Publications

Phase One:Asp NG, Cummings JH, Mensink RP, Prentice A,Richardson DP (2003) Process for the Assessment of Scientific Support for Claims on Foods (PASSCLAIM) – Phase One: preparing the Way. Eur J Nutr 42 (Suppl 1):1-119.

Synthesis and review of existing processesProf. David Richardson (Chair), UKMr. Tage Affertsholt, DKProf. Nils-Georg Asp, SNF – Swedish Nutrition Foundation, SProf. Ake Bruce, National Food Administration, S Prof. Rolf Grossklaus, BgW, DMr. John Howlett, UKDr. Richard Ross, GlaxoSmithKline, UKDr. Hans Verhagen, Unilever Health Institute, NL (now at RIVM, NL)Dr. Volker Viechtbauer, Red Bull, A

Diet-related cardiovascular diseaseProf. Ronald Mensink (Chair), Maastricht University, NLProf. Antti Aro, National Public Health Institute, FINDr. Elly den Hond, Catholic University Leuven, B Dr. Bruce German, Nestle Research Centre, CH Dr. Bruce Griffin, University of Surrey, UKDr. Marja Mutanen, University of Helsinki, FIN Prof. Wilhelm Stahl, Heinrich-Heine-Universität Dusseldorf, DDr. Hans-Ulrich ter Meer, Nutrinova, D

Bone health and osteoporosisDr. Ann Prentice (Chair), MRC Human Nutrition Research, UKProf. Jean-Philippe Bonjour, Hôpitaux Universitaires de Genève, CHDr. Francesco Branca, National Institute of Nutrition, I (now at WHO Europe, DK)Prof. Cyrus Cooper, MRC – Southampton General Hospital, UK Prof. Albert Flynn, University College Cork, IRL Prof. Michele Garabedian, Hôpital Saint-Vincent de Paul, Paris, F Dr. Detlef Muller, Procter & Gamble, D Dr. Peter Weber, F.Hoffmann-La Roche, CH

Physical performance and fitnessProf. Wim Saris (Chair), Maastricht University, NLDr. Jean-Michel Antoine, Danone Vitapole, FDr. Fred Brouns, Cerestar R&D Centre, BDr. Mikael Fogelholm, The UKK Institute, FIN Prof. Mike Gleeson, Loughborough University, UK Dr. Peter Hespel, Catholic University Leuven, BDr. Asker Jeukendrup, University of Birmingham, UKProf. Ron Maughan, University of Aberdeen, UK Dr. Vladimir Stich, Charles University, CZ

Phase Two:Asp NG, Cummings JH, Howlett J, Rafter J, Riccardi G,Westenhoefer J (2004) Process for the Assessment of Scientific Support for Claims on Foods (PASSCLAIM) – Phase Two: moving forward. Eur J Nutr 43 (Suppl 2):1-183.

Body weight regulation, insulin sensitivity and diabetes riskProf. Gabriele Riccardi (Chair), University of Naples, IProf. Peter Aggett, University of Central Lancashire, UK Prof. Furio Brighenti, University of Parma, I Prof. Nathalie Delzenne, Catholic University of Louvain, BProf. Keith Frayn, University of Oxford, UK Dr. Stephan Theis, Sudzucker, DProf. Bengt Vessby, University of Uppsala, S

Diet-related cancerProf. Joseph Rafter (Chair), Karolinska Institute, SDr. Mirjam Govers, Numico Research, NLDr. Paule Martel, INRA, FProf. Beatrice Pool-Zobel, Friedrich-Schiller University of Jena, DProf. Gerhard Rechkemmer, Technical University of Munich, D Prof. Ian Rowland, University of Ulster, UK Dr. Jan Van Loo, Raffinerie Tirlemontoise – Orafti, B

Mental state and performanceProf. Joachim Westenhoefer (Chair), University of Applied Sciences Hamburg, DDr. France Bellisle, INSERM – Unité 341, FProf. John Blundell, University of Leeds, UKDr. Jan de Vries, Friesland Coberco Dairy Foods, NL Dr. Damian Edwards, Masterfoods, UKProf. Wolfgang Kallus, University of Graz, AProf. Berthold Koletzko, University of Munich, D Prof. Hely Tuorila, University of Helsinki, FIN

Gut health and immunityProf. John Cummings (Chair), University of Dundee, UKDr. Jean-Michel Antoine, Danone Vita pole, FProf. Per Brandtzaeg, University of Oslo, NProf. Philip Calder, University of Southampton, UK Prof. Glenn Gibson, University of Reading, UK Dr. Francisco Guarner, Hospital General Vall d'Hebron, EProf. Erika Isolauri, University of Turku, FINProf. Colette Shortt, Yakult, UKDr. Bernhard Watzl, Federal Research Centre for Nutrition, D

Phase Three/Consensus:Aggett PJ, Antoine J-M, Asp N-G, Bellisle F, Contor L,Cummings JH, Howlett J, Müller DJG, Persin C, Pijls LTJ,Rechkemmer G, Tuijtelaars S, Verhagen H (2005) Process for the Assessment of Scientific Support forClaims on Foods (PASSCLAIM) –Consensus on Criteria. Eur J Nutr 44 (Suppl 1): 1-30.

Achievements • The criteria provide a scientific framework that will facil-

itate the assessment of scientific support for claims onfoods.

• This, in turn, will enable the compilation of guidelines onthe preparation of submissions for regulatory review andapproval of claims on foods.

• By establishing a robust standard for the quality of scien-tific data submitted in support of health claims, thecriteria provide a basis for the harmonisation of theregulatory review and approval of such claims.

• The compliance of data submissions with the criteria willprovide consumers with the assurance that claims basedon the data are well founded and justified.

• By establishing a standard for the data to be submitted insupport of claims, the criteria will provide the agri-foodindustry with a stable frame within which new productsto meet consumer needs and expectations for foods withbenefits for health and well-being can be developed.

• Systematic use of the criteria will engender a moreinformed use of scientific data in support of claims.

Consensus GroupProf. Peter Aggett (Chair),

University of Central Lancashire, UK Dr. Jean-Michel Antoine, Danone Vitapole, FProf. Nils-Georg Asp, SNF – Swedish Nutrition Foundation, SDr. France Bellisle, INRA, FDr. Laura Contor, ILSI Europe, BProf. John Cummings, University of Dundee, UKMr. John Howlett, UKDr. Detlef Muller, Procter & Gamble, D Dr. Christoph Persin, Kampffmeyer Mühlen, DDr. Loek Pijls, ILSI Europe, BProf. Gerhard Rechkemmer, Technical University of Munich, D Ms. Sandra Tuijtelaars, ILSI Europe, BDr. Hans Verhagen, Unilever Health Institute, NL

Management teamILSI Europe

Dr. Laura Contor: Project co-ordinator

Dr. Loek Pijls: Scientific supervisor

Ms. Sandra Tuijtelaars, M.Sc.: Project manager

Steering CommitteeProf. Nils-Georg Asp (Chair),

SNF – Swedish Nutrition Foundation, S

Prof. Peter J. Aggett, University of Central Lancashire, UK

Dr. Jean-Michel Antoine, Groupe Danone, F

Dr. France Bellisle, INRA, F

Prof. Bruce German, Nestlé, CH

Dr. Detlef Müller, Procter & Gamble, D

Prof. Gerhard Rechkemmer, Technical University of Munich, D

Dr. Hans Verhagen, Unilever Health, Institute (now at RIVM), NL

For any questions about the PASSCLAIMproject, please contact:

Ms. Sandra Tuijtelaars, M.Sc.Project ManagerILSI Europe83, Avenue E. Mounier, Box 6B-1200 BrusselsTel: +32-2-771.00.14Fax: +32-2-762.00.44Email: stuijtelaars@ilsieurope.be

Or have a look at the project website onhttp://europe.ilsi.org/passclaim

This project has been carried out with financial support of theCommission of the European Communities, specific RTDprogramme “Quality of Life and Management of LivingResources”, QLK1-2000-00086 “A Process for the Assessmentof Scientific Support for Claims on Foods”, undercoordination of the ILSI Europe Functional Food Task Force,which is under the umbrella of the Board of Directors of ILSIEurope. ILSI policy mandates that the ILSI and ILSI branchBoards of Directors must be composed of at least 50% publicsector scientists; the remaining directors represent ILSI’smember companies. The opinions expressed herein are thoseof the authors and do not necessarily represent the views ofILSI or ILSI Europe. Neither do they necessarily reflect theviews of the European Commission and in no way do theyanticipate the Commission’s future policy in this area.

The ILSI Europe Functional Food Task Force IndustryMembers are: Acatris, Arla Foods, Barilla, BASF, BayerCropScience, Beverage Partners Worldwide, Campina,Cerestar, Coca-Cola, Colloides Naturels International, Danisco,DSM Food Specialties, DSM Nutritional Products, FerreroGroup, GlaxoSmithKline, Groupe Danone, VK Mühlen,Kellogg, Masterfoods, McNeil Consumer Nutritionals Europe,Nestlé, Nutrinova, Procter & Gamble, Puleva Biotech,Raffinerie Tirlemontoise – ORAFTI, Raisio Group, Red Bull,Royal Cosun, Südzucker, Swiss Quality Testing Services(SQTS), Unilever, Valio, Wild Flavors Berlin, and Yakult. International

Life SciencesI N S T I T U T E

passclaim

Background and objectives of PASSCLAIM

Much attention is being paid to health claims on foods, includingenhanced function claims, reduction of disease risk claims andalso nutrient function claims. Many food products on the marketcontain claims about health effects beyond the simple provisionof nutrients. One important basis for claims is the increasingnumber of reports of the effects of dietary components on bodyfunctions. However there was no scientific consensus as to howclaims based on these reports should be evaluated at Europeanlevel. With this background, ILSI Europe initiated the ConcertedAction ‘Process for the Assessment of Scientific Support forClaims on Foods’ (PASSCLAIM). PASSCLAIM has set principlesfor assessing the scientific support of health claims.

In 2003, the European Commission proposed its Regulation onnutrition and health claims made on foods, which states thatscientific substantiation of claims should be the main aspect to betaken into account for the use of claims. This regulation alreadyrecognised PASSCLAIM as a considerable work that should betaken into account when assessing claims.

PASSCLAIMProcess for theAssessment of ScientificSupport for Claims on Foods

Final Project Results

A European CommissionConcerted Action

Supported by the European CommissionDG Research, Thematic Programme 1Quality of Life and Management of Living ResourcesKey Action 1 Food, Nutrition and HealthEC-contact: Dr. Jürgen Lucas – Research DG, Food Quality (E. 2)

Coordinated byInternational Life Sciences Institute – ILSI Europe

Duration of the project: from 1 January 2001 to 31 March 2005Contract number: QLK1-2000-00086http://europe.ilsi.org/passclaim

The publications are downloadable from http://europe.ilsi.org/passclaimTo order free paper copies please contact publications@ilsieurope.be

In order to meet the project objectives, eight expertgroups were set up involving experts from academia,regulatory bodies and the food industry involving24 countries (see figure). Representatives of publicinterest groups were also approached. Seven of theexpert groups reviewed the scientific basis for claims invarious areas of health and disease with a focus onmarkers. One group critically evaluated existingapproaches to the scientific substantiation of claims.

The development of criteria for the scientificsubstantiation of claims, based on the results of the expertgroups, was the focus of the plenary meetings andConsensus Group (see figure). A first set of draft interim

criteria was discussed and modified at the first plenarymeeting. The interim criteria were then tested throughpractical application by the second phase expert groupsand further developed at the second plenary meeting.

In the third phase of the PASSCLAIM project, a ConsensusGroup was formed whose role was to refine and clarifythe criteria taking into account the input from the expertgroups, from the working groups and general discussionsat the First and Second Plenary Meetings, and fromindividual comments. The draft set of criteria proposed bythe Consensus Group was reviewed by the Final PlenaryMeeting and resulted in the consensus set of criteria.

Through an iterative process of discussion in expertgroups and workshops, a set of criteria (see box) whichdefine requirements for assessing the quality of scientificdata reporting the impact of foods and food componentson health and well-being have been proposed andprogressively refined.

The context within which a claim and the case made in itssupport should be assessed, involves:

• considering existing legislation and dietary guidelines;

• the need for review in the light of evolving science; and

• the comprehensibility of the claim to consumers.

These aspects are not thought to be part of the scientificcriteria reviewed by PASSCLAIM. They neverthelessprovide the background against which the scientificvalidity of claims should be justified.

2c An adequate duration of exposure and follow upto demonstrate the intended effect.

There are two aspects to this criterion. These areensuring (i) that there has been a suitable periodof exposure to the food or food component(period of intake), and (ii) that the duration ofobservation is long enough for the expected effectto occur, and, if necessary, to show that thebenefit is sustained.

2d Characterisation of the study groups’ backgrounddiet and other relevant aspects of lifestyle.

Humans are exposed to many potentially activesubstances in their diet. Intervention studiesdealing with one potentially functional agentshould determine whether or not the activesubstance is already present in the diet of thepopulation or in sub-groups of interest.Consideration should also be given as to whetherany substance provided by the diet couldpotentially interact with the tested substance toamplify or decrease its effect.

2e An amount of the food or food componentconsistent with its intended pattern ofconsumption.

The amount of food or food component that willbe tested should match its intended use and theway and frequency with which it will be eaten.Where dose response studies are performed, therange of doses must include the amount of foodor food component expected to be consumed.

2f The effect of the food matrix and dietary contexton the functional effect of the component.

This criterion relates to the influence that physico-chemical properties of the food, the diet and theintestinal luminal mileu would have on thestability of the active components and on theefficiency with which it is released from the food.

2g Monitoring of compliance with intake of food orfood component under test.

In any study of diet and health it is essential toknow the actual dietary intake of the subjects andto confirm that they have taken the food or foodcomponent in question in the right amount at theright time and over the specified period.

2h The statistical power to test the hypothesis.

3. When the true endpoint of a claimed benefit cannotbe measured directly, studies should use markers.

Whenever possible the claimed benefit should bemeasured directly. However, even though the ideal ortarget endpoint for human intervention studies ofhealth, performance and well-being may be identified,it may not always be measurable in practice.

4. Markers should be:

- biologically valid in that they have a knownrelationship to the final outcome and their variabilitywithin the target population is known.

Biological validity concerns the extent to which amarker reflects a certain health outcome of interestand the process leading to it. It is not dependent onthe technical competence of any individuallaboratory.

- methodologically valid with respect to theiranalytical characteristics.

Any laboratory performing measurements should becompetent to perform the measurements and tocertify that the values produced can be trusted, i.e.the method should be technically valid in itsperformance by that laboratory.

5. Within a study the target variable should change in astatistically significant way and the change should bebiologically meaningful for the target groupconsistent with the claim to be supported.

At the level of statistical significance, biologicalrelevance can be attached to very small changes in amarker, e.g. blood cholesterol levels, in which at thepopulation level, a few percent change has largeimplications for public health burden of coronary heartdisease. Conversely, a change of several tens ofpercent in immune function parameters on stoolweight, although statistically significant, may not haveany biological relevance.

6. A claim should be scientifically substantiated bytaking into account the totality of the available dataand by weighing of the evidence.

The evaluation of the available data may leave somequestions unanswered. In such cases it should beconsidered whether these questions need to beanswered by additional research, or whether or notthe evidence overall supports the proposed claim.

Synthesis andreview of existing

processes

Diet-relatedcardiovascular

disease

Bone health and

osteoporosis

Physicalperformance and fitness

Expert Groups

Body weightregulation,

insulin sensitivityand diabetes

Diet-relatedcancer

Mental state and performance

Gut health and immunity

Expert Groups

First Plenary Meeting: Interim criteria for claims

Second Plenary Meeting: Revised interim criteria for claims

Consensus Group

Final Plenary Meeting: Consensus criteria for claims

Jan 2001

Sept 2002

Oct 2003

Dec 2004

Mar 2005

Structure of the project

Results

The criteria:• emphasise the need for direct evidence of benefit to

humans in circumstances consistent with the likely useof the food in order for a case to be made;

• recognise the usefulness of markers of intermediateeffects when ideal endpoints are not accessible tomeasurement;

• stress the importance of using only those markers whichare of proven validity; and

• highlight the necessity of ensuring that the magnitudeand character of effects on which claims are based arestatistically and biologically meaningful.

Criteria for the scientificsubstantiation of claims

1. The food or food component to which the claimed effectis attributed should be characterised.

2. Substantiation of a claim should be based on humandata, primarily from intervention studies the design ofwhich should include the following considerations:

2(a) Study groups that are representative of the targetgroup.

2(b)Appropriate controls.

2(c) An adequate duration of exposure and follow up todemonstrate the intended effect.

2(d)Characterisation of the study groups’ background dietand other relevant aspects of lifestyle.

2(e) An amount of the food or food component consistentwith its intended pattern of consumption.

2(f) The effect of the food matrix and dietary context onthe functional effect of the component.

2(g)Monitoring of compliance with intake of food or foodcomponent under test.

2(h)The statistical power to test the hypothesis.

3. When the true endpoint of a claimed benefit cannot bemeasured directly, studies should use markers.

4. Markers should be:

- biologically valid in that they have a known relationshipto the final outcome and their variability within thetarget population is known;

- methodologically valid with respect to their analyticalcharacteristics.

5. Within a study the target variable should change in astatistically significant way and the change should bebiologically meaningful for the target group consistentwith the claim to be supported.

6. A claim should be scientifically substantiated by takinginto account the totality of the available data and byweighing of the evidence.

1. The food or food component to which the claimed effectis attributed should be characterised.

The food or food component should be sufficientlycharacterised and described to enable comparabilitybetween studies, and to ensure that the levels ofexposure can be linked quantitatively to the claimedeffect.

2. Substantiation of a claim should be based on humandata, primarily from intervention studies.

A claim can only be considered substantiated if there is abody of evidence that demonstrates an effect in the targetpopulation. Claims may be supported by studies inanimals, laboratory models or cellular and moleculardata, but direct evidence of benefit in humans isnecessary for a case to be made. There are many forms ofhuman studies, which can broadly be classified intointervention and observational studies. Interventionstudies include the randomised control trials (RCT) inhealthy subjects, or clinical trials in patients, and studieslooking at physiological or psychological effects. Of allstudies, RCTs are thought to provide the best standard ofevidence. More than a single trial is preferable and metaanalysis of multiple RCT is a powerful statistical tool.

The design of the studies should include the followingconsiderations:

2a Study groups that are representative of the targetgroup.

Results gathered from a study group will beextrapolated to the group targeted by the claim. This could be either the whole population, or aspecific sub-group of the population (e.g. elderly,obese, smokers, runners, students, pregnant women).The effects induced by a food or food component inthe study group are expected to occur in the targetedgroup, therefore the physiology or psychology of thestudy group should be representative of thephysiology of the target group.

2b Appropriate controls.

Defining an appropriate control is often not easy indietetic and nutritional studies. The amount of foodconsumed every day is roughly constant and when anew food is added to a diet, another may be left outor eaten in a smaller quantity. Therefore, the additionof a food or food component may induce an effect byitself. The second difficulty is that most foods cannotbe studied in a ‘blinded’ way. It would be difficult tofind a suitable control in a study supporting thebeneficial effects of consuming fruits and vegetables.

Final publication

Human data

Valid markers

Food characteristics

Totalityand

weight of

evidence

Substantiated claim

First Plenary Meeting, December 2002