Sugary drinks and ADHD symptoms: what a new review found

A September 16, 2026 meta-analysis published in Frontiers in Nutrition pooled data on 461,109 children and adolescents to ask whether sugar-sweetened beverage consumption is associated with attention-deficit/hyperactivity disorder (ADHD) symptoms. The researchers identified a statistical association (odds ratio = 1.23, 95% CI: 1.02 to 1.48), but stated that the observational data cannot show causation. Because of possible publication bias, a prediction interval that crosses the null, and the possibility of reverse causation, the findings do not establish that sugary drinks cause ADHD.

Key takeaways:

• On September 16, 2026, researchers published a meta-analysis in Frontiers in Nutrition pooling nine observational studies and 461,109 youths, finding an odds ratio of 1.23 between higher sugary drink intake and ADHD or related symptoms.
• The study authors cautioned against causal interpretations because the prediction interval crossed the null (0.79 to 1.90), evidence pointed to possible publication bias, and reverse causation cannot be ruled out.
• Seven of the nine analyzed studies were cross-sectional, and only one was conducted in the United States, meaning findings cannot be broadly generalized.
• On US labels, the Added Sugars line on the Nutrition Facts panel shows how much sugar a drink adds; ingredient lists name sweeteners such as high fructose corn syrup.

What happened

On September 16, 2026, researchers Shi B, Ma MJ, Yang M, and Li YQ from East China Normal University and Anhui University of Chinese Medicine published a systematic review and meta-analysis in Frontiers in Nutrition. The work investigated the relationship between sugar-sweetened beverage (SSB) intake and ADHD or related behavioral symptoms in children and adolescents aged 1.5 to 16 years.

The review protocol was registered under PROSPERO CRD420251186577, and the researchers reported that financial support was not received for the work, declaring no conflicts of interest. The authors searched databases from their inception to January 2026 for observational studies.

Across the nine identified studies, which included a total of 461,109 participants published between 2006 and 2024, the primary random-effects meta-analysis calculated a pooled odds ratio of 1.23 with a 95% confidence interval of 1.02 to 1.48. In plain statistical terms, children classified in higher consumption categories showed higher odds of exhibiting ADHD symptoms or receiving a diagnosis compared to those in lower consumption categories.

What changed

The review pools existing observational studies; it adds no new data on children. The analysis used a random-effects model with Hartung-Knapp confidence intervals and reported these subgroup results:

• Hyperactivity symptom scales (6 studies): Odds ratio of 1.11 (95% CI: 1.03 to 1.21).
• ADHD-specific clinical assessment or diagnosis (3 studies): Odds ratio of 1.67 (95% CI: 0.49 to 5.72).
• Soft drinks specifically (2 studies): Odds ratio of 1.71 (95% CI: 1.09 to 2.69).
• Aggregated sugar-sweetened beverages (7 studies): Odds ratio of 1.12 (95% CI: 1.02 to 1.24).

The authors also report reasons for caution. Heterogeneity across the studies was moderate (I2 = 63.3%), and the prediction interval crossed the null value (0.79 to 1.90), which they say indicates "uncertainty in effect estimates for future studies."

Furthermore, when the researchers performed a sensitivity analysis excluding a single large cross-sectional study from South Korea (Kim et al., covering 16,831 youths), the pooled estimate was attenuated to an odds ratio of 1.16 (95% CI: 1.00 to 1.34). A Doi plot analysis revealed marked asymmetry, recording an LFK index of 3.37, which points toward possible small-study effects or publication bias across the published literature.

Who is affected

The populations evaluated in the paper span children and adolescents from 18 months up to 16 years of age. However, the geographic distribution of the included datasets is heavily skewed toward East Asia. Of the nine studies, four originated in mainland China, two in South Korea, one in Taiwan, one in Norway, and only one in the United States.

Study designs also impose tight interpretive boundaries. The meta-analysis drew on seven cross-sectional studies, one case-control study and one cohort study. Cross-sectional studies measure exposure and health status at the exact same moment, making it impossible to determine which factor came first.

The single included cohort study, conducted by Kim et al. in South Korea, followed 365,236 children and compared those drinking at least 200 mL per day of SSBs against those consuming less than 200 mL per day at 18 to 24 months of age. That specific cohort reported an odds ratio of 1.17 (95% CI: 1.08 to 1.28) for an eventual ADHD diagnosis by ICD-10 code.

Importantly, studies where artificially sweetened beverages (diet drinks) served as the primary exposure were excluded from the analysis entirely. So the findings say nothing about diet sodas or other drinks made with non-nutritive sweeteners.

What it means for shoppers

For parents, this meta-analysis does not show that soft drinks cause ADHD, and it does not test whether cutting soda changes symptoms. The authors write that the findings "should be interpreted with caution" and "should not be interpreted as evidence of a definitive causal relationship."

One reason is reverse causation, which the authors say "cannot be excluded": they note that children with ADHD show stronger preferences for immediate rewards, so ADHD symptoms could lead to drinking more sugary drinks rather than the other way round.

What a label can tell you is how much sugar a drink adds. Many sodas list a caloric sweetener near the top of the ingredient list:

• High fructose corn syrup (listed second on products such as standard US formulations of Sprite and Fanta Orange, per Open Food Facts records).
• Cane sugar or sucrose.
• Corn syrup solids or dextrose.

Beyond ingredient names, consumers can look at the mandatory Added Sugars line on US labels, which displays added sugar amounts in both grams and percent Daily Value. Some soft drinks also contain caffeine, which this review could not separate from sugar; see Caffeine on Food Labels: How Much Is Too Much? for how caffeine is labelled.

The background

Sugar and children's behaviour have been studied before. The authors cite a 2020 meta-analysis that found dietary sugar alone was not stably associated with ADHD.

Existing sugar guidance is not about ADHD. As set out in the WHO Healthy Diet Factsheet and the official WHO Sugars Intake Guideline, the World Health Organization recommends limiting the intake of free sugars to less than 10% of total daily energy intake, and says limiting further to 5% or less "may provide additional health benefits."

The Frontiers in Nutrition meta-analysis (full text on PubMed Central) is observational and does not change that guidance.

The gap between an observed association and a proven cause comes up elsewhere in nutrition research too, as in Ultra-Processed Food Trials: Weight Gain Clear, Cause Less So.

What to watch next

Several questions remain open:

• Dose: the meta-analysis gives no dose-response estimate per serving or per mL; exposure cut-offs differed across studies.
• Caffeine: the soft-drink subgroup rests on two studies (OR = 1.71, 95% CI: 1.09 to 2.69), and the review does not say whether caffeine in some soft drinks, rather than sugar, drives that signal.
• Mechanism: the authors describe hypotheses only, and no randomized trials were included.
• US relevance: eight of the nine studies came from outside the United States.

Evidence limitations: seven of the nine included studies were cross-sectional, the Doi plot suggested possible publication bias, and the evidence cannot establish dose thresholds or causal mechanisms linking sugar-sweetened drinks and ADHD. This article is not medical advice.