[Federal Register: July 25, 1996 (Volume 61, Number 144)]
[Rules and Regulations]               
[Page 38805-38855]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr25jy96-19]
 

[[Page 38805]]


_______________________________________________________________________

Part II





Department of Agriculture





_______________________________________________________________________



Food Safety and Inspection Service



_______________________________________________________________________



9 CFR Part 304, et al.



Pathogen Reduction; Hazard Analysis and Critical Control Point (HACCP) 
Systems; Final Rule


[[Page 38806]]



DEPARTMENT OF AGRICULTURE

Food Safety and Inspection Service

9 CFR Parts 304, 308, 310, 320, 327, 381, 416, and 417

[Docket No. 93-016F]
RIN 0583-AB69

 
Pathogen Reduction; Hazard Analysis and Critical Control Point 
(HACCP) Systems

AGENCY: Food Safety and Inspection Service, USDA.

ACTION: Final rule with request for comments.

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

SUMMARY: The Food Safety and Inspection Service (FSIS) is establishing 
requirements applicable to meat and poultry establishments designed to 
reduce the occurrence and numbers of pathogenic microorganisms on meat 
and poultry products, reduce the incidence of foodborne illness 
associated with the consumption of those products and provide a new 
framework for modernization of the current system of meat and poultry 
inspection. The new regulations (1) require that each establishment 
develop and implement written sanitation standard operating procedures 
(Sanitation SOP's); (2) require regular microbial testing by slaughter 
establishments to verify the adequacy of the establishments' process 
controls for the prevention and removal of fecal contamination and 
associated bacteria; (3) establish pathogen reduction performance 
standards for Salmonella that slaughter establishments and 
establishments producing raw ground products must meet; and (4) require 
that all meat and poultry establishments develop and implement a system 
of preventive controls designed to improve the safety of their 
products, known as HACCP (Hazard Analysis and Critical Control Points).

DATES: Effective Date: July 25, 1996, however these rules are not 
applicable until the dates listed below.
    Applicability dates: (1) The HACCP regulations set forth in 9 CFR 
Part 417 and related provisions set forth in 9 CFR 304, 327, and 381 
parts will be applicable as follows:
    <bullet> In large establishments, defined as all establishments 
with 500 or more employees, on January 26, 1998.
    <bullet> In smaller establishments, defined as all establishments 
with 10 or more employees but fewer than 500, on January 25, 1999.
    <bullet> In very small establishments, defined as all 
establishments with fewer than 10 employees or annual sales of less 
than $2.5 million, on January 25, 2000.
    (2) The Sanitation SOP's regulations set forth in 9 CFR 416 will be 
applicable on January 27, 1997.
    (3) The E. coli process control testing regulations set forth in 9 
CFR 310.25(a) and 381.94(a) will be applicable on January 27, 1997.
    (4) The Salmonella pathogen reduction performance standards 
regulations set forth in 9 CFR 310.25(b) and 9 CFR 381.94(b) will be 
applicable simultaneously with applicability dates for implementation 
of HACCP.
    Comments: Comments on specified technical aspects of the final 
regulations must be received on or before September 23, 1996. With 
respect to the HACCP final regulations, FSIS requests comments by 
November 22, 1996.

ADDRESSES: Submit one original and two copies of written comments to: 
FSIS Docket Clerk, DOCKET #93-016F, U.S. Department of Agriculture, 
Food Safety and Inspection Service, Room 4352, 1400 Independence 
Avenue, S.W., Washington, DC 20250-3700. All comments submitted on this 
rule will be available for public inspection in the Docket Clerk's 
Office between 8:30 a.m. and 1:00 p.m., and 2:00 p.m. and 4:30 p.m., 
Monday through Friday. The references and baseline surveys cited in 
this document are available for inspection in the FSIS Docket Room.

FOR FURTHER INFORMATION CONTACT: (1) GENERAL: Dr. Judith A. Segal, 
Director, Policy, Evaluation, and Planning Staff, (202) 720-7773; (2) 
MICROBIAL TESTING: Patricia F. Stolfa, Acting Deputy Administrator, 
Science and Technology, (202) 205-0699.

SUPPLEMENTARY INFORMATION:

Obtaining Copies of This Document:

    An electronic version of this document is available on the Internet 
from the Federal Register at www.access.gpo.gov/su__docs/aces/
aces140.html. Paper or diskette copies of this document may be ordered 
from the National Technical Information Service (NTIS), U.S. Department 
of Commerce, 5285 Port Royal Road, Springfield, VA 22161. For a 
complete copy of this document orders must reference NTIS accession 
number PB96-177613 (paper copy) and PB96-502166 (disk copy). For a copy 
of the preamble and rule, the individual appendices, and the impact 
assessment reference the following NTIS accession numbers: PB96-177621 
(preamble and rule only), PB96-177639 (Appendix A), PB96-177647 
(Appendix B), PB96-177654 (Appendix C), PB96-177662 (Appendix D), PB96-
177670 (Appendix E), PB96-177688 (Appendix F), PB96-177696 (Appendix 
G), and PB96-177704 (impact assessment). For telephone orders or more 
information on placing an order, call NTIS at (703) 487-4650 for 
regular service or (800) 553-NTIS for rush service. Dial (703) 321-8020 
with a modem or Telnet fedworld.gov to access this document 
electronically for ordering and downloading via FedWorld. For technical 
assistance to access FedWorld, call (703) 487-4608.

Table of Contents

I. Background
    Overview of FSIS Food Safety Goal and Strategy
    FSIS Regulatory Proposals
    FSIS Regulatory and Inspection Reform Plans
    Change Within FSIS
    The FSIS Pathogen Reduction/HACCP Rulemaking Process
    Seven Information Briefings
    Three Scientific and Technical Conferences
    Public Hearing
    Federal-State Relations Conference
    Scoping Session and Six Issue-Focused Meetings
    Food Safety Forum
    Farm-to-Table Strategy
    General Overview of the Comments and the Final Rule
    HACCP and Performance Standards
    Sanitation SOP's, Antimicrobial Treatments, and Cooling 
Requirements for Raw Meat and Poultry Products
    Timetable for Implementation
    Federally Inspected Establishments
    State-inspected Establishments
    Foreign-inspected Establishments
    Implementation Conferences
    Request for Comments
II. Hazard Analysis and Critical Control Point Systems
    Overview of Final Rule
    History and Background of HACCP
    The Seven HACCP Principles
    HACCP and the FSIS Food Safety Strategy
    Preparing for HACCP Implementation
    Inspection under HACCP
    Implementation Schedule
    Small Business Issues
    Training Considerations
    Mandatory versus Voluntary HACCP
    HACCP from Farm-to-Table
    Total Quality Control (TQC) Establishments and HACCP
    Freedom of Information Act Concerns
    FSIS Enforcement Authority and Whistleblower Protection
    Enforcement and Due Process
    The Final Rule
    Reorganization of HACCP Regulatory Text
    HACCP Systems as a Condition of Receiving Inspection
    Definitions
    Hazard Analysis and HACCP Plan
    Corrective Actions
    Validation, Verification, and Reassessment
    Reassessment
    FSIS Verification
    Records

[[Page 38807]]

    Training
    Adequacy of HACCP Plans
III. Sanitation Standard Operating Procedures
    The Proposed Rule
    The Final Rule
    Comments and Responses
    General
    Development of Sanitation SOP's
    Maintaining Sanitation SOP's
    Recordkeeping
    ``Layering''
    Role of Inspectors
    Relation to HACCP
    Training
    Pre-operation Sanitation Inspection
    Implementation Date
IV. Microbiological Performance Criteria and Standards
    Summary of Proposal
    Role of Microbiological Performance Criteria and Standards in 
FSIS Food Safety Strategy
    Overview of Final Rule
    Process Control Verification Performance Criteria
    Pathogen Reduction Performance Standards
    Process Control Verification: E. coli Performance Criteria and 
Testing
    Rationale for Using E. coli Tests to Verify Process Control
    Use of Baseline Values to Establish E. coli Performance Criteria
    Establishment of E. coli Performance Criteria to Verify Process 
Control
    Sampling Frequency for E. coli Testing
    Sampling and Analytical Methodology
    Recordkeeping
    Use of E. coli Test Results by Establishments
    Use of E. coli Test Results by FSIS
    Implementation Timetable
    Request for Comments
    Pathogen Reduction Performance Standards
    Rationale for Selecting Salmonella
    Basis for Performance Standards and Plans for Future Adjustments
    Determining Compliance with the Standard
    FSIS Testing Strategy
    FSIS Testing Methods
    FSIS Enforcement Strategy
    Implementation Timetable for Pathogen Reduction Performance 
Standards
    Response to Comments
    The Indicator Organism
    Frequency and Cost of Testing
    Legal Authority for Testing Requirement
    Performance Standards for Process Control
    Basis for Target Levels
    Methodology for Meeting Targets
    Sample Size
    Testing Methodology
    Role of Inspectors
    Laboratories
    Alternative Sampling under HACCP
    Relationship to HACCP
V. Other Issues and Initiatives
    Antimicrobial Treatments
    Cooling and Chilling Requirements for Raw Meat and Poultry
    International Trade
    Recordkeeping and Record Retention
    Finished Product Standards for Poultry Carcasses
VI. Economic Impact Analysis and Executive Orders
    Executive Order 12866
    HACCP-based Regulatory Program Produces Net Benefit to Society
    Market Failure Justifies Regulation of Pathogens
    Regulatory Alternatives
    Unfunded Mandates Reform Act
    Regulatory Flexibility Act
    Executive Order 12778
    Paperwork Requirements
    Sanitation Standard Operating Procedures (Sanitation SOP's)
    Time and Temperature
    Microbiological Testing
    HACCP
VII. Final Rules
VIII. Appendix A--Guidelines for Developing a Standard Operating 
Procedure for Sanitation (Sanitation SOP's) in Federally Inspected 
Meat and Poultry Establishments
IX. Appendix B--Model of a Standard Operating Procedure for 
Sanitation
X. Appendix C--Guidebook for the Preparation of HACCP Plans
XI. Appendix D--Hazards and Preventive Measures Guide
XII. Appendix E--FSIS Sample Collection Guidelines and Procedure for 
Isolation and Identification of Salmonella from Raw Meat and Poultry 
Products
XIII. Appendix F--Guidelines for Escherichia coli Testing for 
Process Control Verification in Cattle and Swine Slaughter 
Establishments
XIV. Appendix G--Guidelines for Escherichia coli Testing for Process 
Control Verification in Poultry Slaughter Establishments
XV. Supplement--Final Regulatory Impact Assessment

I. Background

Overview of FSIS Food Safety Goal and Strategy

    The mission of the FSIS is to ensure that meat, poultry, and egg 
products are safe, wholesome, and properly marked, labeled, and 
packaged. Regarding meat and poultry, FSIS currently carries out its 
food safety responsibility primarily by managing an inspection program 
within meat and poultry slaughter and processing establishments. This 
program relies heavily on FSIS inspectors to detect and correct 
establishment sanitation and food safety problems.
    Recent outbreaks of foodborne illness and studies conducted over 
the past decade by the National Academy of Sciences (NAS), the U.S. 
General Accounting Office (GAO), and FSIS itself have established the 
need for fundamental change in the FSIS meat and poultry inspection 
program to improve food safety, reduce the risk of foodborne illness in 
the United States, and make better use of the Agency's resources.
    FSIS has embarked on a broad effort to bring about the necessary 
changes in its program. In the preamble to the ``Pathogen Reduction; 
Hazard Analysis Critical Control Point (HACCP) Systems'' proposed rule, 
published in the Federal Register of February 3, 1995 (Docket #93-016P, 
60 FR 6774; hereafter ``Pathogen Reduction/HACCP proposal''), FSIS 
traced the origins of its current program, described today's food 
safety challenges, and outlined a new food safety strategy for meat and 
poultry products. In that document, FSIS proposed new regulations to 
mandate adoption within meat and poultry establishments of HACCP, a 
science-based process control system for food safety.
    The HACCP requirement and other food safety measures proposed by 
FSIS in the Pathogen Reduction/HACCP proposal were motivated by the 
critical need to fill a gap in the current regulation and inspection 
system and the lack of adequate measures to address the problem of 
pathogenic microorganisms on raw meat and poultry products.
    Such bacteria, including Salmonella, E. coli O157:H7, Campylobacter 
and Listeria monocytogenes, are significant food safety hazards 
associated with meat and poultry products. FSIS estimates that the 
contamination of meat and poultry products with these bacteria results 
annually in as many as 4,000 deaths and 5,000,000 illnesses.

    FSIS stated the goal of its food safety strategy and proposed 
Pathogen Reduction/HACCP regulations as follows: FSIS believes its 
food safety goal should be to reduce the risk of foodborne illness 
associated with the consumption of meat and poultry products to the 
maximum extent possible by ensuring that appropriate and feasible 
measures are taken at each step in the food production process where 
hazards can enter and where procedures and technologies exist or can 
be developed to prevent the hazard or reduce the likelihood it will 
occur (60 FR 6785).

    In establishing this goal, FSIS recognized that no single 
technological or procedural solution exists for the problem of 
foodborne illness and that the Agency's food safety goal would be 
achieved only through continuous efforts to improve hazard 
identification and prevention.
    The food safety strategy FSIS outlined in the Pathogen Reduction/
HACCP proposal included the following major elements: (1) provisions 
for systematic prevention of biological, chemical, and physical hazards 
through adoption by meat and poultry establishments of science-based 
process control systems;

[[Page 38808]]

(2) targeted efforts to control and reduce harmful bacteria on raw meat 
and poultry products; (3) adoption of food safety performance standards 
that provide incentives for innovation to improve food safety and to 
provide a measure of accountability for achieving acceptable food 
safety results; (4) removal of unnecessary regulatory obstacles to 
innovation; and (5) efforts to address hazards that arise throughout 
the food safety continuum from farm to table.
    FSIS also stressed, as a central theme of its strategy, a need to 
clarify and strengthen the responsibilities of establishments for 
maintaining effective sanitation, following sound food safety 
procedures, and achieving acceptable food safety results.

FSIS Regulatory Proposals

    FSIS proposed HACCP as the organizing structure for its food safety 
program because HACCP is the optimal framework for building science-
based process control to prevent food safety hazards into food 
production systems. HACCP also focuses FSIS inspection on the most 
significant hazards and controls.
    To complement HACCP, FSIS proposed to establish, for the first 
time, food safety performance standards for pathogenic microorganisms 
on raw meat and poultry products, initially as ``interim'' targets for 
the reduction of Salmonella contamination of raw carcasses and raw 
ground meat and poultry products. These performance standards would 
measure whether HACCP systems are working effectively to address food 
safety hazards. FSIS proposed to require that establishments conduct 
daily microbial testing for Salmonella to verify achievement of the 
``targets.''
    FSIS also proposed three near-term measures to speed progress on 
controlling and reducing pathogenic microorganisms on raw products 
during the proposed three year phase-in of HACCP. These proposed 
measures were: (1) a requirement that all establishments adopt and 
implement sanitation standard operating procedures (Sanitation SOP's); 
(2) a requirement that all slaughter establishments use at least one 
effective antimicrobial treatment to reduce harmful bacteria; and, (3) 
standards for cooling red meat carcasses to prevent the growth of 
harmful bacteria.

FSIS Regulatory and Inspection Reform Plans

    In the Pathogen Reduction/HACCP proposal, FSIS acknowledged that it 
must do more than mandate HACCP and other new regulatory requirements 
in order to achieve its food safety goals. FSIS must also reform its 
existing regulations, policies, and directives to be consistent with 
HACCP principles and with the Agency's intention to rely more heavily 
on performance standards. Current FSIS regulatory requirements and 
procedures are generally highly detailed and prescriptive. They 
specify, for example, precise cooking time-and-temperature combinations 
for many products. Current regulations often assign to FSIS 
responsibility for the means used by establishments to produce safe 
food in a sanitary environment (e.g., FSIS requires that facility 
blueprints and equipment receive Agency approval before use).
    As part of its regulatory reform initiative, FSIS has undertaken 
the conversion of current command-and-control regulations to 
performance standards. Command-and-control regulations, and the 
Inspection System Guide that FSIS inspectors use to enforce those 
regulations, resulted from the perceived need to achieve uniformity 
among federally inspected meat and poultry establishments. 
Technological advances introduce a new imperative, however. If 
establishments are to innovate, using new technologies to improve food 
safety, they cannot be impeded by a one-size-fits-all regulatory 
system. Under contemporary conditions, affording establishments the 
flexibility to make establishment-specific decisions outweighs the 
advantages of uniformly applicable rules. Recognizing this, FSIS is 
changing inspection to meet the needs of the new regulatory system.
    Under the command-and-control-based system, the inspector assumed 
responsibility for ``approving'' production-associated decisions. Under 
the new system, industry assumes full responsibility for production 
decisions and execution. FSIS, having set food safety standards, 
monitors establishments' compliance with those standards and related 
requirements and under HACCP, verifies process control and pathogen 
reduction and control. The number of inspection tasks will be reduced, 
so that inspectors can focus more attention on areas of greatest risk 
in the meat or poultry production system within each establishment.
    With the shift to HACCP and greater reliance on performance 
standards, establishments will be afforded greater autonomy in 
decision-making affecting their own operations and, in return, be 
expected to take responsibility for setting up site- and product 
appropriate process control measures to achieve FSIS-established 
performance standards. This approach, which is intended to increase 
both the incentives and the flexibility establishments need to innovate 
and improve food safety, requires a complete review and overhaul of the 
``command-and-control'' requirements and procedures in current FSIS 
regulations, policies, and directives.
    HACCP-based food safety strategies and performance standards also 
require important changes in FSIS's approach to inspection. FSIS 
intends to clarify the respective responsibilities of FSIS inspectors 
and establishment management.
    In the Federal Register of December 29, 1995 (60 FR 67469), FSIS 
published an advance notice of proposed rulemaking (ANPR) and 
additional rulemaking proposals describing the Agency's strategy for 
the regulatory and inspectional reform required to achieve the changes 
required for consistency with HACCP. These changes will be accomplished 
before establishments are required to implement HACCP.

Change Within FSIS

    Finally, achieving the Agency's food safety goals will require 
substantial change within FSIS itself, as the roles of establishments 
and Federal inspectors are realigned to accord with the HACCP 
philosophy. The scope of FSIS's food safety activities will also extend 
beyond slaughter and processing establishments to include new 
preventive approaches to hazards that occur during transportation, 
distribution, and retail, restaurant or food service sale of meat and 
poultry products.
    This expansion of the Agency's roles will require substantial 
training and redeployment of employees, and will place an enormous 
strain on agency resources. To meet these challenges, FSIS has 
conducted a top-to-bottom review of its regulatory roles, resource 
allocation and organizational structure. Reports prepared by FSIS 
employees containing analysis and recommendations on these topics were 
described and made available for public comment in the Federal Register 
of September 12, 1995 (60 FR 47346). FSIS will be making the 
fundamental internal changes required to successfully carry out its 
HACCP-based farm-to-table food safety strategy. These changes within 
FSIS, which include a major reorganization of the Agency, will ensure 
that FSIS is using its resources to improve food safety consistent with 
its new regulatory framework.

[[Page 38809]]

The FSIS Pathogen Reduction/HACCP Rulemaking Process

    Recognizing that HACCP and other regulatory requirements contained 
in the Pathogen Reduction/HACCP proposal are part of a broad overhaul 
of the FSIS regulatory program, and involve important changes in the 
responsibilities of meat and poultry establishments, FSIS has conducted 
a thorough and interactive rulemaking process. The Agency's goal has 
been to provide many opportunities for submission by the public of both 
written and oral comments and for interchange between FSIS and 
interested parties on the many major policy and technical issues 
involved in the reform of meat and poultry inspection.
    The initial comment period was 120 days, which FSIS subsequently 
extended for an additional 30 days and later reopened for another 95 
days. During this period, FSIS held seven informational briefings, 
three scientific and technical conferences, a two-day public hearing, a 
scoping session and six issue-focused public meetings, a Federal-State 
conference, and a Food Safety Forum. Extensive oral comments were 
transcribed and included with written comments in the record of this 
rulemaking. A brief summary of the various public meetings follows.

Seven Information Briefings

    Initially, FSIS held informational briefings in seven cities across 
the country to explain the Pathogen Reduction/HACCP proposal to the 
public and to answer questions. A panel of FSIS officials and 
scientists provided information on the proposed regulations and 
answered questions. These briefings were not intended to solicit 
comments, but to help interested parties prepare themselves to comment 
on the Pathogen Reduction/HACCP proposal. These briefings were held:

March 7, 1995; Oakland, California
March 14, 1995; Dallas, Texas
March 16, 1995; Chicago, Illinois
March 21, 1995; Atlanta, Georgia
March 23, 1995; New York, New York
March 30, 1995; Washington, D.C.
May 22, 1995; Kansas City, Kansas

    The Kansas City session included an informational briefing and 
public meeting for owners and representatives of small meat and poultry 
establishments and other affected small businesses to discuss the 
Pathogen Reduction/HACCP proposal. At the meeting, many small business 
owners said that the Pathogen Reduction/HACCP proposal might eventually 
inhibit small businesses from competing with larger entities because 
the resulting additional costs could be borne more easily by larger 
companies. Three Directors of State Meat and Poultry Inspection 
Programs stated their views that the Pathogen Reduction/HACCP proposal 
might have a negative impact upon the small businesses for which they 
provide inspection. Consumers requested that FSIS base its decisions on 
the Pathogen Reduction/HACCP proposal not on industry impacts, but on 
what will best protect the public.

Three Scientific and Technical Conferences

    FSIS held three scientific and technical conferences to foster the 
development of beneficial new food safety technologies, to fill gaps in 
scientific knowledge, and to ensure that the Agency had the best 
scientific information available for the rulemaking. Concerned that the 
typical rulemaking process would not elicit this information, the 
Agency invited experts on relevant subjects to the meetings, which were 
open to all interested parties.
    The first conference, titled ``New Technology to Improve Food 
Safety,'' was held April 12-13, 1995, in Chicago, Illinois. This 
conference explored the available technology that might be introduced 
into the production and manufacturing of meat and poultry products to 
control E. coli O157:H7 and other harmful pathogens in the food supply. 
Participants included members of industry, academia, research 
organizations, and consumers. Additionally, Government representatives 
from non-food Federal regulatory agencies discussed technology 
development and transfer in other industries. FSIS discussed how it 
emphasized and encourages the approval and introduction of new 
technologies.
    The second conference, titled ``The Role of Microbiological Testing 
in Verifying Food Safety,'' was held May 1-2, 1995, in Philadelphia, 
Pennsylvania. This meeting explored scientific issues related to the 
use of microbiological testing for verifying meat and poultry safety. 
Six persons were invited to present discussions relating to the use and 
limitations of microbiological testing in ensuring food safety. Twelve 
representatives from academia, consumer groups, industry, and exporting 
countries also presented talks on the concepts and methods for 
microbiological testing that appeared in the proposed regulation. 
During the comment period following the presentations, 15 people 
commented on the subjects covered at the meeting and in the proposed 
regulation.
    The third conference, titled ``An Evaluation of the Role of 
Microbiological Criteria in Establishing Food Safety Performance 
Standards in Meat and Poultry Products,'' was held May 18-19, 1995, in 
Washington, D.C. It explored the use of microbiological criteria to 
establish food safety performance standards for meat and poultry 
products. Participants generally agreed that HACCP is an effective 
approach to controlling microbiological hazards in foods, and that 
government and industry must work together to establish microbiological 
criteria, sampling plans and training for food safety performance 
standards. Most commenters agreed that the use of an indicator organism 
is effective to facilitate and monitor the reduction of microbiological 
contamination in meat and poultry products. Diverse opinions were 
expressed on which indicator organisms should be chosen for each type 
of product.

Public Hearing

    On May 30 and 31, 1995, FSIS held a public hearing in Washington, 
D.C., on the proposed rule.
    Thirty-seven persons presented comments at the 2-day hearing. 
Issues and viewpoints varied greatly. For instance, requests were made 
to keep carcass-by-carcass inspection, but it was suggested that 
organoleptic inspection is outdated. While there was support for a 
HACCP system, many suggestions were made for changes in specific parts 
of the proposal, particularly microbial testing and antimicrobial 
treatments. Several commenters described their personal experiences 
with foodborne illness. Small business owners and their representatives 
commented on the potential financial burdens that might result from the 
Pathogen Reduction/HACCP proposal.

Federal-State Relations Conference

    As part of the annual meeting of Directors of State Meat and 
Poultry Inspection Programs, FSIS held a ``Federal-State Relations 
Conference,'' August 21-23, 1995, in Washington, D.C. This meeting, in 
which the National Association of State Departments of Agriculture 
participated, provided an opportunity for representatives from State 
government to engage in an open exchange with senior USDA officials on 
the Pathogen Reduction/HACCP proposal. In addition to State Directors, 
the meeting included representatives from State Departments of 
Agriculture, State Health Departments and local food safety enforcement 
agencies; additionally, the Food and Drug Administration (FDA)

[[Page 38810]]

and the Association of Food and Drug Officials were participants. These 
parties recognized a need to better protect the public by optimizing 
the use of available resources. State agency representatives discussed 
the need for better coordination within their own States and with the 
Federal Government to prevent foodborne illness outbreaks. Improved 
food handling education for industry and consumers was seen as one of 
the primary ways to improve farm-to-table food safety.

Scoping Session and Six Issue-Focused Meetings

    By late August, FSIS had received more than 6,800 comments on the 
Federal Register notice, in addition to the input obtained at the 
meetings and the hearing. All this information raised new issues and 
modified Agency thinking in some areas. In order to share new 
information and current thinking with its constituencies, FSIS held six 
issue-focused public meetings on the proposed rule and accepted written 
comments from those unable to attend. The meetings were announced in 
the Federal Register (60 FR 45380; Thursday, August 31, 1995) and held 
at USDA, Washington, D.C., on September 13, 14, 15, 27, 28, and 29, 
1995.
    FSIS framed an agenda for the meetings and provided issue papers 
describing current Agency thinking on the proposed rule. Before the 
issue-focused public meetings, FSIS held a public scoping session on 
August 23, 1995, to ensure that all parties had an opportunity to 
suggest issues for the agenda.
    The issue papers provided at the six issue-focused public meetings 
were published in the Federal Register (60 FR 54450; Tuesday, October 
24, 1995).

Food Safety Forum

    A Food Safety Forum chaired by Secretary Glickman was held on 
November 8, 1995 to discuss food safety reform issues beyond the 
specific issues raised by the proposed Pathogen Reduction/HACCP 
proposal. The forum agenda included topics such as: (1) whether 
legislative changes to the Federal Meat Inspection Act (FMIA) and the 
Poultry Products Inspection Act (PPIA) were needed; (2) how FSIS could 
improve food safety by organizational change, regulatory reform, 
reliance on user fees, effective resource allocation and other means; 
(3) cooperation between USDA and State inspection programs; and (4) 
government and private sector roles in consumer education regarding 
safe food handling practices. A transcript of the forum has been 
included in the record for this rulemaking.

Farm-to-Table Strategy

    In the preamble to its Pathogen Reduction/HACCP proposal, FSIS 
presented a strategy for the control of food safety hazards throughout 
the continuum of animal production and slaughter, and the processing, 
distribution, and sale of meat and poultry products. FSIS has 
historically focused on the manufacturing of meat and poultry products 
through its inspection program, but the Agency's public health mandate 
requires that the Agency also consider pre- and post-processing hazards 
as part of a comprehensive strategy to prevent foodborne illness.
    This farm-to-table food safety strategy is founded on three 
principles:
    <bullet> Hazards that could result in foodborne illness arise at 
each stage in the farm-to-table continuum: animal production and 
slaughter, and the processing, transportation, storage and retail, 
restaurant or food service sale of meat and poultry products. Each 
stage presents hazards of pathogen and other contamination and each 
provides opportunities for minimizing the effect of those hazards.
    <bullet> Those in control of each segment of the farm-to-table 
continuum bear responsibility for identifying and preventing or 
reducing food safety hazards that are under their operational control.
    <bullet> The Agency's public health mandate requires that it 
address foodborne illness hazards within each segment of the food 
production chain and implement or encourage preventative strategies 
that improve the whole system.
    FSIS remains committed to a farm-to-table food safety strategy 
based on these principles. To address hazards arising within slaughter 
and processing establishments, FSIS proposed and is adopting in this 
rule significant new regulatory measures. Improving food safety before 
the animals reach slaughter establishments will require a different 
approach. The preamble to the Pathogen Reduction/HACCP proposal stated 
that FSIS will be cooperating with animal producers, scientists in 
academia, the Animal and Plant Health Inspection Service and other 
government agencies to develop and foster food safety measures that can 
be taken on the farm and through marketing channels to decrease public 
health hazards in animals presented for slaughter. Within this context, 
the voluntary application of food safety assurance programs based on 
HACCP principles can be useful in establishing risk reduction practices 
on the farm and through intermediate marketing stages to control and 
reduce pathogen hazards at slaughter.
    FSIS expects, within the limits of available resources, to serve as 
a facilitator and coordinator of research and other activities designed 
to encourage development and implementation of animal production 
technologies and practices that can improve food safety. FSIS also 
intends to offer its expertise to assist State health and agricultural 
officials, when requested, during outbreak investigations of foodborne 
illnesses to learn more about potential risk factors. FSIS does not 
intend nor is FSIS authorized, to mandate production practices on the 
farm, but does expect that continued public concern about foodborne 
pathogens and adoption of HACCP and food safety performance standards 
within slaughter and processing establishments will increase incentives 
for improving food safety practices at the animal production level.
    The post-processing transportation, storage, and retail, restaurant 
or food service sectors are also important links in the chain of 
responsibility for food safety. In these areas, FDA and State and local 
governments share authority and responsibility for oversight of meat 
and poultry products outside of official establishments. FSIS and FDA 
are collaborating in the development of standards governing the safety 
of potentially hazardous foods, including meat and poultry, eggs, and 
seafood, during transportation and storage, with particular emphasis on 
proper cooling to minimize the growth of pathogenic microorganisms, and 
on disclosure of prior cargoes in transport vehicles. This effort will 
be discussed in a forthcoming advance notice of proposed rulemaking.
    In the retail, restaurant and food service areas, FSIS and FDA are 
working in concert with State and local food regulatory officials to 
foster adoption of updated, uniform, science-based standards, including 
mandates for HACCP process controls for high-risk processing and 
packaging operations. State and local authorities have assumed primary 
responsibility for food safety oversight of retail, restaurant and food 
service operations, but FSIS and FDA, working through the Conference on 
Food Protection and other collaborative mechanisms, provide expertise 
and leadership to support local authorities and foster development of 
sound food safety standards and practices nationwide. FSIS is 
cooperating with FDA to update the Food Code, a set of model ordinances 
recommended for adoption by the

[[Page 38811]]

States, to ensure meat and poultry safety is adequately addressed in 
retail, restaurant and food service settings.
    Even as progress is made in reducing contamination of food by 
harmful bacteria and other safety hazards at the production, processing 
and subsequent commercial stages of the farm-to-table continuum, it 
will remain critically important that individual consumers follow safe 
food handling practices. Proper storage, preparation, and cooking of 
meat and poultry products are essential to achieving the goal of 
reducing the risk of foodborne illness to the maximum extent possible. 
FSIS intends to augment its food handler and consumer education efforts 
by expanding its collaboration with the meat and poultry industry, 
other government agencies, consumer and public interest groups, 
educators, and the media to effectively develop and deliver food safety 
education and information to the public.
    The HACCP requirements and other regulations FSIS is adopting in 
this final rule will ensure that inspected establishments are taking 
appropriate measures to reduce hazards at critical stages where the 
risk of initial contamination is greatest. The public health benefits 
of these measures, however, are only a part of a comprehensive food 
safety strategy that seeks to minimize hazards throughout the farm-to-
table continuum.

General Overview of the Comments and the Final Rule

HACCP and Performance Standards
    The FSIS proposal to require adoption of HACCP in meat and poultry 
establishments was widely endorsed by comments from large and small 
businesses, the scientific and public health communities, consumers, 
and public interest organizations. Commenters strongly supported the 
concept that meat and poultry establishments should systematically 
build science-based food safety measures into their production 
processes following the seven HACCP principles developed by the 
National Advisory Committee on Microbiological Criteria for Food 
(NACMCF). Although many commenters requested clarification of how FSIS 
intends to implement HACCP and conduct inspection under HACCP, the 
principal critical comments concerned costs and the practicality of 
using HACCP in very small establishments. FSIS is adopting the HACCP 
requirements, based on the NACMCF principles, essentially as proposed.
    From a food safety standpoint, the most important objective of this 
rulemaking is to build into food production processes, and into the 
system of FSIS regulation and oversight, effective measures to reduce 
and control harmful bacteria on raw meat and poultry products. This 
will not by itself solve the problem of foodborne illness associated 
with meat and poultry products. Effective measures are needed 
throughout the farm-to-table continuum, but this rulemaking will fill 
the most critical gap in the current system of meat and poultry 
inspection. While products sold in cooked or otherwise ready-to-eat 
forms are currently subject to controls and regulatory standards 
designed to eliminate harmful bacteria, products sold raw are not 
currently subject, as a general matter, to any such controls or 
standards.
    FSIS has concluded that HACCP-based process control, combined with 
appropriate food safety performance standards, is the most effective 
means available for controlling and reducing harmful bacteria on raw 
meat and poultry products. HACCP provides the framework for industry to 
set up science-based process controls that establishments can validate 
as effective for controlling and reducing harmful bacteria. Performance 
standards tell establishments what degree of effectiveness their HACCP 
plans will be expected to achieve and provide a necessary tool of 
accountability for achieving acceptable food safety performance. 
Science-based process control, as embodied in HACCP, and appropriate 
performance standards are inextricably intertwined in the Agency's 
regulatory strategy for improving food safety. Neither is sufficient by 
itself, but, when combined, they are the basis upon which FSIS expects 
significant reductions in the incidence and levels of harmful bacteria 
on raw meat and poultry products and, in turn, significant reductions 
in foodborne illness.
    The proposed interim targets for pathogen reduction based on 
Salmonella generated widely diverse comments. Commenters supported the 
goal of pathogen reduction, and many recognized some role for microbial 
testing and the need for a microbial reduction target or performance 
standard. Some commenters argued that the proposed testing regimen (a 
single sample per species per day) was inadequate for its purpose in 
large establishments, while others argued it was too burdensome in 
small establishments. Some commenters specifically supported the 
proposed Salmonella reduction targets and the daily testing 
requirements. Many, however, criticized the proposed testing 
requirements and considered Salmonella testing less useful than generic 
E. coli testing as an indicator of whether process controls in 
slaughter establishments are effectively preventing fecal 
contamination, the primary pathway for pathogen contamination. At the 
scientific conference on the role of microbial testing held in 
Philadelphia, broad support also was expressed for using generic E. 
coli rather than Salmonella as a process control indicator.
    Based on public comments, FSIS has modified its approach to 
establishing microbial performance standards. FSIS believes that 
testing for generic E. coli is the appropriate and necessary means by 
which meat and poultry slaughter establishments must verify their 
process controls. FSIS reviewed written comments received on the 
original proposal and comments made at the scientific conferences and 
public meetings, as well as available scientific data, and has decided 
to require slaughter establishments to conduct testing for generic E. 
coli to verify process controls. Establishments will be required to 
test for E. coli at a frequency that takes into account their volume of 
production. FSIS is seeking additional scientific and economic data 
that may help to further improve the E. coli testing protocols.
    FSIS is also establishing performance criteria based on national 
microbiological baseline surveys. The criteria are not regulatory 
standards but rather provide a benchmark for use by slaughter 
establishments in evaluating E. coli test results. Test results that do 
not meet the performance criteria will be an indication that the 
slaughter establishment may not be maintaining adequate process control 
for fecal contamination and associated bacteria. Such results will be 
used in conjunction with other information to evaluate and make 
appropriate adjustments to ensure adequate process control for fecal 
contamination and associated bacteria.
    FSIS is also establishing pathogen reduction performance standards 
for Salmonella that will require all slaughter establishments to reduce 
the incidence of Salmonella contamination of finished meat and poultry 
carcasses below the national baseline prevalence as established by the 
most recent FSIS national microbiological baseline data for each major 
species. FSIS will conduct Salmonella testing in slaughter 
establishments to detect whether they are meeting the pathogen 
reduction performance standards, and will require corrective action or 
take regulatory

[[Page 38812]]

action, as appropriate, to ensure establishments are meeting the 
pathogen reduction standards.
    Pathogen-specific performance standards for raw products are an 
essential component of the FSIS food safety strategy because they 
provide a direct measure of progress in controlling and reducing the 
most significant hazards associated with raw meat and poultry products. 
The Salmonella standards being established in this final rule, which 
are based on the current national baseline prevalence of Salmonella 
(expressed as a percentage of contaminated carcasses), are a first step 
in what FSIS expects to be a broader reliance in the future on 
pathogen-specific performance standards. FSIS plans to repeat its 
baseline surveys and collect substantial additional data through other 
means and, on that basis, adjust the Salmonella performance standards 
and possibly set standards for additional pathogens, as appropriate. 
Also, FSIS will continue to explore establishing pathogen-specific 
performance standards based on the levels of contamination (i.e., the 
number of organisms) on a carcass. Future FSIS efforts on such 
performance standards will reflect the fact that achieving the food 
safety goal of reducing foodborne illness to the maximum extent 
possible will require continuous efforts and improvement over a 
substantial period.
Sanitation SOP's, Antimicrobial Treatments, and Cooling Requirements 
for Raw Meat and Poultry Products
    Comments generally supported the objectives of the three near-term 
measures for raw meat and poultry products proposed by FSIS, Sanitation 
SOP's, antimicrobial treatments, and carcass cooling standards, and 
most commenters agreed that Sanitation SOP's should be a required 
element of any meat and poultry establishment's food safety program. 
Many commenters objected, however, to FSIS mandated antimicrobial 
treatments in slaughter establishments and carcass cooling standards 
for red meat prior to the implementation of HACCP. Although most 
comments generally agreed that antimicrobial treatments would play an 
important role in many slaughter establishments' HACCP plans, and that 
proper carcass cooling would be an essential part of any HACCP plan for 
raw meat and poultry products, these commenters argued that mandating a 
particular approach to antimicrobial treatments or carcass cooling 
would be inconsistent with the HACCP concept that establishment 
management is responsible for designing a system of controls 
appropriate for each establishment. They also argued that mandating 
antimicrobial treatments was unnecessary if establishments were 
required to meet pathogen reduction performance standards. Similarly, 
with respect to the proposed requirement that establishments cool red 
meat carcasses following specific cooling rate standards prescribed by 
FSIS, commenters argued that HACCP, reinforced by performance 
standards, would ensure proper carcass cooling. Many commenters said 
that the specific time-and-temperature requirements proposed by FSIS 
were often not feasible, posed worker safety concerns, and would divert 
effort and resources that could be used more productively in preparing 
for implementation of HACCP.
    Based on the comments, FSIS has reconsidered its approach to the 
proposed near-term measures. FSIS believes that its regulatory program 
and the food safety efforts of the meat and poultry industry should be 
focused on making a transition to HACCP as rapidly and effectively as 
possible and that FSIS should not mandate any near-term measures that 
would not be expected to continue as mandatory elements of a HACCP-
based system.
    FSIS has decided to adopt final rules that mandate Sanitation 
SOP's. Good sanitation is a critical foundation for HACCP, and 
Sanitation SOP's are an essential element of the FSIS effort to more 
clearly define establishment and inspector responsibilities, and better 
focus both the establishment management and FSIS on those elements of 
daily sanitation that relate most directly to the risk of product 
contamination. Near-term implementation of Sanitation SOP's will 
facilitate the transition to HACCP.
    FSIS has decided not to mandate antimicrobial treatments in 
slaughter establishments. The Agency expects that antimicrobial 
treatments will play an important role in the design of slaughter HACCP 
plans as establishments institute controls that are effective in 
reducing pathogens and meeting FSIS performance standards. As a general 
matter, however, FSIS does not intend to mandate the specific controls 
that establishments must adopt in their HACCP plans. In the case of 
antimicrobial treatments, FSIS believes that improvement in food safety 
would be better served by providing establishments the incentive and 
flexibility to incorporate antimicrobial treatments in any manner they 
judge most effective for their operations to meet FSIS-established 
performance standards for reducing bacterial contamination.
    With respect to carcass cooling, FSIS continues to believe that, in 
a HACCP environment, appropriate performance standards are needed for 
the cooling of carcasses and raw meat and poultry products to prevent 
the growth of harmful bacteria. After consideration of the comments, 
FSIS has concluded, however, that the specific time-and-temperature 
combinations proposed by FSIS were too restrictive and that a 
scientifically sound and effective strategy for preventing the growth 
of pathogens through proper cooling must apply not only within, but 
also beyond, FSIS-inspected establishments. Thus, instead of including 
requirements for carcass cooling in this final rule, FSIS intends to 
extend this rulemaking to consider alternative approaches to 
performance standards for cooling within establishments. Concurrently, 
FSIS also intends to develop rulemaking covering the adoption of 
standards for cooling of raw products during transportation, storage, 
and retail, restaurant or food service sale. FSIS anticipates adopting 
performance standards designed to minimize the growth of harmful 
bacteria on raw products that establishments will be required to meet 
through their HACCP plans. FSIS will announce in a future issue of the 
Federal Register a three-day public conference to gather further 
scientific information and public comment on these subjects.

Timetable for Implementation

Federally Inspected Establishments
    FSIS proposed an implementation timetable that would have phased in 
the near-term measures and HACCP over a period of time beginning 90 
days and ending three years after publication of the final rule. 
Sanitation SOP's and the other near-term measures, as well as the 
proposed microbial sampling by establishments for Salmonella, were to 
begin 90 days after publication. Slaughter establishments were to be 
held accountable for meeting the Salmonella targets two years after 
publication.
    FSIS proposed to phase in HACCP over a one to three-year period, 
primarily on a process-by-process basis. For example, raw ground 
products would be subject to the HACCP requirements one year after 
publication of the final rule, while all slaughter establishments would 
be required to start HACCP thirty months (2\1/2\ years) after 
publication of the final rule. However, FSIS proposed that 
establishments with annual sales of less than $2.5 million be given 
three years to

[[Page 38813]]

comply with the HACCP requirement, regardless of the processes they 
run.
    Some commenters said the proposed implementation timetable was too 
slow, considering the seriousness of the food safety issues involved 
and the familiarity with HACCP that already exists among many in the 
industry. Other commenters pointed out that many larger establishments 
have already adopted HACCP. Some said the Pathogen Reduction/HACCP 
proposal placed excessive burdens on smaller establishments, which were 
said to be less prepared technically and financially to carry out 
HACCP. Wide support was voiced for implementing HACCP as promptly as 
practicable, taking into account the diversity of businesses involved 
and the different levels of readiness for HACCP.
    FSIS has considered these comments and has also re-evaluated the 
proposed timetable for implementation of all requirements discussed 
above in light of preparations FSIS will itself have to make to 
implement HACCP, including the training of inspection and other agency 
employees. FSIS believes it is important to bring the meat and poultry 
supply under HACCP-based process control and to implement other 
elements of its food safety strategy as rapidly as possible. It is also 
important to have a timetable that is realistic for implementing this 
fundamental transformation in how FSIS regulates meat and poultry 
establishments. FSIS is modifying the timetable for implementation in a 
way that achieves both goals.
    The Sanitation SOP's requirements will take effect 6 months after 
publication of these final rules, rather than 90 days as originally 
proposed.
    Establishments slaughtering livestock or poultry will be required 
to begin process control verification testing for generic E. coli 6 
months after publication of this final rule.
    FSIS will begin holding slaughter establishments and establishments 
producing raw ground products accountable for achieving Salmonella 
pathogen reduction performance standards at the time they will be 
required to implement HACCP under the phase-in schedule described 
below, rather than the single, two-year delayed effective date 
originally proposed. Beginning approximately three months after 
publication of this final rule, FSIS will initiate its pre-enforcement 
Salmonella testing program. This establishment-by-establishment 
Salmonella prevalence survey will provide critical data on the 
performance of establishments; it will inform establishments of their 
performance, and guide FSIS enforcement testing and compliance 
strategies after establishments are required to meet the Salmonella 
performance standards.
    In response to comments, FSIS is modifying the proposed timetable 
for implementing HACCP from one based primarily on production process 
in an establishment to one based on establishment size. Under this 
approach, the pace at which most of the Nation's meat and poultry 
supply comes under HACCP-based process control will be accelerated. 
Most important, slaughter establishments that account for 75% of the 
annual meat and poultry production in the United States will be 
required to implement HACCP 18 months after publication of these final 
rules, rather than 30 months after publication as originally proposed. 
At the same time, very small establishments (those with fewer than 10 
employees or with annual sales of less than $2.5 million, together 
accounting for less than 2% of meat and poultry production) will be 
provided an additional six months beyond the proposed three years to 
implement HACCP.
    Under this timetable, FSIS gains needed time to develop and 
sequence inspector training and other preparatory activities. Also, 
establishments that carry out multiple processes (such as the so-called 
``combo'' establishments that both slaughter animals and grind raw 
products) will be able to implement HACCP on a more coherent 
establishment-wide basis, rather than on a process-by-process basis. A 
detailed description of the implementation timetable and its rationale 
is provided in section II of this preamble.
State-Inspected Establishments
    Both the FMIA and PPIA direct Federal cooperation with States in 
developing and administering intrastate inspection programs that 
include mandatory antemortem and postmortem inspection, reinspection, 
and sanitation requirements which are ``at least equal to'' Federal 
requirements. Consequently, each State receiving matching Federal funds 
for the administration of its intrastate meat and poultry inspection 
program must implement Pathogen Reduction/HACCP programs that are at 
least equal to provisions set forth in this final rule. FSIS will 
coordinate closely with States that maintain federally supported meat 
and poultry inspection programs to ensure that Pathogen Reduction/HACCP 
is implemented in all intrastate establishments.
Foreign-Inspected Establishments
    In order to export meat or poultry to the United States, foreign 
countries must establish a system of inspection that is equivalent to 
the system in this country. Determinations of equivalency made by U.S. 
reviewers of foreign meat and poultry inspection systems are currently 
based upon (1) the presence or lack of specific regulatory requirements 
and (2) how those requirements are enforced. As Pathogen Reduction/
HACCP regulatory provisions are implemented in the U.S. domestic 
market, foreign countries will concurrently be evaluated to ascertain 
whether their inspection systems provide equivalent regulatory 
provisions with adequate levels of enforcement.

Implementation Conferences

    FSIS plans to convene a three-day HACCP implementation conference 
in Washington, DC, about 60 days after publication of this final rule. 
Similar sessions will follow in various cities around the country.
    The purpose of the implementation conferences is to continue, and 
build upon, the dialogue among interested parties that occurred during 
the six days of public meetings FSIS conducted in September 1995 on the 
proposed rule. FSIS anticipates that the following topics will be 
discussed at the implementation conferences: (1) status of FSIS efforts 
to develop generic model HACCP plans and conduct small establishment 
HACCP demonstration projects; (2) the draft guidance materials 
published as Appendices; (3) the revised HACCP implementation schedule 
and certain technical aspects of the regulations being promulgated in 
this final rule; (4) other implementation issues identified by the 
public; (5) methods to achieve the goal of consistent training for FSIS 
and industry employees; and (6) due process and enforcement issues.
    In addition, FSIS plans to conduct two public conferences on 
technical issues related to E. coli testing. The first conference is 
planned to be held approximately 45 days into the 60-day comment period 
following publication of this rule. The public conference will be led 
by a panel of scientists from FSIS and other government agencies who 
will listen to testimony and review comments received on these 
technical issues and share their observations and opinions. FSIS will 
consider their input as well as all comments received as the basis for 
any necessary technical amendments which will be completed at least 30 
days before the

[[Page 38814]]

implementation date. The second conference is tentatively planned for 
approximately 9 months following publication of this rule. This 
conference would be an opportunity for the industry and others to 
discuss with FSIS new information based on about 3 months of testing 
experience that may bear on these same issues and might allow for 
further adjustments of protocols before FSIS inspectors are tasked, 
about three months later, with comparing test results to the national 
criteria as part of their inspection routine. FSIS will publish 
further, more detailed notice of these conferences in future issues of 
the Federal Register.

Request for Comments

    These final rules have benefitted from substantial public comment 
and the dialogue that took place during extensive public meetings with 
interested groups and individuals. Following the close of the comment 
period on November 13, 1995, several industry associations requested 
that these regulations be issued as ``interim'' final rules with a 30-
day opportunity for further public comment prior to the rules becoming 
final. FSIS is denying this request because the HACCP principles and 
other major elements of these final regulations have already been the 
subject of unusually extensive public comment and dialogue, and it is 
important to proceed toward implementation of these new food safety 
measures as promptly as possible.
    FSIS seeks comments, however, on certain technical aspects of these 
final regulations and on the guidelines (published here as Appendices) 
that will play a role in implementation of sanitation SOP's, microbial 
testing, and HACCP. FSIS requests comments no later than September 23, 
1996 on (1) technical issues that are associated with E. coli testing; 
(2) the E. coli performance criteria, and (3) the Sanitation SOP's 
Guideline and Model Sanitation SOP's, published at Appendices A and B, 
respectively.
    Based on comments it receives, FSIS will make any necessary 
revisions in the draft guidelines and technical aspects of the E. coli 
testing regulation prior to the effective date of the affected 
regulatory requirements.
    With respect to the HACCP final regulations, FSIS requests comments 
by November 22, 1996 on (1) the revised HACCP implementation timetable, 
including any factual information that commenters believe would justify 
any adjustments in the announced effective dates; (2) the Hazards and 
Preventive Measures Guide (published at Appendix D) and (3) the 
Guidebook for the Preparation of HACCP Plans (published at Appendix C).

II. Hazard Analysis and Critical Control Point Systems

Overview of Final Rule

    This final rule requires that federally inspected establishments 
implement HACCP systems to address hazards that are reasonably likely 
to occur in their operations. The HACCP systems mandated by this final 
rule focus on attributes affecting product safety, not those affecting 
economic adulteration or quality. On the effective dates of this final 
rule, FSIS will begin verifying HACCP system operations as part of its 
inspection program. Establishments will be required to maintain a HACCP 
plan covering every meat or poultry product produced for human food. 
Processes for which HACCP plans must be developed include slaughter for 
all species; raw ground meat or poultry products; raw product, not 
ground (e.g., meat cuts or whole or cut-up birds); shelf-stable 
nonheat-treated products (e.g., jerky); shelf-stable heat-treated 
products (e.g., edible fats); thermally processed/commercially sterile 
products (e.g., canned soup); fully cooked nonshelf-stable products 
(e.g., canned hams that must be refrigerated); not fully cooked/heat-
treated products (e.g., char-marked beef patties); and nonshelf-stable 
products with secondary inhibitors (e.g., fermented sausage). It should 
be noted that the category of raw, not ground product can include 
products with certain additional processing steps beyond carcass 
dressing, such as cutting up whole carcasses or marinating meat or 
poultry products.

History and Background of HACCP

    HACCP is a conceptually simple system whereby meat and poultry 
establishments can identify and evaluate the food safety hazards that 
can affect the safety of their products, institute controls necessary 
to prevent those hazards from occurring or keeping them within 
acceptable limits, monitor the performance of controls, and maintain 
records routinely. HACCP is the best system currently available for 
maximizing the safety of the nation's food supply.
    HACCP systems have been recommended for use in the food industry 
for more than a quarter century. The HACCP concept has been promoted by 
government and scientific groups and incorporated for many years in 
FSIS's and FDA's regulations on canned foods. Committees of the NAS 
have recommended that government agencies with responsibility for 
controlling microbiological hazards in foods, including FSIS, 
promulgate regulations requiring industry to utilize the HACCP system 
for food protection purposes.
    The NACMCF, which was established in accordance with a NAS 
committee recommendation, endorsed the HACCP system as an effective and 
rational approach to the assurance of food safety. In its March 20, 
1992, publication ``Hazard Analysis and Critical Control Point 
System,'' NACMCF advocated the standardization of the HACCP principles 
and their application by industry and regulatory authorities, with each 
food-producing establishment developing a HACCP system tailored to its 
individual product, processing, and distribution conditions.
    The U.S. General Accounting Office, in a series of reports between 
1992 and 1994, endorsed HACCP as an effective, scientific, risk-based 
system for protecting the public from foodborne illness. On December 
18, 1995, the FDA published final rules requiring the adoption of HACCP 
systems in seafood processing plants (60 FR 65096).
    International and foreign government bodies have also advocated the 
adoption of HACCP systems. The International Commission on 
Microbiological Specifications for Foods (ICMSF), in its 1988 report, 
``HACCP in Microbiological Safety and Quality,'' endorsed the use of 
HACCP systems in food production, processing, and handling. In 1993, 
the Food and Agriculture Organization/World Health Organization Codex 
Alimentarius Commission adopted a HACCP document that now serves as a 
guide for countries to incorporate HACCP principles into their food 
industries. The seven HACCP principles adopted by the Codex 
Alimentarius Commission are identical to those adopted by the NACMCF 
and on which this final rule is based. HACCP principles have been 
embodied in recent European Union regulatory directives and in food 
protection programs conducted by the governments of Canada, New 
Zealand, and Australia.

The Seven HACCP Principles

    The seven HACCP principles recommended by NACMCF in 1992 provide 
the framework for this final rule. While the seven principles are not 
explicitly listed as such in the codified regulatory text, they are 
embodied in the regulatory requirements for a hazard analysis in 
Sec. 417.2(a); the elements of a HACCP plan in Sec. 417.2 (b) and (c); 
the corrective action requirements in Sec. 417.3; the validation, 
verification, and reassessment requirements in Sec. 417.4; and the 
record review and maintenance

[[Page 38815]]

requirements in Sec. 417.5. The seven HACCP principles are discussed 
below.
    Principle No. 1: A hazard analysis of each process must be carried 
out. The purpose of the analysis is to identify and list the food 
safety hazards reasonably likely to occur in the production process for 
a particular product and the preventive measures necessary to control 
the hazards. A food safety hazard is any biological, chemical, or 
physical property that may cause a food to be adulterated or otherwise 
unsafe for human consumption. A listed hazard must be of such a nature 
that its prevention, elimination, or reduction to acceptable levels is 
essential to the production of a safe food.
    Examples of questions to be considered in a hazard analysis 
include: (1) What potential hazards may be present in the animals to be 
slaughtered or the raw materials to be processed? (2) What are the 
avenues that might lead to contamination of finished product with 
pathogenic microorganisms, hazardous chemicals, or other potentially 
hazardous contaminants? (3) What is the likelihood of such 
contamination and what are the means for preventing it? (4) Does the 
food contain any ingredient historically associated with a known 
microbiological hazard? (5) Does the food permit survival or 
multiplication of pathogens or toxin formation during processing? (6) 
Does the process include a controllable processing step that destroys 
pathogens? (7) Is it likely that the food will contain pathogens and 
are they likely to increase during the times and conditions under which 
the food is normally stored before being consumed? (8) What product 
safety devices are used to enhance consumer safety (e.g., metal 
detectors, filters, thermocouples)? (9) Does the method of packaging 
affect the multiplication of pathogenic microorganisms and/or the 
formation of toxins? (10) Is the product epidemiologically linked to a 
foodborne disease?
    Principle No. 2: The critical control points (CCP) of each process 
must be identified. A CCP is a point, step, or procedure at which 
control can be applied and a food safety hazard can be prevented, 
eliminated, or reduced to an acceptable level. All hazards identified 
during the hazard analysis must be addressed. The information developed 
during the hazard analysis should enable the establishment to identify 
which steps in their processes are CCP's.
    Identification of CCP's for controlling microbial hazards 
throughout the production process is particularly important because 
these hazards are the primary cause of foodborne illness. The 
establishment may find the CCP decision tree developed by the NACMCF 
useful in the CCP identification process (see Figure 1). However, the 
use of this technique in identifying CCP's is not required by this 
final rule.
    Principle No. 3: The critical limits for preventive measures 
associated with each identified CCP must be established.

BILLING CODE 3410-DM-P

[[Page 38816]]

[GRAPHIC] [TIFF OMITTED] TR25JY96.000



BILLING CODE 3410-DM-C
A critical limit is the maximum or minimum value to which a process 
parameter must be controlled at a CCP to prevent, eliminate, or reduce 
to an acceptable level the identified physical, biological, or chemical 
food safety hazard. Critical limits are most often based on process 
parameters such as temperature, time, physical dimensions, humidity, 
moisture level, water activity, pH, titratable acidity, salt 
concentration, available chlorine, viscosity, preservatives, or 
survival of target pathogens. Critical limits should be based on 
applicable FSIS regulations or guidelines, FDA tolerances and action 
levels, scientific and technical literature, surveys, experimental 
studies, or the recommendations of recognized experts in the industry, 
academia, or trade associations.
    Establishments are encouraged to establish critical limits more 
stringent than those now required by FSIS regulations or suggested by 
scientific data to ensure that regulatory requirements are routinely 
met, even when minor deviations occur.
    Principle No. 4: The monitoring requirements for CCP's must be 
established. Monitoring is an integral part of HACCP and consists of 
observations or measurements taken to assess whether a CCP is within 
the established critical limit. Continuous monitoring is preferred, but 
when it is not feasible, monitoring frequencies must be sufficient to 
ensure that the CCP is under control.
    Assignment of the responsibility for monitoring is an important 
consideration for each CCP. Personnel assigned the monitoring 
activities should be properly trained to accurately record all results, 
including any deviations, so that immediate corrective actions may be 
taken.
    Principle No. 5: The HACCP plan must include corrective action to 
be taken when monitoring indicates that there is a deviation from a 
critical limit at a critical control point. Although the process of 
developing a HACCP plan emphasizes organized and preventive thinking 
about what is occurring as the meat or poultry product is being 
manufactured, the existence of a HACCP plan does not guarantee that 
problems will not arise. For this reason, the identification of a 
planned set of activities to address deviations is an important part of 
a HACCP plan. In such instances, corrective action plans must be in 
place to determine the disposition of the potentially unsafe or 
noncompliant product and to identify and correct the cause of the 
deviation. The HACCP plan itself might require modification, perhaps in 
the form of a new critical limit, or of an additional CCP.

[[Page 38817]]

    Principle No. 6: Effective recordkeeping procedures that document 
the entire HACCP system must be developed and maintained. A HACCP 
system will not work unless consistent, reliable records are generated 
during the operation of the plan, and those records are maintained and 
available for review. One of the principal benefits of a HACCP process 
control system to both industry and regulatory officials is the 
availability of objective, relevant data.
    Principle No. 7: HACCP systems must be systematically verified. 
After initial validation that the HACCP system can work correctly and 
effectively with respect to the hazards, the system must be verified 
periodically. Periodic verification involves the use of methods, 
procedures, or tests in addition to those used for monitoring, to 
determine whether the HACCP system is in compliance with the HACCP plan 
and/or whether the HACCP plan needs modification and revalidation to 
achieve its food safety objective.
    In the NACMCF explanation of the verification principle, which FSIS 
is following, four processes are involved in the verification of the 
establishment's HACCP system. The establishment is responsible for the 
first three; FSIS is responsible for the fourth. The first is the 
scientific and technical process, known as ``validation,'' for 
determining that the CCP's and associated critical limits are adequate 
and sufficient to control likely hazards. The second process is to 
ensure, initially and on an ongoing basis, that the entire HACCP system 
functions properly. The third consists of documented, periodic, 
reassessment of the HACCP plan. The fourth process defines FSIS's 
responsibility for certain actions (Government verification) to ensure 
that the establishment's HACCP system is functioning adequately.

HACCP and the FSIS Food Safety Strategy

    The food safety goal of FSIS's Pathogen Reduction/HACCP rulemaking 
proposal is to reduce the risk of foodborne illness from meat and 
poultry products to the maximum extent possible by ensuring that 
appropriate and feasible preventive and corrective measures are taken 
at each stage of the food production process where food safety hazards 
occur. There is no single technological or regulatory solution to the 
problem of foodborne illness. Continuous efforts are required by 
industry and government to improve methods for identifying and 
preventing hazards and to minimize the risk of illness.
    FSIS proposed HACCP as the framework for carrying out its 
comprehensive strategy to improve food safety. HACCP, combined with the 
other measures required by this rulemaking, will substantially improve 
the ability of meat and poultry establishments and FSIS to target and 
systematically prevent and reduce food safety hazards and, working 
together, to continuously improve food safety as science and technology 
improve. These measures fill a critical gap in the current system with 
respect to the control and reduction of harmful bacteria on raw meat 
and poultry products and will, over time, significantly reduce the risk 
of foodborne illness.
    FSIS's meat and poultry inspection program currently addresses and 
will continue to address many matters of importance to the safety and 
quality of the food supply, including supervision of industry 
compliance with sanitation standards, exclusion of diseased animals 
from the food supply, examination of carcasses for other visible 
defects that can affect safety and quality, and inspecting for economic 
adulteration. These activities respond to some of the public's most 
basic expectations regarding the safety and quality of the food supply 
and reflect the standards and requirements established by Congress in 
the laws FSIS administers. FSIS is strongly committed to the most 
effective and efficient implementation of these statutory requirements.
    This final rule initiates a fundamental change in the inspection 
program to better meet FSIS's paramount obligation to protect the 
public health. Specifically, it addresses in a substantive way the 
public health problem of foodborne illness associated with the 
consumption of meat and poultry products. It does so in large part by 
better delineating and clarifying the respective roles of industry and 
FSIS to ensure that meat and poultry products are produced in 
accordance with sanitation and safety standards and are not adulterated 
or misbranded within the meaning of the FMIA and PPIA. This rule makes 
clear that the industry is responsible for producing and marketing 
products that are safe, unadulterated, and properly labeled and 
packaged. FSIS is responsible for inspecting products and facilities to 
verify that the statutory requirements are being met and for taking 
appropriate compliance and enforcement actions when the requirements 
are not being met.
    The line between the responsibilities of FSIS and those of the 
industry has often been blurred. This is because of the prescriptive 
nature of the current FSIS inspection program and the tendency for some 
establishments to rely on FSIS inspectors to do what is necessary to 
direct the correction of deficiencies and to ensure that outgoing 
products are safe, and not adulterated or misbranded. Some 
establishments operate on the assumption that if the inspector 
identifies no problem, their meat or poultry products may be entered 
into commerce. This is even more problematic because the current 
inspection system is based primarily on organoleptic methods that 
cannot detect the hazards of pathogenic microorganisms. The line has 
also been blurred because of the excessive reliance of the FSIS 
inspection program on the detection and correction of problems after 
the fact, rather than assurance that problems will be prevented, 
systematically by design, in the first place.
    The changes FSIS will effect with this final rule will eliminate 
this confusion and delineate clearly the respective responsibilities of 
FSIS and industry. The changes constitute a fundamental shift in the 
FSIS regulatory program, which FSIS is convinced will significantly 
enhance the effectiveness of the program and substantially reduce the 
risk of foodborne illness.

Preparing for HACCP Implementation

    For the new FSIS food safety strategy, particularly HACCP, to be 
successful, FSIS must reconsider its current reliance on prescriptive 
command-and-control regulations and instead rely more on performance 
standards. Not only do command-and-control regulations prescribe the 
means by which establishments are to achieve a particular food safety 
objective, but they are susceptible of being enforced in a manner that 
leads to the inspector's substantial involvement in management 
decisionmaking. Performance standards, on the other hand, prescribe the 
objectives or levels of performance (such as pathogen reduction 
standards for raw product) establishments must achieve, but afford 
establishments flexibility in determining how to achieve those 
performance objectives. The shift to performance standards and the 
concomitant increase in flexibility for meat and poultry establishments 
reflect FSIS's commitment to stimulating the innovative capacity of the 
meat and poultry and allied industries to improve the safety of their 
products.
    Command-and-control regulations are generally incompatible with 
HACCP and the FSIS food safety strategy, and conflict with the goal of 
reducing the

[[Page 38818]]

risk of foodborne illness on a continuing basis. They deprive 
establishments of the flexibility to innovate, one of the primary 
advantages of HACCP, and undercut the clear delineation of food safety 
responsibilities between industry and FSIS, on which the FSIS strategy 
is based. Therefore, to prepare for HACCP implementation, FSIS is 
conducting a thorough review of its current regulations and will, to 
the maximum extent possible, convert its command-and-control 
regulations to performance standards. (For a discussion of this 
regulatory reform initiative, see advance notice of proposed rulemaking 
published on December 29, 1995; Docket No. 95-008A; 60 FR 67469).

Inspection Under HACCP

    HACCP-oriented food safety inspection changes FSIS's approach to 
overseeing the safety of meat and poultry products. Under this new 
approach, FSIS will rely less on after-the-fact detection of product 
and process defects and more on verifying the effectiveness of 
processes and process controls designed to ensure food safety. FSIS 
will restructure its inspection tasks and rely on review techniques 
aimed at systems designed for preventing problems that could lead to 
the production of unsafe meat or poultry products. FSIS will carry out 
various activities to ensure that industry HACCP systems meet the 
requirements of this rule, and are functioning as designed.
    Beginning on the effective date of the regulation for a particular 
establishment, FSIS personnel will carry out a general review of an 
establishment's HACCP plan to determine its conformance with the seven 
HACCP principles. This evaluation will take place at the time of start-
up or initial implementation of the HACCP plan for new establishments. 
Subsequently, special teams of FSIS personnel will work in conjunction 
with assigned inspectors to conduct in-depth reviews, on a regular 
basis, of the establishment's current HACCP plan to verify their 
scientific validity and ongoing adequacy for preventing food safety 
hazards. Further, at any time that the HACCP plan is revised or 
amended, FSIS personnel assigned to the establishment will review the 
plan to determine if it is in conformance with regulatory requirements.
    FSIS will also carry out its verification activities by focusing on 
an establishment's ongoing compliance with HACCP-related requirements. 
Inspectors will be assigned to carry out the verification activities 
under HACCP-oriented inspection in much the same way as they receive 
their assignment schedules under the current system. A verification 
activity might include reviewing all establishment monitoring records 
for a process, reviewing establishment records for a production lot, 
direct observation of CCP controls as conducted by establishment 
employees, collecting samples for FSIS laboratory analysis, or 
verifying establishment verification activities for a process.
    As HACCP-based process control is established in meat and poultry 
establishments, with its continuous monitoring by the establishment and 
oversight by FSIS, opportunities to incorporate new technologies and 
continuously improve food safety will be more readily identified. The 
continuous monitoring and verification of production processes and 
controls by the establishment and FSIS, which is an essential feature 
of the HACCP system, will set the stage for further food safety 
improvements.
    Many commenters on the proposal expressed concern that the number 
of inspectors would decline and the quality of Federal inspection would 
diminish with HACCP implementation. FSIS expects HACCP to enhance the 
effectiveness of its meat and poultry inspection, not diminish it. 
Implementation of this final rule will clarify that the meat and 
poultry industries and FSIS have separate responsibilities for safety 
of the food supply. Industry will be required to establish process 
control systems for all forms of meat and poultry slaughter and 
processing and meet appropriate regulatory performance standards. By 
vigorous inspectional oversight of HACCP and reliance on objective test 
results and other observations to verify compliance with performance 
standards, FSIS inspectors will be better able to ensure that products 
leaving FSIS establishments are safe. Also, FSIS will be better able to 
allocate its resources to areas of greatest risk. HACCP implementation 
will move both industry and FSIS toward a more preventive approach to 
ensuring the safety of meat and poultry.
    A cross-section of consumer groups, FSIS employees, and meat and 
poultry establishments stated that each livestock and bird carcass must 
continue to be examined by trained, experienced FSIS inspectors and 
veterinarians, even under a HACCP system. They stated that carcass-by-
carcass inspection is essential to identifying animals with diseases 
that are transmissible to humans and other disease conditions causing 
animals to be unacceptable for human food. About 2,000 commenters 
maintained that HACCP is not, nor should it be, a substitute for 
carcass-by-carcass inspection by Federal inspectors.
    Carcass-by-carcass inspection is a legal requirement that binds 
both FSIS and the industry. It also addresses nonsafety considerations 
that are not addressed by HACCP. Therefore, HACCP cannot substitute for 
carcass-by-carcass examination. However, in light of HACCP, which will 
improve process control in slaughter establishments, FSIS plans to 
examine current tasks related to carcass-by-carcass inspection and 
determine what changes, if any, could improve the effectiveness of 
inspection or result in a more productive use of resources.
    Many commenters representing the meat and poultry industries argued 
that proposed pathogen reduction and HACCP system requirements layer an 
additional set of regulations and an additional program of inspection 
onto the current meat and poultry inspection system. These commenters 
recommended that FSIS review and revise or eliminate current 
regulations, directives and other FSIS guidance prior to finalizing the 
proposal as a means for ensuring they are compatible with pathogen 
reduction and HACCP requirements. Commenters stated that this review 
would not only mitigate inspection burdens imposed on industry by the 
proposal, but would facilitate the smooth implementation of pathogen 
reduction and HACCP requirements, as well.
    FSIS agrees that regulations, directives, and guidelines should be 
consistent with HACCP and is currently reviewing regulations, 
directives, and other guidance materials governing meat and poultry 
inspection. Those regulations, directives, and guidance documents that 
are inconsistent or incompatible with HACCP principles and procedures 
will be amended or revoked. This task will not only ensure consistency 
throughout the regulations, directives, and other documents, but will 
reduce duplication and help focus inspection on the most serious risks 
to food safety.

Implementation Schedule

    FSIS proposed to phase in implementation of HACCP during a 12 to 
36-month period primarily on a process-by-process basis, except that 
all ``small'' establishments (defined as establishments with annual 
sales of less than $2.5 million) would be allowed the full 36 months to 
implement their HACCP plans.
    FSIS received numerous comments on the proposed implementation 
schedule. Many commenters from meat and

[[Page 38819]]

poultry establishments said the proposed period for implementing HACCP 
was too short. These commenters requested more time to develop HACCP 
plans, train employees, and purchase or upgrade equipment. Many 
commenters requested that small businesses be granted more time to 
implement HACCP so they could amortize the costs of hazard analysis and 
plan development, equipment purchases, personnel training and records 
maintenance. A number of commenters suggested alternative timetables 
for implementation, ranging from three to fifteen years.
    Several consumer groups argued that the proposed implementation 
schedule was too slow and would compromise public health because 
serious outbreaks of foodborne illness would continue to occur while 
establishments prepare for HACCP implementation. Some industry 
commenters said they were ready to implement HACCP immediately and 
expressed concern about whether and when the FSIS inspection force 
would be prepared to oversee HACCP implementation.
    Also, several commenters requested a tiered implementation based on 
product risk. These commenters suggested that establishments which 
produce high-risk products, such as slaughter establishments or ground 
beef processors, be required to implement HACCP first and that 
establishments which produce low-risk products, such as canning 
establishments, be required to implement HACCP last.
    Also, some commenters were concerned about the proposed phase-in 
period based on different types of product categories and processes 
because contaminated meat and poultry are known to come from a variety 
of sources. Commenters said that requiring establishments to implement 
HACCP at different times for different processes within an 
establishment would confuse establishment employees, inspection 
personnel and consumers. Consequently, these commenters suggested that 
HACCP be implemented simultaneously by all establishments.
    Other commenters disputed the definition of small business used in 
the proposal. Recommendations for defining a small business included 
using fewer-than-500-employees definition developed by the Small 
Business Administration (SBA), using a definition reflecting volume of 
product or number of animals slaughtered, or using a definition based 
on the level of sales.
    In response to concerns expressed by commenters, FSIS is modifying 
the implementation schedule for HACCP. The revised implementation 
schedule is based on the size of an establishment, that is, a business 
entity producing meat or poultry products at a location. Each 
establishment is required to implement HACCP simultaneously for all 
processes, rather than on a process-by- process basis. Large 
establishments (those having 500 or more employees) are required to 
implement HACCP 18 months after publication of this final rule. 
``Small'' establishments are required to implement HACCP 30 months 
after publication. The definition of ``small'' establishment has been 
changed to correspond with SBA's size standards for business entities, 
and is now an establishment having 10 or more but fewer than 500 
employees. A new category of ``very small'' establishments (those 
having fewer than 10 employees or less than $2.5 million in annual 
sales) will have 42 months to implement HACCP. All individuals employed 
on a full-time, part-time, temporary, or other basis at a given 
establishment must be counted as employees. This requirement 
corresponds with the SBA definition of employee set forth in 13 CFR 
121.404.
    FSIS is committed to bringing the Nation's meat and poultry supply 
under HACCP systems as rapidly as possible. Phasing in HACCP 
implementation is essential due to the logistical effort required to 
manage a fundamental change in work processes, roles, and 
responsibilities for both establishments and FSIS. The revised 
implementation schedule reflects the readiness of establishments of 
varying sizes to implement HACCP, the time needed by industry to 
develop HACCP plans and train employees, and the time needed by FSIS to 
train its employees.
    The principal advantages of the revised implementation schedule are 
as follows:
    1. Large slaughter establishments account for 75 percent of 
slaughter production and thus, most of the Nation's meat and poultry 
supply will come under HACCP-based process control one year earlier 
than originally proposed. Because the greatest risk of contamination 
with pathogenic microorganisms occurs during this initial stage of 
production, FSIS considers this a significant improvement over the 
original schedule in terms of expediting progress on improving the 
safety of meat and poultry products. The revised implementation 
schedule also ensures that approximately 45 percent of processed 
products will be produced under a HACCP system within 18 months. In 
comparison, only 25 percent of processed products would have been 
produced under HACCP systems at the 18-month mark based on the proposed 
implementation schedule.
    2. By shifting initial implementation of HACCP from 12 months to 18 
months after publication of the final rule, FSIS will have sufficient 
time to manage the transition to sanitation SOP's in all 
establishments, which will begin six months after publication of this 
final rule, and to train FSIS employees to implement HACCP. FSIS does 
not believe it could manage this transition and successfully implement 
HACCP in 12 months.
    3. Eighteen months will provide ample time for the large 
establishments to comply. In fact, it is reasonable to assume that many 
of these establishments may implement HACCP before the deadline.
    4. Implementing HACCP on the basis of establishment size will be 
simpler for both FSIS and establishments and much less disruptive for 
establishments with multiple processes. Under the proposal, these 
establishments would have faced multiple implementation dates (e.g., 
establishments that both slaughter cattle and grind beef).
    5. The ``very small'' establishments will have an additional six 
months to implement HACCP. This will enable FSIS to complete the 
demonstration projects planned for ``small'' and ``very small'' 
establishments. The extra time will also ensure the availability of 
``off-the-shelf'' HACCP training programs prepared by private or 
industry-sponsored consultants. Other FSIS implementation aids, such as 
model HACCP plans, audio, video, or computer training aids, and various 
publications such as guidelines, notices and pamphlets will have 
undergone extensive development as well.

Small Business Issues

    FSIS recognizes that many smaller establishments lack the 
familiarity with HACCP that exists already in many larger 
establishments. Therefore, FSIS is planning an array of assistance 
activities that will facilitate implementation of HACCP in ``small'' 
and ``very small'' establishments.
    FSIS is developing 13 generic HACCP models for the major process 
categories, which will be available in draft form for public comment, 
and in final form, at least six months before HACCP implementation. The 
generic models are being developed especially to assist ``small'' and 
``very small'' establishments in preparing their HACCP plans. Because 
each HACCP system is developed by an individual establishment for its 
specific process and practices, the generic models will serve only as 
illustrations, rather than as

[[Page 38820]]

prescriptive blueprints for a specific HACCP plan. They should, 
however, remove much of the guesswork and reduce the costs associated 
with developing HACCP plans.
    FSIS will also conduct HACCP demonstration projects for ``small'' 
and ``very small'' establishments during the two-year period following 
promulgation of this final rule. These projects will be conducted at 
various sites to show how HACCP systems can work for various products 
under actual operating conditions. Some of these demonstrations will 
involve ``very small'' establishments and will address issues unique to 
those establishments. For instance, how does a HACCP system function in 
an establishment with only a single employee? Through these 
demonstration projects, FSIS, State inspection authorities, 
participating establishments, and the industry at large will gain added 
understanding of the problems and techniques of HACCP implementation 
and operation in ``small'' and ``very small'' establishments.
    FSIS is making available to ``small'' and ``very small'' 
establishments various HACCP materials that should assist these 
establishments in conducting their hazard analyses and developing their 
HACCP plans. These guidance materials include a ``Guidebook for the 
Preparation of HACCP Plans'' (Appendix C) and a ``Hazards and 
Preventive Measures Guide'' (Appendix D). These materials should be 
particularly useful to ``small'' and ``very small'' establishments that 
may lack the expertise for conducting hazard analyses and designing 
establishment-specific HACCP plans.
    The ``Guidebook for the Preparation of HACCP Plans'' has been 
designed to provide ``small'' and ``very small'' establishments with a 
step-by-step approach for developing a HACCP plan and includes examples 
and sample forms at each step. The Guidebook can be used alone or in 
combination with the ``Hazards and Preventive Measures Guide.''
    Because the development of an adequate HACCP plan depends on a good 
hazard analysis, the ``Hazards and Preventive Measures Guide'' develops 
HACCP Principle No. 1 in much greater detail than does the ``Guidebook 
for the Preparation of HACCP Plans.'' The hazards guide identifies 
potential biological, chemical, and physical hazards associated with a 
variety of raw materials and common ingredients, as well as major 
processes used in the meat and poultry industry. In addition, the 
hazards guide contains examples of preventive measures for common 
hazards and associated critical limits for those measures. Also 
provided are examples to illustrate approaches to implementing the 
remaining HACCP principles (e.g., monitoring, corrective actions, 
records, and verification procedures) for various hazards and critical 
control points.
    FSIS invites comments and suggestions on how it may further ease 
the transition of ``small'' and ``very small'' establishments to HACCP-
based operations.

Training Considerations

    Many commenters, including consumer groups, FSIS employees, meat 
and poultry establishments, and State governments, agreed that proper 
training in HACCP procedures and plan development is vital for 
successful HACCP implementation. A number of commenters suggested that 
joint training sessions be held for FSIS and establishment employees to 
ensure uniform understanding between inspection personnel and industry. 
Others suggested that FSIS certify acceptable training sites and 
courses of study for establishment employees to coincide with 
government employee training. However, some commenters argued that FSIS 
should not accredit training programs because to do so would limit the 
development of training programs.
    FSIS agrees that effective training of both FSIS and industry 
employees is critical to HACCP's success. FSIS also agrees that 
alternatives are needed to make training practical for various kinds of 
establishments. With these objectives in mind, FSIS is cooperating with 
the private sector to ensure that a wide variety of training options 
are available to industry and FSIS employees. For instance, FSIS is 
encouraging the International Meat and Poultry HACCP Alliance, national 
and local trade associations, State and local officials, the State 
agricultural extension services, and local colleges and universities to 
help establishments incorporate HACCP into their operations. The 
implementation conferences, discussed elsewhere in this preamble, will 
address how to achieve the goal of consistent training for FSIS and 
industry employees.
    Other plans include offering HACCP briefings to industry at many 
locations nationwide. Each session will be led by FSIS HACCP trainers, 
will be held during the evening, be open to industry and other 
interested persons, and include a question-and-answer period. FSIS 
training sessions will be limited to FSIS and State employees because 
of complex logistical and cost considerations.
    USDA's National Agricultural Library has developed and maintains 
the HACCP Training Programs and Resources Database. It is accessible 
via the Internet at ``http://www.nalusda.gov/fnic/foodborne/
foodborn.htm'' or ``gopher://gopher.nalusda.gov/11/infocntr/fnic/
foodborne/HACCP'' and provides listings of available training programs 
(workshops, satellite conferences, etc.), resources (videotapes, 
software, manuals, textbooks, etc.), and consultants (individuals and 
companies). Other Internet servers with HACCP-related information are 
operated by various firms, governments, organizations, and academic 
institutions.
    Several meat and poultry establishments also commented on funding 
for HACCP training, suggesting that FSIS or State inspection programs 
fund establishment employee HACCP training. FSIS is making every effort 
to assist establishments in making the transition to HACCP. However, 
each establishment will be responsible for training its employees.

Mandatory Versus Voluntary HACCP

    Most commenters supported the FSIS proposal to make HACCP mandatory 
in all meat and poultry establishments. However, some commenters 
requested that HACCP be voluntary rather than mandatory to alleviate 
economic burdens, especially on small businesses. Commenters further 
suggested that, at such time as a voluntary HACCP program proved 
successful, FSIS could mandate HACCP or, alternatively, market forces 
and advancing technology could be relied on to ensure its broad 
acceptance in all parts of the meat and poultry industry.
    FSIS has determined that a mandatory HACCP program is the only 
viable option that will effect adequate processing improvements in all 
establishments throughout the meat and poultry industries. Mandatory 
HACCP systems are supported by several prominent organizations, 
including the International Meat and Poultry HACCP Alliance and the 
American Meat Institute, which petitioned FSIS to initiate rulemaking 
to mandate HACCP. HACCP is now and has been voluntary; some 
establishments have it, most do not. The preamble to the proposed rule 
explained FSIS's conclusion, affirmed by most commenters, that HACCP is 
the optimal framework for targeting and reducing the many potential, 
but largely preventable, hazards associated with meat and poultry 
products. The risks of

[[Page 38821]]

foodborne illness associated with meat and poultry products will be 
minimized to the greatest extent possible only if HACCP systems are 
implemented in every establishment.

HACCP From Farm-to-Table

    A large number of commenters requested that HACCP be required 
throughout all phases of food production, from the farm to the 
consumer. These commenters asserted that HACCP plans could be developed 
by producers, slaughterers, processors, retailers, food service 
operators, and restaurants to assess and mitigate food safety risks. 
Furthermore, many commenters claimed that the majority of foodborne 
illness cases can be attributed to mishandling at the consumer level 
and FSIS should therefore strengthen consumer education as well as 
require HACCP.
    There is widespread agreement that ensuring food safety requires 
taking steps throughout the farm-to-consumer continuum to prevent 
hazards and reduce the risk of foodborne illness. FSIS is encouraging 
the active development of food safety measures to minimize public 
health hazards in animals presented for slaughter. A description of 
these farm-to-table efforts is discussed earlier in this document.

Total Quality Control (TQC) Establishments and HACCP

    One commenter requested that establishments currently operating 
under the TQC provisions (9 CFR 318.4(c) and, 381.145(c)) be allowed to 
continue to operate under modified hours. If this is not the case, 
establishments currently under TQC will incur considerable overtime 
costs. The commenter asked why, if HACCP represents an improvement over 
TQC, the establishment operating under HACCP should require more 
inspection coverage than one operating under current TQC provisions.
    This final rule does not alter current policies and practices 
regarding inspectional coverage and overtime charges in establishments 
operating under FSIS-approved TQC systems. HACCP is a safety-oriented 
system of process control that addresses food safety hazards 
differently than any current FSIS inspection systems, including TQC. 
Because TQC systems address considerations unrelated to safety, 
inspection practices developed by FSIS in connection with TQC may or 
may not be applicable to the implementation of HACCP.

Freedom of Information Act Concerns

    Most commenters stated that HACCP records should not be available 
to requestors through the Freedom of Information Act (FOIA). Some said 
HACCP records should be used for verification only and should not be 
included in government files. Others also suggested that access to 
records by FSIS inspection personnel be restricted to records that are 
necessary for HACCP compliance monitoring, such as hazard analyses, 
HACCP plans, CCP monitoring records and corrective action 
documentation. Other commenters wanted to prohibit FSIS personnel from 
copying or removing any records from the establishment. Some commenters 
requested that HACCP records be generally available to the public.
    In the preamble to the proposed regulation, FSIS stated that, as a 
preliminary matter, at least some elements of HACCP plans and 
monitoring records could be classified as trade secrets or commercial 
confidential information and may be protected from public disclosure 
under exemptions provided by FOIA and USDA and FSIS regulations 
promulgated pursuant to FOIA. FSIS specifically invited comment on the 
issue of public disclosure of HACCP records and on whether FSIS has any 
discretion about the releasability of HACCP records that it has in its 
possession.
    Recordkeeping is critical to the successful functioning of HACCP 
systems in meat and poultry establishments. FSIS will have access to 
HACCP records and any other records FSIS regulations require. While the 
records required by this final rule are clearly within the 
establishment's domain and ownership, FSIS will have access to them. 
These records, and FSIS access to them, are necessary to effectuate a 
mandatory system of preventive controls to achieve food safety.
    FSIS will continue to make use of documentation to which it has 
access when necessary to evaluate the operations of official 
establishments. Inspection personnel will normally review the records 
at establishments as part of routine HACCP oversight activities. When 
inspection personnel suspect that an establishment's HACCP system is 
not operating correctly, they will copy appropriate portions of 
establishment records, as needed, for further evaluation and possible 
enforcement action.
    An establishment will not ordinarily be required to submit copies 
of HACCP plans, verification documents, or day-to-day operating records 
to FSIS. Consequently, FSIS will not normally possess establishment 
records that may be of a proprietary nature and the issue of whether 
they are releasable under FOIA should not arise.
    Copies of establishment HACCP records may, however, be acquired by 
inspection personnel to document enforcement actions or otherwise 
assist FSIS in carrying out its responsibilities. The release by FSIS 
of information about establishments and their operations is governed by 
the FOIA. This statute requires Federal agencies to make available to 
the public agency rules, opinions, orders, records, proceedings, and 
information concerning agency organization and operations. FOIA 
provides exemptions from public disclosure for various kinds of 
information, including information concerning trade secrets and 
confidential commercial or financial information, and information 
compiled for law enforcement purposes, the release of which would be 
prejudicial or harmful to law enforcement or to the privacy rights or 
safety of individuals.
    The FOIA disclosure exemption that is most likely to be relevant is 
that covering trade secret and confidential, commercially valuable 
information. FSIS's experience in meat and poultry inspection, its 
experience with HACCP, and its understanding from the cost-benefit 
modeling and other studies undertaken in the preparation of these 
regulations is that HACCP plans will take each establishment some time 
and money to develop, and will be considered by the establishment to be 
confidential. It follows that some HACCP plans will include 
confidential, commercially valuable information, meeting the definition 
of ``trade secret.'' Plans that incorporate unique time-and-temperature 
regimens to achieve product safety, or other parameters that are 
processor-specific and that are the result of considerable research and 
effort, will ordinarily meet this definition.
    Moreover, a plan is valuable to the establishment that produces it 
for no other reason than that it took work to write. The equity in such 
a product is not readily given away to competitors. FSIS also knows 
from its own experience that establishment configurations tend to be 
unique to individual establishments, or at least have unique features. 
While generic plans will have great utility in many circumstances, they 
serve primarily as models for establishments to develop their own 
plans. Establishments will still have to expend time and money to 
tailor HACCP to their individual

[[Page 38822]]

circumstances. Thus, at least some HACCP plans or other records will 
include information to which FSIS has access but which FSIS will not be 
required to disclose publicly under FOIA.
    It should be noted, in this regard, that FOIA is not a 
confidentiality statute, but has as its primary purpose the assurance 
of the public's right of access to Government information. Agencies 
must grant requests that ``reasonably describe'' information sought in 
agency files that is not exempt from mandatory disclosure. For this 
reason, FSIS understands that it cannot make promises of 
confidentiality that exceed the permissible boundaries established 
under FOIA.

FSIS Enforcement Authority and Whistleblower Protection

    A large number of commenters requested that FSIS endorse 
enforcement tools contained in the proposed Family Food Protection Act 
(H.R. 1423, S. 515), including strengthened authority to refuse or 
withdraw inspection from official establishments, assessment by the 
Secretary of civil penalties for violations of the inspection laws, and 
protection of ``whistleblowers'' from harassment, discrimination, 
prosecution, and liability. Within the meaning of the proposed 
legislation, whistleblowers are employees or other persons who assist 
or demonstrate an intent to assist USDA in achieving compliance with 
the laws and regulations, refuse to violate or assist in violating the 
law, or are involved in commencing or testifying in a legal proceeding 
conducted by USDA.
    FSIS has determined that, while additional legislative authority 
would be helpful in certain areas, it is not needed to implement HACCP 
and the other requirements established in this final rule.
    As to whistleblower protection, many comments urged that these 
regulations include such protection for employees of meat and poultry 
slaughtering or processing establishments. Whistleblower protection is 
designed to protect workers from being fired or otherwise discriminated 
against for revealing wrongdoing by their employers. The wrongdoing in 
this case would presumably involve the forced falsification of HACCP 
records or other interference with proper operation of the HACCP 
system.
    One concern raised by these commenters and others about the 
credibility of a HACCP system is that important records can be 
falsified. It is alleged that, without whistleblower protection, it is 
much less likely that FSIS will know about falsifications. It was also 
suggested that there is a need to encourage and protect employees who 
report food safety problems or other violations of the inspection laws.
    While FSIS is confident that it can detect falsification in the 
course of its routine reviews of establishment records, coupled with 
in-plant observations, FSIS also expects that, as is now the case, it 
will be alerted by establishment employees to possible wrongdoing even 
in the absence of whistleblower protection. FSIS has relied on 
information provided by employees of the regulated industries for many 
years. From time to time, information is provided with an expectation 
that the identity of the informant will be kept confidential. FSIS 
provides this protection, to the extent possible. This policy has been 
effective.
    As a legal matter, FSIS is not empowered by the FMIA and PPIA to 
build explicit whistleblower protection into the regulations. In 
contrast to the explicit statutory whistleblower protection accorded 
Government employees, the FMIA and PPIA do not provide for 
whistleblower protection for industry employees of the kind suggested 
by some commenters, and no such explicit protection is included in the 
final rule.
    FSIS believes, however, that certain features of the HACCP 
regulations being adopted and the manner in which FSIS will inspect 
meat and poultry establishments compensate for the lack of formal 
whistleblower protection, for purposes of ensuring food safety. Most 
importantly, each establishment will be required to document, through 
records kept by establishment employees, that the critical limits 
required to ensure food safety are being met and when a failure occurs, 
proper corrective action is taken. The failure to document safety-
related failures and to take necessary corrective action violates HACCP 
regulations and the establishment will be subject to appropriate 
regulatory action. Moreover, the falsification of required HACCP 
records is a serious violation of Federal criminal law and will be 
investigated and pursued aggressively by FSIS.
    Establishments that conscientiously implement HACCP will, in the 
course of normal operations, support employee reports of HACCP 
deviations or other potential hazardous processing conditions and take 
immediate corrective action. HACCP systems in which employees with 
HACCP responsibilities are prevented or deterred from carrying out 
their responsibilities will be considered inadequate, and FSIS will 
pursue appropriate enforcement action.
    By virtue of the extensive presence of FSIS inspectors in meat and 
poultry establishments and the daily access of FSIS inspectors to HACCP 
records, FSIS will be able to verify whether problems are being 
properly documented and addressed and will be able to observe potential 
food safety problems that establishments have not found or are not 
confronting in an appropriate manner. FSIS emphasizes that undetected 
or uncorrected conditions which are likely to cause foodborne illness 
or injury should be reported immediately to FSIS by any person with 
knowledge of their existence.

Enforcement and Due Process

    A significant number of commenters raised concerns about the level 
of discretion inspection personnel will have in suspending 
establishment operations due to alleged deficiencies in either the 
design or the operation of a HACCP plan. Some urged FSIS to make clear 
to inspection personnel that such extreme actions are to be reserved 
only for situations in which continued operation of the establishment 
presents an imminent public health risk. Others strongly argued that 
operations should be suspended or inspection withdrawn when an 
establishment fails to comply with any HACCP requirements. 
Clarification was requested regarding the imposition of penalties and, 
specifically, what circumstances would warrant suspension of operations 
or withdrawal of inspection.
    Generally, the nature of the enforcement action taken will vary, 
depending on the seriousness of the alleged violation. Minor violations 
of the HACCP requirements may be recorded by Agency personnel to 
determine establishment compliance trends. Minor violations may also 
result in intensified inspection to ensure that there is no pattern of 
noncompliance and that there is no underlying food safety concern.
    Conversely, serious, repeated, or flagrant violations will result 
in immediate regulatory action, such as stopping production lines; 
applying ``U.S. Rejected'' tags to involved equipment, lines, or 
facilities; retention of product, and suspension or withdrawal of 
inspection. Because of the importance of recordkeeping to the 
functioning of HACCP systems and the production of foods that are safe 
for human consumption, FSIS views recordkeeping as a serious matter 
with potentially grave implications if records are not properly 
maintained or are falsified.

[[Page 38823]]

    Many commenters were troubled by what they perceived to be limited 
procedural due process afforded to establishments when faced with the 
suspension of inspection due to a finding that the HACCP plan is 
inadequate. FSIS agrees that all findings of inadequacy should be sound 
scientifically and legally, and that suspensions should not be invoked 
in an arbitrary manner. The optimal system would provide an appropriate 
level of protection to establishments without unnecessary delay, 
especially where no factual dispute is likely.
    Based on the comments received on this issue, FSIS has decided not 
to finalize the proposed Rules of Practice at this time. FSIS is 
interested in receiving comments and suggestions on enforcement, 
alternative dispute resolution, and due process issues, and has 
included these topics for discussion at the implementation conferences. 
On the basis of the conference discussions, FSIS will complete any 
required rulemaking covering these issues prior to the first 
implementation date for HACCP.

The Final Rule

Reorganization of HACCP Regulatory Text
    FSIS has reorganized the codified regulatory text proposed in the 
Pathogen Reduction/HACCP proposal and reworded a number of the 
provisions. These changes have been made in response to comments 
received on the proposal, for the sake of greater clarity and ease of 
use, and to conform with FSIS's planned reorganization and 
consolidation of all its meat and poultry inspection regulations. In 
general, the final HACCP regulations are more streamlined than the 
proposed provisions, organized in a more logical form, and less 
prescriptive than the proposed regulations. Also, as part of the FSIS 
and FDA effort to adopt a common approach to food safety (described in 
the January 1996 National Performance Review document ``Reinventing 
Food Regulations''), FSIS has made changes to the proposed regulatory 
text, where applicable, to be consistent with FDA's final rule on HACCP 
systems for seafood (60 FR 65096; December 18, 1995).
    To the extent possible, the HACCP requirements for both meat and 
poultry products have been consolidated in a new part 417.
    Requirements affecting grants or refusals of inspection have been 
moved to a new Sec. 304.3 and a new Sec. 381.22.
    FSIS received approximately 7,500 written and many oral comments on 
the proposed rule from meat and poultry slaughter operations, 
processors, retailers, trade and other associations, consumer 
advocates, the scientific and public health community, Federal and 
State government agencies and foreign governments, employees, and other 
interested parties. While a majority of these commenters supported the 
proposal to require adoption of HACCP by meat and poultry 
establishments, they differed widely regarding plan development, 
implementation, and related issues. Comments on the specific proposed 
regulatory requirements and FSIS's responses, follow.
HACCP Systems as a Condition of Receiving Inspection
    Proposed Sec. 326.7(a)(2) and Sec. 381.602(a)(2) would have 
permitted the issuance of a grant of inspection concurrent with a new 
establishment's development and validation of its HACCP plan. This 
provision is confusing because it is unclear how an establishment can 
develop and validate its HACCP plan ``concurrent'' with the granting of 
inspection when the HACCP plan can only be validated on the basis of 
commercial operations and the establishment can operate commercially 
only under inspection. Therefore, it would be impossible for an 
establishment to validate a HACCP plan prior to receiving a grant of 
inspection, as proposed. A number of commenters noticed this difficulty 
and requested that establishments be allowed a reasonable amount of 
time under commercial production to validate their HACCP plans.
    Commenters also disagreed with the proposed HACCP plan development 
timetable for new establishments or establishments producing new 
products or those conducting product test production runs. Some said 
that new establishments and establishments producing new products or 
conducting test runs subsequent to the applicable HACCP effective date 
should have at least six months or up to two years to finalize HACCP 
plans. Others said that all HACCP plans should be developed before 
start-up with revisions allowed within a reasonable period.
    FSIS is in basic agreement with these comments and is revising the 
basic procedures for granting inspection to allow establishments time 
to validate their HACCP plans. The provisions in Secs. 304.3(b) and 
381.22(b) require that any new establishment conduct a hazard analysis 
and develop a HACCP plan prior to being issued a conditional grant of 
inspection. The establishment must validate its HACCP plan within 90 
days after the conditional grant of inspection is issued. After FSIS 
has determined that the establishment has validated its HACCP plan, a 
permanent grant of inspection will be issued. An establishment already 
receiving inspection may produce a new product for distribution only if 
it has developed a HACCP plan applicable to the product and validates 
the plan within 90 days after beginning production of the product.
    FSIS is requiring that new facilities and products be covered by a 
HACCP plan at the time commercial production begins. Establishment 
management is expected to consider development of HACCP systems as part 
of essential pre-production decisions for new operations. 
Establishments are also expected to modify their HACCP plans as needed 
based upon experience and reported results. FSIS has determined that no 
start-up time is needed in these instances since the establishment will 
not be experiencing any transition from an old system to a new 
processing system.
    FSIS is considering what further changes may be necessary in the 
procedures for granting and inaugurating inspection at official 
establishments to better accommodate HACCP-oriented inspection. FSIS 
plans to publish a notice of proposed rulemaking on this matter in the 
near future.
Definitions
    Proposed Secs. 326.1 and 381.601 have been combined, streamlined, 
and redesignated as Sec. 417.1. Thirteen proposed definitions were 
determined to be commonly understood or unnecessary and have been 
removed. Of the seven definitions remaining, the definitions for 
``critical control point,'' ``critical limit,'' ``HACCP system,'' and 
``responsible establishment official'' have been clarified. For 
example, the definition of ``critical control point'' includes the 
phrase ``as a result'' to indicate that the prevention, reduction, or 
elimination of a food safety hazard occurs because of action taken at 
the critical control point. The definition of ``responsible 
establishment official'' has been expanded to include the individual 
with overall authority or a higher level official of the establishment.
    The revised definitions are consistent with those promulgated in 
FDA's final rule on HACCP systems for seafood. For example, FSIS has 
added a new definition to Sec. 417.1 for the term ``process-monitoring 
instrument.'' This term is defined as ``an instrument or device used to 
indicate conditions during processing at a critical control

[[Page 38824]]

point.'' FSIS determined that this definition would be helpful to 
establishments developing HACCP plans.
Hazard Analysis and HACCP Plan
    The proposal required each establishment to develop and implement a 
HACCP plan which incorporated the seven HACCP principles. A hazard 
analysis was to be conducted to identify biological, chemical and 
physical hazards and a list of steps in the process where potentially 
significant hazards could occur and the preventive measures to be taken 
were to be identified.
    Provisions relating to the hazard analysis and development of the 
HACCP plan were proposed as Secs. 326.2 and 381.602, ``Development of 
HACCP Plan,'' Secs. 326.3 and 381.603, ``HACCP Principles,'' and 
Secs. 326.4 and 381.604, ``Implementation of the HACCP Plan.'' These 
provisions have been modified and incorporated into Sec. 417.2.
    Several commenters argued that in the event the hazard analysis 
identified no significant hazards, the establishment should be exempt 
from developing HACCP plans and operating under a HACCP system. 
Commenters identified lard and meat flavoring manufacturers and canning 
operations as examples of establishments that may identify no hazards.
    To clarify the concept of potentially significant hazards, and to 
be consistent with the FDA final rule on HACCP systems for seafood, the 
final rule requires each establishment to conduct, or have conducted 
for it, a hazard analysis to determine the food safety hazards 
reasonably likely to occur in the production process. A food safety 
hazard that is reasonably likely to occur is defined as one for which a 
prudent establishment would establish controls because it historically 
has occurred, or because there is a reasonable possibility that it will 
occur in the particular type of product being processed, in the absence 
of those controls.
    FSIS agrees that if an establishment's hazard analysis reveals no 
hazards, then no HACCP plan would be required. However, FSIS is 
currently unaware of any meat or poultry production process that can be 
deemed categorically to pose no likely hazards. With regard to the lard 
and meat flavoring examples, FSIS believes that reasonably likely 
biological and physical hazards requiring control measures exist in 
establishments manufacturing these products and that, therefore, HACCP 
plans are required.
    FSIS agrees that the microbial hazards associated with canned meat 
and poultry products are eliminated by complying with the regulations 
in 9 CFR Secs. 318.300-311 and 381.300-311. These regulations are based 
on HACCP concepts and provide for the analysis of thermal processing 
systems and controls to exclude microbial hazards. Accordingly, the 
final rule provides that HACCP plans for thermally processed/
commercially sterile products do not have to address the food safety 
hazards associated with microbiological contamination if the product is 
produced in accordance with the canning regulations. However, because 
the current regulations exclusively address microbial hazards, 
processors of canned meat, meat food and poultry products must develop 
and implement HACCP plans to address chemical and physical hazards that 
are reasonably likely to occur.
    The current canning regulations contain numerous prescriptive 
features, including extensive FSIS involvement in the decisionmaking 
process, that are inconsistent with the philosophy underlying HACCP. In 
the advance notice of proposed rulemaking ``FSIS Agenda for Change: 
Regulatory Review'' (60 FR 67469; December 29, 1995), FSIS stated its 
intention to convert the canning regulations to performance standards, 
which are more consistent with HACCP. Until changes in the canning 
regulations are finalized, canning establishments do not have to 
address microbial hazards in their HACCP plans.
    The provisions of proposed Sec. 326.3(a), (a)(1), and (a)(2), and 
Sec. 381.603(a), (a)(1), and (a)(2) relating to process flow charting 
and the identification of intended uses and consumers of the product 
have been combined in the final rule into Sec. 417.2(a)(2).
    Proposed Secs. 326.2(b) and 381.602(b) would have required that any 
HACCP plan be developed with assistance of a HACCP-trained individual 
employed by the establishment, that the individual's name and resume be 
on file, and that the individual meet other prescriptive requirements. 
These requirements have been removed in response to criticism expressed 
in comments received and for reasons given below in the discussion of 
Sec. 417.7. The new Sec. 417.2(a)(1) permits someone other than an 
establishment employee to conduct the hazard analysis.
    Proposed Secs. 326.3(a) and 381.603(a) would have required a hazard 
analysis to identify any biological (including microbiological), 
physical, or chemical hazards. In Sec. 417.2(a)(3), FSIS lists ten 
areas that should be considered by an establishment when performing its 
hazard analysis. These ten areas are: natural toxins; microbiological 
contamination; chemical contamination; pesticides; drug residues; 
zoonotic diseases; decomposition; parasites; unapproved use of direct 
or indirect food or color additives; and physical hazards. This list of 
possible hazards provides more complete guidance to establishments 
conducting a hazard analysis; it responds to industry comments 
criticizing as ``vague'' the proposed definition of hazard; and it is 
also consistent with the list of hazards in FDA's final rule on HACCP 
systems for seafood.
    Proposed Secs. 326.2(a) and 381.602(a) would have required that 
establishments develop, implement, and operate a HACCP plan for each 
process conducted by the establishment, and provided a list of process 
categories subject to this requirement. Section 417.2(b) provides that 
each establishment develop and implement a HACCP plan covering each 
product produced, whenever its hazard analysis reveals one or more food 
safety hazards that are likely to occur. This requirement is 
substantively the same as the proposal.
    Section 417.2(b)(1) provides a revised list of process categories, 
while Sec. 417.2(b)(2) states that a single HACCP plan may encompass 
multiple products within a single processing category, if the hazards, 
CCP's, and critical limits are essentially the same, and as long as any 
plan features that are unique to a specific product be clearly set out 
in the HACCP plan and observed in practice. For example, an 
establishment's HACCP plan for the processing of cooked sausage might 
cover bologna, knockwurst, and frankfurters that the establishment 
produces.
    Proposed Secs. 326.2(d) and 381.602(d) would have required that the 
HACCP plan be developed in two stages, both to be completed six months 
prior to the phase-in date of the applicable process category or upon 
application for inspection or when a new process is ready for 
implementation. FSIS has eliminated these requirements because they are 
impractical.
    Proposed Secs. 326.2(d)(1) and 381.602(d)(1) would have required 
that every HACCP plan be in a format similar to the NACMCF and FSIS 
generic models. FSIS agrees with those commenters who found this 
proposed requirement to be unnecessary and too prescriptive, and has 
not included this requirement in the final rule.
    Proposed Secs. 326.3 and 381.603 set forth the seven HACCP 
principles accompanied by the corresponding requirements establishments 
must meet when developing HACCP plans. In response to comments that the 
detailed

[[Page 38825]]

provisions were unnecessary, FSIS has set forth in Sec. 417.2(c) a 
simplified list of requirements, based on the seven HACCP principles, 
to be met by establishments when developing HACCP plans. The proposed 
requirements remain, except for the following additions, unchanged.
    Two subparagraphs have been added to new Sec. 417.2(c)(2), 
clarifying the requirements for the identification of CCP's within a 
HACCP plan. This new section requires that establishments list in their 
HACCP plan the CCP's for each of the identified food safety hazards, 
including, as appropriate: (1) CCP's designed to control food safety 
hazards that could be introduced in the establishment, and, (2) CCP's 
designed to control food safety hazards that may have been introduced 
into the product before, during and after its entry into the 
establishment. In response to comments objecting to the proposed 
requirement for establishments to use a decision tree in identifying 
CCP's (proposed Sec. 326.3(b) and 381.603(b)), this requirement has 
been removed from the final rule.
    Proposed Secs. 326.4 and 381.604 would have required that a 
responsible establishment official, formerly defined as ``the 
management official located on-site at the establishment who is 
responsible for the establishment's compliance with this part,'' 
review, approve, and sign the HACCP plan. Section 417.2(d)(1) requires 
that the HACCP plan be signed by the responsible establishment 
official, defined as the individual with overall authority on-site or a 
higher level official of the establishment, possibly off-site. Further, 
in Sec. 417.2(d)(2), FSIS is correcting an oversight in the proposal by 
requiring that the HACCP plan must be signed and dated upon initial 
acceptance by the establishment and at any time the plan is modified. 
The proposal required that the responsible establishment official sign 
the plan upon completion of the hazard analysis and the development of 
the HACCP plan. The HACCP plan must also be signed and dated at least 
once each year after the required reassessment.
    Finally, FSIS explicitly states its statutory authority to enforce 
the HACCP regulations under Sec. 417.2(e), providing that if an 
establishment fails to develop and implement a HACCP plan or to operate 
in accordance with the requirements of this part, the products produced 
by the establishment may be deemed adulterated.
Corrective Actions
    Proposed Secs. 326.3(e) and 381.603(e) would have required that 
each establishment develop corrective actions to be taken when there is 
a deviation from an established critical limit. Under the proposed 
provisions, if a deviation were found, the establishment would describe 
the steps taken to identify and correct the deviation, determine how 
noncompliant product would be handled, ensure that no safety hazards 
exist after the corrective actions are taken, and define measures to 
prevent recurrence. Further, this section required that the 
establishment determine whether its HACCP plan required modification 
and, if so, to modify the plan.
    Many commenters stated that establishments should be empowered to 
make decisions on product safety. Commenters generally maintained that 
the establishment should have primary responsibility for setting the 
CCP's and critical limits and for taking corrective action when there 
is a deviation. Inspectors should verify the overall effectiveness of 
the HACCP plans, including the corrective actions taken by 
establishments. A number of commenters were concerned about the 
possibility that FSIS might take action on a product if a critical 
limit in the establishment's HACCP plan was not met, even if the 
establishment were taking corrective action under the plan. Commenters 
felt that this action by FSIS would be unwarranted. An additional 
concern was that the potential for this type of problem would be 
compounded if the establishment set a critical limit more restrictive 
than necessary for food safety to meet quality standards, for example, 
a higher cooking temperature than necessary to produce a pathogen-free 
product.
    The establishment must take corrective action for any deviation 
from a set critical limit. FSIS will verify that the establishment has 
taken appropriate corrective action as specified in their HACCP plan. 
If an establishment fails to take corrective action as specified in its 
HACCP plan, FSIS may find that the HACCP system is inadequate pursuant 
to Sec. 417.6(c). FSIS agrees that establishments should be empowered 
to make decisions regarding product disposition in accordance with 
corrective actions specified in their HACCP plans. FSIS is requiring 
(Secs. 417.2(c)(5) and 417.3) that establishments describe in their 
HACCP plans the corrective actions that will be taken if a critical 
limit is not met and assign responsibility for taking corrective 
action. Corrective actions must ensure that no product that is 
injurious to health or is otherwise adulterated as a result of the 
deviation enters commerce, that the cause of the deviation is 
identified and eliminated, that the CCP will be under control after the 
corrective action is taken, and that measures to prevent recurrence are 
established.
    FSIS recognizes that preestablished corrective actions may not 
cover every contingency and that unforeseen hazards or deviations may 
occur. Thus, Sec. 417.3 of the regulations provides a series of steps 
to be taken in such situations. These steps include segregating and 
holding affected product and conducting a review to determine the 
acceptability of the product for distribution, ensuring that any 
adulterated product or product otherwise injurious to health does not 
enter commerce, and reassessing HACCP plans to determine if any 
modification is needed.
Validation, Verification, and Reassessment
    Proposed Secs. 326.3(g) and 381.602(g) would have required that 
establishments develop procedures for HACCP plan validation by an 
adequately trained individual, and set forth the related requirements. 
Proposed Secs. 326.4 and 381.604 further detailed the validation 
requirements, stating that during the validation period, establishments 
shall conduct repeated verifications of the plan, hold frequent 
meetings with Program employees, and review records generated by the 
HACCP system. Under the proposal, establishments were to modify their 
HACCP plan following any ingredient change, product reformulation, 
manufacturing process or procedure modification, equipment change, or 
any other such change. Revalidation of an establishment's HACCP plan 
would have been required whenever significant product, process, 
deviations, or packaging changes required modification of the plan.
    Many commenters expressed confusion about the meaning of the terms 
``validation'' and ``verification'' as used in the proposed rule. The 
question of who will be responsible for validating HACCP plans was 
raised by a number of commenters. Some requested a clearer definition 
of the term ``validation'' as well as clarification of who will approve 
and verify a HACCP program. Particular concern was expressed about what 
role local inspection personnel will have in the HACCP plan development 
and approval process. Some said that FSIS should assume more 
responsibility for approving HACCP plans through a prior approval 
system; others argued that no formal acceptance or prior approval of

[[Page 38826]]

HACCP plans by FSIS should be required.
    In the final rule, FSIS has clarified the concepts of 
``validation'' and ``verification'' by delineating the responsibilities 
of FSIS and establishments in separate codified sections. The initial 
validation, ongoing verification, and reassessment procedures to be 
followed by establishments are presented in Sec. 417.4 and FSIS's 
verification procedures are presented in Sec. 417.8.
    Because prior approval of HACCP plans by FSIS would be contrary to 
redefined roles and responsibilities inherent in the HACCP philosophy, 
FSIS will not approve or validate HACCP plans before an establishment 
implements its HACCP system. Each establishment will be responsible for 
developing its HACCP plan and ensuring its adequacy.
    Commenters opposed to FSIS involvement in plan validation offered 
two suggestions: (1) establishments could use an independent third 
party, such as a processing authority or consultant with HACCP 
expertise to validate HACCP plans or (2) HACCP-trained establishment 
employees could validate plans.
    FSIS concurs. Establishments will be required to have validated 
plans and may use independent consultants, process authorities, or 
establishment employees trained in accordance with Sec. 417.7 for plan 
development and validation. FSIS is not prescribing that any particular 
validation method be used.
    Some establishments may choose to use the services of laboratories 
or processing authorities to validate their CCP's, especially if there 
are questions about the effectiveness of traditional controls, or if 
they are considering use of controls which have not been previously 
validated, such as cooking time/temperature combinations. However, many 
establishments will choose to rely on CCP's that have been 
scientifically validated and reported in the literature. In either 
case, FSIS believes that requiring individual establishments to 
validate their HACCP plan ensures that the CCP's and the overall HACCP 
plan work as intended in the establishment to reduce or eliminate 
hazards and prevent the production of unsafe food.
    One industry member observed that his company defines validation as 
documenting that a critical control point eliminates or effectively 
addresses microbiological hazards.
    FSIS agrees that validation includes documenting that critical 
control points effectively address relevant hazards, including such 
microbiological hazards as E. coli O157:H7, Salmonella, and 
Campylobacter, but emphasizes that validation is more than just the 
accumulation of microbiological data verifying each CCP. It involves 
scientifically demonstrating that a HACCP system as designed is 
effective in controlling the food safety hazards identified through the 
hazard analysis.
    One academic commenter advocated inoculation studies using 
pathogens as the best way to assure that a HACCP plan will effectively 
control microbiological hazards. Such studies would be conducted before 
HACCP implementation and should be aimed at demonstrating that selected 
CCP's are appropriately monitored to control specific pathogens. The 
studies would be performed under controlled conditions in off-site 
laboratories or pilot establishments. One advantage of this approach, 
according to the commenter, would be to permit validation studies to be 
conducted by trade associations and other industry groups on a 
collective basis in a way that could benefit both large and small 
establishments.
    FSIS agrees that validation of CCP's is an important part of HACCP 
plan validation, and that laboratory inoculation studies as suggested 
by the commenter can make an important contribution in appropriate 
cases. Inoculation studies can demonstrate the effectiveness of 
particular controls in addressing particular hazards under experimental 
conditions, and can produce data that can be relied upon by many 
establishments to support plan validation. In no case, however, would a 
laboratory inoculation study or any laboratory study be sufficient by 
itself to validate a HACCP plan. An important element of validation is 
the identification or development of data which show that the 
establishment can apply the process or control to get the anticipated 
effect under actual in-plant operational conditions. For some well-
established, widely used processes or technologies, in-plant validation 
can be accomplished by combining existing scientific data from 
laboratory studies, the scientific literature, or other sources, with 
the results of commercial trials using recognized protocols. Where 
processes are well-documented in the scientific literature, it is not 
necessary to require inoculation studies or any other research effort 
as part of the validation process. However, an establishment 
introducing a new technology, applying standard technology in an 
unusual way, or lacking experience with a technology, would have to 
undertake more extensive scientific and in-plant validation of its 
HACCP plan under commercial operating conditions.
    Data assembled to validate a HACCP plan are usually of two types: 
(1) theoretical principles, expert advice from processing authorities, 
scientific data, or other information demonstrating that particular 
process control measures can adequately address specified hazards, such 
as studies establishing the temperatures necessary to kill organisms of 
concern; and (2) in-plant observations, measurements, test results, or 
other information demonstrating that the control measures, as written 
into a HACCP plan, can be operated within a particular establishment to 
achieve the intended food safety objective. This means that the data 
used to validate a HACCP plan may be derived from various sources, 
including the scientific literature, product testing results, 
experimental research results, scientifically based regulatory 
requirements, FSIS guidelines, computer-modeling programs, and data 
developed by process authorities. The nature and quantity of 
information required to validate a HACCP plan will vary depending on 
factors such as the nature of the hazard and the control measures 
chosen to address it.
    FSIS believes that validation data for any HACCP plan must include 
some practical data or information reflecting an establishment's actual 
early experience in implementing the HACCP plan. This is because 
validation must demonstrate not only that the HACCP plan is 
theoretically sound, but also that this establishment can implement it 
and make it work. For example, steam vacuuming has been scientifically 
demonstrated to be effective in removing visible contamination and 
associated bacteria from carcass surfaces. A slaughtering establishment 
using the technology as a control measure at a CCP, however, would 
still have to demonstrate its ability to use the technology effectively 
at the CCP.
    Establishment verification is intended to show that the HACCP 
system is actually working effectively on a day-to-day basis. 
Verification also includes repeatedly reviewing and evaluating the 
various components of the system. Verification activities include 
checking the adequacy of the critical limits; reviewing monitoring and 
recordkeeping procedures (as distinguished from monitoring the CCP's), 
and evaluating the adequacy of corrective actions.
    One consumer group stated that FSIS should require that 
establishments identify the specific microbiological hazards that their 
HACCP plans are

[[Page 38827]]

designed to address, and validate and verify the plans using pathogen-
specific testing to ensure that establishments control these hazards.
    FSIS agrees that establishments must identify the specific 
microbiological hazards their HACCP plans are designed to address and 
that the plan must be initially validated and continually verified as 
effective in addressing those hazards. FSIS also agrees that pathogen-
specific testing can play an important role in both initial validation 
and verification.
    For example, in validating the adequacy of a beef slaughter HACCP 
plan addressing the hazard posed by E. coli O157:H7, laboratory 
inoculation studies involving pathogen-specific testing could be used 
to validate the effectiveness of the specific control measures that an 
establishment is considering for incorporation in its HACCP plan. As 
discussed above, to complete the validation of the control measures for 
E. coli O157:H7, the establishment would also be required to 
demonstrate that the experimentally validated measures can be 
successfully carried out under actual operating conditions, but, for E. 
coli O157:H7 on going verification is unlikely to include in-plant 
testing for the pathogen due to its relatively infrequent occurrence.
    In-plant testing to verify a control measure may be appropriate 
with other pathogens, however. For example, a poultry slaughter 
establishments would be required to validate and verify the 
effectiveness of its HACCP plan in addressing the hazards posed by 
Salmonella and Camplylobactor. Depending on the nature of the control 
measures the establishment selects, in-plant pathogen testing could be 
a necessary and practical component of an on-going verification for 
these pathogens as they are present in sufficient numbers to make in-
plant testing feasible and informative. FSIS intends to work closely 
with industry at large and with specific establishments in particular 
to ensure that HACCP plans are adequately validated and verified for 
microbial pathogens of public health concern.
    Verification of HACCP plans by establishments is designed to 
demonstrate that the HACCP plan is accomplishing process control and 
resulting in the production of safe food on a continuing basis. 
Verification is distinct from ongoing establishment monitoring, which 
is designed to provide a record showing that the written HACCP plan is 
being followed. Establishment verification activities should provide 
practical results specific to the operation of its HACCP plan, and can 
include review of CCP-monitoring records; review of corrective action 
records; calibration of process-monitoring instruments; collection of 
either in-line or finished product samples for microbiological, 
chemical, or physical analysis; and direct observations of monitoring 
activities and corrective actions. Frequencies for conducting 
verification activities will vary, depending on various factors, such 
as the type of process and volume of products, the results of prior 
verification activities, consistency of conformance with the HACCP 
plan, how deviations are handled, and the results of any sampling 
activities.
    The record-verification could include determining whether the 
critical limit for the CCP, as called for in the HACCP plan, matches 
the critical limit indicated in the records. The verification could 
also involve checking to assure that the critical limit as set in the 
establishment's HACCP plan is adequate to prevent a hazard. For 
example, this check might involve determining whether the random 
variations inherent in any process are within the limits (temperature 
ranges, physical contamination) set for the process, and that the 
critical limit is never exceeded or, further, that the probability that 
the critical limit might ever be exceeded is extremely low.
    The visual observations and records verification could include, in 
addition to seeing that the records are being properly maintained, 
assuring that corrective actions have been taken whenever any 
deviations have occurred and that, when taken, the corrective actions 
were sufficient to solve the problem.
    FSIS has made two minor changes from the proposed validation and 
verification requirements. First, FSIS has removed the proposed 
requirement that during validation an establishment hold frequent 
meetings with Program employees. FSIS recognizes that frequent meetings 
may not be necessary or appropriate. Also, Sec. 417.4(a)(2) provides 
that the establishment's ongoing verification activities include direct 
observation of monitoring activities and corrective actions, review of 
records, and calibration of process-monitoring instruments. An 
establishment calibrates its monitoring instruments to determine 
whether they are functioning properly.
Reassessment
    The proposed rule would have required that establishments 
revalidate the HACCP plan whenever significant product, process, 
deviations, or packaging changes required modification of the plan.
    A consumer group stated that establishments should be required to 
examine their plans on a regular basis, whenever any new equipment is 
introduced, new employee training is implemented, or for any other 
significant change in the processing environment. The commenter further 
stated that revalidation should be required of establishments every 
three years even if there has been no significant change in operations. 
Most commenters generally agreed that the industry has the primary 
responsibility to review and modify HACCP plans when necessary and that 
the review and modification process should be flexible.
    FSIS agrees that HACCP plans should be reexamined periodically and 
that the review and modification process should be flexible. The final 
rule requires that each establishment reassess the adequacy of its 
HACCP plan at least annually, and whenever any changes occur that could 
affect the hazard analysis or alter the HACCP plan (Sec. 417.4(a)(3)). 
These changes may include, but are not limited to, changes in: raw 
materials or source of raw materials; product formulation; slaughter or 
processing methods or systems; production volume; personnel; packaging; 
finished product distribution systems; or the intended use or consumers 
of the finished product. The reassessment must be completed by an 
individual trained in accordance with Sec. 417.7. Immediate 
modification of the plan is required if the reassessment reveals that 
the plan is no longer adequate to meet the requirements of part 417. 
FSIS is also requiring that an establishment that does not have a HACCP 
plan reassess its hazard analysis whenever a change occurs that could 
reasonably affect whether a food safety hazard exists.
    FSIS considers annual reassessment appropriate because, as 
commenters have noted, HACCP plans are dynamic and evolving. HACCP 
plans may be modified several times during the months after they are 
first implemented. Further, repeating the entire validation process may 
not be necessary to ensure that the HACCP system is functioning 
correctly after modification.
    The intent of this provision is to provide for periodic 
modification of the HACCP plan to ensure that it is continuously 
effective in controlling and preventing food safety hazards. This 
intent is supported by comments received from various sectors of the 
public. The commenters tended to see periodic review and modification 
of HACCP plans as both desirable and

[[Page 38828]]

expected and that periodic review and modification would allow the 
establishment to apply its experience to continually improve process 
controls.
    FSIS believes that ``reassessment'' encompasses the different types 
of evaluation, from reanalyzing the verification procedures for an 
updated CCP to repeating the validation procedures set forth in 
Sec. 417.4, that may be necessary.
FSIS Verification
    Verification of HACCP plans is also a regulatory responsibility. 
FSIS will verify that HACCP plans comply with the requirements of Part 
417 and have been validated by the establishment. Potential 
verification activities by FSIS may include, but are not limited to, 
sampling activities (targeted and non-targeted, marketplace, rapid 
screening tests for chemical residues); hands-on verification 
(organoleptic inspection, use of temperature or other monitoring 
devices); and review of establishment monitoring records. The frequency 
of FSIS verification activities will vary, depending on a number of 
factors such as the establishment's past performance, risk inherent in 
the processes or products, quantity of product, and likely uses.
    A consumer group stated that as part of its verification 
activities, FSIS should review all pathogen data generated by the 
establishment to determine the adequacy of the establishment's 
conclusions regarding pathogen control. FSIS plans to undertake 
extensive and varied activities to verify that a HACCP plan is working 
as intended, including review of data generated or relied on by the 
establishment to validate its HACCP plan.
    Proposed Secs. 326.7(b) and 381.607(b) set forth FSIS's 
responsibilities with respect to verification activities. These 
provisions have been slightly revised for clarity and are consolidated 
in Sec. 417.8.
Records
    Proposed Secs. 326.6(b) and 381.606(b) listed the types of records 
every establishment would have been required to maintain regarding 
their operations under HACCP. The list included the written HACCP plan, 
hazard analysis, records associated with CCP monitoring, corrective 
actions, verification procedures and results, product codes, identity, 
and slaughter production lot, the dates of the records, and supporting 
documentation for the various features of the HACCP plan. FSIS also 
proposed to require a preshipment review of processing and production 
records associated with the HACCP plan to ensure that the records were 
complete, that all critical limits were met, and, if applicable, that 
corrective actions were taken. The review was to be performed by 
someone other than the person who created the records, preferably by a 
HACCP-trained individual, or by the responsible establishment official. 
FSIS considers the preshipment record review a routine verification 
function under HACCP principle No. 7.
    FSIS also proposed that establishments retain all required records 
on site at all times, except those records concerning monitoring CCP's, 
corrective actions, and verification procedures were to be retained at 
the establishment for no less than one year, and for an additional two 
years at the establishment or other location from which the records 
could be made available to Program employees.
    Regarding the preshipment review of records, several small 
establishments commented that there may not be a person other than the 
person who created the record available to conduct the preshipment 
review. Several large establishments were concerned that a HACCP-
trained individual may not be available to conduct the preshipment 
review. FSIS has modified this requirement by stating that the 
preshipment review shall be conducted by someone other than the person 
who produced the records where practicable. Also, FSIS has retained the 
provision that the review be conducted preferably by an individual 
trained in accordance with Sec. 417.7 or the responsible establishment 
official.
    Some commenters recommended that FSIS allow the use of electronic 
or computerized recordkeeping systems to ease the burden of the 
proposed recordkeeping requirements. In response to these comments, 
FSIS has added a new Sec. 417.5(d) which provides for the maintenance 
of data and information on computers, as long as controls are 
implemented by the establishment to ensure the integrity of the data 
and signatures.
    Commenters also raised concerns regarding proposed record retention 
requirements, maintaining that keeping HACCP records for a minimum of 
three years would be excessive. Commenters requested flexibility in 
deciding how long to retain records; many stated that retention should 
be based on product shelf-life. In response to these commenters, FSIS 
has modified this requirement to provide that records required by 
Sec. 417.5(a)(3) be retained at the establishment for one year if they 
pertain to slaughter activities or refrigerated products, and for two 
years if they pertain to frozen, preserved, or shelf-stable products.
    To further ease the recordkeeping provisions for establishments, 
FSIS will permit the off-site storage of records required by 
Sec. 417.5(a)(3) that are over 6 months old if the records can be made 
available to Program employees within 24 hours of the request. The 
records required by Sec. 417.5 (a)(1) and (a)(2), however, are not 
eligible for off-site storage.
    Proposed Secs. 326.6 and 381.606 would have provided that records 
be made available to Program employees. Section 417.5(f) clarifies that 
all records required by part 417 be available to Program employees for 
review and copying.
    For clarity, FSIS has reworded the recordkeeping provisions to 
require that the establishment maintain the written hazard analysis and 
all supporting documentation, the written HACCP and all decisionmaking 
documents associated with the selection and development of CCP's and 
critical limits, and documents supporting both the monitoring and 
verification procedures selected and the frequency of those procedures. 
Records documenting the monitoring of CCP's and critical limits, 
corrective actions, verification procedures and results, product 
code(s), product name or identity, or slaughter production lot must 
also be maintained. Each record must include the date the record was 
made. To be consistent with FDA's final rule on HACCP systems for 
seafood, FSIS has also added a requirement that records relating to the 
calibration of process-monitoring instruments be maintained.
Training
    FSIS proposed two definitions related to training: ``HACCP-trained 
individual'' and ``recognized HACCP course.'' ``HACCP-trained 
individual'' was defined as ``a person who has successfully completed a 
recognized HACCP course in the application of HACCP principles to meat 
or poultry processing operations, and who is employed by the 
establishment. A HACCP-trained individual must have sufficient 
experience and training in the technical aspects of food processing and 
the principles of HACCP to determine whether a specific HACCP plan is 
appropriate to the process in question.'' A ``recognized HACCP course'' 
was defined as ``a HACCP course available to meat and poultry industry 
employees which satisfies the following: consists of at least 3 days, 1 
day devoted to understanding the seven principles of HACCP, 1 day 
devoted to applying these concepts to this and other regulatory 
requirements of FSIS, and 1 day devoted

[[Page 38829]]

to beginning development of a HACCP plan for a specific process.''
    Some commenters thought that defining a HACCP-trained individual 
was unnecessary, that the role of such a person in operating HACCP 
systems should be analogous to the role of the processing authority in 
canning operations.
    A few commenters questioned the effectiveness of the proposed 
three-day training requirement stating it would not sufficiently 
qualify a person to implement or operate a HACCP system. Some 
commenters asserted that the detailed course composition with no FSIS 
certification of courses was inadequate and too rigid. Others insisted 
that what is needed is a common understanding of the basic principles 
of HACCP and of how HACCP can be applied to specific processes and 
establishments, with no FSIS certification of courses.
    FSIS has revised the regulations, which are now codified in 
Sec. 417.7, to simplify the proposed training requirements. The 
proposed definition and requirements for a HACCP-trained individual 
have been removed. Section 417.7 requires that individuals performing 
certain functions must have successfully completed a course in the 
application of the seven HACCP principles to meat and poultry product 
processing, including a segment on the development of a HACCP plan for 
a specific product. Only those individuals who meet the training 
requirements may perform the following functions:
    <bullet> Development of the HACCP plan as required by 
Sec. 417.2(b);
    <bullet> Reassessment and modification of the HACCP plan as 
required by Sec. 417.3 and/or Sec. 417.4(a)(3).
    The rule has been modified to set a basic standard for HACCP 
training while preserving the flexibility needed by industry to 
implement HACCP systems effectively. The provisions of Sec. 417.7 are 
consistent with FSIS's view that training is central to the success of 
HACCP, that there are many avenues for HACCP training needs, and that 
responsible establishment officials are in the best position to 
determine the training needs for each establishment.
Adequacy of HACCP Plans
    The proposed rule stated that a HACCP plan could be found invalid 
if it does not meet the regulatory requirements, if HACCP records are 
not being maintained to validate the plan or verify process control 
under the plan, or if a processing failure results in production of 
adulterated product.
    The provisions of the final rule relating to the criteria for 
finding a HACCP plan inadequate are essentially the same as in the 
proposal, except that the term ``invalid'' has been replaced with 
``inadequate'' for clarity. Also, the final rule states that a HACCP 
plan may be found to be inadequate if establishment personnel are not 
performing tasks specified in the HACCP plan. One change from the 
proposal concerns the correction of HACCP systems found inadequate 
because of product adulteration. Under the proposed 
Secs. 326.7(c)(3)(ii) and 381.607(c)(3)(ii), the establishment would 
have been required to submit to FSIS, among other things, a written 
plan for chemical or microbiological testing by an external laboratory 
of finished product produced under the modified HACCP plan to show that 
the modified plan corrected the problem. The final rule is more 
flexible because decisions regarding the appropriateness of the HACCP 
system modifications are made by the establishment.
    FSIS will verify that HACCP plans are adequate. The procedure for 
determining the adequacy of the HACCP plan will not be a one-step 
process. Instead, FSIS will take a variety of actions including 
reviewing the HACCP plan and associated records, directly observing the 
HACCP system in operation, and assessing the adequacy of corrective 
actions. After a thorough review is conducted, FSIS will determine 
whether a HACCP plan is adequate. If a plan is found to be inadequate, 
FSIS will take appropriate regulatory action.

III. Sanitation Standard Operating Procedures

The Proposed Rule

    FSIS proposed that all meat and poultry establishments be required 
to develop, maintain, and adhere to written sanitation standard 
operating procedures (Sanitation SOP's). The proposal was based on 
FSIS's belief that effective establishment sanitation is essential for 
food safety and to successful implementation of HACCP. Insanitary 
facilities or equipment, poor food handling practices, improper 
personal hygiene, and similar insanitary practices create an 
environment conducive to contamination of products. There are direct 
and substantial links between inadequate sanitation and the 
contamination of meat and poultry products by pathogenic bacteria. FSIS 
tentatively concluded that Sanitation SOP's were necessary because they 
would clearly define each establishment's responsibility to 
consistently follow effective sanitation procedures and would 
substantially minimize the risk of direct product contamination and 
adulteration.
    FSIS also had determined that Sanitation SOP's would improve the 
utilization of FSIS Inspection Program resources by refocusing FSIS 
sanitation inspection on the oversight of establishment prevention and 
correction of conditions that cause direct product contamination or 
adulteration. After Sanitation SOP's were in place, Agency inspection 
personnel would spend less time enforcing detailed sanitation 
requirements and directing the correction of problems after they occur. 
Instead, FSIS inspectors would focus on oversight of an establishment's 
implementation of Sanitation SOP's and on taking appropriate regulatory 
action when an establishment's Sanitation SOP's were not properly 
executed or when product contamination or adulteration was imminent, 
directly observed, or probably had occurred.
    The concepts underlying the proposed requirements for Sanitation 
SOP's are important and new. In the past, FSIS has not clearly 
articulated the responsibility every establishment has to ensure that 
sanitation requirements are met every day, both before and during 
operations. Although the majority of meat and poultry establishments 
maintain adequate sanitary conditions, some establishments have 
significant sanitation problems that can be resolved only through more 
clearly defining establishment responsibility and accountability for 
the daily observance of sound sanitation practices.
    The proposed requirements for Sanitation SOP's were the result of 
many years of observations by FSIS of establishment sanitation and 
management practices. The persistence of insanitary conditions within 
some meat and poultry establishments was documented in the ``1,000 
Plant Review,'' conducted by FSIS between September 1993 and February 
1995. This project involved unannounced visits to 1,014 inspected 
establishments during which operations were observed and deficiencies 
noted. More than 60 percent of all deficiencies documented by the 
review involved establishment sanitation. The distribution of 
sanitation problems was not, however, uniform in the establishments 
sampled. Fewer than half those establishments visited accounted for 90 
percent of the sanitation deficiencies. Data collected through FSIS's 
Performance Based Inspection System similarly documents that sanitation 
is the most frequent deficiency noted by inspection personnel in 
routine establishment visits.

[[Page 38830]]

    Through analysis of this information, FSIS determined that the 
difference between establishments with consistently sanitary conditions 
and those with chronic sanitation deficiencies is often that the better 
performing establishments have effective quality control and sanitation 
programs, including written Sanitation SOP's, while the marginal 
establishments do not. As a means of bringing all establishments to a 
consistently acceptable level of sanitation, as well as to clarify the 
respective roles of establishments and FSIS in achieving that goal in 
each establishment, FSIS proposed that every meat and poultry 
establishment develop, maintain, and adhere to written Sanitation 
SOP's.
    FSIS proposed that Sanitation SOP's cover the daily preoperational 
and operational sanitation procedures that the establishment would 
implement to prevent direct product contamination or adulteration. 
Additionally, establishments would be required to identify the 
establishment officials who would monitor daily sanitation activities, 
evaluate whether the Sanitation SOP's are effective, and take 
appropriate corrective action when needed. Also, each establishment 
would be required to make daily records showing completion of the 
procedures in the Sanitation SOP's, any deviations and corrective 
actions taken, and maintain those records for a minimum of six months. 
Further, an establishment's Sanitation SOP's and records were to be 
made available to FSIS for verification and monitoring. Finally, the 
proposal provided that any equipment, utensil, room or compartment 
found by an inspection program official to be not in compliance with 
the Sanitation SOP's or insanitary would be tagged ``U.S. Rejected,'' 
and could not be used until it had been reinspected and passed.
    FSIS solicited comments on the proposed regulatory requirements for 
Sanitation SOP's. FSIS also requested comments on how Sanitation SOP's 
should clarify the responsibilities of establishments and what role 
inspection personnel should play in authorizing daily startup of 
operations. Comments also were requested on whether certain Good 
Manufacturing Practices (GMP's) or other sanitation practices should be 
mandatory elements of the Sanitation SOP's.
    The majority of the comments addressing the proposed Sanitation 
SOP's provisions expressed support. Many commenters, however, expressed 
concern about the lack of detail in the proposal regarding the required 
contents of an establishment's Sanitation SOP's and about how 
Sanitation SOP's would be enforced by inspectors. The comments, both 
written and oral, and FSIS's responses are discussed in the 
``Comments'' section, which follows the description of the final rule.

The Final Rule

    After careful consideration of the comments, FSIS is promulgating 
requirements for Sanitation SOP's, essentially the same as proposed, 
though with several changes and additions for both clarity and to grant 
establishments greater flexibility in meeting the Sanitation SOP's 
requirements.
    As proposed, all inspected establishments shall develop, implement, 
and maintain written Sanitation SOP's. The Sanitation SOP's shall 
describe all procedures and establishment conducts daily to prevent 
direct contamination or adulteration of product(s). FSIS has clarified 
that Sanitation SOP's also shall specify the frequency with which each 
procedure in the Sanitation SOP's is to be conducted and identify the 
establishment employee(s) responsible for the implementation and 
maintenance of such procedure(s). While the employee responsible for 
implementation and maintenance of procedures in the Sanitation SOP's 
may be the employee who actually performs such activities, he or she 
instead may be the employee in charge of ensuring that the sanitation 
procedures are carried out. All that is required is that the Sanitation 
SOP's identify the employee(s) responsible for implementation and 
maintenance of the procedures in the Sanitation SOP's. The 
establishment does not need to necessarily identify the employee(s) who 
will actually perform the sanitation procedures. Also, an 
establishment's Sanitation SOP's may have more than one employee 
responsible for implementation and maintenance of sanitation 
procedures. For example, one employee may be responsible for pre-
operational procedures and another may be responsible for operational 
procedures. The rule provides such flexibility.
    Further, FSIS is clarifying in this final rule that establishments 
must explicitly identify pre-operational sanitation procedures in their 
written Sanitation SOP's, distinguishing them from sanitation 
activities to be carried out during operations. This will assist both 
the establishment and FSIS in identifying which sanitation procedures 
are to be carried out each day prior to start-up of operations.
    FSIS is also requiring that Sanitation SOP's be signed and dated by 
``the individual with overall authority on-site or a higher level 
official of the establishment,'' and that the signature shall signify 
that the establishment will implement the Sanitation SOP's. This new 
language grants establishments greater flexibility than did the 
proposed requirement that ``the establishment owner or operator'' be 
responsible for implementation of Sanitation SOP's. Additionally, this 
final rule specifies that Sanitation SOP's must be signed upon 
initiation and upon any modification.
    As in the proposal, the format and content of Sanitation SOP's are 
not specified in the final regulations. Because there are many types of 
inspected establishments that will achieve the required sanitary 
conditions in different ways, this rule gives establishments 
flexibility to customize their sanitation plans. Each meat and poultry 
establishment must analyze its own operations and identify possible 
sources of direct contamination that must be addressed in its 
Sanitation SOP's.
    As proposed, each establishment is required to conduct the pre-
operational and operational procedures as specified in the Sanitation 
SOP's, monitor the conduct of the procedures, and routinely evaluate 
the content and effectiveness of the SOP's and modify the Sanitation 
SOP's accordingly. The Sanitation SOP's must be kept current. The 
establishment must evaluate and modify Sanitation SOP's as needed in 
light of changes to establishment facilities, personnel, or operations 
to ensure they remain effective in preventing direct product 
contamination and adulteration. As upon initial implementation, 
Sanitation SOP's must be dated and signed by the individual with 
overall authority on-site or a higher level official of the 
establishment following any modification.
    Also as in the proposal, FSIS is requiring that each establishment 
initiate corrective action when either the establishment or FSIS 
determines that Sanitation SOP's or their implementation may have 
failed to prevent direct product contamination or adulteration. The 
requirements regarding corrective actions have been more thoroughly 
explained, however, and now specify that corrective actions shall 
include ``procedures to ensure appropriate disposition of product(s) 
that may be contaminated, restore sanitary conditions, and prevent the 
recurrence of direct contamination or adulteration of product(s), 
including

[[Page 38831]]

appropriate reevaluation and modification of the Sanitation SOP's and 
the procedures specified therein.''
    This final rule also adopts the provision in the proposal requiring 
establishments to keep daily records documenting that sanitation and 
monitoring procedures listed in the Sanitation SOP's are performed. 
Establishments also must maintain records documenting any corrective 
actions taken to prevent direct contamination or adulteration of 
products, or when the establishment determines or FSIS notifies the 
establishment that its Sanitation SOP's are inadequate. FSIS has 
clarified that such records must be initialed and dated by the 
designated establishment employee(s) responsible for the implementation 
and monitoring of the Sanitation SOP's procedures.
    In response to comments, FSIS has revised the recordkeeping 
requirements to allow for computer maintenance of records, as long as 
establishments implement controls to ensure the integrity of the 
electronic data. FSIS recognizes that many establishments currently use 
computers for maintaining a variety of types of information, including 
sanitation data. It would be impractical and burdensome to prohibit 
these establishments, or others wishing to use computers, from using 
computers to record and store required sanitation data.
    FSIS proposed that establishments must maintain sanitation records 
for a minimum of six months, but did not specify whether these records 
had to be stored on-site. Several commenters expressed concern about 
the physical location of establishment sanitation records and 
questioned whether sanitation records must be maintained in the 
establishment.
    FSIS requires unimpeded access to all establishment sanitation 
records for oversight and enforcement purposes; these records are to be 
an integral part of the Agency's inspection activities. FSIS 
anticipates that, for most establishments, these records will not be 
voluminous and will not create a significant storage problem. However, 
the Agency recognizes that space may be limited at certain inspected 
facilities and has revised this requirement to allow establishments to 
retain records off-site, provided they are not removed from the 
establishment for at least 48 hours following completion and they can 
be provided to FSIS personnel within 24 hours of being requested.
    In this final rule, FSIS is clarifying that it will verify that the 
Sanitation SOP's are being implemented and maintained, and that they 
are effective. FSIS inspectors will ensure not only that an 
establishment is complying with the requirement to develop, implement, 
and maintain Sanitation SOP's, and to maintain daily records for them, 
but also that the Sanitation SOP's are in fact working. Inspectors will 
review the Sanitation SOP's, the daily records, the conduct of 
procedures specified in the Sanitation SOP's, and the sanitary 
conditions themselves.
    The failure by an establishment to comply with the Sanitation SOP's 
regulations may initiate regulatory action. The full array of 
compliance tools includes process deficiency reports, tagging of 
equipment or areas, retention of product, letters of warning, and 
suspension and withdrawal of inspection. The nature of FSIS's response 
will depend on the circumstances. Minor omissions or errors in 
Sanitation SOP's documentation, not symptomatic of larger ``system'' 
problems, will result in regulatory action commensurate with the 
severity of the violation. For example, process deficiency reports 
might be issued to direct corrective action. However, a pattern of 
violations of the Sanitation SOP's provisions would lead to additional 
responses, with persistent and serious failures resulting in suspension 
or withdrawal of inspection from the establishment. Suspensions and 
withdrawals would be made in accordance with applicable rules of 
practice for those proceedings.
    If FSIS determines that an establishment's Sanitation SOP's fail to 
include procedures to prevent direct product contamination or 
adulteration or that required records are not being kept, the Agency 
may tag affected facilities and equipment and suspend inspection until 
the failure is remedied. Because the tagging of insanitary facilities 
and equipment is based on current statutory authority, the specific 
regulatory provisions for tagging in the proposal are not retained in 
this final rule.
    Verification and compliance activities under the Sanitation SOP's 
provisions are distinguishable from actions taken as a consequence of a 
finding of product adulteration under the sanitation requirements 
elsewhere in the regulations. As a practical matter, however, such 
findings are likely to be connected. A finding of deficient Sanitation 
SOP's or Sanitation SOP's records may prompt additional inspection 
activity directed at determining whether or not product contamination 
or adulteration has occurred. If it has, FSIS will take appropriate 
action to prevent adulterated product from entering commerce and, where 
necessary, seek recall of product that has already entered commerce.
    Finally, the Sanitation SOP's requirements of this final rule are 
set out in a new Part 416, Sanitation. These provisions are formatted 
differently from the proposal to comport with FSIS's announced project 
to reform, reorganize, and recodify the meat and poultry regulations. 
This regulatory reform project is well underway, and will, among other 
things, eliminate unneeded regulations by combining, to the extent 
possible, the currently separate meat and poultry regulations. New Part 
416, like new part 417 on HACCP, covers both meat and poultry products. 
Part 416 will be expanded and supplemented as the Agency proceeds with 
its initiative to review, reform, and reorganize existing FSIS 
regulations concerning sanitation.

Comments and Responses

General
    Support for the proposed requirements for Sanitation SOP's was 
expressed by a wide range of commenters. Most supporters agreed that 
establishment sanitation is essential to product safety and that every 
meat and poultry establishment should be required to have a written 
sanitation plan. Those who opposed mandatory Sanitation SOP's argued 
that current sanitation regulations would be adequate if they were 
better enforced, that Sanitation SOP's would be no more than a 
paperwork exercise, and that they would be an additional burden on 
establishments. FSIS strongly disagrees with the notion that Sanitation 
SOP's will be a mere ``paperwork exercise,'' and believes this 
regulation will, in fact, result in improved sanitation and provide for 
more effective enforcement of the sanitation requirements.
    Substantial evidence exists that insanitary facilities or 
equipment, poor food handling, improper personal hygiene, and similar 
insanitary conditions create an environment in which products become 
contaminated with microorganisms, including pathogens. While sanitation 
has improved greatly throughout the industry over the years, some 
individual establishments still have difficulty getting their 
facilities and equipment ready to start operations each day and keeping 
conditions sanitary during establishment operations. FSIS affirms that 
proper sanitation is an important and integral part of every food 
process and a fundamental requirement of the inspection laws that the 
Agency enforces.

[[Page 38832]]

    In the past, FSIS has enforced the sanitation requirements 
primarily through a combination of prescriptive sanitation regulations, 
detailed guidance materials, and direct, hands-on involvement by 
inspectors in day-to-day pre-operational and operational sanitation 
procedures in inspected establishments. This system achieved sanitation 
goals on a daily basis in individual establishments, but at a 
relatively large public cost because it encouraged establishments to 
shift accountability for sanitation to the FSIS inspector. For example, 
in the past, FSIS inspectors have taken responsibility for checking 
sanitation in every slaughter establishment before it begins daily 
processing. In extreme cases, inspectors have led daily ``bucket 
brigades'' of slaughter establishment employees through pre-operational 
establishment cleanup. In these circumstances, FSIS has, in effect, 
taken responsibility for establishment sanitation conditions. The 
Sanitation SOP's requirement is intended to end this practice. 
Sanitation SOP's make it clear that responsibility for identifying and 
conducting procedures needed to maintain sanitary conditions rests with 
the establishment, not with FSIS.
    Sanitation SOP's are an inspection tool. They will help individual 
inspectors focus their oversight in an establishment on those 
conditions that pose a risk of direct product contamination or 
adulteration, that is, on conditions which pose the greatest 
adulteration hazards to products subject to inspection in that 
establishment. The effectiveness of each establishment's Sanitation 
SOP's in achieving acceptable sanitation will be subject to continuing 
verification by FSIS inspectors through direct observation of 
conditions in the establishment. It is expected that, over time, 
inspectors in most establishments will increasingly be able to rely on 
a review of daily Sanitation SOP's records to determine whether 
establishments are complying with sanitation requirements. However, 
FSIS inspectors will continue to have a full array of regulatory tools 
to ensure the maintenance of sanitary conditions. For instance, FSIS 
inspectors will continue tagging equipment, utensils, rooms, or 
compartments in instances where there is physical evidence of 
insanitary conditions in the production areas of the establishment.
    FSIS anticipates that the development, implementation, and 
maintenance of Sanitation SOP's, as well as the recordkeeping 
provisions, will impose a minimal burden on establishments. Some 
establishments already utilize written Sanitation SOP's. For other 
establishments, compliance with the Sanitation SOP's requirements will 
consist of recording their current sanitation practices. A complete 
discussion of the anticipated costs of implementing the SOP's 
requirements is contained in the Final Regulatory Impact Analysis.
    Sanitation SOP's are an integral part of the Agency's strategy for 
making inspection more effective and more risk-based in its focus. For 
these reasons, FSIS is adopting the proposed requirements for 
Sanitation SOP's and is clarifying that developing, implementing, and 
maintaining Sanitation SOP's and keeping daily Sanitation SOP's 
records, is a condition of inspection.
Development of Sanitation SOP's
    As noted previously, a number of commenters raised concerns about 
the content of the Sanitation SOP's and asked for more specificity. 
Some commenters recommended that FSIS be more specific about what 
procedures must be in the Sanitation SOP's. Other commenters suggested 
that such procedures be fully described and be made mandatory. The 
Agency recognizes these commenters' concerns and therefore is providing 
guidance on how individual establishments may develop their Sanitation 
SOP's in Appendix A and Appendix B to this final rule. Appendix A is a 
guideline on Sanitation SOP's that establishments can use in developing 
their own Sanitation SOP's; Appendix B is a model of an establishment's 
Sanitation SOP's that demonstrates what a completed Sanitation SOP's 
might include. Together, these guidance documents will assist 
establishments to develop Sanitation SOP's that address conditions 
unique to individual establishments and processes and that prevent 
direct product contamination or adulteration. As with all FSIS guidance 
materials, the Agency welcomes comments on how these two documents 
might be improved.
    However, the final rule itself remains nonprescriptive in that it 
requires each establishment to determine for itself what procedures are 
necessary to prevent insanitary conditions that will cause direct 
product contamination or adulteration. Overall, the comments confirmed 
that, while proper sanitation is a common need in every food production 
facility, the means to achieve it are diverse and establishment-
specific. Establishments that now have good sanitation and effective 
process controls are expected to continue using techniques that work in 
their establishment. Other establishments will need to analyze and 
select effective abatement procedures among various alternatives for 
attaining a sanitary processing environment. What works in one 
establishment may or may not work in another.
    The proposed rule also solicited comments as to whether FSIS should 
mandate Good Manufacturing Practices (GMP's) for all or certain 
Sanitation SOP's. FSIS listed illustrations in the proposal of elements 
that might be mandatory elements of Sanitation SOP's. Although some 
commenters expressed support for making GMP's or other practices 
mandatory, many objected to such specific requirements on the basis 
that they would be infeasible. FSIS agrees with those commenters who 
stated that detailed GMP regulations are infeasible because of the 
difficulty in making them specific enough to be useful. FSIS also was 
concerned that such specificity could result in lost flexibility.
    For these reasons, this final rule will not prescribe a single 
format for individual establishment Sanitation SOP's or mandate 
specific GMP's. It will be the responsibility of each establishment to 
consider existing FSIS regulations and guidelines; evaluate its 
facilities, processes, and sanitation conditions; determine what 
sanitation procedures must be implemented to prevent direct product 
contamination or adulteration; and describe these procedures in 
Sanitation SOP's.
Maintaining Sanitation SOP's
    FSIS received several comments regarding the maintenance of 
Sanitation SOP's. Some commenters wanted to know whether if an 
establishment will be able to update its Sanitation SOP's to 
incorporate new technologies. Other commenters wanted to know what type 
of system, if any, FSIS will use to review changes to Sanitation SOP's 
and if a formal request for FSIS review or approval would be required.
    As has been discussed previously, the final rule requires that each 
establishment develop, implement, and maintain its Sanitation SOP's and 
incorporate new sanitation technologies as appropriate. FSIS encourages 
the adoption of new technologies that can improve sanitation and food 
safety. This is an establishment responsibility. Although FSIS will not 
approve Sanitation SOP's, it will provide advice and guidance to 
establishments as they develop and begin to implement Sanitation SOP's.
Recordkeeping
    Commenters also expressed concerns about what was to be in daily 
sanitation

[[Page 38833]]

records and how long and where such records were to be retained. As the 
proposal explained, and this final rule requires, Sanitation SOP's 
records must document the implementation and maintenance of Sanitation 
SOP's, as well as any deviations from Sanitation SOP's procedures, and 
corrective actions taken. As with the development of Sanitation SOP's 
themselves, FSIS will allow each establishment to determine the form 
and format of its daily sanitation records. In many establishments, a 
simple, daily checklist, showing that specific Sanitation SOP's 
procedures were implemented, initialed by the responsible establishment 
employee, is likely to suffice. Other establishments may find a more 
detailed format for its records is more useful. Some establishments may 
wish to use a computer-based system. This final rule provides such 
flexibility.
    Some commenters stated that the proposed six-month retention of 
daily sanitation records was too long. FSIS disagrees and is adopting 
the proposed requirement that establishments retain Sanitation SOP's 
records for six months. Increased product shelf-life and the potential 
need for FSIS personnel to review Sanitation SOP's records many months 
after production make it necessary that establishments retain records 
for six months. Furthermore, sanitation records provide both FSIS and 
establishment management near-term trend data to evaluate how 
establishment sanitation is being carried out under the Sanitation 
SOP's. This feedback should be very useful to establishments in 
determining whether and how their Sanitation SOP's need revision. 
Inspectors will benefit, too, from knowing how the establishment has 
complied with these requirements. Establishment sanitation records will 
also need to be reviewed by the Agency as part of any compliance 
investigation.
    In a related matter, several commenters expressed concern about the 
physical location of establishment sanitation records and questioned 
whether sanitation records must be maintained in the establishment. As 
explained above, FSIS requires unimpeded access to all establishment 
sanitation records for oversight and enforcement purposes. FSIS 
anticipates that, for most establishments, these records will not be 
voluminous and will not create a significant storage problem. However, 
in response to these comments, this final rule will allow 
establishments to retain Sanitation SOP's records off-site provided 
they are not removed from the establishment for at least 48 hours 
following completion and they can be provided to FSIS personnel within 
24 hours of request.
    Some commenters also expressed concern about public accessibility 
to an establishment's Sanitation SOP's records. Like establishment 
HACCP records, these records are kept and maintained by the 
establishment and generally are not Agency records. Occasionally, 
however, such records will be copied and incorporated into Agency 
records for some official purpose. These records will be disclosed to 
third parties only to the extent disclosure is required by the Freedom 
of Information Act and the Privacy Act or other applicable law. 
Proprietary information, personal information, and other information 
exempt from disclosure would be protected.
``Layering''
    Many commenters were concerned that FSIS was layering requirements 
for Sanitation SOP's over existing regulations governing establishment 
sanitation practices, thereby increasing rather than decreasing 
intrusive, command-and-control oversight of all inspected 
establishments. Concern was also expressed that the new requirements 
might conflict with current sanitation regulations.
    FSIS does not consider the Sanitation SOP's requirement to be 
layered over or in conflict with existing regulations. Existing 
regulations establish substantive sanitation-related requirements, 
while the new Sanitation SOP's provisions establish a means by which 
establishments will take responsibility for achieving sanitary 
conditions and preventing direct product contamination or adulteration. 
Sanitation SOP's also will better focus inspection oversight by FSIS 
inspectors on those sanitation measures required to prevent direct 
product contamination or adulteration. As discussed, one of the 
Agency's goals is to reduce inspectors' personal involvement in the 
conduct of routine, day-to-day sanitation procedures.
    FSIS emphasizes that it does not intend or require that an 
establishment's Sanitation SOP's incorporate all elements of the 
existing FSIS sanitation regulations. These regulations contain many 
detailed provisions that do not relate to the prevention of direct 
product contamination. As the text of the Sanitation SOP's regulations 
and the guidance materials at Appendices A and B makes clear, FSIS 
intends and requires only that the Sanitation SOP contain a description 
of the procedures an establishment will follow to address the elements 
of pre-operational and operational sanitation that relate to the 
prevention of direct product contamination.
    For example, under paragraph (a) of Sec. 308.4 of the regulations, 
FSIS requires that ``Dressing rooms, toilet rooms, and urinals shall be 
sufficient in number, ample in size, and conveniently located.'' 
Although compliance with this requirement is important for the 
maintenance of establishment sanitation, and employee hygiene must be 
part of Sanitation SOP's, Sec. 308.4(a) does not concern direct product 
contamination and would not need to be addressed in an establishment's 
Sanitation SOP's. On the other hand, the rule requires that Sanitation 
SOP's specifically address the pre-operational ``cleaning of food 
contact surfaces of facilities, equipment, and utensils'' because these 
procedures are necessary to prevent the direct contamination of 
product. Additionally, the guidance materials in Appendices A and B 
give examples of other procedures necessary to prevent direct product 
contamination that Sanitation SOP's should include, such as 
``Descriptions of equipment disassembly, reassembly after cleaning, use 
of acceptable chemicals according to label directions, and cleaning 
techniques.'' FSIS emphasizes, however, that an establishment does not 
need to reproduce in its written Sanitation SOP's the existing 
regulatory requirements concerning the prevention of direct 
contamination or adulteration of product.
    FSIS also realizes that its existing sanitation regulations contain 
some detailed and prescriptive provisions and that some of those 
regulations may be outmoded and no longer needed in light of the 
Agency's effort to clarify that good sanitation is the responsibility 
of each establishment. FSIS will continue to review, reevaluate, and 
revise, as necessary, all current sanitation regulations, along with 
related issuances and sanitation inspection procedures, to simplify and 
streamline them and make them more compatible with Sanitation SOP's 
requirements. This process was announced and initiated in the advance 
notice of proposed rulemaking published on December 29, 1995 (60 FR 
67469). The review of sanitation regulations is a high priority for the 
Agency. The elements of sanitation that are required to be addressed in 
the Sanitation SOP's will remain as central elements of the FSIS 
sanitation regulations. Establishments will not need to revise their 
Sanitation SOP's because of the simplification and streamlining of 
existing FSIS sanitation regulations.

[[Page 38834]]

Role of Inspectors
     A related concern of many commenters was the role FSIS inspectors 
will play in the development and enforcement of Sanitation SOP's. Some 
commenters expressed concern that during inspection inspectors would 
rely solely on record reviews instead of actually observing 
establishment conditions. Other commenters expressed concerns that 
Sanitation SOP's would merely provide FSIS inspectors with more 
latitude to make intrusive and arbitrary decisions.
    FSIS strongly disagrees with this characterization of Sanitation 
SOP's and the role of the Agency's inspection personnel. Industry's 
responsibility for producing safe meat and poultry and FSIS's 
responsibility for regulatory oversight are fundamentally different. 
Sanitation SOP's are the establishment's commitment to FSIS that they 
will consistently provide a sanitary environment for food production. 
FSIS inspectors will not be tasked with directing an establishment's 
sanitation procedures, nor with ``approving'' the establishment's 
Sanitation SOP's. They will, however, verify that the Sanitation SOP's 
are being implemented and that they are effective in preventing direct 
product contamination and adulteration.
    Oversight of Sanitation SOP's will become an increasingly important 
part of daily inspection activity, while the directing of sanitation 
activities will occur less frequently. Periodic inspection tasks will 
include verifying that Sanitation SOP's meet the regulation's 
requirements, are being implemented and maintained, and are effective 
in producing sanitary conditions. FSIS inspectors' oversight will 
include review of the Sanitation SOP's and required records, direct 
observation of the implementation and monitoring of the Sanitation 
SOP's, and visual observation of sanitary conditions in the production 
areas of the establishment.
    FSIS expects that establishments will rely less on inspectors to 
direct them in maintaining sanitary conditions as establishments rely 
more on adherence to their own Sanitation SOP's. The mix of inspector 
tasks that comprise sanitation inspection also will change. As 
establishments adopt and successfully implement Sanitation SOP's, and 
consistently achieve good sanitation results, FSIS inspectors can spend 
less time ensuring that basic sanitation requirements are being met. 
Conversely, to the extent some establishments do not implement 
effective Sanitation SOP's and consistently achieve good sanitation, 
FSIS inspectors will be obliged to intensify their focus on actual 
establishment conditions and initiate appropriate enforcement actions.
    Ensuring establishments operate under sanitary conditions should be 
made easier for inspectors, and ultimately permit inspectors to spend 
more time on other tasks. One purpose of the Sanitation SOP's 
regulations is to help inspectors, as well as establishments, focus 
their attention on those aspects of establishment sanitation that pose 
the most risk of causing product contamination or adulteration. Under 
the current inspection system, inspectors look at all aspects of 
establishment sanitation, including many that have a relatively low 
probability of causing product contamination. In the future, normal 
oversight activities will focus more on whether an establishment is 
following its Sanitation SOP's and thereby consistently preventing, or 
as appropriate, correcting, conditions that cause direct product 
contamination or adulteration. Some commenters were concerned about the 
effect on establishment operations if inspection personnel, when 
enforcing the Sanitation SOP's requirements, reject one piece of 
equipment, utensil, room or compartment as insanitary. As previously 
stated, inspectors will take prompt action in cases where there is a 
finding of insanitation or the likelihood of product contamination or 
adulteration. The type and intensity of this response will vary. For 
example, establishment operations may be allowed to continue if 
inspection personnel determine that a rejected item, compartment or 
room is not related to other processes or products being produced. 
However, inspection would be withheld in rooms, departments, or 
facilities associated with the production of contaminated or 
adulterated products where the establishment can not show FSIS that 
they have isolated the cause of the contamination or adulteration and 
have taken appropriate action to prevent further contamination or 
adulteration. In a similar vein, commenters also stated that 
establishments should not be penalized for the occurrence of a 
sanitation problem that is effectively abated. These commenters 
suggested that ``U.S. Rejected'' tags should be used only if an 
establishment fails to identify and correct insanitary conditions. If 
the establishment takes proper corrective action, they argued, it 
should be viewed as evidence that the Sanitation SOP's is being 
adequately implemented. FSIS agrees. Establishments that identify and 
correct insanitary conditions in a timely manner and make proper 
disposition of any affected product will be considered to be in 
compliance with the Sanitation SOP's regulations.
    Although FSIS fully expects that the clarification of 
establishments' sanitation responsibilities will lead to better and 
more consistent compliance with sanitation requirements, the Agency 
recognizes that this will not be the case in all establishments. 
Establishments that fail to comply with the requirements in this final 
rule for Sanitation SOP's will be subject to appropriate compliance and 
regulatory action that will, when necessary, include suspension or 
withdrawal of inspection. Further, as noted in the proposal, anyone who 
intentionally falsifies records will be subject to criminal 
prosecution.
    FSIS also recognizes commenters' concerns about its rules of 
practice and due process procedures. FSIS expects that these concerns 
will be addressed through changes to these procedural requirements 
initiated as a result of the Agency's regulatory reform project. These 
subjects are also on the agenda for discussion at FSIS's upcoming 
implementation conferences.
Relation to HACCP
    Another important topic raised by commenters was the link between 
an establishment's Sanitation SOP's and its HACCP plan. This link was 
unclear to some who stated the two were redundant. HACCP plans aim at 
ensuring safety at specific critical control points within specific 
processes, while Sanitation SOP's typically transcend specific 
processes. Sanitation SOP's are important tools for meeting existing 
statutory sanitation responsibilities and preventing direct product 
contamination or adulteration. As such, it is appropriate that they be 
developed and implemented in the near-term prior to implementation of 
HACCP. In a sense, the Sanitation SOP's are a prerequisite for HACCP. 
It is anticipated that some procedures addressed in an establishment's 
Sanitation SOP's might eventually be incorporated into an 
establishment's HACCP plan. Other procedures in an establishment's 
Sanitation SOP's, including those addressing pre-operational sanitation 
procedures for cleaning facilities, equipment, and utensils, will most 
likely remain in the Sanitation SOP's. A sanitation procedure that is 
incorporated into a validated HACCP plan need not be duplicated in the 
Sanitation SOP's.

[[Page 38835]]

Training
    A number of comments expressed concern about the content of 
inspector training, suggesting that inadequate training would result in 
inconsistent enforcement of the rule. Assurance was requested that 
inspectors would be trained to consistently monitor Sanitation SOP's. 
FSIS recognizes that inspectors must be trained to react as regulators 
rather than as quality control consultants or establishment sanitarians 
when a sanitation or other health and safety problem is discovered in 
an establishment. A primary focus of agency training sessions will be 
to attain this goal.
    Also, some commenters asked whether joint FSIS and industry 
training would be offered. FSIS does not plan to allow industry to 
attend Agency training sessions. However, FSIS does plan to hold 
informational briefings for industry personnel. These will be the 
subject of future notices in the Federal Register.
Pre-Operation Sanitation Inspection
    Some commenters asserted that establishments with good Sanitation 
SOP's should be permitted to start daily operations on their own, 
instead of having to wait for an inspector to conduct a pre-operational 
sanitation inspection and allow operations to start. FSIS agrees with 
these commenters. Accordingly, upon the effective date of this rule and 
implementation of Sanitation SOP's, establishments not otherwise 
notified by FSIS may begin daily processing upon completion of pre-
operational sanitation activities without the prior approval of an 
inspector.
    Extending the implementation date for Sanitation SOP's will also 
give FSIS additional time to provide needed training, instruction and 
management support to FSIS inspection personnel tasked with enforcing 
the Sanitation SOP's requirements.
Implementation Date
    Finally, many commenters expressed concern about the amount of time 
they said it would take to prepare and implement effective Sanitation 
SOP's. These commenters requested more lead time to implement these 
requirements. FSIS agrees that some establishments may need more time 
than the 90 days the proposed rule provided for implementing Sanitation 
SOP's requirements. Consequently, FSIS is modifying this aspect of the 
proposal. This final rule will provide establishments six months from 
the effective date of this regulation to develop and implement written 
Sanitation SOP's. This additional time will allow these establishments 
to initially develop and refine their Sanitation SOP's to best meet 
operational needs before the effective date of the Sanitation SOP's 
requirements. Extending the implementation date for Sanitation SOP's 
will also give FSIS additional time to provide needed training, 
instruction, and management support to personnel tasked with enforcing 
the Sanitation requirements.

IV. Microbiological Performance Criteria and Standards

Summary of Proposal

    As part of the Pathogen Reduction/HACCP proposal, FSIS proposed 
interim targets for the reduction of Salmonella for the major species 
and for ground meat and poultry. Further, FSIS proposed to require 
daily testing by slaughter establishments and establishments producing 
raw ground product in order to verify achievement of the Salmonella 
targets on an ongoing basis. The proposal reflected a central tenet of 
the FSIS food safety strategy: to be effective in improving food safety 
and reducing the risk of foodborne illness, HACCP-based process control 
must be combined with objective means of verifying that meat and 
poultry establishments are achieving acceptable levels of food safety 
performance.
    FSIS explained in the preamble to the proposal that food safety 
performance standards, in the form of tolerances or other limits, have 
been an important feature of the food safety regulatory system for 
chemical residues (such as those resulting from the use of animal drugs 
and pesticides) and for pathogenic microorganisms in ready-to-eat meat 
and poultry products (such as Listeria monocytogenes in ready-to-eat 
products and Salmonella in cooked beef). However, performance standards 
have not in the past been incorporated into the regulatory system for 
pathogens on raw meat and poultry products.
    FSIS recognizes that establishing performance standards for 
pathogens on raw products raises different and difficult issues. The 
microbiological safety of a meat or poultry product at the point of 
final sale or consumption is affected by many factors. Most 
significantly, unlike other kinds of contaminants, microbiological 
pathogens can be introduced at many points on the farm-to-table 
continuum, and once in the product, under certain conditions, the 
bacteria can multiply. Some pathogens, such as E. coli O157:H7, are so 
virulent that a small number of organisms can pose a significant 
hazard. Indeed, on that basis the Agency has determined that any amount 
of E. coli O157:H7 will adulterate a meat or poultry product. On the 
other hand, some pathogens, such as Salmonella, ordinarily must 
multiply to relatively large numbers to cause illness, although the 
susceptibility of individuals to illness varies widely. Certain 
segments of the population, such as the very young, the elderly, and 
persons with compromised immune systems, are particularly vulnerable to 
illnesses caused by Salmonella and other foodborne pathogens.
    Therefore, FSIS has not taken the position in this rulemaking that 
some amount of a pathogen necessarily renders a raw meat or poultry 
product unsafe and legally adulterated; the proposed targets for 
pathogen reduction would not have served as a standard for determining 
whether any particular lot of raw product could be released into 
commerce. The proposed targets were intended instead as an initial step 
toward defining levels of food safety performance that establishments 
would be required to achieve consistently over time. The interim 
targets and the required testing by establishments were also intended 
as a first step toward the eventual incorporation of microbial testing 
as an integral part of process-control validation and verification in 
facilities operating under HACCP.
    Salmonella was selected as the target organism because it is the 
most common cause of foodborne illness associated with meat and poultry 
products. It is present to varying degrees in all major species. And, 
interventions targeted at reducing Salmonella may be beneficial in 
reducing contamination by other enteric pathogens.
    As interim targets for pathogen reduction, FSIS proposed that the 
prevalence of Salmonella contamination in each of the major species and 
in raw ground products be reduced by each establishment to a level 
below the current national baseline prevalence as measured by the FSIS 
Nationwide Microbiological Baseline Data Collection Programs and 
Nationwide Microbiological surveys (collectively referred to below as 
the FSIS baseline surveys) or other available data.

Role of Microbiological Performance Criteria and Standards in FSIS Food 
Safety Strategy

    As explained in the ``Background'' section of this preamble, the 
most important objective of this rulemaking is to build into food 
production processes and the FSIS system of regulation and oversight, 
effective measures to reduce and control pathogenic microorganisms

[[Page 38836]]

on raw meat and poultry products. FSIS has concluded that HACCP-based 
process control combined with appropriate microbiological performance 
criteria and standards will achieve this objective.
    Because the current regulatory system lacks any performance 
criteria or standards for harmful bacteria on raw products (other than 
with respect to E. coli O157:H7 on raw ground beef), FSIS inspectors 
have no adequate basis for judging whether establishments producing raw 
meat and poultry products are dealing effectively with the food safety 
hazard posed by harmful bacteria.
    The HACCP requirements discussed in the preceding section of this 
preamble will ensure that all meat and poultry establishments implement 
science-based process controls designed to prevent and reduce the 
significant food safety hazards that arise in their particular 
production processes and products. For slaughter establishments and 
other establishments producing raw meat and poultry products, this will 
mean developing controls that address the hazards posed by pathogenic 
microorganisms as well as other biological, chemical and physical 
hazards. HACCP principles provide the framework by which establishments 
target and reduce harmful bacteria on raw meat and poultry products.
    To be successful in ensuring food safety, however, HACCP must be 
coupled with appropriate performance criteria and standards against 
which the effectiveness of the controls developed by each establishment 
can be validated and verified. For example, controls designed to 
prevent the contamination of processed, ready-to-eat meat and poultry 
products with harmful bacteria would have to be validated as effective 
in meeting the already-existing requirement that such products be free 
of harmful bacteria. Without such performance criteria and standards, 
there would be no objective basis for determining whether a particular 
HACCP plan is adequate for its food safety purpose. Additionally, there 
would be no way to determine whether industry or FSIS had met their 
respective food safety responsibilities.
    In this rulemaking, FSIS for the first time proposed 
microbiological performance standards for raw products. The need for 
some measure of performance in the area of microbiological 
contamination was generally supported by the comments FSIS received on 
its proposal. In response to the comments, FSIS has refined and 
improved its proposed approach, and is establishing microbiological 
performance standards for reduction of Salmonella in raw products, 
coupled with performance criteria for use with E. coli testing to 
verify the effectiveness of process controls in slaughter 
establishments.
    These new provisions are the first steps in what FSIS expects to be 
a long-term effort to ensure that appropriate microbial testing is 
conducted, and appropriate criteria and standards exist, to reduce the 
food safety hazards posed by harmful bacteria on raw meat and poultry 
products. The numerical targets for both the performance criteria and 
the pathogen reduction performance standards are likely to be changed 
as new data become available. The targets currently are set at the 
national baseline prevalence of contamination and reflect what is 
achievable using available technology. FSIS intends to repeat 
periodically its baseline surveys, on which the criteria and standards 
are based. FSIS will collect additional data on Salmonella by testing 
products in establishments pursuant to the performance standards and on 
E. coli through close monitoring of establishments' experience and test 
results associated with that mode of process control verification. 
These new data, together with relevant epidemiologic data, scientific 
research, and new technologies, will be considered by FSIS when 
proposing future revisions to the performance criteria and testing 
requirements for E. coli and the pathogen reduction performance 
standards for Salmonella. New information and data also may support 
different standards and different approaches to microbial testing.
    FSIS is committed to the development and implementation of future 
performance standards, as needed, to achieve the FSIS's public health 
goal of reducing the incidence of foodborne illness associated with 
harmful bacteria on raw meat and poultry products. FSIS is also 
concerned that standards achieve this public health goal in a manner 
that encourages industry innovation and minimizes regulatory burdens on 
the regulated industry. The pathogen reduction performance standards 
promulgated in this regulation will be implemented on the basis of a 
statistical evaluation of the prevalence of bacteria in each 
establishment's products, measured against the nationwide prevalence of 
the bacteria in the same products. These standards will not be used to 
judge whether specific lots of product are adulterated under the law. 
As more research is done and more data become available, and as more 
sophisticated techniques are developed for quantitative risk assessment 
for microbiological agents, it may be possible and appropriate to 
develop performance standards that use a different approach. 
Consideration may also be given to the possibility of establishing 
similar standards for other pathogenic microorganisms. FSIS will 
continue to work with the scientific community in this area.
    The microbiological performance standards set out in this 
rulemaking are part of a fundamental shift in FSIS regulatory 
philosophy and strategy. The current inspection system relies heavily 
on intensive ``command-and-control'' prescription of the means by which 
meat and poultry establishments must achieve statutory objectives 
concerning food safety, sanitation, product wholesomeness, and 
prevention of economic adulteration and misbranding. As explained in 
the ``Background'' section of this preamble, in FSIS's ANPR ``FSIS 
Agenda for Change: Regulatory Review,'' and in the January, 1996, 
National Performance Review report ``Reinvention of Food Regulations,'' 
FSIS plans to shift from this reliance on command and control 
regulations to much greater reliance on performance standards. FSIS 
believes that public health and consumer protection goals can be 
achieved more effectively, in most cases, by converting command-and-
control regulations to performance standards, which provide industry 
with the flexibility to devise the optimal means of achieving food 
safety objectives. FSIS would verify compliance with such performance 
standards through inspection and other forms of oversight.

Overview of Final Rule

    Comments on the proposed rule's microbial testing provisions have 
resulted in a number of changes to those provisions. As discussed in 
the ``Response to Comments'' section, below, FSIS received numerous 
comments supporting the concept of microbiological performance criteria 
or standards, but also received many comments urging alternatives to 
the specific approach proposed by FSIS, including testing for organisms 
other than Salmonella.
    The Agency actively sought out comment and information on the issue 
of target organism(s) to be selected for process control verification 
and pathogen reduction purposes in this regulation. In the proposal, 
FSIS stated that ``the Agency recognizes that there are other foodborne 
human pathogens of public health concern that can be isolated from raw 
meat and poultry product. The Agency would welcome

[[Page 38837]]

comments on the targeting of other pathogens in addition to or in lieu 
of Salmonella'' (60 FR 6800). As noted earlier in this preamble, during 
the comment period FSIS held many meetings to solicit comment on 
various issues, including microbiological criteria and standards. 
Microbiological criteria and standards were discussed in detail at the 
FSIS-sponsored scientific conference held in Philadelphia, 
Pennsylvania, on May 1 and 2, 1995, titled ``The Role of 
Microbiological Testing in Verifying Food Safety.'' This conference was 
open to the public and was announced in the Federal Register on March 
24, 1995 (60 FR 15533). An expert panel at that conference endorsed the 
role of microbiological testing in accordance with appropriate criteria 
or standards, but suggested that mandatory establishment testing focus 
on a quantitative assay for generic E. coli rather than the proposed 
qualitative assay for Salmonella. The panel stated that a quantitative 
assay for the more commonly occurring generic E. coli is a more 
effective process control indicator with respect to the prevention of 
contamination of meat and poultry by feces and associated bacteria.
    FSIS also held a series of six issue-focused public meetings in 
September, 1995. During a preliminary public meeting on August 23, 
1995, at which issues were identified and the meeting agenda was 
established, participants decided that a full day should be devoted to 
further public discussion of pathogen reduction standards and microbial 
testing. The agenda for the six meetings appeared in the Federal 
Register on August 31, 1995 (60 FR 45381). The issues discussed on 
September 27 included: (1) the scientific and policy basis for 
establishing targets; (2) whether Salmonella is the appropriate 
organism for some or all species; (3) whether other pathogens would be 
preferable for some or all animal species; (4) the utility of targets 
for E. coli or other non-pathogenic indicator organisms as a means of 
controlling and reducing pathogenic microorganisms; (5) the advantages 
and disadvantages of targets based on the prevalence of detectable 
contamination vs. targets based on the number of organisms present; and 
(6) the need for pathogen reduction targets for raw ground products in 
general and in establishments that both slaughter animals and produce 
ground product.
    At the September 27, 1995, issue-focused meeting, there was 
additional comment in favor of testing for an organism other than 
Salmonella, such as generic E. coli, that has a strong track record in 
the industry as a good organism to use for process control verification 
testing. There was, however, continued strong support for raw product 
testing targeted at pathogens, such as Salmonella, and support for 
pathogen reduction as the primary goal of such testing.
    At the meetings, FSIS distributed issue papers on the various 
issues being addressed, based in large part on comments already 
received. The issue paper on Pathogen Reduction Performance Standards 
and Microbial Testing stated that the two most common concerns in the 
comments received to that date were the proposed selection of 
Salmonella as the indicator organism and the frequency of proposed 
testing. It stated that although some commenters recommended finalizing 
Salmonella testing, others recommended using E. coli instead of or in 
addition to Salmonella. The issue paper stated the Agency's current 
thinking on the organism to be selected, the need for daily testing at 
every establishment, and the necessity of testing each species 
slaughtered and each ground product produced. In the issue paper FSIS 
stated, among other things, that it was ``seriously considering generic 
E. coli as the process control indicator organism and the adoption of a 
quantitative E. coli standard as a measure of process control with 
respect to the prevention and reduction of fecal contamination in 
slaughter plants.'' FSIS also stated that it was considering setting 
forth pathogen-specific performance standards as a direct measure of 
accountability for controlling and reducing harmful bacteria in raw 
meat and poultry products and that Salmonella targets might be adopted 
as performance standards and enforced by FSIS through its own 
compliance monitoring. The Agency published the issue papers in the 
Federal Register on October 24, 1995 (60 FR 54450).
    Based on the large body of written and oral comments FSIS has 
received on this issue, the Agency has decided not to use Salmonella 
both as a target for pathogen reduction and as an indicator of process 
control. FSIS has decided to adopt pathogen reduction performance 
standards targeting Salmonella, as proposed, except that FSIS, not the 
establishments, will conduct testing for the pathogen to verify 
compliance. FSIS also has decided to require establishments 
slaughtering livestock and poultry to conduct routine testing for 
generic E. coli (instead of the proposed use of Salmonella tests) as an 
ongoing, objective process control indicator for fecal contamination, 
and to establish performance criteria by which results can be 
evaluated.
Process Control Verification Performance Criteria
    Under the FMIA and the PPIA, meat and poultry establishments 
inspected by FSIS are required to maintain sanitary conditions 
sufficient to prevent contamination of products with filth and to 
prevent meat and poultry products from being rendered injurious to 
health (21 U.S.C. 601(m) and 608 (FMIA); 21 U.S.C. 453 (g) and 456 
(PPIA)). A grant of inspection by FSIS is contingent upon an 
establishment meeting this responsibility. FSIS is authorized by law to 
issue regulations establishing appropriate sanitation requirements. 
Meat and poultry products are deemed legally adulterated, whether or 
not they are shown to be contaminated, if prepared, packed, or held 
under insanitary conditions whereby they may have become contaminated 
with filth or may have been rendered injurious to health.
    In slaughter establishments, fecal contamination of carcasses is 
the primary avenue for contamination by pathogens. Pathogens may reside 
in fecal material and ingesta, both within the gastrointestinal tract 
and on the exterior surfaces of animals going to slaughter. Therefore, 
without care being taken in handling and dressing procedures during 
slaughter and processing, the edible portions of the carcass can become 
contaminated with bacteria capable of causing illness in humans. 
Additionally, once introduced into the establishment environment, the 
organisms may be spread from carcass to carcass.
    Because the microbial pathogens associated with fecal contamination 
are the single most likely source of potential food safety hazard in 
slaughter establishments, preventing and removing fecal contamination 
and associated bacteria are vital responsibilities of slaughter 
establishments. Further, because such contamination is largely 
preventable, controls to address it will be a critical part of any 
slaughter establishment's HACCP plan. Most slaughter establishments 
already have in place procedures designed to prevent and remove visible 
fecal contamination.
    There is general agreement within the scientific community that 
generic E. coli is the best single microbial indicator for fecal 
contamination. FSIS, therefore, is requiring that establishments 
slaughtering livestock or poultry begin testing for E. coli (E. coli, 
biotype I, nonspecific as to species, hereinafter referred to simply as 
E. coli) at the

[[Page 38838]]

frequency and following the procedures described in ``Process Control 
Verification; E. coli Performance Criteria and Testing'' section, 
below, 6 months after publication of the final rule. FSIS considers the 
required testing to be essential for meeting current statutory 
requirements for sanitation and the prevention of adulteration. This 
testing also will play an integral role in the successful 
implementation of HACCP in slaughter establishments. In addition, FSIS 
is establishing process control performance criteria for fecal 
contamination based on the frequency and levels of contamination of 
carcasses with E. coli.
    As explained below, FSIS is establishing performance criteria to 
reflect the prevalence and levels of contamination of E. coli on 
carcasses produced nationwide, as determined by FSIS baseline surveys. 
The performance criteria and required testing will provide each 
slaughter establishment and FSIS with an objective means of verifying 
that the establishment is achieving this level of performance and 
maintaining it consistently over time. Test results that show an 
establishment is meeting or exceeding the criteria provide evidence 
that the establishment is maintaining adequate process control for 
fecal contamination.
    FSIS is purposely using the term performance ``criteria'' rather 
than performance ``standard'' in this context because no single set of 
test results can demonstrate conclusively that adequate process control 
for fecal contamination is or is not being maintained. As explained 
below, if test results do not meet the applicable criterion, it raises 
questions about the adequacy of the process control. FSIS intends to 
consider the establishment's results and corrective actions, together 
with other information and inspectional observations, in evaluating 
whether a problem exists that requires regulatory action or other 
measures to protect consumers and ensure compliance with the law.
    Also, as discussed below, although FSIS is proceeding with the 
final rule at this time, it is inviting comment on technical aspects of 
the process control performance criteria and the required testing. FSIS 
requests that comments on the E. coli performance criteria and testing 
requirement be focused on the technical aspects of the rule, i.e., the 
manner in which the criteria are articulated, the sampling frequency, 
and the sampling and testing methodologies.
    FSIS intends to update the criteria periodically to ensure that the 
criteria adequately reflect an appropriate level of performance with 
respect to prevention and removal of fecal contamination and associated 
bacteria from livestock and poultry carcasses.
Pathogen Reduction Performance Standards
    As proposed, FSIS is adopting pathogen reduction performance 
standards using Salmonella as the target organism. The most significant 
difference between the proposal and this final rule is that, as 
explained above, FSIS is not relying on Salmonella to be a process 
control indicator, as well as the target organism for the pathogen 
reduction performance standard. Establishments will not be required by 
this final rule to test for Salmonella, as had been proposed. Instead, 
FSIS will obtain samples from slaughter establishments and 
establishments producing raw ground product or fresh pork sausage and 
test those samples for Salmonella to ensure that the pathogen reduction 
performance standards are being met.
    As proposed, FSIS will require that no establishment can have a 
prevalence of Salmonella contamination, as a percentage of positive 
samples from carcasses and percentage of positive samples from raw 
ground product, greater than the baseline prevalence for each raw 
product as reflected in the FSIS baseline survey for each species or 
other category of raw product. These targets constitute performance 
``standards'' rather than performance ``criteria'' because, following 
an establishment's implementation of HACCP, FSIS will require that the 
establishment meet the standard consistently over time as a condition 
of maintaining inspection.
    The Salmonella pathogen reduction performance standards are not, 
however, lot release standards, and the detection of Salmonella in a 
specific lot of raw product will not by itself result in the 
condemnation of that lot. The performance standards and FSIS's 
enforcement approach, as discussed below, are intended to ensure that 
each establishment is consistently achieving an acceptable level of 
performance with regard to controlling and reducing harmful bacteria on 
raw meat and poultry products.
    FSIS considers systematic reduction of pathogenic microorganisms in 
raw product to be an essential responsibility of meat and poultry 
establishments under the current statutes. As a condition of inspection 
and to avoid the production of product that would be deemed legally 
adulterated, establishments must utilize available process control 
methods and technologies as necessary to achieve applicable pathogen 
reduction standards.

Process Control Verification; E. coli Performance Criteria and Testing

    Establishments that slaughter livestock and poultry currently have 
an obligation to control the slaughter and sanitary dressing process so 
that contamination with fecal material and other intestinal contents is 
prevented. This means that establishments must maintain sanitary 
conditions and use good manufacturing practices to avoid contamination 
with visible feces and ingesta and associated bacteria. When such 
visible contamination occurs, establishments are expected to detect it 
and physically remove it through knife trimming or other approved 
removal procedures. The present FSIS verification activity to 
demonstrate that this has been accomplished is organoleptic inspection. 
FSIS inspectors apply a zero tolerance performance standard for visible 
feces and ingesta on dressed carcasses. As a practical matter, however, 
additional measures must be taken if inspectors are to assess the 
extent to which the invisible bacteria associated with feces and 
ingesta may be present on the carcass.
    FSIS has concluded, based on its proposal and the comments 
received, that the current practice of organoleptic examination by 
inspectors and the physical removal of visible contamination by 
establishments needs to be supplemented with an establishment-conducted 
microbial verification activity. This microbial testing is designed to 
verify, for the establishment and FSIS, that the establishment has 
controlled its slaughter process with respect to prevention and removal 
of fecal material and ingesta and associated bacteria.
Rationale for Using E. coli Tests to Verify Process Control
    E. coli testing is more useful than the originally proposed 
Salmonella testing in verifying that a slaughter process is under 
control. This was expressed in numerous comments on the proposal, 
comments generated in FSIS public hearings, and the results of the 
scientific and technical conference on the Role of Microbiological 
Testing in Verifying Food Safety. The expert panel at that conference 
stated:

    Microbial testing is an essential element for verifying process 
control of raw meat and poultry. A variety of indicators exists, but 
the panel concluded that quantitative measurement of Escherichia 
coli would be more effective than qualitative Salmonella testing. 
When processes are under control for

[[Page 38839]]

E. coli, the potential presence of enteric pathogens will be 
minimized.<SUP>1

    \1\ Expert Panel's Summary Report and Recommendations, 
Scientific and Technical Conference on Role of Microbiological 
Testing in Verifying Food Safety, May 1-2, 1995.
---------------------------------------------------------------------------

    The panel compared selection criteria for the choice of an 
indicator organism and considered alternative microbial targets such as 
E. coli, Enterobacteriaceae, and aerobic plate count, to be used alone 
or in combination with Salmonella testing. In reaching its conclusion 
that E. coli would be the most effective measure of process control for 
enteric pathogens, the panel considered the ideal characteristics of 
microbial indicators for the stated purpose. Important characteristics 
of E. coli are:

    <bullet> There is a strong association of E. coli with the 
presence of enteric pathogens and, in the case of slaughtering, the 
presence of fecal contamination.
    <bullet> E. coli occurs at a higher frequency than Salmonella, 
and quantitative E. coli testing permits more rapid and more 
frequent adjustment of process control.
    <bullet> E. coli has survival and growth characteristics similar 
to enteric pathogens, such as E. coli O157:H7 and Salmonella.
    <bullet> Analysis for E. coli poses fewer laboratory safety 
issues and testing at the establishment site is more feasible than 
such testing with Salmonella.
    <bullet> There is wide acceptance in the international 
scientific community of its use as an indicator of the potential 
presence of enteric pathogens.

    In the panel's view, microbial testing should be used to 
demonstrate process control; they concluded that a proximate indicator 
for enteric pathogens is needed for demonstrating process control with 
respect to fecal contamination. The panel concluded that E. coli would 
be the single most effective indicator for this purpose. The panel's 
conclusion reinforces previous statements by the NAS that ``at present, 
E. coli testing is the best indicator of fecal contamination among the 
commonly used fecal-indicator organisms.'' <SUP>2 FSIS agrees with 
these conclusions.
---------------------------------------------------------------------------

    \2\ Subcommittee on Microbiological Criteria, Committee on Food 
Protection, Food and Nutrition Board, National Research Council. 
1985. ``An Evaluation of the Role of Microbiological Criteria for 
Foods and Food Ingredients.'' National Academy Press, Washington, 
D.C.
---------------------------------------------------------------------------

    If future scientific research identifies another organism or group 
of organisms which would prove as effective in measuring process 
control for fecal contamination, FSIS would consider appropriate 
revisions to the regulations.
Use of Baseline Values to Establish E. coli Performance Criteria
    The presence of some microorganisms on raw meat and poultry is 
unavoidable and highly variable. The goal of process control in a 
slaughter establishment is to minimize initial microbial contamination 
of the carcasses, remove harmful microorganisms that nonetheless may be 
present, control the proliferation of any remaining microorganisms, and 
prevent re-contamination. Process control criteria based on data from 
FSIS's nationwide baseline surveys will aid establishments in achieving 
this goal and complement the transition to HACCP.
    FSIS collects data to develop and maintain a general, ongoing 
microbiological profile of carcasses for selected microorganisms of 
varying degrees of public health concern, and organisms or groups of 
organisms of value as indicators of general hygiene or process control, 
and to document changes in the profiles over time. FSIS's Nationwide 
Microbiological Baseline Data Collection Programs provide for sampling 
over a year's time to account for possible seasonal variations. This 
was the approach taken in collecting data from carcasses for all 
slaughter classes: steer/heifer, cow/bull, broilers, market hogs, and 
turkey. Sampling is designed to represent the vast majority of raw meat 
and poultry products produced, in most cases approximately 99% of the 
product produced. These programs are nationwide in scope. Enough 
samples are taken to enable the Agency to describe the annual 
distribution of test results. The number of samples collected also 
allows for control of sampling variation and non-sampling errors (such 
as missing samples, incomplete data, and inconsistent data). By 
contrast, FSIS's Nationwide Surveys provide a snapshot over a specified 
period of time less than a year. They involve a large enough number of 
samples to ensure a reasonable level of precision for estimates, given 
the prevalence of the microorganisms included in the surveys. This was 
the approach taken in developing baseline data for other raw meat and 
poultry products: ground beef (at inspected establishments and at 
retail), ground chicken, ground turkey, and fresh pork sausage.
    For the current baselines, carcass samples were taken from fresh, 
whole chilled carcasses after slaughter and dressing but before any 
further processing took place. Samples were analyzed fresh, not frozen, 
to gather more accurate data on numbers of microorganisms, especially 
those that are more susceptible to freezing, such as Campylobacter 
jejuni/coli. FSIS personnel collected the samples tested in the surveys 
using standard Agency procedures for taking aseptic samples from animal 
tissues and for ensuring random sample selection.<SUP>3,4
---------------------------------------------------------------------------

    \3\ Food Safety and Inspection Service. 1994. Nationwide Broiler 
Chickens Microbiological Baseline Data Collection Program: Broiler 
Chicken Sample Collection Procedures, 2/18/94. U.S. Department of 
Agriculture, Washington, D.C.
    \4\ Food Safety and Inspection Service. 1993. Nationwide Beef 
Microbiological Baseline Data Collection Program: Cow/Bull Sample 
Collection Procedures, 8/1/93. U.S. Department of Agriculture, 
Washington, D.C.
---------------------------------------------------------------------------

    Reports of FSIS baseline programs and surveys are issued after 
testing results have been compiled and analyzed. Reports have been 
completed for cattle, broiler chickens, hogs, ground beef, ground 
chicken, and ground turkey. The collection and analysis of samples for 
the turkey baseline program and the fresh pork sausage survey will be 
underway soon; criteria for turkeys and fresh pork sausage will be 
determined upon completion of the sampling and analysis of results.
Establishment of E. coli Performance Criteria to Verify Process Control
    Using data from the baseline surveys described in the preceding 
section, FSIS has developed animal species-specific, minimum 
performance benchmarks, or performance criteria, for E. coli on 
carcasses.
    As explained above, these criteria are not enforceable regulatory 
standards. The E. coli performance criteria are intended to assist 
slaughter establishments and FSIS in ensuring that establishments are 
meeting their current statutory obligation to prevent and reduce 
contamination of carcasses by fecal material, ingesta, and associated 
bacteria. The criteria are flexible and are subject to amendment as 
FSIS and the industry gain experience with them and accumulate more 
data on establishment performance. The criteria are intended 
specifically to provide an initial basis upon which slaughter 
establishments and FSIS can begin to use microbial testing to evaluate 
the adequacy of establishment process controls to prevent feces, 
ingesta, and other animal-derived contaminants from contaminating the 
tissues intended for use as food.
    FSIS has designed the criteria so that establishments meeting them 
are achieving results, in terms of E. coli levels, consistent with 
those being achieved by a large majority of the slaughter production in 
the United States, as reflected in the FSIS baseline

[[Page 38840]]

surveys for each species of livestock and poultry.
    The E. coli performance criteria are expressed in terms of a 
statistical procedure known as a ``3-class attributes sampling plan'' 
applied in a moving window. This procedure specifies cutoffs (denoted m 
and M, with m<M) for quantitative E. coli levels so as to define three 
classes of results: acceptable, marginal, and unacceptable. The 
definitions are:

Acceptable--result <ls-thn-eq> m
Marginal--result > m and <ls-thn-eq> M
Unacceptable--result > M

    Under this approach, m and M are defined in relation to the 
distribution of E. coli results for each slaughter class. The Agency 
has used as the starting point for establishing the cutoff for m the 
80th percentile of current industry wide performance, in terms of E. 
coli levels, for each slaughter class. The starting point for 
establishing M is the 98th percentile of industry performance. Thus, if 
the criterion for any species were set precisely at those percentiles, 
a set of test results indicating performance in the 80th to 98th 
percentile range, according to FSIS's Nationwide Microbiological 
Baseline Data Collection Program results, would be deemed ``marginal,'' 
and, as discussed below, would raise a question about the adequacy of 
the establishment's process control. Expressed in another way, 
``marginal'' results would be within the worst 20% of overall industry 
performance in terms of E. coli counts. Similarly, results worse than 
the 98th percentile (M) are within the worst 2% of overall industry 
performance. Any single result exceeding M is, therefore, deemed 
``unacceptable.''

                              Table 1.--Distribution of E. coli by Slaughter Class                              
----------------------------------------------------------------------------------------------------------------
           Percentile                Steer/heifer          Cow/bull            Broilers              Hogs       
----------------------------------------------------------------------------------------------------------------
50th (median)...................  Negative*.........  Negative*.........  29 cfu/ml.........  Negative*         
80th (m)........................  Negative*.........  Negative*.........  80................  10 cfu/cm \2\     
90th............................  Negative*.........  10 cfu/cm \2\.....  180...............  150               
95th............................  10 cfu/cm \2\.....  40................  360...............  880               
98th (M)........................  80................  300...............  1100..............  6,800             
99th............................  290...............  2200..............  3300..............  33,000            
----------------------------------------------------------------------------------------------------------------
* Negative by the method used in the baselines which had a minimum detectable level of 5 cfu/cm \2\ of carcass  
  surface area.                                                                                                 

    Table 1 shows the level at which E. coli has been found on 
carcasses, by slaughter class as a percent of all such product. For 
example, the data show that 80% of broilers tested at or below 80 
colony forming units per milliliter (cfu/ml), while 90% tested at or 
below 180 cfu/ml. More detailed descriptions of the distribution of 
numbers of E. coli found per carcass species are provided in FSIS's 
baseline reports.
    To make the criteria as simple and easy to use as possible, 
consistent with the accepted laboratory practice of diluting samples 
successively by factors of 10 to obtain bacteria counts, FSIS has 
elected to express the criteria in terms of powers of 10 (i.e., 10, 
100, 1000, etc.). As shown in Table 2, this results in m and M being 
the closest power of 10 to the actual numbers estimated for the 80th 
and 98th percentiles from the baseline data.
    Because the Agency's baseline survey work on turkeys is still 
underway, no E. coli criterion is being established at this time for 
that slaughter class.

        Table 2.--m and M Values for E. coli Performance Criteria       
------------------------------------------------------------------------
           Slaughter class                      m                 M     
------------------------------------------------------------------------
Steer/Heifer........................  (<SUP>1)..................          100
Cow/Bull............................  (<SUP>1)..................          100
Broiler.............................  100..................         1000
Hogs................................  10...................      10,000 
------------------------------------------------------------------------
\1\ Negative.                                                           

    It should be noted that ``negative,'' in this context, is defined 
by the sensitivity of the method used in the Baseline Surveys, which 
was 5 cfu/cm<SUP>2 of carcass surface area for cattle and hogs.
    FSIS is requiring the use of an analytic method approved by the 
Association of Official Analytic Chemists or any method validated by a 
scientific body in collaborative trials against the three tube Most 
Probable Number (MPN) method and agreeing with the 95 percent upper and 
lower confidence limit of the appropriate MPN index.
    FSIS has concluded that, at some point, the number of samples 
testing in the marginal range raises a significant question about the 
adequacy of an establishment's process control, and has defined that 
point for purposes of these criteria as more than 3 results above m 
within any consecutive 13 samples tested. This point was established 
based on the following analysis.
    There occasionally will be test results that exceed the acceptable 
level, m, because of variations or aberrations in establishment 
performance, sampling, etc., that do not reflect the state of overall 
process control. FSIS believes that the performance criteria and 
approach to evaluating test results should avoid raising a significant 
process control question on the basis of chance results, but should be 
sensitive enough to provide a reasonably high likelihood of detecting 
performance that falls significantly short of the national baseline 
levels. FSIS has decided that it is appropriate to evaluate test 
results in a manner that ensures that there is an 80% probability that 
establishments actually operating at the acceptable performance level 
will achieve results that are deemed to satisfy the criteria. This is 
the same statistical approach FSIS took in its proposed approach to 
evaluating an establishment's Salmonella test results, using the moving 
window approach to evaluating process control verification tests (see 
pages 6798-6805 of the Pathogen Reduction/HACCP proposal).
    Using this approach, it can be predicted statistically that 
slaughter establishments that are operating at the acceptable 
performance level reflected by m will, with an 80% probability, have 
three or fewer results above m (denoted as c) within every 13 samples 
tested (denoted as n). FSIS will require slaughter establishments to 
record and evaluate E. coli results in a ``moving window'' of 13 
consecutive results. A moving window provides a continuous picture of 
establishment performance and is the preferred statistical approach for 
assessing ongoing processes (as opposed to sampling specific lots of 
product for contaminants). Thus, the presence of more than three 
marginal results within any 13 consecutive samples, or the ``window,'' 
will be indicative of an operation failing to meet the criteria.
    Use of a different probability level, such as a 70% or 90% 
probability of getting acceptable test results if establishments are 
operating at the specified level would result in different values for c 
and n (namely, c=3 and

[[Page 38841]]

n=15 using the 70% probability level, and c=3 and n=10 using the 90% 
probability level). Using 70% as the statistical criterion for setting 
c and n would result in too many chance failures of the criteria, while 
using 90% would make it too difficult to detect potential process 
control problems. It is the judgment of the Agency that use of the 80% 
probability level strikes a reasonable balance.
    In summary, if the results of one test are above M, or if more than 
3 of 13 test results are above m, a significant question is raised as 
to whether the establishment is maintaining adequate process control 
and will trigger further review of establishment process control. FSIS 
stresses again that these E. coli criteria are guidelines, not 
regulatory standards. Ideally, each establishment will develop its own 
equally or more effective criteria for process control based on its own 
data and/or industry-developed benchmarks. FSIS encourages 
establishments, in the context of their HACCP plans, to apply their 
own, establishment-specific criteria to ensure process control.
    FSIS also is inviting comment on the approach it has taken to 
expressing its E. coli performance criteria for verifying process 
control. FSIS recognizes that there is more than one possible approach 
and welcomes comments and suggestions.
Sampling Frequency for E. coli Testing
    FSIS has chosen to use production volume as the basis for 
determining the frequency at which establishments will conduct testing 
for E. coli. In the proposed rule, FSIS proposed to require all 
slaughter establishments and establishments producing ground meat and 
poultry, regardless of size or volume, to conduct one test for 
Salmonella each day. This was based on the premise that verifying that 
a process is ``in control'' is more a function of specific 
establishment characteristics than the amount of product being 
produced. However, commenters suggested and FSIS recognizes that there 
may be striking differences in the ways in which high and low volume 
establishments operate, which can influence the ability of the 
establishment to keep processes in control. High volume establishments 
may receive animals for slaughter from a number of different sources 
for each day's production; there may be several shifts, and production 
personnel are often more transient; there may be multiple supervisors; 
and there may be much greater complexity in the overall slaughter 
process. In contrast, a low volume establishment will have a smaller 
and possibly more stable work-force, often supervised by an owner-
operator, and may employ relatively simple procedures that are 
performed consistently over time. This does not negate the need in low 
volume establishments for microbial verification of a HACCP plan; 
however, under these circumstances it may not be as essential for very 
low volume establishments to undertake daily microbial testing, as 
initially proposed. By adopting a volume-based system, the testing 
frequency will, by definition, be highest in large establishments 
producing the most product, while the number of tests will be minimized 
in smaller establishments.
    The majority of commenters who opposed daily testing stated that 
such a testing requirement would place an unfair cost burden and have a 
negative financial impact on small establishments, as it would require 
the same expenditure for testing by establishments that slaughtered one 
or two animals per day as those slaughtering several thousand daily. It 
was also noted that there is a public health consequence to the 
proposed approach. If a process control problem detectable by microbial 
testing existed in a high volume establishment that tested only once a 
day, a great deal more potentially contaminated product would be 
produced and distributed before enough microbial tests were performed 
to show the problem existed than would be the case in a small volume 
establishment. These issues are addressed by the switch to a volume-
based testing system.
    There is no single method for determining the frequency of 
microbial testing within a volume-based testing system that will be 
equally effective in all establishments. Testing frequencies are 
ideally determined on an establishment-by-establishment basis, taking 
into account a number of variables, including differences in sources of 
raw materials, the type and nature of the process, and the consistency 
of microbial test results over time. Nonetheless, for both public 
health and process control verification reasons, FSIS considers it 
necessary and reasonable to require a minimum frequency of testing 
sufficient to result in completion of at least one E. coli test window 
(13 samples) per day in the highest volume establishments for each 
species. This will provide a daily set of results adequate to verify 
process control in the highest volume establishments. Accumulation of 
results over a longer period of time will be an acceptable basis for 
verifying process control in lower volume establishments.
    Based on these principles and conclusions, the required minimum 
frequencies for E. coli testing for each slaughter species are as shown 
in Table 3.

                  Table 3.--E. coli Testing Frequencies                 
------------------------------------------------------------------------
                                                                        
------------------------------------------------------------------------
Cattle............................  1 test per 300 carcasses.           
Swine.............................  1 test per 1,000 carcasses.         
Chicken...........................  1 test per 22,000 carcasses.        
Turkey............................  1 test per 3,000 carcasses.         
------------------------------------------------------------------------

    The frequencies were derived by first rank-ordering all slaughter 
establishments by species based on total annual production. This 
ranking, which was based on data from FY 1993 and FY 1994, revealed 
that establishment production volumes vary widely and that there are 
appreciable differences in the concentration of business among the 
industries. In cattle slaughter, 12 of 912 establishments accounted for 
over 42% of production, with the smallest of these slaughtering about 
one million head annually. On the small volume end, 620 establishments 
slaughtered fewer than 1000 head annually and together accounted for 
about one-half of one percent (0.5%) of national slaughter production. 
By contrast, there are ten or fewer very low volume establishments 
slaughtering chickens, and production is spread more evenly over the 
240 establishments on the FSIS FY 1994 inventory of establishments. 42 
of 240 slaughter establishments accounted for 40% of production.
    FSIS has selected sampling frequencies so that in the subgroup of 
establishments accounting for 99% of total production for each species, 
the 5% of establishments with the highest production volume would each 
have to conduct a minimum of 13 E. coli tests, or at least one complete 
test window, each day. In addition, with these frequencies, 90% of all 
cattle, 94% of all swine, 99% of all chicken, and 99% of all turkeys 
will be slaughtered in establishments conducting a minimum of one E. 
coli test per day.
    The above frequencies notwithstanding, FSIS has concluded that all 
establishments must conduct sampling at a frequency of at least once 
per week to provide a minimum, adequate basis for process control 
verification using E. coli testing. However, establishments with very 
low volumes, annually slaughtering no more than 6,000 cattle, 20,000 
swine, or a combination of such livestock not to exceed a total of 
20,000 with a maximum of 6,000 cattle, or 440,000 chickens or 60,000 
turkeys (or a combination of such poultry not to

[[Page 38842]]

exceed a total of 440,000, with a maximum of 60,000 turkeys), will be 
required to sample once per week only until a sampling window that 
verifies process control has been completed and the results indicate 
that the slaughter process is under control. Establishments 
slaughtering more than one species would sample the species slaughtered 
in greater number. Once these criteria have been met, these 
establishments will be required to complete a new sampling window that 
verifies process control only once each year, in the 3-month period of 
June through August, or when a change has been made in the slaughter 
process or personnel.
    The Agency is permitting these very low volume establishments to 
conduct as few as 13 tests per year, in part because of their 
relatively simple and stable production environments. The slaughtering 
equipment in many cases may consist merely of a skinning bed, hoist, 
bonesaw (for poultry establishments, a small scalding tank, small 
defeathering device), and/or several types of knives. There are fewer 
personnel and there is less turnover in general. Of course, these 
establishments do change. Should there be any substantial changes in 
installed equipment or personnel, a new sampling window must be 
completed. These establishments must also complete a successful 
sampling window annually, regardless of whether there have been any 
substantial changes, in order to verify that the performance criteria 
continue to be met. Many small, nonsubstantial changes, in aggregate, 
may have an impact on process control. This annual testing must be 
conducted during the summer months of June through August, when there 
is a seasonal peak in the occurrence of foodborne diseases attributable 
to the major bacteria pathogens. Published and summary reports of 
Centers for Disease Control and Prevention (CDC) outbreak and sporadic 
disease surveillance have documented this seasonal trend for Salmonella 
spp.<SUP>5,6 and for Campylobacter jejuni/coli.<SUP>7 Although national 
surveillance for E. coli O157:H7 is relatively new and data are not 
available, Washington State surveillance has documented a similar 
seasonal trend for that pathogen.<SUP>8 The proposed requirement of one 
Salmonella sample per day would have assured testing during this 
period.
---------------------------------------------------------------------------

    \5\ Bean, N.H. and P.M. Griffin. 1990. Foodborne Disease 
Outbreaks in the United States, 1973-1987. J. Food Protection. 
53:804-817.
    \6\ Centers for Disease Control and Prevention. 1995. Salmonella 
Surveillance, Annual Tabulation Summary, 1993-1994. U.S. Department 
of Health and Human Services, Public Health Service, Atlanta, GA.
    \7\ Tauxe, R.V., N. Hargrett-Bean, C.M. Patton, and I.K. 
Wachsmuth. 1988. Campylobacter Isolates in the United States, 1982-
1986. MMWR. 37 (SS-2):1-13.
    \8\ Ostroff, S.M., J.M. Kobayshi, and J.H. Lewis. 1989. 
Infections with Escherichia coli O157:H7 in Washington State. JAMA 
262(3):355-359.
---------------------------------------------------------------------------

    Therefore, the regulation specifies that when sampling and testing 
is done annually, instead of continually, it be conducted within a 13-
sample window between June and August each year. This annual sampling 
must occur during this period, regardless of when other sampling 
windows may have occurred. Completing a successful sampling window 
annually will verify that the slaughter process continues to meet the 
performance criteria or will point to the need to reassess and revise 
the HACCP plan.
    Another reason for this approach to very low volume establishment 
testing is that the total risk of exposure to enteric pathogens from 
product produced at such establishments is assumed to be small and 
roughly proportional to the amount of product produced. Eighty-one 
percent of establishments slaughtering cattle would meet this low 
volume criteria; however, these establishments together supply only 
1.5% of the total national production. Further, establishments meeting 
these low volume criteria constitute 86% of all swine establishments, 
accounting for 1.3% of overall production. Thirteen percent of all 
establishments slaughtering chicken would meet this low volume 
requirement; however, these establishments together supply only 0.05% 
of total national production. Similarly, 42% of all turkey 
establishments are low volume establishments accounting for only 0.1% 
of production.
    FSIS intends that establishments operating under a validated HACCP 
system use microbial testing in their process control verification 
activities, and is requiring that slaughter establishments under HACCP 
use E. coli testing for that purpose. As noted above, however, the 
Agency acknowledges that there may be other, perhaps equally effective 
alternative approaches for determining sampling frequencies for E. coli 
testing for process control verification in slaughter establishments 
with a carefully designed HACCP system. The Agency is aware that 
comparable models have been developed in the context of quality 
assurance programs. These models, however, are part of programs that, 
like HACCP, involve more than mere statistical sampling, and usually 
are much more oriented to specific establishment/process/product 
combinations. Such models cannot easily be transferred to a nationwide 
collection of producers of a product, each with unique characteristics. 
The frequency rule established in this regulation recognizes the 
relevance of establishment characteristics in the area of verification, 
as in other facets of the HACCP plan, and therefore allows slaughter 
establishments to alter frequencies as appropriate for their 
circumstances when they institute HACCP. That is, slaughter 
establishments under HACCP may use a sampling frequency other than that 
provided for in the regulation, if the alternative sampling frequency 
is an integral part of the establishment's HACCP verification 
procedures and if FSIS does not determine, and notify the establishment 
in writing, that the alternative frequency is inadequate to verify the 
effectiveness of the establishment's processing controls. 
Establishments electing to institute HACCP prior to the dates required 
may use an alternative sampling frequency upon presentation to FSIS of 
data demonstrating the adequacy of that sampling frequency for 
verification of process controls to prevent fecal contamination.
    Establishments currently using an alternative E. coli sampling 
frequency for process control purposes, but not yet under a HACCP plan, 
will have to test at the frequencies specified in the regulation unless 
they have been granted an exemption by FSIS. However, after 
consideration of comments received on this rule that may result in 
protocol changes affecting all establishments, and publication of a 
Federal Register document addressing the comments, FSIS will consider 
requests for such exemptions on a case-by-case basis, upon the timely 
submission to FSIS of data demonstrating the adequacy of the 
alternative frequency for verification of process controls to prevent 
fecal contamination.
Sampling and Analytical Methodology
    Carcasses within the same establishment and in different 
establishments must be sampled and analyzed in the same manner if the 
results are to provide a useful measure of process control. Such 
consistency also will facilitate FSIS verification activities. As 
discussed below, the performance criteria are applicable to each type 
of carcass, industry-wide, based on FSIS's national baseline survey 
data. Because each establishment's performance is measured against the

[[Page 38843]]

performance of all surveyed establishments producing the same kind of 
product, it is essential that all like establishments adhere to the 
same basic sampling and analysis requirements.
    Each establishment is responsible for having written sampling 
procedures that are to be followed by a designated employee or agent. 
Samples are to be taken randomly at the required frequency. If an 
establishment runs more than one line, the lines from which samples are 
to be taken also are to be selected randomly. Samples from livestock 
carcasses are to be collected by a nondestructive method that requires 
a commercially available sampling sponge to be rubbed on the carcass 
surface after the carcass has been chilled in the cooler for 12 hours 
or more after slaughter. Establishments are required to take samples 
from three sites on each carcass. These three sites are the same ones 
that were used by FSIS when conducting the baseline studies for cattle 
and swine. On cattle carcasses, establishments will take samples from 
the flank, brisket, and rump areas; on swine carcasses, samples will be 
taken from the ham, ``belly,'' and jowl areas. The sponge is to be 
placed afterwards in an amount of buffer to transfer any E. coli to a 
solution, which then is analyzed for E. coli. Samples from poultry 
carcasses will be collected by taking whole birds from the end of the 
chilling process, after the drip line, and rinsing them in an amount of 
buffer appropriate for the type of bird being tested.
    The sponge sampling technique to be used on swine and cattle 
carcasses has been subject to many studies. A sponge technique has been 
reported by Dorsa et al.<SUP>9 and others, including Gill et 
al.<SUP>10, as an acceptable means of in-plant sampling to detect fecal 
contamination.
---------------------------------------------------------------------------

    \9\ Dorsa, W.J., C.N. Cutter, G.R. Siragusa. 1996. Evaluation of 
Six Sampling Methods for Recovery of Bacteria from Beef Carcass 
Surfaces. Letters in Applied Microbial. 22:39-41.
    \10\ Gill, C.O. J.C. McGinnis, M. Badoni. 1996. Assessment of 
the Hygienic Characteristics of a Beef Carcass Dressing Process. J. 
Food Protection 59(2):136-140.
---------------------------------------------------------------------------

    The excision method for sample collection would not be acceptable 
for routine sampling to verify process control because this defaces the 
carcass, and some establishments would be required to sample 13 
carcasses per day. Instead, for both cattle and swine carcasses, the 
sponge method requires that 100 cm<SUP>2 at each of the three sites be 
sampled by swabbing, for a total area of 300 cm<SUP>2 compared to the 
60 cm<SUP>2 area of excised tissue analyzed in the baseline studies for 
cattle and swine. The results would still be reported on a square 
centimeter basis. The larger sampling area for the swabbing method is 
expected to provide results comparable to the excision technique.
    The exact correlation between the sponging technique and the 
excision technique used during the baseline surveys is being assessed 
by ARS. Currently available results indicate a high degree of 
correlation between the two. These studies and any other new microbial 
sampling data will be made available to the public. This sponging 
technique will also be used in the FSIS Salmonella program. FSIS is 
continuing to improve the sponging technique and welcomes comments.
    FSIS considered providing that samples be taken from only one site 
on livestock carcasses: from the brisket on cattle and the belly area 
on swine. Sampling from one site has advantages. It would be less labor 
intensive. Further, sampling from one site might pose fewer worker 
safety problems than sampling from three sites because, for the latter 
option, a ladder generally is needed to reach the rumps of the 
suspended carcasses. Nonetheless, FSIS has determined that slaughter 
establishments must take samples from the three sites from which 
samples were drawn during the baseline studies or programs in the 
absence of data demonstrating that one-site sampling also will provide 
results comparable to the baseline survey data. The Agency invites 
comments on its requirement that establishments collect samples from 
the specified three sites on swine and cattle carcasses and the 
adequacy of alternative sampling approaches.
    Samples may be analyzed in either the establishment's own 
laboratory or a commercial laboratory. Samples must be analyzed by a 
quantitative method of analysis for E. coli. The method must be 
approved by the Association of Official Analytic Chemists or validated 
by a scientific body in collaborative trials against the three tube 
most probable number (MPN) method and agreeing with the 95 percent 
upper and lower confidence limit of the appropriate MPN index.
    FSIS has developed and is publishing as an appendix to the document 
guidelines that provide additional, detailed information on how best to 
sample, test, record, and interpret results for E. coli under this 
regulation. FSIS invites comment on these guidelines.
Recordkeeping
    Results of each test must be recorded, in terms of colony forming 
units per milliliter (cfu/ml) for poultry carcasses or per square 
centimeter (cfu/cm<SUP>2) for livestock carcasses, on a process control 
chart or table that permits evaluation of the test results in relation 
to preceding tests in accordance with the applicable criteria. These 
records must be maintained at the establishment for 12 months and must 
be made available to Inspection Program employees on request. 
Inspectors will monitor results over time, to verify effective and 
consistent process control.
Use of E. Coli Test Results by Establishments
    As discussed in preceding sections, establishments slaughtering 
livestock or poultry are required to use E. coli testing and evaluation 
of the results to verify the adequacy of their process controls for 
fecal contamination. Any test result in the marginal range (above m) 
indicates to the establishment that there is a potential problem in its 
processing control that may require attention. If the number of test 
results above m exceeds the specific number allowed, c (3, for all 
species), in the specific number of consecutive tests in the moving 
window, n (13 for all species), the establishment has failed to meet 
the performance criteria, and a significant question has been raised 
about the adequacy of the establishment's process controls for fecal 
contamination. Review of the process by the establishment and necessary 
corrective actions are strongly suggested.
    Results above the upper value M are unacceptable and should trigger 
immediate establishment review of slaughter process controls to 
discover the cause of the failure and to prevent recurrence, and, if a 
product has been affected, to consider the status and proper 
disposition of the product as the circumstances dictate.
Use of E. coli Test Results by FSIS
    FSIS personnel, like establishment personnel, will use the E. coli 
test results to help assess how well the establishment is controlling 
its slaughter and dressing processes. FSIS will compare establishment 
test results to the applicable E. coli performance criterion. A single 
failure to meet the criterion does not by itself demonstrate a lack of 
process control or product adulteration, but it will trigger greater 
inspection activity to establish that all applicable sanitation and 
process control requirements are being met and product is not being 
adulterated. Inspectors may make additional visual inspections of 
products and/or equipment and facilities, collect samples for FSIS 
laboratory analysis, and retain or condemn product, as appropriate. In 
addition, Sanitation

[[Page 38844]]

SOP's and HACCP records will be reviewed, as appropriate. Failure to 
meet the criterion may also result in the establishment being selected 
for intensified Agency testing for Salmonella under the pathogen 
reduction performance standard sampling program; and, if the 
establishment produced ground beef, its product could be targeted in 
the E. coli. O157:H7 ground beef testing program.
    The E. coli test results will be used by FSIS, along with all other 
relevant data and observations, including past establishment 
performance, to determine whether a slaughter establishment is meeting 
its process control responsibilities. Repeated failures to meet the 
criterion would lend support to a finding that the establishment's 
process controls are inadequate. Failure to maintain adequate process 
control will result in suspension and withdrawal of inspection, as 
appropriate. Such actions will be made in accordance with rules of 
practice that will be adopted for those proceedings.
    After a slaughter establishment implements HACCP, the E. coli 
testing program will continue as a HACCP verification activity. 
Isolated or occasional failures to meet the E. coli performance 
criterion may indicate that establishment personnel need to take 
corrective actions spelled out in their HACCP plan. Repeated failures 
to meet the criterion will result in FSIS focusing its verification 
oversight on relevant CCP's, which could lead to the need for HACCP 
plan reassessment by the establishment, as well as other inspection and 
compliance related activities that may be appropriate, as discussed 
above.
Implementation Timetable
    Six months from this publication date, establishments that 
slaughter livestock or poultry will be required to begin sampling and 
testing for E. coli at the volume-based rates described above. From 
that time, those establishments that do not test or fail to keep 
records of results as prescribed by the regulation will be subject to 
withdrawal of inspection in accord with the procedures set forth in 9 
CFR 335.13 or 381.234. After another six months, i.e., 12 months after 
publication of this final rule, after establishments have had an 
opportunity to gain experience in conducting this testing, recording 
the results, and using the data to verify and improve process control, 
FSIS personnel will incorporate the review of establishment E. coli 
test results into its inspection routine.
    In considering the timeframe for implementing the E. coli testing 
requirement, FSIS has taken into account the practicality of initiating 
such testing in a large number of establishments, the potential utility 
of the resulting data to establishments as they prepare for HACCP 
implementation, and the added consumer protection of having 
establishments, particularly those scheduled to implement HACCP towards 
the end of the implementation timetable, initiating testing and 
evaluating results against the process control performance criteria. 
FSIS is aware that many establishments, especially large ones, already 
use microbial testing as a means of verifying their process control 
systems; many may already be testing for generic E. coli. Some of those 
establishments may already have HACCP plans in place as well. 
Establishments performing microbiological testing and already working 
under HACCP plans have found that such testing is an important element 
in conducting a hazard analysis, validating HACCP plans, and verifying 
the ongoing effectiveness of HACCP systems.
    For establishments that are not already performing microbiological 
testing and not operating under HACCP plans, the data will be valuable 
in revealing how well or poorly their slaughter process is performing 
in microbiological terms, when compared against the microbial 
characteristics of a large portion of national production, and will 
provide an indication of whether immediate actions are required to 
prevent product adulteration and protect food safety. In addition, such 
data, when accumulated over a period of time, will contribute to the 
conduct of hazard analyses and selection of process control measures. 
Collection of these data will provide benchmarks for each establishment 
as it begins to understand the food safety implications of its 
processes and how to improve them.
    In the meantime, FSIS personnel, using the performance criteria as 
benchmarks for overall industry performance in terms of the number of 
E. coli organisms found on carcasses at a specific point in the 
slaughter process, will be able to review establishment data and other 
evidence to determine if each establishment is achieving an acceptable 
level of performance.
Request for Comments
    The Agency is soliciting additional comment and information on a 
number of technical issues concerning the protocols for E. coli 
testing, and on that basis will consider adjusting those protocols 
prior to the effective date. In particular, two concerns have been 
raised on the issue of the rule's statistical framework: 1) the 
representativeness of the proposed sample collection, and 2) the levels 
and distribution of E. coli on carcasses and the ways in which these 
levels affect the utility of the proposed testing protocol.
    Because poultry slaughter establishments must collect samples with 
a whole bird rinse, the representativeness of the sampling site is not 
an issue; the entire bird is being sampled. FSIS used this technique 
when collecting baseline data and therefore, establishment data should 
be comparable to baseline survey data. Further, greater than 99 percent 
of broiler carcasses in the national baseline survey had detectable E. 
coli. Generic E. coli testing data therefore clearly will be useful to 
poultry slaughter establishments as they initiate HACCP and begin to 
verify the associated process control procedures. E. coli testing 
procedures for poultry required by this rule comport well with the 
available scientific data and discussions held as part of the public 
comment process.
    More difficult issues arose in developing E. coli sampling 
procedures for cattle and swine carcasses. Part of the concern, as 
discussed, stems from the fact that a whole carcass rinse is impossible 
with a large carcass, and thus it is necessary to select specific 
sampling sites. Selections of sites, in turn, may influence results, 
particularly if generic E. coli is not randomly distributed on the 
carcass. Site selection may also influence the usefulness of resultant 
data. For example, the appropriate response to an elevated generic E. 
coli level on the rump of a beef carcass may be different from the 
appropriate response to an elevated generic E. coli level at the site 
of the midline incision. The Agency wants comments on the relative 
merits of a one-site versus three-site sampling approach.
    Another concern revolves around the correlation between non-
destructive and destructive sampling. The baseline surveys used 
destructive sampling, that is, culturing of tissue excised from the 
carcass. FSIS agrees with commenters that reasonable results can be 
obtained with a non-destructive swabbing technique for sampling. 
Preliminary data indicate that results obtained with a destructive and 
non-destructive sampling are comparable, although studies continue.
    Another concern arises from the statistical basis for E. coli 
testing. In

[[Page 38845]]

particular, the levels of generic E. coli on cattle carcasses in the 
national baseline survey were low, with the majority of carcasses 
having no detectable E. coli. This could raise questions about the 
utility of the E. coli test results in evaluating process controls in 
establishments slaughtering cattle.
    The principal utility of process control testing stems from the 
availability to a establishment of results over time from that 
establishment. The tracking of trends and identification of anomalous 
results permits isolation and correction of problem areas that might 
otherwise go unnoticed. FSIS has concluded that testing for generic E. 
coli is the appropriate and necessary means by which meat and poultry 
slaughter establishments must evaluate and verify the adequacy of their 
process controls. FSIS considers systematic measures to prevent and 
remove fecal contamination and associated bacteria, coupled with 
microbial testing to verify effectiveness, to be the state of the art 
in slaughter establishment sanitation. Microbial testing for bacteria 
that are good indicators of fecal contamination and the regular 
availability of test results will help to focus establishments on the 
effectiveness of their measures for preventing and removing fecal 
contamination and will provide information establishments can use in 
maintaining adequate process control. FSIS reached this conclusion upon 
its review of written comments received on the proposal and comments 
made at the scientific conferences and public meetings, as well as 
available scientific data, and has retabulated and reassessed its 
baseline data as it applies to the E. coli testing in the rule.
    In the first reassessment, it was determined that the lower levels 
and more frequent negative test results of E. coli found on livestock, 
particularly steers and heifers, as compared to poultry in the baseline 
survey data does not undercut the utility of the E. coli criteria which 
are also based on the baseline survey data. FSIS tested the performance 
criteria in this rule by applying it to plant-specific test results 
obtained during the baseline surveys. FSIS looked at data from 
establishments for which at least 20 test results were available, and 
listed the results by collection date much as would be done by the 
establishments under the rule. The Agency found that about half of the 
establishments in each of the livestock slaughter categories fully met 
the criteria, which suggests that those establishments have good 
process controls for prevention of fecal contamination. The Agency also 
found that many establishments failed to meet the applicable E. coli 
criterion (any result above M, or more than 3 results above m out of 
the most recent 13 test results): 2 out of 30 steer/heifer 
establishments, 10 out of 34 cow/bull establishments, and 11 out of 31 
market hog establishments failed to meet the criterion at least 20% of 
the time, suggesting that a significant number of livestock slaughter 
establishments should review and make adjustments to their process 
controls.
    The Agency also made an assessment of whether the baselines show 
true differences in E. coli results among establishments that slaughter 
the same categories of livestock. The Agency did a statistical analysis 
of a hypothesis: percents positive are equal among establishments 
slaughtering the same category of livestock. The analysis involved 
comparing E. coli test results of pairs of establishments. This 
comparison showed wide ranges in the percents positive between 
establishments albeit smaller differences among steer/heifer 
establishments. The percents positive ranged between 0.0 to 27.1 for 
steer/heifer establishments, 0.0 to 45.2 for cow/bull establishments, 
and 2.2 to 97.1 for market hog establishments. The hypothesis, 
therefore, was rejected because the data showed significant differences 
in the prevalence of E. coli on carcasses of animals found in 
establishments slaughtering the same categories of livestock.
    The retabulated data developed for these two analyses are available 
for viewing in the FSIS Docket Room (See ADDRESSES) as part of the 
administrative record of this rulemaking.
    FSIS invites comments on the statistical frameworks it has used for 
E. coli testing and performance criteria. The Agency is open to the 
possibility that it might further improve its testing protocols prior 
to the implementation date, and is seeking additional relevant 
scientific and economic data. In particular, in light of the concerns 
noted above, FSIS is seeking additional data relating to the 
distribution of generic E. coli on cattle and swine carcasses, 
differences in E. coli levels within and between establishments, and 
the appropriateness of various data sets for establishing the proposed 
80th and 98th percentile national criteria for generic E. coli levels 
on cattle and swine carcasses.
    FSIS also requests comments and information addressing the 
following questions:

    Are there alternative, equally or more effective risk based 
microbial sampling protocols that could be used for process control 
verification by establishments that slaughter cattle or swine?
    Are there more appropriate anatomical sites for microbial 
testing than those adopted?
    Are there alternative sampling frequencies that would elicit 
results more indicative of process control performance?
    How could the proposed testing protocol be revised to better 
account for differing establishment characteristics and how can FSIS 
minimize the cost to establishments of E. coli testing without 
sacrificing testing effectiveness?
    Are there worker safety concerns regarding sampling from 
difficult to reach carcass sites and, if so, how might they be 
mitigated?
    Given that testing is based on production volume, are there 
effective approaches other than requiring very small establishments 
to conduct a minimal amount of testing during certain months of the 
year?

    FSIS is aware that some individuals, companies, and trade groups 
have conducted research and have data on the various carcass sampling 
sites and associated levels of bacteria at these sites (carcass 
mapping). FSIS welcomes any information concerning E. coli and other 
microorganisms at various sites on carcasses.
    FSIS has opted to establish performance criteria based on the 
levels and distribution of E. coli for the various slaughter classes. 
Some individuals and companies may have established their own criteria 
for process control verification. FSIS welcomes information on the 
rationales, sampling plans and protocols on which any such criteria are 
based, as well as data (or data summaries) collected under such 
protocols.
    FSIS welcomes any new or unpublished research results or 
information that exists concerning the relationship between the 
presence of generic E. coli and the presence of other pathogenic 
microorganisms on cattle and swine carcasses.
    FSIS specifically invites establishments currently conducting 
generic E. coli testing for process control verification to submit data 
regarding their costs, including labor and training costs, as well as 
testing costs per unit. FSIS will use this data to assess the merits of 
alternative testing protocols.
    FSIS invites comments on how, and the extent to which, it should 
summarize and make available to the industry and public E. coli testing 
data made available to it under these regulations. Reports on the 
collective experiences of establishments with various characteristics 
could be useful to the industry, the Agency, and the public at large.
    In light of these issues, in particular those reflecting continuing 
concerns

[[Page 38846]]

about the applicability of the national criteria to all affected 
establishments, the frequency and other parts of the testing protocols, 
and the statistical utility of the establishment's test results as a 
measure of process control, FSIS plans to conduct two public 
conferences. The first conference is planned to be held approximately 
45 days into the 60 day comment period following publication of this 
rule. This public conference will be led by a panel of scientists from 
FSIS and other government agencies who will listen to testimony and 
review comments received on these technical issues and share their 
observations and opinions. FSIS will consider their input along with 
all comments received as the basis for any necessary technical 
amendments, which will be completed at least 30 days before the 
implementation date. The second public conference is tentatively 
planned for approximately 9 months following publication of this final 
rule. This conference would be an opportunity for the industry and 
others to discuss with FSIS new information based on about 3 months of 
testing experience that may bear on these same issues and might allow 
for further adjustments of protocols before FSIS inspectors are tasked, 
about three months later, with comparing test results to the national 
criteria as part of their inspection routine. FSIS will publish 
further, more detailed notice of these conferences in future issues of 
the Federal Register.
Pathogen Reduction Performance Standards
    The pathogen reduction performance standards for Salmonella FSIS is 
establishing in this final rule complement the process control 
performance criteria for fecal contamination and E. coli testing.
    The likelihood of product contamination by Salmonella is affected 
by factors in addition to the incidence or degree of fecal 
contamination, including the condition of incoming animals and cross 
contamination among carcasses during the slaughter process and further 
processing. Under HACCP, establishments will be expected to establish 
controls wherever practicable to address and reduce the risk of 
contamination with harmful bacteria. The pathogen reduction performance 
standards FSIS is establishing for Salmonella are an important step 
toward enabling FSIS and the establishment to verify the aggregate 
effectiveness of an establishment's HACCP controls in reducing harmful 
bacteria.
Rationale for Selecting Salmonella
    In the future, FSIS may develop pathogen reduction performance 
standards targeting a number of pathogens. Initially, however, FSIS has 
developed pathogen reduction performance standards only for one--
Salmonella. Salmonella is an enteric pathogen, which as a group cause 
most preventable illnesses associated with meat and poultry.
    FSIS has selected Salmonella because: (1) it is the most common 
bacterial cause of foodborne illness; (2) FSIS baseline data show that 
Salmonella colonizes a variety of mammals and birds, and occurs at 
frequencies which permit changes to be detected and monitored; (3) 
current methodologies can recover Salmonella from a variety of meat and 
poultry products; and (4) intervention strategies aimed at reducing 
fecal contamination and other sources of Salmonella on raw product 
should be effective against other pathogens.
Basis for Performance Standards and Plans for Future Adjustments
    The pathogen reduction performance standards for Salmonella are 
based on the current prevalence of Salmonella, as determined from 
FSIS's baseline surveys. Current prevalence percentages based on the 
data from these surveys are listed in Table 4 and in the regulations 
(new Secs. 310.25(c)(3)(ii) and 381.94(c)(3)(ii)) under the column 
headed ``Performance Standard.'' This is the performance standard that 
establishments must achieve, not on a lot-by-lot basis, but 
consistently over a period of time through appropriate and well-
executed process control.
    This is the same approach to setting the ``interim targets for 
pathogen reduction'' that FSIS proposed in its Pathogen Reduction/HACCP 
proposal. As explained in the preamble to that proposal, basing the 
performance standard on the national baseline prevalence means that 
some establishments are already meeting or exceeding the standard, 
while other establishments are not. FSIS believes that it is feasible 
for all establishments to meet or exceed the current baseline 
prevalence of contamination with Salmonella, through careful process 
control to prevent contamination and incorporation of readily available 
food safety technologies and procedures to remove contamination. The 
feasibility of achieving this standard is demonstrated by the fact that 
many establishments are already doing so.
    The Agency believes that most establishments maintaining sanitary 
conditions under their Sanitation SOP's and operating under validated 
HACCP plans, as provided for elsewhere in this regulation, will be able 
to meet the pathogen reduction performance standards without major new 
costs. For example, HACCP plans for slaughter establishments are 
expected to address the condition of incoming animals, and may provide 
for more systematic control of relevant processes or interventions, 
such as the cleaning of animals or carcasses before evisceration. HACCP 
systems should, therefore, result in many establishments improving the 
microbial profile of their finished raw products.
    Slaughter establishments concerned that they might not meet the 
pathogen reduction performance standard have available a wide range of 
technologies shown to reduce the levels of pathogens that may be on the 
surface of carcasses. As discussed in some detail in the proposed rule, 
antimicrobial treatments normally include washes or sprays that use 
either hot water or a solution of water and a substance approved by 
FSIS for that use. Such substances include acids (lactic, acetic, and 
citric), trisodium phosphate (TSP), and chlorine. In addition, FSIS has 
recently established that spray-vacuum devices that apply pressurized 
steam or hot water to beef carcasses and immediately vacuum it up also 
are effective in reducing bacteria on carcasses.
    Establishments producing raw ground product from raw meat or 
poultry supplied by other establishments cannot use technologies for 
reducing pathogens that are designed for use on the surfaces of whole 
carcasses at the time of slaughter. Such establishments may require 
more control over incoming raw product, including contractual 
specifications to ensure that they begin their process with product 
that meets the standard, as well as careful adherence to their 
Sanitation SOP's and HACCP plan.
    By basing its Salmonella performance standards on the current 
national baseline prevalence for each major species and product class, 
FSIS is applying a uniform policy principle: all establishments must 
achieve at least the current baseline level of performance with respect 
to Salmonella for the product classes they produce. This policy is 
based on the public health judgment that reducing the percentage of 
carcasses with Salmonella will reduce the risk of foodborne illness, 
and on the regulatory policy judgment that establishing for the first 
time a clear standard for Salmonella, in conjunction with the 
implementation of HACCP, will lead to significant reductions in

[[Page 38847]]

contamination rates. This policy is not based on a quantitative 
assessment of the risk posed by any particular incidence of Salmonella 
contamination or the determination of a ``safe'' incidence or level. 
There is not currently a scientific basis for making such assessments 
or determinations.
    FSIS recognizes that this approach results in a range of 
performance standards among the various product classes (see Table 4). 
For example, the current Salmonella prevalence for broilers is 20 
percent, while the current prevalence for steers and heifers is 1 
percent. This range reflects the current level of performance for each 
class of product, as reflected in the FSIS baseline surveys.
    FSIS intends to revise its Salmonella performance standards 
periodically as new baseline prevalence data become available and in 
furtherance of the Agency's goal of reducing the risk of foodborne 
illness. FSIS will periodically repeat its baseline studies to assess 
the overall progress of the pathogen reduction effort. Also, as 
indicated below in the discussion of the FSIS testing strategy, FSIS 
will be conducting extensive Salmonella testing to ensure compliance 
with the pathogen reduction performance standards. If the data from 
this testing or future baseline surveys justify revision of the 
performance standards, FSIS will promptly publish such revisions for 
public comment in the Federal Register. FSIS anticipates revision of 
these performance standards downward as justified by progress in 
pathogen reduction and demonstrated reductions in the national baseline 
prevalence of Salmonella. In making such adjustments, FSIS will take 
into account the state of scientific knowledge, available technology, 
feasibility, and public health benefits to be achieved. FSIS will also 
consider the current level of industry performance with respect to 
Salmonella prevalence in particular classes of livestock and poultry. 
It is anticipated that such adjustments would more likely occur in 
classes with the highest prevalence. FSIS originally proposed to call 
these performance ``interim'' standards or targets. The final rule 
removes that language.
    Approximately 15 months after the publication of this final rule, 
FSIS will convene a public conference to review available Salmonella 
data and discuss whether they warrant refining the Salmonella 
performance standards. Prior to the conference, FSIS will make 
available the data resulting from the pre- implementation phase of the 
FSIS Salmonella testing program. FSIS also will take advantage of this 
conference to receive public input on the E. coli testing program. FSIS 
will extend an invitation to all interested parties.
    Additionally, FSIS intends to work closely with other Federal 
agencies and the scientific community to improve the scientific basis 
for establishing food safety performance standards for microbial 
pathogens. In particular, the Executive Office of the President, Office 
of Science and Technology Policy, will oversee a task force to 
determine what research and data collection are needed to develop a 
workable approach to quantitative risk assessment for foodborne 
pathogens and determine the most cost-effective way of conducting the 
necessary research. FSIS and other USDA agencies will participate in 
this government-wide task force.
Determining Compliance With the Standard
    The pathogen reduction performance standards specify for each 
species and category of raw product a maximum number of positive test 
results (c) permitted to be found in a specified number of samples (n) 
for each class of raw product before the establishment will be deemed 
to be exceeding the performance standard. The standards were determined 
by first calculating for each category of product tested in the FSIS 
national baseline programs and surveys the percentage of Salmonella 
positives nationwide. This is, in effect, the performance standard that 
must be achieved consistently by each establishment over time. Then the 
number of samples to test (n) and the number of positives to allow from 
among those samples (c) were calculated to provide approximately an 80% 
probability of passing when the establishment is operating at the 
national baseline prevalence of Salmonella positive results, i.e., just 
within the performance standard. As discussed in the preamble to the 
Pathogen Reduction/HACCP proposal and above with respect to E. coli 
testing, the statistical criteria for evaluating Salmonella test 
results balance the need to prevent establishments from failing to meet 
the standard, based on chance results, and the need to ensure both that 
violations are readily detected and that establishments have an 
incentive to improve their performance beyond what is minimally 
required by the standard. The resulting values for the pathogen 
reduction performance standards are shown in Table 4.

                               Table 4.--Pathogen Reduction Performance Standards                               
----------------------------------------------------------------------------------------------------------------
                                                                    Performance                                 
                                                                     standard                     Maximum number
                                                                     (percent        Number of     of positives 
                        Class of product                           positive for       samples       to achieve  
                                                                    Salmonella)     tested (n)     standard (c) 
                                                                        (%)                                     
----------------------------------------------------------------------------------------------------------------
Steers/Heifers..................................................             1.0              82               1
Cows/Bulls......................................................             2.7              58               2
Ground Beef.....................................................             7.5              53               5
Fresh Pork Sausage..............................................             *NA             *NA             *NA
Broilers........................................................            20.0              51              12
Hogs............................................................             8.7              55               6
Ground Turkey...................................................            49.9              53              29
Ground Chicken..................................................            44.6              53              26
Turkeys.........................................................             *NA             *NA             *NA
----------------------------------------------------------------------------------------------------------------
* Not available at this time.                                                                                   

    FSIS has concluded that, for purposes of this rulemaking, it should 
rely only on FSIS baseline data for determinations of the prevalence of 
bacteria on which it is establishing standards. The proposal discussed 
the possibility of relying on other data sources, such as industry 
surveys or other reports in the scientific literature. No such data 
were

[[Page 38848]]

submitted to FSIS in response to the proposal, and FSIS has concluded 
that those alternative data sources are not likely to provide the 
nationwide, objective data that are needed for the Agency's regulatory 
purpose of establishing performance standards. FSIS will consider 
modifications of the scope and approach to these surveys and additional 
data sources, as the needs of public health dictate, but will continue 
to rely only on data that are gathered with appropriate scientific 
rigor.
    FSIS has completed its baseline survey work and has issued reports 
on its findings for Steers/Heifers, Cows/Bulls, Broiler Chickens, 
Market Hogs, Ground Beef, Ground Chicken, and Ground Turkey. Copies of 
these reports are available for inspection in the FSIS Docket Room (see 
ADDRESSES).
    FSIS is currently conducting the fresh pork sausage survey and will 
begin the Baseline Program for turkeys soon. Therefore, performance 
standards for fresh pork sausage and turkeys cannot be established at 
this time. The performance standards for these two classes of products 
will be published for public comment once FSIS's reports on the data 
are available.
    FSIS will determine an establishment's compliance with the 
applicable pathogen reduction performance standard by taking the 
indicated number of samples, generally at the rate of one or more per 
day, testing each sample for Salmonella, and determining whether the 
number of positive results is above the maximum permitted for that 
product in the regulation.
    FSIS has established performance standards for Salmonella on 
carcasses and on raw products derived from meat and poultry. Because 
Salmonella is more likely to be present on raw, ground, or comminuted 
products than on the carcasses from which they are derived, raw, 
ground, or comminuted product ordinarily will be the focus of FSIS 
compliance testing in those establishments that both slaughter and 
produce raw ground product.
    The pathogen reduction performance standard applies to 
establishments, not to individual products. As discussed, 
microbiological testing of raw products for purposes of routinely 
separating adulterated from unadulterated products is impractical at 
this time. The pathogen reduction standard for Salmonella requires 
testing of products not for purposes of determining product disposition 
(although in some circumstances it may contribute to additional 
inspection or compliance activities that do), but rather as a measure 
of the effectiveness of the process in limiting contamination with this 
particular pathogen. If an establishment fails to meet the standard, it 
must institute corrective actions to lower the incidence of Salmonella 
on all such product it produces as measured by subsequent testing, or, 
ultimately, it must cease producing that product. The FSIS enforcement 
strategy is further discussed below.
FSIS Testing Strategy
    FSIS's Salmonella testing program will be implemented in two 
phases, a pre-implementation phase and a compliance phase. The pre-
implementation phase will begin approximately three months after 
publication of the final rule and initially will consist of an 
establishment-by-establishment survey of the slaughter establishments 
represented in the National Microbiological Baseline Data Collection 
Programs. These establishments account for approximately 99 percent of 
the total production volume for each of the major species slaughtered 
nationwide. The testing in each slaughter establishment will be 
conducted in a manner designed to provide a reliable picture of the 
establishment's performance throughout a 12-month period, in relation 
to the pathogen performance standard applicable to the species being 
slaughtered. It is anticipated that initially FSIS will take 
approximately 250 samples per establishment over a one-year period, 
with testing to be completed before the implementation date for the 
standard in each establishment.
    FSIS will also conduct pre-implementation testing in ground product 
establishments and in establishments that account for the remaining one 
percent of production and that were not included in the FSIS baseline 
surveys. This testing will be conducted in a manner and at a level that 
takes into account the size and nature of the establishments involved. 
FSIS will provide more detail on this testing soon in a separate 
notice.
    This pre-implementation testing will inform both the establishments 
and FSIS, prior to the actual enforcement of the performance standards, 
whether each establishment is already meeting the standard, is close to 
meeting the standard, or requires substantial improvement to meet the 
standard. As with all FSIS testing done to check compliance with the 
pathogen reduction standards, the testing results will be provided to 
the establishment by FSIS. These testing results will assist 
establishments in designing and validating their HACCP plans as needed 
to ensure that products meet pathogen reduction performance standards. 
This information also will assist FSIS to more effectively target its 
compliance testing after the standards go into effect, as discussed 
below. This FSIS-generated data on the prevalence of Salmonella on 
inspected products will be available to the public.
    Upon the implementation of HACCP, and upon publication of Federal 
Register documents concerning the pathogen reduction performance 
standards for which baseline survey reports have not yet been 
published, FSIS will initiate phase 2, the compliance phase, of its 
Salmonella testing program in affected establishments. As an integral 
part of its overall responsibility for food safety, FSIS will conduct 
an ongoing testing program to determine compliance with the Salmonella 
performance standard for all classes of livestock and poultry. In 
addition, FSIS will conduct a program of targeted testing where 
warranted. The frequency and intensity of this testing will be 
determined based on past establishment performance, the establishment's 
own generic E. coli test results, FSIS inspectional observations, 
reports of illness associated with product produced at an 
establishment, the results of Salmonella testing during the pre-
implementation phase, previous failures to meet the performance 
standards, and other factors.
    The costs to FSIS of this testing for Salmonella, estimated to be 
approximately 2 million dollars annually, are addressed in the Final 
Regulatory Impact Analysis of this rule.
FSIS Testing Methods
    Details of the sample collection and testing procedures the Agency 
will be using are in Appendix E, ``FSIS Sample Collection Guidelines 
and Procedure for Isolation and Identification of Salmonella from Raw 
Meat and Poultry Products.''

FSIS Enforcement Strategy

    The objective of FSIS's enforcement policy with respect to 
microbial testing is to achieve compliance with the regulations. With 
respect to Salmonella, the Agency's goal is to achieve pathogen 
reduction by ensuring that all slaughter and ground product 
establishments meet the performance standards established by FSIS. FSIS 
intends to achieve this goal through an enforcement strategy based on 
the two-part testing program mentioned above: the ongoing testing, 
which will include all establishments at some fixed interval, 
irrespective of performance;

[[Page 38849]]

and targeted testing focusing on establishments unable to meet the 
Salmonella performance standard when tested by FSIS or for the other 
reasons discussed above.
    The Salmonella enforcement strategy will embody an objective, 
uniform systems approach to ensure that it is administered and applied 
in a fair, equitable, and common-sense manner. The Agency will 
carefully monitor and adjust its enforcement program on an ongoing 
basis to ensure that its enforcement activities reflect these 
principles while ensuring food safety.
    If ongoing or targeted testing in an establishment indicates the 
performance standard is not being met, FSIS will decide whether to 
conduct follow-up testing on the basis of several factors. If an 
establishment with Salmonella test results marginally above the limit 
takes corrective action, FSIS could judge, based on the establishment's 
actions and other factors relevant to ensuring food safety, that 
immediate follow-up testing is not necessary. If, however, that 
establishment were to take inadequate corrective action after failing 
to meet the Salmonella performance standard, or if it simply ignored 
that failure, FSIS will conduct a second series of tests. FSIS will 
invariably conduct further testing at all establishments whose test 
results significantly exceed the standard.
    If an establishment fails the second, targeted series of FSIS-
conducted tests, the establishment will be required to reassess its 
HACCP plan for the tested product, modifying the plan as necessary to 
achieve the Salmonella performance standard. If the establishment fails 
to modify its HACCP plan as necessary, or if it fails the third series 
of targeted tests, FSIS will suspend inspection services. The 
suspension will remain in effect until the establishment demonstrates 
its ability to meet the performance standard.
    The probability of an establishment failing the Agency's pathogen 
reduction standard three consecutive times is less than 1% when the 
establishment prevalence is at the limit of the standard.
Implementation Timetable for Pathogen Reduction Performance Standards
    Slaughter establishments and establishments producing raw, ground, 
and comminuted product subject to these pathogen reduction performance 
standards must meet the Salmonella standard at the time the 
establishment is required to implement HACCP. As explained in section 
II above, HACCP implementation will be phased in based on establishment 
size over a period of 18 to 42 months following the date of publication 
of this final rule. FSIS originally proposed a single two-year delayed 
effective date for its Salmonella performance standards. Many 
commenters argued that it was not reasonable to hold all establishments 
to the same effective date, and, furthermore, that it was more logical 
to hold establishments to compliance with the standard after, rather 
than before, HACCP was in place. This proposition also was strongly 
endorsed by many people who attended an information briefing and public 
meeting held by FSIS in Kansas City, Missouri, on May 22, 1995, 
expressly for small meat and poultry establishments and small 
businesses (60 FR 25869, May 15, 1995). They questioned, among other 
things, the need for and wisdom of a common implementation date for 
large and small establishments.
    Harmonizing the effective dates with implementation of HACCP is 
more consistent with the nature of the pathogen reduction standards as 
measures of what establishments can and should achieve through HACCP-
based process control. It will bring 74% of the nation's slaughter 
production of meat and poultry (by weight) under the performance 
standard 18 months following publication of this final rule. It will 
also facilitate the transition to HACCP, for both the FSIS workforce 
and affected establishments, by requiring all establishments to meet 
the performance standards as they implement HACCP.
Response to Comments
    FSIS proposed to require that all meat and poultry slaughtering 
establishments and establishments producing raw ground product conduct 
daily microbial testing to determine compliance with interim targets 
for the reduction of Salmonella. FSIS proposed to require a single 
qualitative test per day, with daily results to be accumulated over 
time to provide information regarding the performance of an 
establishment's process and to collect data sufficient for process 
control verification. Daily testing was considered the minimal sampling 
necessary to detect process deviations within a realistic time frame.
    The three issues most commonly raised by commenters concerning the 
proposed microbial testing requirements were the proposed selection of 
Salmonella as the indicator organism, the frequency of proposed 
testing, and the disproportionate costs to small establishments. Some 
commenters also argued that the regulatory approach was not justified 
and exceeded FSIS's legal authority.
The Indicator Organism
    Many commenters opposed the use of Salmonella as the indicator 
organism, arguing that its low incidence in beef makes it a poor 
indicator of pathogen reduction in the species, the positive/negative 
test result is a weak measure of process control, and, compared to some 
nonpathogenic alternatives such as generic E. coli, Salmonella tests 
are more difficult, time-consuming, and costly. Others commented that 
testing for Salmonella alone is unacceptable, as there is no direct 
correlation between the presence of this organism and other pathogens 
such as E. coli O157:H7, Listeria, and Campylobacter.
    Various alternative indicator organisms were suggested, including 
generic E. coli (biotype I), total plate counts, Enterobacteriaceae, 
Total Viable Counts (TVC), and Aerobic Plate Counts (APC). Commenters 
who recommended alternatives stated that tests for these organisms 
would be better indicators for process control and fecal contamination 
levels than tests for Salmonella. Still others requested that more 
studies be conducted to determine which type of indicator organism 
would be most useful for verifying process control.
    Some commenters recommended retaining Salmonella as the target for 
pathogen reduction, but suggested adding a requirement for generic E. 
coli testing because it serves effectively as an indicator of fecal 
contamination in all species. A minority of commenters supported the 
proposed use of Salmonella as the indicator organism because of its 
significance as a cause of foodborne illness and because there are 
relatively simple tests available for detecting Salmonella. Some 
commenters recommended requiring testing for Salmonella and additional 
pathogens in selected species or products based on the degree of public 
health risk posed by the pathogen. A number of consumer groups 
requested a pathogen goal of zero for E. coli O157:H7.
    These comments are generally addressed by the FSIS decisions to 
require slaughter establishments to test for generic E. coli as a means 
to verify process control for fecal contamination, and to have FSIS 
conduct testing for Salmonella for pathogen reduction.
    FSIS considers systematic measures to prevent and remove fecal 
contamination and associated bacteria, coupled with microbial testing 
to verify effectiveness, to be the state of the art in slaughter 
establishment sanitation. Further, FSIS believes that testing for 
generic E. coli is the appropriate and necessary means by which meat 
and poultry slaughter

[[Page 38850]]

establishments must verify their process controls. FSIS reviewed 
written comments received on the original proposal and comments made at 
the scientific conferences and public meetings, as well as available 
scientific data, and has decided to require slaughter establishments to 
conduct testing for generic E. coli to verify process controls.
    The Agency has concluded that each kind of testing serves an 
important function. Both play a major part in the Agency's pathogen 
reduction efforts, and working in unison will permit the Agency to use 
its inspection resources more effectively, and efficiently, thereby 
enhancing inspection.
    E. coli testing for process control verification and Salmonella 
testing to enforce the pathogen reduction performance standard both are 
aimed at FSIS's objective to reduce the incidence of disease caused by 
foodborne pathogens. However, E. coli testing and Salmonella testing 
aim at the objective from different directions.
    An ongoing screen for generic E. coli serves both the establishment 
and FSIS as a means of verifying that a slaughter facility's process is 
``in control'' with regard to prevention of fecal contamination of the 
carcasses being produced. In other words, it becomes a marker for 
verifying a slaughter establishment's adherence to the zero tolerance 
for fecal contamination. Such testing provides a standard measure for 
verification of process control at the critical slaughter stage of 
production. Without such a standard measure, there is no objective 
basis upon which either the establishment or FSIS can determine the 
adequacy of process controls, from one establishment to another, in 
preventing fecal contamination. It will permit establishments to make 
ongoing adjustments or changes to their slaughter process when 
necessary to meet the performance criteria. The test results will also 
guide FSIS's ongoing inspection, permitting adjustments in intensity 
and focus as appropriate.
    Generic E. coli testing to verify process control alone, however, 
does not adequately address legitimate public health concerns about 
pathogenic bacteria in and on raw product. E. coli (except for certain 
pathogenic subgroups) is not itself a cause of foodborne disease. It is 
a ``surrogate marker'' or ``indicator'' for fecal contamination, which 
in turn is a source of many pathogens that may contaminate products. 
Fecal contamination, however, does not always correlate with the 
presence of pathogens; high levels of E. coli may be present without 
pathogens, and pathogens may be present without high E. coli levels. 
Because testing for E. coli cannot serve as a surrogate for the 
presence of Salmonella, FSIS's specific public health objective of 
reducing nationwide Salmonella levels on raw meat and poultry products, 
including raw ground products, requires a standard and a testing regime 
that are directed at that pathogen.
    The pathogen reduction performance standard for Salmonella must be 
met by all inspected establishments producing raw meat and poultry 
products. Agency testing for Salmonella is necessary for enforcement of 
that requirement. Slaughter establishments' E. coli testing, a means 
for verifying process control for fecal contamination, should promote 
improved process controls which should, in turn, result in reductions 
of Salmonella and other pathogens. But, E. coli testing cannot measure 
actual reductions and control of Salmonella nor be the basis for Agency 
enforcement of the pathogen reduction standards.
    The test results from both kinds of testing are valuable to the 
Agency in the shift to a HACCP-based regulatory regime, but their value 
comes from the way they work together to verify the effectiveness of an 
overall system of preventive process control. The Agency continues to 
believe that pathogen reduction in inspected establishments requires 
that establishments build into their operations preventive measures and 
systems to reduce the potential for pathogens to be on products to 
begin with, and that such systems must be establishment-produced and 
establishment-specific. The Agency's HACCP and Sanitation SOP's 
regulations are intended to do that. However, these regulations are not 
self-enforcing. The Agency's inspection mandate does not permit it to 
simply assume that an establishment's systems are in fact producing 
uniformly safe and unadulterated products. Pathogen reduction will be 
achieved instead by the combination of HACCP plans validated as 
effective for pathogens of concern, E. coli testing by the 
establishment to provide on-going verification of process control for 
fecal contamination, and Salmonella testing by FSIS to enforce 
compliance with the pathogen reduction performance standards.
Frequency and Cost of Testing
    Many commenters questioned the proposed frequency of daily testing 
for each species and for raw, ground products. The majority of 
commenters who opposed daily testing stated that this testing 
requirement would place an unfair cost burden and have a negative 
economic impact on some establishments, especially small volume 
establishments and establishments producing multiple species and 
multiple ground products that would require multiple tests. These 
commenters stated that under the proposed sampling methodology, a small 
establishment could conceivably conduct more tests per day than a very 
large establishment with a much higher production volume. Also 
mentioned was the fact that many of these establishments do not have 
on-site testing facilities and would have an additional cost of 
shipping samples for testing.
    To minimize the economic impact on establishments, especially small 
establishments, some commenters suggested that FSIS should pay for 
microbial testing. Others recommended less than daily testing or other 
changes to the proposed sampling frequency. Various alternatives to the 
proposed sampling protocol were mentioned, but the sampling scheme 
recommended most often as the most equitable, and the one FSIS is 
requiring, is one based on production volume.
    Although many commenters requested less frequent testing than that 
proposed, others supported the one sample per day testing requirement 
as an efficient means of verifying process control. Still others 
recommended testing even more frequently than once per day. These 
commenters asserted that testing once a day is inadequate to verify 
process control or to screen out product with pathogens. Their main 
concern was that the proposed sampling frequency and moving sum 
statistical procedure would allow inadequate process control to go 
undetected, resulting in large quantities of suspect product being 
produced; recommendations were made for a testing frequency more 
proportional to an establishment's production volume.
    Some commenters requested that exemptions from the proposed daily 
microbial testing be made for small establishments and establishments 
that have consistently complied with their HACCP programs. Others 
requested exemptions for specific products including: raw ground meat 
products; cured products; thermally processed canned foods; frozen 
foods; boxed meat and beef and pork carcasses from other inspected 
establishments; minor species (i.e., sheep, lamb, goats, equines, 
guineas); and raw ground products to be further processed as fully 
cooked, ready-to-eat items, while others stated that exemptions for 
these items would be inappropriate.

[[Page 38851]]

    FSIS has modified the proposal in response to these comments. As 
explained above, FSIS is requiring E. coli testing in slaughter 
establishments where the initial and primary opportunity for fecal 
contamination occurs. FSIS is not requiring E. coli testing of 
processed products. A more limited testing requirement is possible 
because oversight of slaughter establishment verification testing for 
E. coli is not the sole means relied upon by FSIS to detect or prevent 
lack of process control. It is only one of many aspects of 
establishment operations FSIS will inspect in assessing the adequacy of 
an establishment's process controls. In particular, FSIS will 
increasingly rely on its verification that HACCP systems are working as 
intended. HACCP principles require establishments to identify CCP's, 
monitor them to see that they are in control, and take appropriate 
corrective action when monitoring detects a deviation. This is where 
control must be exercised by the establishment and where any lack of 
control will be detected in a establishment operating under a validated 
HACCP system.
    FSIS has reconsidered the proposed requirement of daily testing in 
all slaughter establishments, in part because of the unnecessary and 
disproportionate economic impact that would occur for some small 
establishments. Instead, FSIS is requiring slaughter establishments to 
test carcasses for generic E. coli at frequencies corresponding to 
production volume. In addition, slaughter establishments will have 6 
months, not just 3 months as proposed, after publication of the final 
rule to begin testing carcasses for generic E. coli. Further, very low 
volume establishments may not need to do more than one set of 13 E. 
coli tests annually, and such establishments slaughtering more than one 
species need not test both. These changes will significantly reduce the 
cost impact of mandatory testing for small establishments, while 
providing adequate and useful information to verify process control.
    In addition to requiring testing for generic E. coli by slaughter 
establishments at a frequency relative to the establishment's 
production volume, Salmonella testing will be conducted by FSIS.
    ``Minor species,'' such as sheep, goats, equines, ducks, geese, and 
guineas, are not being addressed at this time because the Agency is 
addressing first the most commonly consumed foods under its 
jurisdiction. FSIS intends to address how best to gather data on and 
develop testing requirements and performance criteria and standards for 
these other food animals at a future date.
Legal Authority for Testing Requirement
    Several commenters have questioned FSIS's legal authority for the 
proposed microbiological testing program. These comments are still 
relevant despite the differences between the proposed and final rules 
for microbiological testing.
    The major change in the final rule is that FSIS is not adopting the 
proposed Salmonella testing regimen. As proposed, results of a series 
of establishment-conducted Salmonella tests would have been used to 
accomplish two goals: to verify process control and to enforce the 
prevalence targets for pathogens in raw products. Instead, FSIS is 
promulgating separate provisions to address these two regulatory goals. 
The first provision requires that slaughter establishments test 
carcasses for E. coli so that the effectiveness of the establishment's 
sanitation and process control measures can be assessed in an 
objective, uniform manner. The second provision sets a pathogen 
reduction performance standard to bring about reductions in the 
prevalence of Salmonella on raw meat and poultry products. This 
standard will be enforced by an FSIS-conducted testing program, and 
will require establishments with prevalence of Salmonella above the 
standard to change their operations to meet that standard. Failure by 
an establishment to achieve the standard could result in Agency 
sanctions, as discussed above. This standard will also encourage 
innovation to reduce pathogens throughout the industry.
    One commenter argues that, because this regulatory strategy is 
precedent-setting, FSIS has a greater than usual burden of articulating 
the legal basis for it. This commenter notes that the testing 
regulation does not rely on a finding that the presence of the targeted 
organisms causes specific lots of product to become adulterated, as is 
the case with E. coli O157:H7 in ground beef. This commenter then 
argues that FSIS is relying upon a vague ``sanitation theory'' as its 
legal basis, and that the Agency has a greater duty to articulate its 
legal basis when new regulations impose new kinds of costs, like 
mandatory E. coli testing, or when the Agency is establishing a new 
regulatory policy.
    This commenter believes that FSIS reliance on a ``sanitation 
theory'' is legally flawed because, if the Agency is unable to tell 
establishments how to correct a failure to meet the established 
targets, it cannot legally require microorganism testing, or impose 
sanctions for failure to meet established standards.
    FSIS has ample statutory authority under the FMIA and PPIA to 
promulgate these microbiological testing provisions. The meat and 
poultry inspection statutes mandate Federal regulatory oversight of 
unusual intensity and comprehensiveness, and they provide the Secretary 
broad rulemaking authorities to implement them. The primary goal of the 
statutes is to prevent adulterated or misbranded meat and poultry 
products from entering into commerce by inspecting meat and poultry 
products and the establishments that produce them before the products 
are introduced into commerce. Such inspections are supplemented by 
compliance actions to remove adulterated or misbranded products from 
commerce and to apply appropriate sanctions against violators of the 
law. FSIS regulations under the FMIA and PPIA may be divided into two 
categories: (1) regulations prescribing the conditions under which, and 
the manner in which, mandatory inspections are conducted; and (2) 
regulations directed more broadly at preventing adulteration or 
misbranding of products, preparation of products in violation of the 
law, and sale of such products in commerce.
    These two regulatory categories are interrelated. The broader 
category is similar to regulations imposed on foods generally by the 
FDA under the Federal Food, Drug, and Cosmetic Act. However, FSIS 
authorities also require compliance with the inspection provisions of 
the acts and regulations by anyone slaughtering poultry or livestock, 
or preparing poultry products, or meat or meat food products for use as 
human food. Thus, the requirements that establishments must meet to 
obtain inspection and to have products marked ``inspected and passed'' 
comprise a unique statutory scheme which provides the Secretary with 
broad rulemaking authorities.
    From their inception, the meat and poultry inspection laws have 
recognized that sanitary conditions in establishments are critical to 
the safety and wholesomeness of the products being produced. Any 
product found to have 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'' is adulterated. 
No product will be granted inspection or marked ``inspected and 
passed'' unless the sanitary conditions and practices required by the 
Secretary are maintained.

[[Page 38852]]

    It is important to distinguish the statutorily required finding 
that a product is not adulterated from the absence of a finding that it 
is adulterated. Only products found not to be adulterated may be marked 
``inspected and passed.'' Even if the evidence does not compel an 
inspector to find that a product is adulterated, it, nonetheless, may 
be enough to prevent him from finding that it is not adulterated. This 
means that products may not be distributed for food use without the 
affirmative determination that they are not adulterated. Products as to 
which such an affirmative determination has not been made must be 
retained at the establishment pending such determination. They are 
being detained because they have not been inspected and passed, not 
because they have been found to be adulterated.
    Thus, FSIS clearly has the authority to require that establishments 
slaughtering livestock or poultry conduct and record tests for E. coli 
on carcasses to measure how well contamination is being avoided. These 
tests provide information by which establishments may evaluate and 
ensure the effectiveness of their sanitary procedures and related 
process controls in preventing product contamination during slaughter 
and dressing.
    Although E. coli testing will not be used to determine the 
disposition of inspected products, it will be an effective indicator of 
the presence of fecal contamination that is not visible and therefore 
not detectable by traditional inspection methods. It will also provide 
FSIS with information necessary to determine how best to conduct 
inspection to ensure that product is not being adulterated.
    Similarly, FSIS has clear authority to establish a Salmonella 
standard for producers of raw meat and poultry to reduce the public's 
exposure to Salmonella and associated pathogens from inspected meat and 
poultry products. The Salmonella standard, like the criteria for E. 
coli on carcasses, is based on the national baseline prevalence of the 
bacteria for the product of concern. However, unlike the E. coli 
criteria, which are, in essence, guidelines, the Salmonella standard 
must be met. Compliance will be determined by Agency testing.
    FSIS is continuing its policy of permitting raw meat and poultry 
products to be marked and labeled ``inspected and passed,'' despite the 
known or suspected presence of some pathogenic bacteria. FSIS 
recognizes that currently there is no available technology (with the 
possible exception of irradiation) to ensure that raw product bears no 
pathogenic microorganisms.
    However, there is overwhelming evidence that raw meat and poultry 
products are frequently contaminated with pathogens and expose 
consumers to avoidable and unacceptable risks of foodborne illness. 
FSIS's statutory mandate to protect consumers from adulterated product 
is not limited to actions associated with inspection. The Secretary may 
also regulate how meat and poultry products are stored and handled by 
anyone who buys, sells, freezes, stores, transports, or imports them, 
to ensure they are not misbranded or adulterated when delivered to the 
consumer.
    The new pathogen reduction standards for Salmonella are necessary 
to establish that raw product is being produced under sanitary 
conditions, has not been prepared, packed or held under insanitary 
conditions, and is not for any other reason unsound, unhealthful, 
unwholesome, or otherwise unfit for human food.
    The fact that the new performance standards and guidelines do not 
specify how the E. coli process control verification performance 
criteria or the Salmonella pathogen reduction standard must be met does 
not undercut the reasonableness or the legal basis of either testing 
program. Process control and the production of product that is not 
adulterated is the responsibility of the establishment, not the 
government. The Agency is responsible for establishing and enforcing 
reasonable standards; it intends to give the industry the maximum 
flexibility to decide how best to meet such standards. It does not 
intend to regulate or prescribe how the standards are to be met. FSIS 
will provide guidance and assistance to the industry, especially small 
businesses. But it is not legally obliged to provide technical services 
to establishments in finding the most efficient and effective way to 
operate within the E. coli criteria and to meet the Salmonella 
reduction standard.
    In summary, FSIS has concluded that the E. coli testing program and 
the Salmonella reduction standard are fully supported by the FMIA and 
PPIA.
Performance Standards for Process Control
    A related comment asserted that FSIS's proposed Salmonella standard 
was not a standard at all, but instead was merely an unenforceable 
criterion because its violation would not alone support seizure or 
condemnation of products. FSIS agrees with the principle that a 
regulatory standard should be enforceable, but does not agree that a 
regulatory ``standard'' must be limited to product-specific 
requirements, or to enforcement by seizure or condemnation of products. 
The Agency acknowledges that historically it has used the term 
``standard'' normally to refer to regulations concerning particular 
products, e.g., standards of identity regulations, but notes that 
current government-wide regulatory reform efforts stress the use of 
``performance standards'' to describe the desired focus of government 
regulations generally. FSIS intends now to issue regulations consistent 
with the notion behind ``performance standards,'' that to the extent 
possible regulations should tell regulated entities what they must 
achieve to comply with the law, while providing maximum flexibility 
regarding how to achieve the standard. Thus, FSIS agrees that one test 
of a ``standard'' might be that violation of that requirement alone 
supports some sort of regulatory sanction, but does not agree that 
``standards'' should be limited to product-specific regulations or to 
enforcement actions directed at specific products. The FMIA and PPIA do 
not limit the Agency to product-specific regulations and enforcement 
activities, and for reasons fully discussed earlier in this preamble, 
the Agency has concluded that standards directed at processes are, at 
this time, the only practical way in which to effectively address the 
hazard presented by microbiological pathogens on raw meat and poultry 
products.
Basis for Target Levels
    Some commenters questioned the validity of microbial target levels 
established by FSIS, while others supported FSIS national baseline 
studies as an effective way to evaluate industry performance. After 
careful review, the Agency considers it reasonable and appropriate to 
use the distribution of results observed for each animal species in the 
FSIS baseline surveys as the basis for both the E. coli criteria and 
the pathogen reduction performance standard for Salmonella. These are 
currently the best available data on the nationwide prevalence and 
level of microbial contamination of raw meat and poultry products. The 
data demonstrate that the E. coli process control verification criteria 
and the Salmonella pathogen reduction standard are being achieved by 
many establishments with today's technology and therefore are 
achievable by all establishments.
    FSIS Nationwide Microbiological Baseline Data Collection Programs 
and its Nationwide Microbiological Surveys provide similar data, but 
the

[[Page 38853]]

``Programs'' generally involve more extensive sampling over a longer 
period, generally 12 months, than the ``Surveys'', which are generally 
limited to 6 months of data collection. They both have provided data 
for an ongoing microbial profile of carcasses and other raw meat and 
poultry products for selected microorganisms or groups of 
microorganisms of various degrees of public health concern of value as 
indicators of general hygiene or process control.
    As explained above, FSIS plans to revise the performance criteria 
and standards as more current baseline data become available from 
future baseline surveys, through establishment E. coli testing, through 
FSIS Salmonella testing, or from other FSIS testing that may be 
appropriate for establishing criteria and standards.
    Although the majority of commenters focused on the issues mentioned 
above, a number of others addressed various aspects of the proposed 
rule such as microbial testing methodology, the concept of end product 
testing, the role of FSIS personnel in test verification, enforcement 
actions for non-compliance, and laboratory qualifications.
Methodology for Meeting Targets
    Some commenters raised objections to use of the ``moving sum'' 
statistical procedure for determining when microbial testing results 
are within the process control. Moving sum procedures are recognized in 
the field of statistical quality control. The American National 
Standard ``Guide for Quality Control Charts'' <SUP>11 identifies two 
principal uses of such charts: assisting judgment as to whether a state 
of control exists and attaining and maintaining control. In order to 
judge whether a state of control exists, operators must analyze 
``collectively an accumulation of quality data.'' In the proposed 
regulation FSIS took this view of the purpose of the moving sum 
procedure: establishments would need to verify that a state of control 
exists with respect to the interim target set by the Agency. FSIS did 
not claim, however, that the procedure would be useful for the second 
purpose, attaining and maintaining control. That requires more timely 
and probably more intense monitoring of process parameters at CCP's.
---------------------------------------------------------------------------

     <SUP>11 American National Standard ANSI Z1.1-1985. ``Guide for 
Quality Control Charts.'' American Society for Quality Control. 
Milwaukee, WI.
---------------------------------------------------------------------------

    The proposed approach to use testing to measure process control was 
designed to inform establishments how they are currently operating with 
respect to the relevant target, and to help them track progress toward 
meeting that target. A simple plot of the moving sum chart would give 
them sufficient feedback for this purpose.
    Some commenters recommended that the moving window verification 
program should use a 90% probability criteria, rather than 80%, to 
reduce the possibility of the testing procedure erroneously identifying 
an establishment as not meeting the pathogen target. The Agency notes 
that the moving sum procedure was designed to measure effectiveness of 
process control with respect to an interim performance standard (called 
a target in the proposal) based on current industry performance (as 
determined by a baseline study). This measure was intended to be the 
first step in holding establishments accountable for meeting acceptable 
levels of performance. As such, the Agency wanted to be able to readily 
identify establishments operating above the target and wanted to 
provide an incentive for establishments to produce at levels better 
than (below) the target. Giving establishments producing at the target 
only an 80% chance of passing was expected to promote this. Giving 
establishments producing at the target a higher chance of passing 
(e.g., 95%) would reduce both the incentive to do better and the 
ability to detect establishments above the target.
Sample Size
    Others specifically addressed the proposed sample size, 
recommending that the same number of samples be used for all species. 
Not all species have the same risks of failure, in part because of the 
varied incidence of pathogens, as was determined in FSIS's baseline 
surveys. The proposed sampling rate was the same for all 
establishments, one per day. Thus the sampling was the same for all 
establishments, only the rules for interpreting results were different. 
The number of results included in the window differed by product class 
because the target percents positive differed by product class. It was 
necessary to employ different-sized windows to maintain a fixed 
probability of passing (80%) at the target for all product classes 
while choosing as short a window as possible and allowing at least one 
positive in the window.
Testing Methodology
    Other commenters asked for clarification on testing methodology. 
Some remarked that using a sponge or swab method to sample carcasses is 
preferable to the proposed excision method because the proposed method 
is time consuming, cumbersome, and expensive, and it may mutilate and 
contaminate the carcass. The Agency agrees and has elected to use non-
destructive sampling methods.
    Others asked for clarification of enforcement actions that would 
result from an establishment not meeting its microbial targets. How the 
rule will be enforced is addressed above.
Role of Inspectors
    Still others asked about the role of inspection personnel in 
verification testing and expressed concern about the amount and type of 
training inspection personnel would receive to analyze test results.
    The final rule makes slaughter process control verification testing 
(E. coli) the responsibility of establishments slaughtering livestock 
or poultry, although FSIS inspectors may also collect samples for E. 
coli testing as needed to carry out their oversight responsibilities. 
FSIS personnel sampling carcasses for Salmonella to ensure that 
establishments are meeting the pathogen reduction performance standard 
will send the samples to an Agency laboratory for analysis. FSIS 
personnel have been involved in collection of samples for FSIS's 
baseline surveys, and have been trained and are highly qualified to 
collect samples for this regulatory program. Inspectors will work with 
other program officials, including scientifically trained experts, in 
analyzing test results and making appropriate regulatory decisions. 
Inspectors will receive training to prepare them for their role in this 
process.
Laboratories
    Some commenters asked for clarification regarding qualifications 
for in-house and outside laboratories. They stated that laboratories 
should be required to use standardized techniques for analyzing test 
results.
    The microbiological test method used by the establishments must be 
AOAC validated techniques, or other methods validated by a scientific 
body in collaborated trials against the three tube most probable number 
(MPN) method and agreeing with the 95 percent upper and lower 
confidence interval, as discussed in the E. coli Methods Section. 
Establishments are responsible for the accuracy of the tests of their 
samples. If the samples are not analyzed by the establishment, the 
establishment, perhaps in concert with a trade association, should 
ensure that the laboratory it chooses is reputable and

[[Page 38854]]

adheres to a Quality Control/Quality Assurance Program.
Alternative Sampling Under HACCP
    Other commenters stated that the proposed microbial testing system 
does not reward very clean establishments by granting reasonable 
reductions in testing when significant periods are pathogen free. They 
recommended that once a facility has implemented its HACCP program, the 
required frequency for mandatory microbial testing should be reduced or 
eliminated altogether.
    In this final rule, a slaughter establishment successfully 
operating under a validated HACCP plan may reduce the specified 
sampling frequency as long as the alternative sampling plan is an 
integral part of the establishment's verification procedures for its 
HACCP system. FSIS does, however, reserve the right to determine that 
the alternative frequency is inadequate to verify the effectiveness of 
the establishment's process controls. In that case, FSIS would notify 
the establishment in writing of its finding, advise that the frequency 
specified in the regulation must be maintained, and specify any 
conditions an acceptable alternative frequency would have to meet to be 
found acceptable to the Agency.
Relationship to HACCP
    Finally, some commenters stated that the proposed end-product 
testing is inconsistent with HACCP principles and that establishments 
should decide for themselves through hazard analysis whether testing is 
needed and at what frequency. Others objected to the concept of end-
product testing because it only measures effectiveness over a small 
percentage of a production lot and has limited value in measuring the 
overall success of a HACCP plan. Still others concluded that placing an 
emphasis on end-product testing gives consumers a false sense of 
confidence about the safety of meat and poultry products. A few 
commenters were concerned about product liability due to product 
recalls stemming from test results.
    The objective of the generic E. coli testing is to verify that 
process control has been maintained by the establishment throughout the 
slaughter and dressing process and that resultant carcasses are 
produced hygienically. If processes are under control for E. coli, the 
potential presence of enteric pathogens will be reduced. End-product 
verification testing of this kind is a well recognized component of 
HACCP-based process control.<SUP>12 The goal of FSIS's Salmonella 
testing program is to verify that pathogen reduction performance meets 
current standards in each establishment and thereby effect a nationwide 
reduction in the incidence of that organism and other enteric pathogens 
on raw meat and poultry products. The end of production is the only 
point that reflects all steps in the production process and, 
ultimately, all elements of the HACCP system. The seventh HACCP 
principle is verification that the HACCP system is working; one cannot 
verify that HACCP is working in slaughter establishments (controlling 
fecal contamination/pathogens) without some end-product testing, so 
end-product testing is not inconsistent with HACCP principles. The two 
different kinds of testing programs: (1) E. coli testing by 
establishments to verify control of fecal contamination; and (2) 
Salmonella testing by FSIS to hold establishments accountable for 
meeting pathogen performance standards, are both forms of end-product 
testing that FSIS considers consistent with HACCP.
---------------------------------------------------------------------------

    \12\ National Advisory Committee on Microbiological Criteria for 
Foods. 1994. ``Hazard Analysis and Critical Control Point Systems.'' 
FSIS, USDA.
---------------------------------------------------------------------------

    End-product testing as part of an overall system of HACCP-based 
process control and performance standards should not give consumers a 
false sense of confidence about the safety of meat and poultry 
products. FSIS recognizes that limited end-product testing alone 
provides little assurance of safety, but, as part of a process control 
system, appropriate end-product testing brings rigor and accountability 
to the system and should appropriately increase consumer confidence in 
the safety of products. By requiring HACCP, FSIS is in fact moving away 
from sole reliance on end-product assessments for lot acceptance, an 
approach that is the opposite of the HACCP system approach to food 
safety. FSIS recognizes that producing safe food requires preventing 
hazards throughout the process rather than relying solely on end-
product testing to ensure safety. Establishments' liability to civil 
lawsuits should not be adversely affected by this rule precisely 
because it is an establishment's process, not individual lots of 
product, that is being assessed, for inspection purposes, on the basis 
of this testing.

V. Other Issues and Initiatives

Antimicrobial Treatments

    FSIS proposed that all slaughter establishments apply at least one 
antimicrobial treatment or other approved intervention to livestock and 
poultry carcasses prior to the chilling or cooling operation. Proposed 
treatment methods included chlorine compounds, hot water, and any 
antimicrobial compound previously approved by FSIS and listed in the 
meat or poultry regulations. Product prepared for export to countries 
that restrict or prohibit the use of antimicrobial treatments would 
have been exempted from this requirement upon application to the 
Administrator.
    While most commenters generally agreed that antimicrobial 
treatments could play an important role in reducing contamination with 
pathogenic microorganisms in slaughter establishments, many commenters 
opposed mandating such treatments. The commenters argued that mandating 
the use of antimicrobial treatments in slaughter operations would not 
be consistent with the HACCP philosophy and the overall shift by FSIS 
to greater reliance on performance standards.
    FSIS agrees with these commenters and has decided not to mandate 
the use of antimicrobial treatments in slaughter establishments. FSIS 
continues to believe that slaughter establishments will find that these 
treatments can play a useful role in reducing pathogens and improving 
the safety of meat and poultry products. Rather than mandating specific 
antimicrobial treatments, FSIS will rely on other requirements in this 
final rule to ensure that slaughter establishments are achieving an 
acceptable level of performance in controlling and reducing harmful 
bacteria on raw product.
    The principle of using antimicrobial treatments as an intervention 
to control pathogens on meat and poultry carcasses was strongly 
endorsed by most commenters. However, few agreed that the treatments 
should be mandatory. A majority of commenters recommended that 
antimicrobial treatments be voluntary interventions. Establishments 
would decide if antimicrobial interventions were needed to control 
specific hazards at one or more critical control points in the 
slaughter process.
    Similarly, a number of commenters tied antimicrobial treatments to 
microbial testing. They argued that carcass treatments should not be 
required in establishments that consistently meet or exceed performance 
standards for microbial contamination.
    Commenters said FSIS should focus its regulatory efforts on 
measurable, attainable goals and not on prescriptive requirements for 
particular processing steps. Several commenters emphasized the need for 
``whole system'' interventions instead of single

[[Page 38855]]

techniques such as antimicrobial treatments. They said these 
interventions work best when they are tailored to species and product 
hazards, individual establishment configurations, and processing 
methods. Furthermore, some commenters cited a danger that 
establishments and inspection personnel would focus on the treatment 
function itself instead of broader food safety goals.
    FSIS generally agrees with these comments. FSIS has concluded that 
its food safety goals can be achieved more effectively and more 
efficiently by requiring HACCP-based process control combined with 
appropriate performance criteria and standards than by mandating 
specific interventions, such as antimicrobial treatments. New 
technological interventions will play a significant role in reducing 
the risk of foodborne illness and should be adopted as part of an 
overall system of HACCP-based process control. FSIS expects that such 
treatments may be used by establishments to meet the process control 
performance criteria and pathogen reduction performance standards FSIS 
is adopting in this final rule.
    A few commenters opposed mandating antimicrobial treatments because 
they believed their use would allow for correction of sloppy carcass 
dressing procedures. These commenters argued that antimicrobial 
treatments, whether mandatory or voluntary, emphasize post-
contamination clean-up rather than prevention.
    FSIS also received many comments which addressed the four proposed 
antimicrobial treatment methods. Many commenters stated that FSIS 
should not restrict establishments to these particular antimicrobial 
interventions.
    A variety of commenters addressed technology issues concerning the 
proposed treatment methods themselves. Many said that too few studies 
have been conducted to show which interventions are most effective and 
efficient for specific pathogens associated with particular species in 
individual slaughter establishment configurations. Some argued that the 
studies FSIS cited in its proposal were too narrow and did not 
adequately demonstrate effectiveness. They said additional studies were 
needed to determine the practicality, efficacy, and expense of various 
antimicrobial treatments in commercial settings. In addition, some 
commenters were concerned that insufficient research was available on 
whether the elimination of competitive micro flora would allow 
uninhibited growth of pathogenic bacteria.
    Individual antimicrobial techniques were also criticized. For 
example, hot water sprays were said to pose dangers to establishment 
personnel applying the treatments at temperatures necessary for 
effectiveness. Hot water sprays raise carcass temperatures with 
consequent melting of surface fat in some species, contribute to 
quality defects such as change in product color and partial cooking, 
and result in higher energy costs. Commenters recognized, however, that 
hot water was the only currently available nonchemical intervention 
that could be implemented at comparatively low cost. Other commenters 
criticized lactic, acetic, and citric acid solution sprays because they 
have low effectiveness as a treatment against E. coli O157:H7. The 
possible carcinogenic effects of chlorine were also mentioned, as were 
concerns about water reuse and possible environmental effects from 
spray effluents.
    Commenters also suggested a variety of alternative antimicrobial 
interventions that could be used by establishments. These interventions 
included irradiation and radiation-emitting electronic devices such as 
x-rays and linear accelerators; high-energy ultraviolet light; pulsed 
light, sonic, infrasonic, and ultrasonic emitters; chemicals such as 
copper sulfate in the pentahydrate form, chlorine dioxide, and hydrogen 
peroxide; procedures such as pre-evisceration washes, water curtains, 
counter current or counter flow scalders, the Peroxi bicarb process, 
automatic warm fresh water rinses, ozonated water, steam 
pasteurization, steam vacuuming, hot wax dipping, and singeing.
    A number of commenters also suggested that FSIS establish protocols 
to evaluate various forms of antimicrobial procedures and treatments. 
FSIS could then publish a regularly updated list of acceptable 
treatments and provide guidelines for their use in a commercial 
setting. It was argued that this process would give establishments the 
flexibility to implement any interventions they deem necessary. Others 
said FSIS should set up a predetermined protocol for antimicrobial 
agents or an expedited review process for new technologies.
    FSIS agrees that issues of effectiveness, product and worker 
safety, product quality, interference with inspection, and 
environmental impact can be raised about most food safety 
interventions, including antimicrobial treatments. Therefore, to 
facilitate industry development of new technologies, FSIS has 
established a process that will facilitate this development.
    On May 25, 1995, FSIS published a notice in the Federal Register 
(60 FR 27714) that presented guidelines for preparing and submitting 
experimental protocols to FSIS for use by establishments wishing to 
conduct trials of new technologies and procedures. In that notice, FSIS 
confirmed its long-standing commitment to foster innovative 
technologies and procedures that more effectively protect meat and 
poultry products from microbiological and other hazards. Specifically, 
FSIS encouraged the development of efficacious, practical and 
manageable technologies and procedures by establishments.
    FSIS also published guidelines (FSIS Directive 10,700.1) for 
establishments to use for submitting written proposals and protocols to 
FSIS for approval to conduct experiments. Agency approval is required 
in cases where the intended technology, procedure or process may affect 
(1) product safety or lead to economic adulteration, (2) worker safety, 
(3) environmental safety, or (4) inspection procedures.
    Similarly, FSIS published a proposed rule in the Federal Register 
(60 FR 67459; December 29, 1995) that will facilitate the review and 
approval of substances intended for use in or on meat and poultry 
products. Under the proposed procedures, FSIS would no longer issue its 
own regulations listing substances it finds suitable for use in meat 
and poultry products. Instead, FDA's regulations would specify whether 
a substance approved for use in foods under the Federal Food, Drug, and 
Cosmetic Act may be used in or on meat or poultry products.
    Many commenters stated that antimicrobial interventions should be 
permitted at any stage in the slaughter process: live animal, pre-hide 
removal, pre- or post-carcass wash, pre- or post-chill, or just prior 
to fabrication.
    Some commenters argued that the proposed treatments would seriously 
compromise the Kosher ritual salting process, while others said the 
interventions would conflict with Confucian and Buddhist-style poultry 
prepared for religious rites.
    A number of commenters questioned the relationship between FSIS's 
policy on zero tolerance for fecal contamination and its antimicrobial 
treatment proposal. In particular, they were concerned about where in 
the process zero tolerance would be measured.
    Finally, several commenters requested a practical definition of 
``feces'' as a means to resolve disagreements between inspectors and 
establishment personnel about trimming contamination.


[[Continued on page 38856]]