By Grant Jacobs 04/03/2016

Below are some loose thoughts on a first reading of the consultation document for the Regulation of Natural Health Products consultation.

I don’t intend this post to be an article, as I usually would write here. It is meant to air a few thoughts I have on reading this document to invite comment. Suggestions will have to be prompt: submissions close 5pm March 4th.

I’ll largely leave aside the fallacy that ‘natural’ remedies are, somehow, a special class of product. Aside from the naturalistic fallacy, there’s that some of the things mentioned are equally ‘non-natural’ and vice versa many active ingredients in otherwise ‘pharmaceutical’ products can be found in natural settings (or a closely related compound can). For example, on page 4 of the consultation document they mention gamma-aminobutyric acid (GABA). GABA can either be extracted from samples or chemically synthesised: both are the same compound.

Before I look at the consultation document, some general thoughts. I’ve previously written suggesting some practical remedies (no pun intended…) for homeopathic products; some of the suggestions apply more widely.

An updated list:

  • Measurements on packages and containers should *only* be in standard IUPAC measures. (“Traditional” units might be presented in an accompanying brochure as part of a marketing spiel.)
  • Measures should be of the final product delivered to the customer, not an initial mixture.
  • Give an evidence level for the active ingredients.
  • Set common licensing standards.
  • Inclusion to be considered as a remedy should come from claims for remedial effect, or sales in a store or practice offering health/medicinal products.

Let’s take a stroll through the consultation document.

Low risk

“Because NHPs are generally low risk,” [NHPs = Natural Health Products; page 1]

While NHPs are usually low risk in and of themselves, two things are worth noting:

  1. They can be low risk because they are ineffective, i.e. they do little or nothing.
  2. Patients can be exposed to risk by delaying effective treatment

Both of these relate to the claims made about the product, not what ingredients are present or not or what the product “looks like”.

For example, Fevamed Relief, from it’s name is aiming to relieve fevers in infants. It is a homeopathic product. Homeopathic products properly do not have any active ingredients: they have placebo effect only. The difficulty with this product isn’t in the product itself, but the claims made.

The consultation document does go on to cover this, but it’s an example of an issue that needs attention. (False claims still fall under the Fair Trading Act 1986.)

Over-the-counter sales only

NHP regulation is intended to cover over-the-counter sales. Products made by a practitioner for a patient following a consultation are exempt from the regime – this includes rongoā Māori and traditional Chinese medicine, as traditionally practised. Products containing less than 20 parts per million of the active ingredient are exempt from the regime as are homeopathic products. [page 1]

This seems a bad mistake in my opinion.

As I related earlier the issues are very often mostly with the claims made, not the ingredients or lack or them.

Inclusion for regulatory control should be based on claiming a remedial effect, not the product, the setting, or “traditional” usage.

It’s worth noting that homeopathic products are over-the-counter sales (and so too, presumably, are many ‘traditional’ remedies).

It’s also worth noting that some products are labelled ‘homeopathic’, but, odd as it may seem, contain ingredients. I’ve seen, for example, an ‘homeopathic’ sun cream with silver oxide – listed but not stated as an active ingredient; see Footnote 2 of Homeopathy – practical remedies to address it?

Exclusion of homeopathic products also reminds me of the NZ Pharmacy Council’s attempt to set a double standard for the products they offer. It leaves me idly wondering if part of the petitioning to set the bill excluding homeopathic products has come from pharmacies, or their owners. If so, that’s awful.

Self-diagnosis [page 7]

I’m a bit wary of their inclusion of “suitable for self-diagnosis” in the allowed claims. I think it’s fair to say that anyone is the worst person to diagnose their illness.

Product notifications

Product notifications will need to include a reference to a website with a summary of the evidence supporting the health claims for a product so that consumers can easily see the basis for the claims. [page 9]

This sounds promising.

One small technical thought: will the website be captured? Websites can be edited over time, so that the website may later not read the same as when the notification was submitted.

The section [page 9] Relevance and representativeness of evidence also looks promising.

Traditional evidence [page 9]

This aspect is neither here nor there.

There is no such thing as traditional evidence, really, only evidence (or not).

There is such a thing as traditional usage – however that in and of itself is not evidence of effectiveness or otherwise.

This aspect invites a logical fallacy, the argument from antiquity. That something has been used for ‘x’ years does not make it ‘safe’ or ‘effective’. Evidence testing it might, but usage doesn’t mean much.

For the same reason, setting an arbitrary time that something falls under ‘traditional’ usage or not is not meaningful. (Indeed it could be used to apply a double standard whereby some things are “accepted” merely from traditional use – essentially by-passing all medical or scientific criteria.)

This fallacy is related to the argument from popularity fallacy, one I’ve touched on previously in Popularity does not mean effectiveness or sensibility.

They give this example on page 10:

For example, a claim might be that kawakawa was traditionally used by Māori to relieve upset stomachs. A product notifier would need to have evidence available to demonstrate that the claim is true. In this case, the claim has been taken from Te Ara: The Encyclopedia of New Zealand, published by the Ministry for Culture and Heritage, at, so a link to that page might be sufficient.

I would suggest this is not sufficient: it illustrates usage, not evidence of effectiveness, safety, etc., or not.

Related to this is that anecdotes are not evidence.

(It’s interesting to note that the consultation document shifts from evidence to use in the middle of page 10. It leaves me wondering if enough attention was paid to the difference between these two.)

Scientific evidence

The Bill establishes that health claims may be made based on scientific evidence. [page 11]

Surely this ought to read,

The Bill establishes that health claims must be made based on scientific evidence.

Other than this the Scientific evidence section looks reasonable, although I would want more detail (that I haven’t time for…)

Summary of evidence [page 11]

I find consulting documents can be obtuse through not laying out their intentions.

Am I to read that the intention is to set up two alternative ways of presenting evidence, either by scientific evidence or by “traditional evidence”?

If so, my opinion this proposed bill is fatally flawed and should be blocked in it’s present form. As I noted earlier, there really is no such thing as “traditional” evidence for effectiveness, safety, etc.

Or do they mean notifications are to include both types of evidence? (i.e. and, rather than or)

My answer to question 17 ‘Are the evidence guidelines clear?’ is clearly going to be no!

Fees [page 15]

I’m going to set aside this, with the exception of noting re exemptions for very low production, that producers that cannot pay $195.50 per annum + $667 every 5 years are unlikely to be running viable businesses in the first place.

I would suggest these at least be limited only to those products with the very lowest risk profile. An important point is that generosity in terms of exemptions should be balanced by the risk. If they’re making relatively high-risk products, they should wear the full compliance costs. This would likely be difficult to manage, I think. (The same issues would apply to manufacturing assessments.)

Labelling [page 19]

The minimum information requirements for labelling should explicitly require that the the quantity or proportion of all active ingredients be given in standard IUPAC measures.

Notification [page 22]

I would be clearer about the health benefit claim. This should include the intended purpose as well as the remedial claims being offered.

It seems to me that this statement could be used in challenging a practitioner’s use of a product, e.g. in a court setting in response to a patient being harmed.

This seems straight-forward for those selling the products: they would simply write down the purpose and remedial claims made on the container or package being sold.

Notification exemptions [page 23]

To my reading they allow,

  • that if a practitioner “makes stuff up” themselves, anything goes
  • if it’s homeopathic, let’s turn a blind eye

In the case of homeopathy, or other product with too little active ingredient, I see no reason to exclude them. Certainly they lack an active ingredient to monitor, but there is no reason that they should not be labelled as any other over-the-counter product would. (It is also complicated by the some products labelled as ‘homeopathic’ in fact have ingredients.) If no active ingredient is present, then that should be indicated to the customer, e.g. ‘Placebo effect only: this remedy contains no active ingredient’. Part of the sims of this bill is that customers be informed.

The case of practitioners who make up remedies essentially ‘on the spot’ has parallels with registered doctors writing prescriptions. There a paper trail exists. While maintaining an equivalent trail may be strictly outside the scope of this bill, does any equivalent exist for those working outside of registered medical practice? The website mechanism described for notifying products would seem to offer a way to do this, so why not use it?

The exemption for the ‘traditional practice’ setting mentioned earlier in the consultation document is not included here.

Determining if your product is a permitted natural health product [page 25]

I would have thought this flowchart ought to start with if the product claims a remedial effect, not what form it is in. Effectively it seems to say that if something “looks like” food, they don’t have to consider it. (I’m assuming that the Food Acts don’t deal with medicinal claims beyond that they meet the Fair Trading Act, which including covering false claims.)

Perhaps this is a hole that products will be driven through?

Essentially the flowchart runs off what a product looks like, rather than what it’s claimed to do.

(For those products well captured by the first box, this will work, but I can see defining what “a therapeutic product” normally “looks like” is going to be very problematic.)


The featured image is of Papaver somniferum, the opium poppy. It features on Wikipedia’s page for natural products as a source of morphine. Photo credit: Catherine Giayvia, sourced from Wikipedia (Creative Commons Attribution-Share Alike 2.0 Generic license).

Other articles on Code for life:

Homeopathy — practical remedies to address it?

Popularity does not mean effectiveness or sensibility

NZ Pharmacy Council wants lower their standards?

Medical DIY…

Undiluted humour: If Homeopathy Beats Science

Sources for medical information for non-medics and non-scientists

Bad science: baking soda, fungi, cancer, nuclear fallout, rosacea

0 Responses to “Regulation of Natural Health Products consultation”

  • Good thoughts Grant, I agree with the points you’ve made here. Particularly on the exclusion of homeopathy, I think that is untenable.

    You raise a good point about products marketed as homeopathic who contain an active ingredient that isn’t listed as the active ingredient. The NZMA’s submission ( makes a similar point:

    ” We note that products that contain fewer than 20 parts per million of the active ingredient (homeopathic products) are also exempt. We query how one can logically exclude homeopathic products or products with fewer than 20 parts per million of active ingredient when they may not be measured in the first instance.”

    My understanding of the summaries of evidence is that they must describe the scientific evidence used to back up health benefit claims, and the traditional evidence used to back up claims of use.

    The bill itself makes this startlingly unclear, and even implies that evidence of use can be used to back up claims of efficacy. The guidelines make the distinction clearer, but in the Society for Science Based Healthcare’s submission we’re asking for further clarity and an assurance that traditional evidence won’t be able to be used in a way to imply it’s evidence of efficacy.