The prediction of human cutaneous irritation has moved away from being primarily based on the use of experimental animals. The most widespread method applied was that based on the original procedures of Draize et al involving the rabbit, but some workers have employed other species such as mice, guinea pigs, or domestic pigs. There are however, inherent problems of extrapolating from animals to humans. There are also practical, economic and ethical reasons for attempting to devise alternatives to Draize type tests. Whilst some progress has been made in terms of alternative in vitro test systems, in vivo methods using human volunteers are more easily interpretable and are able to predict clinically relevant consumer end points, such as erythema, oedema, scaling and other undesirable consequences of exposure to an irritant. In vitro acute toxicology models, using exposures up to 48 hours to test compounds or cosmetic formulations are able to provide some data on likely irritation potential but cannot describe the range of consumer relevant end points described above that can manifest despite low toxicology irritation ranking. We have previously published work examining the optimum methodology for cutaneous irritation testing, including the influence of exposure time and occlusive chamber size in predicting irritation in even very weakly irritant cosmetic products. In this paper, we present research using experimental patch test models with surfactants and blends used in personal care products. In these studies, we sought to determine the correlation between visual scoring of erythema by trained expert assessors, with measurement of erythema using a Chromameter™ and skin temperature using an infra-red non-contact thermometer. In addition, we sought to correlate the degree of irritation with histological changes observed using invivo confocal microscopy of the skin. All volunteers involved in these studies met the inclusion and exclusion criteria and gave written consent. The research was approved by an independent ethics committee. In initial experiments, we used 0.2% solutions of sodium lauryl sulphate in water to induce irritation using an occluded patch test protocol. Chambers were placed on the upper back for 4 consecutive days, with clinical scoring and Chromameter™ measurements made daily before re-application of patches. We saw that there was a positive correlation coefficient of 0.94 between expert visual grading and Chromameter™ readings. In order to determine whether either expert assessment or instrumental measurements were as well correlated with weaker irritants than sodium lauryl sulphate solution, we compared several surfactants, including some considered to be mild in comparison to sodium lauryl sulphate. The studied surfactants included sodium lauryl ethyl sulphate (SLES) and cocamido propyl betaine (CAPB) amongst others. We saw that trained expert assessment of erythema generated by 0.2% solutions applied for 4 days under occlusion ranked the irritancy in the same rank order as with instrumental assessment. Using defined ratios of SLES and CAPB, we were again able to correlate expert grading to instrumental assessment, with a correlation coefficient of greater than 0.94. The relationship between instrumental erythema measurement and expert assessment was less well correlated with fully formulated cosmetic products. In a comparison of marketed products, two that had caused consumer reports of irritation were ranked most irritating by the expert assessors in a patch test, but not by the Chromameter™ measurements. In order to determine whether other measurement end points could improve the instrumental assessment, we also used non-contact infra-red thermometry to measure skin temperature. An additional reason to consider this measurement was for the assessment of irritation where a coloured topical product may make visual or Chromameter™ measurements unreliable, such as for formulations containing dithranol, artificial tanning products and temporary skin tattoos. These data will be presented in this paper, together with clinical evaluation by the dermatologist of histological changes in the skin, visualized using invivo confocal microscopy, associated with the ranking of erythema by the expert assessors. Conclusions will be drawn regarding the relationship between clinically observed erythema, instrumental measurement correlations and histological changes in the stratum corneum and epidermis associated with inflammation and how cutaneous irritation testing can be advanced to increase the safety for consumers using cosmetic and other topical products.