Reliability of naevus counts in identifying individuals at high risk of malignant melanoma.

The majority of deaths from malignant melanoma could be prevented by regular screening of at-risk individuals. However, the feasibility of introducing active screening programmes depends on the reliability of criteria for identifying those individuals who are at increased risk. Several case-control studies indicate that one of the most important risk factors is the number of benign pigmented naevi. This criterion might constitute a practical means of identifying individuals suitable for inclusion in an active screening programme. The aim of this study was to assess the reliability of using naevus counts as an indicator of risk. Two possible methods of assessment were compared: assessment by trained interviewers, and assessment by experienced dermatologists. The results suggest that this criterion may not be as practical or reliable as originally anticipated. Interviewer naevus counts correlated poorly with dermatologists' assessments (kappa = 0.19). Counts by dermatologists of the number of naevi on the left arm differed by up to nine naevi, although most counts were within four naevi of each other. There was reasonable correlation between naevus counts on the left arm and those on the whole body (r = 0.71). However, the relationship was not perfect. Selection of individuals for screening solely based on the number of naevi on the left arm may exclude many people with high total body counts but relatively low left arm counts.