Background Limited research exists on revealing the correlation between laboratory safety tests and real-world safety feedback from consumers of cosmetics. Objective To establish cosmetovigilance by associating patch test results with consumer-reported adverse reaction (CRARs) feedback form e-commerce feedback for cosmetics targeting sensitive skin. Methods Patch tests were performed on a cohort of sensitive-skin cosmetics, and CRARs were extracted from e-commerce comments. A generalized linear model was utilized to analyze the association between patch test results and CRARs, considering functional claims and application modes. Results Among the patch test results of 64 cosmetics, 42 exhibited negative reactions, 21 showed doubtful reactions, and one cosmetic demonstrated a weak positive reaction. The model established with a warning threshold of two doubtful reactions proved to be optimal. Patch test results were associated with sensitization-related and pain-related CRARs, but not with acnegenic-related CRARs. Brightening and acne treatment cosmetics exhibited higher acnegenic and pain rates, while leave-on cosmetics had a higher sensitization rate. Conclusion The data-driven results show that cosmetics with 2 or more doubtful reactions in patch test have a significantly correlation with real-world CRARs, which reminds relevant practitioners to be more cautious about doubtful reactions in patch test.
Yang et al. (Wed,) studied this question.