e-liquids, we found 214 ingredient pairs with a co-occurrence odds ratio greater than 10. Together, these consisted of 62 unique ingredients. Network analysis revealed that ingredients were grouped together based on their flavor and/or chemical structure. We identified two densely connected regions (clusters) in the network. One consisted of six ingredients with sweet-vanilla-creamy flavors. The second cluster consisted of 13 ingredients. While some of these have fruity flavors, others, such as alkyl carboxylic acids and dimethyl sulfide, are known to have unpleasant flavors. Additional data and literature analyses indicated that alkyl carboxylic acids can contribute to a creamy and sweet-fruity taste, whereas dimethyl sulfide can contribute to a more refined fruity taste.

Conclusions: These results exemplify that the flavor of e-liquids is not just the sum of its parts. Big data analyses on product data can be used to detect such patterns, but expert knowledge and additional data are needed for further interpretation. Monitoring of e-liquid flavors as well as ingredients will remain important to regulate e-liquid product attractiveness.

Keywords: EU-CEG; attractiveness; dimethyl sulfide; e-liquids; flavor ingredients.

Grants and funding

FUNDING This work was supported by the Netherlands Food and Consumer Product Safety Authority (NVWA), project 9.7.1.3. The funding source was not involved in the design of the study; in the collection, analysis and interpretation of the data; or the writing of the manuscript.