Case-control study of skin cancer at the Dermatology Department at the University Hospital of Taubaté in the period between January 2005 - December 2006. / Estudo tipo caso controle de câncer da pele no Serviço de Dermatologia do Hospital Universitário de Taubaté no período de janeiro 2005 - dezembro 2006.

Flávia Regina Ferreira

Objective: To identify factors associated to skin cancer in the city of Taubaté-SP using a hospital-based case-control study and logistic regression, with hierarchical pattern. Methods: It was a hospital-based case-control study where the cases were individuals who had skin cancer, residents in the city of Taubaté, assisted at the Dermatology Department at the University Hospital of Taubaté in the period between January 2005 to December 2006 and the controls were individuals assisted at the same Department, also residents in this city, but without this disease and that have never had it. The individuals were combined in according to an Odds Ratio 2.1, for an alpha of 5% and a 80% power of test. This calculation was carried out by the Epi-info 6.0 turning in 127 individuals for each group. There was used a standard questionnaire, applied by the author. The independent variables were: age, sex, phototype, European ancestry, time of residence in Taubaté, occupational solar exposure, time of occupational exposure, time-table of occupational exposure, daily hours of occupational exposure, non occupational solar exposure, time-table of non occupational exposure, daily hours of non occupational exposure, photoprotection, family history of skin cancer and photodamage. Odds Ratio was the measure of effect used, and logistic regression with hierarchical pattern in three levels was applied to control confounding factors. The agreement of the model was appreciated by Hosmer-Lemeshow test and the ROC curve was built. The level of significance adopted was alpha = 5%. Results: There were included 132 cases and 132 controls in this study. In the univariate analysis possible associations were identified for: advanced age, phototype up to II in the Fitzpatricks classification, daily hours of occupational and non occupational solar exposure, photoprotection and family history of skin cancer. After multivariate analysis by logistic regression, with hierarchical pattern remained significant with respective Odds Ratio and confidence intervals (CI 95%): phototype 3.14 (1.79-5.49), number of hours of occupational 1.76 (1.04-2.99) and non occupational 1.80 (0.98-3.29) solar exposure and family history of skin cancer 2.10 (1.13-3.93). The Hosmer-Lemeshow test showed a qui-square of 1.74 (p = 0.97). The ROC curve showed an accuracy of 70 % (CI 95% 63- 76) with p <0.001. Conclusion: The results show that fair skin, family history of skin cancer and the intense occupational and non occupational solar exposure were factors associated with increased risk for cutaneous cancer in the city of Taubaté-SP.

AUTOR(ES)

DATA DE PUBLICAÇÃO

RESUMO

ASSUNTO(S)

ACESSO AO ARTIGO

Documentos Relacionados