diagnostic accuracy (88Æ1%) which was developed for use in the field of dermoscopy. A.Blum Department of Dermatology, University of Tu ¨bingen, Liebermeisterstrasse 25, 72076 Tu ¨bingen, Germany E-mail: a.blum@derma.de References 1 Carli P, Quercioli E, Sestini S et al. Pattern analysis, not simplified algorithms, is the most reliable method for teaching dermoscopy for melanoma diagnosis to residents in dermatology. Br J Dermatol 2003; 148: 981–4. 2 Argenziano G, Fabbrocini G, Carli P et al. Epiluminescence micro- scopy for the diagnosis of doubtful melanocytic skin lesions—comparison of the ABCD rule of dermatoscopy and a new 7-point checklist based on pattern analysis. Arch Dermatol 1998; 134: 1563–70. 3 Pehamberger H, Steiner A, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pig- mented skin lesions. J Am Acad Dermatol 1997; 17: 571–83. 4 Steiner A, Pehamberger H, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. II. Diagnosis of small pig- mented skin lesions and early detection of malignant melanoma. J Am Acad Dermatol 1997; 17: 584–91. 5 Stolz W, Riemann A, Cognetta A et al. ABCD rule of dermatoscopy: a new practical method for early recognition of malignant mela- noma. Eur J Dermatol 1994; 4: 521–7. 6 Argenziano G, Soyer HP, Chimenti S et al. Dermoscopy of pig- mented skin lesions: results of a consensus meeting via the inter- net. J Am Acad Dermatol 2003; 48: 679–93. 7 Blum A, Rassner G, Garbe C. Modified ABC-point-list of dermos- copy: a simplified and highly accurate dermoscopic algorithm for the diagnosis of cutaneous melanocytic lesions. J Am Acad Dermatol 2003; 48: 672–8. Which is the most reliable method for teaching dermoscopy for melanoma diagnosis to residents in dermatology? DOI: 10.1111/j.1365-2133.2004.06122.x SIR, We read with great interest the paper by Carli et al. 1 concerning the reliability of various methods for teaching dermoscopy for melanoma diagnosis to residents in derma- tology. To investigate the diagnostic performance of three different methods, i.e. pattern analysis (the classic procedure based on the simultaneous assessment of multiple dermo- scopic features) and two of the most used simplified algo- rithms (the ABCD rule of dermoscopy and the seven-point check-list), five residents in dermatology were asked to diagnose 200 clinically equivocal melanocytic lesions inclu- ding 44 early melanomas. As already stated in the title of their paper, the authors contended that pattern analysis, not simplified algorithms, is the most reliable method for teaching dermoscopy to residents in dermatology: we wish to challenge this notion. Their conclusion is based on the fact that pattern analysis yielded the best mean diagnostic accuracy (68Æ7%), followed by the ABCD rule (56Æ1%) and the seven-point check-list (53Æ4%). However, the best sensitivity (the number of correctly diagnosed melanomas) was associated with the use of the seven-point check-list (91Æ9%), followed by pattern analysis (85Æ2%) and the ABCD rule (78Æ1%). We agree with the authors that pattern analysis is more reliable than simplified algorithms. This is especially true when pattern analysis is used by experts in dermoscopy because it takes into consideration many variables and different criteria that only an expert in the field can be able to recognize. 2 However, when dermoscopy is used by nonexperts, e.g. residents or dermatologists not specifically trained in the field, the diagnostic procedure can be facilitated by the use of a few alternative algorithms. This is because they are based on a lower number of features to evaluate and on diagnostic scoring systems. 3–6 In a recent re-evaluation of dermoscopic algorithms, the seven-point check-list was found to allow the best mean sensitivity in the hands of nonexperts. 7 In detail, 16 colleagues not yet familiar with dermoscopy were asked to evaluate dermoscopic images of 20 pigmented skin lesions using different diagnostic methods (pattern analysis, ABCD rule, Menzies’ method and seven-point check-list), before and after an internet-based training course in dermoscopy. The results showed a considerable improvement in the dermo- scopic diagnosis of melanoma after the web-based training vs. before. Remarkably, using the seven-point check-list nonex- perts had 100% sensitivity and 69Æ8% specificity after a short web-based training, whereas pattern analysis, ABCD rule and Menzies’ method gave sensitivities of 82%, 78Æ4% and 93Æ4%, and specificities of 78Æ5%, 79Æ6% and 76%, respectively. It is well known that the main purpose for a clinician is not to miss a possible melanoma. That means that all pigmented skin lesions that may be melanoma should be removed, while minimizing the excision of benign pigmented skin lesions. Dermoscopy is known as a technique that increases the performance in differentiating benign from malignant pigmen- ted and nonpigmented skin tumours. Remarkably, dermoscopy is particularly helpful when used for a better differentiation of pigmented skin lesions that have already been judged as equivocal from a clinical point of view. The paper by Carli et al. 1 also used a series of clinically equivocal melanocytic lesions. In this context, although the specificity (the number of benign lesions correctly diagnosed) provided by the seven-point check- list was the worst (35Æ2%) given by the methods tested, this result is referring to a particular set of difficult melanocytic lesions that had already been clinically judged to be equivocal and therefore had been excised to rule out melanoma. By contrast, this method allowed nonexperts to diagnose a high percentage of melanomas (more than 90%)—a perform- ance comparable with that given by experts. 3–5 Therefore we conclude, contrary to Carli et al. 1 that the seven-point check-list and not pattern analysis is the most 512 CORRESPONDENCE Ó 2004 British Association of Dermatologists, British Journal of Dermatology, 150, 506–525