Anthropometric measures and bladder cancer risk: A prospective study in the EPIC cohort

Nina Roswall*, Heinz Freisling, H. B. Bueno-de-Mesquita, Martine Ros, Jane Christensen, Kim Overvad, Marie Christine Boutron-Ruault, Gianluca Severi, Guy Fagherazzi, Jenny Chang-Claude, Rudolf Kaaks, Annika Steffen, Heiner Boeing, Marcial Argüelles, Antonio Agudo, Maŕa José Sánchez, Maria Dolores Chirlaque, Aurelio Barricarte Gurrea, Pilar Amiano, Nick WarehamKay Tee Khaw, Kathryn Erica Bradbury, Antonia Trichopoulou, Helen Maria Papatesta, Dimitrios Trichopoulos, Domenico Palli, Valeria Pala, Rosario Tumino, Carlotta Sacerdote, Amalia Mattiello, Petra H. Peeters, Roy Ehrnström, Paul Brennan, Pietro Ferrari, Börje Ljungberg, Teresa Norat, Marc Gunter, Elio Riboli, Elisabete Weiderpass, Jytte Halkjær

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

27 Citations (Scopus)


Anthropometric measures have been related to risk of several cancers. For bladder cancer, however, evidence is sparse. Comparability of existing studies is hampered by use of different obesity-measures, inadequate control for smoking, and few female cases. This study examined associations between height, weight, waist and hip circumference, waist-hip ratio, waist- height ratio, body mass index (BMI), recalled weight at age 20 and bladder cancer, and investigated effect modification by age, tumor aggressiveness and smoking. The study was conducted in the European Prospective Investigation into Cancer and Nutrition cohort, in 390,878 participants. Associations were calculated using Cox Proportional Hazards Models. During followup, 1,391 bladder cancers (1,018 male; 373 female) occurred. Height was unrelated to bladder cancer in both genders. We found a small but significant positive association with weight [1.04 (1.01-1.07) per 5 kilo], BMI [1.05 (1.02-1.08) per 2 units], waist circumference [1.04 (1.01-1.08) per 5 cm], waist-hip ratio (1.07 (1.02-1.13) per 0.05 unit] and waist-height ratio [1.07 (1.01-1.13) per 0.05 unit] in men. Stratification by smoking status confined associations in men to former smokers. In never smokers, we found no significant associations, suggesting residual confounding by smoking. Results did not differ with tumor aggressiveness and age. Residual analyses on BMI/waist circumference showed a significantly higher disease risk with BMI in men (p = 0.01), but no association with waist circumference. In conclusion, in this large study, height was unrelated to bladder cancer, whereas overweight was associated with a slightly higher bladder cancer risk in men. This association may, however, be distorted by residual confounding by smoking.

Original languageEnglish
Pages (from-to)2918-2929
Number of pages12
JournalInternational Journal of Cancer
Issue number12
Publication statusPublished - 15 Dec 2014
Externally publishedYes


  • Anthropometry
  • Bladder cancer
  • Cohort study
  • Obesity


Dive into the research topics of 'Anthropometric measures and bladder cancer risk: A prospective study in the EPIC cohort'. Together they form a unique fingerprint.

Cite this