13th May 2011, Volume 124 Number 1333

Llifon Edwards, Sarah Jefferies, Bridget Healy, Mark Weatherall, Richard Beasley, Philippa Shirtcliffe

Effective regulation of the complementary and alternative health sector serves the public interest by protecting consumers from unsafe or inadequately trained practitioners, and from products that are unsafe or make misleading claims.1

Most complementary and alternative medicine (CAM) products- also referred to as traditional medicine or TM- are currently marketed as dietary supplements. In New Zealand they can be obtained from pharmacies, health food stores (HFS), homeopathic pharmacies and supermarkets as well as practitioners and non-shop based retailers. CAM products are regulated by an array of legislation, such as the Medicines Act 1981 and the Food Act 1981, depending on whether they are defined as medicines, dietary supplements or food; which category a product falls into may depend on either the product or the desire of the manufacturer.

Currently, there is no specific legislation to regulate CAM practitioners (with the exception of chiropractors) although there are a number of regulatory provisions such as the Health and Disability Commissioner Act 1994. However retail assistants in these stores are not required to hold any formal training or qualifications in CAM.

Since the 1990s the use of CAM has surged worldwide with herbal and other complementary products increasingly being used to treat a variety of medical conditions. Indeed, the latest New Zealand Health Consumers’ Survey in 1997 reported that approximately 50% of the surveyed population had tried at least one form of CAM.2 In the United States, the 2007 National Health Interview Survey, which included a comprehensive survey of CAM use, showed that approximately 38% of adults use CAM.3

A common misperception among patients taking CAM is that these remedies are “natural”, safe and do not interfere with any conventional medicines they might take.4,5 However, not only are there risks of product contamination, adulteration and toxicity from herbal ingredients, but most significantly a risk of interactions with conventional drugs.6 This particularly applies to drugs with a narrow therapeutic index such as warfarin and of all the published interactions between CAM and conventional medicines, warfarin is one of the most common conventional drugs involved.5

Thus, the ability to provide accurate advice regarding the safe use of CAM products by those promoting them is important especially with regards to CAM-drug interactions. Few studies have investigated advice from HFS but in general, the advice given has been found wanting.7

The key issue that this study sought to highlight was the general risk that a consumer may face when they seek help/advice from someone promoting a CAM product. This was related to both the lack of an appropriate referral (to mainstream medicine) in the setting of a potentially serious adverse event from warfarin, and the inherent risks in taking CAM products that may interact with warfarin with potentially significant adverse effect.


Study design—This study was conducted over a 4-week period between May and June 2010. Twenty-one HFS and 21 pharmacies were visited by a 30-year -old male researcher. The HFS were identified by a search of the Yellow Pages telephone directory and were matched with pharmacies on the basis of the closest geographical location to the HFS.
The researcher approached the retail assistant with complaints of tiredness, and difficulty concentrating at work. He would state that he had been taking warfarin over the past 2 months for treatment of a pulmonary embolus, which occurred following a long distance flight back from Europe. He asked for recommendations to assist with these symptoms.
If specifically asked, the researcher stated that he was an otherwise fit and healthy 30-year-old, that he had not been to see his general practitioner (GP) with regards to these symptoms, that he did not have a history of anaemia or abnormal bleeding, ate a well balanced diet, was married, did not feel that he was depressed, and currently worked in an office job doing clerical work.
The investigator was specifically not allowed to ask if any recommended preparation would interact with warfarin, nor for any advice regards what he should do about continuation of the warfarin whilst taking the remedy or awaiting a GP appointment.
This scenario was chosen because tiredness is a common complaint with many potential causes. Warfarin is a commonly used drug, with a narrow therapeutic index and so interactions with CAM products are of particular relevance. In this setting, it is imperative that a referral to the GP is made to further investigate the possibility of occult bleeding.
As warfarin is well recognised to interact with a large number of drugs (with the potential of both increasing and decreasing its effect), we were also looking for some recognition of a potential impact on anticoagulant control.
The researcher would record the name, manufacturer and retail price of any products recommended for purchase immediately after leaving the premises. The ingredients of each product were then confirmed by Internet search for the product.
An interaction between a product and warfarin was considered to be a potential issue if the product contained a substance that has been specifically identified as affecting INR (the standard measure of anticoagulant effect), the metabolism of warfarin, or other related comment such as “may increase the bleeding risk”. Possible interactions were checked using a standard pharmacological text8 and a further text specialising in herb-medicine interaction.9 Advice was defined as appropriate if the client was advised to have more frequent blood test monitoring or notify the anticoagulant clinic or see their doctor.
The study was approved by the Wellington Regional Ethics Committee.
Statistical analysis—Paired contingency table analysis was used to examine the agreement between the matched pairs in advice between pharmacies and health food stores. Statistical analysis was by an exact McNemar's test and a confidence interval for the difference in marginal proportions, representing the proportion of pharmacies that gave particular advice versus the proportion of matched health food stores that gave advice.10 A statistically significant McNemar's test means the marginal proportions of the contingency table are different.
The number of matched pharmacies/health food stores was based on an earlier study comparing the advice from HFS assistants with that of pharmacy assistants given to an individual presenting with symptoms suggestive of moderate to severe asthma who should be referred to a medical practitioner.11 Based on this, it was calculated that the study would need to have 19 store/pharmacy pairs to have 80% power to detect the difference.


21 pharmacies and 21 HFS in the greater Wellington area, matched for location, were visited. In 10 of the pharmacies retail assistants requested pharmacists to either talk to the investigator or check product safety on their behalf and another three suggested consulting with a doctor or the pharmacist. The investigator was advised by retail assistants in all the HFS apart from one in which he was advised by a qualified naturopath.

Consumer was advised to consult his doctor if symptoms not improving or before starting a recommended product—A pharmacy was significantly more likely to advise the consumer to consult a doctor (13/21, 61.9%) than a HFS (3/21, 14.3%), with a difference in marginal proportions of 47.6% (95%CI 22.5 to 72.7), p= 0.006

Number of products recommended—A total of 32 different preparations were recommended, 15 by pharmacies and 34 by HFS (Table 1). A HFS was more likely to recommend more products (none or one product 9/21, 42.9%) than a pharmacy (19/21, 90.5%), with a difference in marginal proportions of 47.6% (95% CI 26.3 to 69.0), p= 0.002.

Table 1. Various products recommended by health food stores (H) and pharmacies (P), grouped according to main components
Multivitamin products
B-complex and other
Men’s Multi (H)

Men’s Care Multi (H)

Centrum (H, P)

B Stress Free (H)

Go Adrenal Support (2H)


Ultra-B (H)

Men’s Ultivite (H)

Multi for Men (H, P)

Executive B (4P)

Ultra-Life (2P)

Age Wise Men’s Daily (P)

Astraforte (H)

Quercetin Complex (H)

Ginseng (H)

Hi-Q (2H)

Fatigue Fighter (H)

Memory Booster (H)

Hairy Lemon (P)

Spirulina (7H)

B-50 complex (H)

B-Max (3H)

B-100 (H)

Mega-B (H,P)

Ener-B (P)

Berocca (P)

Go-B Complex (P)

Vitamin C (P)

L-tyrosine (H)

Flaxomega (H)

Complete Omega (H)

Arctic Cod Liver Oil (H)

Note: More than one product was sometimes recommended. Where a product was recommended by more than one H or P this is indicated by a number in front of the letter.

Advice regarding potential interactions with warfarin and management of warfarin—A pharmacy was significantly more likely to make no product recommendation or make a recommendation for a product with no interaction with warfarin (13/21, 61.9%) than a HFS (4/21, 19.1%), with a difference in marginal proportions of 42.9% (95% CI 14.6–71.1), p=0.023.

Similarly, a pharmacy either did not recommend a product or recommended a product with the correct advice regarding warfarin management in 14/21 cases (66.7%) compared with a HFS 5/21 (23.8%) with a difference in marginal proportions of 42.9% (95%CI 11.7–74.0), p= 0.03. Table 2 summarises those products that contain ingredients that may interact with warfarin.

Table 2. Products recommended that contain ingredients that may interact with warfarin
Products containing Vitamin E (α-tocopherol): may increase INR and caution advised even at recommended doses
Men’s Multi, Spirulina, Men’s Care Multi, Centrum, B Stress Free, CAA, Complete Omega, Arctic cod liver oil, ultra B, Men’s Ultivite, Multi for Men, Executive B, Ultra-Life, Age Wise
Products containing Schizandra chinensis* (may increase metabolism of warfarin) and/or Ginseng# (may decrease effects)
Go Adrenal Support#, Ginseng#, Hi-Q*#, Fatigue Fighter*#, B Stress Free#, Hairy Lemon#, Age Wise#
Products containing Ginkgo biloba (may increase bleeding risk)
Products containing Vitamin K (reduces effectiveness of warfarin)
Men’s Ultivite, Multi for Men, Age Wise
Products containing flax oil or fish oils (increase bleeding time)
Flaxomega, Complete Omega, Arctic cod liver oil


This study shows that there is a significant difference in the advice offered by pharmacies and HFS regarding management of tiredness in someone taking warfarin with nearly two-thirds of the former appropriately recommended referral to mainstream health services compared with less than 15% of HFS.

Similarly, about two-thirds of pharmacies gave good advice regarding adverse interactions between products and warfarin and regarding warfarin management compared with about one-quarter of HFS.

This study is the third in a series looking at the appropriateness of advice given by pharmacies and HFS and calling for better regulation of staff and CAM products.11,12 In the first scenario, which involved a new presentation of moderate to severe asthma, 92% of pharmacy staff recognised the severity of the condition and appropriately referred the consumer to a medical practitioner compared to only 35% of HFS.

A number of products were recommended, particularly by the latter, without any evidence of benefit. In the second scenario, involving a new presentation of severe hypertension, 96% of pharmacy staff recommended an immediate visit to a medical practitioner compared with 4% of HFS staff. Again a number of products were sold that were unlikely to be effective in severe hypertension.

Thus the findings of this paper are broadly in keeping with these studies although it is worth noting that about one-third of pharmacies did not offer sound advice.

A number of studies have examined the issue of co-ingestion of CAM and warfarin. Using postal questionnaires to general practice patients on warfarin, Smith et al found that 19.2% of patients were taking one/more herbal remedies; in 79% this had not been discussed with their doctor.6

Ramsey et al studied patients attending an anticoagulation clinic to start warfarin.5 Of 631 patients, 26.9% were using some form of CAM and 58% of these were using a product that could interact with warfarin. Shalansky et al completed a prospective longitudinal study amongst those prescribed warfarin for at least 4 months.13 Forty-three percent reported taking at least one CAM product with possible warfarin interactions. Similarly in their survey of hospital inpatients and outpatients on warfarin, Leung et al reported that 44.3% were using CAM.14

The major strength of this study was that it used one researcher to present a standardised scenario to 21 matched pharmacies and HFS allowing for a matched analysis. The investigator had to document the content of the conversation immediately after leaving, potentially introducing some recall bias however in practice this was not a major issue. It was also a hypothetical situation so there were no outcomes to monitor.

Ingredient lists for all products were available using a simple internet search, and all ingredients apart from one bacteria in one supplement were checked using authoritative reference texts.

The World Health Organization (WHO) actively recommends the regulation of all complementary and herbal medicine practitioners and all herbal products, particularly in situations where the practice of complementary medicine brings economic benefit.15 This is to ensure the quality of the service received and thus to protect the public from potential harm.

In 2008, the WHO Congress on Traditional Medicine resulted in the Beijing Declaration. This identified six articles including that governments should formulate national policies, regulations and standards as part of comprehensive national health systems to ensure appropriate, safe and effective use of traditional medicines and that governments should establish systems for the qualification, accreditation or licensing of traditional medicine practitioners.

In New Zealand, an extensive review of CAM was undertaken between 2002 and 2004 by the Ministerial Advisory Committee on Complementary and Alternative Health. This recommended that CAM consumers should be further protected through statutory regulation of high risk CAM modality practitioners (chiropractors and osteopaths), and self regulation of low risk CAM practitioners.16

In December 2003, the Australian and New Zealand Governments signed an agreement to establish a joint regulatory scheme for therapeutic products including CAM with the aim of safeguarding public health through regulation of the quality, safety and efficacy of therapeutic products. However negotiations were suspended indefinitely in 2007 following controversy about the proposal to include complementary medicines within the scope of the joint scheme.

In March 2010 a consultation paper was released by the Ministry of Health regarding the development of a Natural Health Products Bill with an expectation that the Bill would be enacted in 2011. Two of the key principles on which proposed legislation is based include that natural health products should be regulated separately from food and medicines, and that suppliers should be able to make low level health claims for products supported by evidence.17 However it does not address the issue of training of HFS (or pharmacy) assistants.

Discussion of all the issues involved in the regulation of CAM is outside the scope of this paper. However, at the crux of these is the fact that CAM is fundamentally a different paradigm to medical science. Consequently there is significant variation in views about how much and what training should be needed to attain regulation status including how to incorporate the intuitive skills and individualised approaches to providing health care that are key aspects of CAM practice.

A “consultation” is not defined in the Medicines Act but interaction between any practitioner and patient is covered by the Code of Health and Disability Services Consumers Rights 1996. It could be argued that a dialogue whereby there is an attempt to establish the patient’s ailment with the aim of making recommendations constitutes a consultation and that there should be an attempt to ascertain any additional circumstances that mean particular treatments should not be administered.

Of course it could also be argued that consumers need to have some skills to carry out their own evaluation if choosing to seek advice in this setting, or alternatively that the doctor who prescribed the warfarin is responsible for ensuring appropriate patient awareness/education.

With regards to efficacy of the products recommended there is also ongoing debate about what level of proof is required to assure the public that a (any) form of health treatment is safe and effective. With regards to potential warfarin-CAM interactions, nearly all the available information is based on in vitro data, animal studies and case reports and definitive cause-effect relationships have not been proven.18


To provide safe and quality advice to consumers, staff who are promoting CAM products need to not only obtain relevant history before recommending various products but also give accurate information regarding possible interactions with current medications and be prepared to refer back to mainstream medical services.

This study showed that this was responsibly done by the majority of pharmacies, although (perhaps surprisingly) a third were found wanting, and by only a minority of HFS. Once again we recommend the implementation of a formal training programme for HFS staff and pharmacy retail staff and better regulation of CAM products to reduce the risk for the consumer.


This study sought to highlight the general risk that a consumer may face when they seek medical advice from someone promoting a complementary and alternative medicine (CAM) product. It is the third in a series by the Medical Research Institute of New Zealand that has found such advice to be wanting, especially from health food stores, and once again calls for better regulation of staff and CAM products. To provide safe and quality advice to consumers, staff who are promoting CAM products need to not only obtain relevant history but also give accurate information regarding possible interactions with current medications. They should also be prepared to refer back to mainstream medical services.



There is currently no specific legislation to regulate either complementary and alternative medicine (CAM) products or the majority of those promoting them. This study sought to highlight the general risk a consumer may face when they seek help/advice from a pharmacy or health food store (HFS).


21 HFS, matched with pharmacies, were visited by a researcher complaining of tiredness, who stated he had been taking warfarin over the previous 2 months. The name, manufacturer and retail price of any products recommended were recorded immediately after leaving the premises. Paired contingency table analysis was used.


A pharmacy was significantly more likely to advise the consumer to consult a doctor (13/21) than a HFS (3/21) with a difference in marginal proportions of 47.6% (95%CI 22.5–72.7), p=0.006. A HFS was more likely to recommend more products, and only about one-quarter gave appropriate advice regarding possible interactions with warfarin and management of anticoagulation compared with two-thirds of pharmacies.


To provide safe and quality advice to consumers, those promoting CAM products need to obtain relevant history and give accurate information regarding possible dug interactions and be prepared to refer back to mainstream medical services. Better regulation of CAM products and those promoting them is called for.

Author Information

Llifon Edwards, Research Fellow, Medical Research Institute of New Zealand (MRINZ), Wellington; Sarah Jefferies, Research Fellow, MRINZ, Wellington; Brigid Healy, Research Fellow, MRINZ, Wellington; Mark Weatherall Associate Professor Medicine, University of Otago, Wellington; Richard Beasley, Consultant Physician, Capital and Coast DHB, Wellington; Philippa Shirtcliffe, Senior Research Fellow, MRINZ, Wellington


Dr Philippa Shirtcliffe, MRINZ, Private Bag 7902, Wellington 6242, New Zealand. Fax: +64 (0)4 3895707

Correspondence Email


Competing Interests



  1. Ministerial Advisory Committee on Complementary and Alternative Health. Complementary and Alternative Medicine: Current policies and policy issues in New Zealand and selected countries (a discussion document) 2003.http://www.newhealth.govt.nz/maccah.htm Last accessed 23rd November 2010.
  2. New Zealand Consumers’ Institute survey. From arsenic to zinc. Consumer 1997;363:20–27.
  3. National Centre for Complementary and Alternative Medicine. What is complementary and alternative medicine?http://nccam.nih.gov/health/whatiscam/ . Last accessed 23rd November 2010.
  4. Winslow LC, Kroll DJ. Herbs as medicines. Arch Intern Med, 1998;158(20):p. 2192–9.
  5. Ramsay N, Kenny M, Davies G, Patel J. Complementary and alternative medicine use among patients starting warfarin. British Journal of Haematology 2005;130:777–780.
  6. Smith L, Ernst E, Ewings P, et al. Co-ingestion of herbal medicines and warfarin. British Journal of General Practice 2004;54:439–441.
  7. Vickers AJ, Rees RW, Robin A. Advice given by health food shops: is it clinically safe? J R Coll Physicians Lond 1998;32(5):426–8.
  8. Sweetman S, ed. Martindale: The Complete Drug Reference. 36 ed. 2009, Pharmaceutical Press: London.
  9. Williamson E, Driver E, Baxter K. Stockley's Herbal Medicine Interactions. In: A guide to the interactions of herbal medicines, dietary supplements and nutraceuticals with conventional medicines. 2009, Pharmaceutical Press.
  10. Agresti A. Categorical data analysis (second edition). John Wiley, Hoboken 2002:644.
  11. Healey B, Burgess C, Siebers R, et al. Do natural health food stores require regulation? N Z Med J 2002;115(1161).http://www.nzma.org.nz/journal/115-1161/ .
  12. Siebers R, Holt S, Healy B, et al. High blood pressure advice given by natural health food stores. N Z Med J 2009;122 (1293).http://www.nzma.org.nz/journal/122-1293/3566/
  13. Shalansky S, Lynd L, Richardson K, et al. Risk of warfarin-related bleeding events and supratherapeutic international normalized ratios associated with complementary and alternative medicine: a longitudinal analysis. Pharmacotherapy 2007; 27: 1237–1247.
  14. Leung V, Shalansky S, Lo M, Jadusingh E. Prevalence of use and risk of adverse effects associated with complementary and alternative medicine in a cohort of patients receiving warfarin. Ann Pharmacother 2009;43:875–81.
  15. WHO Congress on Traditional Medicine and the Beijing Declaration.www.who.int/medicines/areas/traditional/congress/beijing_declaration/ Last accessed on November 23, 2010.
  16. Ministerial Advisory Committee on Complementary and Alternative Health: Advice to the Minister of Health. June 2004.http://www.newhealth.govt.nz/maccah.htm Last accessed November 23, 2010.
  17. Ministry of Health. The development of a Natural Health Products Bill: consultation paper. 2010.http://www.moh.govt.nz/moh.nsf/indexmh/consult-development-natural-health-products-bill-mar10 Last accessed November 23, 2010.
  18. Heck A, DeWitt B, Lukes A. Potential interactions between alternative therapies and warfarin. Am J Health-Syst Pharm 2000;57:1221–1227.