Cancer Causes & Control

, 19:1243

Dietary patterns and risk of bladder cancer: a factor analysis in Uruguay

Authors

    • Grupo de Epidemiología, Departamento de Anatomía Patológica, Facultad de MedicinaHospital de Clínicas
  • Paolo Boffetta
    • International Agency for Research on Cancer
  • Alvaro L. Ronco
    • Departamento de Epidemiología y Métodos Científicos, Facultad de MedicinaCentro Latinoamericano de Economía Humana (CLAEH)
  • Hugo Deneo-Pellegrini
    • Grupo de Epidemiología, Departamento de Anatomía Patológica, Facultad de MedicinaHospital de Clínicas
  • Giselle Acosta
    • Grupo de Epidemiología, Departamento de Anatomía Patológica, Facultad de MedicinaHospital de Clínicas
  • María Mendilaharsu
    • Grupo de Epidemiología, Departamento de Anatomía Patológica, Facultad de MedicinaHospital de Clínicas
Original Paper

DOI: 10.1007/s10552-008-9195-9

Cite this article as:
De Stefani, E., Boffetta, P., Ronco, A.L. et al. Cancer Causes Control (2008) 19: 1243. doi:10.1007/s10552-008-9195-9

Abstract

Objective

To determine the major dietary patterns associated with bladder cancer risk, we conducted a principal components analysis (PCA) in a case–control study from Uruguay.

Methods

A total of 255 newly diagnosed and microscopically confirmed cases of transitional cell carcinoma of the bladder and 501 hospitalized controls were included in the study. Both series were drawn from the four major public hospitals in Montevideo, Uruguay. Cases and controls were frequency matched on age and sex. Controls were submitted to factor (principal components) analysis.

Results

We retained three factors that explained 25.1% of the total variance (including error variance). The first factor was labeled as the sweet beverages pattern. This factor was characterized by high loadings of coffee, tea, and added sugar and was strongly associated with risk of bladder cancer (OR 3.27, 95% CI 1.96–5.45). The second factor was labeled as the Western pattern and displayed high loadings of red meat, fried eggs, potatoes, and red wine. This pattern was directly associated with risk of bladder cancer (OR 2.35, 95% CI 1.42–3.89). Finally, the third factor was labeled as the prudent pattern and showed high loadings of fresh vegetables, cooked vegetables, and fruits. This pattern was not associated with risk of bladder cancer.

Conclusions

According to our study, non-alcoholic beverages were the strongest risk factor for bladder cancer, whereas the Western pattern was also associated with a significant increase in risk of bladder cancer.

Keywords

Bladder cancer Dietary patterns Factor analysis Principal components

Copyright information

© Springer Science+Business Media B.V. 2008