Finding Reliable Information About Chinese Medicine

The problem

Chinese medicine information comes from many places: textbooks, clinics, shops, social media, government agencies and research journals. They do not all answer the same question. A traditional text tells you how a tradition explains something. A clinical trial tells you what happened to a group of people compared with another group. This page helps you tell them apart. It is information only.

Start with sources that have no product to sell

MedlinePlus advises asking who runs a site, what its purpose is (to inform, sell, or promote opinions), and who pays for it. [1] If a business pays, the information may favor its products. [1] It says good sources generally include federal government agencies, medical schools and large professional or nonprofit organizations. [1] NCCIH asks similar questions: who runs it, do the claims seem too good to be true, is it up to date, where does the information come from, and is it selling something. [2]

Starting points we used on this site:

• NCCIH (US National Institutes of Health), which summarizes TCM approaches and safety. [3]


• NHS for plain safety advice on herbal medicines and complementary therapies. [4]


• FDA for warnings on fraudulent and tainted products. [5]


• Cochrane reviews and PubMed for trials and reviews. NCCIH points to both. [3] See Cochrane reviews of Chinese medicine.

What NCCIH says about TCM evidence

NCCIH says some approaches used in TCM, such as acupuncture and tai chi, may help improve quality of life and certain pain conditions, while studies of Chinese herbal products for a range of conditions have had mixed results. [3] Because many herbal studies were of poor quality, no firm conclusions can be made about their effectiveness. [3] That is a more useful starting point than "it works" or "it is all placebo". Your next step is does TCM work?.

Questions for any study or news story

• What type of study? NCCIH says the strongest evidence consists of results from several studies by different investigators, and rarely does one study give a final answer. [2]
• How many people, and who? The number of participants affects the results and what they mean. [2]
• How old is it? Look at the date to see how it fits into the broader evidence. [2]
• People or animals? MedlinePlus suggests asking whether research involved people or animals, how many people, how long, what type of study and who paid. [1]
• Compared with what? See how to read a TCM study and placebo and sham controls.

Claims to question

• "Ancient" or "natural" as proof. NCCIH says there is no official definition of "natural" in medicine and it does not always mean safe. [2]
• Dramatic writing, promises of cures and claims that sound too good to be true. [1]
• One lab or animal result presented as proof for people.
• A traditional explanation presented as biology. See traditional frameworks vs biomedical explanations.

For more on spotting problem products see red flags in TCM practice.

Check before you rely on it

MedlinePlus advises discussing what you find with your health care provider before relying on it, since they can tell you whether it is good for you. [1] Also check a second reliable site for similar information. [1]