Skip to main content

PLANT food supplements: Levels of Intake, Benefit and Risk Assessment

Final Report Summary - PLANTLIBRA (PLANT food supplements: Levels of Intake, Benefit and Risk Assessment)

Executive Summary:
Botanicals, as such or as ingredients of food supplements/traditional medicines, have a long tradition in Europe. Despite the present economic situation, food supplements (including those with botanical ingredients) show an increasing market and acceptance by consumers. On the other hand, there was and there is a growing demand for high quality data to assess risks and benefits of Plant Food Supplements (PFS), in order to guarantee their safety for European consumers.
On these bases, the objective of PlantLIBRA, the EU-financed FP7 project with reference number 245199, was the collection of data from scientific literature or from experimental activities to allow a risk/benefit assessment and support the decision-making by regulators. Furthermore, the different PlantLIBRA stakeholders also aimed to improve the quality assurance of this class of products. In fact, PlantLIBRA was structured to develop, validate and disseminate data and methodologies for risk and benefit assessment and implement sustainable international cooperation.
The most important results from the 4-years activity can be summarized in 7 points:
• The PlantLIBRA PFS Consumer Survey identified the basis for future work providing examples of design, methodology, material, implementation, database construction, data analysis, actual knowledge and its dissemination. The knowledge generated with this survey will benefit all stakeholders in the sector, namely the consumers, policy decision makers, industry, researchers and academics, health professionals etc., by helping them in their decision-making for the consumption of these products or in the decision-making processes of their respective professional areas.
• Collection of information on risk and benefit associated with botanicals, including both data from scientific literature and new knowledge produced by partners thanks to experimental studies. The systematic review of all data allowed the development of new approaches and models to perform risk/benefit assessment offering a new tool (Opasnet) for the scientific world, regulators and other stakeholders.
• Setup of new analytical methods to support the quality control and to help solve several gaps suggested by partners. Particular attention was paid to the improvement of methods for the identification or quantification of botanical ingredients, regulated by legislation (presence of toxic substances, frequent adulteration, etc.)
• Another important tool developed by the PlantLIBRA partners is the database ePlantLIBRA. The primary users of ePlantLIBRA are the regulatory affairs sector (e.g. assessment of PFS supporting health claims/risk assessment), food industry (e.g. evaluation and development of PFS), researchers and epidemiologists. In addition, PFS data and information can be widely used to underpin academic research into links between diets and health. Research outcomes are likely to influence policy at a national and/or international level (e.g. EU, EFSA & globally) and policy may dictate future dietary monitoring programmes or research.
• Consumer perception of food supplements was compared with the perception of other stakeholders using a number of different methodological approaches to ascertain what information exists for consumers and how they use that information.
• The activity on regulatory aspects was based on 4 Policy Advisory Board meetings were PlantLIBRA Work Package (WP) leaders presented their scientific data and stimulated the discussion on harmonization of the food supplement sector. A first result was achieved with the BELFRIT project, which was promoted by a discussion at the second project meeting in Brasov.
• Dissemination of PlantLIBRA results was promoted both in scientific congresses and specific workshops and in training courses organized in Europe, Africa, Asia and South America.

Project Context and Objectives:
1. General aspects
Globally, there is intense demand for safe and health-promoting foods, including food supplements. Though growth expectations are highest for emerging economies in 2014, the EU market is one of the largest in the world. Specifically, Plant Food Supplements (PFS), as confirmed by PlantLIBRA's findings, have a high acceptance by consumers (18.8% of screened survey respondents used at least one PFS), and have potentially a series of advantages: reported efficacy in terms of well-being, a history of mainly safe use, and relatively low costs. Although botanicals have been used for centuries in traditional medicine and, more recently, as food supplements for their health-promoting effects, it is doubtless that data, information, and tools to ensure their safety and reap their potential benefits are not sufficient, particularly when compared to the popularity and market success of the products.
The supply chain of PFS is global. Europe is a significant global player from the farm to retail, with leading suppliers of raw materials, botanical preparations and finished food supplements. As a consequence, the further development and continued success of PFS in Europe could bring substantial growth, both in the internal market and in international trade. Sustainable growth needs to be science-based. If it is, safe products with truthful efficacy claims will bring benefit to consumers.
However, in the case of PFS, scientific and regulatory uncertainty poses barriers to such science-based growth. In order to produce the evidence needed to protect consumers, and exploit market opportunities, PlantLIBRA has proposed, developed, and implemented a new integrated research approach for PFS. Since the European PFS industry is part of a global food chain, international cooperation is required to develop and share common research approaches and priorities, results and for their dissemination.
The project mapped the key societal challenges Health, Demographic change and Wellbeing, as identified in Horizon 2020. Botanicals may contribute to the improvement of health in up to a fourth of Europe's population, and to the reduction of risk factors for chronic diseases, such as cancer and cardiovascular diseases. PlantLIBRA also aimed to contribute to the sustainability of one of Europe's bio-based industries, the botanicals supply chain, within Europe's societal challenge of an agri-food sector for a safe and healthy diet.
In the past three years, changes in policy at the EU level and within Member States also suggest that more research was and will be warranted. EFSA, the European Food Safety Authority, had started work in this area, culminating in a guidance document (2009) and related exercises, in health claims opinions, in versions of the compendium (2007, 2012) and, most recently, in the compound-driven Qualified Presumption of Safety approach (2014). While the work of EFSA is of high quality, the approaches have not been adopted by Member States or routinely at the EU-level; evaluations of claims are on hold since 2012. Supporting indirectly the work of EFSA with data and tools has been a key objective of PlantLIBRA. In fact, PlantLIBRA has disseminated its results to EFSA, and several beneficiaries and experts of PlantLIBRA have been directly involved in EFSA's work on botanicals. At the same time, inspired by PlantLIBRA, Belgium, France and Italy set up a working group to harmonise their regulations on botanicals, with adoption of actual regulations in 2014, the so called BELFRIT project. In 2012, the European Commission and Member States started to debate a pan-EU regulation on botanicals, covering quality, safety and potentially efficacy. PlantLIBRA has delivered scientific information directly to European policymakers through its Policy Advisory Board, with members from almost all Member States, which met three times in plenary sessions and more often in smaller groups, and strengthened the policy debate.
Several EC-funded research projects such as EuroFIR (nutrient databases), MoniQa (contaminant databases), and Beneris/Qalibra (food risk benefit assessment) had defined methodologies and developed tools that have been applied to PFS via PlantLIBRA.
Supporting science-based decision-making and the safe use of PFS have been the driving aims of PlantLIBRA. The project has prioritised the development of strong science, its application to policy and the necessity to investigate and ensure safety for consumers.
As planned, the project has directly achieved, within the four years of the project, objectives related to:
• An accessible Meta-database containing data on PFS, on the composition of botanicals, on biologically active substances, on relevant human studies, on contaminants and on adverse effects. A separate database on consumption was also produced for research purposes.
• An open risk, benefit and risk-benefit assessment platform for botanicals and PFS, with case-studies and the expertise to produce new assessments. Risks and benefits can be qualitatively compared as a result.
• Dissemination to stakeholders to assist science-based decision-making, and to consumers. Training courses were also organized in different countries, with particular attention for international (extra-European) cooperation.

2. Supporting science-based decision-making
In this context, with the overall goal of improving decision-making, and although the evidence on PFS is incomplete and complex to evaluate, PFS are doubtlessly associated with biological effects, with both benefits and risks.
New national regulations suggest more than a thousand of botanicals are permitted. Facing such a vast group of plants, the aim of the project was to retrieve and publish as many reviews on scientific information on benefits as possible; to produce new data on risks; and to develop methods to address plants with much but also little data. Overall, in this context, PlantLIBRA aimed to support scientifically decision-making, that needs to be made on a daily basis by authorities and food chain operators.
Finally, the project aimed to develop a sustainable, immediately accessible databanks available to the authorities making decisions and to industry. The resulting databank is ePlantLIBRA.

3. Ensuring safe use of food supplements
While recognising the relevance of a history of safe use, the project consortium has assumed that safe use also requires valid and adequate risk and safety assessment by competent authorities or private sector risk assessors. As noted above PlantLIBRA addressed this objective developing new methods.
The project was also aware that safe use also implies awareness of risks and benefits by consumers, citizens, the private sector and authorities, so that PlantLIBRA contributed delivering user-friendly and in-language information and, especially, understanding in this area, to inform policy on both benefits and risks.

4. Intake and consumption patterns
The project aimed to collate and review existing data on consumption and intake, which was done revealing limited available information. More significantly, PlantLIBRA aimed to develop specific methodologies and generate actual, novel, crucially missing survey data. The methodology, tested and validated, will be available for further studies. The data have been published, and will be available for further studies.

5. Health promoting properties / health claims/benefit assessment
EFSA's evaluation of claims related to PFS under art. 13 of Reg. 1924/2006 was put on hold in 2012, in light of the different, less stringent approach taken for herbal medicines. It is not known when, and with which approach, the evaluations will resume.
Taking a whole body of evidence approach, with assessments for each type of evidence (in vitro, animal, human) PlantLIBRA aimed to develop new evidence and a methodology to characterise the evidence on benefits and report it to decision-makers. On one hand, our goal was to produce standardized systematic reviews for priority plants. On the other hand, the project planned to identify and in some cases test new markers for intervention studies. The objective was also to identify best practice for the assessment of benefits of PFS.

6. Risk assessment and safety assessment
In the context of risk assessment, PlantLIBRA recognised that adverse events remain a critical, but understudied source of information regarding safety. The objective was therefore to collect retrospectively and prospectively data on adverse events to PFS in Europe. More generally, PlantLIBRA accepted that the EFSA guidance on the safety of botanicals (2009) was the starting point for further steps in this area. In fact, it suggested the need to apply new concepts such as mode of action, margin of exposure, threshold of toxicological concern, to overcome the major bottlenecks in the current methodologies, as well as the matrix effects. These aspects were investigated in the project.

7. Meta-database
Safety of PFS is not only a result of the plant compounds in it, but also of the levels of residues or contaminants that may be present. The RASFF (Rapid Alert System for Food and Feed) of the EC provides information on contamination and other safety issues associated with food supplements, including PFS; the EC-funded MoniQa project had created, and maintains, a database linking RASFF alerts, regulations, analytical methods and other useful information. PlantLIBRA aimed to connect and relate this information to PFS.
To carry out safety and efficacy assessments at the PFS and plant level, a databank is necessary. The aim was to create a comprehensive databank, called ePlantLIBRA.

8. International cooperation and dissemination
A significant portion of the raw materials imported in the EU for producing PFS are sourced in China or other third countries. Concerns about the quality of the raw materials have been voiced and documented, also by industry, with product quality a major concern. One of PlantlLIBRA goal was to work with the exporting industry in those countries and with scientists there, particularly to facilitate capacity building.

Project Results:
1. The PlantLIBRA Plant Food Supplements (PFS) Consumer Survey (WP1)
The PlantLIBRA Plant Food Supplements (PFS) Consumer Survey has collected data from May 2011 to August 2012 on:
• The socio-demographic profile of PFS users
• The usage patterns of PFS products
• The products consumed and their botanical ingredients.

A cross-sectional, 12-month retrospective survey was conducted in 24 cities in six European countries: Finland, Germany, Italy, Romania, Spain and the United Kingdom. An estimated sample size of 2000 screened individuals per country was calculated in order to obtain a final sample of approximately 400 consumers per country (total N= 2400 approximately). Per country, gender and age group quotas were set as follows: 300 adults (18 to 59 years) and 100 older adults (60-and-over years), with 30–50% male and 50–70% female. All individuals were screened by means of a brief questionnaire, which recorded PFS usage in the preceding 12 months. Individuals were considered eligible for inclusion if they were over 18 years old and met the established criteria for a "PFS consumer" (intake of PFS in the last 12 months). Eligible consumers subsequently completed a detailed questionnaire on their PFS usage, providing details of product/plant names, dosage forms, frequency of use, reasons for use, adverse effects, places and patterns of purchase and information sources on products. In addition, respondents were asked to provide sociodemographic data including age, gender, level of education and employment status, as well as self-reported height and weight and further health-related lifestyle information.
Considering the characteristics of the PFS consumer sample, the estimated weighted overall PFS usage prevalence rate was 18.8%, data per country were as follows: Finland 9.6%, Germany 16.9%, Italy 22.7%, Romania 17.6%, Spain 18.0% and the United Kingdom 19.1%.
Across all countries, more than half of the participants (57.5%) were employed, with the percentages slightly lower in Finland (50.9%) and in the United Kingdom (52.4%). The majority of participating consumers were educated to medium level. Respondents were asked a number of questions regarding health-related lifestyle factors. Less than half of the consumers had never smoked (46.6%), less than one quarter were ex-smokers (23.1%) and less than one third were current smokers (30.3%). More than half of the total respondents (59.3%) had not consumed alcohol or had consumed it less than once daily; more than a tenth (12.6%) reported daily alcohol consumption.
The proportion of overweight and obese people in the survey was 49.8%. Some significant differences in levels of physical activity were noted between countries. High levels of activity were reported by 85.5% of Romanian respondents compared to a value of 42.9% across all countries. Three quarters of consumers reported their health status as very good or good (75.5%), while 3.6% reported it as bad or very bad and 21.0% as neither bad nor good.

Overall, products are most often taken ‘‘periodically’’ (37.3%) with respondents also reporting using PFS when experiencing a ‘‘flare up or worsening of a condition’’ (22.2%). Product are also used on a more ‘‘sporadic basis’’ (19.8%) and on ‘‘other non-specified occasions’’ (17.8%).
Respondents reported a total of 1288 products across the six countries. At individual country level, the highest numbers of different PFS were used in Italy (289) and Spain (284).
The number of different botanical ingredients was 491, with the maximum number of different botanicals contained in a single product being 46 and present in a German product
In terms of the number of products used, 83.7% of all consumers reported taking one product in the preceding 12 months, with 12.3% taking two products and 4.0% using more than two products. Overall 51.5% of consumers used a single-botanical product and 32.3% used one multi-botanical product.
The most common dose forms used are capsules (38.3%) and pills/tablets/lozenges (36.8%).
A total of 491 botanicals -used in at least one PFS- were reported across the six participating countries. Based on the survey results, the eleven most frequently used botanicals (numbers of consumers ranging from 194 to 100) in descending order are Ginkgo biloba (ginkgo), Oenothera biennis (evening primrose), Cynara scolymus (artichoke), Panax ginseng (ginseng), Aloe vera (aloe), Foeniculum vulgare (fennel), Valeriana officinalis (valerian), Glycine max (soybean), Melissa officinalis (lemon balm), Echinacea purpurea (echinacea) and Vaccinium myrtillus (blueberry).
Cross-country differences emerge when considering the overall top-40 botanicals more frequently present in PFS products in each of the individual six countries. In the Finnish sample, products containing Glycine max (soybean) are the most frequently used, followed by those containing Echinacea angustifolia and purpurea (echinacea). German consumers reported Ginkgo biloba (ginkgo), Cynara scolymus (artichoke) and Olea europea (olive) as the most frequently used botanicals; whilst in Romania, Ginkgo biloba (ginkgo) was also the ingredient most frequently indicated, followed by Aloe vera (aloe) and Panax ginseng (ginseng). Amongst Italian consumers, Aloe vera (aloe) was the most frequently used botanical, followed by Foeniculum vulgare (fennel) and Valeriana officinalis (valerian). In Spain, PFS containing Cynara scolymus (artichoke) were the most frequently used products, followed by those containing Valeriana officinalis (valerian) and Equisetum arvense (horsetail). In the United Kingdom, Oenothera biennis (evening primrose) was by far the most frequently reported botanical ingredient, followed by Panax ginseng (ginseng) and Hypericum perforatum (St. John’s wort).
The results of this survey represent some of the first data on the usage of PFS at European level, thus addressing the existing deficit of such data by collecting retrospective data directly from consumers in six European countries. The benefits of the data collection instrument used in this study included that it was relatively straightforward to administer, did not alter habitual usage patterns and allowed the classification of individuals into categories of usage.

2. Beneficial properties of PFS: biological markers and methods (WP2)
Epidemiological, clinical and human intervention studies including PFS were examined and collected using systematic review methodology. Benefit areas chosen as important were: cardiovascular health; post-menopausal bone health; menopausal symptoms; gastrointestinal health; inflammation.
Epidemiological, clinical/human intervention studies using PFS were then systematically examined and reviewed for each of the PFS-benefit area combinations. Another objective was the review, application and validation of methodologies to determine the benefits of consumption of PFS. The existing methods were reviewed where gaps were perceived. Assessments and reviews were published in the areas of cardiovascular function, infectious diseases and immune function.
The biomarkers of intake of PFS have been reviewed and validated in human studies. PlantLIBRA partners developed a consensus for "best practice" for future human intervention studies on PFS. It should be noted that biomarkers of exposure are relevant to both benefit and risk, since the effects of PFS are dependent on the dose. The choice of botanicals to include in the reviewing process was based on: 1) preliminary partners' suggestion; 2) missing data as reported by international panels of experts and in particular EFSA experts present in the Consortium; 3) results produced by the PlantLIBRA survey.
Biomarkers of exposure based on measurement in urine have been reviewed; they are in most cases the best source of biomarkers of exposure, and much less invasive than blood sampling. Short and simple human intervention studies with a single quantified dose of selected PFS were carried out to determine further biomarkers of exposure in urine, as required and defined above. This was achieved by feeding Ginkgo biloba (ginkgo), Panax ginseng (ginseng), Camellia sinensis (green tea) or Vitis vinifera (grape seeds) to healthy volunteers and measuring specific metabolites in the urine, some of which were unique to the consumed PFS and so could be used as a marker of intake.
Establishing the proposed beneficial effects of large numbers of existing or new PFS are critical to making a preliminary evaluation of their effectiveness. For this, in vitro assays are often used, and there are many examples in the literature. The project has focused on the most important activities, which were measured in the most physiological relevant way, i.e. taking metabolic changes into account. The panel of assays selected represented the most efficient way to "screen" and select for activities of both existing and new PFS. The effects measured in vitro are not indicative of a protective activity in vivo, but guide the more definitive future in vivo studies. We used novel strategies to improve the quality and predictive power of data derived from in vitro assays. New in vitro methods were developed to mimic metabolism, and these changes modified some of these activities. The final objective was to hold a workshop to define and assess "best practice" benefit methodologies for use in future human intervention and clinical studies, to propose "new" strategies to evaluate and grade the benefit associated with PFS. A consensus workshop with policymakers, stakeholders and consumers was organized before the end of the project (May 2014) to agree the methodology and results of discussion are the object of both scientific and educational publications in preparation.


3. Risks associated to PFS active ingredients (WP3)
PlantLIBRA has investigated how newly developed and validated science based models can be applied to assess the risks of consumption of PFS. It was considered the Margin of Exposure (MOE) concept for the risk and safety assessment of low dose exposure to compounds that are both genotoxic and carcinogenic and present in a variety of PFS was tested and applied. The MOE is the ratio between a point of departure for carcinogenesis, preferable the so-called BMDL10 (the lower confidence limit of the benchmark dose (BMD) causing 10% extra cancer risk above background levels) and estimated human exposure (EDI) (estimated daily intake); a MOE > 10 000 is considered of low concern for risk management. By applying the MOE approach data obtained showed that some alkenylbenzene-containing PFS are of high priority for risk management, whereas others were shown to contain such low levels of the alkenylbenzenes of concern that the resulting intake and thus the MOE were > 10 000 indicating the corresponding PFS to be of no concern. It was found that some of these PFS containing basil, fennel, nutmeg, sassafras, cinnamon or calamus, or their essential oils as the main ingredient, contain relatively high levels of alkenylbenzenes resulting in high EDI and low MOE values. Some PFS had MOE values equal to 1 indicating that the daily dose of alkenylbenzenes resulting from their use would equal the BMDL10 in rodent tumour bioassays. For such cases it seems important that risk management actions be implemented.
Also the concept of Threshold of Toxicological Concern (TTC) was investigated. It was shown that, when evaluating the PFS cited above, the risk management conclusions derived when using the TTC were similar to those based on the MOE approach. It was concluded that when data are available, the MOE is the preferred approach because, in contrast to the TTC, it facilitates ranking of the different PFS in terms of priority for risk management. A Mode Of Action (MOA)-based approach was applied using physiologically based kinetic (PBK) modelling for the alkenylbenzene elemicin in rat and human allowing estimation of possible implications for risk assessment. This was performed to investigate the possibility to use an MOA-based approach to perform a risk assessment on a PFS ingredient for which only limited experimental animal data are available. Using this approach, a risk assessment of elemicin appeared possible. The influence of the matrix of PFS for the risk assessment was considered. It was argued that tumour data used to estimate the MOE values obtained from long-term rodent studies with the pure compounds may not represent a good starting point for the risk assessment (BMDL10 and subsequent MOE calculation) of the same ingredients when dosed in the presence of the natural food matrix. It was demonstrated that PFS may contain ingredients able to inhibit the bioactivation of molecules, such as the genotoxic alkenylbenzenes. Whether such inhibition would also be relevant at realistic low dose human intake remains to be studied on a case-by-case basis, also taking the possible inhibition by other dietary ingredients into account.
New bioassays for risk assessment of PFS were developed and tested, including a panel of in vitro bioassays for studying possible beneficial or adverse health effects of PFS. Green tea (Camellia sinensis) and alkenylbenzenes were selected as the model PFS ingredients to be tested. In these experiments the possible metabolic changes were taken into account. It was concluded that modern mechanistic assays could be used to screen for both beneficial and adverse health effects of PFS. Such assays can often already give an indication of the potential in vivo activity of the PFS, especially when sound data on metabolic conversions (including bioactivation) and exposure data are taken into consideration. In other cases follow-up studies using more complex assays may be needed to assess the mode of action leading to potential beneficial and adverse effects. Together with data on traditional use, these approaches might provide a way to study the potential beneficial and adverse health effects without the use of expensive and ethically controversial animal models.


4. Adverse effects in humans (WP4)
In this area, PlantLIBRA reviewed scientific data and listed the most important plants and PFS involved in adverse effects and poisonings in humans. Special consideration was devoted to misidentification of botanicals or interactions with the bioavailability of pharmaceuticals or nutrients. A systematic literature review was performed, articles were assessed with a particular focus on the WHO classification for causality (association between exposure and effects), extracted and systematically evaluated and introduced into the ePlantLIBRA database (see point 6). In parallel, thanks to a retrospective multi-centre study involving European Poisons Centres, cases of adverse effects or poisoning due to consumption of, misidentification of or interactions with conventional drugs were collected. Methods to measure biological markers of exposures/adverse effects of plants and plant food supplements were searched for and reported. Biological samples (urine, blood) were taken from new clinical cases with adverse effects or poisoning due to consumption of botanicals or PFS, in a prospective multicentre study. New or modified analytical methods have been developed to allow a fast identification of exposure to specific botanical ingredients; among others, two methods (for blood and urine) were set up for a fast identification of exposure to Ginkgo biloba (ginkgo) seeds. A network was created between PlantLIBRA partners and European Poisons Centres, establishing a further alerting system based on the websites of PlantLIBRA (www.plantlibra.eu) and the EAPCCT (www.eapcct.org).

5. Models for Risk and Benefit Assessment (WP5)
The objective was to develop an integrated tiered approach to risk-benefit assessment (RBA) of PFS and to apply it to selected case studies, i.e. botanical preparations used in PFS, on the wiki-based platform Opasnet (http://en.opasnet.org/). The development of the methodology included also the setup on an online tool to estimate the risk of single compounds present in the botanical preparations. Overall, the methodology will represent a tool for improving science-based decision-making on PFS.
Starting from a critical review of RBA methodologies existing in the food and medicine sector and taking into account the specific features of PFS, PlantLIBRA developed a tiered methodology tailored for PFS. Three approaches were reviewed: the EFSA guidance for food RBA published in 2010, the approach developed within the EC-funded project BRAFO and the RBA methodology for herbal medicines of the European Medicines Agency. These approaches were also tested for their applicability to PFS on the five case studies selected: 1) Cinnamomum verum J.S. Presl, dried bark, essential oil, 2) Camellia sinensis (L.) Kuntze, leaves, dried green tea extracts, 3) Foeniculum vulgare Miller sp. vulgare var. vulgare, fruit, essential oil, 4) Ginkgo biloba L., dried leaves, acetone and water extract, 5) Hypericum perforatum L., dried flowering tops, ethanol or methanol extract.
The RBA methodology for PFS was developed taking into consideration the results of the review process. First, the primary unit of the assessment was defined as the botanical preparation, defined as plant species, plant part and preparation method, and the bioactive compounds present. Secondly, the different steps of the procedure were defined and described: 1) scope of the assessment, 2) exposure assessment, 3) identification of hazards and positive health effects, their characterization and the characterization of risks and benefit, 4) evaluation of health impact caused by PFS by a framework for evaluating the quality and strength of the evidence of the health effects and common currencies, depending on data availability.
The conceptual framework of the RBA methodology for PFSs was discussed among partners during in-person meetings. In addition, an eternal expert advice was required because of the controversy existing in relation to RBA of PFS, specifically on the appropriateness of a direct-risk benefit comparison, which appeared not to be accepted by the scientific community. Experts provided their feedback on the topic, which was used to finalise the methodology and proceed with the application to selected case studies.
Highlights from the developed methodology are the setup of a model for estimating intake using data from market research companies; the application of the Observed Safe Level (OSL) approach to identify, from human trials, a dose of botanical not causing adverse effects in human subjects, and the development of an online tool for estimating the risk of single bioactive compounds.
An important result of PLantLIBRA was the development of an online tool to estimate the risk of single compounds present in botanical preparations. The tool was developed on the Opasnet web-platform and it is named "Compound intake estimator". This tool relies on data on composition of botanical preparations under assessment, collected from literature and uploaded on the Opasnet platform, and on health guidance values available for the identified bioactive compounds. With the tool it is possible to calculate statistics such as mean or median value, minimum and maximum values for each compound identified in the botanical preparation under assessment and for each extraction technique reported. The second, and more important, feature is that the "Compound Intake Estimator" allows comparing the intake of single compounds, at calculated level of consumption of the botanical preparation, with related health guidance values such as acceptable daily intake (ADI), tolerable daily intake (TDI), lower confidence limit of benchmark dose (BMDL10), or threshold of toxicological concern (TTC), depending on what data is available for each compound, and to obtain a distribution of the risk estimate for each compound. The tool was developed and applied to three of the case studies selected by the consortium: 1) Cinnamomum verum J.S. Presl, dried bark, essential oil, 2) Foeniculum vulgare Miller sp. vulgare var. vulgare, fruit, essential oil, 3) Ginkgo biloba L., dried leaves, acetone and water extract.


6. ePlantLIBRA (WP6)
The objective "To develop, test and apply a sustainable integrated meta-database of biologically active compounds" has been achieved in the form of the ePlantLIBRA database http://ePlantLIBRA.eurofir.eu/. The database is easily searchable with retrievable data from peer-reviewed publications on chemical composition, botanical information, beneficial bioactivity data and adverse effects, as well as potential contaminants in PFS, into a single platform to enable PFS risk-benefit assessments. ePlantLIBRA is a valuable resource for food regulatory and advisory bodies, risk authorities, epidemiologists and researchers interested in diet and health relationships, as well as product developers within the food industry. Because ePlantLIBRA combines literature on the beneficial and adverse biological effects of PFS in one place, the tool is particularly useful in the risk assessment of botanicals for use in PFS, using the approach described by EFSA. It will also be of use in the public health domain in the estimation of exposure to bioactive compounds in PFS from food consumption surveys. The database has been designed to accommodate continual expansion, as research develops to ensure that it remains a current and useable resource. The basic structure and function of the existing eBASIS (Bioactive Substances in Food Information System), developed by EuroFIR (http://ebasis.eurofir.eu) have been retained in developing the ePlantLIBRA database. However, systems have been updated to enable the input of data on bioactive compounds in PFS. In addition, the biological activity component of the database has been extended to include case-reports of adverse events from the consumption of PFS, as well as literature on beneficial effects. Detailed work has been carried out to connect the plants in ePlantLIBRA with the corresponding commodities in the database (www.moniqa.eu) with links through to appropriate pesticide maximum residue levels (MRLs) and other contaminants. Furthermore, data on global occurrences of safety issues in traded commodities from the HorizonScan database has been linked to the ePlantLIBRA plants to provide current examples of the range of residues and contaminants likely to be encountered in such commodities.
Summarizing, the database contains over 700 references for evaluation yielding over 6500 data inputs covering over 50 botanicals and 40 different PFS. The papers added to ePlantLIBRA during PlantLIBRA project are distributes as listed below:

• For adverse effects in humans: 318 papers sourced, 210 papers evaluated, 41 plant covered, 41 PFS covered;
• For beneficial effects: 218 papers sourced, 211 papers evaluated, 33 plant covered, 41 PFS covered;
• For composition: 350 papers sourced, 280 papers evaluated, 78 plants covered, 35 PFS covered.

Links and information on grey literature, books and related databases have been included in ePlantLIBRA to increase its value to users. Additionally for all plants within ePlantLIBRA scientific name, plant family, synonyms, common name in 15 European languages, colour photograph identification and links to the Germplasm Resources Information Network (GRIN, http://www.ars-grin.gov) have been included. The ePlantLIBRA database has been presented at many events, meetings and webinars and discussions resulting from these sessions have helped shape the future plans for the database. Following discussions with the Policy Advisory Board, Scientific Interest Group and the Sustainability Task Force for a long-term durability plan has been produced including income generation schemes for use of the database including pay-to-view, membership and other fees, as well as options for linking to the EuroFIR and MoniQa database platforms such as micro sites.


7. Analitycal methods (WP7)
The aim was to compile and create data as basis of a respective quality assessment, starting from molecular genetic identification of the plant material via analytical methods in phytochemistry, contaminant and residue control up to the determination of biological markers of exposure. All these methods should be precise, validated, easy to handle and applicable in all stages from herbal (starting) materials via extracts to the final PFS products including in vitro as well as in vivo studies as far as biomarkers are concerned.
DNA based identification ('DNA barcoding') is since a few years the 'gold standard' for authentication of biological material to avoid misidentification, admixtures and adulterations. Although this is very important in herbal materials, for many plant species appropriate methods and SOPs do not yet exist. Starting materials for PFS have to be controlled quite often in the form of crude drug or extract which needs specific preparation. As regards phytochemistry, the project compiled and applied on one side existing methods from the phytochemical literature, Pharmacopoeias and International Standards, on the other side also new methods were developed and tested on validity. When testing the different materials it became obvious that in general a large infraspecific variation exists influencing the quality and efficacy of PFS considerably. This is extremely important if the species in question might contain substances of health concern as e.g. alkenylbenzenes or pyrrolizidine alkaloids (the latter ones recently detected in several herbal materials contaminated with weeds such as e.g. Senecio spp., containing these toxic molecules).
Innovative analytical methods were set up in the field of environmental contaminations, pesticide residues and mycotoxin testing. Electrochemical biosensors open new possibilities of rapid and cheap microchip tools, as well as immunochemical methods (ELISA tests) are applicable for mycotoxin contamination. Monitoring of herbal materials will now be much easier for safety of the consumer. Finally, a thermoluminescence method was developed for the detection if the plant starting material was irradiated (to reduce microbial load of herbal material not proper produced and stored according to GAP for medicinal and aromatic plants).
Due to the above mentioned quality problems of plants and extracts used in PFS it was our aim to collect and investigate samples from different steps of production. For this purpose the respective industry was contacted in Europe, India, China, South Africa and South America, and more than 500 samples in total (herbal material/crude drug, extracts, PFS) were at the disposal for all investigations.
Working in close connection with other workpackages, involved in the identification of biomarkers of exposure for beneficial and toxicological effects, a collection of methods for the quantification of these biomarkers has been prepared through a systematic review of data found in the literature, and new methods suitable for the identification of biological markers have been developed.
A final goal of this area of research was the creation of an international network of laboratories capable to provide reliable expertise in identifying and quantifying bioactive substances, contaminants and irradiation treatments in herbal material and plant preparations. The list of laboratories having expertise in the field of PFS and included in the network is at disposal at www.plantlibra.eu.

8. Consumers' perception (WP8)
The overall aim of this area of research was to map the consumer PFS information environment, to identify and model consumer responses to information provided on the benefits/risks of PFS and to contrast this with stakeholder perceptions. A literature review was undertaken to ascertain what is known about the users of PFS and the determinants of usage. In general PFS use is associated with being female, middle-aged, having better education and a higher income. Consumers' motives for using PFS vary and include the maintenance of overall general health, prevention of disease, a means of compensating for an unhealthy lifestyle or treatment for specific diseases and conditions. A number of key informants were interviewed in three countries, selected from different sectors relevant to PFS. Stakeholders (including mainly companies producing PFS) perceived that PFS are used by a range of consumers for a variety of reasons, with women often being the primary consumer. Motivation for using PFS includes maintaining health, preventing disease, treating illness or compensating for an unhealthy lifestyle. Often the use of these products is driven by their 'natural' origin. Stakeholders identified product packaging as the principal source of information for consumers, though current legislative requirements limit how much information can be provided. Other means of communication included trained staff, advertising, the internet, product catalogues and the print media. Barriers to the effective communication of information included a perceived lack of scientific evidence on the efficacy of PFS, the difficulty in targeting messages for the range of consumers using PFS and the regulatory environment. Stakeholders perceived consumer awareness of the benefits and risks associated with PFS to be poor with many consumers viewing PFS as not potentially harmful, because of their 'natural' origin and consequently may ignore dosage instructions or fail to consider their interaction with conventional medicines. Stakeholders also referred to the confusion some consumers may experience in distinguishing PFS as food or medicine.
The product packaging for five plants was examined in detail with ten products purchased per plant from different retail outlets. The plants were Camellia sinensis folia (green tea), Ginkgo biloba folia, Passiflora incarnata herba (passion flower), Cassia acutifolia (Senna) and Hypericum perforatum (St John's Wort). The patterns that emerged, in terms of the format of the PFS and type of packaging may be conveying certain messages to consumers e.g. capsules/tablets/blister packs may imply medicine whereas tincture, powder, liquid may imply herbal/natural/food. The PFS packaging themes were predominately content or rather plant related with a smaller proportion relating to purpose, i.e. benefit related to most packaging was plant-related rather than benefit related, though this varied between the five plants studied and the three countries. Most of the images found on these products were plants, followed by images of humans/human body parts dominated on packaging, this again varied between the five plants studied and the three countries. Imagery evoking nature and naturalness were found to be a key concept used in the marketing of plant food supplements. Many products used more than one type of image to convey information about the product to the consumer. The analysis of the types of claims on PFS packaging found direct claims (i.e. increase, reduce, lower, maximise) to be most common. The most common type of benefit found was "Help restore natural functioning", this again varied between the five plants studied and the three countries.
In order to better understand how PFS are reflected in the media, 20 articles relating to PFS were selected for analysis in each country from print media. Articles covered a wide range of PFS and included many of the better known supplements.. Generally articles were either 'plant' focused or 'solution' focused with shorter articles reporting on efficacy or lack of efficacy of specific plants, based either on research findings or the opinions of experts, while longer articles encompassed broader and more balanced discussions of the evidence for and against PFS, with a few articles addressing the complexities of new EU regulations. Newspaper articles tended to present factual evidence, whereas magazines tended to begin with the implicit assumption that PFS had some value and that readers were already potentially interested in PFS. All articles referred to the benefits of PFS generally relating to health maintenance or illness prevention. There was less reference to the risks associated with the consumption of PFS and generally those identified were the possible reaction of PFS with conventional drugs or potential side effects of PFS. Across countries there were differences in the sources of information used in the articles.
The role of key elements of PFS packaging in determining perceptions of the risks and benefits associated with PFS was examined in a study with consumers (users and non-users of PFS) and stakeholders (companies involved in import/export or production of food supplements containing botanicals). A number of products based on Ginkgo biloba were used in the study with several packaging cues, (product name, the product image, claim type and warrant type) varied in a systematic way. Overall 81.9% of participants' ratings of the benefits were influenced by factors on the packaging, with significant effects of product name (plant vs. functional) as well as a main effect when there was an explicit claim present. Similarly 69.6% of participants' ratings of product risks were influenced by factors on the packaging with significant group differences relating to both the presence of an explicit claim and the use of a scientific warrant.
In a second study the assumptions and beliefs users and non-users of PFS hold about the role and efficacy of PFS and where this knowledge comes from were elicited via focus groups. Users of PFS perceived them to be natural products and hence unlikely to cause any harm; PFS were regarded as natural alternatives to conventional medicine, particularly in the case of less serious conditions and were also perceived as having a role in compensating for demanding and unhealthy lifestyles or to address dietary deficiencies. Trust emerged as an important issue for both users and non-users of PFS. A primary source of information for users was 'word of mouth', with products often being recommended by family and/or friends. The internet was another popular source, though many reported treating information online with caution.
In keeping with previous findings product packaging that evoked naturalness and simplicity and used the name of the plant in the product title was preferred by users of PFS, rather than complex multi-ingredient products with names invented to suggest intended benefits. Users wanted more comprehensive information on pack particularly relating to benefits, instructions for use, ingredients, warnings and possible interactions.
A number of experimental studies employed memory-based measures to examine consumer's beliefs about plant food supplement products in which packaging was manipulated. The findings indicate that images on plant food supplement labels can implicitly prime the formation of expectations regarding the products' functions. These images can thus act as a source of suggestion, perhaps in some cases making consumers less aware of the real benefits that the products offer.

9. Dissemination (WP9)
Dissemination activity has been primarily aimed to information and training. It's difficult therefore to speak about scientific or technical results or foregrounds. However, we can draw from this experience a series of reflections on the themes and topics that have affected mainly the participants in the meetings organized by PlantLIBRA. All events organized both in Europe and in other three continents (Asia, South America, Africa) went well and the topics have attracted much interest.
It is no coincidence that the aspects of the quality and safety of Botanicals for use in food supplements are at the centre of an intense debate not only at European level which involves researchers, policy makers and industry. The tools and methodologies currently available and in use to assess the quality and safety of a chemical ingredient are only relatively suitable for complex matrices such as the Botanicals are. So further and more detailed studies would be needed to deal in a specific and more scientifically correct way the question of safety and quality of these products.
Plantlibra has just begun a path that must be continued in both the search scope and application. More prepared and competent staff must deal with research but also must be trained to work in enterprises, especially small ones. Only academic preparation is not sufficient. The new professional who must deal with the quality and safety of these products, which occupy an increasingly large area of the market and that can have a significant impact on consumer health, should be able to understand the concrete problems, to have the opportunity to do internships at companies, understand the questions and learn to manage them. Only by comparison and synergy between the world of research and academia and the business world is it fully possible to form such professionals, also acting through the exchange of experiences among different cultures and countries. Especially from international cooperation activities and thanks to events held in non-EU countries, PlantLIBRA had the opportunity to discover the amazing and still largely unknown wealth that the plants are and that could be a source of supply for health products of great importance.


10. Policy implications (WP10)
The objective of this area of activity was to provide the policy foundation of PlantLIBRA. This includes identifying priorities to be addressed from a Regulatory perspective and framing the work undertaken and the outcomes in the policy context, at the level of the EU and internationally. To this end the following activities have been undertaken. Several reports on policy implication have been prepared during the four years of the project.
Input for and conclusions from these reports have been gathered from and discussed with the Work Package Leaders at several occasion, an in particular during the various project meetings organised by the Project. They have also been the core of discussions with the Policy Advisory Board (PAB) at the four meetings that were organised between 2011-2014 and have been presented at several dissemination occasions.
One of the major achievements of PlantLIBRA in this context was the creation and establishment of a Policy Advisory Board (PAB), where all EU regulators and selected international agencies from Argentina, China and United States were invited to participate and contribute feedback on the project. This PAB was successfully established with active participation of botanical officials from 20 EU Member States, China, Norway and United States. Several physical meetings have been organised during which the PlantLIBRA Work Package Leaders presented their work and had an exchange with PAB members.
Although the project was carried out in the EU mainly, it had partners from other regions and the policy implications at the international level were also analysed and reported, and potential impact is evaluated qualitatively, both at the outset and at the end of the project. The final policy implications report, integrating the whole work done, contained two parts: the first part described the policy context and covered a brief update on the legal situation of botanicals in the EU and in selected regions outside the EU; the second part of the report covered the potential policy implications on the basis of the project's outcomes and presents conclusions and recommendations. Both parts have been developed with the help of the PAB and WP Leaders.

Potential Impact:

In a time where the average life span is increasing, plants can be a useful tool to increase the welfare of the elderly and to reduce the use of drugs. A better understanding of how much the world of nature would be able to offer in terms of improving the quality of life is desirable if not necessary.
Food supplements are foods regulated by Directive 2002/46/EC and intended to complement the diet with vitamins, minerals and other substances with a physiological effect. Only vitamins and mineral salts are regulated by Directive 2002/46/EC. As consequence the other substances are freely organized at National level and there are food supplements which contain substances with nutritional or physiological effects, some of which are also used in products classified as medicines in certain Member States, reflecting the different (Member State) interpretation of the boundary/borderline between food and medicine and the lack of harmonization of substance definition and classification. The variety of distribution and sales channels doesn't allow having availability of detailed data.
The most recent and official survey made in 2009, described a EU market valued at between €8.2 billion and €8.6 billion. Almost 50% of this value is referable to food supplements with 'other substances'.
There is a lack of data on the total number of businesses in the sector across the EU, mainly because of its diverse nature. The total number of businesses involved in the market for Plants food supplements (excluding retailers) is probably about 5,000, constituted by several categories of 'economic players' as Manufacturers/suppliers of ingredients; Sub-Contract manufacturers; Product manufacturers and retailers; Product manufacturers and distributors; Distributors. The vast majority of these businesses are small, medium-sized enterprises (SMEs).
To this framework, it is essential to have people capable of dealing with the Botanicals issue, Botanicals specialists who can address the issues related to plants complexity for identification and quality, as well as those of safety and able to evaluate the effects on health and PlantLIBRA contributed collecting and producing new knowledge and expertise. The main conclusions that can be drawn from the project are discussed below.

1. Complexity of the field
Gathering data on the use of PFS is difficult for various reasons. The future work in this area should always take into account the diversity of the market and the existing differences between countries in terms of distribution channels, regulatory environment and the considerable diversity in terms of products and product composition. This work could be simplified if such aspects are addressed on a EU level with a common definition and requirements that enable the same botanicals to be sold in all Member States in the same way. In addition, the notification system that is in place today in the majority of Member States allows recording the composition at product level. If this could be done on the EU level, a database could be established that could be linked to intake survey data.


2. Source of data for risk/benefit assessment
The research undertaken has highlighted that data from human intervention trials is largely missing, whereas much knowledge is available from experimental, animal and observational evidence for many plants, especially for those that have a decades-long traditional use. Models for benefit assessment should therefore be able to judge the extent to which benefits are supported by the totality of the evidence available and raise the importance of in vitro assays. Using a number of in vitro toxicological models, PlantLIBRA has tested the safety of compounds that have been shown to be genotoxic and carcinogenic in animals. Such models are useful to identify causes for concern and may also reduce the need for animal testing.
The consensus workshop organized on this topic has made and will make an impact in the scientific community through its recommendations for in vitro assay design and interpretation. Improved assay design will mean that it is much more likely that human intervention studies will give positive results without performing intermediate studies on animals. The impact would be to reduce the animal experimentation necessary in labs throughout the world. One of the recommendations is that the first step in digestion/metabolism is considered when designing in vitro assays for PFS. Numerous papers have been published. Assessments and reviews were published in the areas of cardiovascular function, infectious diseases and immune function. Previously there was considerable confusion in the grading of evidence, in the role of traditional use in the definition of endpoints, markers, and study design, and reviews coming from PlantLIBRA work have remedied this situation for selected combinations of PFS and benefit areas. This broad panel of papers ensures a wide coverage by many different sections of the scientific community.
The main (potential) impact in the area of risk associated with PFS originated from the results on the risk assessment of the supplements containing alkenylbenzenes, showing that some PFS do contain these genotoxic carcinogenic alkenylbenzenes but at levels that do not raise concern (MOE > 10 000), indicating that the presence of a genotoxic carcinogen as such is not a good indicator for making risk management decisions.
However, some of these PFS, especially the PFS consisting of essential oils that contain high levels of alkenylbenzenes, do raise concerns and the MOE values below 100 indicate a high priority for risk management. Some PFS had MOE values equal to 1 indicating that the daily dose of alkenylbenzenes resulting from their use would equal the BMDL10 in rodent tumour bioassays. For such cases it seems important that risk management actions will be implemented.
The research has also shown that toxicological data are mostly derived from studies with purified compounds, in which possible matrix effects are not considered. Extrapolation to humans should therefore be considered with caution and in the light of the totality of information available.
Considering the adverse effects in humans, data deriving from scientific literature and in particular from case reports showed limitations, since most of them reflected acute exposure.
Data collected by the project added important knowledge for health professionals, food experts, and Poisons Centres specialists and clinical toxicologists. As such, they provide helpful data for the risk assessment of these botanical nutrients and PFS, and contribute to risk management in the food sector.
As PFS-related adverse effects are relatively infrequent issues for Poisons Centres, and because most cases are mild and show a benign clinical course, the periodic occurrence of clinical cases and the severity of some of them, and the increasing popularity of PFS, the access to reliable and detailed data for the clinical evaluation of patients presenting with such complaints is helpful, even crucial in some (critical) cases. Nowadays, only a few comprehensive, structured, and easy-to-access sources of information on this topic are available to poison information specialists. For this reason, the access to the PlantLIBRA database will probably be highly beneficial to Poisons Centres, and may contribute to better care of affected patients.
The alert network established between the PlantLIBRA group members and the EAPCCT (the European association of Poisons Centres) will facilitate the exchange of information between the two, particular if there is a need for urgent dissemination of information ("alerts").
The methodologies developed on risk and benefit assessment can, and to some extent have contributed to, an increasing science-based decision-making by authorities and food chain operators in the field of PFS. The methodologies have to some extent overcome fragmentation across disciplines, and proposed a framework for integrated "risk-benefit sciences" for PFS.

3. Quality control of PFS and analytical methods
PlantLIBRA has contributed considerably to the knowledge on analytical methods that are relevant for quality control of PFS. In particular the inventory of methods and the network of laboratories are useful tools that should be maintained and kept updated after the project closes.
One of the objectives of the project was to contribute to setting-up and implementing a quality assurance system for PFS by developing and compiling analytical methods and respective data concerning quality, safety and beneficial/adverse effects. A high number of well-established and novel methods have been collected, validated and applied to a number of crude preparation, extract and PFS samples. The novel methods developed within the project - genetic and phytochemical identification of the plant material and derived products, innovative electrochemical and biosensor methods for contaminant and residue determination as well as methods for the determination of biomarkers of exposure - will be published and the whole set of information is at the disposal of all parties concerned. Of special public interest might be the problem of confusions and adulterations, the range of quality and contaminations within PFS and finally the identification of biomarkers for beneficial and toxicological effects. All data achieved will continuously be disseminated by oral presentations in conferences, seminars, University lectures and public events as well as through specific health media.
From the socio-economic point of view it might be discussed if this category of products is appropriately placed under food and covered by the general food legislation. There is a high innovative potential also for SMEs working in this sector, much higher than e.g. in traditional herbal medicine, and the impact on well-being and healthcare is obvious since the beneficial effects, but also possible risks of functional (secondary) plant products are better understood over the last years. Due to this fact, the 'hot topic' quality of herbal materials and extracts for PFS, discussed in several meetings of the Europe Botanical Forum, the Society for Medicinal Plant and Natural Product Research and at PlantLIBRA dissemination Conferences resulted in the demand for a respective Guideline which was developed under participation of the parties concerned. Gratefully the European Botanical Forum took charge, and the Guideline "Quality of Botanical Preparations: Specific Recommendations for the Manufacturing of Botanical Preparations as Food Supplements, including Extracts" was issued in 2013 being a great advantage for quality assurance of PFS. As a consequence, a product category between 'food' and 'pharma' could also be taken into consideration when discussing on health claims of PFS in the near future.

4. ePlantLIBRA
The work undertaken has established the e-PlantLIBRA database as an easily searchable platform for retrieving data on chemical composition, botanical information, beneficially bioactivity data and case-reports of adverse effects as well as potential contaminants.
The primary users of ePlantLIBRA will be the regulatory affairs sector (e.g. assessment of PFS supporting health claims/risk assessment), food industry (e.g. evaluation and development of PFS) and researchers and epidemiologists. In addition, PFS data and information are widely used to underpin academic research into links between diet and health. Research outcomes are likely to influence policy at a national and/or international level (e.g. EU, EFSA & globally) and policy may dictate future dietary monitoring programmes or research.
Following discussions with Policy Advisory Board, Scientific Interest Group and the Sustainability Task Group a long-term durability plan has been developed including income generation schemes for use of the database including pay-to-view, membership and other fees and options for linking to the EuroFIR and MoniQa database platforms such as micro sites. An added value suggested by the PAB includes new links to botanical lists such as BELFRIT and Kew Garden Medicinal Plant Names Services (http://www.kew.org/science-conservation/research-data/resources/medicinal-plant-names-services) further links to the opasnet site (http://en.opasnet.org/w/Main_Page) and additional services such as links to conferences and experts.
As well as the ePlantLIBRA database itself, other key services to users and stakeholders via membership will be:
• Improve delivery of PFS data and expertise for research and policy describing bioactive compounds in botanicals and herbal extracts with putative health benefits and adverse effects in PFS in Europe and globally;
• Support cooperation and exchange, and increased awareness of PFS with other information providers, laboratories, regulators and industry to provide a forum for discussion and cooperation in Europe and globally;
• Initiate new coordinated training and support in conjunction with other research projects for users from the research community, health professionals, food and biotech industries, government agencies and departments.
The ePlantLIBRA application is developed in classic Active Server Pages (ASP, using JavaScript for e.g. field validation and generation of dynamic HTML. All input screens are generated by an internal Content Management System, facilitating an advanced help system and online modification of input forms. The system is hosted on a Windows server with IIS and Microsoft SQL Server, hosted by EuroFIR. Together with the development of the ePlantLIBRA database, PlantLIBRA members have presented 18 talks and 7 posters at international conferences, carried out 3 webinars, and prepared 2 media/publicity publications. 3 peer reviewed publications are in progress, 2 submitted, 1 accepted and 1 published. The database can be found at: http://ePlantLIBRA.eurofir.eu/; Log in is by individual user name and password contact Jenny Plumb Jenny.plumb@ifr.ac.uk or Carlos Ramos secretariat@eurofir.org for log in.
A generic open user account that can be used by all to access the database is available User name: ePlantLIBRA@ePlantLIBRA.eu Password ePlantLIBRA

5. Models for risk/benefit assessment
Risk-benefit models are often hampered by the paucity and poor quality of data available.
PlantLIBRA approach is in line with the request to combine risk and benefit assessments, so that health claims are not authorised when safety of a substance has not been assessed. The potential impact of the methodology has been enhanced through frequent interaction with governmental decision-makers through the PAB.
The impact of this study on safe use of PFS is indirect. However, the insistence for a science-based approach (as opposed to a tradition-only approach) emphasises the importance of including carcinogens and genotoxic compounds in the assessment, even if history of safe consumption is available and no data on the botanical preparation as a whole has been produced. This may ultimately lead to the setting of maximum levels for genotoxic compounds in botanical preparations such as fennel tea, and to the banning of some basil extracts. Importantly PlantLIBRA has shown that all PFS botanicals of economic relevance, which were assessed, have enough variability - due to natural variation and extraction techniques – so that safe extracts, with genotoxic compounds at levels of low concern, can be produced.
Therefore, the work done in risk/benefit assessment has not only provided better and tailored knowledge and tools, but has raised the awareness especially among authorities that such tools and knowledge are needed.
When published, the individual case-studies will also provide direct evidence for decision-making (e.g. setting maximum levels of contaminants) and priorities for further research.
The risk and benefit workshop organized during the project, the assessments themselves (to which several investigators participated) and the frequent interactions with scientists of the project -as well as policymakers- have effectively brought together scientists with a background in nutrition, physiological sciences, toxicology and botany to work out together the framework and the details of an individual assessment of botanicals. Therefore, all these scientists now see botanicals in a more multidimensional manner, where benefits, risks, consumer expectations, quality issues and regulations are intertwined and need to be taken into account when performing scientific work on botanicals. As mentioned above, PlantLIBRA has formed consensus around novel models for risk-benefits and explained how they can easily be linked to standardized composition data (ePlantLIBRA) and actual intake data (Survey).
PlantLIBRA evaluated the use of a common currency for risk-benefit assessment for PFS by authorities and in the scientific literature, by means of exploration, comparison, and application to risks, benefits and risk-benefits of alternative options and dissemination of the process through workshops and publications and input to the identification process by policy-makers, through the PAB. However, it has been found that a common currency, such as Disability-adjusted life years (DALYs) can only result in large uncertainties in the risk estimates, even when data are available for a quantitative assessment; this is rarely or ever possible for benefits. Moreover, while the workshop, organized by the project, showed some support for the idea of quantitative estimates even in the case of large uncertainties, policy-makers and stakeholders have not reported that this approach would be informative for their work. A narrative comparison of risks and benefits has gained more support. In this case, the impact of the project relates to further research in qualitative reporting of risks and benefits for policy-making rather than using more resources for attempting a quantitative evaluation.
The PlantLIBRA project has validated the idea that decision-making for PFS occurs in the regulatory process and along the food chain. Regulation in the EU occurs mainly at the national level, and the BELFRIT initiative - that aims to harmonise regulations in the field of botanicals in Belgium, France and Italy and was initiated by PlantLIBRA's Policy Advisory Board (PAB) members in the context of PlantLIBRA meetings - is testimony to the key role that national authorities have at this stage. The project has also identified safety and quality issues that are primarily the industry's responsibility. The BELFRIT project has focused on tradition as evidence of safety; it has been argued that, as in the case of traditional herbal medicines, tradition should be the basis for benefit assessment as well. The risk-benefit framework, which has evolved to take objectively evaluated tradition into account as a potential decision-making factor for policy makers, has shown the potential to bring together different perspectives, while emphasising science as the driver of the assessment. This approach has been appreciated by policy makers and scientists affiliated with scientific bodies, when presented at the PlantLIBRA PAB meetings; it has attracted attention for the treatment of uncertainty. Therefore, when the approach is published and the platform is made available to the public, it can be expected that it will help shape the assessment processes and decision-making mechanisms of authorities in the EU. As originally planned, the risk and benefit assessment platform provided by PlantLIBRA, which can be used in conjunction with the database, will be open for policymakers to use after publication rights have been used.

6. PlantLIBRA and consumers
The activities performed during the Consumer Survey will have an important socio-economic and societal impact derived mainly from the PFS consumer survey, which represents the first successful attempt to collect European PFS consumption data, until now non-existent. The PlantLIBRA PFS Consumer Survey is the ice-breaker that sets the basis for future work providing examples of all the design, methodology, material, implementation, database construction, data analysis, actual knowledge and its dissemination. The knowledge generated with this survey will benefit all stakeholders (companies involved in commercial activities related to PFS) in the sector, namely the consumers, policy decision makers, industry, researchers and academics, health professionals etc., by helping them in their decision-making for the consumption of these products or in the decision-making processes of their respective professional areas. The lessons learned through the conduction of the survey will guide future researchers interested in the study of PFS consumption. The work and results on bioactive compounds of fennel and tea represents an example of what can be done with the information provided by the survey and other PlantLIBRA work (ePlantLIBRA) in a particular country, again very useful for several stakeholders (researchers, industry, etc.).
PlantLIBRA examined the consumer information environment in relation to PFS using a number of different methodological approaches to ascertain what information exists for consumers and how they use that information. A key source of information for consumers is product packaging and stakeholders also view this as a principal means of communication with the consumer. The studies highlighted the importance of packaging in conveying information to consumers, both in terms of the nature of the packaging and the content, including claims. The perception that PFS are natural and thus inherently safe is reinforced by the use of packaging that enhances this through the use of certain images and presentation of the product. Furthermore our results provide insight for the first time into how consumers are using that information to assess the benefits and risks of PFS products. Information regarding the use of PFS appears to often be gained through consumers' social networks. A better understanding of the way information is shared between people within these networks would provide more insights with regard to the use of PFS. Data on patterns of online information seeking and information sharing could be used for this purpose. In countries where professionals play and active role in recommending PFS use there is a need to better understand their recommendation practices.
Participants across studies indicated that having more specific and detailed information about the intended benefits of the products as well as clearer information about exactly how the ingredients achieve these benefits would be helpful in assessing the benefits and risks of PFS. The information environment from which consumers draw PFS information appears to be complex and diverse. This in part relates to the wide diversity of plants being used and purported benefits. More work is needed to understand how consumers can best access up-to-date accurate information about the risks and benefits associated with PFS use and where the responsibility lies for ensuring that information provided to consumers through all channels of information is not misleading and likely to lead to patterns of use that are detrimental to health. There is a need for studies examining the processes by which consumers seek and utilize information regarding PFS. Further controlled studies where product information provision can be systematically linked to product consumption patterns are also warranted. Our work has also used robust memory-based measures, to gain a better understanding of consumer's use of product information. To date food labelling research has primarily relied on assessing consumers' understanding using explicit, often unreliable, self-reports.

7. New and old expertise at disposal
Partners of PlantLIBRA (both academic and SME) have gained expertise in risk and benefit assessment of botanicals that can be put to use in the marketplace. Such expertise, scarcely available on a global scale, was developed both in term of safety evaluators and technical experts (chemists, biochemists, pharmacologists, etc.), who will be at disposal for future research projects or risk evaluations. In this context, PlantLIBRA paid particular attention to young people, who were involved in experimental actifity, scientific meetings and training courses in search of new professional figures, not covered in the current educational systems of any University in Europe. In fact, the market requires people aware of the complexity of the plants taking in account not only the scientific bases but also the history of use of each plant and the expertise in managing (with the new and dedicated tools) the quality, safety and efficacy aspects.
The networks are further tools obtained by the project. The network of laboratories is at disposal for solving specific analytical problems, as well as the network of Poisons Centres could be consulted in case of adverse effects with complex or unclear clinical symptomatology.
The new knowledge and expertise, combined with a harmonization of the use of PFS in Europe, could facilitate the task of the authorities and give the same guarantees to all European consumers.


List of Websites:
www.plantlibra.eu