Supplements · Sweeteners
non-nutritive sweeteners increases diabetes risk
In plain terms: Do artificial sweeteners raise your risk of type 2 diabetes?
Part of: • non-nutritive sweeteners
Genuinely unsettled. Observational studies often find diet-drink users get more diabetes — but that's badly tangled with reverse causation: people already at risk are the ones who switch to diet drinks. Controlled trials don't show sweeteners causing diabetes. The honest answer is that the studies disagree for a reason.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
How the studies fall
What the evidence shows
Artificially-sweetened-beverage consumption is consistently associated with higher type-2-diabetes incidence across large cohorts (and the NutriNet-Santé cohort) — so as an association the claim leans supported.
The evidence (9)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Drouin-Chartier et al. 2019 · Diabetes Care | observational | supports | moderate | Three large US cohorts: increases in ASB consumption associated with modestly higher subsequent T2D risk (attenuated by BMI). |
| Payen de la Garanderie M 2025 · PLoS Med | observational | mixed | moderate | NutriNet-Santé food-additive mixtures (incl. a sweetener/emulsifier cluster) associated with higher T2D incidence. |
| Qin et al. 2020 · Eur J Epidemiol | meta-analysis | supports | moderate | Dose-response MA: ASB associated with higher T2D (and obesity, hypertension). |
| Meng 2021 · Nutrients | meta-analysis | supports | low | Dose-response MA of prospective cohorts: 'the pooled RR for T2D for each additional ASB serving per day was 1.13 (95%CI: 1.03–1.25)' (7 studies, I2 = 78.7%, non-linear); highest-vs-lowest 'ASBs intake contributed to an 18% higher risk of T2D in these cohorts (RR: 1.18; 95% CI: 1.08–1.29)'. Quality low: publication bias not assessable (<10 groups), no adiposity or reverse-causation handling, and the authors conclude that for ASBs 'the quality of the evidence is limited by potential bias and heterogeneity due to the design of the study'. |
| Debras C 2023 · Diabetes Care | observational | supports | high | NutriNet-Santé: total artificial sweeteners (esp. aspartame, acesulfame-K) associated with higher type-2-diabetes risk. |
| Greenwood et al. 2014 · Br J Nutr | meta-analysis | mixed | moderate | MA: ASB–T2D association less consistent than SSB and attenuated by BMI adjustment — possible confounding/reverse causation. |
| Imamura F, O'Connor L, Ye Z, Mursu J, Hayashino Y, Bhupathiraju SN, Forouhi NG 2015 · BMJ 2015;351:h3576 | meta-analysis | mixed | high | Same BMJ meta-analysis, artificially sweetened beverages in 10 of the 17 cohorts. Per one serving/day: RR 1.25 (1.18 to 1.33) before and 1.08 (1.02 to 1.15) after adjustment for adiposity - 'After adjustment, the estimate of 25% greater incidence was attenuated to 8% (2.1% to 15%)' - so a positive association remains on the face of it, and information-bias calibration strengthens it to 1.29 (1.08 to 1.54). But the paper's own bias analyses dissolve it: 'Trim and fill indicated publication bias for both sugar sweetened and artificially sweetened beverages', after which the ASB estimate is 1.22 (0.98 to 1.52), P=0.07; and in the residual-confounding simulation the ASB association 'was attenuated by 96% (1.01, 0.81 to 1.25)' against only 26% for SSB. Hence the authors' verdict: 'We rated the evidence for artificially sweetened beverages and fruit juice to be of low quality' because 'Findings for artificially sweetened beverages were likely to have publication bias and residual confounding'. The operative explanation the paper gives is confounding by adiposity and health status - 'adults at high risk of type 2 diabetes preferentially consumed more artificially sweetened beverages' - with reverse causality named only as a possible additional limitation, not as a finding. Exposure is diet SOFT DRINKS, not sweeteners as such: 'No study assessed subtypes'. A positive point estimate the paper itself argues is bias: no net sign. |
| Ayoub-Charette et al. 2025 · Appl Physiol Nutr Metab | meta-analysis | contradicts | high | Bias-adjusted cohort analyses: NNS associated with LOWER obesity/diabetes — the naïve 'higher risk' reverses under substitution/bias-adjusted methods. |
| Lohner et al. 2020 · Cochrane Database Syst Rev | meta-analysis | mixed | moderate | Cochrane (NNS in diabetes): insufficient evidence to establish benefit or harm on hard outcomes. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.