Human Studies vs Lab Research on Herbal Teas
How studies in cells, animals and people differ for herbal teas, why antioxidant lab results do not prove a drink helps, and how to read each kind.
Content last updated . This page has not been medically reviewed.
Safety first. Natural products can cause side effects and interact with medicines. Read herbal tea and extract safety before you use anything on this page.
Headlines about a tea often start with a laboratory result: "rich in antioxidants" or "reduces inflammation in cells." Those results can be real, and still tell you little about what a cup of tea does in a person. This page explains the difference between lab, animal and human research, using herbal teas as examples. It is information only.
The three kinds of study
| Kind of study | What was tested | What it can tell you | What it cannot tell you |
|---|---|---|---|
| Laboratory (in vitro) | A plant extract on cells, or a chemical test in a tube | Whether a plant contains substances that act in a dish; a reason to study further [1] | Whether people absorb enough, or whether any effect happens in the body |
| Animal | An extract given to animals | Possible effects in a living body, as the hibiscus example below shows [2] | Whether the same happens in people drinking a normal cup |
| Human trial | The tea or product given to people and compared with another group | Whether a measured change happened in that group, preparation and time | Long-term effects, other groups or other preparations |
Label them preclinical (cells or animals) for the first two and keep them apart from human results. [3]
A worked example: hibiscus
The hibiscus (Hibiscus sabdariffa L.) trial describes the path itself. Its authors wrote that in vitro studies showed antioxidant properties and that extracts of the calyces had lowered blood pressure in animal models, so they tested the tea in people. [2][4] In a randomized, double-blind, placebo-controlled trial of 65 adults, the tea lowered systolic blood pressure compared with placebo over 6 weeks. [2] A later review of 26 trials of hibiscus products reported blood-pressure reductions, with a caution that long-term and therapeutic dosing safety requires monitoring. [5] The lab and animal findings were a reason to run the human trial; the human trial is what answers the question for people, and only for the form and group studied.
A worked example: rooibos
A scoping review of rooibos (Aspalathus linearis) in people noted that most of what was known about its anti-inflammatory and antioxidant properties originated from animal and cell culture studies. It found 18 human publications with 488 participants in total, and called the human evidence limited. [3]
Why "antioxidant" is not the same as "helps"
Lab antioxidant tests have become very popular over recent decades. A 2018 review in Food Chemistry said that for bioactive purposes, in vivo models are required, and described these assays as a screening step. [1] A high score in a tube is therefore a reason to study a tea, not evidence that drinking it improves health. Some human trials do measure antioxidant markers in blood, such as a chamomile tea trial in 64 adults with type 2 diabetes. Those are biomarkers, which are not the same as feeling better or avoiding illness, and the authors said a larger sample and longer follow-up may be needed. [6]
Other mismatches to watch for
- Plant versus tea. A result for a capsule, oil or extract does not carry over to tea. NIH notes most ginger studies tested dietary supplements rather than foods. [7]
- Blends. NIH notes some chamomile studies use chamomile plus other herbs, so the plant's own role is hard to know. [8]
- Natural is not proof. NIH points out that not all products from nature have been shown to be effective. [9]
- Where this library says "Proper evidence not found." We use that phrase when we searched and found no usable human study of the tea. See research by tea.
Frequently asked questions
Does a tea that is high in antioxidants improve health?
Not on that evidence alone. Lab antioxidant tests are screening steps, and in vivo models are required for bioactive claims. [1]
Are animal studies of herbal teas useful?
They can point to ideas worth testing. Animal and cell findings are preclinical and do not show the same effect in people. [3]
What is a biomarker?
A lab measure, such as HbA1c or an antioxidant enzyme level, that is not the same as how a person feels or whether they get ill. [6]
Why do tea studies sometimes use extracts?
The sources we read do not say why. NIH shows many studies tested supplements rather than foods. [7]
Who should I ask about a tea I am considering?
A pharmacist or clinician, particularly with medicines, pregnancy, liver or kidney disease, or surgery. [8]
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
Safety and who should ask first
A laboratory or animal result is never a reason to start or stop a tea, and it is no guide to interactions. If you take medicines, are pregnant or breastfeeding, have liver or kidney disease, or have surgery planned, ask a pharmacist or clinician first. NIH reports possible chamomile interactions with warfarin. [8] See how interactions work, the human studies we found, and human, animal and laboratory research.