Sylvetsky Allison C, Mitchell Ellen L, Grilo Mariana F, Um Caroline Y, Wang Ying, Hodge Rebecca A, Patel Alpa V, McCullough Marjorie L
Department of Exercise and Nutrition Sciences, Milken Institute School of Public Health, The George Washington University, Washington, DC, United States.
Department of Population Science, American Cancer Society, Atlanta, GA, United States.
Am J Clin Nutr. 2025 Mar;121(3):663-674. doi: 10.1016/j.ajcnut.2024.11.023. Epub 2024 Nov 26.
Non-nutritive sweeteners (NNS) are used to replace added sugar, yet whether NNS consumers have better or worse diet quality compared to nonconsumers is unclear.
To investigate cross-sectional associations between NNS consumption and diet quality.
Data from participants in the American Cancer Society (ACS) Cancer Prevention Study-3 cohort were used for this analysis. NNS intake was estimated using self-reported consumption of NNS-containing beverages, packets, and yogurt from a validated food frequency questionnaire. Participants were categorized into nonconsumers, <1 serving, 1 to <2 servings and ≥2 servings per day. Diet quality was assessed using the ACS diet score (2020) and Healthy Eating Index (HEI-2015). Analysis of variance was used to compare diet quality scores across NNS consumption groups, and multivariable linear regression was used to examine associations between NNS consumption and diet quality. Multivariable logistic regression was used to evaluate the odds of low diet quality across NNS consumption groups.
Data from 163,679 participants [median age 53 y (IQR 45-60), 78.9% females, mean NNS intake 1.0 ± 1.5 servings/d, mean HEI-2015 score 75.4 ± 10.2] were included. NNS consumers had lower diet quality for ACS diet score (6.8 ± 0.03 among nonconsumers compared with 6.5 ± 0.03, 6.3 ± 0.03, and 6.1± 0.03 for consumers of <1 serving, 1 to <2 servings, and ≥2 servings of NNS per day, respectively, P-trend < 0.0001) and HEI-2015 (76.3 ± 0.1 among nonconsumers compared with 76.7 ± 0.1, 75.6 ± 0.2, and 72.7± 0.2 for consumers of <1 serving, 1 to <2 servings, and ≥2 servings of NNS per day, respectively, P-trend < 0.0001). Odds of low diet quality were higher among NNS consumers and were higher with higher NNS consumption (3%, 17%, and 43% higher odds of low diet quality among those who consumed <1 serving, 1 to <2 servings and ≥2 servings of NNS per day).
NNS consumers have lower diet quality in a large cohort of adults in the United States.