Anthocyanins

Last reviewed October 2026 · Based on 13 studies

Bottom line

Anthocyanins are plant pigments found mainly in berries. Higher intake is linked to lower type 2 diabetes risk in cohort studies, and trials show a small improvement in artery flow. Blood sugar, cholesterol and blood pressure results in trials are mixed or null, so they are not clear benefits for healthy people.

What it helps with

Linked to lower heart attack risk — Mixed results, evidence grade C (Limited). Starts at B (cohort meta-analyses only); downgraded for inconsistency, since a 2019 review found no link to heart attacks, giving C.
Linked to lower type 2 diabetes risk — Benefit, evidence grade B (Moderate). Cohort meta-analyses only, large (about 200,000 people in one) and consistent (I2 about 15%), so the top grade for observational data.
Improves blood vessel flow — Benefit, evidence grade B (Moderate). Starts at A (meta-analysis of 65 RCTs); downgraded once because flow-mediated dilation is a surrogate marker, not heart events, giving B.
Lowers blood sugar — Mixed results, evidence grade C (Limited). Starts at A (meta-analyses of RCTs); downgraded for inconsistency (a 2017 review found lower glucose and HbA1c, the 2026 review of 65 trials found no change in glucose) and for surrogate marker, giving C.
Lowers blood pressure — No effect, evidence grade C (Limited). Starts at A (meta-analyses of RCTs); downgraded for inconsistency (one 12-week trial found a 24-hour systolic fall) and for surrogate marker, giving C.
Lowers LDL cholesterol — Mixed results, evidence grade C (Limited). Starts at A (meta-analyses of RCTs); downgraded for inconsistency (a 2017 review found lower LDL, the 2026 review found no lipid change) and for surrogate marker, giving C.

What it is

A group of flavonoids found mainly in berries [1]. Trial doses ranged from at least 50 mg a day, a level the 2026 review treats as achievable through diet [2], to about 302 mg a day from 26 g of freeze-dried wild blueberry powder [3].

What the research says

The largest recent review pooled 18 cohorts and 65 randomised trials. Comparing the highest with the lowest intake, the cohorts showed about 18% lower heart attack risk, 11% lower type 2 diabetes risk and 9% lower cardiovascular death risk; stroke risk did not change [2]. An earlier meta-analysis of three cohorts (about 200,000 people) found 15% lower type 2 diabetes risk [4]. A 2019 meta-analysis found lower coronary heart disease and cardiovascular death risk, but no link to heart attacks, stroke or total cardiovascular disease [5].

In the trials, chronic intake improved flow-mediated dilation, a test of how well arteries widen, by about 1.4% [2]. Insulin fell slightly, but lipids, blood pressure, glucose and HOMA-IR did not change in healthy participants [2]. A 2017 meta-analysis of 32 trials (1,491 people, healthy and with cardiometabolic disease) found lower fasting glucose and HbA1c, and lower LDL cholesterol [6]. A 2023 meta-analysis of anthocyanin-rich foods in metabolic syndrome found LDL about 8 mg/dL lower, with very low certainty [7].

Doses used in studies

The 2026 review's trials used at least 50 mg a day of anthocyanins [2]. One 12-week trial of 61 healthy older adults used 26 g of freeze-dried wild blueberry powder, about 302 mg of anthocyanins a day [3]. A meal-challenge trial of 45 people with metabolic syndrome used 26 g of freeze-dried blueberries [8].

Safety and interactions

The two blueberry trials we read, with 61 and 45 people, do not mention side effects in their abstracts [3][8]. We found no studies on pregnancy, breastfeeding, kidney disease or interactions with medicines. People who take blood thinners, diabetes medicines or blood pressure medicines should ask a doctor before taking concentrated anthocyanin supplements.

Caveats and open questions

Most disease links come from cohort studies, which show association rather than cause [2][5]. The 2017 and 2026 trial reviews disagree on blood sugar [6][2]. Several reviews rate certainty low or very low [9][7]. The 2026 review was partly funded by a blueberry council, which had no role in study selection or analysis [2].