34650 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

Seven PREIS grantees have requested Objectives and Need for Assistance, Single-source program expansion supplemental funding awards. Their Approach, and Budget and Budget supplement awards are made to the applications were assessed by a review Justification. The applications were following PREIS grantees: panel for completeness and assessed to have scored within a responsiveness in the categories of fundable range.

Supplement award Grantee organization City State amount

Child and Family Resources, Inc...... Tucson ...... AZ $32,314 Children’s Hospital of Los Angeles ...... Los Angeles ...... CA 115,898 Cicatelli Associates Inc...... New York ...... NY 130,000 Demoiselle2Femme ...... Chicago ...... IL 55,959 Education Development Center, Inc...... Newton ...... MA 55,560 Teen Outreach Pregnancy Services ...... Tucson ...... AZ 29,000 The Village for Families & Children, Inc ...... Hartford ...... CT 33,235

Statutory Authority: Section 2953 of the 240–402–1278, email: mical.honigfort@ source of industrially-produced trans Patient Protection and Affordable Care Act of fda.hhs.gov. fatty acids (Ref. 1). As explained in the 2010, Pub. L. 111–148, added Section 513 to SUPPLEMENTARY INFORMATION: tentative determination (78 FR 67169), Title V of the Social Security Act, codified all refined edible oils contain some at 42 U.S.C. 713, authorizing the Personal Table of Contents Responsibility Education Program. trans fat as an unintentional byproduct I. Background of their manufacturing process; Mary M. Wayland, II. Definitions and Scope, and Related however, unlike other edible oils, trans Senior Grants Policy Specialist, Division of Comments With FDA Responses fats are an integral component of PHOs Grants Policy, Office of Administration. III. Discussion of Legal Issues, and Related and are purposely produced in these [FR Doc. 2015–14839 Filed 6–16–15; 8:45 am] Comments With FDA Responses oils to affect the properties of the oils A. GRAS BILLING CODE 4184–37–P and the characteristics of the food to B. Prior Sanctions C. Procedural Requirements which they are added. In addition, the trans fat content of PHOs is significantly DEPARTMENT OF HEALTH AND IV. Discussion of Scientific Issues, and Related Comments With FDA Responses greater than the amount in other edible HUMAN SERVICES A. Intake Assessment oils. Non-hydrogenated refined oils may B. Safety contain trans fatty acids as a result of Food and Drug Administration V. Citizen Petitions high-temperature processing, at levels [Docket No. FDA–2013–N–1317] VI. Environmental Impact typically below 2 percent (Ref. 2). Low VII. Economic Analysis levels (below 2 percent) may also be Final Determination Regarding VIII. Compliance Date and Related Comments found in fully hydrogenated oils (FHOs) Partially Hydrogenated Oils With FDA Responses IX. Conclusion and Order due to incomplete hydrogenation (Ref. AGENCY: Food and Drug Administration, X. References 3). Small amounts (typically around 3 HHS. percent) may be found in the fat I. Background ACTION: Notice; declaratory order. component of dairy and meat products In accordance with the process set out from ruminant animals (Ref. 4). SUMMARY: Based on the available in § 170.38(b)(1) (21 CFR 170.38(b)(1)), FDA’s tentative determination scientific evidence and the findings of we issued a notice on November 8, 2013 identified the significant human health expert scientific panels, the Food and (the November 2013 notice, 78 FR risks associated with the consumption Drug Administration (FDA or we) has 67169), announcing our tentative of trans fat (78 FR 67169 at 67171). The made a final determination that there is determination that, based on currently tentative determination was based on no longer a consensus among qualified available scientific information, PHOs evidence including results from a experts that partially hydrogenated oils are no longer GRAS under any number of controlled feeding studies on (PHOs), which are the primary dietary condition of use in human food and trans fatty acid consumption in humans source of industrially-produced trans therefore are food additives subject to (Refs. 5 and 6), findings from long-term fatty acids (IP–TFA) are generally section 409 of the Federal Food, Drug, prospective epidemiological studies recognized as safe (GRAS) for any use in and Cosmetic Act (the FD&C Act) (21 (Refs. 5 and 6), and the opinions of human food. This action responds, in U.S.C. 348). expert panels (Refs. 7, 8, 9, 10, 11, 12, part, to citizen petitions we received, FDA’s evaluation of the GRAS status 13, and 14). The latter included the and we base our determination on of PHOs centers on the trans fatty acid 2005 recommendation of the Institute of available scientific evidence and the (TFA, also referred to as ‘‘trans fat’’) Medicine (IOM) to limit trans fat findings of expert scientific panels component of these oils. Although we consumption as much as possible while establishing the health risks associated primarily use the word ‘‘oil’’ when consuming a nutritionally adequate diet, with the consumption of trans fat. discussing PHOs in this document, recognizing that trans fat occurs DATES: Compliance date: Affected partially hydrogenated fats (such as naturally in meat and dairy products persons must comply no later than June partially hydrogenated lard), are from ruminant animals and that 18, 2018. included within the definition of PHOs naturally-occurring trans fat is FOR FURTHER INFORMATION CONTACT: (discussed in section II) and therefore unavoidable in ordinary, non-vegan Mical Honigfort, Center for Food Safety within the scope of this order, and diets without significant dietary and Applied Nutrition (HFS–265), Food references to ‘‘oil’’ in this document adjustments that may introduce and Drug Administration, 5100 Paint should be read in most cases to include undesirable effects (Ref. 7). In addition, Branch Pkwy., College Park, MD 20740, fats. PHOs are the primary dietary in the tentative determination FDA cited

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00045 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34651

a peer reviewed, published estimate of governments. Most comments generally When a fat or oil is hydrogenated, the deaths and coronary events that would supported the tentative determination or degree of hydrogenation can be tailored be prevented annually in the United supported aspects of it. FDA also to obtain the desired properties for the States from elimination of remaining received numerous comments stating application. FHOs are produced by uses of PHOs from the food supply (Ref. that although they agreed with FDA’s allowing the hydrogenation process to 15). Given all this evidence, we efforts to further reduce trans fat in the proceed to complete or near complete tentatively determined that there is no food supply, they disagreed with our saturation to obtain a more solid fat. In longer a consensus among qualified tentative determination regarding the practice, the reaction does not proceed experts that PHOs, the primary dietary GRAS status of PHOs. Of the comments to 100 percent completion, even when source of IP–TFA, are safe for human that objected to the tentative producing FHOs, and some degree of consumption, either directly or as determination, many disagreed with unsaturation unavoidably remains in ingredients in other food products. FDA’s scientific analysis and offered the final fat or oil. Non-hydrogenated PHOs have a long history of use as alternative approaches to address trans refined fats and oils generally contain food ingredients. The two most common fat in the food supply. Some comments trans fatty acids as an unavoidable PHOs currently used by the food addressed issues outside the scope of impurity as a result of high-temperature industry, partially hydrogenated the tentative determination (such as processing, at levels typically below 2 soybean oil and partially hydrogenated disruptions to trade, taxation of foods, percent (Ref. 2). The IV of a fat or oil cottonseed oil, are not listed as GRAS or and requests for bans on other is not a direct measure of the TFA as approved food additives in FDA’s substances) and were not considered. content, but is a measure of the degree regulations. However, these and other We reviewed all comments that were of unsaturation. Thus, in a fat or oil that commonly used PHOs (e.g., partially submitted to the docket before arriving has been hydrogenated, a low degree of hydrogenated coconut oil and partially at the decision outlined in this order. unsaturation (i.e., a low IV number) will hydrogenated palm oil) have been We have arranged comments and our correlate to a low level of TFA. FHOs considered GRAS by the food industry responses by topic throughout the with an IV of 4 or less generally contain based on a history of use prior to 1958. remainder of this document. To make it trans fat at levels similar to non- By contrast, the partially hydrogenated easier to identify the comments and our hydrogenated refined fats and oils (less versions of low erucic acid rapeseed oil responses, the word ‘‘Comment,’’ in than 2 percent). By contrast, when the (LEAR oil; § 184.1555(c)(2) (21 CFR parentheses, appears before the hydrogenation process is arrested before 184.1555(c)(2)) and menhaden oil comment’s description and the word near complete saturation, trans fat (§ 184.1472(b) (21 CFR 184.1472(b))) ‘‘Response,’’ in parentheses, appears content is typically higher, and IV is have been affirmed by regulation as before FDA’s response. Each comment is typically greater than 4. GRAS for use in food. Partially numbered to help distinguish between Based on data for FHOs that are hydrogenated LEAR oil was affirmed as different comments. The number currently available on the market, which GRAS for use in food (50 FR 3745 assigned to each comment is purely for are indicative of modern hydrogenation (January 28, 1985)) through scientific organizational purposes and does not technology (Ref. 16), we define FHOs for procedures. Partially hydrogenated signify the comment’s value or the purposes of this order as fats and menhaden oil was affirmed as GRAS for importance. oils that have been hydrogenated to use in food (54 FR 38219 (September 15, The major provisions of this order are: complete or near complete saturation, 1989)) on the basis that the oil is • PHOs are not GRAS for any use in and with an IV of 4 or less, as chemically and biologically comparable human food. determined by a method that is suitable to commonly used partially • Any interested party may seek food for this analysis (e.g., ISO 3961 or hydrogenated vegetable oils such as additive approval for one or more equivalent). FHOs are outside the scope corn and soybean oils. FDA believes specific uses of PHOs with data of this order. For the purposes of this that partially hydrogenated LEAR and demonstrating a reasonable certainty of order, we define PHOs as fats and oils menhaden oils are not currently widely no harm of the proposed use(s). that have been hydrogenated, but not to used by the food industry. We plan to • For the purposes of this declaratory complete or near complete saturation, amend these regulations in a future order, FDA is defining PHOs as those and with an IV greater than 4 as rulemaking. fats and oils that have been determined by a method that is suitable In the November 2013 notice, FDA hydrogenated, but not to complete or for this analysis (e.g., ISO 3961 or requested additional data and scientific near complete saturation, and with an equivalent). These definitions will information related to our tentative iodine value (IV) greater than 4. ensure that IP–TFA content in the food determination and, in particular, • FDA is establishing a compliance supply will be kept to the minimum requested comment on several questions date of June 18, 2018. amount feasible with current (78 FR 67169 at 67174). Interested technology, except as otherwise persons were originally given until II. Definitions and Scope, and Related authorized. January 7, 2014, to comment on the Comments With FDA Responses (Comment 2) We received several notice. However, in response to several (Comment 1) Some comments comments requesting clarification on requests, we extended the comment requested that we define PHOs and the scope of FDA’s tentative period to March 8, 2014 (78 FR 79701 clearly delineate them from FHOs. The determination, including whether it (December 31, 2013)). comments suggested various parameters applies only to PHOs used in human We received over 6000 comments in for defining these fats and oils, food; whether it applies to ingredients response to the November 2013 notice including setting a specification for that contain only naturally occurring announcing our tentative determination, trans fat content (e.g., a percentage) or trans fat, such as those ingredients including over 4500 form letters. In using iodine value (IV; also derived from ruminant sources; and addition to submissions from interchangeably called iodine number). whether it applies to conjugated linoleic individuals, we received comments (Response) FDA agrees with the acid. We also received a citizen petition from industry and trade associations, comments that we should define PHOs (discussed in section V) raising consumer and advocacy groups, health to differentiate them from FHOs, which questions related to partially professional groups, and state/local are outside the scope of this order. hydrogenated methyl ester of rosin.

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00046 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34652 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

(Response) FDA wishes to clarify that III. Discussion of Legal Issues, and (5th Cir. 1975) (‘‘What is required is not this order applies only to PHOs used in Related Comments With FDA unanimous recognition but general human food, not animal feed, and Responses recognition.’’); United States v. Articles applies to PHOs used as a food of Drug * * * Promise Toothpaste, 624 A. GRAS ingredient, which includes those uses F. Supp. 776, at 782–3 (N.D. Ill. 1985) sometimes considered processing aids Section 409 of the FD&C Act provides (‘‘There is nothing in the statute to or food contact substances (e.g., pan- that a food additive is unsafe unless it indicate that Congress intended release agents). By contrast, the use of is used in accordance with conditions ‘generally recognized’ in other than its PHOs as raw materials used to set forth in that section. ‘‘Food additive’’ commonly understood meaning. The synthesize other ingredients is outside is defined by section 201(s) of the FD&C adverb, ‘generally,’ is defined, inter alia, the scope of this order. We do not have Act (21 U.S.C. 321(s)) as any substance to mean . . . extensively, though not specific information on the intake of the intended use of which results or universally’’ (internal quotations industrially-produced trans fat from this may reasonably be expected to result in omitted)). Conversely, general source. There is no requirement that its becoming a component or otherwise recognition of safety does not exist if materials used to make food ingredients affecting the characteristics of any food, there is a lack of consensus among be GRAS themselves; rather, the if such substance is not GRAS or qualified experts that the use of a resultant food ingredient must be safe otherwise excluded from the definition. substance is safe. See, e.g., Coli-Trol 80, for the intended conditions of use. The Certain other substances that may 518 F.2d at 746 (no general recognition use of PHOs as raw materials to make become components of food are also of safety where there was ‘‘no other food ingredients may result in the excluded from the statutory definition recognition of the safety . . . of these incorporation of industrially-produced of food additive, including pesticide products at all’’); Premo Pharmaceutical trans fats into those ingredients. When chemicals and their residues, new Laboratories v. United States, 629 F.2d ingredients are synthesized using PHOs, animal drugs, color additives, and 795, 803–4 (2nd Cir. 1980) (‘‘genuine and the ingredient is being used on the dietary ingredients in dietary dispute among qualified experts’’ basis of a GRAS self-determination, supplements (section 201(s)(1) through precludes finding of general recognition, reevaluation of such a determination (6) of the FD&C Act). and no general recognition existed as a A substance is GRAS if it is generally may be appropriate in light of the health matter of law where there was a ‘‘sharp recognized, among experts qualified by effects from the intake of trans fat that difference’’ of expert opinion); United scientific training and experience to underlie our determination that PHOs States v. Article of Food * * * Coco evaluate its safety, as having been do not meet the GRAS standard. Rico, 752 F.2d 11, 15 n 6 (1st Cir. 1985) adequately shown through scientific (substance was not GRAS as a matter of This order does not apply to procedures (or, in the case of a ingredients that contain only naturally substance used in food prior to January law based on existence of ‘‘genuine occurring trans fat, such as those 1, 1958, through either scientific dispute among qualified experts’’ ingredients derived from ruminant procedures or experience based on regarding safety of use); Promise sources. common use in food) to be safe under Toothpaste, 624 F. Supp. at 783 (court This order does not apply to the use the conditions of its intended use could not conclude whether a ‘‘genuine of conjugated linoleic acid (CLA) as a (section 201(s) of the FD&C Act). dispute’’ existed without considering food ingredient. CLA does not fit the However, history of use prior to 1958 is the substance of the experts’ opinions, definition of PHO. CLAs are a class of not sufficient to support continued such that a triable issue of fact existed fatty acid isomers derived from linoleic GRAS status if new evidence regarding general recognition). See also acid and do not contain nonconjugated demonstrates that there is no longer a United States v. Articles of Drug * * * double bonds in a trans configuration consensus that an ingredient is safe. See 5,906 Boxes, 745 F.2d 105, 119 n. 22 (1st nor are CLAs triglyceride molecules. On § 170.30(l) (21 CFR 170.30(l)) (‘‘New Cir. 1984) (noting certain cases in which the other hand, PHOs are primarily information may at any time require lack of general recognition was mixtures of triglycerides, produced by reconsideration of the GRAS status of a established as a matter of law and others partial hydrogenation and include at food ingredient.’’). in which there was a triable issue of fact least one nonconjugated double bond(s) FDA has defined safe as ‘‘a reasonable regarding general recognition). in a trans configuration (Ref. 16). certainty in the minds of competent Importantly, the GRAS status of a Considering CLA to be distinct from scientists that the substance is not specific use of a particular substance in PHOs is consistent with how FDA has harmful under the intended conditions food may change as knowledge changes. previously defined trans fatty acids for of use’’ (§ 170.3(i) (21 CFR 170.3(i)), and For example, as new scientific data and nutrition labeling purposes, focusing on general recognition of safety must be information develop about a substance the presence of nonconjugated bond(s) based only on the views of qualified or the understanding of the in a trans configuration (see experts (21 CFR 170.30(a)). To establish consequences of consumption of a § 101.9(c)(2)(ii) (21 CFR 101.9(c)(2)(ii))). general recognition of safety, there must substance evolves, expert opinion This order also does not apply to the be a consensus of expert opinion regarding the safety of a substance for a use of partially hydrogenated methyl regarding the safety of the use of the particular use may change such that ester of rosin. Partially hydrogenated substance. See, e.g., United States v. there is no longer a consensus that the methyl ester of rosin does not fit the Western Serum Co., Inc., 666 F.2d 335, specific use is safe. The fact that the definition of PHO. Partially 338 (9th Cir. 1982) (citing Weinberger v. status of the use of a substance under hydrogenated methyl ester of rosin is Hynson, Westcott & Dunning, 412 U.S. section 201(s) of the FD&C Act may composed of resin acids that are 609, 629–32 (1973)). General recognition evolve over time is the underlying basis chemically and structurally distinct of safety does not require unanimous for FDA’s regulation at § 170.38, which from fatty acids found in PHOs. Resin agreement. See, e.g., United States v. provides, in part, that we may, on our acids are terpene-derived aromatic Articles of Drug * * * 5,906 boxes, 745 own initiative, propose to determine compounds that do not have long chain F.2d 105, 119 n. 22 (1st Cir. 1984); that a substance is not GRAS. (See fatty acid components with cis/trans United States v. Articles of Food and generally 37 FR 6207 (March 25, 1972) double bonds (Ref. 16). Drug (Coli-Trol 80), 518 F.2d 743, 746 (proposal of 21 CFR 121.41, the

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00047 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34653

predecessor of § 170.38); 37 FR 25705 experts that PHOs are safe for use in consensus among qualified experts that (December 2, 1972) (issuance of 21 CFR human food. However, there were also such uses are safe, as we do here. We 121.41); 35 FR 18623 (December 8, many comments that disagreed with acknowledge that scientific knowledge 1970) (proposal of 21 CFR 121.3, the FDA’s tentative determination and advances and evolves over time. We predecessor of § 170.30); and 36 FR stated that we did not adequately encourage submission of scientific 12093 (June 25, 1971) (issuance of 21 demonstrate that PHOs are not GRAS. evidence as part of food additive CFR 121.3)). Further, as stated in section (Comment 3) Some comments stated petitions under section 409 of the FD&C I, history of the safe use of a substance that FDA must show a ‘‘severe conflict’’ Act for one or more specific uses of in food prior to 1958 is not sufficient to among experts about the safety of a PHOs for which industry or other support continued GRAS status if new substance in order to determine that interested individuals believe that safe evidence demonstrates that there is no PHOs are not GRAS. conditions of use may be prescribed. We longer expert consensus that an (Response) FDA disagrees that ‘‘severe are establishing a compliance date of ingredient is safe (§ 170.30(l)). conflict’’ is the relevant standard. As June 18, 2018 for this order to allow As noted in section III.A, under discussed in section III.A, general time for such petitions and their review. section 201(s) of the FD&C Act, a recognition of safety does not exist if (Comment 5) One comment stated substance that is GRAS for a particular there is a lack of consensus among that FDA must demonstrate that each use in food is not a food additive, and qualified experts that the use of a and every PHO, and every use of PHOs, may lawfully be utilized for that use substance is safe. We have considered is not safe. without FDA review or approval. all available information and (Response) FDA disagrees. FDA need Currently, a GRAS determination may determined that there is no longer a not demonstrate that PHOs are unsafe to be made when the manufacturer or user consensus among qualified experts that determine that they are not GRAS, only of a food substance evaluates the safety PHOs are safe for human consumption. that there is a lack of consensus among of the substance and the views of To the extent there is disagreement qualified experts regarding their safety. qualified experts and determines that among qualified experts about the safety In addition, our consideration of PHOs the use of the substance is GRAS. This of PHOs for human consumption, this as a class is justified because the approach is commonly referred to as genuine dispute regarding safety available, relevant scientific evidence ‘‘GRAS self-determination’’ or precludes a finding of GRAS. demonstrates an increased risk of ‘‘independent GRAS determination.’’ (Comment 4) Some comments focused coronary heart disease (CHD) Other substances that are GRAS may on the idea that it may be possible to attributable to trans fat (see section be identified in FDA regulations in one establish a threshold below which PHOs VI.B); PHOs are the primary dietary of two ways. Following the passage of may be safely used in the food supply. source of IP–TFA; and there is a lack of the 1958 Food Additives Amendment, One comment argued that there is no consensus among qualified experts that we established in our regulations a list consensus among experts that PHOs are PHOs are safe for use in food at any of food substances that, when used as unsafe below some low threshold level level. indicated, are considered GRAS. We of use. (Comment 6) Some comments stated made clear that this was not a (Response) As discussed later in that, by determining that the use of comprehensive list. This list (commonly section IV.B.1, FDA does not agree that PHOs are not GRAS because they referred to as the ‘‘GRAS list’’) now such a threshold has been identified contain a nutrient that increases risk of appears at 21 CFR part 182. Thereafter, based on the available science. CHD, FDA would be calling into in 1972, we established the GRAS Importantly, even if such a threshold question the regulatory status of other affirmation process through which we could be identified, this alone would food sources of trans fat. affirmed, through notice and comment not meet the requirement of ‘‘general (Response) FDA disagrees. As noted rulemaking, the GRAS status of recognition’’ for uses below the in section II, this order does not apply particular uses of certain substances in threshold without there also being to ingredients that contain naturally food. Regulations affirming the GRAS consensus among qualified experts that occurring trans fat (such as those status of certain substances appear at 21 uses below the threshold are safe. (See ingredients derived from ruminant CFR parts 184 and 186. (As a general United States v. 7 Cartons, 293 F. Supp. sources), fully hydrogenated oils, or matter, we no longer affirm the GRAS 660, 663 (S.D. Ill. 1968) (‘‘an inference edible oils that contain IP–TFA as an status of substances through notice-and- that safety might be shown by scientific impurity. FDA has considered the comment rulemaking. In April 1997, we testing and procedures’’ is insufficient available information and concluded proposed to replace the voluntary GRAS as a matter of law to demonstrate that there is a lack of consensus among affirmation petition process with a general recognition of safety), affirmed qualified experts that PHOs, as the voluntary GRAS notification program, in relevant part, 424 F.2d 1364 (7th Cir. primary dietary source of IP–TFA, are which would not involve rulemaking 1970).) FDA has no basis to conclude safe for use in human food. We may (62 FR 18938 (April 17, 1997)). At the that there is any such consensus. FDA determine that the use of an artificial time of the proposal, we initiated a pilot has previously revoked GRAS status substance is not GRAS without of the GRAS notification program, under similar circumstances (51 FR necessarily making the same which continues to function. A firm 25021 at 25023, July 9, 1986; revoking determination about naturally-occurring may voluntarily submit information on GRAS status of sulfiting agents on fruits versions of the substance. (See, e.g., 35 a GRAS self-determination to FDA for and vegetables intended to be served or FR 7414 (May 13, 1970) (Rescinding review through the GRAS notification sold raw to consumers; explaining that letters that had expressed opinions that program, but is not required to do so.) it was not possible to set a threshold for certain uses of glycine and its salts are FDA received numerous comments on safe use based on available information). GRAS, and stating that such added our tentative determination. Many Moreover, we need not determine that substances are no longer GRAS in related to the GRAS standard and what there is a consensus that low level uses human food); 37 FR 6938 (April 6, 1972) is needed to demonstrate that a are unsafe to find that PHOs are not (Amino Acids in Food for Human substance is not GRAS. Many comments GRAS at low levels; we need only Consumption; Proposed Conditions of agreed with our determination that there determine that based on available Safe Use in Food and Deletion From is not a consensus among qualified scientific evidence there is not a GRAS List) (‘‘[T]he mere natural

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00048 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34654 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

presence of an amino acid in 1958 (Pub. L. 85–929) emphasizes the specifically health risks related to PHOs, unprocessed foods in free or combined broad applicability of sections 201(s), the primary dietary source of IP–TFA. (as protein) form does not qualify it as 409, and 402(a)(2)(C) of the FD&C Act, Nothing in the NLEA suggested that its safe for addition in a pure form as a which apply to ‘‘any substances the passage limited the preexisting food component of a formulated or processed ingestion of which reasonable people additive provisions in the FD&C Act, or food’’), 38 FR 20036 (July 26, 1973) would expect to produce not just cancer that the food additive provisions did not (Amino Acids in Food for Human but any disease or disability’’ (S. Rep. apply to nutrients and chronic Consumption; Conditions of Safe Use in No. 2422, at 11 (1958), as reprinted in multifactorial disease under appropriate Food and Deletion From GRAS List); 47 Vol. 14, Legislative History of the Food, circumstances. On the contrary, as the FR 22545 (May 25, 1982) (Cinnamyl Drug & Cosmetic Act and its comment noted, the NLEA contained a Anthranilate; Proposed Prohibition of Amendments, at 923 (1979)). In fact, we clause stating that ‘‘[t]he amendments Use in Human Food) (acknowledging have previously taken action regarding made by this Act shall not be construed ‘‘the presence of other cinnamyl and health risks related to nutrients using to alter the authority of the Secretary of anthranilate derivatives naturally in these authorities (55 FR 50777 Health and Human Services . . . under food and in natural substances used to (December 10, 1990) (determining the [FD&C Act]’’ (NLEA section 9). flavor food’’ but proposing to prohibit certain Vitamin K Active Substances not The FD&C Act’s nutrition labeling and only cinnamyl anthranilate); 50 FR GRAS); and 38 FR 20036 (July 26, 1973) food additive provisions are two 42929 (October 23, 1985) (Cinnamyl (establishing conditions of safe use for different kinds of authority, with Anthranilate; Prohibition of Use in amino acids for nutritive purposes and different standards, and we may choose Human Food)). deleting them from GRAS list)). We also among available approaches to a public (Comment 7) One comment stated have previously applied these health problem when the FD&C Act that Congress, through the Nutrition authorities to substances presenting provides multiple options. See, e.g., Labeling and Education Act of 1990 increased health risks related to chronic Chevron U.S.A. Inc. v. Natural (NLEA) (Pub. L. 101–535), prescribed multifactorial diseases, such as cancer Resources Defense Council, 467 U.S. labeling as the sole vehicle for achieving (50 FR 42929 (October 23, 1985) 837, 865–6 (1984) (‘‘While agencies are the nutritional policy objective of (prohibiting use of cinnamyl not directly accountable to the people, shifting dietary patterns to reduce the anthranilate in food); and 34 FR 17063 the Chief Executive is, and it is entirely risk of multifactorial chronic diseases (October 21, 1969) (prohibiting use of appropriate for this political branch of such as CHD. The comment argued that cyclamates in food)). the Government to make such policy FDA’s use of its food additive authority With respect to the comment citing a choices—resolving the competing with respect to PHOs and their effect on statement from a final rule on health interests which Congress itself either risk of CHD is not within FDA’s legal claims, FDA does not agree that this inadvertently did not resolve, or authority. Some comments statement shows any change in FDA’s intentionally left to be resolved by the characterized the tentative position, as it was explicitly limited to agency charged with the administration determination as a new approach or a situations that did not meet the food of the statute in light of everyday change in interpretation, arguing that additive definition because the realities’’); United States v. Mead Corp., FDA has not previously addressed components discussed ‘‘have not been 533 U.S. 218, 227 (2001) (‘‘agencies health concerns related to nutrient added to foods.’’ The statement is charged with applying a statute intake through the FD&C Act’s food consistent with FDA’s current necessarily make all sorts of interpretive additive provisions. In support of the understanding of the law. choices’’). There is no ‘‘conflict’’ argument that FDA has changed its Moreover, FDA disagrees with the between the FD&C Act’s nutrition interpretation of the applicability of the argument that FDA must address health labeling provisions and food additive food additive provisions of the FD&C risks related to PHOs through food provisions as the comment suggests. It Act, one comment cited a statement by labeling requirements rather than is also worth noting that we have FDA in rulemaking regarding health through the food additive provisions of previously determined that a use of a claims that ‘‘where the only safety issue the FD&C Act. The NLEA amended the substance is not GRAS while rejecting a is an increased risk of chronic disease FD&C Act to provide, among other labeling-based approach to the health from excessive consumption, the safety things, for certain nutrients and food risks presented by that use (51 FR 25021 provisions of the act would not provide components to be included in nutrition (July 9, 1986) (final rule revoking GRAS regulatory sanctions against such labeling. Section 403(q)(2)(A) and status of sulfiting agents on fruits and components of food, at least if they have (q)(2)(B) (21 U.S.C. 343(q)(2)(A) and vegetables intended to be served or sold not been added to foods’’ (58 FR 2478 (q)(2)(B)) of the FD&C Act state that the raw to consumers); and 50 FR 32830 at 2490 (January 6, 1993)). Secretary of Health and Human Services (August 14, 1985) (proposal to revoke (Response) FDA disagrees with these (the Secretary) (and, by delegation, GRAS status of sulfiting agents on fruits comments. FDA may properly address FDA) can, by regulation, add or delete and vegetables intended to be served or such health risks using the food additive nutrients included in the food label or sold raw to consumers)). authorities in the FD&C Act (sections labeling if he or she finds such action (Comment 8) Some comments stated 201(s), 409, and 402(a)(2)(C) of the necessary to assist consumers in that the expert panels we cited in the FD&C Act). The broad language of the maintaining healthy dietary practices. tentative determination (i.e., the food additive definition in section We have used this authority to require Institute of Medicine/National Academy 201(s) of the FD&C Act covers ‘‘any labeling of trans fat content (68 FR of Sciences (IOM/NAS), American Heart substance’’ added to food, including 41434 (July 11, 2003); see also Association, American Dietetic nutrients. Nothing in the FD&C Act or § 101.9(c)(2)(ii) and § 101.36(b)(2)(i)) (21 Association, World Health Organization, its legislative history suggests that the CFR 101.36(b)(2)(i)). Although we may Dietary Guidelines Advisory Committee, food additive definition should be further address trans fat through and the FDA Food Advisory Committee interpreted in a way that limits its labeling requirements in the future, Nutrition Subcommittee) were not applicability as the comment suggests. labeling is not the only method by experts qualified by scientific training On the contrary, the legislative history which we may address health risks and experience to evaluate the safety of of the Food Additives Amendment of related to trans fats, and more substances in food. The comments also

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00049 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34655

stated that these expert panels were not (§ 170.3(i)), and data showing a respect to the specific relationship convened for the purposes of evaluating potential relationship between a between the state or local law and the the safety of PHOs and did not make nutrient (or any other substance added federal law. FDA believes, however, that determinations regarding the GRAS to food) and disease are safety data. state or local laws that prohibit or limit status of PHOs. Therefore, the Studies reviewed by expert panels use of PHOs in food are not likely to be comments argued that the conclusions showed that trans fatty acids cause in conflict with federal law, or to of these panels do not demonstrate a significant health risks. Such studies are frustrate federal objectives. lack of consensus among qualified safety data. B. Prior Sanctions experts that PHOs are GRAS. (Comment 10) One comment stated (Response) FDA disagrees with these that FDA should hold the manufacturer We stated in our tentative comments. The expert panels we cited initially introducing the food or determination that we were not aware were composed of scientists qualified by ingredient into interstate commerce that FDA or U.S. Department of relevant training and experience to responsible for compliance with a Agriculture (USDA) had granted any review literature on trans fat determination that PHOs are not GRAS, explicit approval for any use of PHOs in consumption, because of their and that distributors should not be food prior to the 1958 Food Additives nationally recognized and established responsible for determining whether Amendment to the FD&C Act, and expertise in the area of food and foods they merely distribute contain requested comments on whether there nutrition. For example, the Food and PHOs. was knowledge of an applicable prior Nutrition Board at IOM/NAS is a (Response) Although we are mindful sanction for the use of PHOs in food (78 recognized national resource for of the need to focus our enforcement FR 67169 at 67174). We received recommendations on health issues, and efforts, those needs do not change the various comments on this topic. We are the Dietary Guidelines Advisory underlying law or FDA’s legal authority. not making a determination regarding Committee members are nationally Food that is adulterated may be subject the existence of any prior sanctions for recognized experts in nutrition and to seizure and distributors, uses of PHO in this order. This order is health. These panels’ evaluations and manufacturers, and other parties limited to our determination regarding conclusions raised significant questions responsible for such food may be subject the GRAS status of PHOs. We intend to about the safety of trans fat, thus to injunction. We recognize that address any claims of prior sanction in showing that there is no consensus manufacturers who have previously a future action. added PHO to food, rather than other among qualified scientific experts that C. Procedural Requirements PHOs are safe, because PHOs are the parties such as distributors who merely primary dietary source of IP–TFA. The receive and sell finished foods, are the Under 5 U.S.C. 554(e) (section 5(d) of safety information reviewed by the members of the food industry who will the Administrative Procedure Act panels is further discussed in section be most directly affected by this order, (APA)), an agency, ‘‘in its sound IV.B.2. We consider that the conclusions and we intend to focus our outreach and discretion, may issue a declaratory order of the panels demonstrate that there is enforcement resources accordingly. to terminate a controversy or remove a ‘‘lack of the proper reputation . . . for However, we remind distributors and uncertainty.’’ The APA defines ‘‘order’’ safety of the food additive among the other members of the food industry that as ‘‘the whole or a part of a final appropriate experts.’’ Coli-Trol 80, 518 they have an obligation to ensure that disposition, whether affirmative, F.2d at 746. Further, whether the panels the food they manufacture, distribute, negative, injunctive, or declaratory in were convened specifically to make a sell, or otherwise market complies with form, of an agency in a matter other than GRAS determination is irrelevant; the the FD&C Act. rulemaking but including licensing’’ (5 purpose of the panels was to review the (Comment 11) Some comments U.S.C. 551(6)). The APA defines available data on health risks associated requested that FDA take a position ‘‘adjudication’’ as ‘‘agency process for with consumption of trans fat. regarding the effect of this order on state the formulation of an order’’ (5 U.S.C. Moreover, the expert panel conclusions and local laws regarding PHOs. 551(7)). are not the only evidence upon which (Response) There is no statutory FDA’s regulations, consistent with the we rely for this determination, and provision in the FD&C Act providing for APA, define ‘‘order’’ to mean ‘‘the final conclusions of an expert panel are not express preemption of any state or local agency disposition, other than the required to establish general recognition law prohibiting or limiting use of PHOs issuance of a regulation, in a proceeding of safety or its absence. in food, including state or local concerning any matter . . .’’ (§ 10.3(a) (Comment 9) Several comments stated legislative requirements or common law (21 CFR 10.3(a)). Our regulations also that the expert panels we cited duties. As with any Federal define ‘‘proceeding and administrative considered nutritional science and not requirement, if a State or local law proceeding’’ to mean ‘‘any undertaking safety. requirement makes compliance with to issue, amend, or revoke a regulation (Response) FDA disagrees that the both Federal law and State or local law or order, or to take or not to take any panels were not considering safety data; impossible, or would frustrate Federal other form of administrative action, panels were considering data from objectives, the State or local under the laws administered by the controlled trials and observational requirement would be preempted. See Food and Drug Administration’’ studies on trans fat consumption that Wyeth v. Levine, 555 U.S. 555 (2009); (§ 10.3(a)). Moreover, our regulations showed adverse effects on risk factors Geier v. American Honda Co., 529 U.S. establish that the Commissioner may (e.g., effects on cholesterol) and 861 (2000); English v. General Electric initiate an administrative proceeding to increased risk of CHD (see section Co., 496 U.S. 72, 79 (1990), Florida Lime issue, amend, or revoke an order (21 IV.B.2 for further discussion on expert & Avocado Growers, Inc., 373 U.S. 132, CFR 10.25(b)). panel reviews). As discussed in more 142–143 (1963); Hines v. Davidowitz, FDA’s regulations also set forth a detail in section III.A, FDA regulations 312 U.S. 52, 67 (1941). We decline to process by which we, on our own define ‘‘safe’’ as ‘‘a reasonable certainty take a position regarding the potential initiative or on the petition of an in the minds of competent scientists for implied preemptive effect of this interested person, may determine that a that the substance is not harmful under order on any specific state or local law; substance is not GRAS. Specifically, the intended conditions of use’’ as such matters must be analyzed with FDA may initiate this process by issuing

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00050 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34656 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

a notice in the Federal Register of PHOs, industry’s ability to comply a substance that is not GRAS, and proposing to determine that a substance with them would not be prevented by a section 402(a)(2)(C) of the FD&C Act is not GRAS and is a food additive change in the regulatory status of PHOs. establishes that food bearing or subject to section 409 of the FD&C Act In addition, our labeling regulations containing a food additive that is unsafe (§ 170.38(b)). The notice must allow a explicitly address ingredient within the meaning of section 409 of the period of 60 days for comment. If, after designations for PHOs (§ 101.4(b)(14) FD&C Act is adulterated. Section 409 of review of comments, FDA determines (21 CFR 101.4(b)(14))). the FD&C Act establishes that a food that there is a lack of convincing This final determination is a 5 U.S.C. additive is unsafe for the purposes of evidence that a substance is GRAS or is 554(e) declaratory order regarding the section 402(a)(2)(C) of the FD&C Act otherwise exempt from the definition of status of PHOs. Consistent with (and therefore adulterated) unless a food additive in section 201(s) of the § 170.38(b)(3), we have reviewed the certain criteria are met, such as FD&C Act, FDA will publish a notice comments received and determined that conformance with a regulation thereof in the Federal Register there is a lack of convincing evidence prescribing the conditions under which (§ 170.38(b)(3)). Such a notice ‘‘shall that PHOs are GRAS. Thus, consistent the additive may be safely used. Section provide for the use of the additive in with § 170.38(c)(3), we are publishing a 409 of the FD&C Act also sets forth a food or food contact surfaces as follows: notice thereof in the Federal Register process by which we administer the (1) It may promulgate a food additive that requires discontinuation of the use review of food additive petitions and regulation governing use of the of these additives. Moreover, we are may establish regulations prescribing additive[;] (2) It may promulgate an providing advance notice of our conditions of safe use for such interim food additive regulation intention to undertake rulemaking with additives. Thus, we have explicit governing use of the additive[;] (3) It respect to the uses of PHOs explicitly statutory authority to review, approve, may require discontinuation of the use permitted for use by regulation and and deny food additive petitions. of the additive[;] (4) It may adopt any other conforming changes. Because it is necessary to determine combination of the above three (Comment 12) Some comments whether the use of a substance is GRAS approaches for different uses or levels of argued that FDA must determine the as part of identifying it as a food use of the additive’’ (§ 170.38(c)). GRAS status of PHOs through notice- additive, it is implicit in this statutory On our own initiative, we began an and-comment rulemaking. structure that we also have the authority administrative proceeding to formulate (Response) FDA agrees that we must to determine whether the use of a a 5 U.S.C. 554(e) declaratory order to conduct rulemaking to revise substance is, or is not, GRAS. The remove uncertainty regarding the GRAS §§ 184.1555(c)(2) and 184.1472(b), statute does not explicitly provide the status of PHOs. Accordingly, we which explicitly permit the use of procedure we must use to make such published a notice in the Federal partially hydrogenated LEAR oil and determinations. Thus, we may choose to Register, consistent with § 170.38(b), partially hydrogenated menhaden oil, use either rulemaking or adjudication. communicating our tentative respectively. FDA will also consider ‘‘The choice between rule-making or determination that PHOs are no longer taking further action to revise declaratory order is primarily one for GRAS for any use in food, and allowed regulations regarding the standards of the agency regardless of whether the 60 days for comments (78 FR 67169 identity for peanut butter (§ 164.150(c)) decision may affect policy and have (November 8, 2013)). We later extended and canned tuna (§ 161.190(a)(6)(viii)), general prospective application.’’ (See the comment period for an additional 60 the regulation regarding ingredient Viacom v. FCC, 672 F.2d 1034, 1042 days (78 FR 79701 (December 31, designations for PHOs (§ 101.4(b)(14)), (2nd Cir. 1982). See also SEC v. 2013)). and nutrition labeling regulations Chenery, 332 U.S. 194, 203 (1947); In the tentative determination, FDA regarding trans fats (§§ 101.9(c)(2)(ii) NLRB v. Wyman-Gordon Co., 394 U.S. noted that two PHOs had been affirmed and 101.36(b)(2)(i)). We note that 759 (1969); NLRB v. Bell Aerospace Co., by regulation as GRAS for use in food although trans fat does occur naturally 416 U.S. 267, 294 (1974); Almy v. (78 FR 67169 at 67171; the partially in some product groups such as dairy Sebelius, 679 F.3d 297, 303 (4th Cir. hydrogenated versions of low erucic foods, it is only likely to be present at 2012); City of Arlington, Texas v. FCC, acid rapeseed oil (LEAR oil; levels at or above 0.5 g per serving in 133 S. Ct. 1863, 1874 (2013); Qwest § 184.1555(c)(2)) and menhaden oil products containing PHOs. Servs. Corp. v. FCC, 509 F.3d 531, 536– (§ 184.1472(b)). We also noted that the We do not agree that we must 37 (D.C. Cir. 2007) (‘‘Most norms that nature of some of the products for determine the GRAS status of PHOs emerge from a rulemaking are equally which there are standards of identity is generally via rulemaking. FDA may capable of emerging (legitimately) from such that PHOs historically have been properly make such a determination in an adjudication, and accordingly used in their manufacture in an order, as we have chosen to do here. agencies have very broad discretion conformance with those standards (78 This is not the first time FDA has issued whether to proceed by way of FR 67169 at 67171). However, we also a declaratory order when determining adjudication or rulemaking’’ (internal noted that no food standard of identity that a substance is not GRAS and is a citations and quotations omitted)). requires the use of PHOs and, therefore, food additive. See 55 FR 50777, 50778 Determining that PHOs are no longer industry’s ability to comply with any (Declaratory Order regarding Vitamin K GRAS for use in human food in a standard would not be prevented by a Active Substances in Animal Food, declaratory order issued as a product of change in the regulatory status of PHOs. issued under 21 CFR 570.38, the informal adjudication is well within As discussed in section III.B, two regulation for animal food that parallels FDA’s discretion under the FD&C Act standards of identity explicitly mention § 170.38 for human food). and the APA. Whether PHOs are GRAS PHOs in allowing partially We have authority to administer the for use in human food is a ‘‘concrete hydrogenated vegetable oil as an statutory provisions of the FD&C Act and narrow question[] of law the optional ingredient; the standards of that are most relevant to this resolution[] of which would have an identity for peanut butter (§ 164.150 (21 determination, namely, are sections immediate and determinable impact on CFR 164.150)) and canned tuna 201(s), 402(a)(2)(C), and 409 of the specific factual scenarios’’ (City of (§ 161.190 (21 CFR 161.190)). Because FD&C Act. Section 201(s) of the FD&C Arlington v. FCC, 668 F.3d 229, 243 (5th these standards do not require the use Act defines a food additive, in part, as Cir. 2012)). (See also Qwest Servs. Corp.,

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00051 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34657

509 F.3d at 536–37; Chisholm v. FCC, enforcement action. This is not a to submit food additive petitions under 538 F.2d 349, 364–66 (D.C. Cir. 1976); statement of policy. This declaratory section 409 of the FD&C Act if industry American Bar Association, A Guide to order has the force and effect of law. believes that it is possible to establish, Federal Agency Adjudication 8 (Jeffrey (Comment 13) Some comments by regulation, safe conditions of use of B. Litwak, ed., 2012) (Agency order to assumed that this order was a statement PHOs. We are establishing a compliance withdraw certain food from the market, of policy, and, on that basis, argued that date of June 18, 2018 for this order to which has particular applicability and this action violates Due Process allow time for submission of such future effect, provided as an example of requirements. petitions and their review and approval, adjudication)). We are issuing this (Response) As explained in our if applicable requirements are met. declaratory order to remove uncertainty response to comment 10, that IV. Discussion of Scientific Issues, and as to the status of PHOs as food assumption is incorrect. Further, FDA’s Related Comments With FDA additives. The order is a product of an order and the process used in its Responses informal adjudication that included formulation raise no Due Process notice to affected parties via publication concern. A. Intake Assessment (Comment 14) Some comments of the tentative determination in the In the November 2013 notice, we argued that FDA did not conduct a full Federal Register and an opportunity for discussed dietary intake of trans fat Regulatory Impact Analysis in issuing affected parties to be heard by from PHOs, estimated in 2010 and the tentative determination. submitting comments to the Agency. updated in 2012 (78 FR 67169 at 67171). Such procedures are appropriate for the (Response) As discussed previously in this section, this final determination is The intake assessment was done for four formulation of declaratory orders. (See, reasons: (1) To determine the impact of e.g., Weinberger v. Hynson, Westcott a declaratory order issued as the result of informal adjudication to remove the 2003 labeling rule and subsequent and Dunning Inc., 412 U.S. 609, 626 reformulations; (2) to assist in our (1973); American Airlines v. Dep’t. of uncertainty regarding the status of PHOs. We have prepared a review of the citizen petitions, which Transportation, 202 F.3d 788, 796–797 are discussed in section V; (3) to (5th Cir. 2000). See also Lubbers, Jeffrey memorandum (Ref. 17) updating our previous estimate of economic impact consider strategies for further trans fat S. and Blake D. Morant, A reduction, if warranted; and (4) to better Reexamination of Federal Agency Use published in the November 2013 notice, using information available to us as well understand the current uses of PHOs of Declaratory Orders, 56 Admin. L. and identify products that still contain Rev. 1097, 1112–1114 (2004) and cases as information we received during the comment period. See discussion in high levels of trans fat. Our cited therein). Moreover, ‘‘adjudicatory determination regarding the GRAS decisions are not subject to the APA’s section VII. Further, we have stated our intention to conduct rulemaking status of PHOs relies on an analysis of notice-and-comment requirements’’ whether PHOs meet the GRAS standard (Blanca Telephone Co. v. FCC, 743 F.3d regarding uses of PHOs in our existing regulations, and such rulemakings will based on available scientific evidence; 860 (D.C. Cir. 2014)). the intake assessment was not the basis Issuance of a declaratory order is also be subject to the procedural requirements pertaining to rulemaking. for this determination. consistent with our regulations In 2012, we estimated the mean trans (§ 170.38(c)(3)), which provide that we (Comment 15) One comment stated that FDA must provide a more detailed fat intake from the use of PHOs to be 1.0 may publish a notice in the Federal grams per person per day (g/p/d; 0.5 Register that requires discontinuation of justification for this action than what was provided in the tentative percent of energy based on a 2,000 the use of these additives, and do not calorie diet 1) for the U.S. population specify that we must do so through determination because it is a change in FDA’s position regarding PHOs and aged 2 years or more. We also estimated rulemaking. Notably, other subsections intake for high-level consumers of § 170.38(c) mention promulgation of industry has a substantial reliance interest in the GRAS status of PHOs. (represented by intake at the 90th regulations, but § 170.38(c)(3), providing percentile), as well as a ‘‘high-intake’’ for prohibition of use, does not. (Response) In the tentative determination (78 FR 67169 at 67172) scenario that assumed consumers Moreover, when we make a consistently chose products with the determination under § 170.38 that a and in this order, FDA has explained the factual findings supporting this highest trans fat levels. We received a substance is not GRAS, we must take number of comments on our intake one (or a combination) of the actions action in detail. In section IV.B, we describe how the scientific evidence, assessment, including comments on listed in § 170.38(c). See Heterochemical assumptions, methodology, and Corp. v. FDA, 741 F. Supp. 382, 384 (E. and consensus among qualified experts regarding the safety of PHOs, has recommendations for future studies. D. N.Y. 1990). (Comment 16) One comment changed over time. We are not changing The purpose of a declaratory order is challenged FDA’s statement that intake our interpretation of the GRAS standard ‘‘to develop predictability in the law by of trans fat did not significantly change authorizing binding determinations or the relevant regulations. We are between 2010 and 2012. The comment which dispose of legal controversies determining that PHOs are no longer indicated that the intake of trans fat without the necessity of any party’s GRAS by applying the GRAS standard from the use of PHOs decreased by acting at his peril upon his own view’’ to current scientific evidence and the roughly 23% in that time period due to (U.S. Department of Justice, Attorney views of qualified experts about the significant reformulation efforts by the General’s Manual on the Administrative safety of PHOs. Moreover, reliance food industry. Procedure Act (1947) at 59, reprinted in interests are implicated whenever FDA (Response) FDA agrees that a Federal Administrative Procedure makes a determination that removes a comparison of the assessments from Sourcebook (William F. Funk et al. ed., substance from the food supply that has 2010 and 2012 demonstrates that ABA Section of Administrative Law and been previously used in food. FDA is reformulation has occurred and intake Regulatory Practice 3rd ed. 2000)). aware of such concerns; however, the has decreased. While the intake Members of industry are not, as some statutory standard for GRAS does not estimates did show a 23 percent comments suggested, faced with a allow FDA to consider the extent to choice between complying with a non- which industry has relied on GRAS uses 1 (1.0 g/p/d × 9 kcal/g × 100)/2,000 kcal/d = 0.5% binding statement of policy and facing of a substance. We encourage industry of energy.

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00052 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34658 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

decrease in trans fat intake between indicating a more inclusive review of contained a PHO was an overestimation 2010 and 2012 (1.3 g/p/d to 1.0 g/p/d), dietary intake of trans fat is warranted. of intake. One comment stated that this this change is small compared to the 3.3 Another comment stated that we did not assumption represents 40% of the g/p/d difference between FDA’s intake consider the cumulative effect of trans estimated daily intake of 1.0 g/p/d. estimate in the 2003 trans fat labeling fat because it did not present data on (Response) FDA disagrees with the final rule of 4.6 g/p/d and the 2010 intake from all sources, including comments. For most of the food estimate of 1.3 g/p/d (about a 72 percent ruminant TFA. products that declared 0 g trans fat on decrease). This was the context for the (Response) Our study was designed to the label, but contained a PHO, a level statement in the tentative determination assess trans fat intake from the use of based on analytical data was used. A that, ‘‘We do not consider this to be a PHOs, because they are the primary value of 0.4 g trans fat/serving was used significant change in the overall dietary source of IP–TFA, and IP–TFA was the for only 2 percent of all of the food intake of trans fat since 2010. However, focus of the intake assessment. As stated codes included in the intake assessment it suggests a continued downward trend in our tentative determination (78 FR (Ref. 16). The value of 0.4 g is the in the dietary intake of trans fat.’’ 67169 at 67172), the IOM’s amount of trans fat estimated to be in (Comment 17) Many comments stated recommendation is that trans fat in the food(s) that corresponds to a that a substantial number of products consumption should be kept as low as given food code that was used in the have been reformulated since the 2012 possible while consuming a intake assessment, and does not intake assessment and that we should nutritionally adequate diet, recognizing represent a percentage of total estimated revise our intake assessment for trans fat that trans fat occurs naturally in meat intake. As a result, we do not expect before issuing our final determination and dairy products from ruminant that using a lower value would on the GRAS status of PHOs. animals and that naturally-occurring significantly affect the overall estimated (Response) FDA agrees that trans fat is unavoidable in ordinary, intake of trans fat from the use of PHOs. reformulation efforts by industry are non-vegan diets without significant The use of 0.4 g trans fat/serving was continuing. However, the 2012 intake dietary adjustments that may introduce reserved for those cases where no other assessment was intended to be a undesirable effects. Therefore, our information was available (i.e., snapshot in time and was based on intake assessment focused only on trans analytical data or an appropriate products containing PHOs that were in fat from the use of PHOs, the primary surrogate). Furthermore, while the market at that time, and was done dietary source of IP–TFA, in which numerically 0.4 g is 40 percent of 1.0 g, for the reasons described previously in trans fat is produced intentionally and it is not appropriate to compare these this section. Given the evidence FDA is an integral component. two parameters. Many factors (i.e., the has reviewed and our determination (Comment 20) One comment urged amount of the particular food that PHOs are not GRAS for any use in FDA to reevaluate the intake of trans fat consumed, the percent of the population human food, an updated intake using the most recent National Health consuming the given food, and the level assessment for trans fats from PHOs is and Nutrition Examination Survey of trans fat in the particular food) were not needed at this time. Our (NHANES) data. The comment used to derive the overall estimated determination that PHOs are not GRAS suggested that the intake of trans fat trans fat intake. for use in human food does not rely on would be lower if the more recent (Comment 22) One comment the intake assessment. NHANES data were used because the suggested that American Oil Chemists (Comment 18) Some comments stated mandatory labeling rule for trans fat Society (AOCS) methods should be used that FDA should not use the ‘‘high became effective on January 1, 2006. for the intake assessment instead of the intake scenario’’ as justification for a (Response) While the 2003–2006 AOAC method 996.06 since the AOAC determination that PHOs are not GRAS. NHANES food consumption data were method is outdated and has not Related comments stated that the intake used in the 2010 and 2012 intake undergone validation. for the highest level consumers should assessments, the levels of trans fat in the (Response) FDA disagrees. This be determined directly rather than using food products were determined based AOAC method is widely used by worst-case scenario assumptions. on products that were available in the industry and other international (Response) FDA disagrees that the market from 2009 to 2012, therefore organizations as a method for high intake assessments provide capturing trans fat reductions due to determining the trans fat content in food justification for our determination product reformulation as a result of the products. Therefore, we considered the regarding the GRAS status of PHOs; the regulation in § 101.9(c)(2)(ii) (effective AOAC method to be appropriate for determination is based on our in 2006) requiring declaration of the analyzing food samples for the purposes assessment of whether any use of PHOs trans fat content of food in the nutrition of our intake assessment. Our choice of in human food meets the GRAS label. The consumption of products in the AOAC method is not intended to standard, based on available scientific the food categories in which PHOs are imply that industry must use this evidence. Our determination did not used would not be expected to change method to analyze food products. rely on the intake assessment. significantly over a few years because (Comment 23) Two comments (Comment 19) Several comments for the most part, foods tend to be indicated that a new intake assessment stated that FDA’s estimate did not commonly consumed with little or no should be performed using modeling to calculate intake from animal products change in consumption patterns over explore potential unintended that contain trans fat, and that FDA short periods of time. Further, we consequences of decreasing the trans fat should update the intake assessment to compared the typical intake of trans fat intake given the possible replacements include the intake of total trans fat from using the 2003–2006 and 2003–2008 for trans fat (e.g., saturated fat, both ruminant sources and IP–TFA. The NHANES food consumption data and carbohydrate) and their impact on CHD comments noted this was necessary to found that there were no significant risk. understand if dietary recommendations differences in the intakes (Ref. 16). (Response) The safety of other are being met. One comment indicated (Comment 21) Several comments substances that are possible that a recent publication suggests that suggested that using a value of 0.4 g replacements for PHOs is outside the the intake of trans fat from ruminant trans fat per serving for foods that scope of this order. However, although sources may be decreasing, thereby declared 0 g trans fat on the label, but we have not updated the intake

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00053 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34659

assessment since 2012, we have used findings reported in the literature since CHD risk. The consistency of the this intake assessment to calculate the 2003, when we had last reviewed the evidence from two different study expected impact of this order on CHD adverse effects of dietary trans fat in methodologies provides strong support events, taking into account possible support of the July 2003 final rule (68 for the conclusion that trans fatty acid replacements for PHOs (see section IV.B FR 41434 at 41442 through 41449). We intake has a progressive and linear effect for detailed discussion). noted that since 2003, both controlled that increases the risk of CHD. (Comment 24) One comment noted feeding trials and prospective Risk factors are variables that that FDA did not examine the use of observational studies published on trans correlate with incidence of a disease or each PHO and the probable fat consumption have consistently condition. Risk factors include social consumption of each use. confirmed the adverse health effects of and environmental factors in addition to (Response) FDA disagrees that we trans fat consumption on risk factor biological factors. A biomarker is a need to examine the intake of each PHO biomarkers (e.g., serum lipoproteins characteristic that can be objectively individually; the intent of the intake including LDL–C) and increased risk of measured and indicates physiological estimate was to evaluate the overall CHD (78 FR 67169 at 67172). We processes. A risk biomarker or risk intake of trans fat from the use of all describe these two types of studies factor biomarker is a biomarker that PHOs for the purposes described (controlled feeding trials and indicates a risk factor for a disease. In previously in this section. Estimating prospective observational studies) in other words, it is a biomarker that trans fat intake from individual PHOs further detail later in this section. We indicates a component of an would be an impractical undertaking, also cited a variety of different kinds of individual’s level of risk for developing and was not necessary for the purposes studies and review articles showing a disease or level of risk for developing of the intake assessment. that, in addition to an increased risk of complications of a disease (Ref. 19). (Comment 25) Two comments stated CHD, trans fat consumption (and, LDL–C, HDL–C, total-C/HDL–C ratio that intake should be evaluated based accordingly, consumption of food and LDL–C/HDL–C ratio are all on the presumption that all products products containing PHOs) has also currently considered to be risk with PHOs as an ingredient contain been connected to a number of other biomarkers for CHD (Refs. 19, 20, 21, trans fat at a specified level (e.g., 0.2 g/ adverse health effects (id.). These effects and 22). LDL–C is a risk factor serving or per reference amount included worsening insulin resistance, biomarker that is also a surrogate customarily consumed). These increasing diabetes risk, and adverse endpoint for CHD; a ‘‘surrogate’’ is a comments suggested that such an effects on fetuses and breastfeeding validated predictor of CHD and can assessment could provide support for an infants, such as impaired growth. substitute for actual disease occurrence alternative approach such as setting an in a clinical trial (Refs. 19, 20, and 21). allowable level of trans fat in foods. Since publication of the November HDL–C, total-C/HDL–C and LDL–C/ (Response) Because we have 2013 notice, we re-reviewed key HDL–C are recognized as major risk concluded that PHOs are no longer literature and expert panel reports factor biomarkers that, although they are GRAS, evaluating intake for alternative published since the 1990s on the not validated surrogate endpoints, are approaches, such as setting an allowable relationship between trans fat predictive of CHD risk (Refs. 19 and 22). level of trans fat in foods, is not planned consumption and CHD risk (Ref. 18). Effect of trans fat intake on blood at this time. Our review focused on the two main lipids in controlled feeding trials. In lines of scientific evidence linking trans controlled feeding trials, a type of B. Safety fat intakes and CHD: (1) The effect of randomized clinical trial, trans fatty In the Federal Register of November trans fat intake on blood lipids in acid intake increased LDL–C (‘‘bad’’ 17, 1999 (64 FR 62746), we issued a controlled feeding trials, a type of cholesterol), decreased HDL–C (‘‘good’’ proposed rule entitled ‘‘Food Labeling: randomized clinical trial; and (2) cholesterol) and increased ratios of Trans Fatty Acids in Nutrition Labeling, observational (epidemiological) studies total-C/HDL–C and LDL–C/HDL–C Nutrient Content Claims, and Health of trans fat intake and CHD risk in compared with the same amount of Claims.’’ The proposed rule would populations. Additionally, we reviewed energy intake (calories) from cis- require that trans fat content be the conclusions of recent U.S. and unsaturated fatty acids. Increases in provided in nutrition labeling, and international expert panels on the LDL–C, total-C/HDL–C and LDL–C/ concluded that dietary trans fats have health effects of trans fat. As HDL–C and decreases in HDL–C are adverse effects on blood cholesterol summarized in our review adverse changes with respect to CHD measures that are predictive of CHD memorandum (Ref. 18), the scientific risk. These adverse effects of trans fat risk, specifically low-density evidence, including combined analyses intake on blood lipids are based on lipoprotein cholesterol (LDL–C) levels of multiple studies (meta-analyses), controlled feeding trials, a study design (64 FR 62746 at 62754). In the Federal supports a progressive and linear cause that is able to reveal cause and effect Register of July 11, 2003 (68 FR 41434), and effect relationship between trans relationships between changes in trans we issued a final rule (the July 2003 fatty acid intake and adverse effects on fat intake and changes in blood lipids. final rule) amending the labeling blood lipids that predict CHD risk, In addition, increases in CHD risk with regulations to require declaration of including LDL–C, high-density increases in LDL–C also demonstrate trans fat content of food in the nutrition lipoprotein cholesterol (HDL–C) and cause and effect. As described in our label of conventional foods and dietary ratios such as total cholesterol (total-C)/ review memorandum (Ref. 18), supplements (68 FR 41434). In the July HDL–C and LDL–C/HDL–C. The combined analyses (meta-analyses) of 2003 final rule, we cited authoritative observational (epidemiological) studies multiple controlled feeding trials reports that recommended limiting demonstrating increased CHD risk demonstrate a progressive and linear intake of trans fat to reduce CHD risk associated with trans fat intake do not relationship between trans fatty acid (68 FR 41434 at 41442). prove cause and effect, but the results intake and adverse effects on blood In the November 2013 notice are consistent with and supportive of lipids including LDL–C, HDL–C, total- containing our tentative determination the evidence from controlled feeding C/HDL–C and LDL–C/HDL–C. The that PHOs are no longer GRAS for any trials of the adverse effect of trans fatty meta-analyses describe consistent use in human food, we summarized acid intake on blood lipids that predict quantitative relationships between trans

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00054 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34660 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

fat intake and blood lipids and show no literature published since 2008 and effects of TFA on total-C/HDL–C plus a evidence of a threshold below which summarized the findings (Ref. 23). The combination of emerging CHD risk trans fatty acids do not adversely affect major human health endpoints factor biomarkers (lipoprotein(a), blood lipids. evaluated for associations with trans fat apolipoproteinB/apolipoproteinA1 and Observational (epidemiological) intake reported in the literature C-reactive protein), as shown through studies of trans fat intake and CHD risk included CHD, all-cause mortality, controlled feeding trials; and Method 4, in populations. Epidemiology is the cardiovascular disease and stroke. Other based on association of TFA with CHD study of the distribution and causes of human health endpoints addressed in risk as shown through prospective disease in human populations. Analytic our search included various types of observational studies. Methods 1 and 2 epidemiology studies are those designed cancer, metabolic syndrome and were also used by FDA in analyzing the to test hypotheses regarding whether or diabetes, and adverse effects on fertility, 1999 and 2003 labeling regulations (64 not a particular exposure is associated pregnancy outcome, cognitive function, FR 62746 at 62768 and 68 FR 41434 at with causing or preventing a specific and mental health. The literature search 41479) and Methods 3 and 4 were based disease outcome. In prospective identified meta-analyses of published on published methods (Ref. 26). We observational (cohort) studies, subjects data; quantitative estimations to predict estimated the change in CHD risk using are classified according to presence or effects of replacing TFA in commercial each of these four methods as applied to absence of a particular factor (such as products; cross-sectional, case-control two different sets of scenarios for usual dietary intake of trans fat) and and prospective observational cohort replacement of IP–TFA, as follows. followed for a period of time to identify studies; and randomized controlled In general, fats and oils in foods have disease outcomes (such as heart attack trials, including controlled feeding carbon chains of various lengths, with or death from CHD). Strengths of the trials. Regarding cardiovascular the carbon atoms in these chains prospective observational study design diseases, the results of the literature connected by single or double bonds. If are that the time sequence of exposure search (Ref. 23) are consistent with the carbon chain contains no double and disease is clearly shown; exposures findings discussed in our November bonds, the fatty acid is called saturated. are identified at the outset of the study; 2013 notice (78 FR 67169 at 67172). If the carbon chain contains a single and measurement of exposure is not Findings associated with higher TFA double bond, the fatty acid is called affected by later disease status. Results intakes included increased risk of CHD, monounsaturated, and if the carbon of four major prospective studies, some adverse effects on biomarkers associated chain contains two or more double with one or more updates during the with CHD, and increased subclinical bonds, the fatty acid is called followup period, consistently show atherosclerosis. Some recent prospective polyunsaturated. Most naturally- higher trans fat intake associated with observational studies also found occurring dietary unsaturated fatty acids increased CHD risk. The association is associations between increased trans fat have double bonds in a ‘‘cis’’ positive and progressive, with no intake and increased risk of stroke, configuration, that is, the two hydrogen indication of a threshold. A 2009 meta- which was a new finding (Refs. 18 and atoms attached to two carbons are on analysis of the major prospective 23). Further understanding of the the same side of the molecule at the studies, based on almost 5,000 CHD apparent association between increased double bond. Thus, the major chemical events in almost 140,000 subjects, found trans fat intake and increased risk of forms of fatty acids in foods are that each additional 2 percent of energy stroke requires additional research, such saturated fatty acids (SFAs), cis- intake from trans fat increased CHD risk as whether the association may differ by monounsaturated fatty acids (cis- by 23 percent compared with the same age, sex, aspirin use, geographic region energy intake from carbohydrate. MUFAs) and cis-polyunsaturated fatty and other risk factors (Refs. 18, 23, and Conclusions of recent U.S. and acids (cis-PUFAs). (By comparison, in a international expert panels on the 24). For the association of trans fat ‘‘trans’’ configuration, the hydrogen health effects of trans fat. As described intake with other human health effects, atoms attached to the carbon atoms at a in our review memorandum (Ref. 18), such as various types of cancer, double bond are not on the same side international and U.S. expert panels, metabolic syndrome and diabetes, and of the double bond). (See definitions in using additional scientific evidence adverse effects on fertility, pregnancy 64 FR 62746 at 62748 to 62749 available since 2002, have continued to outcome, cognitive function and mental (November 17, 1999).) recognize the positive linear trend health, the literature reports remained One set of scenarios focuses solely on between LDL-C and trans fat intake and limited or inconclusive. IP–TFA and the estimated change in the consistent association of trans fat Since publication of the November CHD risk by hypothetically replacing intake and CHD risk in prospective 2013 notice, we also conducted a IP–TFA with each of the major chemical observational studies. The panels have quantitative estimate of the potential forms of macronutrient fatty acids in concluded that trans fats are not health benefits expected to result from foods—i.e., SFAs, cis-MUFAs or cis- essential nutrients in the diet, and have removal of IP–TFA from PHOs from the PUFAs. The other set of scenarios recommended that consumption be kept food supply (Ref. 25). We did this to focuses not only on IP–TFA but also on as low as possible. Recommendations to analyze the expected public health the other fatty acids contained in PHOs. avoid industrial trans fat intake have benefit of removing PHOs from the food This hypothetical set of scenarios come from panels with both clinical and supply. We used four methods for illustrates the estimated change in CHD public health focus. Moreover, estimating changes in CHD risk likely to risk with replacing PHOs in the international and U.S. panels have result from replacement of IP–TFA: marketplace that contain 20 percent, 35 expressed concern regarding population Method 1, based on effects of TFA on percent, or 45 percent IP–TFA, with mean intakes of industrial trans fat LDL–C, a validated surrogate endpoint other likely replacement fats and oils. intakes of 1 percent of energy and lower, biomarker for CHD, as shown through Therefore, this scenario accounts for not recognizing that subgroups may be controlled feeding trials; Method 2, only the replacement of IP–TFA with consuming relatively high levels. based on effects of TFA on LDL–C plus macronutrient fatty acids but also the Since publication of the November HDL–C, a major CHD risk factor replacement of the overall fatty acid 2013 notice, we also conducted a biomarker, as shown through controlled components (or profiles) of the PHOs systematic search of the peer-reviewed feeding trials; Method 3, based on with the fatty acid components (or

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00055 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34661

profiles) found in the various formulations of PHOs, with eight PHOs are replaced by oils high in replacement fats and oils. alternative fats and oils (soybean oil, saturated fat. Consistent with published In the first set of scenarios, we canola oil, cottonseed oil, high oleic analyses, our results show that assumed that the current mean intake of sunflower oil, high oleic soybean oil, estimated changes in CHD risk expected 0.5 percent of total daily calories palm oil, lard, and butter). This to occur with replacement of PHOs (energy) from IP–TFA among U.S. adults approach covers a range of composition depends on the fatty acid profiles of was replaced by the same percent of of replacement fats and oils, from highly both the PHOs and the replacement fats energy from three types of saturated (high in SFAs) to highly and oils (Refs. 25, 26, and 28). We also macronutrient fatty acids, cis-mono- or unsaturated (high in cis-MUFAs and/or note that research indicates removal of polyunsaturated fatty acids and cis-PUFAs), and is based on that trans fat over the past decade has saturated fatty acids) (cis-MUFAs, cis- reported in 2009 by Mozaffarian and generally not been accompanied by PUFAs, and SFAs). As measures of risk Clarke as part of the World Health extensive increases in saturated fat (Ref. reduction, we calculated estimated Organization (WHO) scientific update 29), suggesting that all IP–TFA currently percent changes in CHD risk and on trans fatty acids (Refs. 25 and 26). in the marketplace would not likely be estimated reduction in annual total Among the eight fats and oils, soybean replaced by oils high in saturated fat. cases of CHD, including CHD-related oil and cottonseed oil contain the Among the strengths of our deaths. We based changes in CHD cases highest amounts of cis-PUFAs. Canola quantitative analyses is the use of and deaths on a baseline of 915,000 oil, high oleic acid sunflower oil, and established cause and effect annual new and recurrent fatal and non- high oleic acid soybean oil have the relationships between IP–TFA intakes fatal cases of CHD in U.S. adults, with highest amounts of cis-MUFAs. Butter and adverse changes in CHD biomarker a 41 percent fatality rate (Ref. 27). has the highest amount of SFAs; lard risk factors, including LDL–C and HDL– Results showed an estimated and palm oil are also high in SFAs. We C, derived from high quality, controlled reduction in CHD with replacement of used the same four methods to estimate feeding trials. Our assessments also IP–TFA with each of the fatty acids (cis- risk reduction in this analysis. These relied on a set of emerging risk factors MUFA, PUFA, or SFA), using each of calculations take into account the fatty for CHD, including total cholesterol to the four estimation methods. The acid profiles of the replacement fats and HDL–C ratios, Apo-lipoprotein B to estimated decrease in CHD ranged from oils and the other fatty acids in the Apo-lipoprotein A–I ratios, 0.1 percent to 6.0 percent. This PHOs in addition to IP–TFA. lipoprotein(a) and C-reactive protein corresponded to prevention of 1,180 to Overall, the analysis showed that changes obtained from these same 7,510 annual CHD cases, including 490 removing 0.5 percent of energy from IP– feeding trials. In addition, we relied on to 3,120 deaths, in Method 1 (0.1 TFA by replacing an example PHO information from direct observations of percent to 0.8 percent decrease in CHD containing 35 percent IP–TFA with each CHD outcomes associated with frequent risk based on LDL–C), 9,230 to 15,560 of eight alternative fats and oils would usual intake assessments of trans fatty cases, including 3,830 to 6,460 deaths, reduce CHD risk by 0.4 percent to 1.5 acids and other macronutrient fatty in Method 2 (1.0 percent to 1.7 percent percent across the respective acids in meta-analyses of four large decrease in CHD risk based on LDL–C replacement fats and oils using Method cohorts with long-term followups. These and HDL–C), and 18,660 to 54,900 2, 2.3 percent to 3.0 percent using estimates build on the agency’s previous cases, including 7,740 to 22,770 deaths, Method 3, and 2.7 percent to 6.4 percent quantitative assessment based on short- in Method 3 (2.0 percent to 2.5 percent using Method 4. This would correspond term changes in LDL–C and HDL–C decrease in CHD risk using a to prevention of 3,900 to 58,210 CHD alone (68 FR 41434 at 41466 to 41492). combination of biomarkers) and Method cases including 1,620 to 23,350 CHD We acknowledge that there are always 4 (4.2 percent to 6.0 percent decrease in deaths per year. some uncertainties in assessing risk. CHD risk using observed CHD In a few instances, the analysis in the The estimates we used were based on outcomes). Method 4, based on long- second set of scenarios estimated that 100 percent replacement of IP–TFA by term observations of CHD outcomes in there would be increased CHD risk a group of individual types of fatty acids prospective studies, produced greater when examples of PHOs were replaced or by individual alternative fats and reduction estimates in risk than did entirely with fats or oils high in oils, when actual replacement mixes of Methods 1 and 2, which were based on saturated fat (Ref. 25) using Method 1. fats and oils might vary and individual short-term changes in blood lipid risk This reflects the saturated fatty acids in diets would reflect a combination of factors in controlled feeding trials. This alternative fats and oils replacing the replacement fatty acids and replacement suggests that there may be additional cis-unsaturated fatty acids present in the fats and oils. We assumed a no mechanisms, besides changes in blood PHO in addition to IP–TFA. Method 1 threshold, linear relationship between lipids, through which trans fat alone likely underestimates the overall changes in IP–TFA intakes and changes consumption contributes to CHD risk. change in risk that would result from in biomarker risk factors for CHD Thus, the adverse effects from trans fat replacing PHOs containing IP–TFA because current scientific evidence intake may be greater than predicted because it analyzes only impacts on indicates that the relationship between solely by changes in blood lipids. The LDL–C alone and therefore does not trans fatty acid intake and LDL–C, HDL– greater estimated reduction in CHD in account for the demonstrated adverse C and the total cholesterol to LDL Method 3, compared with Methods 1 effects of IP–TFA on HDL–C, or the cholesterol ratio is progressive and and 2, suggests that the emerging risk adverse effects of IP–TFA on other linear. factor biomarkers in Method 3 may help emerging CHD risk factors. Methods 2, Given these uncertainties, our to identify additional mechanisms 3, and 4 in the second set of scenarios, assessments for the change of CHD risk through which trans fat contributes to which consider other known risk factors at the current U.S. mean daily intake of CHD risk. as well as LDL–C, provides a more 0.5 percent of energy derived from IP– In the second set of scenarios, we thorough estimate of risk reduction than TFA are conservative estimates. The estimated the reduction in risk by considering only LDL–C in isolation, results also suggest that a small shift to replacing the same 0.5 percent of energy and leads us to conclude that there lower CHD risk could prevent large from IP–TFA, along with the other would be an expected benefit to public numbers of annual cases of CHD and component fatty acids in three different health from PHO replacement even if CHD-related deaths. The current U.S.

VerDate Sep<11>2014 19:54 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00056 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34662 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

background rates for CHD are already FDA were not designed to address the did not establish causality between low high, with considerable baseline impact of lowering TFA intake below doses of TFA (less than 1% of caloric variability due to abnormal serum lipid 1% of energy. The comments asserted energy) and increased CHD risk. Other profiles in large percent of U.S. adults that although the expert panel reports comments stated that the review of (33.5 percent have elevated LDL–C) and state that there is no threshold intake available data shows that low levels of other risk factors for CHD (Ref. 25). level for IP–TFA that would not TFA intake (3% of energy or less) have More people may be vulnerable to CHD increase an individual’s risk of CHD or no effect on serum LDL–C and total-C at the current mean intake of IP–TFA adverse effects on risk factors for CHD, levels. Some comments criticized FDA’s from PHOs than the risk reduction a review of the supporting reliance on the Ascherio et al. 1999 estimates as discussed above. documentation accompanying the paper (Ref. 31) and raised issues with In sum, our quantitative estimates reports does not support this statement; this paper and the linear extrapolation demonstrate that large numbers of CHD rather, the comments noted that panel used by the researchers. One comment events and deaths may be prevented reports indicate that due to the paucity suggested that using a different dose- with the elimination of PHOs. We also of evidence in the 0 to 4% energy range, response model is a more appropriate note that our estimates are in line with no evidence-based conclusions could be approach to determine the relationship published results regarding potential made. between PHOs and LDL–C and HDL–C, effects of replacing PHOs (Refs. 26 and (Response) FDA disagrees; the rather than defaulting to a linear 28). In replacing PHOs containing IP– published research described in our function, due to the quantity and type TFA, a more significant reduction in review memorandum (Ref. 18) includes of data available at low intake levels. CHD risk is estimated by replacement six regression analyses of controlled One comment stated that, in general, with vegetable oils containing higher feeding trials summarizing the dose- linear regression is an inappropriate tool amounts of cis-unsaturated fatty acids response relationship of IP–TFA on to determine a safe or unsafe level of a than with those high in saturated fatty blood cholesterol levels, published from dietary substance and questioned the acids, but we expect a risk reduction 1995 to 2010. In addition, a 2010 meta- use of low-dose linear extrapolation in even if IP–TFA is replaced with fats and analysis included 23 trans fat feeding this instance. oils high in saturated fatty acids, based trials and 28 TFA levels, including a (Response) FDA disagrees with these on our conservative risk estimates using low-dose level of 0.4 percent of energy comments. Given that effects of trans fat combinations of the four peer-reviewed (or less than the current mean intake) on LDL–C have been demonstrated at methods with two different sets of likely (Ref. 30). Across these regression doses as low as 0.4 percent and 2.8 scenarios for IP–TFA replacement for analyses, the reported effect of TFA on percent of caloric energy (Table 2 in Ref. each method. Additional details of these LDL–C, a validated surrogate biomarker 18), FDA disagrees that there is no results, and results for replacement of that serves as a direct causal link to evidence of an adverse effect from trans example PHOs containing 20 percent CHD, was very consistent and the fat intake below 3 percent of energy. In IP–TFA and 45 percent IP–TFA, are analyses showed a linear dose-response, addition, results of regression analyses provided in our review memorandum with an increase in LDL–C of about published from 1995 to 2010, including (Ref. 25). 0.038 to 0.049 millimoles per liter Ascherio et al. 1999 (Refs. 26, 30, 31, 32, We have also analyzed the comments (mmol/L) for each 1 percent of energy 33, and 34), are very consistent we received regarding the scientific intake from replacement of cis- regarding the effect of TFA on serum basis for our tentative determination in monounsaturated fat with trans fat lipids, thus indicating that the the November 2013 notice. Comments (Table 3 in Ref. 18). The regression relationship between TFA intake and regarding the safety of PHOs that were analyses also showed a consistent linear CHD risk is progressive and linear with opposed to our tentative determination dose response for HDL–C, with a no evidence of a threshold at which were generally related to one of four decrease of about 0.008 to 0.013 mmol/ effects would not be expected to occur. subject areas: (1) Dose-response L for each 1 percent of energy from Furthermore, we are not aware of any relationship of trans fat intake and replacement of cis-monounsaturated fat published study that supports an abrupt adverse health effects in human studies with trans fat (Table 3 in Ref. 18). reduction in the adverse effects of TFA and whether there is a threshold below Therefore, we conclude that the across the relatively narrow intake range which intake of trans fats is generally available data show that even at low of 0 percent to 3 percent of energy nor recognized as safe; (2) reliance on expert intake levels (e.g., below 3 percent are we aware of any published scientific panel reports and recommendations; (3) energy) there is no identifiable reports that provide a dose-response health benefits and clinical significance threshold, rather the available data model that might reveal a different of replacements for PHOs; and (4) support a conclusion that IP–TFA relationship for TFA intake and CHD alternative approaches. Comments causes a linear increase in blood levels risk that is generally accepted by regarding the safety of PHOs that were of LDL–C, a validated surrogate qualified experts. FDA is aware of an in support of our determination raised biomarker of CHD risk and a linear unpublished meta-regression analysis, concerns about other adverse health decrease in blood levels of HDL–C, a including consideration of the low- effects besides effects on LDL–C, such as major risk biomarker for CHD. If intake range (Ref. 35), suggesting that adverse effects on other risk factors for interested parties are or become aware the data on dietary trans fat intake and CHD (e.g., HDL–C, total-C/HDL–C ratio, of information and data supporting changes in LDL–C may fit a dose- LDL–C/HDL–C ratio, and other lipid establishment of a threshold, such response curve that is non-linear. and non-lipid biomarkers), information and data could be However, this analysis is neither inflammatory effects, harm to submitted to FDA as part of a food published (generally available) nor does subpopulations, and increased diabetes additive petition(s) proposing safe it demonstrate a consensus of expert risk. conditions of use for PHOs. opinion that the use of PHOs at low (Comment 27) Many comments 1. Dose-Response and Evidence of a levels in food is safe as required for disagreed with our conclusion that there general recognition of safety.2 Threshold Level is a linear relationship between TFA (Comment 26) A number of comments intake and LDL–C at low TFA intake 2 FDA also reviewed and considered an stated that the studies relied upon by levels. Some comments stated that we unpublished report of this analysis and its

VerDate Sep<11>2014 19:54 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00057 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34663

Further, we did not rely solely on the analyses discussed previously relationship between a substance and a Ascherio et al. 1999 paper regarding the supported the linear proportionality of disease, and are subject to various forms effect of IP–TFA intake on serum LDL– the data, and the quantitative of bias. C and other lipid biomarkers. Over time, relationships of dose-response are very (Response) Although observational the number of studies covered by the consistent across the analyses (Ref. 18). studies with long-term followup do not published regression analyses or meta- The proportional relationship of trans prove cause and effect, the results are analyses increased from 5 studies and 6 fat intake and blood lipids has also been consistent with and supportive of the TFA levels in 1995 (Ref. 32) through 8 repeatedly affirmed by a series of expert conclusions from the controlled feeding studies and 12 TFA levels in 1999 (Ref. panels (Ref. 18). Therefore, we conclude trial evidence discussed previously in 31) to 23 studies and 28 TFA levels in that the totality of the data supports the this section (which does demonstrate 2010 (Ref. 30). Across these studies, the proportionality of changes in trans fat cause and effect). The consistency of the reported magnitude of the effect of IP– intake and changes in blood lipids (and evidence from two different study TFA on LDL–C and HDL–C levels is therefore, CHD risk) and supports the methodologies is strong support for the very consistent. Furthermore, FDA notes use of a linear regression model to conclusion that trans fatty acid intake that the 2009 National Research Council describe this relationship. has a progressive and linear effect that report, Science and Decisions: (Comment 28) Some comments increases the risk of CHD. Our review Advancing Risk Assessment (Ref. 36), objected to the approach of ‘‘forcing’’ memorandum (Ref. 18) provides a describes conceptual models in which the regression line of the dose-response summary of the scientific evidence from low-dose linearity with no threshold curve through zero (the origin), as done the observational studies on the can arise. Absent evidence of a by Ascherio et al. 1999 (Ref. 31) and association of TFA intake and actual threshold intake level for TFA that does believed this was not appropriate. CHD outcomes in large populations and not increase an individual’s risk of CHD (Response) FDA disagrees. Whether or addresses in detail the study designs or adverse effects on risk factors for not to fix the intercept at zero depends and adjustments for confounding CHD, FDA concludes that a linear low- on the meaning of the data, the research variables. There are four major dose extrapolation is appropriate for question to be answered, and the prospective observational studies (Refs. assessing the dose-response relationship particular study design. (We further 46, 47, 48, 49, 50, 51, and 52), some between TFA intake and risk of CHD (as discuss the methodology for the meta- with one or more updates during the evidenced by effects on LDL–C, a analyses in our review memorandum followup period (e.g., the Nurses’ Health validated surrogate biomarker for CHD, (Ref. 18)). In feeding studies where the Study had followups at 8, 14, and 20 and HDL–C, a risk biomarker (Ref. 18)). total energy intake remains the same for years), that are further discussed in Our conclusion that there is a linear both control and treatment groups, the detail in one of our review memoranda zero intercept means that, with zero relationship (also known as a (Ref. 18). These are prospective (cohort) intake of trans fat, there is no effect of proportional effect, or proportionality) studies, which is the strongest study trans fat on (that is, no change in) the between trans fat intake and CHD risk design for observational studies, and the LDL–C, the LDL–C/HDL–C ratio, or is consistent with the body of evidence results consistently show that higher other serum lipid biomarker being from controlled feeding studies on the trans fat intake is associated with studied. This is the one data point that proportionality of fatty acid intake and increased CHD risk. In several studies, is known to be true by virtue of the blood lipids, beginning with landmark not only was the association of the study design, and many analyses using highest versus lowest level (category) of studies in the 1950s and 1960s (Refs. 18, this approach have been published in trans fat intake with greater CHD risk 37, 38, 39, and 40). Meta-analyses in the peer-reviewed literature (Refs. 30, 31, statistically significant, but also there 1990s and early 2000s showed that the 32, 44, and 45). In these analyses, the was a significant test for linear trend, proportionality in the earlier landmark authors calculated the differences in indicating a positive and progressive studies extended not only to total serum lipid levels between the trans fat association of trans fat intake with CHD cholesterol but to LDL–C, HDL–C, total- diet and the control diet for each risk (or CHD deaths) across levels (low, C/HDL–C ratio and LDL–C/HDL–C ratio controlled feeding trial, with adjustment intermediate, or high categories) of (Refs. 33, 41, and 42). Authors of a 1992 for differences in intake of the other intake (Refs. 46, 48, 49, 50, and 51). In meta-analysis noted, ‘‘a simple linear fatty acids between the two diets, using addition to the analysis of trans fat model in which diets are characterized published dose-response coefficients intake grouped in several levels or solely by their contents of saturated, (Refs. 33 and 42). The serum lipid and categories, in certain studies, numerical monounsaturated and polyunsaturated trans fat intake differences for each trans fat intake, as a continuous fatty acids goes a long way toward study were included in a linear variable, was significantly associated predicting group mean changes in regression model and expressed with with CHD risk, again indicating a serum lipid and lipoprotein levels’’ (Ref. respect to a specific replacement positive and progressive association of 42). Results of an early controlled macronutrient (such as cis- increased trans fat intake with increased feeding trial of trans fat intake and LDL– monounsaturated fatty acids or CHD risk across the range of observed C and HDL–C were questioned because carbohydrate). Therefore, we conclude intake (Refs. 49 and 51). of the high trans fat intake (Ref. 43). that it is logical and appropriate to fit There are also a number of meta- However, when combined with a (not ‘‘force’’) the regression lines analyses of the major prospective subsequent study at a lower dose, through zero because a zero change in studies (Refs. 26, 51, 52, 53, 54, and 55). preliminary data from these two studies trans fat intake results in zero change in In a 2009 meta-analysis, based on suggested that the effect of trans fat blood lipids attributable to trans fat almost 5,000 CHD events in almost intake on LDL–C and HDL–C is intake. 140,000 subjects, each additional 2 proportional (Ref. 18). Subsequent meta- (Comment 29) Some comments percent of energy intake from trans fat criticizing our scientific review stated increased CHD risk by 23 percent executive summary, which were submitted to FDA that prospective observational compared with the same energy intake with the request that they be kept confidential. FDA is including these documents in the administrative (epidemiological) studies which we from carbohydrate (Ref. 52). The record for this matter but is not placing them in the relied on were not designed to magnitude of the increase in CHD risk public docket because they are confidential. demonstrate a cause and effect associated with trans fat intake among

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00058 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34664 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

meta-analyses has remained consistent and did not consider other CHD risk (Response) We used LDL–C, a over time, including the studies with factor biomarkers such as HDL–C, or validated surrogate endpoint biomarker additional updates during the followup total-C/HDL–C or LDL–C/HDL–C ratios. for CHD (Ref. 21), as the primary periods. Further, the prospective studies The paper focused on methodology for endpoint for evaluating the adverse measure actual CHD occurrence in large attempting to identify a tolerable upper effects of IP–TFA intake from PHOs. As groups of people over long time periods, intake level for trans fat. The discussed previously in this section, and describe all CHD risk associated appropriateness of fitting the intercept validated surrogate endpoint biomarkers with trans fat intake, regardless of the through zero in a regression analysis are those that have been shown to be mechanism of action by which trans fat depends on the meaning of the data, the valid predictors of disease risk and may intake may be associated with CHD (i.e., research question to be answered, and therefore be used in place of clinical these studies do not rely on biomarkers the particular study design, and is measurement of the incidence of disease or risk factors but instead measure discussed further in our response to (Refs. 19 and 20). In addition, we actual occurrence of disease). The Comment 28. considered the adverse effects of trans magnitude of the observed CHD risk In addition to the feeding trial data fat intake on other risk factor from TFA intake is greater in the discussed in the 2011 publication, the biomarkers, including HDL–C and the prospective observational studies than authors of the 2011 paper presented LDL–C/HDL–C and total-C/HDL–C from the controlled feeding studies. data from prospective observational ratios. In fact, these other risk factor We also reviewed related studies showing that, compared with biomarkers indicate additional adverse observational studies of TFA intake and the lowest trans fat intake level, there effects of IP–TFA, beyond the primary cardiovascular disease health outcomes was a statistically significant increase in adverse effect of raising LDL–C. that considered all causes of mortality CHD risk at some levels of trans fat Although these other risk factor and cardiovascular disease endpoints intake, but not at others. Based on this, biomarkers are not validated surrogate other than CHD, as well as studies that they stated that, at least theoretically, ‘‘a endpoint biomarkers for CHD, they raise used blood and tissue levels as threshold level could be identified for significant questions about the safety of biomarkers of TFA intake instead of trans and saturated fat,’’ but they were PHOs and are therefore relevant to our dietary questionnaires, and not actually able to identify any specific determination that PHOs are not GRAS. retrospective case control studies (Ref. threshold level. We note that other data For example, HDL–C levels have been 18). The results from these studies from prospective studies that were not shown to be a useful predictor of CHD generally showed trans fat intake or discussed in this paper support the risk (Refs. 22 and 57). Because it has not biomarkers associated with adverse conclusion that there is a direct and been shown that drug therapy to raise health outcomes. The consistent progressive relationship between TFA HDL–C decreases CHD in clinical trials, findings of adverse health effects of intake and CHD risk, and no threshold HDL–C is not considered a validated trans fat from these studies with has been identified. Several studies surrogate endpoint for CHD (Ref. 19). different methodologies strengthen our showed a positive trend for higher CHD We did not primarily rely on the conclusions based on the evidence from risk with higher intake categories of relationship between trans fat intake the major prospective observational TFA that was statistically significant and adverse effects on HDL–C and CHD studies and controlled feeding studies (Refs. 46, 48, 49, 50, and 51) and certain risk, we recognize that a relationship is summarized previously. studies also analyzed numerical TFA known to exist and therefore considered (Comment 30) Several comments intake without using categories (that is, it in our analysis. We discussed this cited a 2011 publication by FDA authors as a continuous variable) and found a issue in detail in the July 2003 final rule (Ref. 56) as evidence of PHO safety and significant positive linear association of (68 FR at 41434 at 41448 through evidence that a threshold can be TFA intake with CHD risk across the 41449). determined below which there is range of usual TFA intake levels of Recent studies have affirmed HDL–C general recognition of safety. The participants in the studies (Refs. 49 and and total-C/HDL–C ratio as risk factors comments argued that these authors 51). These results, not discussed in the that predict CHD (Ref. 18). In a large, reviewed data from clinical trials to paper, are inconsistent with the pooled meta-analysis of prospective assess the relationship between trans fat existence of a threshold. Therefore, we observational studies, including 3,020 intake and LDL–C and total-C and that conclude that there is no currently CHD deaths during 1.5 million person- their regression analysis showed no identifiable threshold below which years of followup, each 1.33 unit association between trans fat there is general recognition that PHOs decrease in the total-C/HDL–C ratio was consumption and either LDL–C or total- may be safely used in human food. associated with a 38 percent decrease in C levels. Also, the comments stated that However, if there are data and risk of CHD death (Ref. 22). Each 0.33 the authors do not ‘‘force’’ the information that demonstrates to a mmol/L decrease in HDL–C was regression line through zero unlike in reasonable certainty that no harm will associated with a 61percent higher risk the Ascherio et al. 1999 paper, relied result from a specific use of a PHO in of CHD death. The authors concluded: upon by FDA in the tentative food, that information could be ‘‘HDL cholesterol added greatly to the determination. submitted as part of a food additive predictive ability of total cholesterol.’’ (Response) FDA disagrees. We note petition to FDA seeking issuance of a They stated: ‘‘Higher HDL cholesterol that the authors of this paper stated that regulation to prescribe conditions under and lower non-HDL cholesterol levels their regression analysis of TFA intake which the additive may be safely used were approximately independently and LDL–C ‘‘supports the IOM’s in food. associated with lower IHD [CHD] conclusion that any intake level of trans (Comment 31) Some comments stated mortality, so the ratio of total/HDL fat above 0 percent of energy increased that FDA made conclusions that any cholesterol was substantially more LDL cholesterol concentration.’’ This incremental increase in trans fat intake informative about IHD mortality than paper did not identify a threshold level increases the risk of CHD based on either, and was more than twice as at which LDL–C began to increase. The endpoints that are not considered informative as total cholesterol’’ (Ref. analysis in the paper was limited to validated surrogate biomarkers for CHD, 22). validated surrogate endpoint biomarkers such as LDL–C/HDL–C ratio in the (Comment 32) One comment stated of CHD, total cholesterol and LDL–C, Ascherio et al. 1999 paper (Ref. 31). that safety evaluation of macronutrients,

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00059 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34665

such as PHOs, is very complex and recognize the progressive linear an estimated 1.5 milligrams per deciliter requires a far more robust assessment of relationship between LDL–C (increase) (mg/dL) and 2.0 mg/dL, respectively the totality of technical and scientific and HDL–C (decrease) and trans fat (Ref. 58). The panel also concluded that evidence. The comment criticized FDA intake, and have concluded that trans replacement of TFA with saturated fatty for relying on ‘‘an isolated physiological fats are not essential nutrients in the acids (SFA), MUFA, or PUFA increases endpoint such as serum lipoproteins’’ as diet and consumption should be kept as HDL–C by an estimated 0.5, 0.4 and 0.5 predictive of CHD, and states that this low as possible. We have compiled a mg/dL, respectively. This panel’s methodology is not appropriate for a detailed summary of the expert panel conclusions were not limited to a GRAS assessment. reports in a review memorandum (Ref. specific TFA dose range and did not (Response) FDA disagrees; the results 18). indicate any threshold TFA intake. The of feeding trials showing changes in (Comment 33) Some comments stated conclusions were based on previously LDL–C, a validated surrogate endpoint that FDA should convene an expert published linear regression analyses biomarker for CHD, and other risk factor panel to specifically address whether (Refs. 26 and 33). biomarkers, are supported by the results evidence exists to indicate the effect of We also disagree that, based on of observational studies showing actual TFA on LDL–C is linear at low intakes generally available information, there is CHD disease outcomes (heart attacks (below 3% energy). Other comments a consensus among qualified experts and deaths) associated with TFA intake stated that there is consensus among that trans fats are safe at some level, and in large populations. The consistency of qualified experts that TFA intake should we note that recommendations from the evidence from two different study be less than 1% of energy, and cited expert panels either: (1) Do not state a methodologies is strong support for the expert panel reviews as evidence. recommended level (Ref. 13); or (2) conclusion that trans fatty acid intake Similar comments stated that PHOs are recommend consideration of further has a progressive and linear effect that safe at current intake levels, and TFA reduction in IP–TFA intake, below increases the risk of CHD. Such health intake is already below levels current levels (Refs. 59, 60, 61, and 62). effects are appropriate for FDA to recommended by nutrition experts. Since 2002, many expert panels have consider when assessing the safety of (Response) We decline to convene considered the adverse effects food ingredients. another expert panel in light of the associated with trans fat consumption. substantial evidence available on the Table 1 provides a list of organizations 2. Expert Panel Reviews and adverse effects of consuming trans fat. that have published reports on trans fat Recommendations FDA notes that a 2013 National and indicates whether they have The November 2013 notice discussed Institutes of Health, National Heart, conducted an evidence review and/or expert panel conclusions and Lung, and Blood Institute (NIH/NHLBI) made formal intake recommendations recommendations, including the 2002/ expert panel conducted a systematic regarding trans fat consumption. The 2005 IOM reports. The conclusions and evidence review and concluded with conclusions and recommendations recommendations of this report have moderate confidence that, for every 1 made by these organizations further since been affirmed by a series of U.S. percent of energy from TFA replaced by demonstrate a lack of consensus and international expert panels. The mono- or polyunsaturated fatty acids regarding the safety of PHOs, as the recent expert panels have continued to (MUFA or PUFA), LDL–C decreases by primary dietary source of IP–TFA.

TABLE 1—LIST OF ORGANIZATIONS THAT HAVE PUBLISHED REPORTS ON TRANS FAT

Evidence Formal trans fat Organization Report title Year review and intake conclusions recommendation

IOM ...... Dietary Reference Intakes for Energy and 2002/2005 X X Macronutrients (Ref. 7). European Food Safety Authority, Scientific Opinion on the presence of trans fatty acids 2004 X ...... Panel on Dietetic Products, Nutrition and in foods and the effect on human health Allergies. of the consumption of trans fatty acids (Ref. 63). FDA Food Advisory Committee, Nutrition Subcommittee Meeting, Summary Minutes 2004 X ...... Subcommittee. (Ref. 14). Dietary Guidelines Advisory Committee Report of the 2005 DGAC (Ref. 64) ...... 2005 X ...... (DGAC). U.S. Dept. of Health and Human Services, Dietary Guidelines for Americans (Ref. 12) 2005 ...... X U.S. Dept. of Agriculture (DHHS/USDA). World Health Organization (WHO) ...... Scientific Update on Trans Fatty Acids (Ref. 2009 X X 60). Food and Agriculture Organization, World Background Papers for Expert Consultation 2009 X ...... Health Organization (FAO, WHO). on Fats and Fatty Acids in Human Nutri- tion (Ref. 59). FAO, WHO...... Expert Consultation on Fats and Fatty 2010 X X Acids in Human Nutrition (Ref. 61). DGAC ...... Report of the 2010 DGAC (Ref. 65) ...... 2010 X ...... DHHS/USDA ...... Dietary Guidelines for Americans (Ref. 13) 2010 ...... X NHLBI ...... Evidence Report on Lifestyles Interventions 2013 X ...... to Reduce Cardiovascular Risk (Ref. 58). American College of Cardiology, American Guideline on Lifestyle Management to Re- 2013/2014 ...... X Heart Association. duce Cardiovascular Risk (Ref. 62).

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00060 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34666 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

3. Safety of Replacements for IP–TFA in 49). These analyses, as well as FDA percentage of trans fat in finished foods PHOs estimates discussed previously in or oils, or set a threshold in foods for the (Comment 34) Several comments section IV, demonstrate that maximum grams (g) of trans fat per questioned whether further reductions replacement of TFA with other serving. Some comments suggested in TFA intake will be clinically macronutrients is expected to result in various specification levels ranging from significant and subsequently affect decreased CHD risk. 0.2 to 0.5 g trans fat per serving or as We also recognize that replacement of public health. a percentage of total fat in foods or oils. (Response) Since publication of the PHOs will result in fatty acids from Another comment urged FDA to other fats and oils replacing not only IP– November 2013 notice, we have establish a reasonable level for trans fat TFA but also the other fatty acids in the quantitatively analyzed the public in food to specifically account for minor PHOs, but disagree that the safety health significance of removing PHOs uses of PHOs as processing aids. implications of these changes have not Some comments urged us to declare from the food supply (Ref. 25), and the been considered. One recent study that certain uses of PHOs in foods are results show that removing PHOs from estimated the change in CHD risk from GRAS, or to issue interim food additive human food would have an expected changes in blood lipids due to replacing regulations for specific low level uses. positive impact on public health. We soybean oil PHOs with application Examples of such uses provided by note that further reductions in IP–TFA specific oils (Ref. 28). Results showed comments included emulsifiers, intake below current levels may result that each of the TFA replacement encapsulates for flavor agents and color in small reductions in LDL–C and small strategies modeled changed the fatty additives, pan release agents, anti- improvements in other biomarkers that acid intake profile in a manner caking agents, gum bases, and use in may not seem clinically significant for predicted to decrease CHD risk, with frostings, fillings, and coatings. The use an individual; however, when differences in the projected decreased of PHOs in chewing gum was considered across the U.S. population, risk due to different replacement oils. specifically noted in some comments as small reductions in CHD risk would be Another recent study estimated the deserving special consideration due to expected to prevent large numbers of effect of the replacement of three the claim that there is no meaningful heart attacks and deaths, as illustrated example PHOs with seven replacement PHO intake from this use. Several in FDA estimates (Ref. 25). Moreover, fats and oils, based on changes in blood comments suggested we issue interim the 2013 Guideline on Lifestyle to lipids and non-lipids and other risk food additive regulations that would Reduce Cardiovascular Risk from the factor biomarkers from controlled allow certain uses of PHOs in food, American College of Cardiology and the feeding trials and on changes in CHD pending completion of studies American Heart Association (Ref. 62) risk from prospective observational evaluating the health effects of low level strongly recommends that clinicians studies (Ref. 26). Results showed that consumption of trans fat that reflect advise adults who would benefit from replacement of PHOs with other fats and current intake levels. Furthermore, one LDL–C reduction to reduce their oils would substantially lower CHD risk comment advised that if we decide to percentage of calories from trans fat (the (Ref. 26). Both studies estimated a treat certain low-level uses of PHOs as report notes that the majority of U.S. greater reduction in CHD risk with food additives, then the GRAS status for adults have one or more risk factors replacement of PHOs with vegetable oils these uses should not be revoked until involving abnormal lipids, high blood containing higher amounts of cis- a food additive approval is issued. pressure or pre-high blood pressure; unsaturated fatty acids than with those In contrast, we also received 33.5 percent of adults have elevated high in saturated fat (Refs. 26 and 28). numerous comments opposed to LDL–C). Therefore, further reduction in FDA also notes that replacement of establishing limits of trans fat in foods. IP–TFA intake below current levels is PHOs containing IP–TFA with other fats Most of these comments noted that expected to be clinically significant and and oils over the past decade has not scientific evidence has shown that no to prevent a large number of heart been accompanied by extensive amount of trans fat in food is safe and attacks and deaths in the United States. increases in saturated fat (Ref. 29), therefore, supported our tentative (Comment 35) Some comments stated which could have diminished the determination. One comment noted that that the safety implications of replacing impact of removing trans fat. trans fat threshold limits in food would TFA with other nutrients (e.g., saturated The safety implications of replacing be too difficult to monitor and enforce, fat, unsaturated fat, carbohydrates) have IP–TFAs in PHOs with other and therefore, should not be established. yet to be determined. macronutrients and replacing PHOs (Response) Regarding the proposals (Response) We recognize that containing IP–TFAs with other fats and for alternate approaches suggesting a removing PHOs from the food supply oils have been addressed in published threshold for trans fat in food or oils or will result in replacing the IP–TFA from studies (Refs. 18, 26, 28, 30, 31, 32, 33, suggesting that FDA declare some uses PHOs with other macronutrients, most 44, 45, and 49) and are also addressed of PHOs as GRAS, no comments likely other fatty acids, but disagree that in our quantitative estimate of decrease provided evidence that any uses of the safety implications of these changes in CHD risk with replacement of IP– PHOs meet the GRAS standard, or have not been considered. The adverse TFA, summarized previously in section evidence that would establish a safe effect of TFA on LDL–C and other blood IV.B (Ref. 25). threshold exposure level. Further, lipids and non-lipids when replacing although the intake from such minor other macronutrients (such as 4. Alternative Approaches and Evidence uses may be low, adequate data (e.g., carbohydrate, saturated fat and cis- for Safety specific conditions of use, use level, unsaturated fat) was extensively In the tentative determination, we trans fat content of the PHOs used) were demonstrated in controlled feeding requested data to support other possible not provided so that intake from these trials and summarized in regression approaches to address the use of PHOs uses could be estimated. Therefore we analyses (Refs. 18, 26, 30, 31, 32, 33, 44, in food, such as setting a specification are not setting a threshold for trans fat. and 45). In prospective observational for trans fat levels in food (78 FR 67169 If industry or other interested studies, reduction in CHD risk was also at 67174). individuals believe that safe conditions associated with replacement of TFA (Comment 36) Several comments of use for PHOs can be demonstrated, it with other macronutrients (Refs. 18 and proposed that we should limit the or they may submit a food additive

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00061 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34667

petition or food contact notification to remove PHOs from the food supply, (‘‘Kummerow citizen petition,’’ which FDA for review. given our conclusion on the GRAS can be found at Docket No. FDA–2009– Interim food additive regulations are status of PHOs. FDA has determined P–0382) requesting that we ban partially appropriate only when there is a that PHOs are not GRAS for any use in hydrogenated fat from the American reasonable certainty that a substance is human food. FDA agrees, however, that diet. The Kummerow citizen petition not harmful. See 21 CFR 180.1(a). As we should work with the food industry cited studies linking intake of IP–TFA to discussed throughout this section, the to review new regulatory submissions or the prevalence of CHD in the United available scientific evidence raises data as new technologies and/or States. The Kummerow citizen petition substantial concerns about the safety of ingredients are developed that may also asserted that trans fat may be PHOs. Based on the currently available serve as alternatives to PHOs, and we passed to infants via breast milk and data and information, FDA cannot will continue to do so. that the daily intake of trans fat related conclude that there is a reasonable to the health of children has been V. Citizen Petitions certainty that PHOs are not harmful, nor ignored since children do not exhibit did any comments provide information As discussed in the tentative overt heart disease (Id. at p. 6). The that would allow FDA to establish determination (78 FR 67169 at 67173), Kummerow citizen petition further conditions of safe use at this time. we received two citizen petitions stated that inflammation in the arteries Therefore, an interim food additive regarding the safety of PHOs. In 2004, is believed to be a risk factor in CHD regulation would not be appropriate. the Center for Science in the Public and studies have shown that trans fatty (Comment 37) Several comments Interest (CSPI) submitted a citizen acids elicit an inflammatory response suggested various changes to our petition (‘‘CSPI citizen petition’’ which (Id.). labeling regulations to encourage can be found under Docket No. FDA– This order constitutes a response, in industry to reformulate products to 2004–P–0279) requesting that we revoke part, to the citizen petitions. As contain less trans fat and help the GRAS status of PHOs, and discussed above in section III.C consumers reduce trans fat intake. In consequently declare that PHOs are food (response to Comment 10), we plan to addition, one comment stated that a 0 g additives. The petition also asked us to amend the regulations regarding LEAR trans fat declaration should not be revoke the safe conditions of use for and menhaden PHOs in a future action, allowed on a label if a PHO is in the partially hydrogenated products that are and we will consider taking future ingredient list. Some comments currently considered food additives,3 to action regarding related regulations. As indicated that a statement prohibit the use of partially discussed in section III.B, we intend to recommending that consumers limit hydrogenated vegetable oils that are address any claims of prior sanction for their intake of trans fat should be added prior sanctioned, and to initiate a specific uses of PHO in a future action. to the Nutrition Facts Panel. A few program to encourage manufacturers comments suggested we set a Daily and restaurants to switch to more VI. Environmental Impact Value for trans fat and consider healthy oils (CSPI citizen petition at pp. We have carefully considered the establishing disclosure or disqualifying 3 through 5, 29 through 30). The CSPI potential environmental effects of this levels of trans fat for nutrient content citizen petition excluded trans fat that action. We have determined, under 21 and health claims. Many comments occurs naturally in meat from ruminant CFR 25.32(m), that this action ‘‘is of a noted that the risk of developing CHD animals and dairy fats, and that forms type that does not individually or is dependent on many factors, and during the production of non- cumulatively have a significant effect on therefore, the association between hydrogenated oils (Id. at pp. 2 through the human environment’’ such that intake of macronutrients, such as PHOs, 3). It also did not include FHOs, which neither an environmental assessment and adverse health outcomes is best contain negligible amounts of trans fat, nor an environmental impact statement addressed through nutrition labeling and PHOs that may be produced by new is required. and consumer education. technologies that result in negligible FDA received some comments on the (Response) FDA disagrees that amounts of trans fat in the final product tentative determination relating to labeling is the best approach to address (Id. at p. 3). The CSPI citizen petition potential environmental impacts of the use of PHOs because FDA has stated that trans fat promotes CHD by removing PHOs from the human food determined that PHOs are not GRAS for increasing LDL–C and also by lowering supply. We considered these comments any use in human food and therefore are HDL–C, and therefore has greater in determining whether extraordinary food additives subject to the adverse effects on serum lipids (and circumstances existed under 21 CFR requirement of premarket approval possibly CHD) than saturated fats (Id., at 25.21. Our discussion is contained in a under section 409 of the FD&C Act. pp. 15 through 18). The CSPI citizen review memorandum (Ref. 66). Although we recognize that the petition also stated that, beyond its VII. Economic Analysis requirement to label trans fat content adverse effects on serum lipids, trans fat led to significant reduction in trans fat may promote heart disease in additional This notice is not a rulemaking. It is levels in products, further changes to ways. Based on these findings, CSPI a declaratory order under 5 U.S.C. labeling are outside the scope of this asserted that PHOs can no longer be 554(e) to terminate a controversy or determination, which relates to considered GRAS. remove uncertainty. We have prepared ingredient safety. In 2009, Dr. Fred Kummerow a memorandum updating our previous (Comment 38) Some comments submitted a citizen petition estimate published in the November suggested that we should work with 2013 notice, using information available industry to encourage voluntary 3 The petition from CSPI provided, as an example, to us as well as information we received partially hydrogenated methyl ester of rosin, which during the comment period. We reductions in PHO use and to foster the is approved as a food additive for use as a synthetic development of innovative flavoring substance (32 FR 7946, June 2, 1967; 21 estimated the 20-year costs and benefits hydrogenation technologies that CFR 172.515) and as a masticatory substance in of removing PHOs from the U.S. human produce PHOs containing low levels of chewing gum base (29 FR 13894, October 8, 1964; food supply, an outcome that could 21 CFR 172.615). Partially hydrogenated methyl result from this order (Ref. 17). We trans fat. ester of rosin is not a PHO as discussed in section (Response) FDA disagrees that a II; accordingly, this this substance is outside the estimated the costs of all significant voluntary program is the best way to scope of this order. effects of the removal, including

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00062 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34668 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

packaged food reformulation and the peer-reviewed literature, and added We estimate the net present value of 20 relabeling, increased costs for substitute this to expected medical expenditure years of benefits to be $140 billion, with ingredients, and consumer, restaurant, savings to determine the expected a 90 percent confidence interval of $11 and bakery recipe changes. We benefits of this order. billion to $440 billion. Expected NPV of monetized the expected health gains We estimate the net present value 20 years of net benefits (benefits from the removal of PHOs from the food (NPV) (over 20 years; Table 2) of reduced by quantified costs) are $130 supply using information presented in quantified costs of this action to be $6.2 billion, with a 90 percent confidence FDA’s safety assessment (Ref. 17) and billion, with a 90 percent confidence interval of $5 billion to $430 billion. interval of $2.8 billion to $11 billion.

TABLE 2—COSTS AND BENEFITS OF PHO REMOVAL, USD BILLIONS

Low High 20-Year net present value of Estimate Mean Estimate

Costs * ...... $2.8 $6.2 $11 Benefits ...... 11 140 440 Net Benefits * ...... 5 130 430 * This does not include some unquantified costs, see the economic estimate memo (Ref. 17) for discussion.

VIII. Compliance Date and Related Several comments expressed concern the rest of industry. Another comment Comments With FDA Responses about adequate availability of stated that small businesses would need We received numerous comments alternative oils, especially palm oil. One at least 5 years due to their limitations about the time needed to reformulate comment stated that the food industry in research and development expertise, products to remove PHOs should FDA would prefer to replace PHOs with inability to command supply of scarce make a final determination that PHOs domestically produced vegetable oils ingredients, and economic pressures of are not GRAS. We also received (e.g., high-oleic soybean oil) rather than labeling changes. A related comment comments about challenges to palm oil, but time is needed to requested that FDA take into reformulation, specific product types commercialize these options. Some consideration the magnitude of private that will be difficult to reformulate, and comments stated that sudden demand label products impacted. Other effects on small businesses. for palm oil would pose challenges for comments stated that small businesses (Comment 39) The comments obtaining sustainably-sourced palm oil, should not be given special recommended compliance dates ranging as the current market would likely not consideration or longer times for from immediate to over 10 years. be able to meet the demand. implementation. Several comments stated that fried foods Other comments indicated that the (Response) Based on our experience should have less time (i.e., 6 months) to time needed for removal of PHOs is and on the changes we have already phase out the use of PHOs. One dependent on the product category. A seen in the market, we believe that 3 comment stated that if the use of low number of comments indicated that the years is sufficient time for submission levels of PHOs were to remain baking industry will have difficulty and review and, if applicable permissible by virtue of being GRAS or replacing the solid shortenings used in requirements are met, approval of food through food additive approval, then the bakery products. Other comments additive petitions for uses of PHOs for estimated time to reformulate would be indicated difficulties in the categories of which industry or other interested 5 years; however, if FDA does not cakes and frostings, fillings for candies, individuals believe that safe conditions authorize low level uses of PHOs, the chewing gum, snack bars, and as a of use may be prescribed. For this timeline would need to be 10 years. In component of what the comments reason, we are establishing a general, the food industry urged FDA to termed minor use ingredients, such as compliance date for this order of June provide sufficient time for all for use in coatings, anti-caking agents, 18, 2018. We recognize that the use of companies to secure a supply of encapsulates, emulsifiers, release PHOs in the food supply is already alternatives and transition to new agents, flavors, and colors. declining and expect this to continue formulations. Some comments stated Several comments indicated that even prior to the compliance date. that FDA should coordinate the other challenges to PHO removal Regarding the use of ‘‘low levels’’ of compliance date with updates to the include the need for new transportation PHOs, no comments provided a basis Nutrition Facts Panel. infrastructure (e.g., terminals, rail cars, upon which we can currently conclude Some comments stated that barges, and storage facilities), packaging that any use of PHO is GRAS (discussed domestically grown oilseed crops must changes, and disruption of international in section IV). We recognize the be planted about 18 months prior to trade. challenges faced by small businesses, their expected usage in order for the A number of comments noted however, considering our determination crop to be grown, harvested, stored, challenges faced by small businesses, that PHOs are not GRAS for any use in crushed, oil extracted, processed, such as access to alternative oils, human food, we conclude that refined, delivered, and used in foods. inability to compete for supply, fewer providing 3 years for submission and One comment stated that the oil resources to commit to research and review of food additive petitions and/or industry will need a minimum of 3 development, and effect of ingredient food contact notifications is reasonable, years to fully commercialize the various costs on growth of the business. Some and will have the additional benefit of oils capable of replacing PHOs in food. comments noted that small businesses allowing small businesses time to A number of comments stated that it represent a relatively small contribution address these challenges. We could take several additional years to to overall IP–TFA intake. One comment understand the difficulties faced by reformulate after the development of the recommended that we allow small small businesses due to limited research new oils. businesses an additional 2 years beyond and development resources and

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00063 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices 34669

potential challenges to gain timely through Friday, and are available 2015. access to suitable alternatives. electronically at http:// 17. Memorandum from R. Bruns to M. The compliance date will have the www.regulations.gov. (FDA has verified Honigfort, June 11, 2015. additional benefit of minimizing market the Web site addresses in this reference 18. Memorandum from J. Park to M. disruptions by providing industry Honigfort, Scientific Update on section, but we are not responsible for Experimental and Observational Studies sufficient time to identify suitable any subsequent changes to the Web sites of Trans Fat Intake and Coronary Heart replacement ingredients for PHOs, to after this document publishes in the Disease Risk, June 11, 2015. exhaust existing product inventories, (Federal Register.) 19. IOM, ‘‘Evaluation of Biomarkers and and to reformulate and modify labeling Surrogate Endpoints in Chronic 1. Tarrago-Trani, M., K. M. Philips, L. E. Disease’’, Washington, DC: National of affected products. Three years also Lemar, et al.,‘‘New and Existing Oils and Academies Press, 2010. provides time for the growing, Fats Used in Products With Reduced 20. Rasnake, C. M., P. R. Trumbo, and T. M. harvesting, and processing of new Trans-Fatty Acid Content,’’ Journal of Heinonen, ‘‘Surrogate Endpoints and varieties of edible oilseeds to meet the the American Dietetic Association, Emerging Surrogate Endpoints for Risk 106:867–877, 2006. expected demands for alternative oil Reduction of Cardiovascular Disease,’’ 2. Kodali, D. R. and G. R. List, Eds., Trans products and to address the supply Nutrition Reviews, 66:76–81, 2008. Fats Alternatives, AOCS Press, chain issues associated with transition 21. Baigent C., A. Keech, P. M. Kearney, et Champaign, IL, p. 34–35, 2005. to new oils. al., ‘‘Efficacy and Safety of Cholesterol- 3. USDA National Nutrition Database for (Comment 40) Several comments lowering Treatment: Prospective Meta- Standard Reference, Release 23, 2010; analysis of Data from 90,056 Participants stated that how FDA defines PHOs and http://www.ars.usda.gov/Services/ FHOs will affect reformulation efforts docs.htm?docid=8964. in 14 Randomised Trials of Statins,’’ and the time needed to reformulate. 4. Kodali, D. R. and G. R. List, Eds., Trans Lancet, 366:1267–1278, 2005. These comments suggested it was Fats Alternatives, AOCS Press, 22. Lewington S., G. Whitlock, R. Clarke, et unclear from the tentative determination Champaign, IL, pp. 4, 2005. al., ‘‘Blood Cholesterol and Vascular 5. Memorandum from J. Park to M. Honigfort, Mortality by Age, Sex, and Blood whether FHOs would be subject to this Pressure: A Meta-analysis of Individual final determination. August 10, 2005. 6. Memorandum from J. Park to M. Honigfort, Data from 61 Prospective Studies with (Response) As discussed in section II, 55,000 Vascular Deaths,’’ Lancet, we have defined PHOs, the subjects of August 19, 2010. 7. IOM/NAS, ‘‘Dietary Reference Intakes for 370:1829–1839, 2007. this order, as fats and oils that have Energy Carbohydrate, Fat, Fatty Acids, 23. Memorandum from J. Park to M. been hydrogenated, but not to complete Cholesterol, and Amino Acids Honigfort, Literature Review, June 11, or near complete saturation, and with an (Macronutrients),’’ chapters 8 and 11, 2015. IV greater than 4 as determined by an National Academies Press, Washington 24. Kiage J. N., P. D. Merrill, S. E. Judd, et appropriate method. We have also DC, 2002/2005; http://www.nap.edu. al., ‘‘Intake of Trans Fat and Incidence of defined FHOs as those fats and oils that 8. American Heart Association, http:// Stroke in the Reasons for Geographic www.heart.org/HEARTORG/ And Racial Differences in Stroke have been hydrogenated to complete or (REGARDS) Cohort,’’ American Journal near complete saturation, and with an GettingHealthy/FatsAndOils/Fats101/ Trans-Fats_UCM_301120_Article.jsp. of Clinical Nutrition, 99:1071–1076, IV of 4 or less, as determined by an 9. Eckel, R.H., S. Borra, A.H. Lichtenstein, et 2014. appropriate method. Thus, FHOs are al., ‘‘Understanding the Complexity of 25. Memorandum from J. Park to M. outside the scope of this order and there Trans Fatty Acid Reduction in the Honigfort, Quantitative Estimate of is no need to allow additional time for American Diet,’’ Circulation, 115:2220– Industrial Trans Fat Intake and Coronary reformulation of products containing 2235, 2007. Heart Disease Risk, June 11, 2015. FHO. 10. Kris-Etherton, P. M., S. Innis, ‘‘Position 26. Mozaffarian, D. and R. Clarke, of the American Dietetic Association and ‘‘Quantitative Effects on Cardiovascular IX. Conclusion and Order Dietitians of Canada: Dietary Fatty Risk Factors and Coronary Heart Disease As discussed in this document, for a Acids,’’ Journal of the American Dietetic Risk of Replacing Partially Hydrogenated Vegetable Oils With Other Fats and substance to be GRAS, there must be Association, pp. 1599–1611, 2007. 11. WHO, ‘‘Diet, Nutrition, and the Oils,’’ European Journal of Clinical consensus among qualified experts Prevention of Chronic Disease,’’ Nutrition, 63:S22–S33, 2009. based on generally available information Technical Series Report 916, pp. 81–85, 27. Go A. S., Mozaffarian, D., Roger, V. L., et that the substance is safe under the Geneva, 2003. al., on behalf of the American Heart intended conditions of use. In 12. USDA and Department of Health and Association Statistics Committee and accordance with the process set forth in Human Services (HHS), Dietary Stroke Statistics Subcommittee, FDA’s regulations in § 170.38, FDA has Guidelines for Americans, 2005, 6th ed., ‘‘Executive Summary: Heart Disease and determined that there is no longer a pp. 29–34, Washington, DC: U.S. Stroke Statistics—2014 Update: A Report consensus that PHOs, the primary Government Printing Office, January from the American Heart Association,’’ 2005. Circulation, 129:399–410, 2014. source of industrially-produced trans 13. USDA and HHS, Dietary Guidelines for 28. Lefevre, M., R. P. Mensink, P. M. Kris- fat, are generally recognized as safe for Americans, 2010, 7th ed., pp. 24–27, Etherton, et al., ‘‘Predicted Changes in use in human food, based on current Washington, DC: U.S. Government Fatty Acid Intakes, Plasma Lipids, and scientific evidence discussed in section Printing Office, December 2010. Cardiovascular Disease Risk Following IV.B regarding the health risks 14. HHS/FDA/Center for Food Safety and Replacement of Trans Fatty Acid- associated with consumption of trans Applied Nutrition Food Advisory Containing Soybean Oil with fat. FDA considers this order a partial Committee, Nutrition Subcommittee Application-Appropriate Alternatives,’’ response to the citizen petitions from Meeting, Total Fat and Trans Fat, April Lipds, 47:951–962, 2012. CSPI and Dr. Kummerow. 27–28, 2004. 29. Mozaffarian, D., M. F. Jacobson, J. S. 15. Dietz, W. H. and K. S. Scanlon, Greenstein, ‘‘Food Reformulations to X. References ‘‘Eliminating the Use of Partially Reduce Trans Fatty Acids’’ [Letter to the Hydrogenated Oil in Food Production editor], New England Journal of The following references have been and Preparation,’’ Journal of the Medicine, 362:2037–2039, 2010. placed on display in the Division of American Medical Association, 108:143– 30. Brouwer, I. A., A. J. Wanders, M. B. Dockets Management (see ADDRESSES) 144, 2012. Katan, ‘‘Effect of Animal and Industrial and may be seen by interested persons 16. Memorandum from D. Doell, D. Folmer, Trans Fatty Acids on HDL and LDL between 9 a.m. and 4 p.m., Monday and H. Lee to M. Honigfort, June 11, Cholesterol Levels in Humans—A

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00064 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES 34670 Federal Register / Vol. 80, No. 116 / Wednesday, June 17, 2015 / Notices

Quantitative Review,’’ PLoS One, Heart Disease in Women: 20 Years of Nutrition & Metabolism, 55:5–7, 2009. 5(3):e9434, 2010. Follow-up of the Nurses’ Health Study,’’ 60. Uauy, R., A. Aro, R. Clarke, et al., ‘‘WHO 31. Ascherio, A., M. B. Katan, P. L. Zock, et American Journal of Epidemiology, Scientific Update on Trans Fatty Acids: al., ‘‘Trans Fatty Acids and Coronary 161:672–679, 2005. Summary and Conclusions,’’ European Heart Disease,’’ New England Journal of 47. Ascherio A., E. B. Rimm, E. L. Journal of Clinical Nutrition, 63: S68– Medicine, 340:1994–1998, 1999. Giovannucci, et al., ‘‘Dietary Fat and S75, 2009. 32. Zock, P. L., M. B. Katan, and R. P. Risk of Coronary Heart Disease in Men: 61. Food and Agricultural Organization of the Mensink, ‘‘Dietary Trans Fatty Acids and Cohort Follow Up Study in the United United Nations (FAO) and WHO, Fats Lipoprotein Cholesterol’’ [Letter to the States,’’ BMJ, 313:84–90, 1996. and Fatty Acids in Human Nutrition. editor], American Journal of Clinical 48. Willett W. C., M. J. Stampfer, J. E. Report of an Expert Consultation. Rome: Nutrition, 61:617, 1995. Manson, et al., ‘‘Intake of Trans Fatty FAO; 2010. 33. Mensink, R. P., P. L. Zock, A. D. Kester, Acids and Risk of Coronary Heart 62. Eckel, R. H., J. M. Jakicic, J. D. Ard, et et al., ‘‘Effects of Dietary Fatty Acids and Disease Among Women,’’ Lancet, al. ‘‘2013 AHA/ACC Guideline on Carbohydrates on the Ratio of Serum 341:581–585, 1993. Lifestyle Management to Reduce Total to HDL Cholesterol and on Serum 49. Hu, F. B., M. J. Stampfer, J. E. Manson, Cardiovascular Risk: A Report of the Lipids and Apolipoproteins: A Meta- et al., ‘‘Dietary Fat Intake and the Risk American College of Cardiology/ Analysis of 60 Controlled Trials,’’ of Coronary Heart Disease in Women,’’ American Heart Association Task Force American Journal of Clinical Nutrition, New England Journal of Medicine, on Practice Guidelines,’’ Circulation, 77:1146–1155, 2003. 337:1491–1499, 1997. 129:S76–S99, 2014. 34. Mozaffarian, D., M. B., Katan, A. Asherio, 50. Pietinen, P., A. Ascherio, P. Korhonen, et 63. European Food Safety Authority (EFSA), et al., ‘‘Trans Fatty Acids and al., ‘‘Intake of Fatty Acids and Risk of ‘‘Opinion of the Scientific Panel on Cardiovascular Disease,’’ New England Coronary Heart Disease in a Cohort of Dietetic Products, Nutrition and Journal of Medicine, 354:1601–1613, Finnish Men. The Alpha-Tocopherol, Allergies on a Request from the 2006. Beta-Carotene Cancer Prevention Study,’’ Commission Related to the Presence of 35. Dourson, M. ‘‘Mode of Action and Dose- American Journal of Epidemiology, Trans Fatty Acids in Foods and the Response Evaluation of the Effect of 145:876–887, 1997. Effect on Human Health of the Partially Hydrogenated Oils on LDL- 51. Oomen, C. M., M. C. Ocke, E. J. Feskens, Consumption of Trans Fatty Acids,’’ Cholesterol,’’ Presented at the SOT FDA et al., ‘‘Association Between Trans Fatty EFSA Journal, 81:1–49, 2004. Colloquia on Emerging Toxicological Acid Intake and 10-year Risk of Coronary 64. Dietary Guidelines Advisory Committee, Science Challenges in Food and Heart Disease in the Zutphen Elderly Report of the Dietary Guidelines Ingredient Safety, November 7, 2014. Study: A Prospective Population-Based Advisory Committee on Dietary 36. National Research Council, Science and Study,’’ Lancet, 357:746–751, 2001. Guidelines for Americans, 2005, Decisions: Advancing Risk Assessment, 52. Mozaffarian D., A. Aro, W. C. Willett, Washington, DC: HHS, 2005; http:// National Academies Press, Washington, ‘‘Health Effects of Trans-fatty Acids: www.health.gov/dietaryguidelines/ DC, 2009; http://www.nap.edu). Experimental and Observational dga2005/report/default.htm. 37. Keys, A., J. T. Anderson, F. Grande, Evidence,’’ European Journal of Clinical 65. Dietary Guidelines Advisory Committee, ‘‘Serum Cholesterol Response to Changes Nutrition, 63:S5–21, 2009. Report of the Dietary Guidelines in the Diet: I. Iodine Value of Dietary Fat 53. Skeaff C. M. and J. Miller, ‘‘Dietary Fat Advisory Committee on Dietary Versus 2S–P,’’ Metabolism, 14:747–758, and Coronary Heart Disease: Summary of Guidelines for Americans, 2010, 1965. Evidence from Prospective Cohort and Washington, DC: USDA, Agricultural 38. Hegsted, D. M., R. B. McGandy, M. L. Randomised Controlled Trials,’’ Annals Research Service, 2010; http:// Myers, et al., ‘‘Quantitative Effects of of Nutrition & Metabolism, 55:173–201, origin.www.cnpp.usda.gov/DGAs2010– Dietary Fat on Serum Cholesterol in 2009. DGACReport.htm. Man,’’ American Journal of Clinical 54. Bendsen N. T., R. Christensen, E. M. 66. Memorandum from M. Pfeil to M. Nutrition, 17:281–295, 1965. Bartels, et al., ‘‘Consumption of Honigfort, June 11, 2015. 39. Keys, A. and R. W. Parlin, ‘‘Serum Industrial and Ruminant Trans Fatty Dated: June 12, 2015. Cholesterol Response to Changes in Acids and Risk of Coronary Heart Dietary Lipids,’’ American Journal of Disease: A Systematic Review and Meta- Leslie Kux, Clinical Nutrition, 19:175–181, 1966. analysis of Cohort Studies,’’ European Associate Commissioner for Policy. 40. Page, I. H., E. V. Allen, F. L. Chamberlain, Journal of Clinical Nutrition, 65:773– [FR Doc. 2015–14883 Filed 6–16–15; 8:45 am] et al., ‘‘Dietary Fat and Its Relation to 783, 2011. BILLING CODE 4164–01–P Heart Attacks and Strokes,’’ Circulation, 55. Chowdhury R., S. Warnakula, S. 23:133–136, 1961. Kunutsor, et al., ‘‘Association of Dietary, 41. Clarke, R., C. Frost, R. Collins, et al. Circulating, and Supplement Fatty Acids DEPARTMENT OF HEALTH AND ‘‘Dietary Lipids and Blood Cholesterol: with Coronary Risk. A Systematic Quantitative Meta-Analysis of Metabolic Review and Meta-analysis,’’ Annals of HUMAN SERVICES Ward Studies,’’ BMJ, 314:112–117, 1997. Internal Medicine, 160:398–406, 2014. 42. Mensink R. P. and M. B. Katan, ‘‘Effect 56. Trumbo, P. R. and T. Shimakawa, Food and Drug Administration of Dietary Fatty Acids on Serum Lipids ‘‘Tolerable Upper Intake Levels for Trans [Docket No. FDA–2012–N–0369] and Lipoproteins. A Meta-Analysis of 27 Fat, Saturated Fat, and Cholesterol,’’ Trials,’’ Arteriosclerosis, Thrombosis, Nutrition Reviews, 69:270–278, 2011. Agency Information Collection and Vascular Biology, 12:911–919, 1992. 57. Willett W. C., ‘‘Dietary Fats and Coronary Activities; Submission for Office of 43. Reeves, R. M., ‘‘Effect of Dietary Trans Heart Disease,’’ Journal of Internal Management and Budget Review; Fatty Acids on Cholesterol Levels’’ Medicine, 272:13–24, 2012. [Letter to the editor], New England 58. National Heart, Lung, and Blood Institute, Comment Request; Regulations Under Journal of Medicine, 324:338–339, 1991. ‘‘Lifestyle Interventions to Reduce the Federal Import Milk Act 44. Katan M. B., P. L. Zock, R. P. Mensink, Cardiovascular Risk: Systematic AGENCY: Food and Drug Administration, ‘‘Trans Fatty Acids and their Effects on Evidence Review From the Lifestyle HHS. Lipoproteins in Humans,’’ Annual Work Group,’’ Bethesda, MD: HHS, Review of Nutrition, 15:473–493, 1995. National Institutes of Health, 2013 ACTION: Notice. 45. Zock P. L. and R. P. Mensink, ‘‘Dietary (http://www.nhlbi.nih.gov/health-pro/ SUMMARY: Trans-Fatty Acids and Serum guidelines/in-develop/cardiovascular- The Food and Drug Lipoproteins in Humans,’’ Current risk-reduction/lifestyle). Administration (FDA) is announcing Opinion in Lipidology, 7:34–37, 1996. 59. Burlingame B., C. Nishida, R. Uauy, et al., that a proposed collection of 46. Oh, K., F. B. Hu, J. E. Manson, et al., ‘‘Fats and Fatty Acids in Human information has been submitted to the ‘‘Dietary Fat Intake and Risk of Coronary Nutrition: Introduction,’’ Annals of Office of Management and Budget

VerDate Sep<11>2014 18:47 Jun 16, 2015 Jkt 235001 PO 00000 Frm 00065 Fmt 4703 Sfmt 4703 E:\FR\FM\17JNN1.SGM 17JNN1 asabaliauskas on DSK5VPTVN1PROD with NOTICES