Dr. Maki Inoue-Choi of the University of Minnesota and colleagues published a report in the journal Cancer Epidemiology, Biomarkers & Prevention linking the intake of sugar-sweetened beverages to the incidence of the most common type of endometrial (uterine) cancer.
Type I endometrial cancer is thought to be associated with factors that increase estrogen exposure these include greater body fatness, postmenopausal estrogen therapy, late menopause, and not having had any children. Type II endometrial cancer, which is more clinically aggressive, is rarer, and is not thought to be estrogen-dependent. These researchers, based on a food-frequency questionnaire completed by nearly 42,000 post-menopausal women in the Iowa Women s Health Study, determined that increased consumption of sugar-sweetened beverages were significantly correlated with the occurrence of Type I, but not Type II endometrial cancer.
The study began in 1986, and the women were followed annually through 2010; any new uterine cancers were reported and specifically typed. At the beginning of the study, the participants were, on average, 62 years old, and reported their BMIs and activity levels as well as their dietary intake.
The results of the investigation revealed that low physical activity levels and the use of an estrogen were associated with a higher intake of sugary beverages (not including fruit juices). When they statistically controlled for age, smoking history, BMI, diabetes and other confounding factors, the researchers found a significant trend of increasing risk of Type I endometrial cancer as the intake of sugary beverages increased. Women in the top 1/5 of intake had a 74 percent increased risk compared to women in the lowest intake level (essentially no intake). No such correlation was found between endometrial cancer and consumption of sugar-free beverages, fruit juice, or sweets/baked goods.
So do these results mean that sugary beverage consumption causes endometrial cancer? Not by any stretch of the imagination. First, this was an observational study, which, although it was large and prospective, cannot prove causation. Second, the data relied on self-reports, which may or may not be very accurate. Third, and most telling, the study s objectives were determined after the data were collected, a form of selective evaluation that violates the main dictates of sound epidemiological research.
The take-home message, says ACSH s Dr. Ruth Kava, is cautionary. These data compared only the highest level of reported intake to the lowest level, and we don t know if intermediate levels of intake would also be linked to an increased occurrence of endometrial cancer. Thus the best advice, as always, is to keep intake of such beverages to a moderate level.
ACSH s Dr. Gilbert Ross had another point of view: This study is basically junk, as it represents true data-dredging, meaning the authors tortured and squeezed their data until out popped some statistically significant link. I hope no one thinks that they set out to study the effects of sugary beverage intake on type I uterine cancer that s the only outcome that fell within their statistical significance trap. Further, in such a study, a relative risk of 74 percent is nothing important, even if it were true.