Abstract
Ultraviolet light from sunlight is implicated in the aetiology of non-ocular malignant melanoma, mostly from Western studies. This study reports the relation of such melanomas from 7 cancer registries in different parts of India with latitude, altitude, ozone levels and ultraviolet (UV) light exposure. A log linear model was fitted using GLIM, assuming Poisson errors in incidence; linear and non-linear regression techniques were also used. Results show slight negative associations of melanoma with latitude, X2 = -3.37, p = 0.07, and statistically nonsignificant ones with atmospheric ozone levels (r = -0.36, n.s.), and a positive association with UV (r = 0.30, n.s.). Melanoma incidence from 4 places relates in a parabolic curve with latitude (y = I-0.09 X2 + 0.003 X2 in males and y = 1.73–0.17 X2 + 0.005 X2 in females). Patients originally from west coastal ethnic/linguistic communities predominate in Bombay. Geographic patterns of melanoma in Indians are similar with those in white Caucasians. Ultraviolet light exposure may also be involved in its aetiology in non-white Caucasians, such as Indians.