Endometrial cancer is associated with increased weight and body size, diabetes, and other conditions that may result from an excess in calories or lack of physical activity. Although a few studies have explored the effect of dietary constituents on the risk of endometrial cancer, the nature of the joint association of these constituents and obesity, energy intake, or energy expenditure with risk is unknown. A population-based case-control study was conducted in Hawaii to examine the association of diet, body size, and physical activity with the risk of endometrial cancer. Subjects included 332 histologically confirmed, primary endometrial cancer cases and 511 controls identified between 1985 and 1993. Cases and controls were residents of Oahu, Hawaii who were between 18 and 84 years of age and were from one of the following ethnic groups: Japanese, Caucasian, Native Hawaiian, Filipino, and Chinese. Cases were identified through the Hawaii Tumor Registry and matched to the controls on age (±2.5 years) and ethnicity. In-person interviews, conducted in the subjects' homes, included dietary, reproductive, menstrual, and medical histories and use of exogenous hormones, physical activity, and other lifestyle variables. Weight, girth, and skinfold measurements were taken at the time of the interview. We found a strong dose-response relation of increased body size to the development of endometrial cancer after adjustment for energy intake. The odds ratio (OR) for endometrial cancer among women in the highest quartile of body mass index (BMI; kg/m2) was more than four times that among women in the lowest quartile. Waist, hip, midarm, and wrist girths were positively associated with the estimated risk of endometrial cancer after adjustment for total calories and other nondietary risk factors, although the trends in the ORs were attenuated after adjustment for BMI. Physically active women had a modest reduction in their risk of disease compared with inactive women. Cases consumed a greater percentage of their calories from fat and a lower percentage of their calories from carbohydrates than did controls. Adjustment for BMI reduced the ORs for the highest compared with the lowest quartile of fat calorie intake from 2.0 (95% confidence interval, 1.3–3.2) to 1.6 (95% confidence interval, 1.0–2.6), suggesting that part of the association is explained by obesity. There was a differential effect of fat on endometrial cancer according to BMI. For all components of fat, the associations with endometrial cancer were either minimal or absent among leaner women (i.e., those with BMI below the median), whereas, among more obese women, two-fold differences in risk were observed between women above and below the median of fat intake. Foods that are high in fat and cholesterol, such as red meat, margarine, and eggs, were positively associated with endometrial cancer, whereas cereals, legumes, vegetables, and fruits, particularly those high in lutein, were inversely associated. These findings suggest that women who avoid being overweight and who consume a diet low in plant and animal fats and high in complex carbohydrates are at a reduced risk of endometrial cancer.
This investigation was supported in part by USPHS Grants P01-CA-33619, R01-CA-58598, R01-CA-55700, and P20-CA-57113 and by contracts N01-CN-05223 and N01-CN-55424 from the National Cancer Institute, NIH, Department of Health and Human Services.