REVIEW
Online Sources of Herbal Product Information Christopher Owens, PharmD, MPH,a Ralph Baergen, PhD, MPH,b Derek Puckettc a
Department of Pharmacy Practice and Administrative Sciences at Idaho State University College of Pharmacy, Pocatello; Department of Philosophy, Idaho State University, Pocatello; cIdaho State University, Pocatello.
b
ABSTRACT Herbal products are commonly used to treat clinical conditions and are often purchased online without the supervision of a healthcare provider. The use of herbals remains controversial because of widespread exaggerated claims of clinical efficacy and safety. We conducted an online search of 13 common herbals (including black cohosh, echinacea, garlic, ginkgo, ginseng, green tea, kava, saw palmetto, and St John’s wort) and reviewed the top 50 Web sites for each using a Google search. We analyzed clinical claims, warnings, and other safety information. A total of 1179 Web sites were examined. Less than 8% of retail sites provided information regarding potential adverse effects, drug interactions, and other safety information; only 10.5% recommended consultation with a healthcare professional. Less than 3% cited scientific literature to accompany their claims. Key safety information is still lacking from many online sources of herbal information. Certain nonretail site types may be more reliable, but physicians and other healthcare professionals should be aware of the variable quality of these sites to help patients make more informed decisions. Published by Elsevier Inc. The American Journal of Medicine (2014) 127, 109-115 KEYWORDS: Complementary and alternative medicine; Dietary supplements; Herbal medicine; Internet; Medicinal plants; Online information sources
Herbal products are a popular form of complementary and alternative medicine used by millions of Americans to treat self-diagnosed clinical conditions or to promote health.1 They are widely marketed and readily available without a prescription in pharmacies, drugstores, and other retail outlets. They also are available for purchase online via Internet retailers. Although some products have been studied in controlled clinical trials, for many, the evidence regarding efficacy and safety is insufficient to provide appropriate clinical guidance.2,3 Complicating the matter is the mistaken notion on the part of many consumers that herbal products are safer than conventional pharmaceuticals, have few or no side effects, and have little potential for interactions with other herbals or medications.4 Funding: DP’s participation was funded in part by a Career Path Internship Grant from Idaho State University. Conflict of Interest: None. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Christopher Owens, PharmD, MPH, College of Pharmacy, Idaho State University, 970 South, 8th Ave, Pocatello, ID 83209-8333. E-mail address: [email protected]
0002-9343/$ -see front matter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.amjmed.2013.09.016
Herbal products have been classified as dietary supplements since the passage of the Dietary Supplement Health and Education Act of 1994. A dietary supplement is defined as a nonfood, nondrug product taken by mouth and containing at least 1 identified dietary ingredient, such as a vitamin, mineral, herb or botanical, amino acid, enzyme, or metabolite.5,6 Product labeling for dietary supplements, which includes medical claims and safety information, is regulated less stringently than for drugs. Labeling includes both the information on a product’s packaging itself and any information used in its advertisement.7 Specific disease claims are prohibited. For example, a supplement may not explicitly claim to prevent urinary tract infections; however, a structure-function claim such as “helps maintain a healthy urinary system” is allowable. The Food and Drug Administration (FDA) has established criteria by which disease claims are differentiated from structure-function claims.8 A disclaimer by the FDA also must be included on every dietary supplement label making clear that the product “is not intended to diagnose, treat, prevent, or cure any disease.”5 Enforcement of labeling requirements is an ongoing issue related to FDA manpower and resource limitations.9 The growing use of herbal products and related expenditures by the American public has been the focus of much
110
The American Journal of Medicine, Vol 127, No 2, February 2014
study10,11; the most current and comprehensive assessment The general search was used to evaluate information on of Americans’ use of these products and practices found that nonretail sites and the shopping search to evaluate inforapproximately 40% of adult Americans reported using some mation on retail sites. form of complementary and alternative medicine in the Our analysis included the most commonly used herbal previous year for health-related reasons.12 The compleproducts in the United States, according to the most recent mentary and alternative medicine industry occupied a nationwide survey12: cranberry, echinacea, flaxseed oil, substantial portion of the health ginseng, ginkgo biloba, garlic, market by 2007, with Americans green tea, grapeseed extract, saw CLINICAL SIGNIFICANCE spending approximately $34 palmetto, and soy. We also billion out-of-pocket for these included 3 top-selling herbal Fewer than 1 in 10 retail Web sites therapies. Dietary supplements products with known safety concontain information regarding herbal represented approximately $18 cerns: black cohosh, kava, and products’ potential for adverse effects, billion of this overall expendiSt John’s wort. The herbal product drug interactions, or other basic safety ture, with $4 billion being spent name was used as the Google precautions. on herbal products.13 search term, plus the word “supplement” in the case of cranberry, Given the widespread use of Nonretail Web sites, in general, have garlic, and green tea to differenthese products without medical more reliable information and more tiate culinary uses of these prodsupervision and the less stringent often include citations from medical ucts from those specifically being labeling requirements, concern literature; however, certain types of sought out for their purported exists on the part of many in nonretail sites are preferable. health-related benefits. the medical community relating Physicians must be aware of the variable Web site content was recorded to the adequacy of protection for on a standardized data-collection consumers of dietary supplequality of different sources of online spreadsheet. For the retail search, ments, especially as it relates to herbal information and help patients we noted whether the site included the validity of the claims made by make informed decisions correct plant species name, statemanufacturers and retailers.14,15 ments regarding product stanSeveral studies have shown that dardization, the FDA disclaimer, a list of potential adverse much of the label information is misleading or inaccurate effects, medical contraindications, safety of use in certain and that key safety information is often omitted.16-19 circumstances (pregnancy, lactation, pediatrics), known inThe frequent use of the Internet has increased the concern teractions, and a recommendation to consult with a healthwith respect to the potential for inadequate labeling of care professional before use. The presence of references or dietary supplements, including exaggerated reports of efficitations of medical or scientific literature also was noted; cacy and minimal safety warnings. The Pew Research such sources did not need to be peer-reviewed. For the Center recently reported that 79% of all American adults use nonretail search, the safety information assessed was the Internet, and this number has remained relatively consimilar; we also assessed whether a mechanism of therastant since 2006. Of these, approximately 83% report using peutic action was included, as well as recommended dosage, the Internet to look for health-related information, including and whether any indication of efficacy was mentioned. for dietary supplements.20 We also categorized the nonretail sites according to their By building on the body of published literature regarding affiliation with government, consumer group, educational Internet sources of information for herbal products,16-19,21-24 institution, or “enthusiast,” including those administered by the objective of the current study was to examine the state of individual laypersons or nonmedical organizations of herbal online information with respect to top-selling herbal produsers. An “other” category included news outlets, wiki sites, ucts, now a decade after the last large systematic review.16 and references to articles in medical journals. By using online searches that are typical of today’s conWeb site content was analyzed in comparison with sumer for herbal products and herbal product information, publicly available, authoritative safety and efficacy inforwe conducted the largest Internet search of herbal products mation included on the Herbs-at-a-Glance monographs to date and included the 10 most commonly used herbal available from the National Center for Complementary and products, as well as 3 additional top-selling herbals with Alternative Medicine (NCCAM) Web site, (http://nccam. known safety concerns. nih.gov). These monographs include basic information, such as plant species name, recognized and purported clinMATERIALS AND METHODS ical uses, therapeutic efficacy according clinical trial data, important drug interactions, and other safety precautions. We used the most commonly used Internet search engine, A monograph is available for each of the herbal products Google (www.google.com), and conducted 2 different included in this review. searches (general and shopping) to analyze Web site content Web site search results from the top 50 hits of each that would likely be encountered by the typical consumer product’s general and shopping searches were evaluated. going online to learn about or purchase an herbal product.
Owens et al
Online Sources of Herbal Information
Suggested pages or advertisements by Google were disregarded; only the first 50 genuine search sites or “hits” were examined. A research assistant (DP) coded the information on the data-collection sheets for retail sites from August to November 2012. The nonretail Web site review was conducted from November 2012 to January 2013. The pharmacist reviewer (CO) made all final categorization decisions.
RESULTS A total of 1300 English-language Web sites were identified for initial analysis, 650 nonretail sites from the general search and 650 retail sites from the shopper search. Of these, 121 were excluded from the general search because they were determined to be retail sites whose sole purpose was to sell product, leaving a total of 1179 Web sites for review. For the 650 retail sites, all directly sold at least 1 herbal product or were linked to a vendor that did. Of these, 394 (60.6%) made mention of a standardized extract or included information regarding the standardization process for the active ingredient(s). A total of 90 (13.8%) retail Web sites included information that violated established FDA criteria by making claims regarding diagnosing, treating, preventing, or curing a disease. Some claims were clear assertions that the herbal product was useful in the treatment of a specific disease, including the name of the medical condition. Sites selling soy, black cohosh, and green tea extract containing products made disease claims most often (38%, 22%, and 20%, respectively). Examples of sample disease claims and patient testimonials that could be considered disease claims are shown in Figure 1. The required FDA disclaimer was included on 54 retail sites (8.3%). Adverse effect and drug interaction information were rarely provided; specific safety information related to the use of products by pregnant or lactating individuals was included on 46 retail sites (7%), and cautions related to use in pediatric patients was found on 40 sites (6%). Only 68 Web sites recommended that consumers consult a physician or pharmacist before using the product (10.5%), and 18 retail Web sites (3%) included references to scientific literature. Of the 650 nonretail Web sites initially identified from the general search, 121 were excluded from the analysis because they directly sold products and were therefore determined to be retail sites. The remaining 529 sites were considered informational and categorized by type (Table 1). Most nonretail sites (283 or 53.5%) were administered by alternative medicine organizations or individual enthusiasts (eg, Livestrong, Soylife, herbs.org). These sites included chat forums and appeared to provide information meant for consumers of herbal products by consumers of these products. The next largest categories of Web sites were those directed at consumers of medical information (eg, WebMD, Rx List, Drugs.com) and general consumer sites (About.com, Consumer Reports); these totaled 88 (16.6%) and 49 (9.3%) of sites, respectively. Web sites affiliated with universities or educational medical centers (eg, Mayo
111 Clinic) totaled 33 sites (6.2%) and government-operated Web sites (eg, NCCAM) totaled 30 sites (5.7%). The remaining 46 Web sites (8.7%) were simple reference sites (Dictionary.com, medical dictionaries, Wikipedia), news outlets, or associated with medical journals. Inclusion of adverse effects and drug interaction information was significantly more common on nonretail sites and was reported on approximately half of all nonretail sites reviewed, 251 sites (47.4%) and 259 sites (48.9%), respectively (Figure 2). Precaution information related to safety in pregnancy/lactation or use in pediatric patients also was more common on nonretail sites, 206 (38.9%) and 100 (18.9%), respectively. A total of 302 nonretail sites (57.2%) included statements of recommendations that individuals consult with a healthcare professional before using the product. A total of 172 nonretail sites (32.5%) also included references to the medical literature or citations of other authoritative information. All of these were significantly more common on nonretail versus retail Web sites (P < .001). Thirteen nonretail sites (2.5%) included explicit statements denying any adverse or side effects whatsoever. Nonretail sites provided a purported mechanism of action of the active ingredient(s) for the herbal product on 129 Web sites (24%), listed at least 1 appropriate therapeutic use on 499 sites (94.3%), and included appropriate dosing information on 262 Web sites (49.5%). Descriptions of therapeutic efficacy (or inefficacy) were included on most nonretail sites, with reports of positive efficacy being included on 264 Web sites (49.9%), mixed efficacy (where both positive and negative statements were included) was reported on 190 Web sites (35.9%), and negative efficacy was reported on 20 Web sites (3.7%). The information related to the 3 herbal products with specific safety warnings (black cohosh, kava, and St John’s wort) was compared between retail and nonretail sites. Nonretail sites reported adverse effects, drug interactions, and general warnings significantly more often than retail sites. Nonretail sites also were more likely to recommend consulting with a healthcare provider. (Table 2).
DISCUSSION This review of online herbal product information is a significant update from previous systematic analyses in several ways. It includes the most popular herbal products used by Americans in the most recent nationwide survey and demonstrated that many of the concerns raised by past reviews are still important issues that need to be addressed by healthcare providers and regulatory agencies. Its use of the most popular Internet search engine in 2 ways mimics the typical search pattern of the average consumer of both herbal information and herbal products for purchase and use. Retail sites continue to make illegal disease claims, but do so less commonly than 10 years ago. Our review found that just more than 13% of retail sites make such claims compared with 55% of sites in the analysis by Morris and
112
The American Journal of Medicine, Vol 127, No 2, February 2014
Figure 1
Sample of Disease Claims from Retail Web Sites.
Avorn.16 This is encouraging if it means that the FDA criteria related to disease claims versus structure-function claims are being more consistently observed on the part of retailers. Other reasons for this difference may be related to the subjective nature of such an analysis and could be due to our application of the criteria to the Web site content. Our approach was to attempt to read the site information as a lay consumer would and to include only those
statements as disease claims if they included a specific disease or diagnosable medical condition name (eg, cancer, depression, menopause) or list of clinical signs or symptoms associated with a specific disease or medical condition (eg, tumor, hot flashes) that would be recognizable by the typical consumer. This was difficult in that consumers reading the information included on Web sites may have vastly different levels of medical literacy.
Owens et al Table 1
Online Sources of Herbal Information
113
Nonretail Web Sites Analyzed by Type Nonretail Web Site Type
Black cohosh Cranberry Echinacea Flaxseed Garlic Gingko Ginseng Grapeseed Green tea Kava Saw palmetto Soy isoflavones St John’s wort
University
Government
Consumer
Enthusiast
Other
Total No. Sites
4 1 4 2 2 4 4 1 0 2 2 2 5
3 2 2 2 2 2 5 1 1 3 3 0 4
15 10 11 17 10 15 7 9 4 7 14 2 16
19 16 25 19 28 24 27 24 24 26 16 24 11
6 2 4 3 0 3 5 3 6 3 2 6 3 Total
47 31 46 43 42 48 48 38 35 41 37 34 39 529
Nonretail sites were categorized as university if they were affiliated with an educational institution or educational medical center with address ending in .edu, government Web sites ended in .gov, consumer sites included those for general consumers (eg, About.com) and medical consumers (drugs.com, WebMD), enthusiast Web sites were administered by lay persons or nonmedical organizations, and other included all other general information sites (eg, Wikipedia).
Of particular interest was the inclusion of patient comments and testimonials on many retail sites and some nonretail sites administered by enthusiasts. Such comments may be viewed as making a disease claim and may be viewed as claims made by Web site administrators, even though the words are not their own. Such claims likewise deserve careful scrutiny because they also may affect consumer behaviors and beliefs. The inclusion of this type of information constitutes at least a tacit endorsement on the part of the Web site administrator and is further cause for concern related to the type of information available on the Internet being used by patients to make health decisions. Disease claims were most commonly made for soy, black cohosh, and green tea extract. These products are commonly used for menopausal symptoms (soy and black cohosh) and weight loss or cancer (green tea). In these cases, structurefunction claims that describe the purported health benefit
Figure 2 Safety content comparison between retail and nonretail sites.
without explicitly mentioning a disease or diagnosable medical condition are difficult to make. Therefore, it appeared that these Web sites used language that necessarily included references to the medical conditions of perimenopause, obesity, or cancer. Other herbal supplements can more easily make legal structure-function claims by clothing their health statements in phrases that are technically legal, but still lead individuals to believe and expect health benefits related to recognizable medical conditions. Even if outright disease claims are made less frequently for most of these products, important safety information is still often omitted, and when combined with the unsubstantiated structure/function health claims that are made, this presents an ongoing danger to patients who may purchase and use these products in combination with their prescribed medications and without the knowledge of their healthcare providers. The known safety concerns of black cohosh and kava (hepatotoxicity) and St John’s wort (drug interactions) are especially problematic because retail sites only rarely include any information to warn patients of these potential issues. Although retail sites do often include information such as the scientific name of the plant species and the amount of standardized ingredient per pill, this information may only serve to lull a consumer into a false sense that the product is safe and effective, given the authoritative way the product is described and labeled. The manufacturers and vendors of these products know when and how to use science and scientificsounding jargon as marketing tools, but when combined with the omission of key safety information, this only increases the likelihood of consumer use and potential harm. Not all herbal information on the Internet is misleading. Nonretail sites are frequently found when conducting an online general search for information on herbal products. Many of these sites contain authoritative information that
114
The American Journal of Medicine, Vol 127, No 2, February 2014
Table 2
Comparison of Web Site Information on Products with Known Safety Warnings
Adverse effects Interactions Warnings Recommendation to discuss with a provider
Black Cohosh
Kava
St John’s Wort
Retail/Nonretail
Retail/Nonretail
Retail/Nonretail
0%/74% 4%/40% 4%/68% 8%/72%
0%/51% 12/59% 20%/54% 20%/56%
0%/74% 6%/67% 2%/67% 6%/72%
All differences were statistically significant (P < .001).
can help consumers make more informed decisions about the use of these products. A review of the most popular hits on a general search will include information from the NCCAM, the Mayo Clinic, and other medical- or educational institutionesupported sources that more often provide balanced information with respect to efficacy and safety. There also is often a recommendation on these Web sites for individuals to consult their healthcare providers before using a supplement. However, it should be noted that there also are many sites that will be found in a general search that are administered by amateur herbalists and enthusiastic proponents of these products. Such sites may appear to be or are portrayed as “medical information” but in fact are maintained by individuals whose credentials are questionable or absent. Taken together, however, the information contained on many nonretail Web sites is more often referenced and at a minimum contains the warnings and safety precautions for special populations (pregnant patients, children) that afford some consumer protection, when compared with information on retail sites. Although there is good news with respect to the amount and type of herbal product information available on the Internet, the proliferation of herbal retailers occurs at an ever-increasing rate. Policing the Internet is an impossible task because of ongoing shortages of federal manpower and resources to review and take action against inappropriate or harmful information on Web sites.14,25 It therefore becomes the purview of healthcare professionals to take on the vital role of educating patients regarding valid sources of information and how to appraise the information they do find on the Internet critically. The NCCAM’s “Time to Talk” campaign is an important step that helps to provide tools to facilitate conversation about the use of complementary and alternative medicine between health care providers and patients.1 National efforts like this must be coupled with local educational activities to reach additional audiences— including alternative medical providers and the entire diverse community of consumers—to ensure that herbal products are incorporated appropriately into the overall healthcare landscape.26
CONCLUSIONS Although there are benefits to the use of herbal products in certain clinical conditions, until efficacy and safety data are
better established and information sources accurately reflect what is known about these products and their place in therapy, it is the responsibility of government agencies and healthcare professionals to take an even greater role in the critique and dissemination of information that is available to individuals who may be considering the use of such products for their health.
References 1. National Center for Complementary and Alternative Medicine. NCCAM Third Strategic Plan: 2011-2015. National Center for Complementary and Alternative Medicine. Available at: http://nccam.nih. gov/about/plans/2011?nav¼gsa. Accessed March 27, 2013. 2. Fontanarosa PB, Rennie D, DeAngelis CD. The need for regulation of dietary supplements: lessons from ephedra. JAMA. 2003;289: 1568-1570. 3. Wolsko PM, Solondz DK, Phillips RS, Schachter SC, Eisenberg DM. Lack of herbal supplement characterization in published randomized controlled trials. Am J Med. 2005;118:1087-1093. 4. Ernst E. Harmless herbs? A review of the recent literature. Am J Med. 1998;104:170-178. 5. Office of Dietary Supplements. Dietary Supplement Health and Education Act of 1994. Office of Dietary Supplements. Available at: http://ods. od.nih.gov/About/DSHEA_Wording.aspx. Accessed April 15, 2013. 6. Dietary Supplement Health and Education Act. Pub L No. 103-417, 108 Stat 4325 (codified at 21 USC § 301 [1994]). 7. Harris IM. Regulatory and ethical issues with dietary supplements. Pharmacotherapy. 2000;20:1295-1302. 8. US Department of Health and Human Services, US Food and Drug Administration. Regulations on statements made for dietary supplements concerning the effect of the product on the structure or function of the body; final rule, 65 Federal Register 1000(2000) (codified at 12 CFR §101). 9. Crawford SY, Leventis C. Herbal product claims: boundaries of marketing and science. J Consum Marketing. 2005;22:432-436. 10. Eisenberg DM, Kessler RC, Foster C, Norlock FE, Calkins DR, Delbanco TL. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N Engl J Med. 1993;328:246-252. 11. Davis RB, Eisenberg DM, Tindle HA. Trends in use of complementary and alternative medicine by US adults: 1997-2002. Altern Ther Health Med. 2005;11:42-49. 12. Barnes PM, Bloom B, and Nahin R. Complementary and alternative medicine use among adults and children, United States, 2007. National Center for Health Statistics. http://nccam.nih.gov/sites/nccam.nih.gov/ files/news/nhsr12.pdf. Published December 10, 2008. Accessed June 30, 2013. 13. National Center for Complementary and Alternative Medicine. Statistics on CAM Costs. National Center for Complementary and Alternative Medicine. Available at: http://nccam.nih.gov/news/camstats/costs? nav¼gsa. Accessed March 20, 2013. 14. DeAngelis CD, Fontanarosa PB. Drugs alias dietary supplements. JAMA. 2003;11:1519-1520.
Owens et al
Online Sources of Herbal Information
15. Temple NJ. The marketing of dietary supplements in North America: the emperor is (almost) naked. J Alt Complement Med. 2010;16:803-806. 16. Morris CA, Avorn J. Internet marketing of herbal products. JAMA. 2003;11:1505-1509. 17. Jordan MA, Haywood T. Evaluation of internet websites marketing herbal weight-loss supplements to consumers. J Alt Complement Med. 2007;13:1035-1043. 18. Walji M, Sagaram S, Sagaram D, et al. Efficacy of quality criteria to identify potentially harmful information: a cross-sectional survey of complementary and alternative medicine web sites. J Med Internet Res. 2004;6:e21. 19. Bonakdar RA. Herbal cancer cures on the Web: noncompliance with Dietary Supplement Health and Education Act. Fam Med. 2002;34: 522-527. 20. Zickuhr K. Generations 2010/Pew Research Center’s Internet and American Life Project. Pew Research Center. Available at: http://www. pewinternet.org/Reports/2010/Generations-2010.aspx. Accessed April 24, 2013.
115 21. Ashar BH, Miller RG, Getz KJ, Pichard CP. A critical evaluation of Internet marking of products that contain ephedra. Mayo Clin Proc. 2003;78:944-946. 22. Martin-Facklam M, Kostrzewa M, Schubert F, Gasse C, Haefeli W. Quality markers of drug information on the Internet: an evaluation of sites about St. John’s wort. Am J Med. 2002;113: 740-745. 23. Nazeri A, Massumi A, Wilson JM, et al. Arrhythmogenicity of weightloss supplements marketed on the Internet. Heart Rhythm. 2009;6: 658-662. 24. Thurairaja R, Barrass B, Persad R. Internet websites selling herbal treatments for erectile dysfunction. Int J Impot Res. 2005;17: 196-200. 25. Bent S, Ko R. Commonly used herbal medicines in the United States: a review. Am J Med. 2004;116:478-485. 26. Bodeker G, Kronenberg F. A public health agenda for traditional, complementary, and alternative medicine. Am J Public Health. 2002;92:1582-1591.
Online Sources of Herbal Product Information Christopher Owens, PharmD, MPH,a Ralph Baergen, PhD, MPH,b Derek Puckettc a
Department of Pharmacy Practice and Administrative Sciences at Idaho State University College of Pharmacy, Pocatello; Department of Philosophy, Idaho State University, Pocatello; cIdaho State University, Pocatello.
b
ABSTRACT Herbal products are commonly used to treat clinical conditions and are often purchased online without the supervision of a healthcare provider. The use of herbals remains controversial because of widespread exaggerated claims of clinical efficacy and safety. We conducted an online search of 13 common herbals (including black cohosh, echinacea, garlic, ginkgo, ginseng, green tea, kava, saw palmetto, and St John’s wort) and reviewed the top 50 Web sites for each using a Google search. We analyzed clinical claims, warnings, and other safety information. A total of 1179 Web sites were examined. Less than 8% of retail sites provided information regarding potential adverse effects, drug interactions, and other safety information; only 10.5% recommended consultation with a healthcare professional. Less than 3% cited scientific literature to accompany their claims. Key safety information is still lacking from many online sources of herbal information. Certain nonretail site types may be more reliable, but physicians and other healthcare professionals should be aware of the variable quality of these sites to help patients make more informed decisions. Published by Elsevier Inc. The American Journal of Medicine (2014) 127, 109-115 KEYWORDS: Complementary and alternative medicine; Dietary supplements; Herbal medicine; Internet; Medicinal plants; Online information sources
Herbal products are a popular form of complementary and alternative medicine used by millions of Americans to treat self-diagnosed clinical conditions or to promote health.1 They are widely marketed and readily available without a prescription in pharmacies, drugstores, and other retail outlets. They also are available for purchase online via Internet retailers. Although some products have been studied in controlled clinical trials, for many, the evidence regarding efficacy and safety is insufficient to provide appropriate clinical guidance.2,3 Complicating the matter is the mistaken notion on the part of many consumers that herbal products are safer than conventional pharmaceuticals, have few or no side effects, and have little potential for interactions with other herbals or medications.4 Funding: DP’s participation was funded in part by a Career Path Internship Grant from Idaho State University. Conflict of Interest: None. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Christopher Owens, PharmD, MPH, College of Pharmacy, Idaho State University, 970 South, 8th Ave, Pocatello, ID 83209-8333. E-mail address: [email protected]
0002-9343/$ -see front matter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.amjmed.2013.09.016
Herbal products have been classified as dietary supplements since the passage of the Dietary Supplement Health and Education Act of 1994. A dietary supplement is defined as a nonfood, nondrug product taken by mouth and containing at least 1 identified dietary ingredient, such as a vitamin, mineral, herb or botanical, amino acid, enzyme, or metabolite.5,6 Product labeling for dietary supplements, which includes medical claims and safety information, is regulated less stringently than for drugs. Labeling includes both the information on a product’s packaging itself and any information used in its advertisement.7 Specific disease claims are prohibited. For example, a supplement may not explicitly claim to prevent urinary tract infections; however, a structure-function claim such as “helps maintain a healthy urinary system” is allowable. The Food and Drug Administration (FDA) has established criteria by which disease claims are differentiated from structure-function claims.8 A disclaimer by the FDA also must be included on every dietary supplement label making clear that the product “is not intended to diagnose, treat, prevent, or cure any disease.”5 Enforcement of labeling requirements is an ongoing issue related to FDA manpower and resource limitations.9 The growing use of herbal products and related expenditures by the American public has been the focus of much
110
The American Journal of Medicine, Vol 127, No 2, February 2014
study10,11; the most current and comprehensive assessment The general search was used to evaluate information on of Americans’ use of these products and practices found that nonretail sites and the shopping search to evaluate inforapproximately 40% of adult Americans reported using some mation on retail sites. form of complementary and alternative medicine in the Our analysis included the most commonly used herbal previous year for health-related reasons.12 The compleproducts in the United States, according to the most recent mentary and alternative medicine industry occupied a nationwide survey12: cranberry, echinacea, flaxseed oil, substantial portion of the health ginseng, ginkgo biloba, garlic, market by 2007, with Americans green tea, grapeseed extract, saw CLINICAL SIGNIFICANCE spending approximately $34 palmetto, and soy. We also billion out-of-pocket for these included 3 top-selling herbal Fewer than 1 in 10 retail Web sites therapies. Dietary supplements products with known safety concontain information regarding herbal represented approximately $18 cerns: black cohosh, kava, and products’ potential for adverse effects, billion of this overall expendiSt John’s wort. The herbal product drug interactions, or other basic safety ture, with $4 billion being spent name was used as the Google precautions. on herbal products.13 search term, plus the word “supplement” in the case of cranberry, Given the widespread use of Nonretail Web sites, in general, have garlic, and green tea to differenthese products without medical more reliable information and more tiate culinary uses of these prodsupervision and the less stringent often include citations from medical ucts from those specifically being labeling requirements, concern literature; however, certain types of sought out for their purported exists on the part of many in nonretail sites are preferable. health-related benefits. the medical community relating Physicians must be aware of the variable Web site content was recorded to the adequacy of protection for on a standardized data-collection consumers of dietary supplequality of different sources of online spreadsheet. For the retail search, ments, especially as it relates to herbal information and help patients we noted whether the site included the validity of the claims made by make informed decisions correct plant species name, statemanufacturers and retailers.14,15 ments regarding product stanSeveral studies have shown that dardization, the FDA disclaimer, a list of potential adverse much of the label information is misleading or inaccurate effects, medical contraindications, safety of use in certain and that key safety information is often omitted.16-19 circumstances (pregnancy, lactation, pediatrics), known inThe frequent use of the Internet has increased the concern teractions, and a recommendation to consult with a healthwith respect to the potential for inadequate labeling of care professional before use. The presence of references or dietary supplements, including exaggerated reports of efficitations of medical or scientific literature also was noted; cacy and minimal safety warnings. The Pew Research such sources did not need to be peer-reviewed. For the Center recently reported that 79% of all American adults use nonretail search, the safety information assessed was the Internet, and this number has remained relatively consimilar; we also assessed whether a mechanism of therastant since 2006. Of these, approximately 83% report using peutic action was included, as well as recommended dosage, the Internet to look for health-related information, including and whether any indication of efficacy was mentioned. for dietary supplements.20 We also categorized the nonretail sites according to their By building on the body of published literature regarding affiliation with government, consumer group, educational Internet sources of information for herbal products,16-19,21-24 institution, or “enthusiast,” including those administered by the objective of the current study was to examine the state of individual laypersons or nonmedical organizations of herbal online information with respect to top-selling herbal produsers. An “other” category included news outlets, wiki sites, ucts, now a decade after the last large systematic review.16 and references to articles in medical journals. By using online searches that are typical of today’s conWeb site content was analyzed in comparison with sumer for herbal products and herbal product information, publicly available, authoritative safety and efficacy inforwe conducted the largest Internet search of herbal products mation included on the Herbs-at-a-Glance monographs to date and included the 10 most commonly used herbal available from the National Center for Complementary and products, as well as 3 additional top-selling herbals with Alternative Medicine (NCCAM) Web site, (http://nccam. known safety concerns. nih.gov). These monographs include basic information, such as plant species name, recognized and purported clinMATERIALS AND METHODS ical uses, therapeutic efficacy according clinical trial data, important drug interactions, and other safety precautions. We used the most commonly used Internet search engine, A monograph is available for each of the herbal products Google (www.google.com), and conducted 2 different included in this review. searches (general and shopping) to analyze Web site content Web site search results from the top 50 hits of each that would likely be encountered by the typical consumer product’s general and shopping searches were evaluated. going online to learn about or purchase an herbal product.
Owens et al
Online Sources of Herbal Information
Suggested pages or advertisements by Google were disregarded; only the first 50 genuine search sites or “hits” were examined. A research assistant (DP) coded the information on the data-collection sheets for retail sites from August to November 2012. The nonretail Web site review was conducted from November 2012 to January 2013. The pharmacist reviewer (CO) made all final categorization decisions.
RESULTS A total of 1300 English-language Web sites were identified for initial analysis, 650 nonretail sites from the general search and 650 retail sites from the shopper search. Of these, 121 were excluded from the general search because they were determined to be retail sites whose sole purpose was to sell product, leaving a total of 1179 Web sites for review. For the 650 retail sites, all directly sold at least 1 herbal product or were linked to a vendor that did. Of these, 394 (60.6%) made mention of a standardized extract or included information regarding the standardization process for the active ingredient(s). A total of 90 (13.8%) retail Web sites included information that violated established FDA criteria by making claims regarding diagnosing, treating, preventing, or curing a disease. Some claims were clear assertions that the herbal product was useful in the treatment of a specific disease, including the name of the medical condition. Sites selling soy, black cohosh, and green tea extract containing products made disease claims most often (38%, 22%, and 20%, respectively). Examples of sample disease claims and patient testimonials that could be considered disease claims are shown in Figure 1. The required FDA disclaimer was included on 54 retail sites (8.3%). Adverse effect and drug interaction information were rarely provided; specific safety information related to the use of products by pregnant or lactating individuals was included on 46 retail sites (7%), and cautions related to use in pediatric patients was found on 40 sites (6%). Only 68 Web sites recommended that consumers consult a physician or pharmacist before using the product (10.5%), and 18 retail Web sites (3%) included references to scientific literature. Of the 650 nonretail Web sites initially identified from the general search, 121 were excluded from the analysis because they directly sold products and were therefore determined to be retail sites. The remaining 529 sites were considered informational and categorized by type (Table 1). Most nonretail sites (283 or 53.5%) were administered by alternative medicine organizations or individual enthusiasts (eg, Livestrong, Soylife, herbs.org). These sites included chat forums and appeared to provide information meant for consumers of herbal products by consumers of these products. The next largest categories of Web sites were those directed at consumers of medical information (eg, WebMD, Rx List, Drugs.com) and general consumer sites (About.com, Consumer Reports); these totaled 88 (16.6%) and 49 (9.3%) of sites, respectively. Web sites affiliated with universities or educational medical centers (eg, Mayo
111 Clinic) totaled 33 sites (6.2%) and government-operated Web sites (eg, NCCAM) totaled 30 sites (5.7%). The remaining 46 Web sites (8.7%) were simple reference sites (Dictionary.com, medical dictionaries, Wikipedia), news outlets, or associated with medical journals. Inclusion of adverse effects and drug interaction information was significantly more common on nonretail sites and was reported on approximately half of all nonretail sites reviewed, 251 sites (47.4%) and 259 sites (48.9%), respectively (Figure 2). Precaution information related to safety in pregnancy/lactation or use in pediatric patients also was more common on nonretail sites, 206 (38.9%) and 100 (18.9%), respectively. A total of 302 nonretail sites (57.2%) included statements of recommendations that individuals consult with a healthcare professional before using the product. A total of 172 nonretail sites (32.5%) also included references to the medical literature or citations of other authoritative information. All of these were significantly more common on nonretail versus retail Web sites (P < .001). Thirteen nonretail sites (2.5%) included explicit statements denying any adverse or side effects whatsoever. Nonretail sites provided a purported mechanism of action of the active ingredient(s) for the herbal product on 129 Web sites (24%), listed at least 1 appropriate therapeutic use on 499 sites (94.3%), and included appropriate dosing information on 262 Web sites (49.5%). Descriptions of therapeutic efficacy (or inefficacy) were included on most nonretail sites, with reports of positive efficacy being included on 264 Web sites (49.9%), mixed efficacy (where both positive and negative statements were included) was reported on 190 Web sites (35.9%), and negative efficacy was reported on 20 Web sites (3.7%). The information related to the 3 herbal products with specific safety warnings (black cohosh, kava, and St John’s wort) was compared between retail and nonretail sites. Nonretail sites reported adverse effects, drug interactions, and general warnings significantly more often than retail sites. Nonretail sites also were more likely to recommend consulting with a healthcare provider. (Table 2).
DISCUSSION This review of online herbal product information is a significant update from previous systematic analyses in several ways. It includes the most popular herbal products used by Americans in the most recent nationwide survey and demonstrated that many of the concerns raised by past reviews are still important issues that need to be addressed by healthcare providers and regulatory agencies. Its use of the most popular Internet search engine in 2 ways mimics the typical search pattern of the average consumer of both herbal information and herbal products for purchase and use. Retail sites continue to make illegal disease claims, but do so less commonly than 10 years ago. Our review found that just more than 13% of retail sites make such claims compared with 55% of sites in the analysis by Morris and
112
The American Journal of Medicine, Vol 127, No 2, February 2014
Figure 1
Sample of Disease Claims from Retail Web Sites.
Avorn.16 This is encouraging if it means that the FDA criteria related to disease claims versus structure-function claims are being more consistently observed on the part of retailers. Other reasons for this difference may be related to the subjective nature of such an analysis and could be due to our application of the criteria to the Web site content. Our approach was to attempt to read the site information as a lay consumer would and to include only those
statements as disease claims if they included a specific disease or diagnosable medical condition name (eg, cancer, depression, menopause) or list of clinical signs or symptoms associated with a specific disease or medical condition (eg, tumor, hot flashes) that would be recognizable by the typical consumer. This was difficult in that consumers reading the information included on Web sites may have vastly different levels of medical literacy.
Owens et al Table 1
Online Sources of Herbal Information
113
Nonretail Web Sites Analyzed by Type Nonretail Web Site Type
Black cohosh Cranberry Echinacea Flaxseed Garlic Gingko Ginseng Grapeseed Green tea Kava Saw palmetto Soy isoflavones St John’s wort
University
Government
Consumer
Enthusiast
Other
Total No. Sites
4 1 4 2 2 4 4 1 0 2 2 2 5
3 2 2 2 2 2 5 1 1 3 3 0 4
15 10 11 17 10 15 7 9 4 7 14 2 16
19 16 25 19 28 24 27 24 24 26 16 24 11
6 2 4 3 0 3 5 3 6 3 2 6 3 Total
47 31 46 43 42 48 48 38 35 41 37 34 39 529
Nonretail sites were categorized as university if they were affiliated with an educational institution or educational medical center with address ending in .edu, government Web sites ended in .gov, consumer sites included those for general consumers (eg, About.com) and medical consumers (drugs.com, WebMD), enthusiast Web sites were administered by lay persons or nonmedical organizations, and other included all other general information sites (eg, Wikipedia).
Of particular interest was the inclusion of patient comments and testimonials on many retail sites and some nonretail sites administered by enthusiasts. Such comments may be viewed as making a disease claim and may be viewed as claims made by Web site administrators, even though the words are not their own. Such claims likewise deserve careful scrutiny because they also may affect consumer behaviors and beliefs. The inclusion of this type of information constitutes at least a tacit endorsement on the part of the Web site administrator and is further cause for concern related to the type of information available on the Internet being used by patients to make health decisions. Disease claims were most commonly made for soy, black cohosh, and green tea extract. These products are commonly used for menopausal symptoms (soy and black cohosh) and weight loss or cancer (green tea). In these cases, structurefunction claims that describe the purported health benefit
Figure 2 Safety content comparison between retail and nonretail sites.
without explicitly mentioning a disease or diagnosable medical condition are difficult to make. Therefore, it appeared that these Web sites used language that necessarily included references to the medical conditions of perimenopause, obesity, or cancer. Other herbal supplements can more easily make legal structure-function claims by clothing their health statements in phrases that are technically legal, but still lead individuals to believe and expect health benefits related to recognizable medical conditions. Even if outright disease claims are made less frequently for most of these products, important safety information is still often omitted, and when combined with the unsubstantiated structure/function health claims that are made, this presents an ongoing danger to patients who may purchase and use these products in combination with their prescribed medications and without the knowledge of their healthcare providers. The known safety concerns of black cohosh and kava (hepatotoxicity) and St John’s wort (drug interactions) are especially problematic because retail sites only rarely include any information to warn patients of these potential issues. Although retail sites do often include information such as the scientific name of the plant species and the amount of standardized ingredient per pill, this information may only serve to lull a consumer into a false sense that the product is safe and effective, given the authoritative way the product is described and labeled. The manufacturers and vendors of these products know when and how to use science and scientificsounding jargon as marketing tools, but when combined with the omission of key safety information, this only increases the likelihood of consumer use and potential harm. Not all herbal information on the Internet is misleading. Nonretail sites are frequently found when conducting an online general search for information on herbal products. Many of these sites contain authoritative information that
114
The American Journal of Medicine, Vol 127, No 2, February 2014
Table 2
Comparison of Web Site Information on Products with Known Safety Warnings
Adverse effects Interactions Warnings Recommendation to discuss with a provider
Black Cohosh
Kava
St John’s Wort
Retail/Nonretail
Retail/Nonretail
Retail/Nonretail
0%/74% 4%/40% 4%/68% 8%/72%
0%/51% 12/59% 20%/54% 20%/56%
0%/74% 6%/67% 2%/67% 6%/72%
All differences were statistically significant (P < .001).
can help consumers make more informed decisions about the use of these products. A review of the most popular hits on a general search will include information from the NCCAM, the Mayo Clinic, and other medical- or educational institutionesupported sources that more often provide balanced information with respect to efficacy and safety. There also is often a recommendation on these Web sites for individuals to consult their healthcare providers before using a supplement. However, it should be noted that there also are many sites that will be found in a general search that are administered by amateur herbalists and enthusiastic proponents of these products. Such sites may appear to be or are portrayed as “medical information” but in fact are maintained by individuals whose credentials are questionable or absent. Taken together, however, the information contained on many nonretail Web sites is more often referenced and at a minimum contains the warnings and safety precautions for special populations (pregnant patients, children) that afford some consumer protection, when compared with information on retail sites. Although there is good news with respect to the amount and type of herbal product information available on the Internet, the proliferation of herbal retailers occurs at an ever-increasing rate. Policing the Internet is an impossible task because of ongoing shortages of federal manpower and resources to review and take action against inappropriate or harmful information on Web sites.14,25 It therefore becomes the purview of healthcare professionals to take on the vital role of educating patients regarding valid sources of information and how to appraise the information they do find on the Internet critically. The NCCAM’s “Time to Talk” campaign is an important step that helps to provide tools to facilitate conversation about the use of complementary and alternative medicine between health care providers and patients.1 National efforts like this must be coupled with local educational activities to reach additional audiences— including alternative medical providers and the entire diverse community of consumers—to ensure that herbal products are incorporated appropriately into the overall healthcare landscape.26
CONCLUSIONS Although there are benefits to the use of herbal products in certain clinical conditions, until efficacy and safety data are
better established and information sources accurately reflect what is known about these products and their place in therapy, it is the responsibility of government agencies and healthcare professionals to take an even greater role in the critique and dissemination of information that is available to individuals who may be considering the use of such products for their health.
References 1. National Center for Complementary and Alternative Medicine. NCCAM Third Strategic Plan: 2011-2015. National Center for Complementary and Alternative Medicine. Available at: http://nccam.nih. gov/about/plans/2011?nav¼gsa. Accessed March 27, 2013. 2. Fontanarosa PB, Rennie D, DeAngelis CD. The need for regulation of dietary supplements: lessons from ephedra. JAMA. 2003;289: 1568-1570. 3. Wolsko PM, Solondz DK, Phillips RS, Schachter SC, Eisenberg DM. Lack of herbal supplement characterization in published randomized controlled trials. Am J Med. 2005;118:1087-1093. 4. Ernst E. Harmless herbs? A review of the recent literature. Am J Med. 1998;104:170-178. 5. Office of Dietary Supplements. Dietary Supplement Health and Education Act of 1994. Office of Dietary Supplements. Available at: http://ods. od.nih.gov/About/DSHEA_Wording.aspx. Accessed April 15, 2013. 6. Dietary Supplement Health and Education Act. Pub L No. 103-417, 108 Stat 4325 (codified at 21 USC § 301 [1994]). 7. Harris IM. Regulatory and ethical issues with dietary supplements. Pharmacotherapy. 2000;20:1295-1302. 8. US Department of Health and Human Services, US Food and Drug Administration. Regulations on statements made for dietary supplements concerning the effect of the product on the structure or function of the body; final rule, 65 Federal Register 1000(2000) (codified at 12 CFR §101). 9. Crawford SY, Leventis C. Herbal product claims: boundaries of marketing and science. J Consum Marketing. 2005;22:432-436. 10. Eisenberg DM, Kessler RC, Foster C, Norlock FE, Calkins DR, Delbanco TL. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N Engl J Med. 1993;328:246-252. 11. Davis RB, Eisenberg DM, Tindle HA. Trends in use of complementary and alternative medicine by US adults: 1997-2002. Altern Ther Health Med. 2005;11:42-49. 12. Barnes PM, Bloom B, and Nahin R. Complementary and alternative medicine use among adults and children, United States, 2007. National Center for Health Statistics. http://nccam.nih.gov/sites/nccam.nih.gov/ files/news/nhsr12.pdf. Published December 10, 2008. Accessed June 30, 2013. 13. National Center for Complementary and Alternative Medicine. Statistics on CAM Costs. National Center for Complementary and Alternative Medicine. Available at: http://nccam.nih.gov/news/camstats/costs? nav¼gsa. Accessed March 20, 2013. 14. DeAngelis CD, Fontanarosa PB. Drugs alias dietary supplements. JAMA. 2003;11:1519-1520.
Owens et al
Online Sources of Herbal Information
15. Temple NJ. The marketing of dietary supplements in North America: the emperor is (almost) naked. J Alt Complement Med. 2010;16:803-806. 16. Morris CA, Avorn J. Internet marketing of herbal products. JAMA. 2003;11:1505-1509. 17. Jordan MA, Haywood T. Evaluation of internet websites marketing herbal weight-loss supplements to consumers. J Alt Complement Med. 2007;13:1035-1043. 18. Walji M, Sagaram S, Sagaram D, et al. Efficacy of quality criteria to identify potentially harmful information: a cross-sectional survey of complementary and alternative medicine web sites. J Med Internet Res. 2004;6:e21. 19. Bonakdar RA. Herbal cancer cures on the Web: noncompliance with Dietary Supplement Health and Education Act. Fam Med. 2002;34: 522-527. 20. Zickuhr K. Generations 2010/Pew Research Center’s Internet and American Life Project. Pew Research Center. Available at: http://www. pewinternet.org/Reports/2010/Generations-2010.aspx. Accessed April 24, 2013.
115 21. Ashar BH, Miller RG, Getz KJ, Pichard CP. A critical evaluation of Internet marking of products that contain ephedra. Mayo Clin Proc. 2003;78:944-946. 22. Martin-Facklam M, Kostrzewa M, Schubert F, Gasse C, Haefeli W. Quality markers of drug information on the Internet: an evaluation of sites about St. John’s wort. Am J Med. 2002;113: 740-745. 23. Nazeri A, Massumi A, Wilson JM, et al. Arrhythmogenicity of weightloss supplements marketed on the Internet. Heart Rhythm. 2009;6: 658-662. 24. Thurairaja R, Barrass B, Persad R. Internet websites selling herbal treatments for erectile dysfunction. Int J Impot Res. 2005;17: 196-200. 25. Bent S, Ko R. Commonly used herbal medicines in the United States: a review. Am J Med. 2004;116:478-485. 26. Bodeker G, Kronenberg F. A public health agenda for traditional, complementary, and alternative medicine. Am J Public Health. 2002;92:1582-1591.
