IRRITATION OUTCOME FOR POSITIVE/NEGATIVE CONTROLS 521 Numerical score 0 = No evidence of irritation 1 = Minimal erythema, barely perceptible Table II Scoring Scale 2 = Moderate erythema, readily visible or minimal edema or minimal papular response 3 = Strong erythema or erythema and papules 4 = Definite edema 5 = Erythema, edema, and papules 6 = Vesicular eruption 7 = Strong reaction spreading beyond test site Letter grade A = Slightly glazed appearance B = Marked glazing C = Glazing with peeling and cracking F = Glazing with fissures G = Film of dried serous exudate covering all or portion of the patch site H = Small petechial erosions and/or scabs relative humidity, RH) were obtained from the National Climatic Data Center for the US locations and the Environment Canada Meteorological Services for the Canadian location. Average RH representing each study day was calculated using the average dew point and average temperature, with appropriate equations when information was not directly available. Since humans tend to relate discomfort to the combination of hu­ midity and temperature, RH and dew point were utilized representing the humidity levels. Table I summarizes the average temperature (°C), RH(%) and dew point (°C) for each study. STATISTICAL ANALYSIS The actual patch test scores are a combination of a numerical and letter score consistent with the definitions given in the scoring scale (Table II). In order to perform statistical analyses, scores containing letter grades were converted to numerical equivalents. These were converted as follows: A=0 B= 1 C=2 and F, G, and H=3. These equivalents are considered additive to any numerical score (e.g., 2C=2+2=4). However, an upper limit of 3 is selected since the intent of a 14-day cumulative irritation test was to compare treatments that are relatively mild. Thus, for any value greater than 3, a value of 3 is used in the statistical analysis. Furthermore, following the development of a strong reaction, no further exposure was made and a score of 3 was carried forward for all scoring days for the remainder of the test. The daily irritation scores from completed study subjects were averaged. The overall cumulative irritation scores were the summation of the 14-day daily average. The overall cumulative irritation scores for each study are shown in Table I. Analysis of variance (ANOV A) was utilized to compare the differences in overall irri­ tation, temperature, RH, and dew points among the three geographical sites and to compare the overall irritation level between the positive and negative control test materials (Table III). Regression analysis was performed to see if the overall irritation score correlated with the climatic factors (i.e., temperature, RH, and dew point) (Ta­ ble IV). RESULTS Approximately 85% (71-97%) of the enrolled subjects completed the study. The pri­ mary reason for early subject withdrawal was scheduling conflict that is unrelated to the

522 JOURNAL OF COSMETIC SCIENCE Table III ANOV A Test Results among Study Centers Mean ± SD Arizona Florida Manitoba p-value Cumulative irritation (0.1 % SLS) 30.3 ± 2.7 29.9 ± 1.0 29.9 ± 2.5 0.941 Cumulative irritation (saline) 1.2 ± 1.0 0.1 ± 0.1 1.2 ± 1.3 0.084 p-value 0.05 0.05 0.05 Temperature (QC) 24.1 ± 10.4 21.0 ± 5.52 -2.4 ± 20.5 0.009 Relative humidity(%) 32.7 ± 17.8 71.7 ± 5.3 72 ± 10.7 0.000 Dew point (QC) 4.3 ± 4.4 15.5 ± 5.4 -3.98 ± 19.2 0.028 Table IV Regression Analysis Test Results between Irritation Scores and Climatic Factors Irritation scores (y) vs climatic factors (x) Regression equation r2 F value p-value 0.1 % SLS vs average temperature (QC) y = -0.05x + 30.75 0.19 3.76 0.077 0.1 % SLS vs RH(%) y = 0.0lx + 29.27 0.02 0.25 0.621 0.1 % SLS vs dew point (QC) y = 0.07x + 30.41 0.20 3.76 0.071 Saline vs average temperature (QC) y = -0.02x + 1.07 0.10 1.76 0.205 Saline vs RH(%) y = -0.0lx + 1.62 0.09 1.46 0.244 Saline vs dew point (QC) y = -0.04x + 1.07 0.38 9.11 0.009 test products. The completion rates were not significantly different between locations. The repetitive exposure to 0.1 % SLS under occlusion began showing skin irritation responses on Days 3 and 4 (when average panel scores were 0.5), and the irritation scores tended to plateau after Day 8 (when average panel scores were 2.5) for each study and study center (Figure 1). Repeated exposure to saline under occlusion illustrated minimal signs of irritation (when average panel scores were approximately 0.5 or less) for the entire study duration for each study and study center (Figure 2). As occlusion by itself may induce cutaneous irritation (4), daily average irritation scores of approximately 0.5 on average is reasonably expected, as that represents less than one half of the subjects eliciting a barely perceptible minimum sign of erythema. All studies, regardless of climate conditions, concluded that 0.1 % SLS induced statis­ tically significantly higher irritation levels than saline (p 0.05). The overall cumulative irritation scores for the positive and the negative control test materials were not sig­ nificantly different between study centers (p=0.941 and P=0.084), respectively (Ta­ ble III). The average temperature, RH, and dew point at each study center were significantly different between selected study centers (Table III). Manitoba studies were conducted under an average temperature of -2.4 ° C that was significantly lower than that of Arizona and Florida (between 21 °-24°C for these centers). Florida studies were con­ ducted under an average dew point of 15.5°C that was significantly higher than that of Arizona and Manitoba (between 4.0 and -4.0°C for these centers, respectively). Arizona studies were conducted under an average RH of 33% that was significantly lower than that of Florida and Manitoba (approximately 72% for both centers.).