[Federal Register: November 10, 2003 (Volume 68, Number 217)]
[Page 63802-63804]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]



Food and Drug Administration

[Docket No. 2003N-0509]

Agency Emergency Processing Under Office of Management and Budget 
Review; Experimental Study of Health Claim Disclaimers on Foods

AGENCY: Food and Drug Administration, HHS.

ACTION: Notice.


SUMMARY: The Food and Drug Administration (FDA) is announcing that a 
proposed collection of information has been submitted to the Office of 
Management and Budget (OMB) for emergency processing under the 
Paperwork Reduction Act of 1995 (the PRA). The proposed collection of 
information is an experimental study of health claims on food product 
labels to evaluate the communication effectiveness of various possible 
labeling statements (i.e., disclaimers) to convey differing levels of 
scientific support for health claims. The study examines the 
communication effectiveness of disclaimers in realistic label use 
situations for a range of health claims and associated food products 
that may bear such health claims.

DATES: Fax written comments on the collection of information by 
December 10, 2003. FDA is requesting approval of this emergency 
processing by December 10, 2003.

ADDRESSES: OMB is still experiencing significant delays in the regular 
mail, including first class and express mail, and messenger deliveries 
are not being accepted. To ensure that comments on the information 
collection are received, OMB recommends that comments be faxed to the 
Office of Information and Regulatory Affairs, OMB, Attn: Fumie Yokota, 
Desk Officer for FDA, FAX: 202-395-6974.

FOR FURTHER INFORMATION CONTACT: Peggy Robbins, Office of Management 
Programs (HFA-250), Food and Drug Administration, 5600 Fishers Lane, 
Rockville, MD 20857, 301 827-1223.

SUPPLEMENTARY INFORMATION: FDA is requesting emergency processing of 
this proposed collection of information under section 3507(j) of the 
PRA (44 U.S.C. 3507(j) and 5 CFR 1320.13). The information is critical 
to the agency's mission of regulating food labeling. Currently FDA is 
operating under interim procedures for reviewing qualified health 
claims on conventional foods and dietary supplements (``Guidance for 
Industry and FDA: Interim Procedures for Qualified Health Claims in the 
Labeling of Conventional Human Food and Human Dietary Supplements,'' 
that published in the Federal Register of July 10, 2003 (68 FR 41387-
41390)). This interim approach was necessitated by various developments 
since the passage of the Nutrition Labeling and Education Act (NLEA), 
including successful legal challenges based on the First Amendment. The 
interim procedures provide guidance to industry regarding how the 
agency will respond to qualified health claims until the agency can 
promulgate notice-and-comment rulemaking. However, guidance documents 
do not establish legally enforceable responsibilities and are intended 
only as recommendations.
    The interim procedures strain the agency's limited resources for 
reviewing qualified health claims. Qualified health claims greatly 
increase the number of potential health claims and as a result the 
agency anticipates a far greater number of health claim petitions. The 
agency included criteria for prioritizing petitions in order to 
maximize the public health benefit of its interim qualified health 
claim procedure, which will necessitate delays for some petitions. The 
interim guidance also creates uncertainty for industry, since qualified 
health claims are permitted through a letter of enforcement discretion, 
and are not authorized through a regulation. This is likely to inhibit 
some companies from submitting petitions during the interim period. FDA 
prefers that this interim period be as short as possible.
    Consumer data are important to the development of new regulations 
for health claims. A central consideration in the development of a new 
regulatory framework for qualified health claims is the importance of 
ensuring that such claims can be made in a way that is not misleading 
to consumers. The agency recognizes that it is unknown whether 
consumers can distinguish between differing levels of scientific 
support and there are no consumer data currently available to assess 
the effectiveness of wording options proposed for conveying the 
different levels. The interim guidance relies on limited prior 
experience under a temporary policy of enforcement discretion, using ad 
hoc health claim disclaimers.
    Given the uncertainties and constraints inherent with interim 
guidance and the absence of relevant consumer data to address questions 
raised by the new approaches to health

[[Page 63803]]

claims under consideration, we are seeking emergency approval of the 
proposed study in order to provide needed consumer data in time to 
assist the agency in developing new regulations for qualified health 
    FDA invites comments on these topics: (1) Whether the proposed 
collection of information is necessary for the proper performance of 
FDA's functions, including whether the information will have practical 
utility; (2) the accuracy of FDA's estimate of the burden of the 
proposed collection of information, including the validity of the 
methodology and assumptions used; (3) ways to enhance the quality, 
utility, and clarity of the information to be collected; and (4) ways 
to minimize the burden of the collection of information on respondents, 
including through the use of automated collection techniques, when 
appropriate, and other forms of information technology.

Experimental Study of Health Claim Disclaimers on Foods

    FDA is requesting OMB approval of an experimental study of health 
claims and disclaimers on food labels to help the Center for Food 
Safety and Applied Nutrition formulate decisions and policies affecting 
labeling requirements for qualified health claims. Several possible 
approaches to implementing this qualified health claim scheme that 
differ in terms of the specific language and form of disclaimers used 
to convey level of scientific certainty are evaluated in terms of the 
ability of the proposed approach to accurately convey the actual level 
of scientific uncertainty for the stated claim.
    The recent report of the FDA Task Force (Consumer Health 
Information for Better Nutrition Initiative Task Force Final Report, 
July 10, 2003) describes a four-level rating scheme for evaluating 
petitioned claims (consisting of unqualified claims that meet the 
standard of significant scientific agreement as defined by NLEA and 
three levels of qualified claims supported by decreasing levels of 
scientific evidence). The proposed consumer research is designed to 
test approaches to conveying levels of scientific uncertainty through 
the use of disclaimers that are linked to this four-level rating scheme 
for petitioned health claims.
    The proposed study is intended to evaluate the effectiveness of 
several possible options for communicating the strength of scientific 
evidence for a given health claim across a range of health claims of 
varying scientific certainty. The evidence should provide empirical 
support for possible policy decisions about the need for disclaimers to 
minimize consumers' misunderstanding and misapplication of qualified 
health claims and the optimal language and the form such disclaimers 
should take. The impact of disclaimers is examined across a range of 
measures that capture what is conveyed about the state of scientific 
certainty for the claim as well as the impact of the qualified health 
claim on attributions about the food product that displays the claim.
    FDA will conduct an experimental study using shopping mall 
intercept samples. The mall intercept methodology allows controlled 
presentation of visual materials, experimental manipulation of study 
materials, and the random assignment of participants to experimental 
conditions. The experimental manipulation of label conditions and 
random assignment to conditions allows for statistical estimates of the 
effects of different approaches to conveying level of scientific 
support and allows quantitative comparisons of the effectiveness of 
different forms and wording options for health claim disclaimers. 
Random assignment ensures that mean differences between conditions can 
be tested using established techniques such as analysis of variance and 
multiple regression analysis to yield statistically valid estimates of 
effect size.
    The study design is based on the controlled presentation of 
realistic product labels that carry health claims for four nutrient/
disease health claims. The four health claims that are tested vary in 
terms of the degree of scientific evidence underlying the health claim. 
Label conditions consist of different forms and specific wordings for 
disclaimers that accompany the nutrient/disease health claim as well as 
various control conditions that assess how consumers view the product 
and the scientific evidence in the absence of an explicit health claim 
on the product label.
    Participants will be recruited using standard mall intercept 
methods, implemented in 6 geographically dispersed shopping malls. 
Participants are adults, aged 18 and older who do half or more of the 
grocery shopping for their household. Each site will have the same 
number of replicates of the experimental design that include all 
counterbalancing factors.
    Four different schemes for communicating strength of science are 
tested: Point-Counterpoint (claim, followed by disclaimer), Embedded 
language (disclaimer first), Report Card (A-D letter ratings) and 
Graphic (graphic device to illustrate the rating scheme). Each scheme 
adopts the four-level strength of science ranking system described in 
the Interim Guidance.
    The study includes four control conditions, representing important 
types of label statements and label users that constitute benchmarks 
for assessing the direction and magnitude of effects due to 
communications about the strength of scientific evidence for the health 
claims: (1) ``Tombstone'' control with no nutrient content or health 
claim, (2) nutrient content claim, but no health claim, (3) ``full 
information control'' in which the participant is provided with a 
summary of the scientific evidence for the claim prior to observing 
food labels and (4) expert controls, based on separate information 
gathered from nutrition experts knowledgeable about the diet-disease 
    The key measures for this study are the perceived strength of 
science for the claim that is conveyed by the label condition and 
product perception questions about the labeled food product (expected 
health benefits, perceived nutrition ratings) that identify the 
practical impact of the product label.
    FDA will use the information from this study to guide the 
development of regulatory policy options related to qualified health 
claims. The agency acknowledges the lack of empirical data about how 
consumers understand and respond to statements they see in product 
labeling. The information gathered in this study can be used by the 
agency to assess likely consumer responses to various options for 
qualifying health claims based on varied levels of scientific evidence.
    FDA estimates the burden of this collection of information as 

[[Page 63804]]

                                  Table 1--Estimated Annual Reporting Burden\1\
                             Annual Frequency
         Number of              [chyph]per            Total Annual               Hours per          Total Hours
    [chyph]Respondents           Response           [chyph]Responses          [chyph]Response
1,920                       1                   1,920                     .30                     576
\1\ There are no capital costs or operating and maintenance costs associated with this collection of

    The approaches and wording options for qualified health claims of 
central interest to the agency requires a complex experimental design. 
To ensure adequate power to identify differences, the minimum cell size 
is 60 participants. This will be sufficient to identify small to medium 
effects (i.e., r =.15 to .30) for all main effects and first order 
interactions with power = (1 - beta), well in excess of .80 at the .05 
significance level.

    Dated: November 4, 2003.
Jeffrey Shuren,
Assistant Commissioner for Policy.
[FR Doc. 03-28196 Filed 11-7-03; 8:45 am]