Demographic, behavioural and physician-related determinants of early melanoma detection in a low-incidence population.

Summary Background Knowledge of the factors that influence early detection of melanoma is important in developing strategies to reduce associated mortality. Objectives To identify sociodemographic, behavioural and medical care-related factors associated with melanoma thickness in a low-incidence population but with a high case fatality. Patients and methods In a multicentre, retrospective, survey-based study of 202 patients with a recent diagnosis of invasive melanoma (< 1 year), we collected data on demographic and behavioural factors, attitudes towards prevention, access to medical care, frequency of skin self-examination (SSE) and physician skin examination (PSE) in relation to melanoma thickness. Results Thinner tumours (≤ 1 mm, 80 melanomas) were associated with female sex (P ≤ 0·049), nonnodular (superficial spreading melanoma, lentigo maligna melanoma, acral lentiginous melanoma) histological subtypes (P < 0·001), absence of ulceration (P ≤ 0·001), and location other than lower extremity or trunk location (P ≤ 0·004). Patients married at the time of diagnosis or who performed SSE during the year prior to diagnosis were more likely to have thinner tumours than those who did not [odds ratio (OR) 3·45, 95% confidence interval (CI) 1·48–8·04 and OR 2·43, 95% CI 1·10–5·34, respectively]. Full-body skin examination by a physician was not significantly associated with thinner melanoma (OR 1·99, 95% CI 0·66–6·07). Conclusions SSE was shown to be an important factor in the detection of thin melanoma, in contrast to partial or full-body PSE, which did not show any statistically significant effect on tumour thickness.