Does it matter whether you take an herb as tea, tincture, or capsule?

The forms. A tea (infusion) steeps leaves or flowers in hot water. A decoction simmers tougher material such as roots, bark, or seeds. A tincture is an extract in alcohol and water; homeopathic mother tinctures, for example, are made by macerating plant material in a hydroalcoholic solvent⁶. Capsules and tablets hold powdered herb or a concentrated extract, sometimes 'standardized' to a marker compound.

What gets extracted. In a lab comparison of ten plant species, infusions generally yielded more phenolic compounds than decoctions, and a 15-minute decoction beat longer simmering; chemical richness is not proof of benefit, though⁹. A review comparing granules with traditional decoctions of Chinese herbal medicine found no significant difference in effectiveness in 55 of 56 trials, but most were low quality¹⁰.

Dose and strength. A tea's strength depends on the amount of herb, the water, and the steeping time; a capsule's dose is printed on the label, but labels can be wrong. Of 59 echinacea products bought in Denver, 10 percent had no measurable echinacea and only 52 percent matched the labeled species, and 'standardized' products were no more reliable³. German St John's wort tablets and capsules differed widely in hyperforin and hypericin, even per milligram of extract, and some varied from batch to batch¹¹.

Why form matters for results. Trials usually test a defined extract. Peppermint oil in enteric-coated capsules improved irritable bowel symptoms in a meta-analysis¹, although a later trial found no difference from placebo¹²; a 2006 review found clinical trials of peppermint tea absent². Brewed green tea supplied about 90 to 300 mg of EGCG a day, while supplements ranged up to 1,000 mg, and 800 mg a day or more as a supplement raised liver enzymes⁴. A hyperforin-rich St John's wort extract lowered digoxin levels, while tea and oil preparations did not⁵.

Safety and quality. 'Traditional' does not guarantee safe: water-based kava extracts have also been linked to liver injury⁷, and 20.7 percent of Ayurvedic products bought online contained detectable lead, mercury, or arsenic⁸. For testing, see how to tell if a supplement is pure; for medicine clashes, herb-drug interactions; and for one herb's forms in detail, peppermint.

Same herb, different preparations: St John's wort. Mueller and colleagues randomized 96 healthy volunteers to placebo or one of 10 St John's wort products for 14 days while taking digoxin, a heart medicine⁵. A hyperforin-rich extract (LI 160; hyperforin is one of St John's wort's main active compounds) lowered digoxin's 24-hour exposure (AUC) by 24.8 percent (95% CI 21.3 to 28.3) and its peak concentration by 37 percent; 4 g of powder with comparable hyperforin lowered AUC by 26.6 percent, and 2 g of powder with half the hyperforin by 17.7 percent. A hyperforin-free extract, low-dose powders, tea, juice, and oil extract did not change digoxin levels significantly. The authors concluded that the interaction tracks the dose of hyperforin, so tea and extract are not interchangeable. Wurglics's analysis of eight German products (five batches each, ten dosage units per batch) found widely differing hypericin and hyperforin content even after correcting for the amount of extract, and pronounced batch-to-batch variability in some products¹¹. A result from one product therefore cannot be assumed for another.

Tea versus supplement: green tea. The European Food Safety Authority panel estimated that EU adults get 90 to 300 mg of EGCG a day from green tea infusions (up to 866 mg for the highest consumers), whereas supplements supply 5 to 1,000 mg⁴. It judged traditionally prepared infusions safe at usual intakes, noted rare liver injury after infusions that was probably idiosyncratic, and found interventional trials showing that 800 mg a day or more of EGCG (the main catechin in green tea) as a supplement significantly raised serum transaminases, a sign of liver stress, compared with controls.

Capsule versus tea: peppermint. Khanna's meta-analysis of nine trials (726 patients) of enteric-coated peppermint oil capsules (coated to dissolve in the intestine) found global improvement of irritable bowel symptoms more likely than with placebo (relative risk 2.23; 95% CI 1.78 to 2.81; five studies, 392 patients) and better abdominal pain (2.14; 1.64 to 2.79); side effects were mild and mostly heartburn¹. A later six-week US trial gave enteric-coated peppermint oil (180 mg three times daily) or placebo to people with moderate to severe IBS: both groups improved substantially (90.8 points on the symptom score with peppermint oil, 100.3 with placebo) with no significant difference between them (P = 0.97)¹². McKay and Blumberg's 2006 review found human studies of peppermint leaf limited and clinical trials of peppermint tea absent².

Decoction versus granules. Luo's systematic review covered 56 clinical trials (9,748 participants; 42 randomized), all conducted in China, that compared Chinese herbal granules with decoctions; 55 of 56 found no significant difference in effectiveness and no severe adverse events were reported, but only seven trials described adequate randomization and two allocation concealment, so the authors call the equivalence preliminary¹⁰. Fotakis's metabolomic study of ten species compared infusions and decoctions: infusions had higher phenolic content and antioxidant capacity, spearmint was highest and chamomile lowest, and 15-minute decoctions beat longer ones⁹. These are chemistry results, not clinical outcomes.

Label accuracy and contamination. Gilroy bought every single-herb echinacea product on sale in the Denver area in August 2000 (59 products): 6 (10 percent) had no measurable echinacea, the species matched the label in 31 (52 percent), and only 9 of 21 'standardized' products (43 percent) met their label standard; labeled milligrams correlated only weakly with measured constituents (r = 0.35)³. Saper randomly selected 230 Ayurvedic medicines from 673 sold via the internet; of 193 analyzed, 20.7 percent contained detectable lead, mercury, or arsenic, with similar rates in US- and Indian-made products (21.7 and 19.5 percent), and every one exceeded at least one acceptable daily intake standard⁸. Both are single-time snapshots of particular product categories.

Solvents and safety. Teschke compared 14 patients with liver injury: nine linked to ethanol or acetone kava extracts and five to traditional water extracts or kava-herb mixtures. The clinical picture was similar, which suggests the plant material rather than the solvent⁷. Alcohol in tinctures matters on its own terms: in Boatto's experiment 9 of 30 volunteers had positive breath-alcohol readings (0.11 to 0.82 g/L) within a minute of drinking a homeopathic mother tincture, and none did 15 minutes later⁶.

Bottom line. Confidence is 3 of 5. The examples above show that preparation can change the dose, the chemistry, and the interaction risk of the same herb, and that product quality varies in every form. What is missing is a systematic comparison of tea, tincture, and capsule across herbs, so check which form the evidence for your herb actually used.