What traditional use establishes, and what it does not

The argument that a preparation has been used for thousands of years is the most common defence of traditional medicine and the most common target for dismissal.

Both responses are too quick. Traditional use is evidence about some things and not about others.

What it does establish, reasonably

Absence of gross acute toxicity at traditional doses in traditional preparations. If a plant, prepared in a particular way and taken in particular quantities, had killed people reliably and quickly, that would have been noticed and the practice would have stopped.

This is genuine evidence and it is worth something. It is why the regulatory schemes for traditional herbal products in several jurisdictions accept evidence of traditional use in place of clinical trials for safety, subject to manufacturing standards.

Note the qualifiers: gross, acute, at traditional doses, in traditional preparations. Each matters.

Acceptability and tolerability. A preparation people have continued to take is one they can tolerate.

Cultural and practical knowledge. Which plants, which parts, how prepared, in what combinations, at what times. This body of knowledge represents accumulated observation and it is genuinely valuable, including as a starting point for investigation.

What it does not establish

Efficacy. This is the central point. Long use is compatible with a treatment doing nothing.

Bloodletting was practised across many cultures for around two thousand years, for a wide range of conditions, by highly trained practitioners. It was ineffective for nearly all of them and harmful for many.

Its longevity was sustained by the same mechanisms that sustain any ineffective treatment: most illnesses resolve on their own, symptoms fluctuate and improvement after treatment is attributed to it, expectation affects reported symptoms, and practitioners remember successes.

These mechanisms operate everywhere, in every tradition, including modern medicine — which is precisely why controlled trials were developed.

Chronic and delayed harms. Traditional observation is poor at detecting harms that appear years later, that are statistical rather than obvious, or that affect a small proportion of users.

Aristolochic acid is the clearest example. Plants containing it were used traditionally in several systems, and they cause kidney failure and urothelial cancer, with a delay of years. The connection was identified only through modern epidemiology, after cases appeared in a European clinic.

Heavy metal exposure, discussed elsewhere on this site, is similarly invisible to traditional observation.

Safety of modern preparations. A concentrated extract at a standardised dose in a capsule is not the traditional preparation. Traditional use of a decoction of leaves does not establish the safety of a high-dose extract of a specific compound.

Safety in modern populations. Traditional use occurred in populations not taking prescribed medication, and the interaction concerns discussed elsewhere did not exist.

The mechanisms that sustain ineffective treatment

Worth being explicit, because understanding them explains how long traditions can persist without efficacy.

Natural history. Most acute conditions resolve. Treatment given during the illness receives credit for the recovery.

Regression to the mean. People seek treatment when symptoms are worst. Symptoms fluctuate, so the next measurement tends to be better, treatment or not.

Expectation effects, which are substantial for subjective outcomes and which do not require deception.

Confirmation bias, in practitioners and patients alike.

Attribution of failure elsewhere — the patient came too late, did not follow instructions, or the case was too severe.

These are not accusations against any tradition. They are general features of human reasoning about causation from uncontrolled observation, and they are the reason controlled comparison exists.

The productive position

Traditional use is a good reason to investigate something. It is a hypothesis generator of demonstrated value — a substantial proportion of pharmaceuticals derive from plants identified through traditional use.

Aspirin, quinine, artemisinin, digoxin, morphine, and several chemotherapy agents all trace to traditional knowledge. Artemisinin in particular was identified from an ancient Chinese text and its development was recognised with a Nobel Prize.

What made these useful was the investigation: identifying the compound, establishing the dose, testing efficacy, characterising safety.

The tradition supplied the lead. The testing supplied the medicine.

What follows practically

Treat traditional use as reassuring about gross acute safety at traditional doses, and as a reason for interest rather than as evidence of benefit.

Be more cautious with concentrated modern preparations than the traditional history warrants.

And be sceptical of the argument in both directions — those who say long use proves efficacy, and those who say a tradition can be dismissed without examining what it contains.

The second position has cost medicine several useful drugs, and the first has cost people rather more.

Traditional use does not establish safety of modern concentrated preparations or of use alongside prescribed medication. Discuss any supplement with your doctor.