Design.-Case-control study.

Setting.-Outpatient clinics in referral hospitals.

Patients.-Cases were 716 consecutive patients with newly diagnosed melanoma identified at 2 melanoma centers between January 1, 1991, and December 31, 1992. Stratified random sampling of patients from outpatient clinics was used to identify 1014 participating controls of the same age, sex, race, and geographic distribution as the melanoma cases. All study subjects underwent an interview, a complete skin examination, photography of the most atypical nevi, and, if the patient was willing, a biopsy of the most atypical nevus.

Main Outcome Measures.-Number and type of nevi on the entire body were systematically reported. All diagnoses of clinically dysplastic nevi were confirmed by expert examiners.

Results.-Risk for melanoma was strongly related to number of small nevi, large nondysplastic nevi, and clinically dysplastic nevi. In the absence of dysplastic nevi, increased numbers of small nevi were associated with an approximately 2-fold risk, and increased numbers of both small and large nondysplastic nevi were associated with a 4-fold risk. One clinically dysplastic nevus was associated with a 2-fold risk (95% confidence interval, 1.4-3.6), while 10 or more conferred a 12-fold increased risk (95% confidence interval, 4.4-31). Congenital nevi were not associated with increased risk of melanoma.

Conclusions.-Although nondysplastic nevi confer a small risk, clinically dysplastic nevi confer substantial risk for melanoma. On the basis of nevus number and type, clinicians can identify a population at high risk of this epidemic cancer for screening and intervention.