Abstract
AIM: Obesity has been reported to be associated with increased cancer risk. Body mass index (BMI) and waist circumference (WC) are representative measures of adiposity, but BMI does not accurately reflect body composition. We aimed to evaluate the association between BMI, WC, and cancer risk in Korean old people. METHODS: We used Korean National Health Insurance Service data of the elderly population (65-80 years) who underwent a health examination in 2009 without a history of malignancy. The relative hazard ratios (HRs) for cancer of the 247,625 elderly subjects were analyzed according to their BMI and WC categories. RESULTS: During a median follow-up duration of 11.296 years, 43,369 cancer cases developed. In the fully adjusted model, the HRs for cancer incidence were significantly lower in the higher quartiles of BMI (Q1, 1; Q2, 0.920 [0.894-0.946]; Q3, 0.901 [0.873-0.930]; Q4, 0.880 [0.846-0.914], p < 0.001). Meanwhile, there was a sequential increase of cancer risk in the higher quartiles of WC (Q1, 1; Q2, 1.038 [1.008-1.070]; Q3, 1.074 [1.041-1.108]; Q4, 1.146 [1.103-1.190], p < 0.001). Even in those of normal BMI, higher WC significantly increased the cancer risk. The association between a one-standard-deviation (SD) increase in WC and cancer risk was more prominent in elderly men and the subjects with impaired fasting glucose (p for interaction < 0.05). CONCLUSIONS: In old individuals, increased BMI was significantly associated with a reduced cancer risk, whereas higher WC significantly increased it. WC, rather than BMI, should be applied as an anthropometric indicator in cancer prediction in older adults.