![Numbers of reports of suspected adverse drug reactions associated with herbal medicines received by the UK Committee on Safety of Medicines/Medicines and Healthcare products Regulatory Agency's yellow card scheme for the period 1964 to July 25th, 2003. Blue bars (i.e. pre-1996) represent licensed herbal medicines; black bars (1996 onwards) represent reports for both licensed and unlicensed herbal medicines following extension of the scheme to unlicensed herbal medicines in October 1996; the dotted line above the 2003 bar represents an estimate of the total number of reports for the full year. Source: Adverse Drug Reactions On-line Information Tracking. [43]](https://www.researchgate.net/profile/Joanne-Barnes-2/publication/5280710/figure/fig1/AS:339691565010984@1458000188892/Numbers-of-reports-of-suspected-adverse-drug-reactions-associated-with-herbal-medicines_Q320.jpg)
Content uploaded by Joanne Barnes
Author content
All content in this area was uploaded by Joanne Barnes
Content may be subject to copyright.
Drug Safety 2003; 26 (12): 829-851
L
EADING
A
RTICLE
0114-5916/03/0012-0829/$30.00/0
Adis Data Information BV 2003. All rights reserved.
Pharmacovigilance of
Herbal Medicines
A UK Perspective
Joanne Barnes
Centre for Pharmacognosy and Phytotherapy, School of Pharmacy, University of London,
London, United Kingdom
There is an increasing awareness at several levels of the need to develop
Abstract pharmacovigilance practices for herbal medicines. The current model of pharma-
covigilance and its associated tools have been developed in relation to synthetic
drugs, and applying these methods to monitoring the safety of herbal medicines
presents unique challenges in addition to those described for conventional medi-
cines. Several problems relate to the ways in which herbal medicines are named,
perceived, sourced, and utilised. Other important challenges arise from the current
regulatory framework for herbal medicines in the UK.
In the UK at present, the Committee on Safety of Medicines/Medicines and
Healthcare products Regulatory Agency’s (CSM/MHRA) ‘yellow card’ scheme
for adverse drug reaction (ADR) reporting is the main method of monitoring the
safety of herbal medicines. Despite recent initiatives to stimulate reporting of
suspected ADRs associated with herbal medicines, such as extending the scheme
to unlicensed herbal products, and including community pharmacists as recog-
nised reporters, numbers of herbal ADR reports received by the CSM/MHRA
remain relatively low. Under-reporting, an inevitable and important limitation of
spontaneous reporting schemes, is likely to be significant for herbal medicines,
since users typically do not seek professional advice about their use of such
products, or report if they experience adverse effects. The herbal sector in the UK
has initiated various spontaneous reporting schemes, based on the yellow card
scheme, but targeted mainly at herbal-medicine practitioners. It is important that
these schemes have a link with the CSM/MHRA so that potential signals are not
missed. Several other tools used in pharmacovigilance of conventional medicines,
such as prescription-event monitoring, and the use of computerised health-record
databases, currently are of no use for evaluating the safety of herbal and other
non-prescription medicines.
Proposed European Union legislation for traditional herbal medicinal products
will require manufacturers of products registered under new national schemes to
comply with regulatory provisions on pharmacovigilance. In the longer term,
other improvements in safety monitoring of herbal medicines may include modifi-
cations to existing methodology, patient reporting and greater consideration of
830 Barnes
pharmacogenetics and pharmacogenomics in optimising the safety of herbal
medicines.
Pharmacovigilance has been defined as “the health. Against this background, this paper aims to
provide a critical overview of the current state of
study of the safety of marketed drugs under the
pharmacovigilance of herbal medicines in the UK,
practical conditions of clinical usage in large com-
and to discuss the particular challenges that this area
munities”.[1] It involves monitoring drug safety and
presents. The paper is written from a UK perspec-
identifying adverse drug reactions (ADRs) in
tive, but also has wider international relevance,
humans, assessing risks and benefits, and respond-
particularly to those countries with spontaneous re-
ing to and communicating drug safety concerns;
porting schemes and a healthcare system similar to
recently, it has been suggested that there could be
those of the UK.
more emphasis on extending knowledge of safety
rather than focusing on demonstrating harm.[2]
1. Herbal Medicines: Challenges
The above definition makes no distinction be- for Pharmacovigilance
tween pharmacovigilance of conventional and herb-
The unique characteristics of herbal medicines,
al medicines.1 Indeed, there is no need, nor is it
and the ways in which herbal medicines are named,
desirable, to separate the two; pharmacovigilance
perceived, sourced, utilised and regulated raise im-
should embrace all preparations used medicinally
portant issues and challenges for pharmacovigi-
regardless of their regulatory status, pharmaceutical
lance.
composition, cultural use and philosophical frame-
Some issues arise because most herbal medicines
work. Hence, the same aims and activities of
can be obtained without a prescription from various
pharmacovigilance apply to herbal medicines. How-
outlets, not only pharmacies. Thus, the problems
ever, pharmacovigilance activities largely have been
that apply to pharmacovigilance of conventional
focused on conventional medicines, and the current
non-prescription medicines, for example, that gener-
model of pharmacovigilance and its science and
ally their use does not involve a prescriber and is not
processes have developed in relation to synthetic
recorded or monitored through the National Health
drugs. Applying the existing model and its tools to
Service (NHS), also apply to herbal medicines. Oth-
monitoring the safety of herbal medicines presents
er problems are specific to herbal medicines and
unique challenges in addition to those described for
present difficulties over and above those described
conventional medicines, and it is important that
for conventional prescription and non-prescription
these are understood by all stakeholders.
medicines.
There is an increasing awareness at several levels
of the need to develop pharmacovigilance practices 1.1 Characteristics of Herbal Medicines
for herbal medicines; the WHO, for example, has
produced draft guidelines on this.[3] Awareness has Herbal medicinal products (also known as
arisen not only because of the extensive use of phytomedicines or phytotherapeutic preparations)
herbal medicines, but also because in recent years are "medicinal products containing as active sub-
there have been several high-profile herbal safety stances exclusively herbal drugs or herbal drug
concerns which have had an impact on the public preparations",[4] i.e. they contain as active ingredi-
1For simplicity, the term conventional medicines is used here to describe licensed medicinal products typically
comprising a single characterised chemical entity, but also includes more complex products such as vaccines. It is
recognised there are licensed herbal medicinal products and that in the UK some of these are considered conventional
medicines, for example, formulations of ispaghula husk licensed for use as bulk-forming laxatives and in hypercholes-
terolaemia, and standardised formulations of sennosides, licensed for use as stimulant laxatives.
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 831
Table I. Formal definitions and other terms used in relation to herbal medicines
Term Definition
Herbal remedy “a medicinal product consisting of a substance produced by subjecting a plant or plants to drying, crushing
or any other process, or of a mixture whose sole ingredients are two or more substances so produced, or of
a mixture whose sole ingredients are one or more substances so produced and water or some other inert
substances.”[5]
Herbal substance “all mainly whole, fragmented or cut plants, plant parts, algae, fungi, lichen in an unprocessed, usually dried
(herbal drug) form but sometimes fresh. Certain exudates that have not been subjected to a specific treatment are also
considered to be herbal substances. Herbal substances are precisely defined by the plant part used and the
botanical name according to the binomial system (genus, species, variety and author).”[6]
Herbal preparation “preparations obtained by subjecting herbal substances to treatments such as extraction, distillation,
(herbal drug preparation) expression, fractionation, purification, concentration and fermentation. These include comminuted or
powdered herbal substances, tinctures, extracts, essential oils, expressed juices and processed exudates.”[6]
Herbal medicinal product “any medicinal product, containing as active ingredients one or more herbal substances or one or more
herbal preparations, or one or more such herbal substances in combination with one or more such herbal
preparations; in addition, the product may contain vitamins or minerals or other non-biological
substances for which there is well-documented evidence for its safety; the action of the non-herbal
substances must be ancillary to that of the herbal active ingredients.”[7]a
Herbal constituent a specific chemical compound found in a herbal ingredient, e.g. hyperforin found in the aerial parts (herb) of
St John’s wort.
Herbal ingredient a specific individual medicinal plant and the plant part, present in a herbal medicine, e.g. St John’s wort
herb present in St John’s wort tablets.
a The text in bold is a recent amendment, the wording of which is still to be confirmed at a European level.
ents only crude and/or processed plants and/or plant quantitatively between different batches of herbal
starting materials because of one or more of the
parts; an isolated chemical constituent which
following factors:
originates from plant material is not a herbal medi-
cine. The term ‘herbal medicines’ is also used gener- •inter- or intraspecies variation in constituents
ally to describe both relatively crude preparations, •environmental factors, such as climate, and
such as herbal tinctures, usually supplied by herbal- growing conditions
medicine practitioners (medical herbalists), and •time of harvesting – the profile of constituents
manufactured or finished herbal medicinal products, can vary even over the course of a day
usually formulated as tablets or capsules and avail- •post-harvesting factors, such as storage condi-
able for purchase without a prescription. Legal defi- tions and drying.[8]
nitions relating to herbal medicines and explana- The method of processing crude herbal material,
tions of other terms used are given in table I. for example, the type of extraction, can also influ-
In contrast with conventional medicines, herbal ence the precise chemical composition of a herbal
medicines are chemically rich complex mixtures preparation or product. Many herbal medicinal
comprising several hundreds of constituents, often products contain several herbal ingredients, and
more. For many herbal medicines, the chemical medical herbalists usually prescribe combinations of
constituents are unknown, and even for those with herbal tinctures often supplied as a mixture, in both
well-documented phytochemistry, there are few for cases, further adding to the chemical complexity of
which the specific constituents responsible for phar- the herbal medicine taken by the patient. The chemi-
macological activity are fully understood. The pro- cal complexity of herbal medicines creates difficul-
file of constituents is not uniform throughout a plant, ties in determining their clinical pharmacokinetics,
and for many plants, only a specific plant part, or pharmacodynamics and toxicology and, equally,
parts, such as roots or leaves, is (or should be) used where a safety concern has been identified in associ-
medicinally. Moreover, the precise profile of con- ation with a particular herbal medicine, establishing
stituents is likely to vary both qualitatively and which constituent(s), even which herbal ingredi-
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
832 Barnes
ent(s) with combination herbal medicines, are impli- The names that are used for herbal medicines
present a further problem (see also section 2.1.1.,
cated is problematic.
ADR Reporting Form). Often, common or vernacu-
For the reasons given above, it is likely that there lar names are used to describe herbal medicines, but
will be variations in the chemical composition of these vary widely, and may be used to describe more
herbal medicines containing the same herbal ingre- than one species, and cannot be precise.
dient but produced by different manufacturers; this Contrary to popular belief, herbal medicines are
will apply to both licensed herbal medicinal prod- not ‘safe’ because they originate from natural
ucts and unlicensed herbal medicines. Several stud- sources; some plants are highly poisonous, and
ies have found important differences in the pharma- many others have inherently toxic constituents. For
ceutical quality of different products on the US example, metabolites of ‘unsaturated’ pyrrolizidine
market. For example, variations in the content of alkaloids, such as senecionine, are hepatotoxic in
major constituents in St John’s wort (Hypericum humans, and carcinogenic and mutagenic in ani-
perforatum) products (which in several cases also mals.[8] Senecionine is found in liferoot (Senecio
differed markedly from concentrations stated on the aureus) and in other Senecio species, such as
label),[9] and variations in and unacceptably high Senecio scandens, which has been reported as an
concentrations (in several cases >25 000 parts per ingredient in a traditional Chinese medicine product
million) of ginkgolic acids, which potentially are Qianbai Biyan Pian found in the UK.[12,13] Other
allergenic, in ginkgo products.[10] Standardisation on known intrinsically toxic groups of constituents,
content of certain constituents is an approach used their effects and examples of plant sources include
by some manufacturers to achieve more consistent aristolochic acids (nephrotoxic and carcinogenic),
pharmaceutical composition, but its usefulness is found in Aristolochia species throughout the plant,
limited at present since the specific active constitu- sesquiterpene lactones (allergenic), found in fever-
ents are known only for a few herbal medicines. few (Tanacetum parthenium) and other species in
the Asteraceae family, and furanocoumarins
Because of the variations that can exist between (phototoxic), found in angelica (Angelica
different manufacturers’ products and preparations archangelica) and other species belonging to the
of the same herbal ingredient, evidence of safety Apiaceae family.[8]
(and efficacy) should be considered in this light;
strictly speaking, evidence is product- or extract- 1.2 Utilisation of Herbal Medicines and
specific, and should be extrapolated only to those Related Issues
products or extracts which have been shown to be
pharmaceutically equivalent and bioequivalent.[11] The use of herbal medicines is a popular health-
This is largely impractical at present, given the care approach among patients and consumers in the
limited data available for herbal medicines; never- UK, although there are few reliable estimates of the
theless, the differences between different prepara- prevalence of use. Estimates of herbal-medicine use
tions of a herbal ingredient should not be ignored. In among adults in England come from a cross-section-
many cases, because of the nature of herbal medi- al survey (n = 5010; response rate: 59%) carried out
cines, a group of related constituents, rather than a in 1998 which found that 19.8% (95% CI 18.3–21.3)
single constituent, is likely to be responsible for an had purchased an over-the-counter herbal medicinal
observed adverse effect. In this case, it may be product and that 0.9% (95% CI 0.6–1.3) had con-
appropriate to group together different preparations sulted a medical herbalist in the previous year.[14]
and products containing the same herbal ingredient There are no longitudinal data for prevalence of use
or group of constituents in order to detect signals. In of herbal medicines for the UK at present, although
some ways, this is similar to investigating ‘class market research data indicate increasing sales –
effects’ with conventional medicines. sales of licensed and unlicensed herbal medicinal
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 833
products were worth £75 million in 2002, an in- tant, or if a consultation does occur, pharmacy staff
crease of 57% over the previous 5 years.[15] Studies may not have sufficient knowledge to feel confident
carried out in other developed countries, such as about providing information and advice on herbal
Australia and the US, also suggest increasing preva- medicines.[23] A small proportion of users of herbal
lence of use of herbal medicines among the general medicines seeks treatment from a herbal-medicine
adult population.[16,17] Extrapolating estimates of practitioner, but at present, there is no legal require-
herbal medicine use from such studies suggests that ment for such practitioners to have undertaken train-
large numbers of people are being exposed to herbal ing in herbal medicine or to belong to a professional
medicines; this in itself is of concern for the public organisation for herbal-medicine practitioners, and
health. while many herbal-medicine practitioners will have
taken these steps, some will not.
In the UK, herbal medicines are used by a wide
range of individuals for both acute and chronic A related issue is that some users of herbal medi-
conditions. Many herbal medicinal products are pur- cines may not disclose this use to a healthcare pro-
chased for maintenance of general health and well- fessional;[21] equally, healthcare professionals do not
being, and for use in the prevention and treatment of ask their patients routinely whether they are using
minor, common ailments. Use is not necessarily herbal medicines, even when receiving reports from
based on evidence, nor limited to symptoms and patients of suspected ADRs associated with conven-
conditions suitable for self-treatment. Herbal medic- tional medicines, and rarely record information on
inal products are also used by individuals with seri- herbal-medicine use on patient records.[24,25] It is
ous chronic diseases, including cancer, AIDS, multi- possible, therefore, that undisclosed herbal-medi-
ple sclerosis, and asthma, and many other condi- cine use could be an alternative explanation for
tions, by older patients, and pregnant or breast- reports of suspected ADRs associated with conven-
feeding women, and are administered by parents/ tional medicines.
guardians to children.[18] Similarly, medical herbal- Disclosure of herbal-medicine use to healthcare
ists use herbal medicines to treat a variety of condi- professionals is particularly important where pa-
tions.[19] Some patient groups, such as children and tients start, stop, or are already receiving treatment
older people, are at increased risk of adverse drug with conventional medicines and, equally, individu-
effects, and there is no reason why this should not als consulting medical herbalists should disclose
also apply where they use herbal medicines. Other their current use of conventional medicines, because
groups, for example, pregnant women, may use there may be a potential for drug-herb interactions.
herbal medicines in preference to conventional med- Information on the extent to which concurrent use of
icines because they are perceived to be safer, with- herbal and conventional medicines occurs is limited,
out realising that little is known about the effects of although preliminary data suggest that it may be
herbal medicines taken during pregnancy. extensive. In a cross-sectional survey of comple-
Typically, users of herbal medicinal products do mentary-therapy use among adults in the US (n =
not seek professional advice in selecting herbal 2055 respondents; 60% weighted overall response
medicines, but rather rely on friends’ or relatives’ rate), 44% were regular users of prescription medi-
recommendations, and information in the popular cines and, of these, 18.4% were using concurrently a
media.[20,21] Herbal medicines are widely available herbal or high-dose vitamin preparation.[16] In a
for purchase over the internet and from retail outlets small study conducted in the UK, 59% of herbal-
in which there is no trained healthcare professional medicine users identified in pharmacies and health-
available (see section 1.3.1).[22] Even where herbal food stores claimed that they had used herbal medi-
medicinal products are purchased from pharmacies, cines concurrently with conventional medicines,
a consumer or patient may not have any interaction mostly prescription medicines, in the previous
with a pharmacist or trained pharmacy counter assis- year.[21]
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
834 Barnes
In summary, the ways in which herbal medicines under section 12 of the Medicines Act 1968, or as
are described (named), perceived and obtained, to- unlicensed food supplements without making
gether with users’ behaviour towards herbal medi- medical claims and regulated under food, not medi-
cines and issues relating to healthcare professionals’ cines, legislation.[8] Herbal medicines meeting the
and herbal practitioners’ practice present opportuni- definition of a herbal remedy (see table I) and com-
ties for herbal medicines to be used inappropriately, pounded and supplied by ‘herbal practitioners’ on
even unsafely, and for suspected ADRs to go unde- their own recommendation currently are exempted
tected and unreported. from licensing requirements under section 12 (1).[5]
This exemption was initially intended to give ‘herb-
1.3 Regulation of Herbal Medicines al practitioners’ the flexibility to prepare remedies
for their patients, although the term is not defined
1.3.1 Current Regulatory Framework and, at present, there is no statutory regulation of
As with all medicines, the origins of regulation herbalists in the UK. A Herbal Medicine Regulatory
and pharmacovigilance for herbal medicines lie in Working Group has been established to consider
the thalidomide tragedy of the 1950s and 1960s. appropriate legislation and reform of section 12 (1)
This was the milestone which led, of course, to the and is expected to report during 2003.[26]
establishment of the Committee on Safety of Drugs Many herbal medicinal products are sold under
(now the Committee on Safety of Medicines, CSM), the section 12 (2) exemption, which exempts from
an ‘early-warning system’ (the ‘yellow card’ licensing requirements those herbal remedies con-
scheme) for doctors to report their suspicions on sisting solely of dried, crushed or comminuted (frag-
adverse effects of drugs, and legislation in the form mented) plants sold under the plant or botanical
of the Medicines Act 1968 requiring pharmaceutical name and with no written recommendations as to
companies to satisfy the competent authority (now their use.[5] In other words, such products must not
the Medicines and Healthcare products Regulatory contain any non-herbal ‘active’ ingredients, must
Agency, MHRA) of the quality, safety and efficacy not use proprietary names and must not make
of their new medicines before marketing. medical claims. Some manufacturers are unaware
There are around 600 licensed herbal medicinal of, or ignore, these conditions and illegal unlicensed
products on the UK market, although most of these herbal products can be found for sale, as currently
are not ‘new’ marketing authorisations, but are there is no requirement for manufacturers to consult
products which initially were granted product the competent authority before placing an unli-
licences of right (PLRs) as they were on the market censed herbal medicinal product on the market.
when the medicines licensing system was set up in MHRA does have the statutory power to decide
1971. When PLRs for herbal medicinal products whether a particular marketed unlicensed product
were reviewed by the competent authority, manu- satisfies the definition of a relevant ‘medicinal prod-
facturers of those intended for use in minor self- uct’ and, therefore, is subject to the usual provisions
limiting conditions were permitted to rely on biblio- of regulations relating to Medicines for Human Use,
graphic evidence to support efficacy and safety, unless it meets criteria for exemption (i.e. as pro-
rather than being required to carry out new tests and
vided in section 12 of the Medicines Act). Herbal
controlled clinical trials.[8] So, although many herbal
medicines available in the UK include some tradi-
medicinal products have product licences, the prod-
tional Chinese medicines (TCM) and Ayurvedic
ucts have not necessarily undergone the stringent
medicines (both of which often also contain non-
testing required to obtain a full marketing authorisa-
tion today, but rather have relied on evidence from herbal substances), and these are subject to the same
long-standing use. legislation as are ‘Western’ herbal medicines.[27]
There are further restrictions on certain toxic herbal
Other herbal medicines available in the UK are
ingredients, such as Aristolochia species, found in
sold either as herbal remedies exempt from licensing
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 835
some TCM products, and on other herbal ingredi- In addition, the range of possible regulatory ac-
ents that may be confused with toxic herbal ingredi- tions that the competent authority can take in res-
ents.[28] However, unlicensed herbal products con- ponse to a herbal safety concern is limited for unli-
taining these banned species continue to be censed herbal medicinal products and, for some
found.[13] responses, requires the voluntary co-operation of
herbal-medicines manufacturers. For example, after
In the UK, licensed medicinal products, includ- important interactions between St John’s wort and
ing licensed herbal medicinal products, are classi- certain prescription medicines emerged around
fied as prescription-only medicines (POMs; general- 1999/2000, MHRA took the decision that provision
ly may only be sold or supplied from a registered of warnings on St John’s wort products was an
pharmacy in accordance with a prescription given appropriate part of the regulatory response, but this
by an appropriate practitioner), pharmacy medicines required the co-operation of manufacturers of unli-
(P; may only be sold or supplied from a registered censed St John’s wort products. At the same time,
pharmacy and by or under the supervision of a marketing authorisation holders of conventional
pharmacist), and general sales list medicines (GSL; medicines believed to interact with St John’s wort
may be sold or supplied at registered pharmacies or products were obliged to make variations to product
other businesses which can be closed to exclude the information for their relevant products. Similarly,
public).[5] Most licensed herbal medicinal products when an association between use of kava-kava (Pip-
are GSL medicines. Potentially hazardous herbal er methysticum) preparations and liver toxicity was
substances (e.g. Digitalis leaf) are controlled as being investigated by the CSM, the herbal sector
POMs; certain other plants or plant parts (e.g. agreed to withdraw kava-kava products from sale.
yohimbe bark; Pausinystalia yohimbe) are control- Voluntary withdrawal worked reasonably well ini-
led as P medicines, but some of these (e.g. belladon- tially, but as the period of evaluation drew on, some
na herb; Atropa belladonna) can be sold or supplied retail outlets began selling kava-kava products
by ‘herbal practitioners’ (see earlier in this section) again. Community pharmacists, however, had a pro-
if certain conditions are met (e.g. limits on maxi- fessional and ethical responsibility not to do so.[30,31]
mum dose, maximum daily dose and/or strength). In
practice, these restricted herbal substances are rare- Other issues relevant to pharmacovigilance arise
ly, if ever, sold or supplied by pharmacists, but because manufacturers of unlicensed herbal medici-
some, such as Ephedra species, where permitted, are nal products are not required to demonstrate to
utilised by herbal practitioners in their practice. MHRA the quality, safety and efficacy of their
products before marketing. The importance of phar-
The current regulatory framework presents sev- maceutical quality for the safety (and efficacy) of
eral major problems for pharmacovigilance of herb- herbal medicinal products is well-recognised,[8,32,33]
al medicines. Whereas manufacturers of licensed but manufacturers are required only to demonstrate
herbal medicinal products are required to comply pharmaceutical quality standards for their licensed
with regulatory provisions on pharmacovigilance as herbal medicinal products. Some manufacturers of
set out in Directive 2001/83/EC,[29] manufacturers
unlicensed herbal medicinal products may have ap-
of unlicensed herbal products and those sold under
propriate quality control and quality assurance pro-
exemptions from licensing are not required to do so,
cedures for their products, but others do not, and the
i.e. for these products, which comprise the majority
pharmaceutical quality of many unlicensed herbal
of herbal medicines on the UK market, manufactur-
medicinal products is of real concern. In addition to
ers have no obligation to keep records of suspected
difficulties with assuring pharmaceutical quality due
ADRs associated with these products, nor to report
to the variation in chemical composition, quality
these suspected ADRs to the competent authority.
problems with unlicensed herbal products include
This is also the case where herbal medicines are
supplied to patients by medical herbalists. intentional or accidental substitution of species, con-
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
836 Barnes
tamination with restricted or toxic substances, in- user) associated with the use of certain herbal medi-
cines.[35]
cluding prescription medicines, and differences be-
tween labelled and actual contents.[8,13] It is essen- In addition, the efficacy of many herbal medi-
tial, therefore, when assessing reports of suspected cines has not been evaluated in randomised clinical
ADRs associated with a particular unlicensed herbal trials. Even for well-tested herbal medicines, such as
medicine to establish whether the herbal ingredi- certain extracts of St John’s wort herb which have
ent(s) implicated are what the product actually con- been assessed in around 30 randomised clinical
tains, and whether the product could be adulterated trials in depression, only a small number of clinical-
or contaminated. Ideally, a sample of the suspected trial participants has been exposed to a specific
herbal medicine should be retained for pharmaceuti- manufacturer’s product. Furthermore, there are few
cal analysis if necessary. long-term clinical trials of herbal medicines intend-
ed for long-term use. For comparison, conventional
There is a general lack of objective information medicines have been tested in around 1500 patients
on the safety of many herbal medicines. This has before they reach the market. The lack of informa-
arisen in part because under the current regulatory tion on the safety and efficacy of herbal medicines
framework there is little incentive for manufacturers makes it difficult to carry out benefit-risk assess-
to carry out pre-clinical tests and clinical trials. ments.
Postmarketing surveillance studies involving certain In summary, in the UK, the current regulatory
herbal medicinal products have been conducted by framework allows unlicensed herbal medicines,
some manufacturers (usually those based in Ger- which may be of inadequate pharmaceutical quality
many), but this is the exception. Generally speaking, and for which there is a lack of information on safety
there is a lack of information on the types and aspects, to be placed on the market and obtained by
frequency of adverse effects, including interactions consumers and patients from a range of retail outlets
with other medicines, foods, alcohol, disease and so without a prescription or other involvement of a
forth, and other aspects relevant to safety for herbal healthcare professional. Manufacturers are under no
medicines, such as their active constituents, obligation to carry out pharmacovigilance of such
pharmacokinetics, pharmacology, use in special pa- products. By contrast, conventional P or GSL medi-
tient groups (e.g. children, older people, individuals cines are permitted to be sold or supplied without a
with renal or hepatic disease, pregnant or breast- prescription because they have a history of relative
feeding women), effects of long-term use, and so on. safety.
It is often argued that herbal medicines have a long
history of traditional use and that this provides evi-
1.3.2 Proposed New Regulatory Framework
dence for their safety (and efficacy). However,
while the ‘test of time’ may have identified inherent- The need for a new regulatory framework for
ly toxic plants, it cannot, for example, identify herbal medicinal products was first discussed in the
delayed adverse effects, effects that may arise from late 1980s and, for several reasons, today it is recog-
use in patients with ‘modern’ illnesses, such as HIV/ nised widely that the existing regulatory framework
AIDS, and safety issues arising from how herbal does not adequately protect the public health. In
medicines are utilised today, for example, concur- particular, the current system does not give consum-
rently with conventional medicines.[34] Certainly, ers and patients adequate protection against poor-
there are examples of type A reactions (those that quality and unsafe unlicensed herbal medicinal
typically are dose dependent and related to the phar- products. It also discriminates against manufactur-
macological effects of the medicine) and type B ers of licensed herbal medicinal products as their
reactions (typically unrelated to dose, idiosyncratic) costs are likely to be higher because of the need to
and other types of ADRs (e.g. delayed effects in the comply with the principles of good manufacturing
user or offspring remote from medicine use in the practice and other regulatory provisions.
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 837
Table II. Key features of proposed European Union (EU) Directive 2002/0008 (amends 2001/83/EC) on traditional herbal medicinal
products[6,7]
Establishes a Committee on Herbal Medicinal Products which will be part of the EMEA and will take over the tasks of the CPMP with
regard to authorisations or registrations of herbal medicinal products by member states. Other tasks will include producing EC herbal
monographs, and establishing a ‘positive list’ of herbal substances (to include indication, route of administration, strength and so on)
allowed under the directive.
Requires EU member states to set up a specified simplified national registration procedure for traditional herbal medicinal products that
could not fulfil medicines licensing criteria.
Main features of requirements for registration include:
• products for oral, external or inhalation use only
• minor indications only (suitable for self-diagnosis and self-treatment)
• evidence that the herb has been used traditionally for at least 30 years, including at least 15 years within the EC;
period of traditional use can include the transition period
• reliable identification of raw materials and use of appropriate quality herbal ingredients, i.e. compliance with Ph Eur
standards where they exist and with manufacturer’s own specification otherwise
• systematic quality assurance and quality control throughout the manufacturing process: compliance with principles of GMP
and other relevant European guidelines; qualified person responsible for release of batches onto the market;
manufacturer’s licence or wholesale dealer’s licence where appropriate; inspection of premises
• provision of bibliographic data on safety with an expert report
• labelling, information and advertising requirements in accordance with 2001/83/EC and relevant national regulations
• compliance with pharmacovigilance requirements in accordance with 2001/83/EC
• transition period of at least 5 years, probably 7 years, once the directive comes into force
CPMP = Committee on Proprietary Medicinal Products; EC = European Community; EMEA = European Agency for the Evaluation of
Medicinal Products; GMP = good manufacturing practice; Ph Eur = European Pharmacopoeia.
Against this background, a draft European Union herbal medicinal products to provide systematic in-
formation with their products.
(EU) directive (2002/0008, which amends 2001/83/
EC) has been produced which aims to establish a The proposed directive will have an important
harmonised legislative framework for authorising impact on pharmacovigilance of herbal medicines,
the marketing of traditional herbal medicinal prod- once it comes into force. Manufacturers of tradition-
ucts.[6] The directive will require EU member states al herbal medicinal products registered under the
to set up a simplified national registration scheme UK national scheme established under the directive
will be required to comply with relevant existing
for traditional herbal medicinal products meeting
pharmaceutical legislation, including the provisions
defined criteria. The key features of the directive at on pharmacovigilance.[29] Several of these may pose
the time of writing are summarised in table II, problems for manufacturers with little or no experi-
although the directive is still under discussion and ence in this area. For example, the requirement to
amendments may yet be made. In essence, the pro- have constant access to an appropriately qualified
posed directive will require manufacturers wishing and experienced person responsible for pharma-
to obtain registrations for their traditional herbal covigilance, implementation of the use of Medical
medicinal products under a national scheme to de- Dictionary for Regulatory Activities (MedDRA)
monstrate the quality and, to some extent, the safety and connection to and compliance with EudraVigi-
of their products, whereas the usual efficacy and, to lance. There has been some discussion about the
some extent, safety requirements will be replaced by possibility of making some allowances for manufac-
evidence of traditional use. Another major change is turers with regard to the qualified person responsible
that manufacturers of products registered under the for product quality; it is not clear whether any simi-
directive will be required to comply with informa- lar allowances will be made with respect to the
tion and labelling requirements. Currently there is qualified person responsible for pharmacovigilance.
no requirement for manufacturers of unlicensed According to the current time-scale, the directive is
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
838 Barnes
expected to come into force around the end of 2004; schemes appear to function reasonably effectively
following this, there will be a transition period of at as a pharmacovigilance tool for herbal medicines in
least 5 years (the precise duration is currently under countries such as Germany where herbal medicinal
discussion but probably will be 7 years).[36] products are regulated as medicines, frequently pre-
scribed by physicians and well known to other
Other developments at the European level are
healthcare professionals, particularly pharma-
concerned, at least in part, with pharmacovigilance
cists.[38] However, spontaneous reporting is likely to
of herbal medicines. The European Agency for the
be far less effective in countries such as the UK
Evaluation of Medicinal Products’ Herbal Medici-
where herbal medicines are marketed mainly as
nal Products Working Party (previously the ad hoc
unlicensed products with no obligation for manufac-
Herbal Medicinal Products Working Group, set up
turers to report suspected ADRs to the competent
in 1997) has several pharmacovigilance issues on its
authority, and where herbal medicines are used
agenda.[37]
mostly in self-treatment without any supervision
2. Methods for Pharmacovigilance of from a healthcare professional.
Herbal Medicines
2.1.1 UK National Spontaneous Reporting Scheme
Some standard methods used in pharma- The CSM/MHRA’s national spontaneous report-
covigilance, particularly spontaneous reporting ing scheme for suspected ADR reporting by health-
schemes, are used to monitor the safety of herbal care professionals (also known as the ‘yellow card’
medicines, although these methods are less well- scheme because of the form used to report suspected
established than for conventional medicines. Other ADRs to the CSM/MHRA) has applied to licensed
methods, such as prescription-event monitoring, medicines, including licensed herbal medicines,
have not yet been applied to exploring the safety of since its inception in 1964. However, the inclusion
herbal medicines. All available pharmacovigilance of licensed herbal medicines in the scheme was not
tools have important limitations with regard to their well-publicised until October 1996, over 30 years
use in investigating the safety of herbal medicines, later, when the scheme was extended to include
in addition to those already recognised, and it is reporting for unlicensed herbal medicines.[39] This
likely that modified, even novel, methods are re- move followed a 5-year study of traditional reme-
quired. This section discusses the available meth- dies and food supplements, carried out by a UK
ods, with a focus on spontaneous reporting schemes, Medical Toxicology Unit,[40] which identified sus-
and the particular challenges that herbal medicines pected ADRs associated with these types of prod-
present for each. ucts. The extension allowed those with official re-
porter status – at the time, doctors, dentists and
2.1 Spontaneous Reporting Schemes coroners only – to submit reports for unlicensed
herbal medicines, but did not (and could not) place
The future of spontaneous reporting schemes in
any statutory obligation on manufacturers to report
pharmacovigilance has been questioned,[2] although
suspected ADRs associated with their unlicensed
it is likely that this point was raised in relation to
herbal products.
conventional medicines for which other well-estab-
lished tools, such as computerised health-record In April 1997 and November 1999, the scheme
databases, can be used for pharmacovigilance pur- underwent further extensions to allow reporting of
poses. By contrast, spontaneous reporting for herbal suspected ADRs by all hospital and community
medicines is in the early stages of its development pharmacists, respectively.[41] Community pharma-
and, at present, in the absence of other tools and/or cists were encouraged by the CSM and MHRA to
resources, is the main method of generating and concentrate on areas of limited reporting by doctors,
detecting signals of potential safety concerns asso- namely licensed and unlicensed herbal products, and
ciated with herbal medicines. Spontaneous reporting other non-prescription medicines.[41] This extension
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 839
140
120
100
80
60
40
20
0
Number of reports
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
Year received
Fig. 1. Numbers of reports of suspected adverse drug reactions associated with herbal medicines received by the UK Committee on Safety
of Medicines/Medicines and Healthcare products Regulatory Agency’s yellow card scheme for the period 1964 to July 25th, 2003. Blue bars
(i.e. pre-1996) represent licensed herbal medicines; black bars (1996 onwards) represent reports for both licensed and unlicensed herbal
medicines following extension of the scheme to unlicensed herbal medicines in October 1996; the dotted line above the 2003 bar represents
an estimate of the total number of reports for the full
y
ear. Source: Adverse Dru
g
Reactions On-line Information Trackin
g
.[43]
followed a 1-year pilot scheme for community phar- the yellow card scheme was extended to unlicensed
macist ADR reporting, carried out in the four CSM herbal medicines and when its inclusion of herbal
regions during 1997–98 and involving around 3200 medicines was first well publicised) to 2002 inclu-
pharmacies, which showed that community pharma- sive, 467 reports of suspected ADRs associated with
cists, compared with general practitioners (GPs), herbal medicines were received (see figure 1). Most
submitted a greater proportion of reports of suspect- frequently, these reports related to products contain-
ed ADRs associated with herbal medicines (the ing the herbal ingredients St John’s wort (Hyper-
numbers of herbal ADR reports as a proportion of icum perforatum), ginkgo (Ginkgo biloba), pepper-
the total number of reports submitted by pharmacists mint (Mentha piperita), Echinacea species, senna
and GPs were 4/96 (4.2%) and 8/1975 (0.4%), re- and valerian (Valeriana officinalis). It is not known
spectively; p < 0.001).[42] However, numbers of whether the low numbers of reports of suspected
herbal ADR reports submitted by both groups of ADRs associated with herbal medicines simply re-
reporters were very low and represented an average flect a low frequency of adverse effects with herbal
of only one and two reports per CSM region for medicines, or whether there are other explanations,
pharmacists and GPs, respectively. for example, substantial under-reporting.
Despite these initiatives to stimulate reporting of The number of herbal ADR reports received in-
suspected ADRs associated with both licensed and creased over the period 1999–2002, with a peak in
unlicensed herbal medicines, numbers of herbal the year 2000 around the time that reports emerged
ADR reports submitted to the CSM/MHRA remain of suspected interactions between St John’s wort
very low. From 1964 until the end of 1995, 832 and certain prescription medicines. In part, this
reports were received.[43] For the period 1996 (when simply reflected an increase in numbers of reports of
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
840 Barnes
suspected ADRs associated with St John’s wort – of the NHS’s 22 NHS Direct telephone call cen-
tres.[48] The impact that these new reporter groups
60% (82/140) of herbal reports for the year 2000
and initiatives will have on the reporting of suspect-
(42% for 1999 and 13% for 1998), and 40% (138/
ed ADRs associated with herbal medicines is not yet
345) of all herbal reports received during this period
known. The results of the pilot scheme for patient
related to St John’s wort, with around 40 reports in
ADR reporting via NHS Direct will be relevant for
total describing drug interactions with St John’s herbal medicines, given that many users of herbal
wort – but there was also a (small) general increase medicines select and self-treat with such prepara-
in numbers of herbal ADR reports submitted during tions without the advice or supervision of a health-
this period.[43]
care professional, and that some users may not re-
It is not yet clear whether this just reflects year- port suspected ADRs associated with herbal medi-
to-year variation, or whether it has been sustained. cines to their doctor or a pharmacist.[20] Evaluations
Typically, a year-on-year increase in the number of of pilot schemes have indicated that the complete-
herbal suspected ADR reports submitted could be ness of all reports submitted by community pharma-
expected, with further increases with the addition of cists and GPs was similar,[42] although in the pilot
new recognised reporters and initiatives aimed at scheme for ADR reporting by nurses there was some
stimulating herbal ADR reporting. However, the evidence that completeness of doctors’ reports was
inclusion of pharmacists does not yet appear to have slightly better than that of nurses.[49]
had a marked impact in this regard, since the addi-
tional reports of suspected ADRs associated with ADR Reporting Form
herbal medicines submitted during 1999–2002 were The minimum information required for a report
not submitted solely by pharmacists but also by of a suspected ADR is the same for both convention-
other recognised reporters.[43,44] For comparison, in al and herbal medicines, and a standard yellow card
1964 when the yellow card scheme began, there is used to collect data, regardless of the type of
were 1414 submitted reports of suspected ADRs preparation implicated. In the year 2000, a modified
associated with conventional medicines. This in- yellow card was introduced which included in the
creased steadily until 1977 when there was a surge section for “Other drugs” the prompt “(including
in reporting (10 921 reports for the year) around the self-medication and herbal remedies)”.[50] Apart
time practolol was associated with oculomucocuta- from this, the yellow card does not mention herbal
neous syndrome and withdrawn from the market, medicines specifically, and its current design has
and when initiatives to encourage reporting of sus- several deficiencies with regard to prompting for
pected ADRs were introduced, for example, the and collecting information on herbal medicines.
introduction of the newsletter Current Problems in The section ‘Suspected drug’ presents several
Pharmacovigilance.[45] Annual reporting rates have problems. First, the reporter is asked to provide the
since increased, with some fluctuations, currently to brand (proprietary) name of the suspected drug(s).
around 20 000 reports per year (which includes the While licensed herbal medicinal products are likely
small number of herbal ADR reports), giving a total to have brand names, unlicensed herbal products
of over 450 000 reports to the end of 2002. legally are not permitted to use them – only the
Further extensions to the reporter base for the vernacular and/or botanical name, such as St John’s
yellow card scheme occurred in October 2002 when wort or Hypericum perforatum should be used, al-
all nurses, midwives and health visitors became though this is ignored by some manufacturers. For
recognised reporters.[46] At the same time, electronic unlicensed herbal medicines it would be more ap-
reporting of suspected ADRs over the internet was propriate to request the name of the herbal ingredi-
launched in an attempt to facilitate reporting,[47] and ent(s) and the name of the manufacturer/supplier.
in April 2003, a pilot scheme was introduced to Identifying the manufacturer is particularly impor-
allow patient reporting of suspected ADRs via one tant for reasons mentioned earlier, namely because
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 841
the composition of products containing the same existing reporting card could be made so that impor-
tant details on herbal medicines can be requested.
herbal ingredient can vary both qualitatively and
quantitatively between manufacturers. Also, there Signal Detection and Assessment
may be other problems with the pharmaceutical At present, because of the relatively small num-
quality (e.g. contamination) of unlicensed herbal ber of reports of suspected ADRs associated with
products which should be considered when assess- herbal medicines held on the MHRA’s ADROIT
ing ADR reports. Ideally, the form should also in- (Adverse Drug Reaction On-line Information Track-
clude space to indicate whether a sample of the ing) database reports, signals are detected simply by
suspected product(s) is available. numbers of reports. It may be possible to obtain
To identify specifically the herbal ingredient(s) proportional reporting ratios for some suspected
implicated, the binomial botanical name (genus and ADRs associated with certain herbal medicines,
species) should be given. For example, ‘echinacea’ such as St John’s wort for which around 150 reports
is insufficient, since three different Echinacea spe- in total have been received since 1996. In this case,
cies (E. purpurea, E. pallida and E. angustifolia) are the comparison is made against the rest of the data-
used medicinally and these differ in their phyto- base, rather than only against the subset of herbal
chemical composition. In addition, the specific plant ADR reports. The assumptions made in proportional
part used should also be stated, since one or more analysis, and the importance of considering the ef-
plant parts may be used medicinally and, again, the fect of selected backgrounds, has been discussed in
phytochemical composition can vary. For example, the context of conventional medicines.[51] Since
both the root and the herb (aerial parts) of E. purpur- there are additional biases and other issues in
ea and nettle (Urtica dioica) are used medicinally. pharmacovigilance of herbal medicines, what is an
However, there is no specific request for these de- appropriate comparator requires consideration. This
tails on the yellow card. subject is, however, beyond the scope of this article.
Following confirmation of a signal relating to a
Other relevant information not specifically re- herbal safety concern, the next stages in its evalua-
quested includes the method of processing the crude tion are also difficult with respect to herbal medi-
herbal material (e.g. type of extract), since this can cines. In the UK, quantifying the risk is probably
also influence the precise chemical composition impossible as there is no reliable way of determining
and, therefore, the potential toxicity of a herbal the number of individuals exposed to the herbal
preparation,[38] the strength of the preparation (e.g. medicine of interest. Benefit-risk analysis is prob-
drug :extract ratio), and the formulation of the lematic because of the limited clinical data on safety
product (e.g. tablets, tincture). Also, many herbal and efficacy of herbal medicines, and identifying at-
medicinal products contain several herbal ingredi- risk groups is also difficult because the user profile
ents, some include non-herbal ingredients, such as for herbal medicines is poorly defined. A particular
vitamins and minerals, and herbal practitioners often problem is that a specific herbal medicine can have
prescribe several herbal tinctures together supplied numerous uses and may be taken by healthy individ-
as a mixture. With respect to these preparations, one uals for ‘general well-being’, as well as by patients
or more herbal ingredient(s) may be the suspected with chronic disease. These problems are further
agent(s), yet there is limited space on the current compounded if the variation in different prepara-
yellow card to provide this level of detail. tions of the same herbal ingredient is considered.
Although it is not desirable to introduce different The concerns regarding kava-kava (Piper
reporting forms for different types of preparations, it methysticum) and hepatotoxicity illustrate the pro-
could be argued that herbal medicines present a cess of assessing and responding to safety issues
special case and that a more specialised reporting relating to unlicensed herbal medicinal products. A
form is required. Alternatively, modifications to the signal concerning kava-kava and liver toxicity was
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
842 Barnes
first raised in 2000 following a cluster of cases kava-kava preparations by patients consulting
medical herbalists, were not known; reports in-
reported in Switzerland, and was strengthened a
volved different types of kava-kava preparations;
year or so later following further spontaneous re-
only a very low number of reports was received in
ports from Switzerland and Germany.[52] The UK
the UK; the quality and completeness of the reports
CSM undertook an initial evaluation, including cau-
was poor, and some reports were duplicated; there
sality assessment, and found that the risks of kava- are few clinical trials of kava-kava products and a
kava appeared to outweigh its benefits. No regulato- lack of clear evidence of efficacy; regulatory options
ry action was taken at that time, although the herbal in responding to the signal were limited, and alterna-
sector instigated a voluntary withdrawal of products tives, such as including warning information with
containing kava-kava while the safety concerns products, would have required the voluntary co-
were investigated further. operation of manufacturers of unlicensed kava-kava
The next stage involved further data collection products and MHRA would have had no means of
and evaluation. The CSM set up a working group to enforcement.
assess the issue and requested additional data on
benefits and risks of kava-kava from the herbal Strengths and Weaknesses
sector and regulatory authorities. When the CSM Spontaneous reporting schemes have recognised
next considered the issue in July 2002, a total of 68 advantages and limitations, and several of these may
reports originating from several countries had been be even more important with regard to herbal medi-
received, although only three originated in the cines (see table III). In particular, under-reporting is
UK.[52] The severity of the liver damage described in a well-recognised, important and inevitable limita-
the reports varied from abnormal liver function test tion of any spontaneous reporting scheme, but for
results to liver failure and death; six patients re- several reasons it may be an even greater problem
ceived liver transplants. Different preparations of for herbal medicines.
kava-kava were available (e.g. different types of Under-reporting of suspected ADRs associated
extracts) and consideration was given as to whether with herbal medicines could occur at several levels.
only certain types of kava-kava preparation might be
associated with liver toxicity. However, there ap-
peared to be no relationship between the method of
processing/type of extract, strength or dose, and the
adverse reactions. Thus, on the basis of the data
available, the CSM advised that the possible bene-
fits of preparations containing kava-kava do not
outweigh the risks, that kava-kava had the potential
to cause hepatotoxicity which could be serious in
nature, and that kava-kava should be prohibited in
unlicensed medicines. On January 13, 2003, a statu-
tory order came into effect in the UK prohibiting the
sale, supply and import of unlicensed medicines
containing kava-kava. Product licences for licensed
kava-kava products were revoked.[53]
Some of the difficulties in assessing safety con-
cerns with unlicensed herbal medicines were evi-
dent here. For example, the number of unlicensed
herbal products containing kava-kava available in
the UK, their extent of use, and the extent of use of
Table III. Summary of advantages and limitations of spontaneous
adverse drug reaction (ADR) reporting schemes with respect to
herbal medicines
Advantages
Monitor all drugs, including all herbal medicines, all the time and
for all consumers and patients
Provide early warnings of undocumented drug safety concerns;
important for herbal medicines as information on safety is limited
Relatively cheap to run; important as the herbal sector may not
have the resources to conduct large-scale postmarketing
surveillance studies
Limitations
Under-reporting; likely to be greater for herbal medicines
Poor quality of data available to or provided by reporter; yellow
card does not cater specifically for recording information on
herbal medicines as suspected drugs
Biases in reporting
Cannot estimate frequency of an ADR as do not provide accurate
information on number of individuals exposed to the drug of
interest; probably not possible to obtain denominators for
unlicensed herbal medicinal products
Suspected ADRs may be identified/reported outside the formal
system (e.g. to herbalists, health-food stores)
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 843
First, because of the perception that herbal medi- cines information pharmacists working in a Medi-
cines Information Centre in Wales found that al-
cines are ‘safe’, users of these preparations may not
though they encouraged only 41% of enquirers
associate an adverse event with their use of a herbal
about ADRs to complete yellow cards, they were
medicine, particularly if they are taking other (con-
more likely to give encouragement where an ‘alter-
ventional) medicines. If the user does make an asso-
native’ medicine was involved rather than a conven-
ciation between use of a herbal medicine and an
tional medicine.[57] In addition, all these studies re-
adverse event, they may take steps to resolve the
vealed deficiencies in community pharmacists’
problem themselves (for example, stop taking the
knowledge on other aspects of ADR reporting, such
preparation) and/or may not inform a healthcare
as the level of certainty required regarding a causal
professional.[20] Under-reporting can also occur at
relationship.
the level of the healthcare professional, since doc-
tors, pharmacists and other recognised reporters To date, there are very few studies that provide
could filter out reports of suspected ADRs described any information on the extent of under-reporting of
by patients.[54] Reasons for under-reporting among suspected ADRs associated with herbal medicines.
healthcare professionals are well-documented, al- In one cross-sectional survey of community phar-
though studies exploring this area have been carried macists who were not involved in the CSM/MHRA
out in the context of conventional medicines, and it pilot scheme for community pharmacist ADR re-
is not known if these same reasons apply to under- porting (see earlier in this section), respondents
reporting for herbal medicines. were asked to describe any reports of suspected
ADRs associated with complementary medicines
Several studies involving community pharma- that they had received or identified over the previous
cists indicate that many pharmacists are unaware 12 months.[24,25] In total, among 818 respondents, 44
that they should report suspected ADRs associated reports of suspected ADRs associated with herbal
with herbal medicines. A cross-sectional survey car- medicines were described, an average of one report
ried out in 1998 of over 1300 community pharma- per 19 pharmacists. By contrast, the CSM/MHRA
cists (response rate: 67%) not involved in the CSM/ pilot scheme, which ran over approximately the
MHRA pilot scheme for community pharmacist same period covered by the survey, and involved
ADR reporting found that 47% of respondents were around 3200 pharmacies, received only four re-
not aware that the yellow card scheme applied to ports.[42,58] Conclusions cannot be drawn from these
herbal medicines at all, 37% were aware it applied to crude comparisons, since these studies used differ-
licensed herbal medicines, and only 16% knew it ent methodologies, involved pharmacists/pharma-
applied to both licensed and unlicensed herbal medi- cies in different regions of the UK, and so on. They
cines.[25] This finding is not so surprising since these do, however, raise the hypothesis that there is signif-
pharmacists were not recognised reporters at the icant under-reporting by pharmacists of suspected
time of the study and would not have received ADRs associated with herbal medicines.
training materials on ADR reporting. Of more con-
cern is that studies conducted since all community It is recognised that pharmacists can make an
pharmacists became recognised reporters and were important contribution to ADR reporting for herbal
encouraged to focus on reporting suspected ADRs medicinal products, but it is likely that greater vigi-
associated with herbal and other non-prescription lance on the part of the pharmacist and initiatives to
medicines have continued to find that many com- encourage herbal ADR reporting by pharmacists are
munity pharmacists are unaware of the need to re- required. Against this background, there have been
port suspected ADRs associated with herbal medi- several recent papers in a journal received by all UK
cines,[55,56] particularly unlicensed herbal medi- pharmacists,[18,44,59] and a fact-sheet on ADR report-
cines.[55] There may also be biases favouring ADR ing by pharmacists has been produced by the
reporting for herbal medicines. An audit of medi- Science Committee of the Royal Pharmaceutical
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
844 Barnes
2.1.3 Herbal-Sector-Initiated Spontaneous
Society of Great Britain (the professional and regu- Reporting Schemes
latory body for all pharmacists in the UK) which
provides guidance and reminds pharmacists of their At present, herbal medicine practitioners are not
recognised as reporters by the CSM/MHRA yellow
professional and ethical responsibilities in this re-
card scheme. Several herbal-medicine practitioner
gard.[60]
and other herbal-sector organisations have initiated
A further limitation of spontaneous reporting their own ADR reporting schemes for herbal medi-
schemes is that herbal medicines are widely avail- cines based on the CSM/MHRA scheme. While this
able from a range of outlets without the need for is a responsible and potentially useful step forward
interaction with a healthcare professional and, there- where these schemes have developed a link with the
fore, suspected ADRs associated with herbal medi- CSM/MHRA or WHO/UMC, ad hoc schemes are
cines may be identified by or reported to an individ- not encouraged because there is a risk that reports
ual (e.g. herbalist) who is outside the formal system will be dispersed and signals may be not be detected
for ADR reporting. Health-food stores are a major as early as possible, or may be missed. As with any
outlet for herbal medicinal products, but it is not spontaneous reporting scheme, herbal-sector-initiat-
known if staff in these outlets receive reports of ed schemes are also likely to be prone to limitations
suspected ADRs associated with such products, and such as under reporting. It is not known whether
if they do, what action, if any, they take. reasons for under-reporting of suspected herbal
ADRs by the herbal sector are different to those for
herbal ADR reporting by conventional healthcare
2.1.2 WHO/Uppsala Monitoring Centre Traditional
Medicines Project professionals. It is possible that there may be con-
cerns among the herbal sector that the availability of
ADR reports, including herbal ADR reports, herbal medicines and their freedom to practise herb-
from the CSM/MHRA yellow card scheme and (in al medicine may be threatened if significant num-
2003) those from 70 other countries with national bers of herbal ADR reports are submitted.
ADR monitoring schemes are fed into the WHO/
The National Institute of Medical Herbalists
Uppsala Monitoring Centre (UMC). The UMC
(NIMH), the major organisation for medical herbal-
recognises the problems inherent in ADR reporting
ists in the UK, requests from its members reports of
for herbal medicines and has established a Tradi- suspected ADRs associated with herbal treatments.
tional Medicines project to stimulate reporting in Reports are submitted on a modified ‘yellow card’
this area and to standardise information on herbal form, which has some additional data fields relevant
medicines, particularly with regard to nomencla- to herbalists’ prescriptions. The NIMH sends an
ture.[61] For example, a special set of herbal anatomi- annual summary of reports received to MHRA. For
cal-therapeutic-chemical (ATC) codes has been de- the period January 1994–November 2001, 23 re-
veloped which is fully compatible with the regular ports were received by the NIMH.[63] Most reports
ATC classification system for conventional medi- described reactions experienced by patients who had
cines.[62]
received a combination of several herbs, which is
The UMC database, established in 1968, held typical of medical herbalists’ treatment approach. A
over 2 million reports of suspected ADRs by 1999, similar scheme has been set up by the Register of
of which around 0.5% involve herbal medicines.[62] Chinese Herbal Medicine (RCHM) which also uses
For the period 1968–1997, almost 9000 reports in- a modified yellow card form to collect data from its
volving herbal medicines were received by the practitioners of Chinese herbal medicine. The
UMC. The UK is among the top five countries in RCHM scheme also has a link with MHRA. At the
terms of absolute numbers of herbal suspected ADR time of writing, nine reports (three in 2000) had
been received by the RCHM from its members.[64]
reports submitted.
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 845
Other schemes have been established which are concern if, for example, a herbalist submits a report
not restricted to herbal-medicine practitioners. of a suspected ADR associated with a herbal medi-
Phytonet is a password-protected, internet-based cine to this scheme alone.
system for gathering reports of suspected ADRs 2.1.4 Intensive Monitoring Schemes
associated with herbal medicines that was set up by Extensions to the CSM/MHRA yellow card
a UK university on behalf of the European Scientific scheme were launched in 1997 and 1998 to stimu-
Co-Operative on Phytotherapy in 1996.[65] Phytonet late reporting of suspected ADRs associated with
uses an electronic form based on the CSM/MHRA medicines used in the treatment of HIV infection,
yellow card, but differs from the schemes described and of ADRs occurring in children.[45] The HIV
above in that it accepts reports from healthcare reporting scheme, launched in 1997, targets special-
professionals, herbal practitioners, patients and the ist healthcare professionals working with HIV-in-
public. Submitted reports are assessed by an expert fected individuals and encourages reports of sus-
panel and, where appropriate, fed into the WHO/ pected ADRs on a modified yellow card form which
UMC. Few reports have been received, however, does not request the patient’s name. The paediatric
and support is needed to revive the system. As there reporting scheme also targeted healthcare profes-
is no obligation for manufacturers to report suspect- sionals but did not involve a modified yellow card.
ed ADRs associated with their unlicensed herbal Numbers of reports increased following the intro-
products, the British Herbal Medicine Association duction of these schemes, although the underlying
(BHMA), whose members include many herbal- reporting rate increased only with the HIV scheme;
medicines manufacturers, has addressed this in its ongoing promotion of the scheme, for example,
voluntary code of practice for its members.[66] The through a regular newsletter is stated to be impor-
code includes the requirement that manufacturers tant.[45] To date, CSM/MHRA have not introduced
send reports of suspected ADRs associated with an intensive monitoring scheme for herbal medi-
their unlicensed herbal products to the BHMA, cines and, given that the impact of initiatives to
which may, at its discretion, forward such reports to stimulate reporting of herbal ADRs has so far been
MHRA. At the time of writing, the BHMA had not limited, it may be an appropriate time to set up such
received from its members any reports of suspected a scheme together with a programme of ongoing
ADRs associated with unlicensed herbal medicinal promotion aimed at maintaining its effectiveness.
products. The University of Westminster, London, The herbal sector has not yet set up any such
UK, in conjunction with a supermarket retailer, is schemes for pharmacovigilance purposes, although
developing an ADR reporting scheme for consum- a pilot study run by the Medicinal Plant Research
ers, nutritional advisers and other health practition- Group collected detailed data, including outcomes,
ers to report suspected ADRs associated with herbal from around 30 herbalists on all their patients treat-
and ‘complementary’ medicines.[67] There is at least ed for irritable bowel syndrome.[68]
one other scheme in the UK, set-up by a herbal
supplier in Leicester, UK, which is inviting purchas- 2.2 Prescription Event Monitoring
ers of its products to submit reports of suspected
ADRs. It is not clear what are the aims of this The methodology of prescription event monitor-
scheme, given that it is stated that the information ing (PEM) in monitoring the safety of newly mar-
will be used only by the herbal supplier. This is of keted prescription drugs is well-established.[69]2 The
2PEM is a hypothesis-generating, non-interventional, observational form of monitoring for newly marketed medi-
cines carried out by the Drug Safety Research Unit, Southampton, UK. Current PEM methodology involves sending a
‘green form’ to GPs who have prescribed the medicine being studied; these data are obtained from the UK Prescription
Pricing Authority. The green form comprises a simple questionnaire, which requests data on all health events the patient
who was prescribed the drug experienced during treatment. These forms are usually sent to the GPs around 6 months
after the patient was first prescribed the medicine under study.
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
846 Barnes
valuable contribution that PEM has made to strengths and limitations of case-control and cohort
pharmacovigilance of conventional medicines is studies are well documented,[72] but as with other
clear, but the existing method is of no use at present study designs, some of the problems are com-
for pharmacovigilance of herbal medicines because pounded when these study designs are applied to
they are rarely prescribed. herbal medicines. For example, to establish and
verify both cases’ and controls’ exposure to the
A protocol for modified PEM methodology has
herbal medicine(s) of interest is particularly prob-
been developed by the Drug Safety Research Unit,
lematic since herbal medicines are rarely prescribed;
Southampton, UK, in collaboration with the NIMH,
even where herbal medicinal products are purchased
the University of Southampton, the Medical Toxi-
from pharmacies, pharmacists do not routinely re-
cology Unit at Guy’s and St Thomas’ Hospital
Trust, London, and the School of Pharmacy, Univer- cord use of herbal and other non-prescription medi-
sity of London. This approach involves using herb- cines on computerised patient medication
alists to provide adverse event data on green forms records.[24,25] In addition, for reasons explained ear-
for patients treated with a specific herbal medicine. lier, there are likely to be variations in different
Where patients give permission, a green form re- manufacturers’ products and, therefore, defining ex-
questing adverse event data would also be sent to posure precisely will be difficult at best.
their GP. There are limitations to this method, such Case-control and cohort studies involving con-
as whether sufficient patient numbers could be ventional prescribed medicines can be carried out
achieved and, particularly, that the herb of interest is using UK computerised health-record databases
not ‘newly marketed’ so there may be preconcep- such as the General Practice Research Database and
tions about its safety profile. Nevertheless, the pro- the Medicines Monitoring Unit database, but such
tocol represents a step forward in attempting to tools currently are of no use for studies involving
develop methods for pharmacovigilance of herbal herbal medicines since herbal medicines are rarely
medicines. Funding is being sought to carry out a prescribed and information on non-prescription
pilot study of the modified PEM methodology. medicines, including herbal medicines, is not re-
Another potential approach, based on PEM con- corded on GPs’ patient records.
cepts, is to use community pharmacists to recruit a As with case-control and cohort studies, experi-
cohort of purchasers (where consent is given) of a mental studies can be applied to investigating the
specific herbal medicinal product who would then safety of herbal medicines. At present, notwith-
be followed up over time and adverse event data standing recognised limitations, such as sample size
collected. The feasibility of this approach has been and ethical considerations, well-designed and well-
demonstrated in a pilot study using a conventional conducted randomised clinical trials (RCTs) over-
non-prescription medicine,[70,71] but needs to be come some of the difficulties that herbal medicines
evaluated as a method for pharmacovigilance of present for other pharmacoepidemiological studies.
herbal medicinal products. For example, precisely establishing exposure is sim-
pler since compliance checks can be carried out, and
2.3 Other Pharmacoepidemiological RCTs are unlikely to use herbal medicinal products
Study Designs (containing the same herbal ingredient) from differ-
ent manufacturers, so product variation and, usually,
The methodology for case-control and cohort batch-to-batch variation in products, is eliminated.
studies is well established and these study designs There is always the possibility of course that clin-
can be used to investigate safety concerns with ical-trial participants could take purchased herbal
herbal medicines, although few studies have been medicines in addition to the study medication.
carried out to date. One study explored the relation-
ship between colorectal cancer and use of prepara- Systematic reviews and meta-analyses of adverse
tions containing anthranoid laxatives.[38] The event data from RCTs of specific herbal medicines
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 847
have been carried out, but this introduces other risks of adverse effects associated with prescription
problems. Many existing RCTs of herbal medicinal medicines and conventional non-prescription medi-
products are of poor or limited methodological qual- cines,[73,74] but given that herbal medicines are wide-
ity, and/or published reports of studies do not follow ly perceived to be safe, the hypothesis that users of
Consolidated Standards of Reporting Trials (CON- herbal medicines may underestimate risks needs to
SORT) guidelines. In addition, clinical trials of a be tested. Furthermore, once the proposed directive
particular herbal ingredient usually will have been on traditional herbal medicinal products comes into
carried out using several different manufacturers’ force, manufacturers of products registered under
products, but systematic reviews and meta-analyses the new national scheme will be required to provide
often ignore variations between products. systematic information with their products, includ-
ing information on adverse events and special warn-
ings. The impact of this on users’ perceptions of the
3. Communication of Herbal risks associated with herbal medicines will also re-
Safety Concerns quire evaluation.
The action taken by MHRA to communicate
The importance of the timing, content and information on interactions between St John’s wort
method of delivery of messages regarding safety
and certain prescription medicines after this issue
concerns has been discussed extensively, and the
emerged in the year 2000 provides an example of the
requirements for successful communication of safe-
ty concerns should apply equally to herbal medi- process of communicating information on herbal
cines. However, communicating information on safety concerns. Following its decision that manu-
herbal safety concerns presents additional difficul- facturers should include warning information on
ties for several reasons. ‘Dear Doctor/Pharmacist’ product packaging, MHRA used various ways of
letters can be sent, but healthcare professionals are communicating the message. ‘Dear Doctor/Pharma-
unlikely to know which of their patients are using cist’ letters were sent, and pharmacists in particular
herbal medicines and, therefore, will be unable to were asked to provide advice to consumers and
pass on safety messages to specific individuals. patients on interactions between St John’s wort and
Medical herbalists may keep some records of their conventional medicines. A telephone help-line was
patients’ treatment, but as there is no statutory regu- set up, and information for patients was posted on
lation for herbal-medicine practitioners, lists of all the MHRA website. However, it is difficult to assess
individuals practising herbal medicine are not avail- the effectiveness of these measures. Since February
able. 2000 when the information was made public, the
Moreover, most users of herbal medicines obtain CSM/MHRA yellow card scheme has continued to
these medicines from outlets where there is no receive reports of suspected interactions between St
healthcare professional present and without seeking John’s wort and conventional medicines (>30 from
professional advice. Methods aimed at reaching the February 2000–April 2003), for example, reports of
public directly (e.g. the internet) and the popular breakthrough bleeding and unintended pregnancy in
media are often the only ways of communicating women taking St John’s wort products concurrently
herbal safety information to such individuals. There with oral contraceptives.[43]
is a lack of research on how herbal-medicine users It is likely that there is scope for improving
interpret information on risks associated with herbal communication with the public on herbal safety
medicines. It should not be assumed that users’ issues. Recognising this, an area on the MHRA
understanding of risk associated with herbal medi-
website has been set up which is dedicated to pro-
cines is the same as that for prescription medicines
viding early information on herbal safety con-
or conventional non-prescription medicines. It has
cerns.[13]
been shown that individuals may overestimate the
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
848 Barnes
4. The Future for Pharmacovigilance of Another effect of the directive may be to shift the
Herbal Medicines emphasis of research involving herbal medicines. At
present, most research in the herbal medicines area
The potential for herbal medicines to have a is aimed at discovering the pharmacological activi-
significant negative impact on the public health ties of medicinal plants and providing evidence of
needs to be kept in perspective. Nevertheless, a clinical efficacy; rather less effort is focussed on
parallel can be drawn between the lack of a formal investigating safety. However, since the proposed
medicines regulatory system before the thalidomide traditional herbal medicinal products directive does
disaster and the current situation in the UK as re- not require manufacturers to demonstrate efficacy
gards herbal medicinal products in that the sector (other than by way of traditional use), there may be
largely is unregulated. Most herbal medicinal prod- more interest among manufacturers and researchers
ucts, including herbs from China, South America in extending knowledge of the safety of herbal med-
and many other countries, which are new to the UK, icines. While research into the safety of herbal medi-
are sold without any requirement to demonstrate to cines is to be welcomed, research into efficacy is
the licensing authority evidence of quality, safety also needed in order to develop herbal medicinal
and efficacy. Post-thalidomide, new initiatives in products with favourable benefit-risk profiles.
drug safety monitoring initially followed further Statutory regulation of herbal-medicine practi-
high-profile drug safety problems.[75] Likewise, sev- tioners, as recommended by The House of Lords’
eral recent high-profile herbal safety concerns, such Select Committee on Science and Technology’s re-
as renal failure and urothelial cancer associated with port on complementary/alternative medicine[79] is
exposure to Aristolochia species,[76] drug interac- also expected to be implemented over the next few
tions with St John’s wort,[77] and hepatotoxicity years. Once this has been achieved, it seems reason-
associated with kava-kava,[78] have contributed to able to expect that the yellow card scheme would be
the increasing awareness of the need to monitor the extended to include state-registered herbal-medicine
safety of herbal medicines. Against a background of practitioners as recognised reporters who would be
increasing use of herbal medicines, particularly by encouraged to report suspected ADRs associated
patients using conventional drugs concurrently and with herbal medicines.
those with serious chronic illness, it is likely that In the longer term, modified, even novel tools for
new safety concerns will continue to emerge. monitoring the safety of herbal medicines may be
However, improvements in the safety and developed. Pharmacy-record linkage is used in The
pharmacovigilance of herbal medicines can be ex- Netherlands for pharmacovigilance purposes, but no
pected, if the proposed EU directive for traditional such tool exists currently in the UK. A new Depart-
herbal medicinal products is implemented as ment of Health report,[80] however, discusses the
planned.[6] Manufacturers of traditional herbal me- possibility of community pharmacists being able to
dicinal products registered under national schemes access a common electronic health record which
established under the directive will be required to will be created for all patients and, presumably, to
adhere to quality standards, to provide bibliographic
add community pharmacy data to it. While such a
evidence of the safety of their products, and to
system probably would apply only to prescription
comply with regulatory provisions on pharma-
medicines initially, with technological advances it
covigilance. These improvements may not happen
might also be developed into a computerised record-
immediately across all manufacturers, since some
linkage database that could be used to monitor the
may take advantage of the transition period (which
safety of herbal and other non-prescription medi-
will be at least 5 years, probably 7 years, from the
cines. Consideration should also be given as to
date the legislation comes into force) before submit-
whether consumers and patients could have a greater
ting their dossiers to MHRA in order to apply for
product registrations. role in pharmacovigilance of herbal medicines,
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 849
8. Barnes J, Anderson LA, Phillipson JD. Herbal medicines: a
possibly by their inclusion as recognised reporters in guide for healthcare professionals. London: Pharmaceutical
spontaneous reporting schemes and by collecting Press, 2002
9. De Los Reyes GC, Koda RT. Determining hyperforin and
data directly from patients in studies based on modi- hypericin content in eight brands of St John’s wort. Am J
fied PEM methodology. Health Syst Pharm 2002; 59: 545-7
The future for ensuring the safety of herbal medi- 10. Kressman S, Muller WE, Blume HH. Pharmaceutical quality of
different Ginkgo biloba brands. J Pharm Pharmacol 2002; 54:
cines may lie, at least in part, with pharmacogenetics 661-9
and pharmacogenomics. The importance of genetic 11. Loew D, Kaszkin M. Approaching the problem of bioequiva-
lence of herbal medicinal products. Phytother Res 2002; 16:
factors in determining an individual’s susceptibility 705-11
to ADRs is well documented,[81] and this applies to 12. Woodfield R. Senecio species in unlicensed herbal remedies
herbal medicines as well as to conventional drugs. [letter]. London: Medicines Control Agency, 2002
13. Medicines and healthcare products Regulatory Agency. Herbal
However, optimising treatment, including reducing safety news [online]. Available from URL: http://
the potential for ADRs, on the basis of a patient’s medicines.mhra.gov.uk [Accessed 2003 Jul 29]
genotype has barely been discussed in the context of 14. Thomas KJ, Nicholl JP, Coleman P. Use and expenditure on
complementary medicine in England: a population based sur-
herbal medicines. vey. Complement Ther Med 2001; 9: 2-11
15. Mintel International Group Ltd. Complementary medicines,
Acknowledgements UK. April 2003. London: Mintel International Group Limited,
2003
The author thanks Mrs Leigh Henderson, Medicines and 16. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative
medicine use in the United States, 1990-1997: results of a
Healthcare products Regulatory Agency (MHRA) for provid- national follow-up survey. JAMA 1998; 280 (18): 1569-75
ing data from the ADROIT system, and Dr Linda Anderson, 17. MacLennan AH, Wilson DH, Taylor AW. The escalating cost
MHRA, and the referees for their comments. and prevalence of alternative medicine. Prev Med 2002; 35:
166-73
No sources of funding were used to assist in the prepara- 18. Barnes J. Herbal therapeutics (1): an introduction to herbal
tion of this manuscript. The author has no conflicts of interest medicinal products. Pharm J 2002; 268: 804-6
directly relevant to the contents of this review. The views 19. Barnes J, Ernst E. Traditional herbalists’ prescriptions for com-
expressed are those of the author alone and do not necessarily mon clinical conditions: a survey of members of the UK
National Institute of Medical Herbalists. Phytother Res 1999;
represent the views of the MHRA or the individuals men- 12: 369-71
tioned above. 20. Barnes J, Mills SY, Abbot NC, et al. Different standards for
reporting ADRs to herbal remedies and conventional OTC
medicines: face-to-face interviews with 515 users of herbal
References remedies. Br J Clin Pharmacol 1998; 45: 496-500
1. Mann RD, Andrews EB, editors. Pharmacovigilance. Chicester: 21. Gulian C, Barnes J, Francis S-A. Types and preferred sources of
Wiley, 2002 information concerning herbal medicinal products: face-to-
2. Waller PC, Evans SJW. A model for the future conduct of face interviews with users of herbal medicinal products [ab-
pharmacovigilance. Pharmacoepidemiol Drug Saf 2003; 12: stract]. Int J Pharm Prac 2002; 10 Suppl.: R33
17-29 22. Vickers AJ, Rees RW, Robin A. Advice given by health food
3. World Health Organization. Draft WHO guidelines on safety stores: is it clinically safe? J R Coll Phys Lond 1998; 32 (5):
monitoring and pharmacovigilance of herbal medicines. Gene- 426-8
va: World Health Organization, 2003 23. Quinn CG, Waterman P. A comparison of the teaching of herbal
4. European Agency for the Evaluation of Medicinal Products medicine in the United Kingdom and Europe. University of
(EMEA) Committee for Proprietary Medicinal Products Strathclyde, 1997
(CPMP). Note for guidance on quality of herbal medicinal 24. Barnes J, Abbot NC. Experiences with complementary medi-
products. CPMP/QWP/2819/00 (EMEA/CVMP/814/00). cines: a survey of community pharmacists [abstract]. Pharm J
London: EMEA, July 26, 2001 [online]. Available from URL: 1999; 263: R37-43
http://www.emea.eu.int/pdfs/human/qwp/281900en.pdf [Ac- 25. Barnes J. An examination of the role of the pharmacist in the
cessed 2003 Aug 13] safe, effective and appropriate use of complementary medi-
5. The Medicines Act. London: The Stationery Office, 1968 cines [PhD thesis]. University of London, 2001
6. Commission of the European Communities. Proposal for 26. Department of Health. Herbal Medicine Regulatory Working
amending the directive 2001/83/EC as regards traditional herb- Group [online]. Available from URL: http://www.doh.gov.uk/
al medicinal products. 2002/0008 (COD). Brussels: European herbalmedicinerwg [Accessed 2003 Jul 29]
Commission, 2002 27. Medicines Control Agency. Traditional ethnic medicines: pub-
7. Commission of the European Communities. Amended proposal lic health and compliance with medicines law. November
for a Directive of the European Parliament and of the Council 2001. London: Medicines Control Agency, 2001
amending the directive 2001/83/EC as regards traditional herb-
al medicinal products. 2002/0008 (COD). Brussels: European 28. The Medicines (Aristolochia and Mu Tong) (Prohibition) Order
Commission, 2003 2001 (SI 2001/1841). London: The Stationery Office, 2001
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
850 Barnes
29. Commission of the European Communities. Directive 2001/83/ 52. Medicines Control Agency. Consultation MLX 286: proposals
EC. Brussels: European Commission, 2001 to prohibit the herbal ingredient kava-kava (Piper methys-
ticum) in unlicensed medicines. London: Medicines Control
30. Adcock H. Medicines Control Agency proposes ban for kava- Agency, 2002 Jul 19
containing products. Pharm J 2002; 269: 128 53. The Medicines for Human Use (Kava-Kava) (Prohibition) Or-
31. Royal Pharmaceutical Society of Great Britain. Medicines, eth- der 2002 (SI2002/3170). London: The Stationery Office, 2003
ics and practice: a guide for pharmacists. London: Royal 54. Van Grootheest K, de Graaf L, de Jong-van den Gerg LTW.
Pharmaceutical Society of Great Britain, 2003 Jul: 92 Consumer adverse drug reaction reporting: a new step in
32. De Smet PAGM, H¨
ansel R, Keller K, et al., editors. Toxicologi- pharmacovigilance? Drug Saf 2003; 26 (4): 211-7
cal outlook on quality assurance of herbal remedies. In: Ad- 55. Wingfield J, Walmsley J, Norman C. What do Boots pharma-
verse effects of herbal drugs. Vol 1. Berlin: Springer Verlag, cists know about yellow card reporting of adverse drug reac-
1992 tions? Pharm J 2002; 269: 109-10
33. Busse W. The significance of quality for efficacy and safety of 56. Green CF, Mottram DR, Raval D, et al. Community pharma-
herbal medicinal products. Drug Inf J 2000; 34: 15-23 cists’ attitudes to adverse drug reaction reporting. Int J Pharm
34. Ernst E, De Smet PA, Shaw D, et al. Traditional remedies and Prac 1999; 7: 92-9
the “test of time”. Eur J Clin Pharmacol 1998; 54 (2): 99-100 57. Biscoe R, Houghton JE, Woods FJ. An audit of the level of
35. De Smet PAGM. Health risks of herbal remedies. Drug Saf encouragement given by medicines information pharmacists to
1995; 13 (2): 81-93 enquirers of suspected adverse drug reactions to complete a
36. Woodfield R. Proposed directive on traditional herbal medicinal yellow card report: perspectives in patient safety [abstract].
products: progress in European negotiations [letter]. London: 28th UK Medicines Information Conference Proceedings;
Medicines and Healthcare products Regulatory Agency, 2003 2002 Sep 19-21; Chester, UK
37. Press release (revised). Meeting of the Working Party on Herbal 58. Pharmacists’ adverse drug reaction reporting to start on April 1.
Medicinal Products. February 24-25, 2003, European Agency Pharm J 1997; 258: 330-1
for the Evaluation of Medicinal Products, London. Document 59. Cox A. Embracing ADR reporting could improve pharmacists’
reference: EMEA/CPMP/HMPWP/1090/03Rev.1 standing [letter]. Pharm J 2002; 269: 14
38. De Smet PAGM. An introduction to herbal pharmacovigilance. 60. Moffat T. Adverse drug reaction (ADR) reporting by pharma-
In: De Smet PAGM, Keller K, H¨
ansel R, et al., editors. cists. London: Royal Pharmaceutical Society of Great Britain,
Adverse effects of herbal drugs. Vol. 3. Berlin: Springer- 2003. In press
Verlag, 1997 61. Farah MH, Edwards R, Lindquist M, et al. International moni-
39. Extension of the yellow card scheme to unlicensed herbal reme- toring of adverse health effects associated with herbal medi-
dies. Curr Prob Pharmacovigilance 1996; 22: 10 cines. Pharmacoepidemiol Drug Saf 2000; 9: 105-12
40. Shaw D, Leon C, Kolev S, et al. Traditional remedies and food 62. World Health Organisation Collaborating Centre for Interna-
supplements: a 5-year toxicological study (1991-1995). Drug tional Drug Monitoring. Draft guidelines for herbal ATC clas-
Saf 1997; 17 (5): 342-56 sification. Uppsala: The Uppsala Monitoring Centre, 2002
41. Extension of the yellow card scheme to pharmacists. Curr Prob 63. Broughton A. Yellow card reporting scheme. Eur J Herb Med
Pharmacovigilance 1997; 23: 3 2001; Dec: 3-6
42. Davis S, Coulson R. Community pharmacist reporting of sus- 64. Ward T. Register for Chinese Medical Herbalists. 2002 Sep 18
pected ADRs: (1) the first year of the yellow card demonstra- 65. European Scientific Co-operative on Phytotherapy (ESCOP).
tion scheme. Pharm J 1999; 263: 786-8 PhytoNET [online]. Available from URL: http://www.es-
43. Medicines and Healthcare Products Regulatory Agency. Ad- cop.com [Accessed 2003 Jul 29]
verse drug reaction on-line information tracking (ADROIT) 66. British Herbal Medicine Association. Code of good practice:
system, Jul 25, 2003 unlicensed herbal remedies. Bournemouth: British Herbal
44. Major E. The yellow card scheme and the role of pharmacists as Medicine Association, 1997 Mar
reporters. Pharm J 2002; 269: 25-6 67. Nutraceuticals adverse events database. Executive summary
45. Davis S, Raine JM. Spontaneous reporting: UK. In: Mann RD, [online]. Available from URL: http://www.wmin.ac.uk/sih/re-
Andrews E, editors. Pharmacovigilance. Chicester: Wiley, search/proj_nutraceuticals.html [Accessed 2003 Jul 25]
2002: 195-207 68. Logue M, Pendry B, Waters E, et al. Irritable bowel syndrome:
46. Committee on Safety of Medicines and Medicines and Health- herbal research update [abstract]. Phytotherapy Research Con-
care Products Regulatory Agency. The yellow card scheme: ference; 2002 Apr, Glasgow
extension of the yellow card scheme to nurse reporters [on- 69. Shakir SAW. PEM in the UK. In: Mann RD, Andrews EB,
line]. Available from URL: http://medicines.mhra.gov.uk/ editors. Pharmacovigilance. Chicester: Wiley, 2002: 333-44
aboutagency/regframework/csm/csmhome.htm [Accessed 70. Layton D, Sinclair HK, Bond CM, et al. Pharmacovigilance of
2003 Jul 25] over-the-counter products based in community pharmacy:
47. MCA launches web version of yellow card scheme. Pharm J methodological issues from pilot work conducted in Hamp-
2002; 269: 631 shire and Grampian, UK. Pharmacoepidemiol Drug Saf 2002;
48. Patients able to report ADRs via NHS Direct. Pharm J 2003; 11: 503-13
270: 608 71. Sinclair HK, Bond CM, Hannaford PC. Pharmacovigilance of
49. Morrison-Griffiths S, Walley TJ, Park BK, et al. Reporting of over-the-counter products based in community pharmacy: a
adverse drug reactions by nurses. Lancet 2003; 361: 1347-8 feasible option? Pharmacoepidemiol Drug Saf 1999; 8: 479-91
50. Updated “yellow card” launched. Pharm J 2000; 265: 387 72. Strom BL. How should one perform pharmacoepidemiology
51. Gogolak VV. The effect of backgrounds in safety analysis: the studies? Choosing among the available alternatives. In: Strom
impact of comparison cases on what you see. BL, editor. Pharmacoepidemiology. 3rd ed. Chicester: Wiley,
Pharmacoepidemiol Drug Saf 2003; 12: 249-52 2000: 401-13
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
Pharmacovigilance of Herbal Medicines 851
73. Berry DC, Knapp PR, Raynor DK. Is 15% very common: 79. House of Lords Select Committee on Science and Technology,
informing people about the risks of medication side effects. Int Session 1999-2000, 6th report. Complementary and alternative
J Pharm Prac 2002; 10: 145-51 medicine. London: The Stationery Office, 2000
74. Berry DC, Raynor DK, Knapp P, et al. Patients’ understanding 80. Department of Health. A vision for pharmacy in the new NHS.
of risk associated with medication use: impact of European London: Department of Health, 2003 Jul
Commission guidelines and other risk scales. Drug Saf 2003;
26 (1): 1-11 81. Pirmohamed M, Park BK. Genetic susceptibility to ADRs.
75. Edwards IR, Olsson S. WHO programme: global monitoring. Trends Pharmacol Sci 2001; 22 (6): 298-305
In: Mann RD, Andrews EB, editors. Pharmacovigilance.
Chicester: Wiley, 2002: 169-82
76. Cosyns J-P. Aristolochic acid and ‘Chinese herbs nephropathy’: Correspondence and offprints: Dr Joanne Barnes, Centre for
a review of the evidence to date. Drug Saf 2003; 26 (1): 33-48
77. Henderson L, Yue QY, Bergquist C, et al. St John’s wort Pharmacognosy and Phytotherapy, School of Pharmacy,
(Hypericum perforatum): drug interactions and clinical out- University of London, 29-39 Brunswick Square, London,
comes. Br J Clin Pharmacol 2002; 54: 349-56 WC1N 1AX, UK.
78. Kava-kava and hepatotoxicity [news item]. Curr Prob Pharma-
covigilance 2002; 28: 6 E-mail: joanne.barnes@ulsop.ac.uk
Adis Data Information BV 2003. All rights reserved. Drug Safety 2003; 26 (12)
