SPF TESTING METHODOLOGIES 183 followed by its execution by a staff member. The subject was given a copy of the fully executed Informed Consent Document. ULTRAVI OLET RADIATION (UVR) SOURCE Xenon A rc Solar Simulators from Solar Light Company, Philadelphia, PA (150 or 300 W), were used as the source of UVR (12). The spectral output for the 150 and the 300 W were essentially identical (13,14). The lamp output was measured with a UV intensity meter (Model PMA2100, Solar Light Company) thrice daily. Solar simulators were equipped with 1-mm UG11 and WG320 fi lters, providing a spectral output in the ultraviolet range (290–400 nm) comparable with that of natural sunlight and meeting both 2011 FDA-Final Rule and ISO 24444 standards (see example spectral output for single-port solar simulator (Figure 1 and Table II) and for multiport with six light guides (Figure 2 and Table III). Irradiation beams were a minimum of 1 cm2 with a beam uniformity of 10%, and they exhibited less than 20% time-related fl uctuation. All solar simulators were calibrated and adjusted to deliver energies within 10% variance. SPF DET ERMINATION An MED is defi ned by ISO 24444 (7) as “the lowest dose of UVR that produces the fi rst perceptible unambiguous erythema with defi ned borders appearing over most of the fi eld of UV exposure, 16 to 24 h after UV exposure.” An MED is defi ned by 2011 Final Rule (6) as “The quantity of erythema-effective energy (expressed as Joules per square meter) Figure 1. Typical spectral output of a 150-W xenon arc single-port solar simulator.

JOURNAL OF COSMETIC SCIENCE 184 required to produce the fi rst perceptible, redness reaction with clearly defi ned borders.” Although these two defi nitions vary slightly, the implementation of the defi nitions is es- sentially identical. Each evaluator of erythema was qualifi ed through training and evalu- ation testing. A subje ct’s SPFi, defi ned as the ratio of the MED on protected skin (MEDp) to the MED on unprotected skin (MEDu), was calculated for each subject as follows: MEDi protected skin MEDpi SPFi MEDi unprotected skin MEDui SUMMAR Y O F DATA A total of 2,607 observations encompassing 664 subjects were collected from February 2016 to September 2017. Of these observations, 104 observations (3.99%) on 86 subjects were collected from clinical trials labeled as invalid. An invalid observation occurs when all test sites exhibit an erythema score of at least 1, no test sites exhibit an erythema score of 1, or the erythema scores do not follow the irradiation sequence (2011 FDA-Final Rule method only). These 104 observations were not included in the data analysis. The result- ing sample consisted of 2,503 valid observations encompassing 652 subjects. The arith- metic average SPF value for all 2,503 observations was 15.6 ± 2.5, before stratifying by the method used and before incorporating any statistical model of the data. Within the Table II Typical Spectral Output of a 150 W Xenon Arc Single-Port Solar Simulator Total irradiance (250–1,600 nm) 1.20E-01 W/cm2 UVC irradiance (250–290 nm) 2.02E-07 W/cm2 UVB irradiance (290–320 nm) 1.18E-02 W/cm2 UVA irradiance (320–400 nm) 1.02E-01 W/cm2 UVA2 irradiance (320–340 nm) 2.88E-02 W/cm2 UVA1 irradiance (340–400 nm) 7.36E-02 W/cm2 Visible + NIR irradiance (400–1,600 nm) 5.55E-03 W/cm2 %UVC 0.0002% – %UVB 9.81% – %UVA 85.55% – %Visible + NIR 4.63% – SED 12.2 s Erythemal effective irradiance 8.21E-04 W/cm2 % Erythemal contribution of FDA (June 2011)/ISO 24444 SPF method (November 2010) 290 nm ( 0.1%) 0.02% 290–300 nm (1.0–8.0%) 4.77% 290–310 nm (49.0–65.0%) 57.71% 290–320 nm (85.0–90.0%) 87.74% 290–330 nm (91.5–95.5%) 93.32% 290–340 nm (94.0–97.0%) 95.48% 290–400 nm (99.9–100.0%) 99.98% %UVA2/Total UV (≥20%) (320–340 nm/290–400 nm) 25.25% %UVA1/Total UV (≥60%) (340–400 nm/290–400 nm) 64.45% Total irradiance 250–1,400 nm for FDA ( 1,500 W/m2) 1,196 Total irradiance 250–1,500 nm for ISO ( 1,600 W/m2) 1,197