Health evidence, carefully

Four kinds of evidence

A label that calls rooibos a natural source of antioxidants is making a statement about composition. A laboratory assay that measures activity in a test tube is a second kind of statement. A human study that measures a change in a blood marker is a third. Evidence that drinking rooibos changes what happens to people over years would be a fourth, and it is the kind that is largely missing. Industry pages tend to run the first three together and let the reader supply the fourth. This page keeps them apart.

What the human study establishes

The study most often cited is by Marnewick and colleagues, published in 2011. Forty adults at cardiovascular risk drank six cups of oxidised rooibos a day for six weeks, with comparison periods, and the researchers measured biochemical and oxidative-stress markers in their blood. The study reports changes in some of those markers. That is what it establishes. It does not establish protection against heart disease, any effect on diabetes, or a recommended dose for anyone, and a six-week marker study in forty people cannot. Its findings are a reason for further research, not a claim for a pack.

Claims that need restraint

The Memorial Sloan Kettering monograph on rooibos states that human evidence is limited. It summarises laboratory findings, notes possible interactions with some treatments, and records case reports of liver effects. A case report identifies a signal in one person; it does not establish how often something happens in ordinary consumers, and neither does its absence. The monograph also flags caution for hormone-sensitive conditions. People undergoing treatment should discuss concentrated or very frequent herbal-product use with their treating clinician rather than with a tea company.

The presence of a compound such as aspalathin, or a strong result in an antioxidant assay, does not establish that drinking the infusion protects against cancer, resolves skin complaints, settles infant colic, or is suitable for infants or in pregnancy. Tea, concentrated extracts, supplements and topical products have different exposures and different regulatory contexts, and evidence from one does not transfer to another.

Where the strong claims come from

A 2012 government agronomy brochure carries strong medical statements about rooibos, and industry pages repeat older folk uses. This library keeps them as historical context only. The industry's own typical nutritional table shows an infusion with no energy, no fat, no sugar, no fibre and a trace of sodium: in other words, a pleasant, caffeine-free drink. That is the honest product description, and it is enough.

For a label or a product page

• Prefer identity, origin, preparation, flavour and substantiated composition to any statement about effects.
• Before using a nutrition or medical claim in any market, obtain that market's legal claim requirements and an evidence review for the specific claim and the specific product form.
• Keep extract and supplement claims separate from infusion claims; the exposure is different.
• If a supplier offers you a claim, ask for the study behind it and read what the study measured.

Nothing on this page is medical advice. It is a description of what the cited evidence contains, written so that a buyer or a reader can tell a measurement from a promise.