Test design and results of a method performance characterization study for SPF and UVA-PF testing.

OBJECTIVE

This paper presents the experimental design, the statistical model, performance criteria and results of a method performance study conducted in order to characterize the performance of alternative methods for SPF and UVA-PF testing. The performance of alternative methods is characterized against the performance of the ISO 24444 and ISO 24443 methods.

METHODS

A factorial experimental design was implemented. For a given method, such a design allows good statistical reliability of precision estimates with as few as four participating laboratories. To cover the wide range of sunscreen products on the market, 32 sunscreens were tested. These 32 products were classified into eight product groups. The product groups differ in terms of SPF level, application viscosity and type of UV filters and matrix. In co-operation with a WG7 ISO ad hoc group performance criteria for the alternative methods were developed for reproducibility precision, between-laboratory variation, bias and the variation of bias across products and across product groups.

RESULTS

Performance characteristics for the reference method along with reproducibility and repeatability limits are provided. The results for the participating alternative methods will be presented in separate articles, following this paper.

CONCLUSIONS

The precision of ISO 24444 and ISO 24443 was characterized for each of the eight product groups. The main sources of variation were identified for each of the two reference methods. Reproducibility and repeatability limits were derived and are provided.