Different standards for reporting ADRs to herbal remedies and conventional OTC medicines: face-to-face interviews with 515 users of herbal remedies
Joanne Barnes Department of Complementary Medicine, Postgraduate Medical School, University of Exeter, 25 Victoria Park Road, Exeter EX2 4NT.
Aims To determine whether adverse drug reactions (ADRs) to herbal remedies would be reported differently from similar ADRs to conventional over-the-counter (OTC) medicines by herbal-remedy users.
Methods Face-to-face interviews (using a structured questionnaire) with 515 users of herbal remedies were conducted in six pharmacy stores and six healthfood stores in the UK. The questionnaire focused on the likely course of action taken by herbal-remedy users after experiencing an ADR associated with a conventional OTC medicine and a herbal remedy.
Results Following a ‘serious’ suspected ADR, 156 respondents (30.3%) would consult their GP irrespective of whether the ADR was associated with the use of a herbal remedy or a conventional OTC medicine, whereas 221 respondents (42.9%) would not consult their GP for a serious ADR associated with either type of preparation. One hundred and thirty-four respondents (26.0%) would consult their GP for a serious ADR to a conventional OTC medicine, but not for a similar ADR to a herbal remedy, whereas four respondents (0.8%) would consult their GP for a serious ADR to a herbal remedy, but not for a similar ADR to a conventional OTC medicine. Similar differences were found in attitudes towards reporting ‘minor’ suspected ADRs.
Conclusions Consumers of herbal remedies would act differently with regard to reporting an ADR (serious or minor) to their GP depending on whether it was associated with a herbal remedy or a conventional OTC medicine. This has implications for herbal pharmacovigilance, particularly given the increasing use of OTC herbal remedies. The finding that a high proportion of respondents would not consult their GP or pharmacist following ADRs to conventional OTC medicines is also of concern.
In the UK, herbal remedies (or ‘phytomedicines’) are increasingly being used by the general public on a self-selection basis to replace or complement conventional medicines . The use of herbal remedies is widespread across Europe—in 1991, the total over-the-counter (OTC) market for herbal remedies was £1.45 billion . Another source estimated that, in 1992, the herbal market of the European Community was worth US$2.4 billion . More recently, the market for licensed herbal medicines in the UK was estimated to be worth £38 million in 1996, representing over half of the total market for complementary remedies .
One of the reasons for the popularity of herbal remedies is the belief among many users and suppliers of herbal remedies that these preparations are natural and therefore ‘safe’ . This, however, is a misconception—herbal remedies can produce adverse drug reactions (ADRs) [5, 6], some of which can be serious and even fatal [7, 8]. However, because users believe that such remedies are ‘safe’, individuals experiencing ADRs may not associate these with their use of herbal remedies . A further complication is that, in the UK, the majority of herbal remedies are self-prescribed , and many individuals may be reluctant to tell their general practitioner (GP) that they are using them . Even if ADRs are reported by patients, their GPs may not be fully briefed about the use and effects (adverse or otherwise) of herbal remedies.
The European Union (EU) has commissioned research into this problem through its BIOMED (Biomedical and Health Research) programme. As part of that programme, this study was designed to determine whether ADRs to herbal remedies would be reported differently from similar ADRs to conventional OTC medicines, and to identify experiences of ADRs to herbal remedies and how they are perceived by consumers.
Experienced interviewers, recruited and trained for this task by a market research company, were provided with study questionnaires, photographs and lists of examples of herbal remedies, other complementary remedies and conventional OTC medicines, and a list of examples of ADRs. These materials were used to assist interviewers and interviewees in identifying what was and what was not a herbal remedy. The list of ADRs was used as a prompt if consumers were unsure what was meant by a ‘side-effect’, or if they answered that they had never experienced any ‘side-effects’ to herbal remedies. Interviewers were instructed to position themselves at an appropriate distance from the herbal remedies counter in the respective stores, and to approach customers who had purchased herbal remedies or those who had browsed the herbal remedies section. Customers were asked if they would be willing to be interviewed as part of a study on herbal remedies. Those agreeing to be interviewed were asked if they ever use herbal remedies; if they answered, ‘No’ the interview was terminated. If a customer answered, ‘Yes’, the interviewer continued with the questionnaire (written consent was not obtained); interviews took around 8 min. All questionnaires were analysed at the University of Exeter.
Two types of outlet—Boots the Chemists Ltd (BTC) and Holland & Barrett (H & B), representing a pharmacy setting and healthfood store setting, respectively—were chosen. Interviews were conducted in six BTC stores with a high turnover of herbal remedies (Manchester, Leeds, Newcastle, Milton Keynes, London, Cardiff) and in six H & B stores near the selected BTC stores (Manchester, Leeds, Newcastle, Milton Keynes, London, Swansea).
The study was conducted in September, 1996. An interviewer was present in the selected stores on 2 consecutive days for 8h per day in BTC stores, and 4h per day in H & B stores. The study was weighted more towards BTC customers than H & B customers to reflect market share .
A structured questionnaire for customer interviews was designed and developed for this survey by researchers at the University of Exeter. Copies of the questionnaire are available on request.
Respondents were asked what herbal remedies they used, how often, and for what condition. The same questions were asked for conventional OTC medicines. Respondents were also asked if they ever used vitamins, minerals or dietary supplements, or other ‘natural’ health remedies (e.g. homoeopathic medicines, aromatherapy oils). Data on how respondents choose their herbal remedies and from where they obtain them, were also collected. In addition, respondents were asked if they had ever experienced any ‘side-effects’ after taking herbal remedies and, if so, were asked to provide the following details: name of herbal remedy; associated ‘side-effect’; severity (mild, moderate or severe); if they reported the ‘side-effect’ and to whom; if they stopped taking the remedy because of the ‘side-effect’. Respondents were also asked for demographic information (gender, age, occupation); social grade and ethnic group were assessed by the interviewer.
The key part of the interview sought to obtain information on what action respondents would take if they experienced (a) a ‘serious side-effect’ (for the purposes of this survey, this was defined as symptom(s) that were ‘worrying or alarming’), and (b) a ‘minor side-effect’ (defined as symptom(s) that ‘caused some discomfort, but were not alarming’) to a conventional OTC medicine and to a herbal remedy. Respondents were allowed to select one or more of the following responses: continue taking [the preparation] and see if symptom(s) resolved; stop taking immediately; consult your doctor; consult your pharmacist; consult another health care practitioner; other action.
Prior to conducting the full survey, a pilot survey was conducted. Thirty-two herbal remedy users were interviewed by one interviewer in the BTC store in Leeds. Following the pilot study, a minor alteration was made to the questionnaire (the order of the two questions on the action respondents would take following a serious and a minor ADR was reversed).
Data from the full survey were entered via a semiautomated Foxbase application into a spreadsheet for analysis. The data from the pilot study were not included in the final analysis.
Six hundred and ninety individuals agreed to be interviewed. Of these, 175 (25.4%) stated that they did not use herbal remedies and therefore these interviews were terminated. Five hundred and fifteen face-to-face interviews with users of herbal remedies were conducted: 336 in BTC stores and 179 in H & B stores. Females predominated (82% overall). The ethnic origin of respondents was predominantly Caucasian (91%); Afro-Caribbean (2%), Indian/Pakistani (2%) and Chinese/Japanese (1%) ethnic groups were also represented. The age distribution of respondents was: <20 years, 2%; 20–29 years, 15%; 30–39 years, 20%; 40–49 years, 24%; 50–59 years, 19%; 60 years, 20%. There were no marked differences in age distribution between the two types of stores.
Sixty-two per cent of all respondents (58.0 and 68.2% for BTC and H & B respondents, respectively) used one or more herbal remedies regularly, whereas 38% (42.0 and 31.8% for BTC and H & B respondents, respectively) used one or more herbal remedies occasionally (respondents were allowed to name a maximum of three remedies). Eighty-one per cent of all respondents (83 and 77% for BTC and H & B respondents, respectively) were also regular or occasional users of conventional medicines; 78% (79 and 77% for BTC and H & B respondents, respectively) stated that they used vitamins, minerals and/or food supplements; 49% (51 and 45% for BTC and H & B respondents, respectively) stated that they used other ‘natural’ health remedies (e.g. essential oils used in aromatherapy, homoeopathic remedies).
Respondents choose their herbal remedies on the basis of a friend's or family member's recommendation (31% of replies), on the basis of their own knowledge (40% of replies) and on the basis of a pharmacist's recommendation, on a prescription or recommendation from their doctor, and on the recommendation, or supplied by, a herbal medicine practitioner (6% of replies for each). There were no marked differences between BTC respondents and H & B respondents with regard to choosing herbal remedies except that H & B respondents were more likely than BTC respondents to choose herbal remedies recommended by a complementary health practitioner other than a herbalist (23/179 vs 11/336 for H & B vs BTC respondents, respectively; χ2=17.37; P<0.001).
Attitudes towards reporting ADRs
Table 1 shows the numbers of respondents who would take a particular course of action after experiencing a suspected ADR to a herbal remedy and to a conventional OTC medicine. The data are presented in a manner that allows the numbers of respondents who would act differently for ADRs to herbal remedies than for similar ADRs to conventional OTC medicines to be identified.
Number of respondents (% of total) that would choose a particular course of action after experiencing a suspected ADR to (i) a conventional OTC medicine and (ii) a herbal remedy. Participants responded yes or no for each type of preparation. The four groups represent respondents who answered yes for both preparations, no for both, and those who gave different answers depending on whether the ADR was associated with a conventional OTC medicine or a herbal remedy.
Following a serious ADR, 156 respondents (30.3% of all replies) would consult their GP irrespective of whether the ADR was associated with the use of a herbal remedy or a conventional OTC medicine; 221 respondents (42.9%) would not consult their GP for a ‘serious’ ADR associated with either type of preparation. One hundred and thirty-four respondents (26.0%) would consult their GP for a serious ADR to a conventional OTC medicine, but not for a similar ADR to a herbal remedy, whereas four respondents (0.8%) would consult their GP for a serious ADR to a herbal remedy, but not for a similar ADR to a conventional OTC medicine. Similar differences were found in the attitudes of herbal-remedy users towards reporting ‘minor’ ADRs associated with herbal remedies and for similar ADRs to conventional OTC medicines.
Subgroup analysis of respondents interviewed in BTC stores and those interviewed in H & B stores revealed the following differences between the two groups. Following a serious ADR, significantly more BTC respondents than H & B respondents would consult their GP irrespective of whether the ADR was associated with a conventional OTC medicine or a herbal remedy (113/336 vs 43/179 for BTCvs H & B respondents, respectively; χ2=5.11; P<0.05). For minor ADRs, the result was reversed—H & B respondents were more likely than were BTC respondents to consult their GP irrespective of whether the ADR was associated with a conventional OTC medicine or a herbal remedy (19/179 vs 14/336 for H & B vs BTC respondents, respectively; χ2=8.10; P<0.01). However, BTC respondents were more likely to consult a pharmacist than were H & B respondents for minor ADRs irrespective of whether the ADR was associated with a herbal remedy or a conventional OTC medicine (31/336 vs 4/179 for H & B vs BTC respondents, respectively; χ2=9.01; P<0.01).
Subgroup analysis of ‘young’ respondents (<30 years of age) and ‘old’ respondents (>50 years) did not reveal a significant effect of age on ADR reporting for either serious or minor ADRs.
Perceptions and experience of ADRs associated with herbal remedies
Thirty-one respondents stated that they had experienced ADRs to herbal remedies. However, six of these reports referred to non-herbal complementary remedies. A further three reports cannot definitely be called herbals (‘Vitalax’, ‘Keratine’ and a product the name of which was written illegibly). Excluding the latter three reports gives a total of 22 respondents (4.3%) who reported having experienced an ADR associated with the use of a herbal remedy. Of these, four respondents (0.8% of total) rated the adverse effect as ‘severe’, eight (1.6%) rated it as ‘moderate’, nine (1.7%) rated it as ‘mild’ and one entry was missing. Of the four reports rated ‘severe’, two were reported (both to a doctor); two were not reported. One ‘severe’ report (‘asthma symptoms’ associated with the use of Royal Jelly† reported to a doctor) gives cause for concern. In total, four (18%) of the 22 respondents who reported ADRs claimed to have informed their GP (we have not received replies to our letters attempting to verify these reports), 16 (73%) stopped taking the medicine concerned and six (27%) did not stop taking the preparation.
This is the first study to provide evidence that herbal remedy users would be less likely to consult their GP for suspected ADRs (serious or minor) to herbal remedies than for similar ADRs to conventional OTC medicines. This has implications for herbal pharmacovigilance and implies that many suspected ADRs to herbal remedies will go unmonitored.
There may be several reasons for this finding. Herbal remedies are largely used on a self-treatment basis and some users may not realize that they can consult their GP about ADRs to such products. Others may be reluctant to admit herbal-remedy use to their GP by consulting him/her for suspected ADRs, while some users may feel it is more appropriate to consult the herbal practitioner from whom the remedies were obtained. In a study of unconventional medicine use involving 1539 adults in the US, Eisenberg et al. reported that of 34% respondents who reported using at least one unconventional therapy in the previous year, 72% did not inform their doctor of their use of the therapy .
There is an increasing amount of research into patients’ attitudes towards complementary therapies and the reasons why people choose to use such therapies as well as, or instead, conventional medicine . This is an important and complex area which is likely to have implications for ADR reporting. The present study appears to have uncovered differences between two groups of users of herbal remedies with regard to their attitudes towards reporting ADRs. BTC respondents would be more likely than H & B respondents to consult their GP for a serious ADR irrespective of whether it was associated with a herbal remedy or a conventional OTC medicine, and would be also more likely to consult a pharmacist for a minor ADR irrespective of whether it was associated with a herbal remedy or a conventional OTC medicine. Why there should be this difference in willingness to consult a health professional is a matter of debate. It may reflect H & B customers disenchantment with the orthodox approach to health care, or it may simply be that a health care professional is present on site in BTC stores.
Whatever the reason, the findings of the present survey raise concerns not only with regard to reporting of ADRs associated with herbal remedies, but also for those associated with conventional OTC medicines. Even for a ‘serious’ ADR, only 290 respondents (56.3%) would consult their GP; for ‘minor’ ADRs associated with conventional OTC medicines, only 108 respondents (21.0%) would do so. Also, less than 10% of respondents would consult a pharmacist for a ‘serious’ or ‘minor’ ADR associated with a conventional OTC medicine.
Even those suspected ADRs which are reported to a GP and which meet ADR-reporting criteria may not be reported on to national pharmacovigilance centres. In the UK, hospital physicians have been shown to grossly under-report ADRs that meet CSM criteria ; there is no evidence to suggest that GPs are any more diligent in this area. Furthermore, deficiencies in the reporting process may be even more likely to occur with herbal remedies . The UK Medicines Control Agency's (MCA) and Committee on Safety of Medicine's (CSM) Yellow Card scheme already requests reports relating to suspected ADRs to all (i.e. both licensed and unlicensed) herbal remedies , yet (perhaps because they are not aware of this request) reporting by doctors is still limited . In April 1997, the MCA extended the Yellow Card scheme to include hospital and (in certain regions) community pharmacists ; community pharmacists are seen as having a critical role in areas of limited reporting by doctors, e.g. over-the-counter medicines, and licensed and unlicensed herbal products .
There is an increasing awareness of the need to monitor the safety of herbal remedies . Our findings lend support both to the MCA/CSM decision to extend its Yellow Card reporting scheme to pharmacists, and to the European Scientific Co-operative for Phytotherapy's (ESCOP) pharmacovigilance system for herbal remedies (PhytoNet). In the latter, suspected ADRs to herbal remedies may be reported via a password-protected, Internet-based reporting system which is being targeted at all health professionals who use herbal remedies.
In conclusion, consumers of herbal remedies would act differently with regard to reporting an ADR (serious or minor) to their GP, depending on whether the ADR was associated with a herbal remedy or a conventional OTC medicine. This implies that many ADRs to herbal remedies may go unmonitored. Our findings illustrate the need for greater public awareness that ADRs to herbal remedies can occur, and that such events should be reported to an appropriate authority. Professionals also need to be aware of the potential for herbal remedies to cause ADRs and routinely question their patients about their use of such remedies.
In the longer term, further research to investigate the safety and efficacy of herbal remedies is needed so that the risk/benefit ratio of using a particular herb for a specific condition can be determined.
This study was conducted as part of the European Union as part of a BIOMED (Biomedical and Health Research) programme entitled, Determining European standards for the safe and effective use of phytomedicines; J. Barnes holds the Boots Research Fellowship in Complementary Medicine.
We would like to express our appreciation of the contribution to data collection of the late Mr Jeff Shaw of the market research company Bemrose Shaw Berridge and Partners Ltd, Derby.