Calorie Control Council Statement: The Use of Low-Calorie Sweeteners in Pregnancy

ATLANTA (June 18, 2012) — The Calorie Control Council is pleased that the French Agency for Food, Environmental and Occupational Health and Safety (ANSES) today released a report that concludes that low-calorie sweeteners do not cause adverse effects during pregnancy.  ANSES does state, however, that on the basis of one study additional research on the use of low-calorie sweeteners during pregnancy is needed.   This recommendation ignores the weight of the scientific literature supporting the safety of low-calorie sweetener use during pregnancy.

• Multi-generational studies have found no adverse effects on the mother or developing baby related to the use of low-calorie sweeteners.
• Less than five percent of the women in the study experienced pre-term labor and one-third of those were medically induced. Moderate pre-term delivery (defined as delivery between 32 and 34 weeks) was less than one percent and early pre-term delivery (prior to 32 weeks) was less than half a percent.
• Weight gain was not studied during pregnancy, nor was diabetes.  It is possible that women who gained excess weight during pregnancy were more likely to drink diet soda and those who gained excess weight may have been more likely to deliver early.
• Because this study only included diet soda (and low-calorie sweeteners are used in tea, coffee, etc.,) a potential bias is introduced.
• The associations in the study are statistically weak (especially with such a large sample size). Weak associations are often the result of chance, bias, etc.
• Use of diet soda was captured at just one point during pregnancy (between 20 and 24 weeks).  Thus, is it possible that participants may have started or stopped using diet soda both before and after the information was collected.
• The Council questions the merging of the three highest intake groups into one group in data analysis since such a procedure tends to enhance statistical significance. Given the large sample size, this merging of groups might predispose their analysis to biologically meaningless statistical significance.
• Even though smoking and socioeconomic status were entered in the multivariate analysis as adjustment variables, there was a relatively “high” prevalence of smoking during pregnancy and, women using diet beverages also smoked on a daily basis and this accounted for 12-31 percent of those women. Smoking has been linked to adverse outcomes in pregnancy including pre-term labor. Further, these same women were more likely to be part of a lower socioeconomic group (94 percent or higher), which can also play a role in prenatal care and appropriate treatment.
• The study is epidemiological in nature and cannot show cause and effect. The authors note, “As with all observational studies, we cannot exclude that our findings may be a result of unidentified and unadjusted confounders.”