Culinary Spice Plants in Dietary Supplement Products and Tested in Clinical Trials^,,

Abstract

Dried plant parts used as culinary spices (CSs) in food are permitted as dietary ingredients in dietary supplements (DSs) within certain constraints in the United States. We reviewed the amounts, forms, and nutritional support (structure/function) claims of DSs that contain CS plants listed in the Dietary Supplement Label Database (DSLD) and compared this label information with trial doses and health endpoints for CS plants that were the subject of clinical trials listed in clinicaltrials.gov. According to the DSLD, the CS plants occurring most frequently in DSs were cayenne, cinnamon, garlic, ginger, pepper, rosemary, and turmeric. Identifying the botanical species, categorizing the forms used, and determining the amounts from the information provided on DS labels was challenging. CS plants were typically added as a component of a blend, as the powered biomass, dried extracts, and isolated phytochemicals. The amounts added were declared on about 55% of the labels, rendering it difficult to determine the amount of the CS plant used in many DSs. Clinicaltrials.gov provided little information about the composition of test articles in the intervention studies. When plant names were listed on DS labels and in clinical trials, generally the common name and not the Latin binomial name was given. In order to arrive at exposure estimates and enable researchers to reproduce clinical trials, the Latin binomial name, form, and amount of the CS plant used in DSs and tested in clinical trials must be specified.

Introduction

Plants are used as culinary spices (CSs)⁴, flavorings, and colorants in foods worldwide, and many of these same plants are also used as dietary ingredients in dietary supplements (DSs) and as drug substances (1). In the United States, the FDA regulates products in interstate commerce based on their intended use, as determined by the form in which they are consumed, the route of administration, and the claims made about them in labeling.

Regulations surrounding CS plants are different when they are used as dietary ingredients in DSs rather than as spices to impart flavor and color to conventional foods (foods). FDA regulations define the term spice as “any aromatic vegetable substance in the whole, broken, or ground form, except for those substances which have been traditionally regarded as foods, such as onions, garlic and celery; whose function in food is seasoning rather than nutritional; that is true to name; and from which no portion of any volatile oil or other flavoring principle has been removed” (2). The regulations also provide guidance on the common and acceptable names for use in labeling spices and foods in which they are used, and a list of spices that are generally recognized as safe for their intended use (2–4).

Examples of botanicals that are used as CSs include rosemary, sage, and bay leaves; flax and fenugreek seeds; dried fruits of black pepper; cayenne pepper and paprika; cinnamon bark; licorice and chicory root; ginger and turmeric rhizomes; garlic bulbs; clove flower buds; and saffron flowers (5). Because culinary use applies to only foods, health benefit claims related to spices cannot be made on food product labels.

When the intended use of a product is to supplement the diet, CS plants are treated as dietary ingredients in DSs and categorized as “herbs or other botanical” (6). The Dietary Supplement Health and Education Act of 1994 (DSHEA) created the regulatory category called dietary supplements and defined dietary ingredients as “a vitamin; a mineral; an herb or other botanical; an amino acid; a dietary substance for use by man to supplement the diet by increasing total dietary intake; or a concentrate, metabolite, constituent, extract; or combination of any of the above dietary ingredients” (1). Dietary supplement products contain dietary ingredients and are intended for ingestion in tablet, capsule, powder, softgel, gelcap, and liquid form; are not represented as a conventional food or as the sole item of a meal or of the diet; and are labeled as dietary supplements.

There are specific regulations governing how herbs or botanicals are to be declared on DS labels (7). Structure/function claims (“statements of nutritional support”) are permitted on DS labels (8). Structure/function claims describe the role of a nutrient or dietary ingredient intended to affect the normal structure or function of the human body, for example, “calcium builds strong bones.” In addition, they may characterize the means by which a nutrient or dietary ingredient acts to maintain such structure or function, for example, “fiber maintains bowel regularity” or “antioxidants maintain cell integrity.”

In addition, plant extracts and isolated phytochemicals are permitted in DS formulations. Herbs and botanicals marketed in a DS before 15 October 1994 are considered grandfathered ingredients, whereas nongrandfathered ingredients are new dietary ingredients (NDIs) (9). NDIs require a premarket notification before being introduced into the US marketplace. NDIs that are not chemically altered may be exempt from this notification requirement.

Use of CS plant products appears to be increasing, reflected by the increased demand for these products. Although current numbers for sales of spices in the United States are not available, the global seasonings and spices market was $12 billion in 2012 and is expected to reach an estimated value of $17 billion in 2019 (10). To take just 1 example, in 2014, mainstream market sales of turmeric/curcumin as the primary ingredient in DSs grew by 60% and sales of botanical DSs grew 6.8% and represented 17% of the $36.7 billion DS market in the United States (11).

Researchers might assume that if a spice is being tested in a clinical trial it must be the same as that available for culinary uses or as a DS, but this may not be so. Our first objectives were to identify the CS plants most commonly occurring as dietary ingredients in DSs and to compare the forms used in DSs to the forms used in clinical trials. Second, we wished to identify claims made about their purported benefits on DS labels and compare these to the outcomes measured in clinical trials. Because the intent of this study was only to compare the forms used, recommended serving sizes (doses) and claims made on DS labels with the forms, doses used, and outcomes measured in clinical trials, we limited our research to information from 2 NIH publicly accessible databases, the Dietary Supplement Label Database (DSLD) and clinicaltrials.gov. The DSLD is a voluntary public use database that lists CS plants as reported on the labels of DSs sold in the United States (12). Clinicaltrials.gov is a database of publicly and privately supported clinical studies of human participants conducted around the world (13).

Methods

We used names from FDA’s list of generally recognized as safe spices to search the DSLD for CS plants commonly occurring in DSs (4). At the time of this study, the DSLD contained label-derived information from ∼30,000 products. After frequently used CS plants in DSs were identified, the DSLD dietary ingredient and expanded search features were used to compile information on ingredient forms (biomass, extract, isolated phytochemical) and amounts for each CS plant used in DSs. A list of structure/function claims, the most common type of health-related claims on DS labels, was also compiled. The search results from the DSLD were downloaded as comma-separated value (CSV) files, and the data analysis tool in Excel was used to calculate the mean (and SE) of the listed label amount of the whole botanical (powdered biomass) and all other forms (extracts and isolated phytochemicals).

Clinicaltrials.gov was used to find registered trials investigating the health effects of the common CS plants identified through the DSLD search and classified as DS studies. Clinicaltrials.gov is an international registry of clinical studies (13). The advanced search feature was used to narrow the search results to all trials (open and closed) using DSs. Screening and absorption, distribution, metabolism and excretion studies and studies testing DSs with disease endpoints were eliminated. The search results from clinicaltrials.gov were then downloaded as CSV files and converted to Excel files to compile the results. The online study record was used to complement information about a trial intervention not included in the CSV download.

Results

In the DSLD, the 7 spice plants that appeared most frequently as dietary ingredients on DS labels were cayenne, cinnamon, garlic, ginger, black pepper, rosemary, and turmeric. Table 1 provides the mean amount of each spice plant and the structure/function claims printed on the DS labels, when present. Isolated phytochemical compounds, extracts, and/or whole botanicals were used in DS product formulations. The amount of each botanical CS dietary ingredient was declared on ∼55% of product labels; the percentages for each CS plant ranged from 37% (cayenne) to 77% (garlic). The amounts reported in Table 1 reflect only the label declaration of the added amount of each CS ingredient. Identifying the botanical species and categorizing the forms used from the information provided on DS labels available in the DSLD was challenging, as described below, and so search results are presented in 2 broad categories as “whole” and as “all other forms (extracts and isolated compounds)” in Table 1.

TABLE 1.

Amounts declared and claims made for culinary spice plants listed on dietary supplement labels in the DSLD¹

Dietary ingredient	Products in DSLD,2 n	Number that declared amount on label, %	Serving,3 mg	Claims made on product labels
Cayenne4	116	37	113 ± 22.5	Circulation support; cleansing/detox formula; may help regulate cholesterol and lipid levels stimulates healthy digestive function; vascular health
Whole5			148 ± 39.0
All other forms			101 ± 35.8
Cinnamon	241	49	364 ± 35.3	Increasingly well-known for enhancing insulin sensitivity; may assist with glucose control and decreases fasting glucose levels when taken consistently; supports sugar and fat metabolism, used throughout history by herbalists to promote healthy digestion and metabolism
Whole			450 ± 52.4
All other forms			239 ± 43.2
Garlic	504	77	257 ± 18.4	Allicin acts as an antioxidant and helps support cardiovascular health; clinically proven to support heart health in 15 independent studies; supporting cardiovascular health, supports healthy cholesterol levels and overall cardiovascular health
Whole			254 ± 28.2
All other forms			260 ± 26.8
Ginger	961	42	127 ± 8.5	A natural anti-emetic; cleansing formula to gently flush away excess waste and toxins; demonstrated natural anti-inflammatory activity, ginger settles an upset stomach and supports healthy tone and peristalsis in the intestines
Whole			129 ± 13.9
All other forms			142 ± 14.5
Pepper	378	59	8 ± 0.7	Clinically shown to increase bioavailability of vitamin C by up to 40%; clinically proven bioperine improves nutrient bioavailability; directly impacts fat metabolism, promotes absorption of nutrients in the GI tract
Whole			6 ± 0.8
All other forms			8 ± 0.6
Rosemary	153	57	134 ± 40.4	Contains numerous anti-aging constituents and significant 5-lipoxygenase modulators; intestinal comfort formula to support GI function, soothe discomforts, and help reduce occasional gas; nature’s rich source of anti-aging constituents, including phytonutrients that optimize memory
Whole			159 ± 65.4
All other forms			96 ± 30.5
Turmeric	650	70	182 ± 10.7	Also supports healthy heart and brain function; benefits joints and other tissues by helping neutralize free-radicals; helps support healthy liver function, some of the very best antioxidants for protecting the central nervous system/health of the breast; supports the liver’s detoxification processes; demonstrated natural anti-inflammatory activity
Whole			204 ± 14.5
Curcuminoids			176 ± 32.2
All other forms			180 ± 19.2

¹DSLD, Dietary Supplement Label Database; GI, gastrointestinal.

²Represents the number of products containing the specific spice plant in the database.

³Values are the mean label-declared amount per serving in mg ± SE.

⁴Includes all added forms.

⁵Includes all powder forms of the whole botanical.

Table 2 lists the form, doses, and outcome measure(s) used in clinical trials corresponding to each CS plant as determined from clinicaltrials.gov. Difficulties in identifying the form, dose, and plant species used in the clinical trials posted on clinicaltrials.gov were similar to those encountered when trying to obtain that information from DS labels. In almost all of the clinical trials reviewed, the isolated phytochemicals, extracts, or proprietary blends, extract rather than powdered plant material, was tested as the active intervention. In addition, the study record seldom provided adequate details about the CS plant being tested or how it was processed.

TABLE 2.

Form of culinary spice plant, dose tested, and outcome measures in clinical trials listed in clinicaltrials.gov¹

Botanical intervention2	Clinical trials, n	Daily dose, mg	Outcome measure
Cayenne
Dihydrocapsiate	3	9	Adaptive thermogenesis
Chili pepper extract	1	6	Body weight and body composition
Cayenne	1	900	Chemopreventive agent for prostate cancer
Capsaicin	1	NP	Energy expenditure
Cinnamon
Fermented cinnamon	1	30,000	Abdominal fat
Cinnamon extract	12	1000–9000	Regulate blood glucose, glycated hemoglobin; gastric emptying; postprandial glucose
Cinnamon extract	1	1000	Ovulatory cycles
Cinnamon	1	2800	Endothelial function
Garlic
Aged garlic	6	1200–6000	Variety of cardiovascular disease measures
Aged garlic extract	1	1500	Antioxidant effect in heavy smokers
Aged garlic extract	1	1200	Endothelial function
Aged garlic extract	1	2560	Immune system; cold and flu symptoms
Ginger
Ginger	6	2000	Treating nausea in patients receiving chemotherapy
Ginger	1	NP	Chronic osteoarthritis
Ginger	5	500–2000	Treating nausea
Ginger	1	2800	Endothelial function
Pepper
Bioperine + curcumin or resveratrol	4	5–25	Mild cognitive impairment in healthy and in schizophrenia adults; enhance bioavailability
Black pepper	1	1500	Energy expenditure in overweight older adults
Black pepper	1	2800	Endothelial function
Rosemary
Rosemary extract	1	NP	Weight loss
Rosemary	1	2800	Endothelial function
Turmeric
Turmeric extract	1	168	Osteoarthritis of the knee
Curcumin + bioperine	3	∼5000	Mild cognitive impairment in healthy and in schizophrenia adults
Curcumin	3	1500	Osteoarthritis of the knee
Curcumin	8	180–3000	Management of age-related cognitive decline
Curcumin	2	1800	Reduce severity of irritable bowel
Curcumin	1	1000	Management of depression
Curcumin	1	200	Improve post-exercise recovery
Curcumin	1	2000	Prevention of radiation-induced dermatitis
Curcumin	1	NP	Reduce number of polyps

¹Interventions listed in clinicaltrials.gov, where the intervention was categorized as a dietary supplement. NP, not provided or unable to determine the amount tested.

²Botanical or its active constituent.

Although many DS label claims were based solely on the common or traditional use of the CS plant, some proprietary dietary ingredients and blends used in DS product formulations had clinical studies listed in clinicaltrials.gov to support their label claims. The doses used in clinical trials in our clinicaltrials.gov search were generally higher than the average amounts listed on DS labels, except when the botanical was the only or the main dietary ingredient in the product (see Tables 1 and 2).

Discussion

Over half of product labels listed amounts of CS plant, possibly because FDA labeling regulations give manufacturers considerable flexibility on how dietary ingredients that do not have official daily values (nondaily values) can be declared on DS labels (8). Manufacturers are required to list the amount of the dietary ingredient when it is listed individually by name within the Supplement Facts panel. (Manufacturers would, however, be required to know the amount of each ingredient in the product in order to comply with current good manufacturing practices for dietary supplements.) If a botanical dietary ingredient is part of a proprietary blend (which can consist of botanical and nonbotanical ingredients), manufacturers can provide the total amount of the blend and list the blend’s individual dietary ingredient in decreasing order of prominence, i.e., the proprietary blend is considered the ingredient and the amounts of each individual ingredient within the blend need not be declared.

There may be marketing advantages to using the proprietary blend concept on labels. Manufacturers gain control over the exact composition of their blend, making it more difficult for competitors to duplicate their formulas. Manufacturers can also give these blends creative names, which are often trademarked, to make products stand out from similar products and to make labels more appealing to consumers.

One specific example of the complexity of marketed products and their attendant labels is turmeric and curcumin. Curcumin is a bright yellow polyphenol that belongs to a class of a structurally related class of compounds called curcuminoids. These are thought to be the major bioactive constituents in turmeric and its extracts. Curcumin is often used as a synonym for curcuminoid blends on DS labels, but this common usage is incorrect because curcumin is only 1 of the 3 major marker compounds to be expected in a curcuminoid blend (curcumin, demethoxycurcumin, and bisdemethoxycurcumin) (14). In addition to being chemically incorrect, this common usage is implausible from a manufacturing and quality control standpoint, because pure curcumin is unstable as a sole isolated compound (15). Thus when curcumin was listed on DS labels, we treated it as a curcuminoid blend and assumed all components to be derived from turmeric.

We found 16 active trademarked names for curcuminoid blends on the US Patent and Trademark Office website. These names are also found on DS labels. Examples of 3 of the names found on labels are Curcumin Synergy, Meriva Bioavailable Curcumin, and Curcumin C3 Complex (16). The US Patent and Trademark Office grants a trademark if a name is unique within a given application or use, not if there are chemical differences in the composition of blends or products. The compositional differences in these blends are proprietary information held by the manufacturer and are not information available to the public.

Identifying the plant species used in DS formulations and in clinical trials was also challenging. Cinnamon is a good example of the difficulties in comparing information obtained from DS labels to the information provided by researchers. Many issues can arise with the use of common and scientific names. Cinnamon is a common name used to refer to more than 1 species of cinnamon, e.g., Chinese, Ceylon, Saigon, and Indonesian. In addition, there is more than 1 synonym by which each of these species is known, due to the existence of many taxonomy and scientific naming conventions. There is also inconsistency in what the regulations permit as the accepted common name in the labeling of products. Thus confusion can ensue when researchers and manufacturers unfamiliar with these naming conventions try to navigate various regulations and references that do not use the same naming conventions.

The US DS labeling regulations require that the common or usual name used on labels of botanical DS ingredients should be consistent with the standardized common names in the 1992 edition of Herbs of Commerce (7, 8). If there is no common name or if the manufacturer opts not to use that name, the regulation requires that the Latin binomial name of the plant be used in the labeling of botanical DSs. In 2003 the FDA issued a proposed rule to replace the 1992 edition of Herbs of Commerce with the 2000 edition (17). Although this proposed rule was not finalized, the FDA does not object to the use of the 2000 edition. However, there is not always agreement between the 1992 and the updated 2000 editions (18). For example, “cinnamon” is the common name for Chinese cinnamon (Cinnamomum cassia or Cinnamomum aromaticum) in the 1992 edition, whereas it is the common name for Ceylon cinnamon (Cinnamomum verum or Cinnamomum zeylanicum) in the 2000 edition (18). In the 1992 edition the common name for Ceylon cinnamon is “Ceylon cinnamon,” whereas in the 2000 edition the common name for Chinese cinnamon is “cassia.” As a result, if a DS declares “cinnamon” on the label it could either be Chinese or Ceylon.

Often the form, dose, and plant species used were incompletely described in the clinical trials. In part this was because most studies using DSs did not meet the “applicable clinical trial” criteria that trigger FDA data requirements. The study sponsor or principal investigator is responsible for registering and maintaining the information entered into clinicaltrials.gov (13). Although the information in clinicaltrials.gov is presented in a standardized format, the information in the study record is not standardized. As a result information contained in some study records was incomplete and often did not provide sufficient detail about the test product, such as the Latin name(s) of the botanical(s) tested. Trials registered on the clinicaltrials.gov website are not limited to US trials or US products, and so regulatory requirements under which trials are conducted and reported will vary according to each country’s regulatory paradigm.

The clinical trials reviewed focused on active constituents or extracts, and not on the powdered CS plant material. Neither did study records provide details about the substances being tested nor how they had been processed. In order to reproducibly delineate the health outcomes for each species as Swanson and others (19–22) have noted, it is important to specify the species and note quality standards used to evaluate the materials used in clinical trials. The importance of describing the quality standards applied to the intervention was emphasized by Brickman (20), who noted that different quality standards often exist for botanical ingredients intended to be used as food, as DS ingredients, and as phytomedicines. Of the studies we reviewed, 41% were industry funded. Although product formulations are held as proprietary information by manufacturers, it is important to provide the species of the botanical intervention when registering trials even if the amount(s) used are not declared to ensure that the claims made on product labels correspond to the plants tested in clinical trials. For example, our review of the published literature examining the role of cinnamon supplements in glycemic control in type 2 diabetes shows that these studies tested mainly Chinese cinnamon (RB Costello, JT Dwyer, LG Saldanha, RL Bailey, E Wambogo, Office of Dietary Supplements/NIH, unpublished results).

Another problem we encountered was the use of common names that did not accurately and clearly reflect the identity and composition of the intervention. Similar to information found on DS product labels, nearly all the clinical trials listing the intervention as curcumin in the study record on clinicaltrials.gov actually used curcuminoid blends, as we discovered in a web search of information from the suppliers of these products.

Clinical studies listed in clinicaltrials.gov were sometimes used to support DS structure/function label claims for the plant, dietary ingredient, or blend. Although companies are not required to submit substantiation data or obtain premarket FDA approval before making structure/function label claims, US law requires that label statements must be truthful and not misleading and companies that make such claims must have data that substantiate these claims and share this data, if requested, with the FDA. These data may not necessarily be in the public domain. Although clinicaltrials.gov provides ready access to studies that evaluate botanical DSs, the results from these studies are generally not posted on the website, and so it is difficult to know whether the doses used in the trials are effective.

The value of this review is that it highlights the challenges researchers face in working with information on botanicals declared on DS labels and in clinical trials. As previous reviews on caffeine and green tea have demonstrated, this review again shows that the amounts of botanical or botanical constituent, form (or composition), and identity of ingredients in botanical DSs were difficult to ascertain even on labels that complied with the DS label regulations (23, 24). These challenges make estimating intakes and exposures to herbs and botanicals difficult for epidemiological studies that use data solely from DS labels. Intake estimates of various plants and plant-derived products require a number of assumptions. Because permissible common names for DS ingredients are not consistent with those for foods (in this case, plants used as culinary spices), assumptions about the identity of plant species reported by consumers are necessary. Assumptions must be made and methods are needed to arrive at the intake of a botanical, when that botanical is listed as an ingredient in a proprietary blend. In our study, ∼45% of product labels did not declare the amounts of the each spice plant. Assumptions must be made and methods may be necessary to convert amounts of extracts to the whole biomass, if researchers wish to combine intakes of botanicals from DSs with intakes from food to arrive at estimates of total daily intake.

Assumptions about the identities and quantities of components of test product (and placebo or control material) used in clinical trials are also necessary to compare and consolidate study results. Although there are many naming conventions for botanicals, researchers should provide the Latin binomial name for the plant(s) they are researching because these are unique names unlike common names, which are not unique or consistent. A variety of online, searchable resources are available to researchers to find synonyms for medicinal and herbal plant species used in the United States and internationally (25–27). There are also a number of documents, such as Reporting randomized, controlled trials of herbal interventions: an elaborated CONSORT statement, available to researchers that provide guidance on how to report and review studies involving botanicals (22, 28–30). Researchers should review these resources to ensure proper reporting and evaluation of botanical interventions.

Our study had several limitations. The report represents only a snapshot of the botanical research and botanical supplement product marketplace. It provides a comparison of the amounts and forms of a subset of plants and plant ingredients found on DS labels with similar information found in a registry of clinical trials designed to evaluate the health effects of spice plants and spice plant ingredients. The limitations of the comparison are that we did not search the literature for published clinical trials that were not registered in clinicaltrials.gov or all products sold in the United States that were not listed in DSLD. There are currently over 65,000 DSs in commerce in the United States and about 3,000 plant species in commerce (18). Clinicaltrials.gov currently lists 204,041 studies with locations in all 50 states in the United States and in 191 countries (13).

Our analysis suggests that improved reporting of the forms and amounts of CS plants on DS labels and used in clinical trials is needed. This will enable researchers to arrive at more accurate estimates of exposures and to determine the health properties of these botanicals.