[Federal Register: October 6, 2005 (Volume 70, Number 193)]
[Proposed Rules]               
[Page 58569-58601]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr06oc05-29]                         


[[Page 58569]]

-----------------------------------------------------------------------

Part III





Department of Health and Human Services





-----------------------------------------------------------------------



Food and Drug Administration



-----------------------------------------------------------------------



21 CFR Part 589



Substances Prohibited From Use in Animal Food or Feed; Proposed Rule


[[Page 58570]]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Part 589

[Docket No. 2002N-0273] (formerly Docket No. 02N-0273)
RIN 0910-AF46

 
Substances Prohibited From Use in Animal Food or Feed

AGENCY:  Food and Drug Administration, HHS.

ACTION:  Proposed rule.

-----------------------------------------------------------------------

SUMMARY:  The Food and Drug Administration (FDA) is proposing to amend 
the agency's regulations to prohibit the use of certain cattle origin 
materials in the food or feed of all animals. These materials include 
the following: The brains and spinal cords from cattle 30 months of age 
and older, the brains and spinal cords from cattle of any age not 
inspected and passed for human consumption, the entire carcass of 
cattle not inspected and passed for human consumption if the brains and 
spinal cords have not been removed, tallow that is derived from the 
materials prohibited by this proposed rule that contains more than 0.15 
percent insoluble impurities, and mechanically separated beef that is 
derived from the materials prohibited by this proposed rule. These 
measures will further strengthen existing safeguards designed to help 
prevent the spread of bovine spongiform encephalopathy (BSE) in U.S. 
cattle.

DATES:  Submit written or electronic comments by December 20, 2005. 
Submit written comments on the information collection provisions by 
November 7, 2005.

ADDRESSES: You may submit comments, identified by [Docket No. 2002N-
0273 or RIN 0910-AF46], by any of the following methods:
Electronic Submissions
    Submit electronic comments in the following ways:
     Federal eRulemaking Portal: http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.regulations.gov. 

Follow the instructions for submitting comments.
     Agency Web site: http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/dockets/ecomments. 

Follow the instructions for submitting comments on the agency Web site.
Written Submissions
    Submit written submissions in the following ways:
     FAX: 301-827-6870.
     Mail/Hand delivery/Courier [For paper, disk, or CD-ROM 
submissions]: Division of Dockets Management (HFA-305), Food and Drug 
Administration, 5630 Fishers Lane, rm. 1061, Rockville, MD 20852.
    To ensure more timely processing of comments, FDA is no longer 
accepting comments submitted to the agency by e-mail. FDA encourages 
you to continue to submit electronic comments by using the Federal 
eRulemaking Portal or the agency Web site, as described in the 
Electronic Submissions portion of this paragraph.
    Instructions: All submissions received must include the agency name 
and Docket No(s). or Regulatory Information Number (RIN) for this 
rulemaking. All comments received may be posted without change to 
http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/ohrms/dockets/default.htm, including any personal 

information provided. For detailed instructions on submitting comments 
and additional information on the rulemaking process, see the 
``Comments'' heading of the SUPPLEMENTARY INFORMATION section of this 
document.
    Docket: For access to the docket to read background documents or 
comments received, go to http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/ohrms/dockets/default.htm 

and insert the docket number(s), found in brackets in the heading of 
this document, into the ``Search'' box and follow the prompts and/or go 
to the Division of Dockets Management, 5630 Fishers Lane, rm. 1061, 
Rockville, MD 20852.

FOR FURTHER INFORMATION CONTACT:  Burt Pritchett, Center for Veterinary 
Medicine (HFV-222), Food and Drug Administration, 7519 Standish Pl., 
Rockville, MD 20855, 240-453-6860, e-mail: burt.pritchett@fda.gov.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Background
    A. Bovine Spongiform Encephalopathy
    B. Current Animal Feed Safeguards in the United States
    C. Risk of BSE in North America
    D. Additional Measures Considered to Strengthen Animal Feed 
Safeguards
    1. Comments on November 6, 2002 Advance Notice of Proposed 
Rulemaking (ANPRM)
    2. Actions in Response to Washington State Case
    3. Comments on July 14, 2004 ANPRM
II. Proposed Measures to Strengthen Animal Feed Safeguards
    A. FDA Response to Comments to the 2004 ANPRM
    B. Additional Measures to Further Strengthen Feed Protection
    C. Basis for Proposing to Apply Additional Measures to All Animal 
Food and Feed
    D. Cattle Materials Proposed to be Prohibited From Use in All 
Animal Food and Feed
    E. Disposal of Cattle Materials Prohibited in Animal Feed
III. Description of Proposed Rule and Legal Authority
    A. Definitions
    B. Proposed Requirements
    C. Proposed Recordkeeping and Access Requirements
    D. Conforming Changes to 21 CFR 589.2000--Animal Proteins 
Prohibited in Ruminant Feed
    E. Legal Authority
IV. Analysis of Economic Impacts
    A. Summary of Proposed Regulatory Impact Analysis
    B. Need for Regulation
    C. Benefits
    D. Costs
    E. Government Costs
    F. Sensitivity Analysis
    G. Regulatory Flexibility Analysis
V. Paperwork Reduction Act
VI. Environmental Impact
VII. Federalism
VIII. Comments
IX. References

I. Background

A. Bovine Spongiform Encephalopathy

    BSE belongs to the family of diseases known as transmissible 
spongiform encephalopathies (TSEs). In addition to BSE, TSEs also 
include scrapie in sheep and goats, chronic wasting disease (CWD) in 
deer and elk, and Creutzfeldt-Jakob disease (CJD) in humans. The agent 
that causes BSE and other TSEs has yet to be fully characterized. The 
most widely accepted theory in the scientific community is that the 
agent is an abnormal form of a normal cellular prion protein. The 
abnormal form of the prion protein is less soluble and more resistant 
to heat degradation than the normal form. The abnormal prion does not 
evoke any demonstrated immune response or inflammatory reaction in host 
animals. BSE is diagnosed by postmortem microscopic examination of an 
animal's brain tissue and by detection of the abnormal form of the 
prion protein in an animal's brain tissue. There is currently no 
available test to detect the disease in a live animal.
    Since November 1986, there have been more than 180,000 confirmed 
cases of BSE in cattle worldwide. Over 95 percent of all BSE cases have 
occurred in the United Kingdom, where the epidemic peaked in 1992/1993, 
with approximately 1,000 new cases reported

[[Page 58571]]

per week. In addition to the United Kingdom, the disease has been 
confirmed in native-born cattle in 22 European countries and in some 
nonEuropean countries, including Japan, Israel, Canada, and the United 
States.
    Epidemiological studies have characterized the outbreak of BSE in 
the United Kingdom as a prolonged epidemic arising at various 
locations, with all occurrences due to a common source, and have 
suggested that feed contaminated by a TSE agent was the cause of the 
disease outbreak (Ref. 1). The subsequent spread of BSE was associated 
with the feeding of meat-and-bone-meal from rendered BSE-infected 
cattle to non-infected cattle (Ref. 1). It appears likely that the BSE 
agent was transmitted among cattle at an increasing rate by ruminant-
to-ruminant feeding until the United Kingdom ban on such practices went 
into effect in 1988 (Ref. 2).
    Agricultural officials in the United Kingdom have taken a series of 
actions to eliminate BSE. These actions include making BSE a reportable 
disease, banning mammalian meat-and-bone meal in feed for all food-
producing animals, prohibiting the inclusion of animals more than 30 
months of age in the animal and human food chains, and destroying all 
animals showing signs of BSE. As a result of these actions, most 
notably the feed bans, the rate of newly reported cases of BSE in the 
United Kingdom has decreased sharply and continues on a downward trend.
    In 1996, a newly recognized form of the human disease CJD, referred 
to as variant CJD (vCJD), was reported in the United Kingdom. 
Scientific and epidemiological studies have linked vCJD to exposure to 
the BSE agent, most likely through human consumption of beef products 
contaminated with the agent. To date, approximately 150 probable and 
confirmed cases of vCJD have been reported in the United Kingdom, where 
there had likely been a high level of contamination of beef products. 
It is believed that in the United States, where measures to prevent the 
introduction and spread of BSE have been in place for some time, there 
is far less potential for human exposure to the BSE agent. The Centers 
for Disease Control and Prevention (CDC) leads a surveillance system 
for vCJD in the United States. To date, CDC, has not detected vCJD in 
any resident of the United States that had not lived in or traveled to 
the United Kingdom for extended periods of time. In 2002, a probable 
case of vCJD was reported in a Florida resident who had lived in the 
United Kingdom during the BSE epidemic. Epidemiological data indicate 
that the patient likely was exposed to the BSE agent before moving to 
the United States.

B. Current Animal Feed Safeguards in the United States

    In the Federal Register of June 5, 1997 (62 FR 30936) (the 1997 
ruminant feed final rule), FDA published a final rule to provide that 
animal protein derived from mammalian tissues is prohibited for use in 
ruminant feed. Although BSE had not been identified in the United 
States at that time, the 1997 ruminant feed final rule was put in place 
to prevent the establishment and amplification of BSE in the United 
States through animal feed and thereby minimize risk to humans and 
animals. The 1997 ruminant feed final rule created a new Sec.  589.2000 
(21 CFR 589.2000), Animal proteins prohibited in ruminant feed, and 
established a system of controls to ensure that ruminant feed did not 
contain animal protein derived from mammalian tissues. The 1997 
ruminant feed final rule set out requirements for persons who 
manufacture, process, blend, or distribute certain animal protein 
products or ruminant feeds containing such products.
    The 1997 ruminant feed final rule (Sec.  589.2000) prohibits the 
use of mammalian-derived proteins in ruminant feed, with the exception 
of certain proteins believed at that time not to pose a risk of BSE 
transmission. These exceptions to the definition of ``protein derived 
from mammalian tissues'' included: Blood and blood products; gelatin; 
inspected meat products which have been cooked and offered for human 
food and further heat processed for feed (such as plate waste and used 
cellulosic food casings), referred to herein as ``plate waste'' milk 
products (milk and milk protein); and any product whose only mammalian 
protein consists entirely of porcine or equine protein. The 1997 
ruminant feed final rule does not prohibit ruminant animals from being 
fed processed animal proteins derived from nonmammalian species (e.g., 
avian or aquatic animals). The 1997 ruminant feed final rule permits 
the manufacture of non-ruminant feed containing prohibited mammalian 
protein and ruminant feed on the same premises, provided that separate 
equipment is used in the production of ruminant feed or that documented 
adequate clean-out procedures are used between production batches.
    Following the discovery of a BSE positive cow in Washington State 
in December 2003, FDA provided guidance on the use of materials from 
BSE positive cattle. In Guidance for Industry, ``Use of Material from 
BSE Positive Cattle in Animal Feed,'' published in the Federal Register 
in September 2004 (69 FR 58448), FDA stated its view that under section 
402(a)(5) of the Federal Food, Drug, and Cosmetic Act (the act) (21 
U.S.C. 342(a)(5)), animal feed and feed ingredients containing 
materials derived from a BSE-positive animal are considered adulterated 
and should be recalled or otherwise removed from the marketplace.

C. Risk of BSE in North America

    In April 1998, the United States Department of Agriculture (USDA) 
contracted with the Harvard Center for Risk Analysis (HCRA) at Harvard 
University and the Center for Computational Epidemiology at Tuskegee 
University to conduct a comprehensive investigation of the BSE risk in 
the United States. The report, (Ref. 3) widely referred to as the 
Harvard Risk Assessment or the Harvard Study, is referred to in this 
document as the Harvard-Tuskegee Study. The study was completed in 2001 
and released by USDA. Following a peer review of the Harvard-Tuskegee 
Study in 2002, the authors released a revised risk assessment in 2003 
(Ref. 4).
    The Harvard-Tuskegee Study reviewed available scientific 
information related to BSE and other TSEs, assessed pathways by which 
BSE could potentially occur in the United States, and identified 
measures that could be taken to protect human and animal health in the 
United States. The assessment concluded that the United States is 
highly resistant to any proliferation of BSE, and that measures taken 
by the U.S. Government and industry make the United States robust 
against the spread of BSE.
    The Harvard-Tuskegee Study concluded that the most effective 
measures for reducing potential introduction and spread of BSE are as 
follows: (1) The ban placed by USDA's Animal and Plant Health 
Inspection Service on the importation of live ruminants and ruminant 
meat-and-bone meal from the United Kingdom since 1989 and all of Europe 
since 1997 and (2) the feed ban instituted in 1997 by FDA to prevent 
recycling of potentially infectious cattle tissue. The Harvard-Tuskegee 
Study further indicated that, if introduction of BSE had occurred via 
importation of live animals from the United Kingdom before 1989, 
mitigation measures already in place would have minimized exposure and 
begun to eliminate the disease from the cattle

[[Page 58572]]

population even assuming less than complete compliance with the feed 
ban.
    The Harvard-Tuskegee Study also identified pathways or practices 
that, if addressed, would further decrease the already low risk of 
spread BSE if it were introduced into the United States. These include 
the following: (1) Failing to comply with FDA's ruminant feed 
regulations that prohibit the use of certain proteins in feed for 
cattle and other ruminants; and (2) rendering of animals that die on 
the farm (considered the highest risk cattle), and then incorporating 
(through illegal diversion or cross-contamination) the rendered product 
in ruminant feed. The Harvard-Tuskegee Study's independent evaluation 
of the potential additional risk mitigation measures predicts that a 
prohibition against rendering of animals that die on the farm would 
reduce potential new cases of BSE in cattle following a hypothetical 
introduction of 10 infected animals by 80 percent (from 4.3 to 0.77 
cases) as compared to the base case scenario, (i.e., present state of 
the U.S. cattle population, along with government regulations and 
prevailing agricultural practices, and an assumption of less than 
complete compliance with the feed ban) (Ref. 4). Further, the study 
evaluated the impact of a specified risk materials (SRMs) ban that 
would prohibit high risk materials such as the brain, spinal cord, 
vertebral column and animals that die on the farm, from inclusion in 
human and animal food. The analysis predicts that this measure would 
reduce potential new BSE cases in cattle following a hypothetical 
introduction of ten infected animals by 90 percent (from 4.3 to 0.53 
cases).
    In 2003, following the detection of BSE in a native-born cow in 
Canada, the HCRA evaluated the implications of a then-hypothetical 
introduction of BSE into the United States (Ref. 5), using the same 
simulation model developed for the initial Harvard-Tuskegee Study. The 
results of this assessment were consistent with the conclusions of the 
earlier study--namely, that the United States presents a very low risk 
of establishing or spreading BSE should it be introduced.
    On December 23, 2003, USDA announced that a dairy cow in Washington 
State had tested positive for BSE. The results were confirmed on 
December 25, 2003, by the Veterinary Laboratories Agency in Weybridge, 
England. Immediately after the diagnosis was confirmed, USDA, FDA, and 
other Federal and State agencies initiated an epidemiological 
investigation (Ref. 6), and began working together to trace any 
potentially infected cattle, trace potentially contaminated rendered 
product, increase BSE surveillance, and take additional measures to 
address risks to human and animal health. The epidemiological 
investigation and DNA test results confirmed that the infected cow was 
born and most likely became infected in Alberta, Canada, before 
Canada's 1997 implementation of a ban on feeding mammalian protein to 
ruminants.
    On January 22 through 24, 2004, the Secretary of Agriculture 
convened an international panel of experts on BSE. The panel, referred 
to as the International Review Team (IRT), was asked to: (1) Assess the 
epidemiological investigation conducted in response to the BSE case in 
Washington State, (2) provide expert opinion about when the active 
phase of the investigation should be terminated, (3) consider the 
response actions of the United States to date, and (4) provide 
recommendations about actions that could be taken to provide additional 
meaningful human or animal health benefits in light of the North 
American experience. The IRT provided its report on February 4, 2004.
    In May 2004, USDA contracted with HCRA to update the BSE risk 
assessment model to reflect its January 2004 rulemaking to prohibit 
SRMs and certain other cattle material in human food. HCRA was also 
asked to update the parameters in the model for compliance with FDA's 
feed ban. HCRA was also asked to model the impact that the IRT 
recommendation would have on the BSE risk to humans and cattle.
    In December 2004, Canada announced that a third North American cow 
tested positive for BSE. An ongoing epidemiologic investigation found 
that this animal, an 8-year-old, nonambulatory dairy cow, originated in 
Alberta, Canada and was born before the Canadian feed ban went into 
effect in August 1997. Shortly thereafter, in January 2005, another cow 
in Alberta was found to be positive for BSE. This case involved a beef 
cow born in March 1998, 6 months after the Canadian feed ban went into 
effect. Based on preliminary information, Canada believes that the most 
likely source of infection in this animal was feed produced before 
implementation of Canada's feed ban (Ref. 7).
    In June 2005, USDA announced that a 12-year-old beef cow, born and 
raised in Texas, was confirmed BSE positive. The BSE-positive cow most 
likely became infected before FDA's implementation of the 1997 ruminant 
feed final rule. It was determined that no part of the animal entered 
the human food or animal feed chains.

D. Additional Measures Considered to Strengthen Animal Feed Safeguards

1. Comments on November 6, 2002, Advance Notice of Proposed Rulemaking 
(ANPRM)
    In the Federal Register of October 5, 2001 (66 FR 50929), FDA 
announced its plan for an October 30, 2001 public hearing in Kansas 
City, MO, to solicit comments from the public on the 1997 ruminant feed 
regulation. Recognizing that new information had emerged since 
publication of the feed rule in 1997, FDA requested comments on whether 
changes to the rule or other additional measures were necessary (Ref. 
8). Information obtained from the public hearing and from the Harvard-
Tuskegee Study was used in the publication of an ANPRM (2002 ANPRM) in 
the Federal Register of November 6, 2002 (67 FR 67572). This ANPRM 
sought comment from affected industries and the public on possible ways 
to strengthen the 1997 ruminant feed regulation. The ANPRM specifically 
asked for comments on a number of questions related to the following 
five aspects of the BSE feed regulation: (1) Excluding brain and spinal 
cord from rendered animal products, (2) prohibiting the use of poultry 
litter in cattle feed, (3) assessing the improper use of pet food as a 
feed for ruminants, (4) preventing cross-contamination, and (5) 
eliminating the plate waste exemption.
    The predominant view of those who submitted comments in response to 
the ANPRM was that the BSE risk in the United States was low enough 
that no new feed controls were needed. Most said that the current feed 
ban provided more than adequate protection against BSE, that there was 
no scientific justification for additional regulations, that compliance 
with the 1997 ruminant feed final rule was extremely high, and that 
over 19,900 USDA surveillance samples in 2002 alone failed to detect 
BSE in U.S. cattle. They also cited the Harvard-Tuskegee Study 
conclusion that existing control measures made the risk to U.S. cattle 
and to U.S. consumers from BSE very low.
    In the 2002 ANPRM, FDA said that the Harvard-Tuskegee Study 
identified the removal of high-risk bovine tissues, such as brain, 
spinal cord, intestine, and eyes, from human food and from rendered 
material for all animal feed as a way to reduce the potential exposure 
of cattle and humans to the BSE agent. The 2002 ANPRM then asked for 
comments on the following three questions related to SRMs: (1) Should 
high risk materials be excluded from rendered products?; (2) how 
feasible would it be for the rendering industry

[[Page 58573]]

to implement such an exclusion?; and (3) what will be the adverse and 
positive economic, environmental, and health impacts from an exclusion?
    Comments in support of an SRM ban included one comment from USDA 
citing conclusions from the Harvard-Tuskegee Study that this action 
would significantly reduce the amount of infectivity in the animal feed 
chain, and would reduce risks resulting from ``leaks'' in the feed ban. 
Other comments stressed the infectivity of these tissues, and the 
recommendation by the World Health Organization (WHO) that countries 
exclude these tissues from the animal and human food chain (Ref. 9).
    Comments opposing an SRM ban said that the measure would be 
redundant because the 1997 ruminant feed final rule already prohibits 
this high-risk material in ruminant feed. Therefore, the ban would only 
be beneficial if BSE were present in the United States and there were 
significant non-compliance with the feed ban. The comments also cited 
the conclusions of the Harvard-Tuskegee Study that the risks of BSE in 
the United States are low. One comment said that restrictions on SRMs 
in animal feed should be decoupled from restrictions for human food 
because of the substantial reduction in infectivity obtained during 
rendering. Another comment said that an SRM ban would give only the 
perception of a risk reduction, not a real reduction, and that it would 
send the message to our trading partners that our BSE risks are such 
that more controls are needed. Australia asked that, if an SRM ban is 
implemented, the ban not apply to Australia because of its widely 
recognized status as a low-risk BSE country.
    Numerous comments addressed the feasibility and the adverse 
economic impacts of an SRM ban. One comment pointed out that it is not 
feasible to remove central nervous system (CNS) tissue from decomposing 
carcasses. Comments from a trade association said that an SRM ban would 
require costly restructuring of facilities that would force many small 
rendering plants out of business, depriving some parts of the country 
access to rendering as a means of animal disposal. A June 2002 Sparks 
Report estimated disposal costs of an SRM ban to be $54 million, based 
on the assumption that the ban would apply to all cattle because of the 
difficulty of determining the age of cattle at slaughter (Ref. 10). 
According to an earlier 1996 Sparks Report, the cost of disposal of 1.7 
billion pounds of CNS tissue and dead stock would exceed $400 million. 
Another estimate for disposal was $50 million for the beef industry 
alone. One comment said that feed costs account for 70 percent of 
poultry production cost, and that renderers would pass on the costs of 
excluding brains and spinal cords to the poultry industry.
    Several comments mentioned the environmental impact of an SRM ban. 
One comment stated that a total ban on SRMs in rendered animal products 
would create a waste stream with no economic value. Another comment 
said that a ban on SRMs would encourage improper disposal of dead stock 
because there are no federal regulations on disposal of dead animals.
2. Actions in Response to Washington State Case
    In response to the BSE case identified in Washington State, USDA 
published an interim final rule in the Federal Register of January 12, 
2004 (69 FR 1861), excluding high-risk tissues from human food. The 
interim final rule prohibited the use of SRMs and certain other cattle 
material in USDA-regulated human food. USDA defined SRMs as brain, 
skull, eyes, trigeminal ganglia, spinal cord, vertebral column 
(excluding the vertebra of the tail, the transverse processes of the 
thoracic and lumbar vertebrae, and the wings of the sacrum), and dorsal 
root ganglia (DRG) of cattle 30 months of age and older, and the 
tonsils and distal ileum of the small intestine of cattle of all ages. 
To ensure effective removal of the distal ileum, USDA requires that the 
entire small intestine be removed and disposed of as inedible product. 
In its January 12, 2004, interim final rule, USDA took the additional 
step of making cattle that are unable to rise from a recumbent 
position, referred to in this document as nonambulatory disabled 
cattle, ineligible to be slaughtered for human consumption.
    On January 26, 2004, FDA announced its intention to implement 
additional measures to strengthen existing BSE safeguards for FDA-
regulated products. These measures included the issuance of an interim 
final rule to implement additional measures related to animal feed. The 
interim final rule would have implemented four specific measures 
related to animal feeds. These measures included the elimination of the 
exemptions for blood and blood products and ``plate waste'' from the 
1997 ruminant feed rule, a prohibition on the use of poultry litter in 
ruminant feed, and a requirement for dedicated equipment and facilities 
to prevent cross-contamination.
    However, on February 4, 2004, IRT released its report on measures 
related to BSE in the United States. The report recommendations 
included a somewhat different set of measures for reducing the risks 
associated with animal feed than the measures FDA had announced that it 
intended to implement through an interim final rule. Although FDA 
believed its previously announced measures would serve to reduce the 
already small risk of BSE spread through animal feed, the broader 
measures recommended by the IRT, if implemented, could make some of the 
previously announced measures unnecessary. FDA believed that additional 
information was needed to determine the best course of action in light 
of the IRT recommendations and decided to publish an ANPRM, which 
requested comments on the recommendations of the IRT, as well as on 
other measures under consideration to protect the animal feed supply.
    Consistent with measures implemented by USDA to exclude high-risk 
cattle tissues from human food (69 FR 1861), FDA published an interim 
final rule on July 14, 2004 (69 FR 42255), prohibiting a similar list 
of risk materials from FDA-regulated human food, including dietary 
supplements, and cosmetics.
3. Comments on July 14, 2004, ANPRM
    In the Federal Register of July 14, 2004 (69 FR 42287), FDA 
published an ANPRM (2004 ANPRM) jointly with USDA in which FDA 
announced its tentative conclusion that it should propose banning SRMs 
in all animal feed. In this ANPRM, FDA asked for comment on this 
measure and also on the IRT's recommendations to require dedicated 
equipment or facilities for feed manufacture and transport, and its 
recommendation to prohibit the use of all mammalian and poultry protein 
in ruminant feed. Finally, FDA also asked for comment on the set of 
measures that the agency had announced in January 2004. Comments 
submitted in response to the 2004 ANPRM that relate to SRMs are 
summarized in sections I.D.3a through I.D.3f by general topic area.
    a. Need for SRM ban. As with the comments received in response to 
the 2002 ANPRM, many comments questioned the need for an SRM ban at the 
time of the 2004 ANPRM. Several comments argued that the comparison 
made by the IRT between the BSE situations in Europe and the United 
States is inappropriate. One reason given for the invalid comparison 
was that there were an estimated 3 to 4 million undiagnosed BSE cases 
in the United Kingdom, compared to two diagnosed cases in North America 
in cattle born before feed restrictions were implemented. Another 
comment said

[[Page 58574]]

that the United States did, in fact, learn from the European experience 
and implemented controls early so that potential animal exposure to the 
BSE agent in the United States remains exceedingly small compared to 
the massive exposure in the United Kingdom. One comment submitted by 
the agriculture department of a state with a large agriculture industry 
said that its findings from 600 inspections do not support the premise 
of the IRT's recommendation that an SRM ban is needed to address 
problems of cross-contamination and on-farm misfeeding. The state 
indicated that, in these inspections, it did not observe any prohibited 
materials or feed containing prohibited materials on farms where 
ruminant feeds were being mixed.
    Other comments said that the reduction in risk obtained through an 
SRM ban would be minimal, mostly citing the effectiveness of the 
current firewalls in reducing BSE infectivity in the cattle population. 
One comment said that the Harvard-Tuskegee Study conclusion that an SRM 
ban will reduce potential cattle exposure to BSE infectivity by 88 
percent sounds more impressive than it really is. Reducing a very small 
risk by 88 percent does not necessarily provide significant risk 
reduction.
    Finally, many comments questioned FDA's decision to ban SRMs from 
animal feed before the results of USDA's enhanced BSE surveillance 
program are known. USDA's one-time effort to test as many high-risk 
cattle as possible was started on June 1, 2004, and was expected to be 
completed by the end of 2005. One comment pointed out that the IRT's 
recommendations for defining SRMs are predicated on the outcome of this 
aggressive surveillance program.
    In support of FDA's tentative conclusion that it should propose to 
ban SRMs from all animal feed, many comments cited the conclusion of 
the Harvard-Tuskegee Study that an SRM ban will provide additional risk 
reduction, and also cited the recommendation of the IRT that SRMs 
should be excluded from all animal feed, including pet food. One 
comment said that an SRM ban would restore confidence in U.S. beef 
exports.
    b. Definition of SRMs. SRMs are typically defined as the tissues in 
which BSE infectivity has been demonstrated in experimentally or 
naturally infected animals. SRMs are further defined by the OIE 
Terrestrial Animal Health Code based on the age of the animal and the 
BSE risk status of a country. In the 2004 ANPRM, FDA asked how SRMs 
should be defined for animal feed, specifically, if the SRM list should 
be the same list as for human food. FDA also asked what information is 
available to support having two different lists.
    Comments from one organization included data from the Harvard-
Tuskegee Report on the relative infectivity of specific tissues. These 
data were based on pathogenesis studies carried out in the United 
Kingdom and showed the fraction of total infectivity of each tissue to 
be: Brain 64.1 percent; spinal cord 25.6 percent; dorsal root ganglia 
3.8 percent; trigeminal ganglia 2.6 percent; distal ileum 3.3 percent; 
tonsil < 0.1 percent; and eyes < 0.1 percent. The comment used the data 
to make the point that 90 percent of infectivity could be removed by 
excluding only the brain and spinal cord. A different comment citing 
the same data pointed out that the infectivity distribution represents 
more than a worst case scenario because, in the pathogenesis study, the 
BSE dose administered orally to calves was substantially greater than 
would reasonably be expected under field conditions. This second 
comment went on to point out that FDA's interim final rule on food and 
cosmetics said that in cattle infected under field conditions, BSE 
infectivity had been demonstrated only in the brain, spinal cord, and 
retina of the eye at the end stages of the disease.
    Many comments recommended that the human food list of SRMs be used 
to define which SRMs should be excluded from animal feed. Several 
comments recommended expanding the list beyond the human food list by 
applying it to tissues from cattle 12 months of age or older, or to 
tissues from all cattle. Others advocated eliminating bovine or animal 
protein from ruminant feed altogether. Reasons given by the comments 
for these recommendations were the large risk reduction that could be 
achieved and the desirability of being consistent with the requirements 
for human food.
    Those who submitted comments in support of a more limited SRM list 
mostly did so to minimize the volume of material that would require 
nonfeed disposal. The comments stated that reducing this volume of 
material that would require nonfeed disposal would lessen the adverse 
impact of an SRM ban on the livestock, meat, and animal feed 
industries. One company used the Harvard model to simulate three 
different SRM scenarios and then submitted data showing that limiting 
the SRM list to brain and spinal cord (while also prohibiting use of 
dead stock and downers over 30 months of age), eliminating vacuum 
rendering, and keeping the existing feed ban in place, achieved a risk 
reduction equivalent to that obtained by banning the full human list of 
SRMs.
    The following are other suggestions provided in comments submitted 
in response to the 2004 ANPRM for reducing the volume of SRM material 
needing alternative disposal: (1) Allow the use of SRMs from animals 
that test negative for BSE, (2) designate only the head as an SRM which 
reduces by 64 percent the potential BSE infectivity in feed, (3) allow 
the use of intestines from veal calves whose carefully controlled diets 
consist of low-risk formulas, and (4) allow mechanically separated beef 
from pet food plants to be used if SRMs are removed before meat is 
mechanically separated from bones.
    c. Cattle not inspected and passed for human consumption. The term 
``cattle not inspected and passed for human consumption'' is used in 
this document to mean cattle that were not inspected and passed for 
human consumption by the appropriate regulatory authority. For the 
purposes of this document, this term also includes nonambulatory 
disabled cattle, i.e., cattle that could not rise from a recumbent 
position or that could not walk, including, but not limited to, those 
with broken appendages, severed tendons or ligaments, nerve paralysis, 
fractured vertebral column, or metabolic conditions. This proposed 
definition is consistent with the use of the terms ``inspected and 
passed and nonambulatory disabled cattle'' as defined in USDA's interim 
final rule on human food (69 FR 1862) and FDA's interim final rule on 
human food and cosmetics (69 FR 42255). For the purposes of this 
proposed rule, nonambulatory disabled cattle are included in the 
definition of cattle not inspected and passed, since nonambulatory 
disabled cattle cannot be passed for human consumption.
    A number of questions were included in the 2004 ANPRM regarding the 
use of materials from cattle not inspected and passed for human 
consumption as previously defined. Comments received discussed both the 
advantages and disadvantages of excluding these animals from being 
rendered for use in animal feed.
    Advantages mentioned included the additional risk reduction that 
would be provided by the measure. A number of comments cited the 
Harvard-Tuskegee Study, which showed that removing dead stock from the 
feed chain would reduce potential exposure of cattle to the BSE agent 
by 88 percent. However, other comments noted that such a ban would 
result in dead stock being disposed of on the farm, impacting USDA's 
surveillance program and

[[Page 58575]]

increasing environmental problems due to improper disposal of animal 
carcasses. Concerns were also expressed about lack of infrastructure 
for non-feed disposal of dead stock, and the serious economic impact of 
diverting these animals to alternative disposal.
    In response to the question in the 2004 ANPRM about effective 
removal of SRMs from dead stock and nonambulatory disabled cattle, 
several comments stated that such removal would not be economically or 
technically feasible. Other comments stated that SRM material could be 
effectively removed because there is no substantial difference between 
the processing of dead and nonambulatory animals at rendering 
facilities and the processing of healthy cattle at slaughter plants. 
One other comment mentioned instances where some USDA-inspected 
deboning facilities already remove SRMs from dead cattle at the request 
of pet food manufacturers. This comment also said that, based on their 
experience, SRMs can be removed from dead cattle in all but the hottest 
months of the year when the rate of decomposition increases. Another 
comment said that removing SRMs from dead stock may increase exposure 
of plant employees to pathogens and zoonotic diseases.
    One comment noted that the European experience has shown that 
cattle at highest risk for BSE are dead cattle, downer cattle, and 
ante-mortem condemned cattle over 30 months of age. This comment said 
that, while it is possible to remove the meat from these carcasses for 
use in pet food, they are not aware of any way of verifying the removal 
of SRMs from dead and nonambulatory cattle (short of active government 
oversight) that would allow this material to be rendered for use in 
feeds for non-ruminant animals. Another comment suggested that as an 
option for reducing the amount of material for disposal, dead stock 
under 30 months of age be allowed to be rendered for feed use. This 
comment also said that USDA could test dead stock over 30 months of 
age, allowing material from negative animals to be used in feed.
    d. Small intestine. The 2004 ANPRM also requested information to 
evaluate the IRT recommendation that the entire intestine from cattle 
of all ages should be excluded from the human and animal food chains. 
With publication of its interim final rule on January 12, 2004, USDA 
required that the entire small intestine be disposed of as inedible. 
Likewise, FDA prohibited the use of the entire small intestine in FDA-
regulated human food and cosmetics, even though the agency only 
considers the distal ileum portion of the small intestine to be a 
specified risk material (69 FR 42259).
    However, based on comments received in response to the FDA interim 
final rule on human food and cosmetics, FDA concluded that processors 
have the technology to effectively remove the distal ileum portion from 
the rest of the small intestine. Thus, FDA amended the human food and 
cosmetics interim final rule to state that the small intestine is not 
considered prohibited cattle material if the distal ileum is removed by 
a procedure that removes at least 80 inches of the uncoiled and trimmed 
small intestine as measured from the caeco-colic junction and 
progressing proximally towards the jejunum or by a procedure that the 
establishment can demonstrate is equally effective in ensuring complete 
removal of the distal ileum (70 FR 53063, September 7, 2005). This 
amendment is consistent with USDA requirements (70 FR 53043, September 
7, 2005).
    Many comments in response to the 2004 ANPRM stated that inclusion 
of the entire small intestine from cattle less than 30 months of age in 
the list of prohibited material would double the volume of SRMs from 
slaughter requiring alternative disposal while only marginally 
decreasing infectivity. Several comments stated that only the distal 
ileum should be included in the list of SRMs and noted that it is 
easily identified for separation at slaughter.
    One comment questioned the need to designate the intestinal tract 
as SRM, pointing out that the distal ileum accounts for only 5 percent 
of infectivity, which is reduced by two logs during rendering. Another 
comment said that it was unnecessary to designate any portion of the 
intestinal tract of cattle less than 30 months of age as SRM because 
these animals were born 4 1/2 years after the feed ban was implemented, 
and are therefore low risk animals. Several comments said that, if 
packers can demonstrate a satisfactory technique, they should be 
allowed to remove only the distal ileum rather than the entire small 
intestine.
    One comment expressing concern about the BSE risk associated with 
bovine intestines said that research in the United Kingdom found 
positive immunostaining for the resistant form of the prion protein 
along the length of the intestine, which provides evidence that the 
entire intestine should be considered SRM.
    e. Infrastructure for alternative disposal. We received a number of 
comments addressing the issue of disposal infrastructure. One comment 
noted that the IRT recognized that an infrastructure was not in place 
to dispose of SRM material and that the IRT had suggested that a staged 
implementation may be necessary to allow this infrastructure to 
develop. One comment said that before an SRM ban is implemented a 
comprehensive plan for disposal of this material needs to be developed. 
Another comment noted that in Texas, SRMs are considered special waste, 
and that no landfill in the state is capable of accommodating a large 
volume of this material. Additional comments indicated that this 
concern was also true for other states, including Nebraska and Utah.
    Two organizations submitted slaughter and cattle mortality data to 
emphasize the amount of waste that would be generated by regulations 
that would exclude this material from being rendered for use in animal 
feed. One of these organizations said that it is deeply concerned that 
FDA fails to recognize that a suitable disposal infrastructure does not 
exist to deal with the very large quantities of SRMs that would be 
generated on a daily basis. Its estimate for the volume of waste 
generated from slaughter and cattle mortalities was 2 billion pounds 
per year. The other organization submitted similar comments saying that 
the U.S. system is currently unprepared to manage the waste disposal 
challenges certain to arise if significant quantities of livestock 
mortalities and slaughter byproducts require disposal by means other 
than rendering. The comments further stated that the disposal and 
environmental challenges resulting from the ban would be faced 
immediately, but the solutions to these challenges would arise only 
after significant time and financial investment across the livestock 
sector. The comments also said that there is an absence of direct 
regulatory control over alternative methods of disposing of the 
enormous quantities of this unpleasant material.
    Another comment suggested that renderers should be allowed to 
dedicate lines to SRM material and SRM-free material within a single 
facility. Equipment for receiving, grinding, cooking, processing, and 
conveying could be dedicated lines, while the facility itself, 
including the utilities, odor control, and wastewater treatment systems 
be shared. Further, another comment suggested FDA work with the 
rendering industry to develop cleanout procedures that would allow a 
plant to process both SRMs and SRM-free material. These procedures 
would be helpful to allow for seasonal deer rendering, for cleaning up 
after accidental cross-contamination, and for

[[Page 58576]]

converting a facility back to SRM-free rendering.
    One comment addressed the use of rendered SRM material as an 
alternative fuel source for cement kilns, indicating that ruminant meat 
and bone meal and fat are being used as a fuel source in Europe and 
Japan. According to the comment, these materials burn efficiently, and 
the heat from the kiln leaves virtually no organic residue.
    f. Verification of SRM removal and SRM marking. One comment stated 
that, in the absence of a practical test for verification of SRM 
removal, the documentation required by HACCP plans should be sufficient 
to show that SRMs at slaughter are excluded from animal feed channels. 
Thus, inspections of records could be used to verify SRM removal. Also, 
the comment stated that FDA can verify SRM removal by shifting 
resources from inspections of thousands of feed mills and farms to the 
much smaller number of slaughter plants and renderers.
    One comment stated that rendering plants are capable of keeping raw 
materials from various sources separated and capable of using 
production, inventory, and shipping records to document the movement of 
both SRM and SRM-free materials. Such management practices can be 
verified by inspection, much like those conducted at USDA-inspected 
cattle slaughter facilities. The comment went on to say that, if a 
rendering plant is dedicated to rendering only SRMs, such a plant will 
have to be inspected to determine how it disposes of SRMs.
    Two comments suggested that raw or SRM-derived rendered materials 
can be effectively marked using automatic dosage pumps to dispense 
markers like glyceroltriheptanoate (GTH). GTH is a C7 synthetic fatty 
acid not found in nature. A gas chromatography (GC) method for its 
detection is available. Charcoal was mentioned as another potential 
marker for use in rendered products.

II. Proposed Measures to Strengthen Animal Feed Safeguards

A. FDA Response to Comments to the 2004 ANPRM

    FDA agrees with the numerous comments saying that it is important 
to keep the BSE risk in the United States in proper perspective. FDA 
acknowledges that the risk is likely low, and acknowledges that it is 
inappropriate to compare the BSE situation in the United States to the 
situation in Europe. However, FDA disagrees with comments concluding 
that for these reasons no additional measures are needed. Even though 
strong control measures have been put in place and compliance with the 
current BSE feed regulation is high by renderers, protein blenders and 
feed mills, the Agency is concerned, as discussed further below, about 
such issues as the presence of high risk material in the non-ruminant 
feed supply and cross-contamination of ruminant feed during the 
rendering or feed manufacturing process. For example, without fully 
dedicated equipment, it may not be possible to verify that there is 
zero carryover of feed or feed ingredients in equipment, even where a 
firm's cleanout procedures have been judged to be adequate. In 
addition, resource constraints limit FDA's ability to assure full 
compliance by all segments of the industry that are subject to the 
current BSE feed regulation. For example, resources are not available 
to the FDA and its state counterparts to fully verify compliance on 
over 1 million farms where cattle are being fed.
    FDA does not agree with comments that the agency should wait until 
USDA completes its enhanced BSE surveillance program before deciding if 
additional feed controls are needed. As stated in the July 2004 ANPRM, 
FDA had tentatively decided based on clear evidence that the BSE agent 
had been introduced into the North American animal feed supply, and 
based on the recommendation of the IRT, that SRMs should be removed 
from all animal feed. Results from the enhanced surveillance that was 
being conducted concurrent with our rulemaking process indicated that 
BSE had been introduced into the United States, but was present at a 
very low level. These results reinforced FDA's decision that the 
measures being proposed are appropriate.
    With respect to the definition of SRMs, FDA agrees that prohibiting 
the full list of SRMs would achieve greater risk reduction than 
prohibiting a partial list, but also agrees with comments saying that 
the infrastructure does not currently exist to handle the volume of 
material that would require non-feed disposal if the full list of SRMs 
were diverted from animal feed use. Therefore, FDA agrees that focusing 
on brain and spinal cord is an effective approach for achieving 
additional animal and public health protection while minimizing the 
economic, environmental, and public health concerns associated with 
disposal of the full list of SRMs. FDA, however, seeks comments on 
whether a full SRM ban is warranted.
    Comments were mixed on the feasibility of removing SRMs from dead 
stock. FDA therefore concluded that some firms would elect to remove 
SRMs and render the remainder of the carcass, and that this could 
lessen difficulties associated with alternative disposal. FDA does not 
agree that allowing test-negative animals to be rendered for animal 
feed use is appropriate. Unlike Europe, rapid screening tests in the 
United States have been used only for surveillance purposes. These 
tests have not been used as food or feed safety tests because currently 
available tests can detect BSE only in the late stages of disease. 
Finally, although FDA agrees that vacuum rendering is less effective at 
inactivating TSEs than atmospheric rendering, the Agency disagrees that 
vacuum rendering should be prohibited. Modeling results submitted with 
the comment showed that such a prohibition would result in an 
additional one percent reduction in risk. In light of other measures 
being proposed and the fact that few plants use vacuum rendering, FDA 
does not believe that prohibiting this rendering process would 
appreciably improve animal or public health protection.

B. Additional Measures to Further Strengthen Feed Protection

    The United States and Canadian feed regulations implemented in 1997 
were necessary because of uncertainty about whether BSE infectivity had 
already been introduced into North America before new import 
restrictions on live cattle and meat and bone meal from Europe were put 
in place. It is now clear from the five North American BSE cases that 
the BSE agent was introduced into the North American animal feed supply 
at some point in time. While FDA continues to believe that compliance 
with the feed regulation has provided strong protection against the 
spread of BSE, the agency believes that the recent cases are an 
indication that additional animal feed protections are needed to remove 
residual infectivity that may be present in the animal feed supply. FDA 
also believes that of all the options considered since publication of 
the 2002 ANPRM, excluding the highest risk tissues from all animal feed 
is the best approach to address the risks of BSE in the United States. 
In the 2004 ANPRM, FDA announced its tentative conclusion that it 
should propose a prohibition on the use of SRMs in all animal feed.
    The decision to propose banning SRMs from all animal feed led to 
the following questions: (1) Which material to exclude, (2) what 
alternative disposal methods could be used, (3) what the economic and 
environmental impacts of diverting material to alternative disposal 
would be, and (4) how an SRM ban could be enforced. As the IRT 
reported,

[[Page 58577]]

exclusion of large volumes of raw material is a massive burden for all 
countries affected by BSE. FDA received valuable information pertaining 
to these issues in comments submitted in response to the 2004 ANPRM.
    In reaching a decision about what specific additional measures 
should be proposed at this time, FDA considered the magnitude of the 
BSE risk in the United States. While the recent North American cases 
clearly show the BSE agent was introduced, the USDA enhanced BSE 
surveillance program indicates that the prevalence of the disease in 
the United States is very low. As of July 2005, USDA has tested over 
418,000 high-risk cattle under its enhanced BSE surveillance program 
(Ref. 11), and has found one positive animal in addition to the cow 
identified in Washington State in December 2003. Therefore, FDA 
believes that the additional measures being proposed are appropriate at 
this time. The agency proposes to prohibit from use in all animal feed 
the brains and spinal cords from cattle 30 months of age and older, the 
brains and spinal cords from all cattle not inspected and passed for 
human consumption, and the entire carcass of cattle not inspected and 
passed for human consumption from which brains and spinal cords were 
not removed. The agency also proposes to prohibit from use in all 
animal feed mechanically separated beef and tallow that are derived 
from materials prohibited by the rule. However, the rule proposes to 
exempt tallow from this requirement if it contains no more than 0.15 
percent insoluble impurities.

C. Basis for Proposing to Apply Additional Measures to All Animal Food 
and Feed

    The current U.S. ruminant feed regulation prohibits the use of 
certain mammalian-origin proteins in ruminant feed, but allows the use 
of these materials in feed for non-ruminant species. FDA believes that 
the presence of high-risk materials in the non-ruminant feed supply 
presents a potential risk of BSE to cattle in the United States. 
European experience showed that, in countries with high levels of 
circulating BSE infectivity, controls on only ruminant feed were not 
sufficient to prevent further transmission of BSE. Until SRMs were 
removed from all animal feed, a significant number of new cases 
continued to be found in cattle born in the United Kingdom after 
implementation of a ruminant-to-ruminant feed ban (Ref. 12). These new 
cases were attributed to either cross-contamination during feed 
manufacture and transport, or to intentional or unintentional 
misfeeding on the farm.
    The 1997 ruminant feed regulation requires feed manufacturers and 
distributors that handle both ruminant feed and feed ingredients and 
materials prohibited in ruminant feed to control cross-contamination by 
either: (1) Maintaining separate equipment or facilities or (2) using 
adequate clean-out procedures or other means adequate to prevent carry-
over of prohibited material into feed for ruminant animals. FDA has 
been concerned about the adequacy of such clean-out procedures and 
sought public comment on this issue in the 2002 ANPRM. Although many 
firms using the clean-out option have written procedures in place, 
evaluating their adequacy is difficult because of wide variation in 
equipment and practices used by the feed industry, and because there is 
currently no definitive test method to detect prohibited proteins.
    Further increasing FDA's concerns about cross-contamination are 
preliminary data from an unpublished study showing that the minimum 
infectious dose for BSE may be lower than previously thought. Interim 
results at approximately 5 years post exposure of an oral challenge 
experiment have demonstrated transmission of BSE to 1 out of 15 animals 
that received 0.01 gram of brain tissue from a BSE-infected animal 
(Ref. 13). The lowest dose previously tested was 1.0 gram of brain 
tissue which showed transmission of BSE in 7 out of 10 animals in the 
trial group. This finding of a lower minimum infectious dose for BSE 
would suggest that the risk from cross-contamination is greater than 
previously thought. We seek comment on this interpretation of theses 
interim results.
    Instances of cattle being exposed to prohibited material through 
noncompliance with the 1997 feed bans have been identified in both 
Canada and the United States. The investigation by the Canadian Food 
Inspection Agency of the BSE case identified in May 2003 found several 
instances where cattle might have had access to non-ruminant feed 
containing prohibited material. In the United States, FDA inspections 
have identified situations where cattle could have been exposed to 
material prohibited in ruminant feed as a result of ruminant feed being 
contaminated with non-ruminant feed, or non-ruminant feed not being 
properly labeled.
    In fiscal year 2004 and the first half of fiscal year 2005, federal 
and state inspections identified 41 instances (0.4 percent of 
inspections) of cross-contamination or commingling problems in firms 
that handle animal feeds containing prohibited mammalian protein (Ref. 
14). During this same period, inspections identified 165 instances (1.7 
percent of inspections) in which non-ruminant feeds containing 
prohibited material were not properly labeled with the caution 
statement ``Do Not Feed to Cattle or Other Ruminants''. Firms receiving 
mislabeled feed would not be aware of the need to take steps to prevent 
cross-contamination of ruminant feed with such products. Furthermore, 
inspections during this period identified 604 instances (6.3 percent of 
inspections) in which firms handling animal feeds containing prohibited 
mammalian protein did not meet the recordkeeping requirements. These 
instances involved a variety of recordkeeping deficiencies, including 
not maintaining sales records for feeds received or distributed, not 
establishing written protocols for avoiding commingling, and not fully 
documenting clean-out measures utilized. Such deficiencies are 
typically corrected by the involved firms without further action by the 
agency. However, the occurrence of these deficiencies nonetheless 
supports the need for additional measures to address concerns about the 
presence of high-risk materials in the non-ruminant feed supply. 
Without sales records, it is difficult to verify the source of feed or 
feed ingredients or to track distributed feeds when conducting recalls 
in response to known instances of product contamination. Without 
appropriate documentation of procedures related to commingling or 
cross-contamination, it is difficult to verify that workers are 
informed of such procedures or that the procedures are adequate.
    FDA has issued warning and untitled letters to firms addressing 
noncompliance with the current ruminant feed ban regulation and a feed 
manufacturer has been permanently enjoined in connection with 
noncompliance with the current feed ban regulation.
    FDA is also concerned about intentional and unintentional 
misfeeding of non-ruminant feed to ruminants on the farm. Financial 
incentives for intentional misfeeding could occur any time inexpensive 
sources of prohibited protein are locally available to the feeder. The 
use of salvaged pet food that contains ruminant meat and bone meal is 
an example. There may be other incentives to intentionally feed non-
ruminant feed to cattle. For example, the Florida Department of 
Agriculture and Consumer Services issued a statement cautioning against 
the misuse of pet

[[Page 58578]]

food as feed for show cattle as a way to increase the shine in the 
cattle coat (Ref. 15). Unintentional feeding could occur on the farm 
from feeding ruminants and non-ruminant in close proximity to each 
other. If intentional or unintentional uses occur, this proposed rule 
would protect cattle by removing the highest risk material from the 
non-ruminant feed being used in cattle feed. Assuring that misfeeding 
does not occur on the farm is particularly difficult due to the large 
number of cattle feeding operations in the United States, and FDA's 
extremely limited resources to inspect these operations, which number 
over 1 million.
    Therefore, although overall compliance with the 1997 ruminant feed 
rule has been high for renderers, protein blenders, and feed mills, 
removal of the highest risk tissues from animal feed channels should 
serve to address noncompliance with the rule that could result in 
cattle exposure to prohibited material through cross-contamination, 
mislabeling, or intentional or unintentional misfeeding.

D. Cattle Materials Proposed to be Prohibited From Use in All Animal 
Food and Feed

1. Brain and Spinal Cord From Cattle 30 months of Age and Older
    The USDA interim final rule published on January 12, 2004, provides 
a full description of the scientific rationale for identifying the list 
of tissues and selection of the 30-month age criterion used in its 
definition of SRMs. FDA has adopted an identical definition of SRMs in 
its interim final rule regarding FDA-regulated human food and 
cosmetics. In the preamble of its July 14, 2004 interim final rule 
regarding human food, including dietary supplements, and cosmetics, FDA 
includes a detailed discussion of its rationale for the SRM definition. 
As discussed in the preamble to the USDA and FDA interim final rules, 
infectivity is not present in most tissues that harbor BSE infectivity 
until more than 30 months after the animal was exposed to the agent. 
Although the epidemiological and experimental data indicate that BSE 
can develop in animals less than 30 months of age, the evidence 
available to date indicates that this was a very rare occurrence, and 
was associated with high levels of circulating infectivity at the peak 
of the BSE epidemic in the United Kingdom. The agency continues to 
believe that the rationale for the 30-month age criterion described 
previously for human food and cosmetics is appropriate and proposes 
that it be applied to animal feed as well.
    In response to a question posed in the 2004 ANPRM as to which 
tissues should be defined as SRMs for animal feed, FDA received 
suggestions ranging from defining all animal protein as SRMs to 
limiting the SRM definition to the head only. FDA considered 
prohibiting from animal feed the same materials defined as SRMs that 
are currently prohibited from use in food for humans, but decided that 
proposing to require the removal of brain and spinal cord is the most 
appropriate approach at this time.
    In reaching the decision to propose to exclude only the brain and 
spinal cord from animal feed, FDA considered information regarding the 
tissue distribution of BSE infectivity. Under field conditions, BSE 
infectivity has been found in the brain, spinal cord, and retina of the 
eye in animals with clinical disease (Ref. 16). The Scientific Steering 
Committee (SSC) of the European Union (Ref. 17) has also reported on 
the proportion of total infectivity in various tissues.\1\ According to 
the report, in an animal with clinical BSE, approximately 64 percent of 
the infectivity is in the brain, 26 percent is in the spinal cord, 4 
percent is in the dorsal root ganglia, 2.5 percent is in the trigeminal 
ganglia, and 3 percent is in the distal ileum. The eyes are estimated 
to contain less than 1 percent of the infectivity. Although available 
data are limited on the distribution of tissue infectivity, data from 
both naturally infected and experimentally infected cattle support the 
finding that the brain and spinal cord are the tissues with the highest 
level of infectivity.
---------------------------------------------------------------------------

    \1\ A more recent report (Comer and Huntly, 2004, Journal of 
Risk Research, 7, (5) 523-543) attributes 84.3 percent of 
infectivity to brain and spinal cord and 9.6 percent to distal 
ileum. We chose not to use the data from this more recent report 
because its author (personal communications) explained that the 
newer data suggesting that the level of infectivity in the distal 
ileum at 6 to 18 months of age is higher than earlier estimates also 
suggest that it is lower than earlier estimates at 32 months of age.
---------------------------------------------------------------------------

    Because available data indicate that the brain and spinal cord 
contain about 90 percent of BSE infectivity (Ref. 17), FDA believes 
that the most appropriate course of action is to concentrate efforts on 
excluding these highest risk tissues from animal feed. In deciding to 
propose to prohibit brain and spinal cord only, rather than the same 
list of materials previously defined as SRMs, FDA also considered the 
following: (1) Surveillance data indicate the current risk of BSE to 
U.S. cattle is very low, (2) the existing ruminant feed regulation 
provides strong protection against BSE, and (3) the new measures 
considered in this proposed rule represent a secondary level of 
protection to address failures in compliance that may occur with the 
existing ruminant feed rule. FDA believes that the existing ruminant 
feed rule provides the primary line of defense by prohibiting the use 
in ruminant feed of all material with potential BSE infectivity. The 
measures proposed by this rule will effectively reinforce existing 
ruminant feed protection measures by removing the tissues with the 
highest infectivity from all animal feed. As a result, these measures 
greatly minimize BSE risks if cross-contamination of ruminant feed with 
non-ruminant feed, or diversion of non-ruminant feeds to ruminants, 
were to occur.
2. Cattle Not Inspected and Passed for Human Consumption
    As noted earlier in this document, the term ``cattle not inspected 
and passed for human consumption'' includes cattle not inspected and 
passed for human consumption by the appropriate regulatory authority as 
well as nonambulatory disabled cattle.
    European surveillance data indicate that cattle found dead or 
culled onsite, where the carcass was submitted to rendering (fallen 
stock), and cattle with health-related problems unfit for routine 
slaughter (emergency slaughter) have a greater incidence of BSE than 
healthy slaughter cattle. Surveillance data in the European Union in 
2002 showed that there were 27.95 positive animals per 10,000 emergency 
slaughter bovine animals tested and 6.15 positive animals per 10,000 
fallen stock bovine animals tested compared to 0.31 positive animals 
per 10,000 healthy slaughter animals tested (Ref. 18). In Switzerland, 
the odds of finding a BSE case in fallen stock and emergency slaughter 
cattle were found to be 49 and 58 times higher, respectively, compared 
to the odds of finding a BSE case through passive surveillance (Ref. 
19). These findings suggest that cattle not inspected and passed for 
human consumption are more likely to test positive for BSE than healthy 
cattle that have been inspected and passed for human consumption.
    Because cattle not inspected and passed for human consumption are 
included in the population of cattle at highest risk for BSE (Refs. 18 
and 19), and processes are currently not established in the rendering 
industry for verifying the age of such cattle through inspection, the 
agency is proposing to define brains and spinal cords from all cattle 
not inspected and passed for human consumption, regardless of age, to 
be cattle materials prohibited in animal feed. As noted previously, the

[[Page 58579]]

term cattle not inspected and passed for human consumption is defined 
in this proposed rule to include nonambulatory disabled cattle as 
defined by FDA in its interim final rule on human food and cosmetics 
and USDA in its interim final rule on human food. If the brains and 
spinal cords are removed from these animals, FDA is proposing that the 
remaining material can still be used in animal feed. FDA notes that for 
cattle not inspected and passed that are diseased or that die other 
than by slaughter, the entire carcass of such animals is adulterated 
under section 402(a)(5) of the act. FDA has traditionally exercised 
enforcement discretion with regard to the use of such animals in animal 
feed. For example, see Compliance Policy Guide 675.400. FDA intends to 
continue exercising such discretion for the use in animal feed of the 
remaining material from cattle that are diseased or that die other than 
by slaughter when the brain and spinal cord are removed. Because 
comments to the ANPRM were mixed on the feasibility of removing SRMs 
from cattle mortalities, FDA requests further comment on which tissues 
should be removed from this class of animals and the feasibility of 
removing them.
    In deciding to propose to allow these remaining materials to be 
used in animal feed, FDA considered the following: (1) brain and spinal 
contain about 90 percent of BSE infectivity (Ref. 17), (2) surveillance 
data indicate the current risk of BSE to U.S. cattle is very low, (3) 
the existing ruminant feed rule provides strong protection against BSE, 
and (4) the new measures considered in this proposed rule represent a 
secondary level of protection to address failures in compliance that 
may occur with the existing ruminant feed rule. FDA believes that the 
existing ruminant feed rule provides the primary line of defense by 
prohibiting the use in ruminant feed of all material with potential BSE 
infectivity. If the brains and spinal cords are not removed from such 
animals, FDA proposes that all parts of ``cattle not inspected and 
passed for human consumption'' be prohibited.
3. Mechanically-Separated Beef (MS)
    Mechanically-separated (MS) beef is a finely comminuted meat food 
product resulting from the mechanical separation and removal of most of 
the bone from attached skeletal muscle of cattle carcasses and parts of 
carcasses. This proposed definition of MS beef is consistent with, but 
not identical to, the definition of the term used by USDA in its 2004 
interim final rule (69 FR 1862) prohibiting its use in human food and 
by FDA in its 2004 interim final rule (69 FR 42255) prohibiting its use 
in human food, including dietary supplements and cosmetics. Those 
definitions provide that MS beef means a meat food product that meets 
the specification in 9 CFR 319.5. This USDA regulation applies to MS 
beef for human food use. Because there is MS beef produced solely for 
animal feed use that would not fall within the USDA specification, the 
definition of MS beef as proposed in this rule is meant to refer to 
beef that is the product of the mechanical separation process, 
regardless of whether it meets the USDA specifications for MS species 
in 9 CFR 319.5. The definition of MS beef is not meant to include 
product produced by Advanced Meat Recovery (AMR) systems used in the 
meat industry.
    Although MS beef was not considered in the 2002 ANPRM, 2004 ANPRM, 
or in the IRT report, FDA has included this material in this animal 
feed proposed rule to ensure that any such material that is used in 
animal feed is not contaminated with the other material prohibited by 
this proposed rule. A comment submitted in response to the 2004 ANPRM 
said that FDA should allow mechanically separated beef to be used for 
pet food if SRMs are removed from material going into the mechanical 
deboning equipment that separates meat from bone, because some pet food 
operations are very similar to slaughter establishments and are capable 
of removing SRMs.
    Because the mechanical separation process may result in the 
contamination of the MS beef product with spinal cord, FDA proposes to 
designate MS beef as cattle materials prohibited in animal feed if it 
is derived from carcasses or parts of carcasses from which cattle 
materials prohibited in animal feed were not previously removed.
4. Tallow
    Tallow is an animal-derived hard fat that has been heat processed; 
most tallow is derived from cattle. Any risk of BSE transmission from 
tallow is a result of protein that is present as an impurity in the 
tallow. Taylor et al. (Refs. 20 and 21) found, in rendering studies 
with abnormal prion protein, that the prion protein did not 
preferentially migrate into the fat fraction, but remained with the 
protein fraction. Therefore, there is no reason to believe that tallow 
is likely to contain unusually high amounts of prion protein as a 
constituent of the insoluble impurities fraction that remains in tallow 
after rendering. Taylor et al. (Refs. 20 and 21) also reported that the 
various rendering processes used for tallow production in the United 
Kingdom were sufficient to produce tallow that did not result in 
infection when injected into the brains of mice, even though the 
starting material was highly spiked with the scrapie agent. Wilesmith 
et al. (Ref. 22) noted that the geographical variation in the incidence 
of BSE in the United Kingdom was not consistent with the use of tallow 
in cattle feed and concluded that the most likely source of infection 
in cattle was BSE-contaminated meat and bone meal.
    The Office International des Epizooties (OIE), the world 
organization for animal health, categorizes tallow with insoluble 
impurities of no more than 0.15 percent as protein-free tallow. OIE 
guidelines recommend that tallow that meets this standard can be safely 
traded regardless of the BSE status of the exporting country (Ref. 23). 
FDA's Transmissible Spongiform Encephalopathy Advisory Committee 
(TSEAC) considered the safety of tallow and tallow derivatives in 1998 
(Ref. 24). Members of the committee indicated that tallow is a food 
with negligible or no risk of transmitting BSE to humans or animals.
    For the purposes of this proposed rule, tallow is defined as the 
rendered fat of cattle obtained by pressing or by applying any other 
extraction process to tissues derived directly from discrete adipose 
tissue masses or to other carcass parts and tissues. The 1997 ruminant 
feed final rule did not include tallow, fats, oils, and grease in the 
definition of animal proteins prohibited in ruminant feed because they 
are not proteins and were not considered to contain BSE infectivity. 
The agency said that infectivity studies conducted on some of these 
products (e.g., tallow) had demonstrated that they were at low risk of 
transmitting the TSE agent and; thus, it was unnecessary to restrict 
their use in ruminant feed (62 FR 30935). While the agency is not aware 
of any new scientific information suggesting that infectivity is 
present in tallow itself, the agency is concerned about potential BSE 
risks that tallow poses as a result of protein that is present as an 
impurity. These impurities may be of greater concern now because, as 
previously noted, new preliminary data suggest that the minimum 
infectious dose for BSE may be substantially lower than previously 
thought. We seek comment on this interpretation of the preliminary 
results.
    The agency is proposing to prohibit the use of tallow in animal 
food or feed that is derived from cattle materials prohibited in animal 
feed. However, the agency proposes to exempt from this requirement 
tallow that contains no

[[Page 58580]]

more than 0.15 percent insoluble impurities. The proposal would require 
that impurities be measured by the method entitled ``Insoluble 
Impurities'' (AOCS Official Method Ca 3a-46), American Oil Chemists' 
Society, which is incorporated by reference in accordance with 5 U.S.C. 
552(a) and 1 CFR part 51, or another method equivalent in accuracy, 
precision, and sensitivity to A.O.C.S. Official Method Ca 3a-46. In 
response to the 2004 ANPRM, comments were submitted to the agency 
requesting that the primary method for the impurity determination for 
tallow be one other than the method in the Food Chemicals Codex. 
Comments stated that the domestic tallow industry primarily uses a 
method of AOCS to measure insoluble impurities. In comparison to the 
Food Chemicals Codex method, comments stated that the AOCS method is 
less expensive, requires less solvent, and has lower solvent disposal 
costs. In addition, it does not require specialized equipment or 
supplies. FDA agrees with these comments, and proposes that the primary 
method for the impurity determination for tallow be the method from 
AOCS rather than the method in the Food Chemicals Codex.
    This proposed requirement for tallow would apply to all animal 
feed, including feed for ruminants. Since the existing ruminant feed 
rule Sec.  589.2000 (21 CFR 589.2000) does not include provisions 
relative to tallow, this proposal represents a new requirement for 
ruminant feed as well as for feed for non-ruminants. To make clear that 
this proposed requirement would apply to ruminant feed, FDA is 
proposing to amend Sec.  589.2000 to include the tallow requirements.
    FDA is also proposing to exempt tallow derivatives from the 
requirements of this rulemaking. Tallow derivatives are produced by 
subjecting tallow to chemical processes (hydrolysis, 
transesterification, and saponification) that involve high temperature 
and pressure. FDA's TSEAC considered tallow derivatives in 1998 (Ref. 
24), and determined that the rigorous conditions of manufacture are 
sufficient to reduce the BSE risk in tallow derivatives to 
insignificant levels. In addition, according to OIE guidelines tallow 
derivatives produced by hydrolysis, saponification, or 
transesterification using high temperature and pressure can be safely 
traded regardless of the BSE risk status of the country of origin (Ref. 
23).

E. Disposal of Cattle Materials Prohibited in Animal Feed

    FDA agrees with comments from the affected industry that a 
comprehensive plan would be needed to safely dispose of approximately 
2.5 billion pounds of material if FDA decided to prohibit all dead 
stock and the full list of SRMs, as defined in the USDA interim final 
rule (69 FR 1862) and the FDA interim final rule (69 FR 42255), from 
being rendered for use in animal feed. The 2.5 billion pounds of cattle 
material includes approximately 1.4 billion pounds of material from 
cattle slaughtered for human consumption and 1.1 billion pounds of 
material from cattle not inspected and passed for human consumption 
that are currently being rendered for use in animal feed. FDA is 
concerned about the feasibility of establishing a new infrastructure to 
safely dispose of this large quantity of material, as well as the time 
it would take to implement these processes.
    Limiting the list of SRMs as proposed by this rule reduces the 
volume of slaughter byproducts that would require alternative disposal. 
First, this proposal does not require the diversion from use in animal 
feed the small intestine and tonsils from the 28 million head of cattle 
under 30 months of age that are slaughtered annually. Second, only the 
brain and spinal cord (weighing 1.3 pounds per animal) rather than the 
head, spinal column, and small intestine, (weighing 88.5 pounds per 
animal) are diverted from the estimated 7 million head of cattle over 
30 months of age that are slaughtered annually in the U.S. FDA believes 
that this more limited amount of material from slaughter operations can 
be disposed of through landfill, incineration, or alkaline digestion.
    Based on comments received, FDA acknowledges that there is some 
uncertainty regarding the amount of material that will require 
alternative disposal as a result of the proposed requirements 
pertaining to cattle not inspected and passed for human consumption 
(i.e., dead stock and nonambulatory disabled cattle). FDA is including 
in this proposed rule the option to remove brain and spinal cord from 
cattle not inspected and passed for human consumption so that most of 
this material could continue to be rendered for use in animal feed. As 
previously noted, FDA intends to continue exercising enforcement 
discretion for the use in animal feed of the remaining material from 
cattle that are diseased or that die other than by slaughter when the 
brain and spinal cord are removed. As discussed in more detail in 
Section IV, Analysis of Economic Impacts, FDA acknowledges that while 
the proposed rule will result in additional material from these animals 
being disposed of by means other than rendering, FDA believes such 
increases will be modest. FDA seeks comment and further information on 
the feasibility of removing brain and spinal cord from cattle not 
inspected and passed for human consumption and on the impact of this 
proposed rule on the number of these cattle that would be disposed of 
by rendering.
    In summary, FDA believes that the measures proposed by this rule 
can be more feasibly implemented than a full SRM ban, and can add 
substantially to the protection provided by the current BSE feed 
regulation. With this approach, the resulting volume of material 
requiring special disposal would be manageable in the short term. This 
approach is also consistent with the advice of the IRT that a staged 
approach may be necessary in implementation of an SRM ban. Further, FDA 
believes that other feed controls that FDA previously considered, such 
as dedicated facilities, are not needed if these high-risk tissues are 
excluded from animal feed channels. Therefore, at this time FDA is not 
proposing rulemaking to address other feed control recommendations of 
the IRT or the additional planned measures announced by FDA on January 
26, 2004.

III. Description of Proposed Rule and Legal Authority

    FDA is proposing to establish a new Sec.  589.2001 (21 CFR 
259.2001), Cattle materials prohibited in animal food or feed. While 
the existing Sec.  589.2000 outlines requirements related to ruminant 
feeds only, proposed Sec.  589.2001 outlines requirements intended to 
apply to food or feed for all animal species. The terms and 
requirements of proposed Sec.  589.2001 are described in section IV.A 
of this document.

A. Definitions

    The proposed Sec.  589.2001(a) defines the following terms for the 
purposes of this regulation:
    (1) Cattle materials prohibited in animal feed includes: (i) the 
brains and spinal cords of cattle 30 months of age and older; (ii) the 
brains and spinal cords of cattle of any age not inspected and passed 
for human consumption; (iii) the entire carcass of cattle not inspected 
and passed for human consumption from which brains and spinal cords 
were not removed; (iv) mechanically separated beef that is derived from 
cattle materials prohibited

[[Page 58581]]

under (i), (ii), or (iii) above; and (v) tallow that is derived from 
cattle materials prohibited under (i), (ii), or (iii) above. Tallow 
that is derived from cattle materials prohibited under (i), (ii), or 
(iii) above that contains no more than 0.15 percent insoluble 
impurities and tallow derivatives are not considered cattle materials 
prohibited in animal feeds.
    (2) Cattle not inspected and passed for human consumption means 
cattle of any age that were not inspected and passed for human 
consumption by the appropriate regulatory authority. This term includes 
nonambulatory disabled cattle. Non-ambulatory disabled cattle are 
cattle that cannot rise from a recumbent position or that cannot walk, 
including, but not limited to, those with broken appendages, severed 
tendons or ligaments, nerve paralysis, fractured vertebral column, or 
metabolic conditions.
    (3) Mechanically separated beef means a meat food product that is 
finely comminuted, resulting from the mechanical separation and removal 
of most of the bone from attached skeletal muscle of cattle carcasses 
and parts of carcasses.
    (4) Renderer means any firm or individual that processes slaughter 
byproducts, animals unfit for human consumption, or meat scraps. The 
term includes persons who collect such materials and subject them to 
minimal processing, or distribute them to firms other than renderers 
(as defined in paragraph (a)(1)) whose intended use for the products 
may include animal feed, industrial use, or other uses. The term 
includes renderers that also blend animal protein products.
    (5) Tallow means the rendered fat of cattle obtained by pressing or 
by applying any other extraction process to tissues derived directly 
from discrete adipose tissue masses or to other carcass parts and 
tissues.
    (6) Tallow derivative means any product obtained through initial 
hydrolysis, saponification, or trans-esterification of tallow; chemical 
conversion of material obtained by hydrolysis, saponification, or 
trans-esterification may be applied to obtain the desired product.

B. Proposed Requirements

    Proposed Sec.  589.2001(b)(1) provides that no animal food or feed 
shall be manufactured from, processed with, or otherwise contain cattle 
materials prohibited in animal feed. Proposed Sec.  589.2001(b)(2) 
provides new requirements for renderers that handle cattle material 
prohibited in animal feed. Proposed Sec.  589.2001(b)(3) provides new 
requirements for renderers that handle any cattle material.
1. Proposed Requirements for Renderers That Manufacture, Process, 
Blend, or Distribute Cattle Materials Prohibited in Animal Feed
    The proposed Sec.  589.2001(b)(2) requires that renderers that 
handle cattle materials prohibited in animal feed use separate 
equipment or containers to handle such material once it has been 
separated from other cattle materials. This requirement is intended to 
ensure that equipment used to manufacture, process, blend, store, or 
transport cattle materials prohibited in animal feed or products that 
contain or may contain cattle materials prohibited in animal feed do 
not serve as a source of cross-contamination for materials intended for 
animal feed. In addition, proposed Sec.  589.2001(b)(2) requires 
renderers that manufacture, process, blend, or distribute cattle 
materials prohibited in animal feed or products that contain or may 
contain cattle materials prohibited in animal feed must: (1) Label the 
prohibited materials in a conspicuous manner with the statement ``Do 
not feed to animals''; (2) mark the prohibited material with an agent 
that can be readily detected on visual inspection; and (3) establish 
and maintain records sufficient to track the prohibited materials to 
ensure such material is not introduced into animal feed, and make the 
records available for inspection and copying by FDA. These proposed 
requirements are intended to ensure that cattle materials prohibited in 
animal feed do not enter the animal feed chain and thus have no 
opportunity for inclusion in animal food or feed. FDA believes that 
such material must be both labeled and marked to ensure that it does 
not enter the feed channels since without such measures this material 
would be indistinguishable from cattle materials not prohibited by this 
proposed rule. Marking the material will provide a readily detectable 
method on visual examination by which all persons in the animal feed 
chain can be made aware that the a product is prohibited material or 
contains prohibited material. Marking also will serve as a way to make 
the status of the material known if, for some reason, the label ``Do 
not feed to animals'' is separated from the product.
2. Proposed Requirements for Renderers that Manufacture, Process, 
Blend, or Distribute Any Cattle Materials
    Proposed Sec.  589.2001(b)(3) requires that renderers that handle 
any cattle materials shall: (1) Establish and maintain records 
sufficient to demonstrate that material rendered for use in animal feed 
was not manufactured from, processed with, or does not otherwise 
contain, cattle materials prohibited in animal feed, (2) make copies of 
records available for inspection and copying by FDA, and (3) be in 
compliance with requirements under Sec.  589.2000 regarding animal 
proteins prohibited in ruminant feed.

C. Proposed Recordkeeping and Access Requirements

    The proposed recordkeeping requirements associated with this rule 
are focused on renderers because FDA believes this is the point at 
which cattle material prohibited in animal feed could enter the animal 
feed channel. Renderers, as defined in this proposed rule, receive 
cattle materials from slaughter facilities or receive entire cattle 
carcasses that were not inspected and passed for human consumption and 
further process that material so that it may be used in animal feed. 
FDA believes this is the critical control point in the feed and feed 
ingredient processing channel at which the exclusion of cattle material 
prohibited in animal feed must be documented. Once material is removed 
from cattle and further processed, we may not be able to obtain the 
information necessary to determine whether it is cattle material 
prohibited in animal feed. There is currently no way to reliably test 
feed or feed ingredients for the presence of the BSE agent or for the 
presence of cattle materials prohibited in animal feed.
    This proposed rule requires that no animal feed or feed ingredient 
be manufactured from, processed with, or otherwise contain cattle 
materials prohibited in animal feed. However, FDA does not believe it 
is necessary for persons, other than renderers, that are involved in 
the manufacture or processing of feed or feed ingredients to maintain 
records documenting the exclusion of cattle materials prohibited in 
animal feed. FDA believes, for the reasons cited previously, that it is 
critical that such records be maintained at the point of the renderer. 
However, FDA believes that requiring the maintenance of such records at 
all manufacturing and processing points downstream would be redundant 
and provide little additional information of value. FDA seeks comments 
on the need to require that records be maintained by persons other than 
renderers.
    Because at this time there is no way to test reliably for the 
presence of the BSE agent or the presence of the cattle materials 
prohibited in proposed Sec.  589.2001(b)(1), renderers must depend

[[Page 58582]]

on records to ensure that the materials prohibited by this proposed 
rule are excluded from material intended for use in animal feed and 
that such material is appropriately disposed. Similarly, without 
adequate records kept by renderers and access to the records by the 
agency, FDA may not know whether renderers have complied with the 
requirements. We are proposing in Sec.  589.2001(b)(2)(iv) that 
renderers that manufacture, process, blend, or distribute cattle 
materials prohibited in animal feed establish and maintain records 
sufficient to demonstrate that such material was not introduced into 
animal feed. Furthermore, we are proposing in Sec.  589.2001(b)(3)(i) 
that renderers that manufacture, process, blend, or distribute cattle 
materials establish and maintain records sufficient to demonstrate that 
material rendered for use in animal feed was not manufactured from, 
processed with, or does not otherwise contain, cattle materials 
prohibited in animal feed.
    Proposed Sec.  589.2001(d) requires that the records required by 
this proposed rule be maintained for a minimum of 1 year. The 1-year 
record retention period is consistent with the existing requirements 
for ruminant feeds in Sec.  589.2000(h). We believe that for the 
purposes of the recordkeeping requirements, 1 year is appropriate in 
light of the time that the products will be in the animal feed 
production and distribution systems. Extending the record retention 
period would have little practical value in determining the source of 
BSE in an animal. This is also considering the potentially long time 
period from ingestion of the BSE agent in feed to manifestation of 
clinical signs and lesions and the lack of a reliable estimate for the 
latency period.
    The proposed rule does not specify the types of records that would 
need to be maintained in order to comply with the recordkeeping 
requirements. The agency seeks comments on what type of records would 
be appropriate and whether further detail is needed in the regulation 
regarding specific record requirements such as the specific data 
elements that must be included in such records.

D. Conforming Changes to Sec.  589.2000--Animal Proteins Prohibited in 
Ruminant Feed

    The requirements related to tallow in the proposed Sec.  589.2001 
are intended to apply to all animal feed, including feed for ruminants. 
Since the existing ruminant feed rule (Sec.  589.2000) does not include 
provisions relative to tallow, this proposal represents a new 
requirement for ruminant feed as well as for feed for non-ruminants. 
Therefore, due to concerns about protein impurities present in tallow, 
FDA is proposing to amend Sec.  589.2000 to include tallow in the 
definition of ``protein derived from mammalian tissues'' and to add 
language that excludes from the definition of ``protein derived from 
mammalian tissues'' tallow containing no more than 0.15 percent 
insoluble impurities and tallow derivatives as specified in proposed 
Sec.  589.2001.

E. Legal Authority

    FDA is issuing this proposed regulation on animal feed under the 
food adulteration provisions in sections 402(a)(2)(C), (a)(3), (a)(4), 
(a)(5), 409, and 701(a) of the Federal Food, Drug, and Cosmetic Act 
(the act) (21 U.S.C. 342(a)(2)(C), (a)(3), (a)(4), (a)(5), 348, and 
371(a)). The term ``food'' is defined to include articles used for food 
``for man or other animals.'' See section 201 of the act (21 U.S.C. 
321(f)). We note that the material that would be prohibited under this 
proposed rule from use in animal feed continues to meet the definition 
of food. Therefore, this material would be adulterated or misbranded 
under the act based on violations of the proposed rule, as well as any 
animal feed or feed ingredients that were manufactured from, processed 
with, or otherwise contained, the prohibited material.
    Under section 402(a)(3) of the act, a food is deemed adulterated 
``if it consists in whole or in part of any filthy, putrid, or 
decomposed substance, or if it is otherwise unfit for food.'' 
``Otherwise unfit for food'' is an independent clause in section 
402(a)(3). The statute does not require that a food be filthy, putrid, 
or decomposed for it to be ``otherwise unfit for food.'' In FDA's 
interim final rule on the Use of Materials Derived from Cattle in Human 
Food and Cosmetics (69 FR 42256 at 42264), we concluded that a food can 
be ``otherwise unfit for food'' based on health risks, and sought 
comments on that interpretation. Because of the possibility of 
intentional or unintentional use of the materials that would prohibited 
under this proposed rule in ruminant feed and the risk of BSE to 
ruminants and humans from these materials, we have tentatively 
concluded that these materials would be ``otherwise unfit for food'' 
under section 402(a)(3) of the act. We seek comment on this 
interpretation.
    Under section 402(a)(4) of the act, a food is deemed adulterated 
``if it has been prepared, packed, or held under insanitary conditions 
whereby it may have become contaminated with filth, or whereby it may 
have been rendered injurious to health.'' The failure to ensure that 
animal feed is prepared, packed, or held under conditions in which 
cattle materials prohibited in animal feed under this proposed rule do 
not contaminate animal feed would constitute an insanitary condition 
whereby the feed may have been rendered injurious to health. Thus, this 
insanitary condition would render the animal feed adulterated under 
section 402(a)(4) of the act.
    Under section 402(a)(5) of the act, food is deemed adulterated ``if 
it is, in whole or in part, the product * * * of an animal which has 
died otherwise than by slaughter.'' Some cattle are not inspected and 
passed because they are diseased or have died before slaughter. 
Material from these cattle that are diseased or that die otherwise than 
by slaughter that is used as animal feed would render that feed 
adulterated under section 402(a)(5) of the Act. FDA has traditionally 
exercised enforcement discretion with regard to the use of such animals 
in animal feed. For example, see Compliance Policy Guide 675.400. FDA 
intends to continue exercising such discretion for the use in animal 
feed of the remaining material from cattle that are diseased or that 
die other than by slaughter when the brain and spinal cord are removed.
    We are also relying on the adulteration provision in section 
402(a)(2)(C)(i) of the act. Section 402(a)(2)(C)(i) deems a food 
adulterated if it is or bears or contains a food additive that is 
unsafe under section 409 of the act. Section 201(s) of the act, (21 
U.S.C. 321(s)), defines as a food additive any substance whose intended 
use results or may reasonably be expected to result in it becoming a 
component of food unless, among other things, it is the subject of a 
prior sanction (explicit approval for a specific use by USDA or FDA 
before September 6, 1958), or is generally recognized as safe (GRAS). 
Section 409 of the act provides that a food additive is unsafe unless 
it and its use conform to a food additive regulation or an exemption 
under section 409(j).
    Prior sanctions are described in part 570 (21 CFR part 570). FDA is 
not aware of any prior sanctions that relate to the present animal feed 
use of the cattle material that would be prohibited in animal feed 
under this proposed rule. Any person who intends to assert or rely on 
such sanction is required to submit proof of the existence of the 
applicable prior sanction. The failure of any person to come forward 
with proof of such an applicable prior sanction in response to this 
notice will constitute a waiver of

[[Page 58583]]

the right to assert or rely on such sanction at any later time.
    A determination that a substance added directly or indirectly to a 
food is GRAS, for its intended use is generally based on scientific 
information regarding the composition of the substance, its use, method 
of preparation, methods for detecting its presence in food, and 
information about its functionality in food as determined by experts 
qualified by scientific training and experience to evaluate the safety 
of such a substance (Sec.  570.30). A substance added to food becomes 
GRAS as a result of a common understanding about the substance 
throughout the scientific community familiar with the safety of such 
substances. The basis of expert views may be either scientific 
procedures, or, in the case of a substance used in food before January 
1, 1958, experience based on common use in food (Sec.  570.30(a)). 
Substances that are GRAS based on use before January 1, 1958, must be 
currently recognized as safe based on their pre-1958 use (See United 
States v. Naremco, 553 F. 2d 1138 (8th Cir. 1977; compare United States 
v. Western Serum, 666 F. 2d 335 (9th Cir. 1982)).
    General recognition of safety based upon scientific procedures 
requires the same quantity and quality of scientific evidence as is 
required to obtain approval of a food additive regulation for the 
ingredient (21 CFR 570.30(b)). (See United States v. Naremco, 553 F.2d 
at 1143). A substance is not GRAS if there is a genuine dispute among 
experts as to its recognition (An Article of Drug * * * Furestrol 
Vaginal Suppositories, 294 F. Supp 1307 (N.D. Ga. 1968), aff'd 415 F.2d 
390 (5th Cir. 1969)). It is not enough, in attempting to establish that 
a substance is GRAS, to establish that there is an absence of 
scientific studies that demonstrate the substance to be unsafe; there 
must be studies that show the substance to be safe (United States v. An 
Article of Food * * * CoCo Rico, 752 F.2d 11 (1st Cir. 1985)). 
Conversely, a substance may be ineligible for GRAS status if studies 
show that the substance is, or may be, unsafe, or if there is a 
conflict in studies.
    Expert opinion that cattle materials that would be prohibited in 
animal feed under this proposed rule are GRAS would need to be 
supported by scientific literature and other sources of data and 
information, establishing that there is a reasonable certainty of no 
harm from the material under the intended conditions of use. Expert 
opinion would need to address topics such as whether BSE infectivity 
can be detected, and whether it is reasonably certain that the BSE 
agent will not be transmitted through cattle materials that would 
prohibited in animal feed under this proposed rule. The burden of 
establishing that a substance is GRAS is on the proponent of the 
substance. (See CoCo Rico, supra.)
    For the reasons discussed in other sections of this document, the 
agency is tentatively concluding that cattle materials prohibited in 
animal feed under this proposed rule are not GRAS by qualified experts 
for use in animal food and, therefore, would be food additives. Section 
402(a)(2)(C)(i) and (ii) of the act deems food adulterated ``if it is 
or it bears or contains any food additive which is unsafe within the 
meaning of section 409 * * * .'' Under section 409(a), a food additive 
is unsafe unless a food additive regulation or an exemption is in 
effect with respect to its use or its intended use. Therefore, in the 
absence of a food additive regulation or an exemption, the cattle 
materials that would be prohibited in animal feed under this proposed 
rule would be adulterated under section 402(a)(2)(C)(i) of the act 
because it bears or contains an unsafe food additive, and their 
presence in animal feed would render the food adulterated.
    Under section 701(a) of the act, FDA is authorized to issue 
regulations for the act's efficient enforcement. The proposed 
regulation would require measures to prevent animal food from being 
unfit for food, being or bearing an unsafe food additive, being the 
product of an animal that died otherwise than by slaughter. The 
measures will also be required to prevent animal food from being held 
under insanitary conditions whereby it may have been rendered injurious 
to health. These proposed measures would allow for the efficient 
enforcement of the act. Under the proposed regulations, renderers would 
be required to establish and maintain records to track cattle materials 
prohibited in animal feed to ensure that such material is not 
introduced into animal feed and make such records available to FDA for 
inspection and copying. Once material is removed from cattle, we may 
not be able to obtain the information necessary to determine whether it 
is prohibited cattle material. Because at this time there is no way to 
test reliably for the presence of the BSE agent or the presence of the 
cattle materials prohibited in proposed Sec.  589.2001(b)(1), renderers 
must depend on records to ensure that their products do not contain 
cattle materials prohibited from animal feed. In addition, without 
adequate records, FDA cannot know whether renderers have complied with 
the regulations that prohibit the use of certain cattle material in 
rendered products intended for animal feed. For example, we would not 
know from examination of a spinal cord whether the source animal was 
over 30 months of age at the time of slaughter or whether the cattle 
had been inspected and passed. Therefore, the proposed recordkeeping 
and records access requirements are necessary for the efficient 
enforcement of the proposed rule. Under the proposed rule, failure to 
comply with the recordkeeping and records access requirements would 
render the cattle material and any animal feed manufactured from, 
processed with, or otherwise containing, the cattle material 
adulterated under section 402(a)(4) of the act.
    Furthermore, the proposed marking provision in Sec.  589.2001 is 
necessary for the efficient enforcement of the act. Because there is 
currently no reliable method to determine which cattle materials would 
be the prohibited materials, marking is necessary to ensure compliance 
with the proposed requirement that animal feed is not manufactured 
from, processed with, or otherwise contains the prohibited cattle 
materials. Under the proposed rule, failure to comply with this marking 
requirement would render the cattle material and any animal feed 
manufactured from, processed with, or otherwise containing, the cattle 
material adulterated under section 402(a)(4) of the act.
    FDA is issuing the proposed labeling requirement under sections 
403(a)(1) and 201(n) of the act (21 U.S.C. 343(a)(1)). Section 
403(a)(1) provides that a food is deemed misbranded if its labeling is 
false or misleading in any particular. Section 201(n) provides that:
     * * * in determining whether the labeling of a product is 
misleading, there shall be taken into account (among other things) 
not only representations made or suggested by statement, word, 
design, device, or any combination thereof, but also the extent to 
which the labeling * * * fails to reveal facts material in light of 
such representations or material with respect to consequences which 
may result from the use of the article to which the labeling * * * 
relates under conditions of use prescribed in the labeling * * * or 
under such conditions of use as are customary or usual.
    The proposed rule would require cattle material prohibited in 
animal feed to be labeled ``Do not feed to animals.'' We believe this 
statement is material with respect to the consequences that may result 
from the use of this material within the meaning of section 201(n) of 
the act. As discussed in other sections of this document, the use of 
the material

[[Page 58584]]

that would be prohibited under this proposed rule presents a risk of 
BSE. Furthermore, there are no available definitive tests to detect 
this material in feed. Therefore, under this proposed rule, the failure 
to include this labeling statement would render the cattle material or 
feed containing the prohibited cattle material misbranded under section 
403(a)(1) of the act. We are also proposing that such statement be made 
in a conspicuous manner. Under section 403(f) of the act, (21 U.S.C. 
343(f)), a food is misbranded if ``any word, statement, or other 
information required by or under authority of this Act to appear on the 
label or labeling is not prominently placed thereon with such 
conspicuousness * * * and in such terms as to render it likely to be 
read and understood by the ordinary individual under customary 
condition of purchase and use.'' Therefore, under the proposed rule, 
the failure to include the statement ``Do not feed to animals'' in a 
conspicuous manner would render the cattle materials or any feed 
containing the cattle materials misbranded under section 403(f) of the 
act.

IV. Analysis of Economic Impacts

    FDA has examined the impacts of the proposed rule under Executive 
Order 12866, the Regulatory Flexibility Act (5 U.S.C. 601-612), and the 
Unfunded Mandates Reform Act of 1995 (Public Law 104-4). Executive 
Order 12866 directs agencies to assess all costs and benefits of 
available regulatory alternatives and, when regulation is necessary, to 
select regulatory approaches that maximize net benefits (including 
potential economic, environmental, public health and safety, and other 
advantages; distributive impacts, and equity).
    Section 202(a) of the Unfunded Mandates Reform Act of 1995 requires 
that agencies prepare a written statement, which includes an assessment 
of anticipated costs and benefits, before proposing ``any rule that 
includes any Federal mandate that may result in the expenditure by 
State, local, and tribal governments, in the aggregate, or by the 
private sector, of $100,000,000 or more (adjusted annually for 
inflation) in any one year.'' The current threshold after adjustment 
for inflation is $115 million, using the most current (2003) Implicit 
Price Deflator for the Gross Domestic Product. FDA does not expect this 
proposed rule to result in any 1-year expenditure that would meet or 
exceed this amount.
    FDA tentatively finds that the proposed rule does not constitute an 
economically significant regulatory action as defined in Section 
3(f)(1) of Executive Order 12866. We base this conclusion on both a 
study of the impacts on industry of the proposed rule (on file at the 
Division of Dockets Management (see ADDRESSES) conducted for FDA by the 
Eastern Research Group (ERG)), a private consulting firm, and the 
discussion in the remainder of this section (Ref. 25). The agency has 
further tentatively determined that the proposed rule may have a 
significant impact on a substantial number of small entities. This 
proposed rule imposes no mandates on government entities, and would not 
be expected to require the expenditure of over $115 million in any 1 
year by the private sector. As such, no further analysis of anticipated 
costs and benefits is required by the Unfunded Mandates Reform Act.
    The following regulatory impact analysis begins with a summary of 
the proposed rule and the expected benefits and costs. Next, in section 
V.B of this document, we discuss the need for the regulation. In 
section V.C, we discuss the benefits of the proposed rule, while in 
section V.D, we discuss the costs. In section V.E, we discuss the costs 
to the government. Finally, in section V.F, we discuss the regulatory 
flexibility analysis.

A. Summary of Proposed Regulatory Impact Analysis

    The proposed regulation would prohibit the use of certain cattle 
materials in any animal feed. The cattle materials prohibited in animal 
feed (CMPAF) would include the brain and spinal cord of all cattle 30 
months of age or older, as well as the brain and spinal cord of cattle 
not inspected and passed for human consumption regardless of age, the 
entire carcass of cattle not inspected and passed if brain and spinal 
cord is not removed (again, regardless of age), as well as other 
materials. For the purposes of this proposed rule, the term ``cattle 
not inspected and passed for human consumption'' includes nonambulatory 
disabled cattle. Tallow derived from any of the prohibited materials 
named previously would also be banned from use in animal feed unless it 
contains no more than 0.15 percent insoluble impurities. MS beef from 
any of the prohibited materials named above would be prohibited from 
use in animal feed. Additional provisions of the proposed rule would 
require renderers that handle cattle materials prohibited in animal 
feed to use separate equipment or containers to handle this material 
once it has been separated from other cattle materials. Such renderers 
will also be required to follow certain procedures for labeling and 
marking prohibited material and recordkeeping and records access.
    The benefits of the proposed rule include the elimination of the 
vast majority of the risk of spreading BSE to other cattle from 
intentional or unintentional use of non-ruminant feed for ruminants or 
cross-contamination of ruminant feed with non-ruminant feed or 
ingredients intended for non-ruminant feed. FDA believes that the 
proposed rule would effectively remove from use in non-ruminant feeds 
those cattle tissues that account for approximately 90 percent of 
potential BSE infectivity. Although the animal and public health 
benefit associated with the additional BSE risk reduction is paramount, 
the U.S. economy may also benefit from increased exports to the extent 
that the proposed rule, if finalized, persuades foreign governments 
that U.S. beef products are safe to import. Although we are unable to 
quantify these benefits, they are potentially large, because the 
expected loss of exports from the discovery of one infected cow in 
Washington State in December 2003 amounted to approximately $3.4 
billion in the first year (Ref. 26).
    The total costs to industry of complying with the proposed rule 
range from roughly $14 million to $24 million per year annualized over 
10 years assuming a 7-percent discount rate (at a 3-percent discount 
rate, the cost would range from $14 million to $23 million). These 
estimated costs are the sum of the costs including: (1) The ban on the 
use of certain tissues from cattle 30 months of age or older and cattle 
not inspected and passed for human consumption in any animal feed and 
(2) feed substitution costs. We discuss the proposed brain and spinal 
cord prohibitions as direct costs to the affected firms (including 
disposal costs, where applicable) and the firms' lost revenues from the 
ban on these raw materials used in feed product inputs. Then, we 
discuss the costs incurred by feed substitution costs. Table 1 of this 
document shows a summary of these costs.
    The proposed ban on the use of certain tissues from cattle 30 
months of age or older and cattle not inspected and passed for human 
consumption in any animal feed would require slaughterers and renderers 
that process cattle 30 months of age or older and firms that process 
dead, down, disabled, and diseased cattle to separate the CMPAF from 
the remaining cattle offal that could still be used for animal feed. We 
estimate that, for slaughterers, the separation of these materials from 
cattle

[[Page 58585]]

30 months of age or older and cattle not inspected and passed for human 
consumption regardless of age would require about $555,000 in one-time 
capital costs (or $79,000 annualized at 7 percent and $65,000 
annualized at 3 percent, over 10 years) (see table 1 of this document). 
We estimate that the annual cost of the additional labor to separate 
these CMPAF from other cattle offal is estimated to cost about $597,000 
annually. Although compliance costs for these activities would be borne 
initially by slaughterers, and are presented as such by ERG, a portion 
of the costs are likely to be passed along to cattle producers and 
consumers. For renderers, capital investments and labor for separation 
and segregation of CMPAF would range from about $1.88 million to $4.65 
million annually.
    Our analysis does not project a specific disposal route for CMPAF 
due to the uncertainty inherent in disposing of such low volumes of 
material. Instead, it describes various disposal methods that may be 
employed and estimated a $12 per 100 lbs. (cwt) of CMPAF disposal cost 
(including transportation costs) for the low-cost end of the range of 
disposal methods. The cost to dispose of the CMPAF is estimated to 
range from $7.72 million to $9.97 million annually. Additional on-farm 
disposal of dead and nonambulatory disabled cattle is expected to 
increase compliance costs from about $1.02 million to $2.53 million 
annually (including labor and equipment). The annual revenues foregone 
from meat and bone meal (MBM) sales due to the prohibition of CMPAF in 
animal feeds are estimated at $1.41 million to $2.78 million, and 
foregone tallow sales are estimated at $1.37 million to $2.62 million. 
This includes the value from CMPAF from cattle 30 months of age or 
older and cattle not inspected and passed for human consumption 
regardless of age, as well as from whole carcasses of cattle not 
inspected and passed for human consumption that could not be rendered 
due to this proposed rule.
    We considered including a provision in this proposed rule that 
would limit the use of all tallow in animal feed to that which contains 
no more than 0.15 percent insoluble impurities, not just tallow derived 
from the materials proposed to be prohibited in animal feed that 
contains no more than 0.15 percent insoluble impurities. Analysis of 
this alternative concluded that it would result in annualized costs of 
about $1.78 million. These costs would consist of capital and operating 
costs for polishing centrifuges that would be needed by a small segment 
of independent renderers. We have not included a provision requiring 
that all tallow meet the 0.15 percent limit in the proposal because the 
CMPAF ban would effectively negate the risk of infectivity in non-
CMPAF-derived tallow. We invite public comments and data on the need 
for, and impacts of, a provision that would require all tallow used in 
animal feeds meet the 0.15 percent limit.

                   Table 1.--Total Costs ($ Millions)
------------------------------------------------------------------------
                                                           Annualized
       Cost Item         One-Time Cost   Annual Costs       Costs\1\
------------------------------------------------------------------------
Slaughter Facilities    ..............  ..............  ................
------------------------------------------------------------------------
Capital Investments     $0.56           N/A             $0.08
------------------------------------------------------------------------
Labor                   ..............  $0.60           $0.60
------------------------------------------------------------------------
Lost Value of MBM       ..............  $1.41--$2.76    $1.41--$2.78
 (cattle 30 months of
 age or older, cattle
 not inspected and
 passed)
------------------------------------------------------------------------
Lost Value of Tallow    ..............  $1.37--$2.62    $1.37--$2.62
 (cattle 30 months of
 age or older, cattle
 not inspected and
 passed)
------------------------------------------------------------------------
Disposal of cattle not  ..............  ..............  ................
 inspected and passed
------------------------------------------------------------------------
Labor                   ..............  $0.12--$0.29    $0.12--$0.29
------------------------------------------------------------------------
Equipment               ..............  $0.9--$2.23     $0.9--$2.23
------------------------------------------------------------------------
Renderer Facilities     ..............  ..............  ................
------------------------------------------------------------------------
Capital Investments     $3.11-$7.67     $0.04--$0.11    $0.49--$1.20
------------------------------------------------------------------------
Labor                   ..............   $1.40--$3.45   $1.40--$3.45
------------------------------------------------------------------------
Disposal of CMPAF from  ..............  $7.72--$9.97    $7.72--$9.97
 cattle 30 months of
 age or older, cattle
 not inspected and
 passed
------------------------------------------------------------------------
CMPAF Marking (High     ..............  $0.01           $0.01
 Estimate)
------------------------------------------------------------------------
Recordkeeping/Labeling   $0.10           $0.05           $0.06
------------------------------------------------------------------------
Feed Substitution       ..............  $0.30--$0.46     $0.30--$0.46
------------------------------------------------------------------------
Proposed Rule Total      $3.76           $13.91--$22.5   $14.44--$23.75
 Costs                                   6
------------------------------------------------------------------------
\1\ Annualized costs equal to annual costs plus one-time costs at 7
  percent over 10 years. Using a 3 percent rate, annualized costs equal
  $23,535,000.

    FDA believes that this proposal, when evaluated in terms of its 
incremental cost-effectiveness at reducing risks from BSE, is more 
consistent with efficient science-based risk management than other 
regulatory approaches that it identified in the 2004 ANPRM. This 
proposal limits use of animal tissues for which infectivity is high 
relative to

[[Page 58586]]

tissue weight. Weight is a key determinant of the incremental costs 
from excluding tissues from rendering for animal feed. The approach 
adopted in this proposal is likely to be relatively cost-effective 
because it is directed primarily at those tissues for which infectivity 
is likely to be high relative to control compliance costs.
    In the 2004 ANPRM, FDA stated it was considering prohibiting a 
larger list of cattle tissues (the full SRM list) from use in all 
animal feeds. Under this option, SRMs would be defined as the skull, 
brain, eyes, spinal cord, trigeminal ganglia, vertebral column 
(excluding the vertebrae of the tail, the transverse processes of the 
thoracic and lumbar vertebrae, and the wings of the sacrum) and dorsal 
root ganglia of all cattle over 30 months of age or older, including 
the tonsils and distal ileum of all cattle regardless of age. 
Additionally, this option would prohibit the small intestine of all 
cattle, all material from nonambulatory disabled cattle, all material 
from cattle that are not inspected and passed for human consumption, 
and MS beef. Lastly, tallow derived from other prohibited materials and 
containing more than 0.15 percent insoluble impurities would also be 
prohibited from use in all animal feeds under this SRM option. As 
detailed later in the analysis of alternatives, we have not included 
all of these measures in this proposed rule because we believe the 
proposed rule adequately addresses the risk from the presence of the 
highest risk cattle material in the animal feed chain. We also note 
that the proposed rule offers a more cost-effective approach to 
achieving nearly the same level of protection against the spread of BSE 
with regard to the presence of high-risk material in the non-ruminant 
feed supply.
    The approach described in the 2004 ANPRM is itself a refinement of 
an approach announced early in 2004. In January 2004, shortly after 
USDA reported finding a BSE-infected cow in Washington State, HHS 
announced its intention to amend the current animal feed regulations by 
adding several materials to the list of substances prohibited from use 
in ruminant feed (Ref. 27). These materials included mammalian blood 
and blood products; inspected meat products that have been cooked, 
offered for human food, and then further heat-processed for feed (such 
as plate waste and used cellulosic casings); and poultry litter. 
Further, FDA planned to require establishments that manufacture, 
process, blend, or distribute both products containing mammalian-
derived proteins and ruminant feed to use separate equipment or 
facilities in their manufacture, processing and handling.
    Preliminary analysis of the regulatory approach described in the 
January 2004 announcement (Ref. 27) suggests that it is relatively less 
effective in risk reduction compared to the CMPAF and SRM bans because 
it would not remove the highest risk tissue (brain and spinal cord) 
from animal feed channels. Instead, the approach described in the 
January 2004 announcement would continue to allow the highest risk 
cattle material in non-ruminant feed, but includes measures intended to 
prevent cross-contamination of ruminant feed. Although we have not been 
able to quantify the risk reduction associated with the approach 
announced in January 2004, it is comparable in costs to the full SRM 
ban described in the 2004 ANPRM. As a result we are not proposing it 
here.
    In developing this proposed rule we also considered other 
alternatives (not included here), including combinations of bans of 
various cattle tissues, from cattle of various ages (>30 months and <30 
months) and various states (slaughtered for human food, deads, 
downers). All of these resulted in costs over $100 million per year 
with potential infective tissue reductions between 80 percent and 99 
percent, when compared to the base case scenario.
    Table 2 of this document lists the proposed rule (the CMPAF ban), 
the SRM ban, and one of the options mentioned previously, namely a ban 
on brain and spinal cord from slaughter cattle 30 months of age or 
older, and a ban on the entire carcass of all dead and downed cattle. 
The table lists both the expected costs of these options, and our best 
estimate of the percent reduction in cattle tissues known to harbor BSE 
infectivity. The proposed rule would reduce cattle oral ID50s (the 
amount of infective material that would result in a case of BSE in 50 
percent of the cattle that consumed it) that are available for use in 
animal feed by about 90 percent as much as a ban on the full list of 
SRMs (option 3), while imposing only 7 to 10 percent of the costs of 
the SRM option (0.07 = $14 million/$195 million; 0.10 = $24 million/
$240 million). The second option would reduce the cattle oral ID50s by 
more than 90 percent (a less than 10 percent increase over option 1), 
but would impose costs that are about five to nine times greater than 
option 1, though still only about 50 percent to 70 percent of the costs 
of option 3. Based on the level of protection provided against the 
spread of BSE and its cost-effectiveness, we believe the proposed rule 
to be the most appropriate. FDA seeks further comment and scientific 
and risk information on this analysis of additional regulatory options 
for strengthening animal feed safeguards.

          Table 2.--Cost-Effectiveness of Alternative Policies
------------------------------------------------------------------------
Option (Description of Banned       Infectivity         Annual Cost ($
      Tissues/Materials)           Reduction\1\           millions)
------------------------------------------------------------------------
CMPAF list from (1) Cattle 30   90%                   $14--$24
 months or older, (2) deads,
 (3) downers and (4), MS beef
 if CMPAF not removed from
 carcass, dedicated equipment/
 container requirement;
 tallow restriction (proposed
 rule)
------------------------------------------------------------------------
Brain and spinal cord from      >90%                  $115--$135\2\
 cattle 30 months or older,
 carcasses of all deads and
 downs, MS Beef
------------------------------------------------------------------------
Full SRM list from cattle 30    >99%                  $195--$240
 months or older, tonsils and
 distal ileum from cattle of
 all ages, carcass of all
 deads and downers, MS beef,
 tallow restriction
------------------------------------------------------------------------
\1\ Percent of ID50s from an infected animal that would be banned from
  use in animal feed.
\2\ Detailed cost estimate of this alternative is not included in the
  regulatory flexibility analysis section of this document.


[[Page 58587]]

B. Need for Regulation

    Executive Order 12866 directs agencies to assess the need for any 
significant regulatory action and an explanation of how the regulation 
will meet that need. In this instance, FDA tentatively concludes that 
private incentive systems for both suppliers and purchasers in markets 
for cattle, rendering, and ruminant feed