Changes in dietary glycemic index and glycemic load in Australian adults from 1995 to 2012.

Background: Australians have used the glycemic index (GI) since 1995; however, there are no data on changes in carbohydrate quality over time.Objectives: The aim was to compare average dietary GI and glycemic load (GL), and contributing carbohydrate foods, in the 2 most recent national dietary surveys.Design: Dietary data from adult participants of national nutrition surveys conducted in 1995 (the 1995 Australian National Nutrition Survey; n = 8703) and 2012 (the 2011-2012 National Nutrition and Physical Activity Survey; n = 6278), collected by a single 24-h recall, were analyzed. The differences in mean dietary GI and GL between surveys were compared by using 1-factor ANOVA. The main sources of dietary GL in the 2 surveys were also assessed. Multiple linear regression was performed to examine the contributions of the food groups to interindividual variations in dietary GI and GL.Results: Overall, dietary GI and GL decreased by 5% and 12%, respectively, from 1995 to 2012 (GI on glucose standard: 56.5 ± 6.2 compared with 53.9 ± 6.8, respectively; GL: 153.3 ± 62.1 compared with 135.4 ± 58.5, respectively; both P < 0.001). Breads were the main contributor to GL at both time points. Potatoes and sweetened beverages contributed less, whereas cereal-based dishes contributed more in 2012 than in 1995. The top 20 GL-contributing food groups explained less interindividual variation in dietary GI (R2: 0.376 compared with 0.290) and GL (R2: 0.825 compared with 0.770) in 2012 than in 1995.Conclusion: Although the average dietary GI and GL declined between 1995 and 2012, trends in specific carbohydrate foods suggest that Australians are avoiding potatoes and sugary beverages in favor of a greater variety of carbohydrate foods, particularly cereal products.