Cinnamon and blood sugar: the trials used capsules

Pooled trials in people with type 2 diabetes show cinnamon lowering blood sugar, and they disagree with each other so badly that the pooled number describes none of them. Almost everything we know about it was measured out of a capsule rather than off a spoon. Across twenty-eight trials in 3 054 adults, fasting glucose fell by 15.26 mg/dL and HbA1c by 0.56, and the scale of the disagreement is itself measured: heterogeneity ran to 88 percent for fasting glucose and 100 percent for the figure after a meal. A separate pooling of twenty-four trials found the same direction with a different statistic. The doses were weighed. Only 2 g a day or less moved HbA1c, and the fatty liver trial used two 750 mg capsules a day. Nobody has tested the teaspoon on your porridge. One finding does cross from the capsule to the kitchen. Coumarin, the compound that makes cassia cinnamon a safety question, is absorbed from cinnamon tea and from cinnamon in rice pudding about as readily as from a capsule, so the spice gets no safety discount for arriving as food. Which jar you have matters, and the label rarely tells you. Our own USDA record is filed simply as cinnamon, ground, with no species on it at all.

The numbers, per 100 g

What your body absorbs

• Coumarin is absorbed from cinnamon about as readily as from a capsule, so the spice gets no safety discount for being food. — 24 healthy volunteers, 23 analysed (n = 24) Evidence B source
  Relative absorption was 62.8% for isolated coumarin, 56.0% for cinnamon in capsules, 66.1% for cinnamon tea and 54.7% for cinnamon in rice pudding; tea gave the fastest absorption and the highest peak plasma concentration; the authors conclude the tolerable daily intake for coumarin can be applied to coumarin exposure from cinnamon-containing meals

What it does to your health

• A separate pooling of twenty-four trials found the same direction using a different statistic. — 24 randomised controlled trials in patients with type 2 diabetes (n = 24 trials) Evidence A source
  Fasting blood sugar SMD -1.32 (95% CI -1.77 to -0.87, p < 0.001), HOMA-IR SMD -1.32 (-1.77 to -0.87, p < 0.001) and HbA1c SMD -0.67 (-1.18 to -0.15, p = 0.011); serum insulin did not change significantly (SMD -0.17, -0.34 to 0.01, p = 0.058)

Safety, limits and interactions

• As a spice cinnamon is safe, as a daily supplement it is something to keep an eye on. — 38 clinical trials, 20 case reports, 7 case series and 160 spontaneous adverse event reports Evidence C source
  Only 5 of the 38 clinical trials reported adverse events at all; the most frequent were gastrointestinal disorders and allergic reactions, self-limiting in the majority of cases

Who this works differently for

• In women with polycystic ovary syndrome cinnamon moved insulin resistance too, on five trials. — 5 randomised controlled trials in women with polycystic ovary syndrome (n = 5 trials) Evidence A source
  HOMA-IR fell by a standardised mean difference of -0.84 (95% CI -1.52 to -0.16, p = 0.010)
• Twelve weeks of 1.5 grams a day improved liver enzymes in people with fatty liver disease. — 50 patients with non-alcoholic fatty liver disease (n = 50) Evidence B source
  In the cinnamon group HOMA index, fasting blood glucose, total cholesterol, triglycerides, alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transpeptidase and high sensitivity C-reactive protein all fell significantly (P < 0.05) while HDL did not change (P = 0.122); LDL fell in both groups

What is still unknown

• Across twenty-eight trials cinnamon lowered fasting glucose by fifteen milligrams per decilitre, with heterogeneity near total. — 28 randomised controlled trials of 30 to 120 days in 3 054 adults and older adults with type 2 diabetes (n = 3054) Evidence A source
  Fasting blood glucose -15.26 mg/dL (95% CI -22.23 to -8.30, I² 88%), postprandial glucose -39.22 mg/dL (-63.90 to -14.55, I² 100%), HbA1c -0.56 (-0.99 to -0.13, I² 94%) and HOMA-IR -0.76 (-1.13 to -0.39, I² 22%); in capsule form total cholesterol, LDL and triglycerides also fell; only doses of 2 g a day or less lowered HbA1c

The bottom line

Use cinnamon because you like what it does to food. If you are thinking about it for blood sugar, three things should slow you down. The pooled effects look large on paper and the trials behind them barely agree, which is what an I squared of 88 or 100 percent means: something worked somewhere, and we cannot say for whom. More is not better here either, because only doses of 2 g a day or under lowered HbA1c. And the review of adverse events ends on a sentence worth reading twice. Cinnamon is safe as a spice and a flavouring, and medicinal use at larger doses or for longer may bring significant adverse effects and should be clinically monitored. Coumarin is the safety question people ask about cassia, and the one thing we can show you is that food does not blunt it. Relative absorption was 56 percent from cinnamon capsules, 66 percent from cinnamon tea and 55 percent from cinnamon in rice pudding, against 63 percent for isolated coumarin on its own. How much coumarin is too much, and what it harms, we cannot tell you. No source we were able to check gave us a figure, so there is no number here and we are not going to imply one. A daily cinnamon supplement is a question for whoever treats you. Every trial pooled on this page ran in people who already had type 2 diabetes, and nobody has tested the spice on anyone else. Women with polycystic ovary syndrome have five small trials in which insulin resistance fell. Five small trials is a reason to watch the field rather than to start.

Sources

1. cin-r2-03 · controlled crossover trial
   The relative extent of absorption measured as urinary excretion of the main metabolite 7-hydroxycoumarin (7OHC) was found to be 62.8% for isolated coumarin in a capsule (reference), 56.0% for cinnamon in capsules, 66.1% for cinnamon tea, and 54.7% for cinnamon in rice pudding (means, n=23, observation period 8 hours).
   Mol Nutr Food Res, 2011 · checked 2026-09-22
2. cin-r2-02 · systematic review and dose-response meta-analysis of randomised controlled trials
   Findings from 24 RCTs revealed that cinnamon supplementation had a statistically significant reduction in fasting blood sugar (SMD: -1.32; 95% CI: -1.77, -0.87, p < 0.001), Homeostatic Model Assessment for Insulin Resistance (SMD: -1.32; 95% CI: -1.77, -0.87, p < 0.001), and hemoglobin A1C (SMD: -0.67; 95% CI: -1.18, -0.15, p = 0.011) compared with the control group in patients with T2DM.
   Phytother Res, 2024 · checked 2026-09-22
3. cin-r2-04 · systematic review of adverse events
   The available data suggests that despite the safety of cinnamon use as a spice and/or flavoring agent, its use may be associated with significant adverse effects in medicinal uses with larger doses or longer duration of use and should be clinically monitored.
   Clin Nutr, 2019 · checked 2026-09-22
4. cin-r2-05 · systematic review and meta-analysis of randomised controlled trials
   After pooling data, cinnamon supplementation significantly reduced homeostatic model assessment for insulin resistance (HOMA-IR) scores in women with PCOS (SMD: -0.84, 95% CI: -1.52, -0.16, p = .010).
   J Food Biochem, 2021 · checked 2026-09-22
5. cin-r2-06 · randomised double-blind placebo-controlled trial
   In the treatment group (P < .05), significant decreases in HOMA (Homeostatic Model Assessment) index, FBS (fasting blood glucose), total cholesterol, triglyceride, ALT (alanine aminotransferase), AST (aspartate aminotransferase), GGT (gamma glutamine transpeptidase), and high-sensitivity C-reactive protein were seen, but there was no significant change in serum high-density lipoproteins levels (P = .122).
   Nutr Res, 2014 · checked 2026-09-22
6. cin-r2-01 · systematic review and meta-analysis of randomised controlled trials
   Participants consuming cinnamon showed a significant reduction in fasting blood glucose (FBG) (WMD: -15.26 mg/dL; 95% CI: -22.23 to -8.30; I2 = 88%), postprandial glucose (WMD: -39.22 mg/dL; 95% CI: -63.90 to -14.55; I2 = 100%), HbA1c (WMD: -0.56 mg/dL; 95% CI: -0.99 to -0.13; I2 = 94%), and HOMA-IR (WMD = -0.76, 95% CI: -1.13 to -0.39; I2 = 22%) compared with the control group.
   Nutr Rev, 2025 · checked 2026-09-22