From a spokeswoman for the TGA:

The authors of the MJA article have misrepresented the TGA’s regulatory oversight of complementary medicines, of which herbal medicines are a sub-set.

 From Byard et al:

Naturally we disagree – We feel there is need for a more open debate about the oversight of herbal medicines. If the regulatory framework can be improved then it should be. We note we are not the only ones that have been critical of the TGA’s oversight of herbal medicines (see articles by Ken Harvey and story on the TGA by the ABC’s Checkout).

TGA:

For example, in their abstract [Byard et al] say: ’The lack (our emphasis) of regulation and monitoring of traditional herbal preparations in Australia …” The article then goes on to discuss aspects of the regulation of these products in Australia. Readers who only read the abstract, which is a common practice, particularly in library searching programs, would be led to believe there is no regulation of these products in Australia.

Byard et al:

We are referring here to the “light touch” regulation and the emphasis on (the very limited) post marketing regulation which is nowhere near the levels for standard pharmaceuticals, despite the potential for these products to have material that can have significant adverse effects and the general expectations of consumers. If the regulation was effective then the ARTG products purchased in Adelaide would not have the contamination/adulteration profiles we found.

TGA:

The article (p86) states that the recent Sansom report … made a series of recommendations … permitting the listing on packaging only of ingredients that have been approved for medicinal purposes. There is no such recommendation because the law already requires every active ingredient and its quantity to be displayed on the product label.

Byard et al:

We think this is a misreading of our article. What we actually wrote was “The Sansom Report has recently made a series of recommendations relating to the regulation of complementary medicines in Australia which included continued evaluation of ingredients and the listing of only those ingredients which have been approved for use”. This was in regard to the database of approved products not packaging.

TGA:

Page 88 of the MJA article states … “In Australia, the sale of herbal medicines is profitable and their use is advertised and strongly advocated in popular media, and is not balanced by package warnings.” This is not correct.

Byard et al:

One only has to look at the promotion of products that contain St. John’s Wort and the wholly inadequate health warnings on the packaging to see that it is correct (detailed below).

TGA:

There are legal requirements for herbal medicines that contain certain ingredients to carry warning statements on the packaging, including St John’s wort which is the example cited. In our experience over many years of monitoring herbal medicines it is very rare that such warnings are not present on labels where they are required. The reference cited by the authors is to a survey of website information, not to packaging of products legally available in Australia.

Byard et al:

The website information reflects the packaging information, in some cases being identical, or slightly more informative that the packaging. We have personally compared actual labels to the website safety information to check this for Australian and overseas products. As mentioned for St. John’s Wort, this is wholly inadequate.

TGA:

Page 88 also misleads the reader by stating: “a system that relies on the honesty of the manufacturer with regard to their composition…” In fact, as well as the sponsor of the goods making a legal declaration as to the composition of their products, TGA conducts on-site inspections of all manufacturers of herbal medicines to verify information provided in the form of declarations and to ensure proper quality control procedures are in place.

Byard et al:

In a 2011 Department of Health and Ageing audit of the TGA’s handling of CAMs in Australia, it was revealed that non-compliance of HCAMs has been high for the last 5 years, and may be as high as 90% – Rowlands, D., et al., Therapeutic Goods Regulation: Complementary Medicines, Department of Health and Ageing, Editor. 2011-12, Australian National Audit Office. The salient point here is that the level of non-compliance is hig h and that there is more that could be done to enforce compliance.

TGA:

We also conduct hundreds of desk-top reviews of information each year to verify compliance with standards and the declarations made when the goods are first listed in the ARTG.

Byard et al:

Can these desk-top reviews determine if a manufacturer has changed the composition of the medicine, substituted different herbs, has undeclared products or is sourcing herbs that have substantial levels of contamination?

TGA:

The authors further mislead the reader (p 86) by stating: “With more than 1 in 20 complementary medicines failing essential (our emphasis) composition quality control in Australia…” The authors are in no position to comment on the nature of the deficiencies we have found in our post-market reviews. Most of the deficiencies we have found in this category are in fact minor and technical in nature with no consequences for public safety. Any findings of that sort would be followed up with stronger regulatory actions such as product recalls.

Byard et al:

We would sincerely hope that compositional quality control is essential. Such a level of failure in standard pharmaceuticals, even if minor or technical, would be of great concern. A perusal of the complementary medicine cancellations is instructive in this regard. If the TGA is not conducting detailed audits that speak to the composition of the products are these audits fit for purpose? The point we make is that there are tools available to screen for composition, adulterants and other chemicals – why are they not being more rigorously explored?

TGA:

On the same theme of quality control the authors again give misleading commentary. On p86 they state: “Further, the presence of Aust–L numbers on herbal medicines can be misleading, as for consumers it implies a higher level of quality control than is accurate.” Aust–L medicines are subject to stringent quality control requirements, imposed in various ways. As mentioned, the manufacturing sites are inspected by TGA auditors to international medicines quality standards.

Byard et al:

The general public do not understand the AUST-L/AUST-R system, and wrongly assume that AUST-L medicines have the same level of quality control as prescription pharmaceuticals. While the manufacturers are required to use Good Manufacturing Practice, the issues revolve around the source, composition and substitution of primary herbal materials. A manufacturer’s GMP certificate will not help if the primary herb is contaminated with Heliotropium, or if a toxic herb has been accidentally substituted. The TGA should also keep in mind the Pan Pharmaceuticals issue, where a certified manufacturer was producing herbal medicines with wildly varying constituents, resulting in adverse events.   As well, materials that are produced overseas are unlikely to have their manufacturers being the subject of regular TGA audits. Furthermore, if, as we have stated, audits do not examine actual composition of products then we question if these audits are fit for purpose.

TGA:

These types of medicines must only contain ingredients that have been pre-evaluated for safety and quality by the TGA, and may only contain levels of ingredients or their components that have been determined to be low risk.

Byard et al:

And yet it is clear that a not insignificant number of medicines do not comply with these requirements.

TGA:

Labelling standards apply so that all active ingredients must be specified on the label and any relevant cautionary statements must also be on the label.

Byard et al:

The cautionary statements are of a very general nature and usually do not speak to the specific harms. For example, the labelling on a variety of St. John’s wort products, an herbal which can produce significant and life-threatening adverse events in combination with other drugs, does not mention that it should not be taken with antidepressants, the most likely significant source of life threatening adverse events (and the most likely reason people are taking St. John’s Wort is depression, so the chance of co-medication is high), while mentioning the contraceptive pill.

TGA:

All the components of the listed medicine must be identified using Australian approved terminology, which for herbal ingredients includes the ‘complete’ name, which identifies the botanical species name, plant part and plant preparation.

Byard et al:

The issue that has been identified is that the contents do NOT match the labels and that additional ingredients are included and/or ingredients substituted without being declared.

TGA:

TGA carries out surveys of complementary medicines from time to time in which we perform our own laboratory analysis on samples are purchased from retail outlets.

Byard et al:

The TGA undertakes both random and targeted surveys of a fraction of the available complementary medicines, let alone primarily traditional herbal medicines. There is no survey report more recent than 2014 on the TGA website. Given the levels of formulation/manufacturing/quality issues identified by the TGA itself, combined with our surveys and reports of adverse events in the community, there is sufficient evidence of non-compliance to warrant more proactive approaches.

TGA:

We screen for the presence of pharmaceuticals and for heavy metals and check labels for compliance with labelling standards. We also often perform other routine quality control tests such as uniformity of weight (a measure of product consistency) and disintegration or dissolution (measure of whether or not ingredients are likely to be released in-vivo in reasonable amounts within reasonable time). In many hundreds of tests performed over the years it is rare that we have found an issue that presents a significant safety concern.

Byard et al:

The TGA itself found formulation/manufacturing/quality issues in 7% of the medicines it audited in 2014. These levels of non-compliance are concerning, as is a perusal of the complementary medicine cancelling notifications. As well, the issues are ‘rare’ using the tests currently employed. Without more detail of how the TGA are testing for composition, it is not clear that the contaminations of concern will be detected.

TGA:

The reference to use of DNA and other ‘advanced’ methods of testing gives an unbalanced view as to the utility of these methods in testing of herbal medicines because the vast majority of ARTG-listed herbal medicines are solvent extracts of plants with no DNA or other genetic information being present for identification purposes. There may be a role for DNA testing on herbal material but its use for routine quality control purposes has not been agreed by the international scientific community.

Byard et al:

The majority of ARTG products we have tested do yield DNA and as such the statement that there is “no DNA for identification” purposes is inaccurate. In particular the DNA tests can pick up issues with quality control and contamination which would not be available via other methods. We urge the TGA to engage more actively with researchers in this area. Offers have been made to work collaboratively with the TGA in this area but were declined.

TGA:

In their conclusion (p89) the authors state: “Light touch regulation of the industry means that dosages can be inconsistent, contents inaccurately listed, and toxic and pharmaceutical contaminants and adulterants may be present in herbal medicines.” This is grossly misleading with respect to herbal medicines supplied legally in Australia following listing in the ARTG. The authors largely base their opinion on the finding published by them as cited in reference 1 of the MJA paper. The one sample (but a different batch) of an ARTG-listed product reported at reference 1 to contain a pharmaceutical substance was tested by TGA and found not to contain any pharmaceutical substances.

Byard et al:

Without details of the assays we cannot comment on the TGA results. However in our studies the same pharmaceuticals in samples were detected independently by two separate laboratories. Moreover, the rates of contamination and adulteration in our studies is consistent with the TGA’s own studies on compositional quality.

TGA:

Similarly samples of Aust-L products found to have unacceptable levels of heavy metals were tested with the finding that one sample had 1.8 parts per million of mercury where the action limit is 1 ppm. Our assessment of the information at reference 1 in the MJA article is that the samples tested by those authors are not representative of ARTG listed medicines generally. We understand that many of the samples tested at reference 1 were not included in the ARTG.

Byard et al:

That these levels of contamination are found at all is a serious issue and should not be dismissed.  Note that our paper is specifically about herbal products and not ARTG listed medicines generally, which included vitamins and minerals. The fact that medicines without an ARTG listing can be purchased on the open market in Australia is another indication of failure of regulatory oversight, not a vindication of the current regulatory scheme.

TGA:

The thrust of the MJA paper, that regulation of complementary medicines in Australia is inadequate, is contrary to the finding of the recent independent review (referenced by them) which found that most submissions to the review were generally supportive of the current regulatory framework but that there were opportunities to improve consumer confidence in the efficacy of listed medicines. Importantly, there were no negative findings in relation to the current regulatory arrangements in relation to the oversight of quality matters, which is the issue raised by the MJA article.

Byard et al:

The thrust of the paper is that the regulatory framework can be substantially improved and that without improvements that the public are in the dark as to potential risks. This is in accordance with recommendation 49 of the Samson report “that the NRA develops a more comprehensive post-market monitoring scheme for listed medicinal products” and more broadly recommendations 35, 36, 39 and 40 (specifically those parts of the recommendation regarding safety and quality pages ix to xi). The reluctance of the TGA to accept that there are new, innovative and more robust ways to test herbal products pre and post market is puzzling.