Dermoscopy

A conservative estimate based on metaanalysis of nine prospective studies with consecutively recruited patients suggests that dermoscopy is roughly nine times more sensitive for detecting melanoma (odds ratio 9.0, 95% confidence interval 1.5-54.6) than naked-eye examination.1 Sensitivity was 18% higher with dermoscopy than with naked-eye examination (0.87 v. 0.69, p = 0.008), with no difference in specificity.1 Four studies have shown that these estimates apply not only to specialists but also to generalists trained in the procedure.2 For images illustrating the use of dermoscopy in the diagnosis of melanoma, see Appendix 1, available at www.cmaj .ca/lookup/suppl /doi :10 .1503/cmaj.140008/-/DC1. Primary care physicians given a one-day training course in techniques for skin cancer detection were randomly assigned to evaluate lesions suggestive of skin cancer during a 16-month prospective trial using either dermoscopy or nakedeye examination.5 Patients with lesions suggestive of skin cancer were correctly identified in 79.2% of cases in the group examined using dermoscopy, compared with 54.1% of cases in the group examined with naked-eye examination (p = 0.002).5