Shaw et al have raised several questions concerning our report on survival superiority of females with melanoma. We considered Clark's level of invasion in the analysis. Although there was a slightly greater prevalence of higher Clark's levels IV and V among male patients, adjustment for Clark's level in the proportional hazards model did not change the finding of higher mortality rates and shorter disease-free interval among males. Clark's level was not selected for inclusion in the final multivariant models because it had a minimal effect on the male-female comparisons of survival and disease-free interval, ie, there was no evidence in these data that women had more favorable outcomes because of lower-stage disease at diagnosis.
Ulceration was not included in the data set used in these analyses. Measurement of ulceration is inexact and we elected to not include it even though we agree with Shaw et al that ulceration