Comparison of Different Approaches to Assess Carbohydrate Quality of Grain Foods: Glycemic Index Versus Carbohydrate Metrics. (14th June 2022)

• Record Type: Journal Article
• Authors: Fontanelli, Mariane; Batista, Lais; Rogero, Marcelo; Micha, Renata; Mozaffarian, Dariush; Fisberg, Regina; Sarti, Flavia
• Is Part Of: Current developments in nutrition. Volume 6(2022)Supplement 1
• Page Start: 482
• Page End: 482
• Publication Date: 2022-06-14
• Subjects: Nutrition -- Periodicals; Nutritional Physiological Phenomena; Nutrition; Periodicals; Periodicals; Fulltext; Internet Resources; Periodicals; 612.3
• DOI: 10.1093/cdn/nzac059.010
• Languages: English
• ISSNs: 2475-2991

Abstract

Abstract: Objectives: Carbohydrate quality remains a concept under development in the field of Nutrition, and different methods have been proposed to assess healthfulness of carbohydrate-providing foods. The study investigated the relationships between carbohydrate metrics and glycemic index (GI) of grain foods. Methods: The investigation was based on data from the population-based study 2015 Health Survey of Sao Paulo, including a probabilistic sample of residents in Sao Paulo, Brazil. Foods reported by 1741 participants in 24 h recalls were categorized according to USDA food groups, and grain foods were identified using the grain products group. The following metrics were applied per 10 g of carbohydrate: 1) ≥1 g fiber (10:1), 2) ≥1 g fiber and <1 g free sugars (10:1:1), 3) ≥1 g fiber and <2g free sugars (10:1:2); and 4) ≥1 g fiber and per 1 g of fiber < 2 g free sugars (10:1|1:2). We used kappa statistics to assess agreement between metrics, Mann-Whitney test to compare GI values (glucose reference) across metrics, and chi-squared test to compare proportions of foods meeting the metrics and GI categories. Results: A total of 244 grain foods were identified and the highest number of foods met the 10:1 (15.6%), followed by 10:1|1:2 (13.1%), 10:1:2 (12.3%), and 10:1:1 (9.8%) [kappa = 0.86]. The median (IQR) GI of grain foods was 66.6 (5.4), and foods meeting the 10:1 had lower median GI than those not meeting this metric (61.4 vs 66.8, P = 0.043), as well as grain foods meeting the 10:1:1 (50.7 vs 66.7, P = 0.020). Grain foods were classified in low (13.9%), medium (70.9%), and high GI (15.2%). The highest proportion of low GI foods was identified by the 10:1:1 (54.2%), followed by 10:1|1:2 (46.9%), 10:1:2 (43.3%) and 10:1 (42.1%), and foods not meeting the metrics identified 9.7%−10.7%. Regarding high GI foods, the highest proportion was identified by the 10:1:2 (36.7%), 10:1|1:2 (34.4%), 10:1 (28.9%), and 10:1:1(25%), while foods not meeting the metrics identified 12.3–14.1%. Conclusions: The 10:1 and 10:1:1 metrics identified lower GI grain foods, although a higher proportion of high GI foods was observed among foods meeting the metrics, especially the 10:1:2 and the 10:1|1:2. Considering the perspective of postprandial glycemic response, findings support the use of the 10:1 and 10:1:1 to identify higher carbohydrate-quality grain foods. Funding Sources: São Paulo Research Foundation.