Contact Dermatitis 2006: 55: 357–363 Printed in Singapore. All rights reserved Copyright # Blackwell Munksgaard 2006 CONTACT DERMATITIS The effect of population diversity on skin irritation LISA PETERS 1 ,MARIE MARRIOTT 1 ,BHASHKAR MUKERJI 2 ,PINKY INDRA 2 ,JANARDHANA VAIDYANATHAN IYER 2 , ADIP ROY 2 ,MATT ROWSON 1 ,SHARMIN AHMED 1 ,KAREN COOPER 1 AND DAVID BASKETTER 1 1 Safety and Environmental Assurance Centre, Unilever, Colworth Laboratory, Sharnbrook, Bedfordshire, MK44 1LQ, UK, and 2 Environmental Safety Laboratory, B. D. Sawant Marg, Chakala, Andheri (E), Mumbai 400 099, Maharashtra, India The impact of many human variables on the response to skin irritating substances has been studied to varying degrees, including the impact of age, sex, and atopic status. However, the importance of ethnic origin has been more difficult to investigate, leading to a relative paucity of compelling data, either for or against the existence of differences. A primary reason for this lack is that studies on different ethnic groups often have to be undertaken in different locations thus introducing variables, e.g. time, environmental conditions that confound interpretations. In the present work, an attempt has been made to eliminate all variables except ethnicity by conducting a study on 2 distinct popu- lations (Punjabis and Tamils) at the same location on the same day with a single assessor of the skin reactions, using sodium lauryl sulfate as the skin irritant. The skin reactions were assessed visually, and it was demonstrated that the modality of the reactions in these 2 populations had clear differ- ences, but that the dose–response profiles were very similar. Thus, although the irritant response was expressed differently (e.g. erythema was much less evident in the darker Tamil population), the overall outcome was that the populations reacted in an equivalent manner. Key words: ethnic differences; Indian; patch test; skin irritation; sodium lauryl sulfate; visual assessment. # Blackwell Munksgaard, 2006. Accepted for publication 16 August 2006 It is well recognized that irritant contact derma- titis (ICD) responses to skin irritants commonly show interindividual variation (1, 2). When con- ducting human volunteer studies in order to gen- erate risk assessment data on cosmetic products, a random population is commonly chosen in which to assess their skin irritation potential. Examples of such studies include the predictive human patch test and repeat open application use methods. Typically, effects of interindividual variation are controlled by inclusion of a suitable positive control and sufficient panel size to enable effective statistical analysis. There are numerous published reports indicat- ing that population diversity differences play a large part in the variation in irritant response to a known stimulus. For instance, Negroid skin is considered to be less sensitive than Caucasian skin, particularly when only visual assessment is used to compare differences in ICD according to different ethnicity (3). Consequently, organiza- tions such as the European Cosmetic Trade Asso- ciation (COLIPA) recommend that the testing of finished cosmetic products is conducted using volunteers from the same population in which they are intended to be marketed in (e.g. in terms of sex, age, ethnic origin, and skin type) (4). Historically, however, many of the reports sug- gesting a difference in ICD elicited in white and black skin have only compared erythematous reactions. It is well documented that this assess- ment parameter is notoriously difficult to assess in Negroid skin and hence could be deemed to give a false impression of skin irritation potential (5). Indeed, many dermatological disorders present differently in dark skin and hence assessment of skin reactions should take into account more than just an erythema assessment (6). To avoid poten- tial misinterpretation due to only using visual assessment of erythema to assess ICD, other stud- ies have incorporated bioengineering techniques (typically transepidermal water loss measure- ments and laser Doppler flowmetry) in order to compare different populations (7–9). However, use of other visual assessment parameters (such as assessing oedema, wrinkling, glazing, skin ero- sion/scabbing separately) to compare these differ- ences has not been explored.