Laboratory Accreditation for Analyses of Foods

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Parts 1, 11, 16, and 129

[Docket No. FDA-2019-N-3325]

RIN 0910-AH31

Laboratory Accreditation for Analyses of Foods

AGENCY:

Food and Drug Administration, HHS.

ACTION:

Final rule.

SUMMARY:

The Food and Drug Administration (FDA, the Agency, or we) is amending its regulations to establish a program for the testing of food in certain circumstances by accredited laboratories, as required under the Federal Food, Drug, and Cosmetic Act (FD&C Act). Establishing this program will help FDA improve the safety of the U.S. food supply and protect U.S. consumers by helping ensure that certain food testing of importance to public health is conducted subject to appropriate oversight and in accordance with appropriate model standards to produce reliable and valid test results.

DATES:

This rule is effective February 1, 2022. The incorporation by reference of certain publications listed in the rule is approved by the Director of the Federal Register as of February 1, 2022.

ADDRESSES:

For access to the docket to read background documents or comments received, go to

https://www.regulations.gov

and insert the docket number found in brackets in the heading of this final rule into the “Search” box and follow the prompts, and/or go to the Dockets Management Staff, 5630 Fishers Lane, Rm. 1061, Rockville, MD 20852, 240-402-7500.

FOR FURTHER INFORMATION CONTACT:

With regard to the final rule:

Stacie Hammack, Chemist, Food and Feed Laboratory Operations, Office of Regulatory Affairs, Food and Drug Administration, 60 8th Street NE, Atlanta, GA 30309, 301-796-5817;

Stacie.Hammack@fda.hhs.gov.

With regard to the information collection:

Domini Bean, Office of Operations, Food and Drug Administration, Three White Flint North 10A-12M, 11601 Landsdown Street, North Bethesda, MD 20852, 301-796-5733,

PRAStaff@fda.hhs.gov.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive Summary

A. Purpose and Coverage of the Final Rule

B. Summary of the Major Provisions of the Final Rule

C. Legal Authority

D. Costs and Benefits

II. Table of Abbreviations/Commonly Used Acronyms in This Document

III. Background

A. Need for the Regulation

B. Summary of Comments to the Proposed Rule

C. General Overview of Final Rule

D. Incorporation by Reference

IV. Legal Authority

V. Comments on the Proposed Rule and FDA's Response

A. Introduction

B. General Comments

C. Comments Regarding General Provisions

D. Comments Regarding General Requirements

E. Comments Regarding FDA Recognition of Accreditation Bodies

F. Comments Regarding Requirements for Recognized Accreditation Bodies

G. Comments Regarding FDA Oversight of Recognized Accreditation Bodies

H. Comments Regarding LAAF-Accreditation of Laboratories

I. Comments Regarding Requirements for LAAF-Accredited Laboratories

J. Comments Regarding FDA Oversight of LAAF-Accredited Laboratories

K. Comments Regarding Requesting FDA Reconsideration or Regulatory Hearings of FDA Decisions Under This Subpart

L. Comments Regarding Electronic Records and Public Disclosure Requirements

M. Comments on Conforming and Technical Amendments and FDA Response

VI. Effective Date

VII. Economic Analysis of Impacts

VIII. Analysis of Environmental Impact

IX. Paperwork Reduction Act of 1995

X. Federalism

XI. Consultation and Coordination With Indian Tribal Governments

XII. References

I. Executive Summary

A. Purpose and Coverage of the Final Rule

This rule is part of FDA's implementation of the FDA Food Safety Modernization Act (FSMA) (Pub. L. 111-353), through which the Agency intends to better protect public health by, among other things, adopting a modern, preventive, and risk-based approach to food safety regulation. In this document we establish the Laboratory Accreditation for Analyses of Foods (LAAF) program as required by FSMA section 202(a), which added section 422 to the FD&C Act (21 U.S.C. 350k). Under the LAAF program, FDA will recognize accreditation bodies that will accredit laboratories to the standards established in this final rule. Laboratories accredited to the LAAF standard (“LAAF-accredited laboratories”) are authorized to conduct certain food testing as described in this rule.

The program structure is portrayed in the following diagram:

1

1

For a description of how the program structure diagram has been revised, see (Response 11).

ER03DE21.000

You are subject to this rule if you are an accreditation body seeking recognition to accredit laboratories under this subpart, a recognized accreditation body, a laboratory seeking accreditation to conduct food testing under this subpart, or an accredited laboratory conducting food testing under this subpart. This rule also applies to owners or consignees that must have certain food testing conducted by a laboratory accredited under this subpart. Although participation in this program is voluntary for accreditation bodies and laboratories, only recognized accreditation bodies may accredit laboratories to conduct the testing of food covered under this subpart.

This program for the testing of food by accredited laboratories establishes oversight, uniformity, and standards necessary to help ensure that the results of certain food testing of importance to public health are reliable and accurate. Establishing this program will substantially improve our capability to protect U.S. consumers from unsafe food.

B. Summary of the Major Provisions of the Final Rule

This rule contains model standards that laboratories must meet in order to participate and conduct certain food testing covered by this subpart. The rule will establish a publicly available registry listing accreditation bodies and laboratories that have been recognized or accredited under this program. Results of food testing conducted by laboratories under the program must be sent directly to FDA. Laboratories accredited under this program (“LAAF-accredited laboratories”) are required to submit to FDA analytical reports as specified in this final rule.

This rule contains eligibility requirements for accreditation bodies to qualify for FDA recognition and requirements that accreditation bodies must meet once recognized, such as requirements related to assessing and overseeing laboratories, conflicts of interest, reporting, and records. The rule contains eligibility requirements for laboratories to qualify for LAAF-accreditation by a recognized accreditation body and requirements that laboratories must meet once LAAF-accredited, such as requirements related to conflicts of interest, analysis, reporting, and records. These requirements will help ensure the effectiveness of the recognized accreditation bodies and LAAF-accredited laboratories under this program. This rule contains procedures we will follow to recognize accreditation bodies under this program and procedures for accreditation bodies to follow to LAAF-accredit laboratories under this program. This rule contains regulatory procedures and requirements relating to our oversight of recognized accreditation bodies and LAAF-accredited laboratories.

This rule applies when food testing is conducted in certain circumstances. “Food testing” and “testing of food” include the analysis of human or animal food, as well as testing of the food growing or manufacturing environment (

i.e.,

“environmental testing”).

C. Legal Authority

Section 422(a)(1)(A) the FD&C Act, which was added by section 202(a) of FSMA, directs us to establish a program for the testing of food by accredited laboratories. Therefore, section 422 of the FD&C Act provides FDA with authority for these final regulations, which outline requirements for participants in the program for the testing of food by LAAF-accredited laboratories. FDA also derives authority for these requirements from section 701(a) of the FD&C Act (21 U.S.C. 371(a)), which authorizes FDA to issue regulations for the efficient enforcement of the FD&C Act.

D. Costs and Benefits

The rule will require that testing of food in certain circumstances be performed by a laboratory that is LAAF-accredited by a recognized accreditation body, and for the testing results to be submitted directly to us. The costs of the rule primarily will be incurred by participating accreditation bodies, participating laboratories, shell egg producers, sprouts producers, bottled drinking water manufacturers, owners and consignees of certain import-related food, and FDA. Rarely, certain firms will have participating laboratories conduct tests for other reasons including as part of a corrective action plan after an order suspending registration, as part of evidence for a hearing prior to issuance of a mandatory recall order, as part of evidence for an appeal of an administrative detention order, and as required under a directed food laboratory order (formerly, a food testing order). We will incur costs to, among other things, establish and maintain the program for recognizing accreditation bodies that apply to participate in our program, evaluate participating accreditation bodies and review the performance of participating laboratories, and review associated documents and reports. The present value of the costs of the rule ranges from $38 million to $66 million when discounted by 7 percent over 10 years and from $43 million to $77 million when discounted by 3 percent over 10 years. Annualized costs over 10 years range from $5.8 million to $9.6 million when discounted by 7 percent, and from $5.9 million to $9.7 million when discounted by 3 percent.

The rule will generate some quantified and unquantified benefits. Quantified benefits include a reduction in the number of foodborne illnesses from fewer false negative test results for import-related food covered under the rule and for shell eggs, sprouts, and bottled drinking water testing covered under the rule. We anticipate cost savings from the clarification of the process for compiling, submitting, and reviewing analytical reports for import-related food covered under this rule, including reduced reporting burden. There would be less revenue lost from fewer false positive test results for import-related food covered under the rule and for tests of shell eggs, sprouts, and bottled drinking water testing covered under the rule. The present value of the benefits of the rule ranges from $46 million to $88 million when discounted at 7 percent over 10 years and ranges from $56 million to $106 million when discounted at 3 percent over 10 years. Annualized benefits over 10 years range from $6.6 million to $12.5 million when discounted by both 7 and 3 percent.

II. Table of Abbreviations/Commonly Used Acronyms in This Document

Abbreviation/acronym

What it means

AAVLD

American Association of Veterinary Laboratory Diagnosticians.

ANSI

American National Standards Institute.

AOAC

AOAC International.

APA

Administrative Procedure Act.

CFR

Code of Federal Regulations.

CPSC

Consumer Product Safety Commission.

CVM

Center for Veterinary Medicine.

DWPE

Detention Without Physical Examination.

EO

Executive Order.

E. coli

Escherichia coli.

FDA

United States Food and Drug Administration.

FD&C Act

Federal Food, Drug, and Cosmetic Act.

FOIA

Freedom of Information Act.

FR

Federal Register.

FRIA

Final Regulatory Impact Analysis.

FSMA

FDA Food Safety Modernization Act.

FSVP

Foreign Supplier Verification Program.

HACCP

Hazard Analysis and Critical Control Point.

IBR

Incorporation by Reference.

IEC

International Electrotechnical Commission.

ILAC

International Laboratory Accreditation Cooperation.

IOM

Investigations Operations Manual.

ISO

International Organization for Standardization.

LAAF

Laboratory Accreditation for Analyses of Foods.

MRA

Mutual Recognition Arrangement.

NIST

National Institute of Standards and Technology.

NRTE

Not Ready to Eat.

NTTAA

National Technology Transfer and Advancement Act of 1995.

OMB

Office of Management and Budget.

ORA

Office of Regulatory Affairs.

PLAP

Private Laboratory Analytical Package.

PRA

Paperwork Reduction Act of 1995.

PRIA

Preliminary Regulatory Impact Analysis.

SAHCODHA

Serious Adverse Health Consequences or Death to Humans or Animals.

U.S.C.

United States Code.

Vet-LIRN

Veterinary Laboratory Investigation and Response Network.

WTO

World Trade Organization.

III. Background

A. Need for the Regulation

FSMA is transforming the nation's food safety system by shifting the focus from responding to foodborne illness to preventing it. Congress enacted FSMA in response to dramatic changes in the global food system and in our understanding of foodborne illness and its consequences, including the realization that preventable foodborne illness is both a significant public health problem and a threat to the economic well-being of the food system. FSMA provides us with new enforcement authorities designed to achieve higher rates of compliance with risk-based, prevention-oriented safety standards and to better respond to and contain problems when they do occur. In addition, FSMA gives us important new tools to better ensure the safety of imported foods and encourages partnerships with State, local, tribal, and territorial authorities. In implementing FSMA, we prioritized the development of seven foundational rules that provide the framework for risk-based preventive controls and enhance our ability to oversee their implementation by industry for both domestic and imported food. We have finalized these foundational rules and begun their implementation while also developing additional programs required by FSMA, including this program for food testing by accredited laboratories.

FSMA, in establishing section 422 of the FD&C Act, underscores that food testing can play a role in detecting and responding to food safety problems. Section 422(b)(1) of the FD&C Act requires that food be tested by laboratories accredited to the standards we are establishing in this final rule in four circumstances:

• In response to a specific testing requirement under the FD&C Act or implementing regulations, when applied to address an identified or suspected food safety problem;

• As required by the Secretary of Health and Human Services (Secretary), as the Secretary deems appropriate, to address an identified or suspected food safety problem;

• In support of admission of an article of food under section 801(a) of the FD&C Act (21 U.S.C. 381(a)); and

• Under an import alert that requires successful consecutive tests.

With one exception, section 422(b)(2) of the FD&C Act requires the results of food testing conducted under section 422(b)(1) to be sent directly to FDA, thereby allowing FDA to review the test results.

Direct receipt of food testing results in these circumstances is of particular importance to the Agency and to public health. This rule applies to food testing conducted under specific testing requirements in the FD&C Act and implementing regulations that “address an identified or suspected food safety problem”, and in directed food laboratory orders that we will issue “as required by the Secretary, as the Secretary deems appropriate, to address an identified or suspected food safety problem.” Further, owners and consignees often engage private laboratories to test their food products and submit the results of the testing, along with associated analysis and data, to us to show that the imported food complies with the FD&C Act. If we determine that the food testing results are valid and that they demonstrate the detained food product does not violate the FD&C Act, we will release the food from detention and allow it to proceed into the United States. We use the detention without physical examination (DWPE) procedure when there exists a history of the importation of violative products, or products that may appear violative, or when other information indicates that future entries may appear violative. Import alerts inform FDA field staff and the public that we have enough evidence to allow for DWPE of products that appear to be in violation of FDA laws and regulations. Concerns periodically have arisen regarding importers' manipulation or substitution of the samples a private laboratory tests, and practices such as “testing into compliance,” in which multiple samples from a shipment are tested, but only those results that would allow the shipment to enter the United States are submitted to us. See,

e.g.,

“The Safety of Food Imports: Fraud & Deception in the Food Import Process; Hearings Before the Senate Committee on Governmental Affairs, Permanent Subcommittee on Investigations,” September 10, 1998 (statement of “Former Customs Broker”) (Ref. 1, pages 26-34 and 137-140).

B. Summary of Comments to the Proposed Rule

We published a proposed rule for “Laboratory Accreditation for Analyses of Foods” (the proposed rule) in the

Federal Register

on November 4, 2019 (84 FR 59452). The comment period was extended twice (85 FR 11893 (February 28, 2020); 85 FR 19114 (April 6, 2020)). Upon close of the comment period on July 6, 2020, we had received approximately 70 comment submissions that covered almost every aspect of the proposed rule.

C. General Overview of the Final Rule

We have made changes in the final rule in response to public comments; these changes are discussed in greater detail in section V below. Additionally, we have revised the final rule to improve clarity and readability. We also have reorganized the final rule as described in the following table.

Table 1—Summary of Section Numbering Changes in the Final Rule

Final rule

General provisions

Proposed rule

General provisions

§ 1.1101 What documents are incorporated by reference in this subpart?

N/A.

§ 1.1102 What definitions apply to this subpart?

§ 1.1102 What definitions apply to this subpart?

§ 1.1103 Who is subject to this subpart?

§ 1.1103 Who is subject to this subpart?

General Requirements

General Requirements of this Subpart

§ 1.1107 When must food testing be conducted under this subpart?

§ 1.1107 Under what circumstances must food testing be conducted under this subpart by an accredited laboratory?

§ 1.1108 When and how will FDA issue a directed food laboratory order?

§ 1.1108 When and how will FDA issue a food testing order?

§ 1.1109 How will FDA make information about recognized accreditation bodies and LAAF-accredited laboratories available to the public?

§ 1.1109 How will FDA make information about recognized accreditation bodies and accredited laboratories available to the public?

§ 1.1110 What are the general requirements for submitting information to FDA under this subpart?

N/A.

FDA Recognition of Accreditation Bodies

Recognition of Accreditation Bodies

§ 1.1113 What are the eligibility requirements for a recognized accreditation body?

§ 1.1113 What requirements must an accreditation body meet to be recognized by FDA?

§ 1.1118 What are the general requirements for recognized accreditation bodies to remain recognized?

§ 1.1114 How does an accreditation body apply to FDA for recognition or renewal of recognition?

§ 1.1128 How does an accreditation body apply to FDA for recognition or renewal of recognition?

§ 1.1115 How will FDA evaluate applications for recognition and renewal of recognition?

§ 1.1129 How will FDA review applications for recognition and applications for renewal of recognition?

§ 1.1116 What must a recognized accreditation body do to voluntarily relinquish or not renew its recognition?

§ 1.1132 What must a recognized accreditation body do if it wants to voluntarily relinquish its recognition or does not want to renew its recognition?

§ 1.1117 How may an accreditation body request reinstatement of recognition?

§ 1.1133 How does an accreditation body request reinstatement of recognition?

Requirements for Recognized Accreditation Bodies

Requirements for Recognized Accreditation Bodies

N/A—(contents combined with § 1.1113)

§ 1.1118 What are the general requirements for recognized accreditation bodies to remain recognized?

§ 1.1119 What are the conflict of interest requirements for a recognized accreditation body?

§ 1.1119 What requirements apply to how a recognized accreditation body must protect against conflicts of interests?

§ 1.1120 How must a recognized accreditation body assess laboratories seeking LAAF-accreditation and oversee LAAF-accredited laboratories?

§ 1.1120 How must a recognized accreditation body evaluate laboratories seeking accreditation and oversee the performance of laboratories it accredits?

§ 1.1121 When must a recognized accreditation body require corrective action, suspend a LAAF-accredited laboratory, reduce the scope of or withdraw the LAAF-accreditation of a laboratory?

§ 1.1121 What appeal procedures must a recognized accreditation body provide for appeals of decisions to not grant accreditation?

§ 1.1122(h) Appeals procedures.

§ 1.1122 What procedures must a recognized accreditation body provide for appeals of decisions to suspend, reduce the scope of, withdraw, or deny LAAF-accreditation?

§ 1.1122 When must a recognized accreditation body withdraw or reduce the scope of the accreditation of a laboratory, and when may a recognized accreditation body put an accredited laboratory on probation?

§ 1.1123 What reports, notifications, and documentation must a recognized accreditation body submit to FDA?

§ 1.1123 What reports and notifications must a recognized accreditation body submit to FDA?

§ 1.1124 What are the records requirements for a recognized accreditation body?

§ 1.1124 What records requirements must a recognized accreditation body meet?

§ 1.1125 What are the internal audit requirements for a recognized accreditation body?

§ 1.1125 What internal audit requirements must a recognized accreditation body meet?

FDA Oversight of Recognized Accreditation Bodies

Procedures for Recognition of Accreditation Bodies

§ 1.1130 How will FDA oversee recognized accreditation bodies?

§ 1.1130 How will FDA oversee recognized accreditation bodies?

§ 1.1131 When will FDA require corrective action, put a recognized accreditation body on probation, or revoke the recognition of an accreditation body?

§ 1.1131 When will FDA revoke the recognition of an accreditation body or put a recognized accreditation body on probation?

LAAF-Accreditation of Laboratories

Accreditation of Laboratories

§ 1.1138 What are the eligibility requirements for a LAAF-accredited laboratory?

§ 1.1138 What requirements must a laboratory meet to become accredited by a recognized accreditation body?

§ 1.1146 What are the general requirements for accredited laboratories to remain accredited?

§ 1.1139 How does a laboratory apply for LAAF-accreditation or extend its scope of LAAF-accreditation?

§ 1.1158 How does a laboratory apply for accreditation or modification of its scope of accreditation by a recognized accreditation body?

§ 1.1140 What must a LAAF-accredited laboratory do to voluntarily relinquish its LAAF-accreditation?

§ 1.1163 What if a laboratory wants to voluntarily relinquish its accreditation?

§ 1.1141 What is the effect on a LAAF-accredited laboratory if its recognized accreditation body is no longer recognized by FDA?

§ 1.1164 What is the effect on accredited laboratories if their accreditation body voluntarily or involuntarily loses its recognition?

§ 1.1142 How does a laboratory request reinstatement of LAAF-accreditation?

§ 1.1165 How does a laboratory request reinstatement of accreditation?

Requirements for LAAF-Accredited Laboratories

Requirements for Accredited Laboratories

Content added to § 1.1138

§ 1.1146 What are the general requirements for accredited laboratories to remain accredited?

§ 1.1147 What are the impartiality and conflict of interest requirements for a LAAF-accredited laboratory?

§ 1.1147 What impartiality and conflict of interest requirements must accredited laboratories meet?

Content moved to § 1.1138

§ 1.1148 What quality assurance requirements must accredited laboratories meet?

§ 1.1149 What oversight standards apply to sampling?

§ 1.1149 What oversight standards apply to sampling?

§ 1.1150 What are the requirements for analysis of samples by a LAAF-accredited laboratory?

§ 1.1150 What requirements apply to analysis of samples by an accredited laboratory?

§ 1.1151 What requirements apply to the methods of analysis a LAAF-accredited laboratory uses to conduct food testing under this subpart?

§ 1.1151 What requirements apply to the methods of analysis an accredited laboratory uses to conduct food testing under this subpart?

§ 1.1152 What notifications, results, reports, and studies must a LAAF-accredited laboratory submit to FDA?

§ 1.1152 What notifications, results, and reports must accredited laboratories submit to FDA?

§ 1.1153 What are the requirements for submitting abridged analytical reports?

N/A.

§ 1.1154 What other records requirements must a LAAF-accredited laboratory meet?

§ 1.1153 What other records requirements must an accredited laboratory meet?

FDA Oversight of LAAF-Accredited Laboratories

Procedures for Accreditation of Laboratories

§ 1.1159 How will FDA oversee LAAF-accredited laboratories?

§ 1.1159 How will FDA oversee accredited laboratories?

§ 1.1160 How will FDA review test results and analytical reports?

§ 1.1160 How will FDA review submitted test results and analytical reports?

§ 1.1161 When will FDA require corrective action, put a LAAF-accredited laboratory on probation, or disqualify a LAAF-accredited laboratory from submitting analytical reports?

§ 1.1161 When will FDA put an accredited laboratory on probation or revoke the accreditation of a laboratory?

§ 1.1162 What are the consequences if FDA puts a LAAF-accredited laboratory on probation or disqualifies a LAAF-accredited laboratory?

§ 1.1162 What are the consequences if FDA puts an accredited laboratory on probation or revokes the accreditation of a laboratory?

Requesting FDA Reconsideration or Regulatory Hearings of FDA Decisions Under This Subpart

Requesting FDA Reconsideration, FDA Internal Review, or Regulatory Hearings of FDA Decisions Under This Subpart

§ 1.1171 How does an accreditation body request reconsideration by FDA of a decision to deny its application for recognition, renewal, or reinstatement?

§ 1.1171 How does an accreditation body request reconsideration by FDA of a decision to deny its application for recognition, renewal, or reinstatement?

§ 1.1173 How does an accreditation body or laboratory request a regulatory hearing on FDA's decision to revoke the accreditation body's recognition or disqualify a LAAF-accredited laboratory?

§ 1.1173 How does an accreditation body or laboratory request a regulatory hearing on FDA's decision to revoke the recognized accreditation body's recognition or revoke the accredited laboratory's accreditation?

§ 1.1174 How does an owner or consignee request a regulatory hearing on a directed food laboratory order?

§ 1.1174 How does an owner or consignee request a regulatory hearing on a food testing order?

Electronic Records and Public Disclosure Requirements

Electronic Records and Public Disclosure Requirements under This Subpart

§ 1.1199 Are electronic records created under this subpart subject to the electronic records requirements of part 11 of this chapter?

§ 1.1199 Are electronic records created under this subpart subject to the electronic records requirements of part 11 of this chapter?

§ 1.1200 Are the records obtained by FDA under this subpart subject to public disclosure?

§ 1.1200 Are the records obtained by FDA under this subpart subject to public disclosure?

Also, in one location in the proposed rule we inadvertently misstated the title of this subpart (the third codified instruction, 84 FR 59452 at 59501). Throughout the final rule we correctly state the subpart title (“Laboratory Accreditation for Analyses of Foods”).

D. Incorporation by Reference

FDA is incorporating by reference two consensus standards, which were approved by the Office of the Federal Register in accordance with 5 U.S.C. 552(a) and 1 CFR part 51. Both standards are widely accepted globally. The consensus standards may be examined at FDA's Dockets Management Staff (see

ADDRESSES

).

The standards listed below are available for purchase from the International Organization for Standardization (ISO), Chemin de Blandonnet 8, CP 401, 1214 Vernier, Geneva, Switzerland, +41 22 749 01 11,

central@iso.org

(

https://www.iso.org/store.html

) or from any other source from which the user is assured that the copy to be received is an accurate version of the standard.

ISO/IEC 17011:2017, Conformity assessment—Requirements for accreditation bodies accrediting conformity assessment bodies, Second edition, November 2017 (Ref. 2). ISO/IEC 17011:2017 specifies the general standards for accreditation bodies assessing and accrediting conformity assessment bodies (“conformity assessment bodies” are organizations providing testing, inspection, management system certification, personnel certification, or product certification). Its incorporation by reference should allow us to use a framework that is familiar to accreditation bodies and the laboratory industry.

ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories, Third edition, November 2017 (Ref. 3). ISO/IEC 17025:2017 sets general standards for the competence of testing laboratories, including general management requirements such as impartiality and quality assurance. It is

very familiar to the testing laboratories that may be interested in applying to conduct food testing under this subpart.

IV. Legal Authority

We are issuing this final rule under the FD&C Act and FSMA. As noted, section 202(a) of FSMA, “Laboratory Accreditation for Analyses of Foods”, amends the FD&C Act to create a new provision, section 422, under the same name. Section 422 of the FD&C Act directs us to establish a program for the testing of food by accredited laboratories and provides several requirements for the program.

Additionally, section 701(a) of the FD&C Act gives FDA the authority to publish regulations for the efficient enforcement of the FD&C Act. The requirements discussed in this final rule will allow FDA to efficiently enforce section 422 of the FD&C Act. Thus, our legal authority for this final rule is derived primarily from section 422 and section 701(a) of the FD&C Act. Further, we also note that this rule is consistent with section 404 of FSMA, which states that nothing in FSMA should be construed in a manner that is inconsistent with the agreement establishing the World Trade Organization (WTO) or any other treaty or international agreement to which the United States is a party.

Section 379j-31 of the FD&C Act (21 U.S.C. 743) is one of many statutory provisions that provide authority for FDA's regulations contained in part 1 (21 CFR part 1). We inadvertently omitted that citation from the authority citation in the proposed rule, but have included it in the final rule.

V. Comments on the Proposed Rule and FDA Response

A. Introduction

We received approximately 70 comment submissions on the proposed rule by the close of the comment period, each containing one or more comments on one or more issues. We received comments from consumers, food associations, accreditation bodies, laboratory associations, laboratories, consumer groups, and other organizations.

In the remainder of this document, we describe the comments that are within the scope of this rulemaking, respond to them, and explain any revisions we made to the proposed rule.

We have numbered each comment to help distinguish between different comments. We have grouped similar comments together under the same number, and, in some cases, we have separated different issues discussed in the same comment and designated them as distinct comments for purposes of our responses. The number assigned to each comment or comment topic is purely for organizational purposes and does not signify the comment's value or importance or the order in which comments were received.

Note that summaries of and responses to comments on the estimated costs and benefits of the proposed rule and other topics covered by the Preliminary Regulatory Impact Analysis (PRIA) may be found in the Final Regulatory Impact Analysis (FRIA) (Ref. 4).

B. General Comments

Many comments made general remarks supporting or opposing the proposed rule without focusing on a particular proposed provision. Further, several comments made overarching comments that pertain to the rule more generally, focusing on issues throughout the rule such as program structure, FDA's role, terminology, and implementation. In the following paragraphs, we discuss and respond to such general comments.

(Comment 1) We received many comments expressing general support for the proposed rule, most expressing the view that the LAAF program would help to ensure the safety of food. Some of these comments stress the importance of accurate and reliable food testing results, and the role of valid results in enhancing food safety. Some comments focus on the advantages of setting quality standards and establishing accountability for food testing laboratories. Some comments opine that the laboratory accreditation program will increase U.S. consumer confidence in the safety of the food supply. Other comments maintain that the program will result in fewer illnesses, thus reducing healthcare costs. Other comments express support for implementation of FSMA section 202 and the underlying goals of the laboratory accreditation program,

e.g.,

improved safety of imported food, trustworthy testing results. A few comments opine that the rule would lead to more efficient food imports by clarifying what information needs to be in a laboratory analytical report, which should in turn expedite FDA review of those reports. These comments assert that such efficiencies are particularly valuable when the imported food is perishable, such as produce. Some of these comments further suggest that a more efficient review process for FDA could allow FDA to focus its limited resources on imports that generally are not subject to testing under this subpart.

(Response 1) We appreciate the comments in support of the proposed rulemaking and moving forward to implement the LAAF program. We agree that the program established by the final rule will help ensure the safety of food and should increase U.S. consumer confidence in the food supply. We also agree that requiring analyses to be performed by LAAF-accredited laboratories that meet the standards set forth in the final rule will make tests consistently more accurate and prevent illnesses. Further, setting model standards for LAAF-accredited laboratories will improve the reliability and accountability of test results on which we rely to make regulatory decisions regarding certain foods.

We agree with comments predicting fewer illnesses as a result of this final rule. For additional discussion of the cost benefit analysis associated with this final rule, see section VII. We also agree there will be efficiencies gained for industry and FDA from clarifying the requirements in an analytical report and from the process that allows submission of abridged analytical reports.

(Comment 2) Some comments question whether the LAAF program established by this final rule would make a food safety impact because only a small fraction of food testing laboratories are likely to participate.

(Response 2) Although the laboratory accreditation rule does not set mandatory standards for all food testing laboratories, the program will make an important difference for the food testing subject to the rule, as the testing situations covered by the rule all involve heightened food safety concerns. Therefore, the food testing covered by the rule addresses the specific circumstances in which accurate and reliable test results are especially important to protect public health. We also anticipate that some owners or consignees who are not covered by the rule may choose to use a LAAF-accredited laboratory because these laboratories will have met the program standards; this would create a benefit incidental to the program. Finally, we expect that creating model laboratory standards based on ISO/IEC 17025:2017 accreditation may encourage other laboratories to work toward these standards, including accreditation.

(Comment 3) Some comments are generally supportive of the proposed rule but state that FDA already regulates food safety, and because it is unclear how much safer food would be as a result of the proposed rule, the resources necessary for this program may be better spent elsewhere. A subset

of these comments states that the proposed rule would make food safety regulations more complicated for small food businesses and would also burden small food businesses with additional costs.

(Response 3) As described in section 422 of the FD&C Act, this final rule will establish a program for the accreditation of laboratories the use of which will be required in certain circumstances where heightened food safety concerns exist. We estimate the benefits outweigh the costs of the rule. For additional information on the estimated costs and benefits of this final rule, see section VII and the FRIA (Ref. 4). As mentioned in the preceding response, there may be other benefits incidental to the LAAF program.

Some comments express concern that this rule may complicate the regulatory landscape for small business owners and consignees that are also subject to other food safety regulations. It is true that some small owners and consignees will be required to use a LAAF-accredited laboratory for the testing described in § 1.1107. However, this rule does not create new testing requirements; it merely requires certain tests that are already occurring to be conducted by a LAAF-accredited laboratory. Further, in some cases the regulation creating the underlying testing requirement addresses this issue in its application to small businesses. For example, § 1.1107(a)(1)(ii) provides that certain shell egg tests required by the egg safety rule (see part 118 (21 CFR part 118)) are covered by this final rule. However, the egg safety rule does not apply to producers with less than 3,000 laying hens at a particular farm (see § 118.1(a)). Accordingly, those small egg producers are unaffected by this provision of the final rule. We also expect that the online registry of LAAF-accredited laboratories, described in § 1.1109, will make it easy for all owners and consignees to locate laboratories LAAF-accredited to conduct the tests covered by this subpart.

Regarding the concern that this final rule will burden small owners and consignees with additional costs, see the discussion below in section VII and the FRIA (Ref. 4).

(Comment 4) Some comments express support for specific aspects of the proposed rule, including the provisions protecting against conflicts of interest, and state that the program would improve transparency and consistency in the food testing that falls within its scope. Some comments contend that there have been situations in which a food is described in terms such as “safe” based on biased testing conducted by the food's producer.

(Response 4) We appreciate the supportive comments regarding the conflict of interest provisions. FDA anticipates that the model laboratory standards being established in this final rule, as well as the program requirements for LAAF-accreditation of laboratories by recognized accreditation bodies, will increase the reliability of tests conducted under this subpart. Ensuring that both accreditation bodies and laboratories are free from conflicts of interest is critical to the integrity of food testing conducted under this subpart. For more information on the conflict of interest requirements applicable to recognized accreditation bodies, see the discussion of § 1.1119 below; for more information on the conflict of interest requirements applicable to LAAF-accredited laboratories, see the discussion of § 1.1147 below.

(Comment 5) Some comments support the establishment of laboratory standards and appreciate the transparency of the public registry that will list recognized accreditation bodies and LAAF-accredited laboratories but express concern that laboratories would conform to the standards only while being actively monitored by the Agency. These comments encourage the Agency to address this risk.

(Response 5) We acknowledge a hypothetical risk that LAAF-accredited laboratories might conform to standards only while being actively monitored by FDA; however, we believe that the model laboratory standards and reporting requirements we are establishing in this final rule, as well as oversight of LAAF-accredited laboratories by both recognized accreditation bodies and FDA, will adequately address this risk. For example, under this subpart, FDA will recognize accreditation bodies that will LAAF-accredit laboratories to conduct certain testing of food under this subpart. Recognized accreditation bodies' assessment of LAAF-accredited laboratories involves onsite and remote assessments as described in § 1.1120 of the rule. FDA may conduct an onsite or remote review of a LAAF-accredited laboratory at any reasonable time to review performance (see § 1.1159(c)). LAAF-accredited laboratories must submit quality control results with each analytical report (see §§ 1.1152(d)(8), 1.1153(c)(2)), so FDA will be able to review the quality control results to ensure that methods are performed correctly. Further, for LAAF-accredited laboratories that submit abridged analytical reports, FDA may audit these reports by requesting that additional documentation or a full analytical report be submitted within 72 hours of the request (see § 1.1153(d)(2)).

In sum, in this final rule, FDA is establishing requirements for accreditation bodies and laboratories that will provide sufficient oversight of LAAF-accredited laboratories such that we expect consistent quality test results to be the norm.

(Comment 6) A few comments philosophically disagree with defining and regulating food at all, and thus oppose the establishment of a program to require any laboratory testing of food.

(Response 6) Congress defined “food” in section 201(f) of the FD&C Act (21 U.S.C. 321(f)) and by statute has authorized FDA to regulate food, including in section 422 of the FD&C Act, which directs FDA to establish this program.

(Comment 7) Some comments ask what effect the final rule will have on existing food testing laboratories. Other comments express a concern that some individuals may perceive that test results from laboratories not participating in the LAAF program are suspect or less valuable.

(Response 7) Food testing laboratories are not required to participate in this program; however, owners and consignees will be required to use a LAAF-accredited laboratory for the food testing covered by this rule, such as testing to support removal from import alert and the shell egg testing required by part 118 (see § 1.1107). Laboratories that wish to conduct the food testing covered by this rule will need to apply to a recognized accreditation body and must satisfy the standards established in this final rule in order to voluntarily participate in the program. A LAAF-accredited laboratory engaged by an owner or consignee to conduct the food testing covered by this final rule will conduct the test and send the results directly to FDA, in accordance with the requirements of this subpart.

Food testing laboratories that do not wish to conduct the testing described in § 1.1107 are not required to participate in the program.

We do not expect this program to decrease confidence in food laboratories that choose not to become LAAF-accredited, in part due to the very large number of food testing laboratories that exist and conduct all sorts of food testing for myriad customers and purposes. We view the program as beneficial to the food testing industry, as an explicit goal of the statute is to increase the number of qualified food testing laboratories. See section 422(a)(3) of the FD&C Act.

(Comment 8) Some comments advocate for expanded roles for the laboratories that participate in this program. Some of these comments suggest that LAAF-accredited laboratories could conduct tests for FDA's surveillance sampling program and argue that sufficient capacity exists in the United States for ISO/IEC 17025:2017-accredited laboratories to conduct all DWPE and FDA surveillance sampling and testing. Under the surveillance sampling program, FDA focuses its sampling and testing efforts on a few commodities at a time with the goals of keeping contaminated products from reaching consumers and facilitating a greater understanding of hazards. For more information on FDA's surveillance sampling, see

https://www.fda.gov/food/sampling-protect-food-supply/microbiological-surveillance-sampling.

These comments also suggest that FDA should create a program whereby private laboratories meet the standards of FDA laboratories, such that FDA could rely on those private laboratories for its testing needs and therefore focus its resources elsewhere. Finally, these comments suggest that independent accredited laboratories could also conduct sampling and testing on imported food, most of which is not sampled and tested by FDA prior to entry.

(Response 8) This final rule establishes the LAAF program, the scope of which is specified in FD&C Act section 422(b)(1) and described in § 1.1107. All the tests that will be conducted by LAAF-accredited laboratories are currently being conducted by non-FDA laboratories (

e.g.,

private laboratories). Expanding the scope of this program to include testing currently conducted by FDA laboratories, such as surveillance sampling, was not proposed because it is not contemplated by the statute. Any future expansion of this program will be accomplished via rulemaking and will include an opportunity for public comment.

(Comment 9) Some comments offer general support for this subpart, stating that it will improve the defensibility of the resulting test data by ensuring that all participating laboratories operate in accordance with a robust quality management system. These comments suggest that as we continue to develop the LAAF program, we consider two documents that were developed to improve the defensibility of human and animal food laboratory data: The Partnership for Food Protection document, “Human and Animal Food Testing Laboratories Best Practices Manual,” (Ref. 5) and the Association for Public Health Laboratories document, “Best Practices for Submission of Actionable Human and Animal Food Testing Data Generated in State and Local Laboratories” (Ref. 6). The former document is based on ISO/IEC 17025:2017 and its purpose is to “promote mutual acceptance and assurance of quality laboratory data shared among Federal, State, local, territorial, and tribal human and animal food regulatory agencies.” (Ref. 5). The latter document, focused on unaccredited laboratories, provides information on the minimum elements of a quality management system.

(Response 9) FDA appreciates this support and information. As an active member of the Partnership for Food Protection initiative, FDA is particularly familiar with the former document. We consider both documents to be helpful resources for the intended audiences.

1. FDA's Role and Related Terminology

In the proposed rule, FDA sought to define “accreditation” to mean, “a determination by a recognized accreditation body that a laboratory meets the applicable requirements

of this subpart

to conduct food testing under this subpart using one or more methods of analysis” (emphasis added). We then proceeded to use the word “accreditation” to mean that a laboratory had been approved to conduct testing under this subpart. For example, we wrote that the proposed rule “would establish certain model laboratory standards that accredited laboratories must meet to remain accredited” (84 FR 59452 at 59478). By way of another example, we wrote that the proposed provision on duration of accreditation under this subpart, “clarifies that an accredited laboratory's accreditation continues” until there is a voluntary or involuntary separation from the program (id. at 59489).

Consequently, when we used phrases such as, “FDA may revoke accreditation,” we intended to communicate that FDA could cause the involuntary separation of a laboratory from this program. For example, we wrote that “if we revoke the accreditation in whole of a laboratory, the laboratory would be immediately ineligible to conduct food testing under this rule” (id. at 59491).

We did not propose to define the term “assess.” However, we generally used it interchangeably with “evaluate.” For example, we entitled one section, “[h]ow must a recognized accreditation body evaluate laboratories seeking accreditation and oversee the performance of laboratories it accredits?” (Proposed § 1.1120, 84 FR 59452 at 59469). By way of additional examples, we also wrote, “[a]s the ISO/IEC 17025 revision is still relatively new, FDA is not able to adequately assess the accreditation of entities that only conduct sampling at this time” (id. at 59476); we said it was critical that we receive sufficient supporting information “for us to understand the test results and to assess the validity of the underlying testing” (id. at 59482) and we asserted authority to “exercise some ability to oversee accredited laboratories, via requesting records and, if appropriate, conducting onsite assessments” (id. at 59490).

(Comment 10) Numerous comments request that FDA address and clarify the roles and relationships among the Agency, recognized accreditation bodies, and LAAF-accredited laboratories under this subpart.

Several comments contend that the Agency should not use the words “assess” or “accredit” to describe Agency actions toward laboratories. Similarly, comments argued that FDA could not revoke a laboratory's “accreditation.” We understand several comments to be suggesting that the words “accredit” and “assess” have particular meaning in the accreditation body and laboratory community, and in the context of food testing, that meaning is always and necessarily related to the voluntary consensus standard ISO/IEC 17025:2017. For example, some comments state that FDA should limit its onsite “assessments” of laboratories to matters pertaining to this subpart. Comments explain that failure by FDA to use key terms as they are understood in the industry will lead to market confusion,

e.g.,

regarding the ISO/IEC 17025:2017 accreditation status of laboratories.

Some comments express concern that FDA may be under the impression that it can affect the ISO/IEC 17025:2017 accreditation of laboratories, either by “assessing” against the ISO/IEC 17025:2017 standard or by withdrawing a laboratory's ISO/IEC 17025:2017 accreditation. Comments argue that such a role is contrary to the Congressional intent underlying section 422 of the FD&C Act. Comments state that Congress did not intend for FDA to be an accreditation body. Some comments contend that FDA's role in the rule as proposed would be redundant of or “above” the role of the recognized accreditation bodies. Some comments express concern that FDA would be able to coerce a recognized accreditation body into withdrawing a laboratory's ISO/IEC 17025:2017 accreditation.

Some comments suggest that FDA's role should be administering a program that evaluates data or program integrity. Some comments suggest that FDA reframe its relationship with the laboratories in terms of an agreement to list and de-list the laboratories on our online registry. Some comments recommend that FDA grant each laboratory a license to conduct testing under this subpart. In this framework, comments state that FDA's role with regard to the laboratories would be limited to the review of test results and analytical reports submitted to FDA by the laboratories. Some comments suggest that FDA should perform some level of review, even if brief, of laboratory applications approved by recognized accreditation bodies. Finally, some comments offer to work with FDA to more clearly define roles and responsibilities under this program.

(Response 10) We agree that substantial revisions and considerable clarification are in order.

In proposing to define “accreditation,” to reflect a positive assessment by a recognized accreditation body

under this subpart,

we failed to sufficiently appreciate that in the context of food testing, many parties may perceive “accreditation,” to mean accreditation to ISO/IEC 17025:2017. Similarly, when we used the word, “assess,” we did not intend to communicate, “assess against ISO/IEC 17025:2017.” Instead, we used the word as consistent with its more general use: The Cambridge Dictionary defines “assess” as, “to judge or decide the amount, value, quality, or importance of something.” (Ref. 7).

Accordingly, it was not our intent to communicate that FDA had the authority to assess laboratories against the ISO/IEC 17025:2017 standard. For example, when we said in the proposed rule that we had the authority to conduct an “onsite assessment” of a laboratory participating in this program, we did not mean that our visit would be for the purpose of assessing against ISO/IEC 17025:2017. Nor did we intend to communicate that we had the authority to withdraw ISO/IEC 17025:2017 accreditation, or to pressure or demand an accreditation body to take such an action. We agree such a role would not be appropriate or consistent with section 422 of the FD&C Act.

To communicate our intent more effectively, we have taken several steps. First, we removed the definition of “accreditation” and no longer refer to laboratories that have been approved by a recognized accreditation body to conduct testing under this subpart as merely “accredited.” Instead, we use the more precise term “LAAF-accredited,” where “LAAF” is an acronym for the title of this subpart, “Laboratory Accreditation for Analyses of Foods.” We added a definition for “LAAF-accreditation” to § 1.1102. Where we do use the word, “accredited” in this final rule without further qualification, we generally mean accredited to ISO/IEC 17025:2017.

Second, we no longer use the verb “assess” to refer to an action that FDA takes regarding laboratories. We reserve the word “assess” to refer to the action a recognized accreditation body takes toward a laboratory. We employ the word “evaluate” to mean an activity FDA takes with regard to an accreditation body seeking to become recognized or already recognized under this subpart. Largely accepting the suggestion of some comments, we describe our relationship with regard to the laboratories under this subpart as “reviewing” the performance of LAAF-accredited laboratories.

Third, we do not use the word “revoke” in the final rule to mean an action FDA may take to remove a LAAF-accredited laboratory from this program. Instead, although an accreditation body may withdraw or reduce the scope of LAAF-accreditation, we say that FDA may “disqualify” a laboratory from conducting testing under this subpart. We note that although “disqualify” was used in the proposed rule in connection with permission to submit abridged analytical reports, we have revamped that process such that there is no longer a disqualification period. In the final rule, “disqualify” is used to describe the action FDA may take to remove a laboratory from the program; we say that FDA may “disqualify a LAAF-accredited laboratory from submitting analytical reports under this subpart” (see § 1.1161). For further information on the process related to submitting abridged analytical reports, see the discussion of § 1.1153 below at Response 124.

We agree in part with the comments suggesting that FDA perform some level of review of laboratory applications approved by recognized accreditation bodies. Although we have just explained that it is not appropriate for FDA to assess or accredit laboratories ourselves, we nevertheless have a responsibility to ensure that the laboratories we list on our website have been properly assessed by a recognized accreditation body. To that end, we will require the accreditation bodies to submit certain information to us concerning their assessment of a laboratory, including the resulting certificate listing the scope of LAAF-accreditation (see § 1.1123(d)). We decline the suggestion to reframe FDA's relationship with LAAF-accredited laboratories in terms of FDA granting a license to such laboratories, or in terms of entering into a listing agreement with the laboratories. We note that some comments suggest that such a construct could prove helpful in relation to FDA granting permission for certain laboratories to submit abridged analytical reports. Nevertheless, we have determined that such a construct would present complications (

e.g.,

could be legally cumbersome for the FDA to “license” laboratories) and is unnecessary to achieve the goals of this program.

We have implemented the revised terminology described here throughout the final rule. We also have tried to avoid describing the proposed rule using the now-discarded terminology (

e.g.,

FDA “assessing” a laboratory), even if that is the language we originally used in the proposed rule, because we wish to reduce confusion and communicate more clearly. We thank the commenters for their feedback on this important topic and we look forward to contributions of all interested shareholders as we implement the LAAF program.

2. Program Structure

(Comment 11) In the proposed rule, FDA proposed evaluating and recognizing accreditation bodies, and then those accreditation bodies would assess and LAAF-accredit laboratories. We received several comments on this proposed structure. Some comments express support because the rule relies on the current accreditation body-laboratory conformity assessment structure and leverages existing public-private partnerships in the United States.

Alternatively, some comments contend that the structure was unnecessary or ineffective. Some of these comments advocate that laboratories should simply send their analytical reports to FDA and the Agency would ensure the testing of food was properly conducted. Some comments contend that the only requirement should be that accreditation bodies are signatories to the International Laboratory Accreditation Cooperation (ILAC), and then let the accreditation bodies assess the laboratories for LAAF-accreditation, applying the accreditation bodies' usual standards. Some comments argue that FDA should not have any authority over accreditation bodies, because such authority would result in two entities overseeing the laboratories, which these

comments view as both confusing and intrusive.

(Response 11) The structure of the LAAF program is specified by the statute, per section 422(a)(1)(B) and (a)(2) of the FD&C Act. FDA will recognize accreditation bodies, which in turn will accredit laboratories. Further, there are advantages and efficiencies to relying on the structure of the existing conformity assessment industry (

i.e.,

accreditation bodies assess laboratories) for the structure of this program. For example, this familiarity may make it easier for these stakeholders to participate in the program. At the same time that we are glad to leverage widely accepted international voluntary consensus standards as foundational requirements, we are supplementing those standards with certain requirements that we have determined will help ensure the integrity of the testing under this program. As a reminder, all the testing that we are requiring be conducted by a LAAF-accredited laboratory is occurring in the context of increased food safety concern (see § 1.1107(a). For example, under § 1.1107(a)(4), testing to support the release of food detained at the border because it is or appears to be adulterated or misbranded, is covered by this rule. Accordingly, we have determined that it is appropriate to impose some requirements in addition to those of the international voluntary consensus standards.

Regarding the concern that FDA's exercise of authority over recognized accreditation bodies for purposes of this program will be confusing and intrusive, we have structured the program such that FDA evaluates the recognized accreditation bodies, and the accreditation bodies assess the laboratories against the model standards established in this rule, including conformity to ISO/IEC 17025:2017. FDA will not be assessing laboratory applicants.

As shown in section I.A. above, we have revised the program structure diagram from the proposed rule (see 84 FR 59452 at 59453) to reflect changes made in the final rule. The program structure diagram incorporates revised program terminology throughout (

i.e.,

“LAAF-accredited”; see discussion at Response 10). We also include a second box representing FDA to better illustrate our roles of recognizing accreditation bodies and reviewing results and supporting information submitted by LAAF-accredited laboratories.

(Comment 12) Some comments opine that the framework of the proposed rule is inappropriate. These comments contend that it is not appropriate for FDA to oversee accreditation bodies because FDA is not an ILAC signatory. These comments further state that only accreditation bodies should oversee the laboratories they accredit and that therefore FDA's involvement would be both unnecessary and confusing. These comments recommend that FDA simply maintain a list of ILAC-signatory accreditation bodies, and have laboratories accredited by those listed accreditation bodies submit test results to us.

(Response 12) We disagree that the framework of the rule, and FDA's oversight of both recognized accreditation bodies and LAAF-accredited laboratories, is inappropriate. Section 422 of the FD&C Act directs FDA to establish this program and, in relevant part, provide for the recognition of laboratory accreditation bodies that meet criteria established by the Secretary (see section 422(a)(2) of the FD&C Act). The Agency has established that being an ILAC signatory is a necessary, but not sufficient, condition to being recognized by FDA to LAAF-accredit laboratories. We have determined it necessary and appropriate to set additional standards for accreditation bodies, such as the conflict of interest requirements in § 1.1119. FDA must also evaluate the work of the accreditation bodies to ensure the integrity of the program. Further, the statute directs the Agency to periodically review a recognized accreditation body's compliance with the requirements of the program.

Similarly, section 422(a)(6) of the FD&C Act directs the Agency to develop model standards that a laboratory must meet to be LAAF-accredited to conduct testing under this subpart. We have adopted ISO/IEC 17025:2017 accreditation as a baseline requirement, but given the specific circumstances in which food testing is required to be conducted by a LAAF-accredited laboratory and since we use the results of such tests to inform regulatory decisions and protect public health, we have included FDA oversight of LAAF-accredited laboratories among the components of the program (see section 422(a)(6)(B) of the FD&C Act).

Therefore, FDA oversight of recognized accreditation bodies is not only appropriate, but it is also required by statute. Further, FDA has determined that oversight of LAAF-accredited laboratories submitting test results to FDA is appropriate given the Agency's use of the test results. The alternative framework proposed by the comment is not a viable option for a comprehensive and effective program that is sufficiently protective of public health.

(Comment 13) A few comments encourage FDA to reassess our proposal to place laboratories or accreditation bodies in probationary status, which is noted on the public registry, after finding one or more nonconformances. These comments suggest that we consider the variety of circumstances that may surround nonconformance, including that the entity may be in the process of actively addressing the nonconformance. The comments express a concern that publication of probationary status on the online registry may negatively and unfairly impact the entity, as the entity may be in the process of addressing the issue that resulted in a non-conformance.

(Response 13) We agree that entities should have an opportunity to address concerns before those concerns cause the entity to be placed on probation, particularly as probation will be noted on the online registry. Accordingly, we have revised the final rule such that generally an entity will be notified of deficiencies and provided an opportunity to take corrective action prior to being placed on suspension or probation. Consistent with our decision to incorporate by reference ISO/IEC 17011:2017 and ISO/IEC 17025:2017, we have decided to leverage the corrective action processes described in those standards to provide such an opportunity.

Under these ISO/IEC standards, the corrective action process requires the entity to do more than simply correct a non-conformity. Instead, the entity is required to consider the non-conformity from a process perspective, including identifying the cause of the non-conformity and considering whether internal process changes are needed to prevent its recurrence. FDA's view is that that this focus on looking for and addressing any systemic weaknesses in the entity's procedures, rather than simply remedying a single error or lapse, will serve to strengthen both the accreditation bodies and the laboratories that participate in this program, and therefore the LAAF program itself.

Section 1.1121(a) of the final rule states that if a recognized accreditation body observes a deficiency in a LAAF-accredited laboratory, the recognized accreditation body may require corrective action using the procedures described by ISO/IEC 17025:2017 section 8.7 (Ref. 3). Similarly, we have revised §§ 1.1131 and 1.1161 regarding FDA oversight actions regarding recognized accreditation bodies and LAAF-accredited laboratories, respectively, such that generally entities will be provided an opportunity to take

corrective action prior to being placed on probation.

Some problems may warrant immediate action by a recognized accreditation body to suspend, reduce the scope of, or withdraw the LAAF-accreditation of a laboratory or by FDA to immediately disqualify a LAAF-accredited laboratory. For additional information, see § 1.1121 (“When must a recognized accreditation body require corrective action, suspend a LAAF-accredited laboratory, reduce the scope of, or withdraw the LAAF-accreditation of a laboratory?”); § 1.1131 (“When will FDA require corrective action, put a recognized accreditation body on probation, or revoke the recognition of an accreditation body?”); and § 1.1161 (“When will FDA require corrective action, put a LAAF-accredited laboratory on probation, or disqualify a LAAF-accredited laboratory from submitting analytical reports?”).

Finally, note that we have revised the final rule to refer to “suspension” of LAAF-accredited laboratories by recognized accreditation bodies instead of “probation” as proposed. The final rule retains and limits the term “probation” to refer to an action that FDA may take with respect to a recognized accreditation body or a LAAF-accredited laboratory in certain circumstances (see §§ 1.1131 and 1.1161). For more information on this terminology change, see Comments 58, 71, and 82 and Responses.

3. Implementation

(Comment 14) Several comments address implementation. In section VII of the proposed rule, we proposed that implementation would occur in a stepwise fashion; we would focus first on accreditation bodies and subsequently, laboratories. See 84 FR 59452 at 59495. We proposed that after the program attains sufficient laboratory capacity, we would publish a notice in the

Federal Register

giving 6 months' notice that owners and consignees would be required to use laboratories approved for participation in this program. All comments on this aspect of our proposal endorse a stepwise approach to implementation. These comments also agree with providing notice to affected entities via a

Federal Register

document. Some comments encourage the Agency to also issue

Federal Register

notices to announce when we will commence accepting applications from accreditation bodies, and when recognized accreditation bodies are able to start accepting applications from laboratories.

(Response 14) We appreciate comments supporting our proposed implementation steps. As we stated in the preamble to the proposed rule, implementation of the LAAF program will necessarily occur in a stepwise fashion. We will announce when accreditation bodies may apply for recognition. When we have recognized a sufficient number of accreditation bodies, we will announce that laboratories may apply to the recognized accreditation bodies for LAAF-accreditation. When we have sufficient LAAF-accredited laboratory capacity for the testing covered by § 1.1107, we will publish a document in the

Federal Register

giving owners and consignees 6 months' notice that they will be required to use a LAAF-accredited laboratory for such testing.

We decline to commit to publishing notices in the

Federal Register

to announce that we are ready to accept applications from accreditation bodies and that laboratories may apply to recognized accreditation bodies. There are a variety of methods to communicate effectively with stakeholders and the interested public; at the appropriate time we will determine which methods best advance the Agency's interest in transparency and the needs of the LAAF program.

(Comment 15) Some comments recommend that in addition to the stepwise approach discussed in the previous comment and response, we also take a phased-in approach to implementation. That means that FDA would only require testing under the rule for the various categories of tests described in § 1.1107 as sufficient laboratory capacity is attained for each. Some comments suggest that we refrain from requiring testing under the rule until we have achieved sufficient laboratory capacity for a majority of the tests covered by the rule.

Some comments maintain that there will be sufficient laboratory capacity for the DWPE-related testing covered by the final rule, because as we noted in the proposed rule, 10 laboratories that conduct the majority of such testing already are ISO/IEC17025-accredited (see 84 FR 59452 at 59457). These comments state that there are “hundreds” of ISO/IEC 17025-accredited independent food laboratories in the United States that potentially could participate in the program, which would expand capacity. These comments expect that the program we are establishing in this final rule would also increase incentives for ISO/IEC17025 accreditation and therefore expand capacity even further.

Some comments question whether, and some comments ask when, sufficient laboratory capacity will be reached for all the tests covered by this final rule. Other comments inquire how FDA will determine when sufficient laboratory capacity has been reached. Some comments urge that when FDA considers whether there is sufficient laboratory capacity, we take into account whether laboratories can perform the testing in a timely manner. Other comments suggest that when we consider capacity, we take into account laboratory location relative to owners and consignees. Some comments predict that it will take a long time to achieve sufficient laboratory capacity, and some comments request that we explain what will happen if sufficient laboratory capacity is not attained for a particular category of testing. Some comments encourage FDA to identify the LAAF-accredited laboratories publicly once sufficient capacity is reached.

Further, some comments express skepticism that the program would ever be able to attain sufficient capacity to implement the bottled drinking water followup testing covered by the rule (see § 1.1107(a)(1)(iii)). These comments state that such followup tests occur rarely and suggest that no water testing laboratory will find it worthwhile to participate in this program for the relatively little bottled drinking water followup testing business it might gain by doing so.

Other comments focus on laboratories that currently test shell eggs and maintain that many such laboratories are not currently ISO/IEC 17025-accredited. These comments question whether those laboratories would choose to become ISO/IEC 17025-accredited in order to participate in this program, as, according to these comments, such laboratories would be unlikely to test any commodities covered by this final rule other than shell eggs. These comments state it is unclear how quickly additional laboratories would be able to get approved for participation in the program and predict there could be a logistical problem of bottlenecking if sufficient laboratory capacity for a particular test is not attained. These comments encourage FDA to consult with the National Poultry Improvement Plan at the U.S. Department of Agriculture and other Agencies that have experience testing agricultural products. Finally, these comments ask that FDA allow adequate time for a sufficient number of laboratories to become LAAF-accredited to conduct the shell egg testing described in § 1.1107(a)(1)(ii) before we require owners and consignees to have those tests conducted under this program.

(Response 15) We agree that given the breadth of matrices and methods covered by the rule it may be necessary to separately consider whether sufficient laboratory capacity has been attained for the variety of tests described in § 1.1107. As discussed in the preceding comment and response, the first implementation step is for FDA to receive, review, and evaluate applications from accreditation bodies. Once we have recognized a sufficient number of accreditation bodies, we anticipate that many laboratories will be interested in becoming LAAF-accredited, but it is impossible for us to predict various relevant factors including how many laboratories will apply, the methods for which they will be successful, and the associated timeframes. Perhaps sufficient laboratory capacity will be promptly attained for all tests covered by the rule; that would allow us to issue a single

Federal Register

document notifying owners and consignees that in 6 months they must use a LAAF-accredited laboratory for all tests described in § 1.1107. That outcome is not assured, however, and therefore we may phase in implementation as suggested by some comments. To the extent that some comments suggest we wait to implement any of the rule until we have attained sufficient capacity for a majority of all the tests covered by the rule, we decline the suggestion due to the many variables that are not entirely within our control (the number of laboratories that apply as soon as they are able, the number and capacity of recognized accreditation bodies that will be assessing the initial laboratory applications, etc.).

We appreciate the comments contending that there will be more than sufficient laboratory capacity for all the testing under this rule. This program represents the least amount of change for those private laboratories that are already ISO/IEC 17025-accredited and have been conducting the tests that support admission of a food under section 801(a) of the FD&C Act and removal from DWPE under an import alert and sending their test results and associated analyses to FDA, some for many years. Further, as indicated by some comments, the data we analyzed for the proposed rule indicated that many of the laboratories that have been conducting tests to support admission of a food and removal from DWPE under import alerts are already ISO/IEC 17025-accredited; the cost for such laboratories to become LAAF-accredited is relatively low. We agree with comments maintaining that our reliance on ISO/IEC 17025 as a foundational requirement for LAAF-accreditation provides an incentive for laboratories to become ISO/IEC 17025-accredited and we note that an explicit goal of section 422 is to increase the number of laboratories qualified to conduct testing under this subpart (see section 422(a)(3) of the FD&C Act).

Determining whether the program has attained sufficient laboratory capacity may appear to be a simple comparison of the number of a particular type of test that is needed, to the number of laboratories LAAF-accredited for that method. The reality is far different. Test demand cannot be predicted with certainty; in part it is a result of the prevalence of circumstances presenting heightened food safety concerns (

e.g.,

the number and breadth of import alerts; how much food product is or appears to be violative when offered for import) and in part it is a result of business choices outside of our control or knowledge (

e.g.,

how much food subject to DWPE is offered for import; whether a shell egg producer's environment tests positive for

Salmonella

Enteritidis and whether the producer then chooses to test its shell eggs or divert them to treatment (see §§ 118.5(a)(2)(ii) and (b)(2)(ii); 118.6(a)(2)). Some laboratories are much bigger than others, and bigger laboratories presumably can conduct more tests than smaller laboratories, so simply knowing how many laboratories are LAAF-accredited for a given method does not present a complete picture of capacity. We acknowledge that location is a relevant factor in choosing a laboratory, in large part due to the time and cost implications of shipping samples to a laboratory that is relatively far away, but the degree to which this factor is relevant to laboratory capacity may vary depending on the test at issue (

e.g.,

size of sample, whether there are time and temperature requirements, the degree to which a product is perishable). Similarly, although timeliness may be an important factor for one sort of food test, it may be less critical in other food testing contexts. Other factors may also be relevant, and as noted above, it is infeasible for us to predict them all.

FDA is committed to implementing this program promptly and, as in other FSMA contexts, in a practical manner. In determining laboratory capacity we will take all relevant information and factors into account. We remain committed to providing owners and consignees 6 months' notice via a document in the

Federal Register

before requiring them to use a LAAF-accredited laboratory for the testing covered by this rule. We will not preclude the possibility that we may issue more than one

Federal Register

document as laboratory capacity is attained for various tests described in § 1.1107.

The publication of this final rule in the

Federal Register

arguably marks the beginning of the implementation of this program. Although we expect to reach sufficient laboratory capacity for all the tests covered by this rule, we decline the invitation of some comments to predict how long it will take to achieve that milestone. If sufficient laboratory capacity is not reached for a particular category or subcategory of the tests described in § 1.1107, then the immediate result would be that we not require owners and consignees to use a LAAF-accredited laboratory to conduct those particular tests.

We anticipate a sufficient number of LAAF-accredited laboratories for the bottled drinking water tests covered by this final rule (see § 1.1107(a)(1)(iii)). For a related discussion, please see Comment and (Response 87.

Some comments claim that the laboratories that currently conduct shell egg testing tend not to be accredited to ISO/IEC 17025. These comments express concern that such laboratories may not become LAAF-accredited, which may result in a bottleneck effect (due to insufficient laboratory capacity). First, as discussed earlier in this response, FDA does not intend to require owners and consignees to use a LAAF-accredited laboratory for the testing described in § 1.1107 until the program has attained sufficient laboratory capacity for the relevant testing, even if that means that a LAAF-accredited laboratory is required for some categories or subcategories of testing described in § 1.1107 sooner than for other categories or subcategories. Accordingly, the implementation of this program should not result in a bottleneck for shell egg testing.

The research supporting the FRIA for this final rule (Ref. 4), and the information we gleaned from our consultations with the National Poultry Improvement Plan, is consistent with comments' claim that the majority of laboratories that currently conduct the shell egg testing described in § 1.1107(a)(1)(ii) are not accredited to ISO/IEC 17025. Although we believe some of those laboratories will pursue ISO/IEC 17025 and LAAF-accreditation as a result of this final rule, we have no way of knowing with certainty.

We estimate that once this final rule is fully implemented, FDA will receive about 3,771 analytical reports of shell egg testing per year (Ref. 4). Due to the testing regime required under the FDA

egg safety rule, each analytical report will consist of 50 tests (each shell egg sample of 1,000 eggs is separated into 50 pools of 20 eggs each). (See § 118.6.) Accordingly, we expect that more than 188,000 FDA-required shell egg tests currently conducted each year to comply with § 118.6 will eventually be conducted by LAAF-accredited laboratories. If the laboratory market responds rationally, a sufficient number of laboratories will react to the business opportunity those shell egg tests create and choose to become LAAF-accredited. If a sufficient number of laboratories that currently conduct shell egg tests choose not to become LAAF-accredited, then other laboratories will emerge to seize this opportunity. The costs of becoming LAAF-accredited for laboratories new to shell egg testing will be lowest for those laboratories that are already accredited to ISO/IEC 17025; it would therefore be reasonable to expect such laboratories to pursue LAAF-accreditation to conduct shell egg testing. The FRIA in section II.F.3.f. accounts for the costs for some shell egg producers to switch laboratories if the one they are currently using is not LAAF-accredited (Ref. 4).

Shell egg testing is only required if the poultry house has tested positive for

Salmonella

Enteritidis, and the producer chooses not to divert the eggs to treatment. The central purpose of this final rule is to help ensure that the results of certain food testing that takes place amidst just this sort of heightened food safety concern, are reliable and accurate. No comments suggest that shell egg testing should be excluded from the coverage of this final rule, or subject to less stringent standards. We expect to avoid the logistical problem identified by these comments. And as noted above, we are committed to providing 6 months' notice via a

Federal Register

document before shell egg producers are required to use a LAAF-accredited laboratory to conduct the testing described in § 1.1107(a)(1)(ii).

C. Comments Regarding General Provisions

Table 2—Changes to General Provisions

Final rule

Proposed rule

Note

§ 1.1101 What documents are incorporated by reference in this subpart?

N/A

New section for centralized incorporation by reference (IBR).

§ 1.1102 What definitions apply to this subpart?

§ 1.1102 What definitions apply to this subpart?

See preamble table below for specific changes to § 1.1102.

§ 1.1103 Who is subject to this subpart?

§ 1.1103 Who is subject to this subpart?

See preamble discussion below for specific changes to § 1.1103.

1. What documents are incorporated by reference in this subpart (§ 1.1101)?

In the proposed rule, we proposed to incorporate by reference two international voluntary consensus standards: ISO/IEC 17011, Conformity assessment—Requirements for accreditation bodies accrediting conformity assessment bodies, Second edition, November 2017 (Ref. 2), for accreditation bodies, and ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories, Third edition, November 2017 (Ref. 3), for laboratories.

This final rule implements section 422 of the FD&C Act against the backdrop of the broader Federal policies on consensus standards and conformity assessment under the National Technology Transfer and Advancement Act of 1995 (NTTAA) (Pub. L. 104-113). The NTTAA, together with the Office of Management and Budget (OMB) Circular A-119, revised January 27, 2016 (81 FR 4673), directs Federal Agencies to use voluntary consensus standards in lieu of government-unique standards except where inconsistent with law or otherwise impractical. OMB Circular A-119 states that the use of voluntary standards, whenever practicable and appropriate, is intended to eliminate the cost to government of developing its own standards; decrease the cost of goods procured and the burden of complying with Agency regulation; provide incentives and opportunities to establish standards that serve national needs, and encourage long-term growth for U.S. enterprises and promote efficiency and economic competition through harmonization of standards; and further the policy of reliance upon the private sector to supply the government with cost-effective goods and services (Ref. 8).

As directed by OMB in Circular A-119, the National Institute of Standards and Technology (NIST), in the

Federal Register

of September 29, 2020 (85 FR 60904), issued updated policy guidance on Federal conformity assessment activities. The Federal conformity assessment guidance is codified at 15 CFR part 287 and applies to all Federal Agencies that set policy for, manage, operate, or use conformity assessment activities or results (85 FR 60904 at 60905). The guidance advises Agencies on using conformity assessment to meet government needs in a manner that is efficient and cost-effective for both the Agency and its stakeholders (15 CFR 287.1(a)). In keeping with these national policies, FDA has determined that it is appropriate and will be beneficial to both the Agency and the public if we rely on voluntary consensus standards to provide the baseline requirements for both accreditation bodies and laboratories wishing to participate in the LAAF program.

In the proposed rule, the incorporation by reference information was repeated throughout the codified text (

e.g.,

§ 1.1113(b) (ISO/IEC 17011:2017); § 1.1138(a)(2) (ISO/IEC 17025:2017)). On our own initiative, for readability we have revised the final rule to include a centralized incorporation by reference section at § 1.1101. Note that throughout the codified, after the year of each standard, we included the letter “E” to clarify that we are incorporating the standard in English (

e.g.,

“ISO/IEC 170211:2017(E)).” However for readability, we did not repeat the “E” after each mention of the standards throughout the preamble.

We received a few comments regarding the proposal to incorporate by reference the two consensus standards. These comments are addressed below.

(Comment 16) Several comments support our reliance on existing international voluntary consensus standards: ISO/IEC 17011:2017 for accreditation bodies and ISO/IEC 17025:2017 for laboratories.

(Response 16) Voluntary consensus standards such as ISO/IEC 17011:2017 and ISO/IEC 17025:2017 are developed by organizations with the involvement of interested parties representing various roles, concerns, and perspectives, via a robust process that seeks to achieve consensus (Ref. 9). As noted in the immediately preceding

section, Federal law and policy direct us to use voluntary consensus standards rather than creating our own unique standards whenever practical and consistent with our legal obligations. Further, section 422(a)(6) of the FD&C Act specifically directs the FDA to “consult existing standards” in the course of developing model standards for this rulemaking.

Comments do not suggest that we consider any other standard for accreditation bodies wishing to participate in this program. And although some comments recommend that we permit the participation of laboratories that meet certain industry-specific standards (see Comment 87 and Comment 88), no comment suggests a standard other than ISO/IEC 17025:2017 as a baseline requirement. We appreciate support for our position that ISO/IEC 17011:2017 and ISO/IEC 17025:2017 are the most appropriate globally recognized and widely used standards for the LAAF final rule.

2. What definitions apply to this subpart (§ 1.1102)?

Table 3—Revisions to the Proposed Definitions in § 1.1102

Term

Revision

Accreditation

Term revised to “laboratory accreditation for analyses of foods (LAAF) accreditation” to clarify that decisions regarding accreditation under this subpart are limited to the LAAF program.

Accredited laboratory

Term revised to “LAAF-accredited laboratory.”

Analyst

No change.

Corrective action

New term that we define as an action taken by an accreditation body or laboratory to investigate and eliminate the cause of a deficiency so that it does not recur.

Food

No change.

Food testing, testing of food

No change.

Food testing order

Term revised to “directed food laboratory order” to more accurately describe the order. Revised the definition to strike reference to § 1.1107(a)(2); the definition now states the order is issued only under § 1.1108.

Owner or consignee

Definition revised to refer to the circumstances in § 1.1107(a) instead of repeating the circumstances in § 1.1107(a) in the definition.

Recognition

Definition revised to refer to LAAF-accreditation of laboratories.

Recognized accreditation body

Definition revised to refer to the accreditation body's authority with respect to LAAF-accredited laboratories.

Representative sample

Definition revised to clarify that accuracy is to a “statistically acceptable degree” in response to comments and a grammatical revision made on our own initiative.

Sampler

Definition revised to reference the individual who collects a sample.

Sampling firm

New term that we define as an entity that provides sampling services.

Scope of accreditation

Term revised to “scope of LAAF-accreditation” and definition revised to delete the second sentence of the definition to remove the phrases, “in-whole” and “in-part” from the definition and throughout the rule.

We proposed to apply the definitions in section 201 of the FD&C Act unless otherwise specified. Additionally, we proposed to codify several terms used in the LAAF regulations. We received several comments on this section. As discussed in the following paragraphs, we have revised many of the terms and proposed definitions in response to comments received, as well as on our own initiative. Where we disagree with comments or decline a suggested revision, we offer an explanation in response. Some definitions were finalized as proposed.

The definitions for terms used in the laboratory accreditation for analyses of foods regulations are codified in § 1.1102.

Accreditation, Accredited Laboratory

We proposed to define

accreditation

and

accredited laboratory

to relate to determinations regarding a laboratory under this subpart. On our own initiative, we moved the phrase, “under this subpart” in the definition of the term, “LAAF-accredited laboratory” to clarify that food testing is conducted under this subpart as opposed to using methods of analysis under this subpart, as proposed.

(Comment 17) A number of comments express concern with the proposed definitions of “accreditation” and “accredited laboratory,” suggesting that they may result in confusion with similar terms already being used by industry. Some comments recommend aligning the definitions of “accreditation” and “accredited laboratory” under this regulation with their meaning in the conformity assessment industry to avoid potential confusion. Others propose that we differentiate the terms under this regulation from those used elsewhere and suggest the more specific terms, “Section 422 accreditation” and “Section 422 accredited laboratory” as potential options.

(Response 17) We acknowledge the potential for confusion regarding the terms, “accreditation” and “accredited laboratory” under this subpart with the use and understanding of these terms by industry. Accordingly, we have revised the terms to be specific to the LAAF program. Therefore, the terms have been revised to “LAAF-accreditation” and “LAAF-accredited laboratory” respectively in § 1.1102 and throughout the rule to clarify the impacts and limitations of accreditation decisions under this subpart. See also Comment and Response 10.

Analyst

We received no comments on the proposed definition of “analyst” and therefore have finalized the definition as proposed.

Corrective Action

We have added a definition for corrective action to clarify that in this subpart, it means, “an action taken by an accreditation body or laboratory to investigate and eliminate the cause of a deficiency so that it does not recur.” For additional discussion, see Comment and Response 31.

Food

In the proposed rule, we defined “food” as having the meaning given in section 201(f) of the FD&C Act, except that food does not include pesticides (as defined in 7 U.S.C. 136(u)). The proposed definition would align with the definition of “food” in the “Accreditation of Third-Party Certification Bodies to Conduct Food Safety Audits and to Issue Certifications” (21 CFR 1.600

et seq.

) (Accredited Third-Party Certification Program) and the “Foreign Supplier Verification Programs for Food Importers” (21 CFR 1.500

et seq.

) (FSVP) regulations.

(Comment 18) Some comments express support for the proposed definition of “food,” which the comments characterize as being the same as the definition in section 201(f) of the FD&C Act.

(Response 18) We appreciate the support for our proposed definition of “food” and we are retaining it without change. We note that for the purposes of this subpart, we are not giving the term, “food,” the same meaning as in section 201(f) of the FD&C Act. Under section 201(f), “food” is not defined to exclude pesticides, whereas the definition in this subpart expressly indicates that food does not include pesticides. As we stated in the proposed rule, we have not identified a need for “food” to include pesticides for purposes of this final rule, and no comment suggests otherwise.

Food Testing, Testing of Food

We proposed to define “food testing” and “testing of food” to mean the analysis of food product samples or environmental samples.

(Comment 19) Numerous comments indicate support for the inclusion of environmental testing within the definition for “food testing” and “testing of food” in the proposed rule. These comments assert that both food product and environmental testing are important to protecting public health. Conversely, multiple comments oppose the proposal to include environmental testing within the definition of “food testing” and “testing of food.” Some of these comments suggest that because FSMA section 202 did not explicitly mention environmental testing, the statute only permits the testing of food product samples, and not environmental samples, within the scope of this regulation. Other comments suggest that the definition of “food testing” and “testing of food” should be consistent in scope with the statutory definition of “food” in section 201(f) of the FD&C Act and limited to the analysis of food product samples only. Some comments further specify that although they oppose the inclusion of environmental testing within the definition for “food testing” and “testing of food,” they recognize the utility of environmental monitoring in ensuring food safety. Similarly, some comments state that the food industry has conducted environmental testing for a long time and argue that industry does not need this final rule to cover environmental testing to continue conducting such testing.

(Response 19) After carefully considering the comments and the statute, we define “food testing” and “testing of food” to mean, “the analysis of food product samples or environmental samples.”

As discussed in the proposed rule, the terms, “food testing” and “testing of food,” used in section 422 of the FD&C Act, are not defined in the statute (84 FR 59452 at 59460). We find these terms ambiguous and rely on context for their interpretation. Section 202(a) of FSMA is located in Title II of FSMA, which is titled “improving capacity to detect and respond to food safety problems.” Further, in describing some of the testing to be covered by this subpart, section 422(b)(1)(A) of the FD&C Act twice includes testing that addresses, “an identified or suspected food safety problem.” This context indicates the critical importance of “food testing” and “testing of food” being interpreted to include the analysis of environmental samples, so that this final rule will cover an important method of detecting and responding to identified and suspected food safety problems. We acknowledge and appreciate those comments asserting that including environmental testing is important to addressing food safety concerns and protecting public health. We also note that even some comments that oppose defining “food testing” and “testing of food” to include environmental testing state that such testing plays a valuable role in identifying potential pathways for contamination and helping to ensure food safety.

We agree with aspects of comments that acknowledge the importance of testing food production environments (

e.g.,

the environment where food is grown, harvested, packed, held, processed, or manufactured). The term, “environment” includes food contact surfaces such as utensils and table surfaces. Pathogens in the environment can be (and unfortunately, sometimes are) transmitted to food. Therefore, environmental testing is sometimes used as a followup test to verify that cleaning and sanitizing designed to eliminate an identified pathogen, was sufficient to eradicate that pathogen. Environmental testing may also be employed to determine the source of an identified pathogen (

e.g.,

in circumstances where a food product tested positive for a pathogen but it is not yet known how the food became adulterated). It is important that FDA be able to utilize this subpart to help ensure valid testing in the context of those sorts of heightened food safety concerns.

Some comments indicate that Congress used the term, “environmental testing” in other parts of the statute and could have done so here. Although we do not disagree with that statement, we note that Congress also used the term, “product testing,” in other parts of the statute, and could have done so here. We do not believe the absence of these phrases implies a lack of statutory authority to include both product and environmental testing within the scope of this final rule. Furthermore, the inclusion of both types of testing within the scope of the final rule serves a central purpose of section 422 of the FD&C Act, which is to improve FDA's access to reliable and accurate results of public health significance, thus improving our capability to protect U.S. consumers from unsafe food.

Some comments contend that the statutory definition of “food” limits our definitions of “food testing” and “testing of food,” to product samples. As we acknowledged in the preamble to the proposed rule, that is one, but not the only, reasonable interpretation of the statute. For the reasons discussed, we are adopting a different and more public health-protective interpretation and therefore finalize the definition of “food testing” and “testing of food” without change.

Finally, we appreciate that many in the food industry have long monitored their production environment through environmental testing. We applaud and encourage the continued practices of firms that conduct robust environmental monitoring programs. As discussed further in Response 35, this final rule does not cover routine environmental testing.

Food Testing Order

We proposed to define “food testing order” as an order issued by FDA under §§ 1.1107(a)(2) and 1.1108 requiring food testing to be conducted under this subpart by or on behalf of an owner or consignee. Although we did not receive specific comments regarding the proposed definition, we received many comments about the food testing order provisions in proposed §§ 1.1107 and 1.1108. We discuss those comments in section V.D. below; however, we are also making a change to the related terminology. We have revised the term, “food testing order” to “directed food laboratory order” throughout the rule to more accurately reflect the order and its impact. To reduce confusion, we generally use the term, “directed food laboratory order,” throughout this document, even when referring to discussions in the proposed rule.

On our own initiative, we revised the definition to strike the reference to § 1.1107(a)(2) and now state the order is issued solely under § 1.1108, as this provision directly describes FDA's issuance of such orders.

Owner or Consignee

We proposed to define “owner or consignee” as a person with an ownership interest in the food or environment samples in the circumstances described in proposed § 1.1107. On our own initiative, we have revised the definition to refer more generally to the circumstances described in § 1.1107 instead of repeating the circumstances in the definition.

Recognition

We proposed to define “recognition” to mean a determination by FDA that an accreditation body meets the applicable requirements of the LAAF program and is authorized to accredit laboratories under this subpart. As a result of revising the terms, “accreditation” and “accredited laboratory” to be specific to the LAAF program, we have revised the definition of “recognition” to reflect that a recognized accreditation body will LAAF-accredit laboratories to conduct food testing under this subpart.

(Comment 20) Some comments state that having a definition for “recognition” specific to this regulation may result in confusion, as the term is already used by the conformity assessment industry in other contexts outside of this regulation.

(Response 20) In contrast to the many comments that argue that our proposed use of the terms “accreditation,” “accredited laboratory,” and “assessment,” created confusion, only a small number of comments claim that our proposed use of the term, “recognition,” would create the potential for confusion. Further, these comments provide no specific examples of how the term, “recognition,” would be confusing, and do not offer alternative terms or definitions.

In addition, the FDA Foods Program uses the term, “recognition,” in the same way as proposed in our Accredited Third-Party Certification Program (see 21 CFR 1.600), and has not heard from those program participants that the term has proved problematic. For more information on the Accredited Third-Party Certification Program, see

https://www.fda.gov/food/importing-food-products-united-states/accredited-third-party-certification-program.

Therefore, we are retaining the definition of the term, “recognition” in the final rule.

Recognized Accreditation Body

We proposed to define “recognized accreditation body” as an accreditation body that FDA has determined meets the applicable requirements of this subpart and is authorized to accredit laboratories under this subpart. We have revised the definition to state that the recognized accreditation body is authorized to LAAF-accredit laboratories under this subpart. This change aligns with our overall revisions to terminology throughout the rule.

Representative Sample

We proposed to define “representative sample” to mean “a sample that accurately, to a scientifically acceptable degree, represents the characteristics and qualities of the food product or environment the sample was collected from.”

(Comment 21) Several comments contend that the proposed definition of “representative sample” is vague and impractical. Some comments suggest we clarify that determining whether a sample is “representative” involves an assessment of various factors. Others suggest that FDA clarify the Agency's expectations regarding “representative sample” by specifying sampling protocols within import alerts or including specific procedures and sampling plans for different foods and analyses within the final rule. Some comments suggest the addition of a definition for “representative sampling,” based on the concern that if sampling is not performed appropriately, results may be invalidated.

Some comments specify that the phrase, “to a scientifically acceptable degree” is difficult to understand and vague; these comments suggest that we replace the phrase, “to a scientifically acceptable degree,” with the phrase, “based on a scientific risk-based rationale.” These comments also suggest we add a second sentence to the definition to explain that the suggested phrase, “includes consideration of the environment, food matrix, and analyte of interest, among other factors.”

(Response 21) We agree that whether a food testing sample is representative depends on a variety of factors. Relevant factors include what is being sampled, the population from which the sample is taken, the dispersion pattern of potential adulterants, and adherence to any time and temperature controls, to name just a few. We also appreciate the desire for clarity expressed in the comments suggesting that we specify sampling protocols for the samples that will be tested under this final rule. However, the purpose of defining “representative sample” in this subpart is not to prescribe how to achieve a representative sample either generally or specifically for the testing conducted under this program. Instead, it is to accurately communicate the concept of a representative sample. We considered altering the definition, but because every food product and environmental testing circumstance is slightly different, and as already noted, there are many relevant factors that also vary, our attempts to add specificity to the definition resulted in unnecessarily complex language or the introduction of some inaccuracy. Accordingly, although we understand that some comments describe the proposed definition as vague and impractical, we are retaining it with limited changes because we conclude that it broadly satisfies the purpose for which it was created. We also consider the definition to be similar to and consistent with definitions that are accepted nationally and internationally. (See,

e.g.,

Codex Alimentarius Commission, General Guidelines on Sampling document CAC/GL-50-2004, § 2.2.3: “A representative sample is a sample in which the characteristics of the lot from which it is drawn are maintained. It is in particular the case of a simple random sample where each of the items or increments of the lot has been given the same probability of entering the sample” (Ref. 10).

Some comments suggest that the proposed phrase, “to a scientifically acceptable degree,” is difficult to understand and vague, and suggest instead the phrase, “based on a scientific risk-based rationale.” We agree that the proposed phrase could be improved. However, we do not believe the proffered alternative phrase is the best choice, because it would not always be applicable and also, is less common in the laboratory industry and therefore not widely understood. Instead, we have replaced “to a scientifically acceptable degree,” with, “to a statistically acceptable degree,” which we believe communicates with more precision than the proposed phrase the need for samples to be selected based on a statistical sampling design. A sample that represents the whole to a statistically significant degree will yield information about the average composition of the whole, and therefore enable valid, accurate test results.

We decline the suggestion to add a second sentence to the definition to explain the phrase at issue but have already agreed with the concept it expressed, which is that determining whether a sample is representative involves considering a host of varying factors. We also decline the suggestion to add a definition of “representative sampling,” to this subpart. Although we certainly agree that sampling techniques are critical to obtaining a representative

sample, this final rule does not set standards for those techniques and therefore our discussion of them is not so extensive as to justify the need to define the term.

On our own initiative, we also made grammatical changes to this definition.

See our discussion of § 1.1149 below for additional information on sampling requirements and resources.

Sampler

We proposed to define “sampler” as an individual or individuals who perform sampling.

(Comment 22) A few comments disagree with the proposed definition of “sampler,” and state that a sampler may also be an entity (for example, in the case of laboratories that are commercially liable for the performance of the persons collecting the samples). These comments suggest that FDA include definitions for both “sampler” (an entity) and “sample collector” (individual(s)) within the final rule to clarify this distinction.

(Response 22) We agree that it would be clearer to use two distinct terms throughout the rule regarding activities related to sampling. First, we have clarified the definition of the term, “sampler” to mean an individual who collects samples. Second, we have added a new term, “sampling firm,” which we define as an entity that provides sampling services. Accordingly, we have revised the final rule to use the term, “sampling firm” where appropriate.

Scope of Accreditation

We proposed to define this term to refer to the methods of analysis for which the laboratory is accredited. The proposed definition went on to state that “[r]eferences in this subpart to accreditation `in-whole' refers [sic] to all methods in the accredited laboratory's scope of accreditation and references to accreditation `in-part' refers [sic] to only certain methods in the accredited laboratory's scope of accreditation.” 84 FR 59452 at 59502. We received no comments on this proposed definition; however, we have revised the proposed term and definition to be consistent with our terminology changes throughout the final rule. The term has been revised to “scope of LAAF-accreditation” and the definition of the term has been revised to refer to “. . . the methods of analysis for which the laboratory is LAAF-accredited.”

We have omitted the proposed second sentence in the definition which removes the terms, “in-whole” and “in-part.” Instead, in the final rule we generally employ the construct that changes in LAAF-accreditation relate to specific methods, or apply to all methods, within a laboratory's scope of LAAF-accreditation. Additionally, in the final rule, to better align with the ISO/IEC conformity assessment paradigm, we consistently use the word, “withdraw” to refer to the action a recognized accreditation body takes to remove all methods within the laboratory's scope of LAAF-accreditation, and we use the phrase, “reduce the scope of LAAF-accreditation” to refer to recognized accreditation body actions which remove only certain methods from the laboratory's scope of LAAF-accreditation.

Additional Definitions

On our own initiative, we have included a definition for the term “street address” which appears throughout the final rule. We define the term to mean the full physical address, including the country. We go on to clarify that, for purposes of this rule, a post office box number alone is insufficient; however, a post office box number may be provided in addition to the street address.

We received comments requesting that we include and define additional terms in the final rule. We address these comments below.

(Comment 23) Multiple comments suggest adding a definition for “identified or suspected food safety problem,” stating that doing so would help to clarify when it would be necessary to use a LAAF-accredited laboratory for testing.

(Response 23) For the reasons stated in the preamble to the proposed rule, we decline the recommendation to include a specific definition for “identified or suspected food safety problem” (see 84 FR 59452 to 59462). Instead, we proposed codifying the specific circumstances in which use of a LAAF-accredited laboratory would be required under this subpart. As discussed below in section V.D, we have revised some of the circumstances in response to public comments and have added additional discussion in the preamble.

(Comment 24) Some comments suggest adding definitions for “quality assurance” and “raw data,” stating that similar terms are used by other programs, entities, and regulations—such as FDA's Good Laboratory Practice for Nonclinical Laboratory Studies at 21 CFR part 58—that may serve as a basis for developing a definition under this subpart.

(Response 24) We decline to add definitions for these terms to the final rule.

Quality assurance is a critical pursuit that must undergird both recognized accreditation body and LAAF-accredited laboratory processes. Indeed, we consider the integral nature of quality assurance in ISO/IEC 17011:2017 and ISO/IEC 17025:2017 to be among the standards' greatest strengths (Ref. 2, Ref. 3). In this final rule we are establishing requirements consistent with our perspective that quality assurance must be nurtured (

e.g.,

incorporation of the corrective action process for both recognized accreditation bodies and LAAF-accredited laboratories, submission by recognized accreditation bodies of their internal audit reports, proficiency test requirements for each method within the laboratories' scope of LAAF-accreditation at least every 12 months). Nevertheless, we decline the suggestion to define “quality assurance” in this subpart because we conclude a definition is neither necessary nor would it meaningfully add to the final rule. We prefer instead to include in our standards provisions that will require the quality assurance processes and actions we deem necessary for this program.

We note that the term, “quality assurance” appeared in § 1.1148 of the proposed rule (“What quality assurance requirements must accredited laboratories meet?”). In the final rule, we have omitted the specific section regarding quality assurance requirements and incorporated those requirements into § 1.1138, which addresses the eligibility requirements for LAAF-accredited laboratories.

The term, “raw data” is not used so extensively in the final rule as to warrant a definition. In fact, it only appears once in the codified text, in § 1.1152(d)(8), where we require as part of a full analytical report, “[a]ll original compilations of raw data secured in the course of the analysis.” We explain the term in two ways. First, section 1.1152(d)(8) includes some examples of raw data, and second, in our discussion of that provision at Response 119, below, we have expounded on our thinking regarding this requirement. We consider these forms of explanation to be sufficient in the context of this subpart.

(Comment 25) Some comments state that the term, “specific major food testing discipline” is used throughout the proposed rule and suggest that a definition for the term be added to the regulation for additional clarity.

(Response 25) We included the term, “specific major food testing discipline” in proposed § 1.1152(d) regarding permission to submit abridged

analytical reports. To clarify the term, we have included detail in the final rule at § 1.1153(a) regarding the three major food testing disciplines under this rule for purposes of submitting abridged analytical reports. We identified these in the preamble to the proposed rule regarding § 1.1152(d) (see 84 FR 59484 (Nov. 4, 2019)) using slightly different terms: “microbiology, chemistry, and physical (filth).” In the final rule at 21 CFR 1.1153(a), we have codified the specific major food testing disciplines that will be used to categorize analytical reports for purposes of determining permission to submit abridged analytical reports as “biological, chemical, and physical.”

3. Who is subject to this subpart (§ 1.1103)?

Proposed § 1.1103 listed the entities subject to the subpart: recognized accreditation bodies, entities seeking to become recognized accreditation bodies, LAAF-accredited laboratories, entities seeking to become LAAF-accredited laboratories, and owners and consignees who are required to use LAAF-accredited laboratories for the food testing under this program.

We have made minor changes throughout this section to reflect revised program terminology. Specifically, we have modified the term, “accreditation” to “LAAF-accreditation” in this section and throughout the rule. Additionally, we have made minor editorial changes on our own initiative to improve clarity. Comments regarding this section are discussed below.

(Comment 26) Some comments request clarification of which owners and consignees will be covered by this final rule, stating that there may be multiple owners and consignees in the context of imported food.

(Response 26) FDA-regulated products imported into the United States must comply with the same FDA laws and regulations that apply to domestic products. Entries are submitted to U.S. Customs and Border Protection which then refers entries of FDA-regulated products to FDA for review. Imported items may not be distributed into commerce until FDA has determined admissibility.

If FDA detains a food product at the border under section 801(a) of the FD&C Act because the food is or appears to be adulterated or misbranded, but FDA has not yet refused admission, the owner or consignee of the food may introduce testimonial evidence that the food is admissible. Owners and consignees often engage laboratories to test the food and submit to FDA the results of the testing, as testimony to support admission. If FDA determines that the food testing results are valid and that they demonstrate the detained product does not violate the FD&C Act, FDA will release the food from detention and allow it to proceed into the United States. The testing of detained product at the direction of such owners and consignees is covered by this final rule (see § 1.1107(a)(4)).

The DWPE procedure allows FDA to detain an imported product without physically examining it at the time of entry. FDA employs the DWPE procedure when there is a history of product that violates or appears to violate the FD&C Act, or when other information indicates that future entries may be violative. Import alerts inform FDA staff and the public that we have enough evidence to allow for DWPE of particular products. Testing to support removal from an import alert is also covered by this final rule (see § 1.1107(a)(5)). For more information on FDA's import program generally see

https://www.fda.gov/industry/import-program-food-and-drug-administration-fda;

for more information on DWPE, see

h

ttp

s

://www.fda.gov/media/71776/download.

It is true that for a particular food shipment or entry being offered for import into the United States, multiple parties may be considered owners and/or consignees of the entry or of particular products within that entry (

i.e.,

line items or lines). However, there is generally only one importer of record for each entry,

2

and it is the importer of record that is ultimately responsible for ensuring that the product(s) complies with the FD&C Act and implementing regulations at the time of entry. (See § 1.83(a), where the term, “owner or consignee” is defined for the purposes of articles offered for import.) The importer of record may negotiate or contract with another party such that the other party agrees to engage the laboratory to test the product. Such arrangements are purely between the parties to the shipment; at the end of the day the importer of record remains the party ultimately responsible for the compliance of that entry and therefore is ultimately responsible for amassing any testimonial evidence (

e.g.,

test results and associated analytical documentation) in support of admission of the food.

2

There may not be an importer of record for some informal entries. (Informal entries, as defined by U.S. Customs and Border Protection regulations, are usually valued at less than $2,500 (value subject to change) (19 CFR 143.21), and usually do not require a bond. Some products are restricted from informal entry (for example, high risk products), regardless of value.) For such shipments that are not accompanied by an importer of record when making entry, the owner or consignee of the line(s) will serve as the responsible party when presenting evidence to FDA in support of admission of the food.

D. Comments Regarding General Requirements

Table 4—Revisions to General Requirements

Final rule

Proposed rule

Notes

§ 1.1107 When must food testing be conducted under this subpart?

§ 1.1107 Under what circumstances must food testing be conducted under this subpart by an accredited laboratory?

Revised section title to simplify language and incorporate revised terminology.

§ 1.1108 When and how will FDA issue a directed food laboratory order?

§ 1.1108 When and how will FDA issue a food testing order?

Revised section title to reflect revised terminology.

§ 1.1109 How will FDA make information about recognized accreditation bodies and LAAF-accredited laboratories available to the public?

§ 1.1109 How will FDA make information about recognized accreditation bodies and accredited laboratories available to the public?

Revised section title to reflect revised terminology.

§ 1.1110 What are the general requirements for submitting information to FDA under this subpart?

N/A

New section which consolidates requirements from throughout the proposed rule.

1. When must food testing be conducted under this subpart (§ 1.1107)?

Proposed § 1.1107(a) stated that food testing must be conducted under this subpart whenever food testing is conducted by or on behalf of an owner or consignee in any of the following five circumstances: (1) In response to explicit testing requirements that address an identified or suspected food safety problem in existing FDA regulations covering sprouts (21 CFR 112.146(a), (c) and (d)), shell eggs (§§ 118.4(a)(2)(iii), 118.5(a)(2)(ii), 118.5(b)(2)(ii), 118.6(a)(2), 118.6(e)), and bottled drinking water (§ 129.35(a)(3)(i) (21 CFR 129.35(a)(3)(i))) (regarding the requirement to test five samples from the same sampling site that originally tested positive for

Escherichia coli

(

E. coli

)); (2) as required by FDA in a directed food laboratory order (issued under § 1.1108 of this rule); (3) to address an identified or suspected food safety problem and presented to FDA as part of evidence for a hearing under section 423(c) of the FD&C Act (21 U.S.C. 350

l

) prior to the issuance of a mandatory food recall order, as part of a corrective action plan under section 415(b)(3)(A) of the FD&C Act (21 U.S.C. 350d) submitted after an order suspending the registration of a food facility, or as part of evidence submitted for an appeal of an administrative detention order under section 304(h)(4)(A) of the FD&C Act (21 U.S.C. 334(h)(4)(A)); (4) in support of admission of an article of food under section 801(a) of the FD&C Act; and (5) to support removal from an import alert through successful consecutive testing.

Section 1.1107(b) of the proposed rule stated that when food testing is conducted under paragraph (a), analysis of samples must be conducted by a laboratory that is LAAF-accredited for the appropriate method(s). Proposed paragraph (c) stated the requirement for food testing on articles of food offered for import into the United States to be conducted after the articles have arrived in the United States unless FDA has provided prior written authorization to the owner or consignee that a sample(s) of the article(s) taken prior to arrival in the United States is or would be representative of the article(s) offered for import.

We revised the proposed rule section title, “Under what circumstances must food testing be conducted under this subpart by an accredited laboratory?” to “When must food testing be conducted under this subpart?” in the final rule. We have made changes throughout this section to incorporate revised terminology. We also have made non-substantive revisions to paragraph (a)(2) (to add the word, “issued”), to paragraph (a)(3) to add an inadvertently omitted word (“of”), and to paragraph (c) to improve clarity and readability. Comments regarding this section are discussed below.

(Comment 27) We received several comments regarding the proposed policy to allow all testing under this subpart to be conducted “by or on behalf of an owner or consignee.” Some comments contend that laboratories operated by owners or consignees (“in-house” laboratories) should be ineligible to conduct some or all tests described in § 1.1107. Other comments voice agreement with the proposal.

(Response 27) After considering the comments in light of the statute, we are retaining the proposed policy such that in-house laboratories may become LAAF-accredited to conduct any or all the testing described in § 1.1107 as long as those laboratories meet all the laboratory requirements of this subpart. Please see the discussion of this issue in Response 101 where we address the general eligibility of these laboratories, as well as the impartiality and conflict of interest requirements contained in § 1.1147.

(Comment 28) We received a few comments asking us to clarify the foods to which the testing requirements in the final rule will apply. Some of these comments ask whether any commodities would be exempt from the final rule and state that seafood, juice, and low-acid canned foods are exempt from certain requirements of the “Current Good Manufacturing Practice, Hazard Analysis, and Risk-based Preventive Controls for Human Food” (preventive controls for human food) regulation (part 117 (21 CFR part 117)). Other comments inquire whether the final rule would apply to any commodities other than sprouts, shell eggs, and bottled drinking water.

(Response 28) Proposed § 1.1107(a) described the specific circumstances under which food testing would need to be conducted under this subpart by a LAAF-accredited laboratory. Sprouts, shell eggs, and bottled drinking water are the only commodities for which specific testing requirements contained in existing regulations are currently covered by the final rule (see § 1.1107(a)(1)(i) through (iii)). The remaining circumstances in § 1.1107(a) could require food testing under this subpart for any food or environment within FDA's jurisdiction. We note that hazards addressed by hazard analysis and critical control point (HACCP) regulations for seafood (21 CFR part 123) and juice (21 CFR part 120), and those addressed by regulations for low-acid canned food (21 CFR part 113), are exempt from certain requirements of the preventive controls for human food regulation because those commodities and hazards are covered by commodity-specific HACCP or other regulations that predate the preventive controls for human food regulation. Seafood, juice, and low-acid canned foods are not exempt from this final rule. If seafood, juice, low-acid canned foods, or any article of food or environment within FDA's jurisdiction are covered by any of the circumstances described in § 1.1107(a)(2) through (5), then food testing must be conducted under this subpart by a LAAF-accredited laboratory. For a discussion of program implementation, see Response 14.

(Comment 29) Some comments agree with our proposal regarding the scope of testing that would be covered by the final rule. Some comments express alignment with the general notion of FDA requiring the use of LAAF-accredited laboratories in circumstances where heightened food safety concerns exist. Other comments support the proposed requirement that testing prescribed by certain explicit testing requirements in FDA regulations to address an identified or suspected food safety problem should be covered by this final rule. Specifically, some comments support the inclusion of the bottled drinking water testing required in § 129.35(a)(3)(i) and agree that other bottled drinking water testing required by FDA regulations does not constitute testing in connection with an “identified or suspected food safety problem” and therefore was properly excluded from coverage in the proposed rule.

(Response 29) Section 422 of the FD&C Act prescribes several circumstances in which testing must be conducted by a LAAF-accredited laboratory. First, section 422(b)(1)(A)(i) of the FD&C Act requires testing under this subpart to be conducted, “in response to a specific testing requirement under this Act or implementing regulations, when applied to address an identified or suspected food safety problem.” As discussed in the proposed rule, we proposed to interpret section 422(b)(1)(A)(i) to apply to provisions of the FD&C Act or its implementing regulations that explicitly require food testing. 84 FR 59452 at 59462. We identified nine explicit testing requirements in our regulations that we tentatively concluded address an identified or suspected food safety problem because each of those testing requirements was a followup test after a

routine test indicated the presence of a pathogen or indicator organism (

i.e.,

an organism that indicates conditions in which an environmental pathogen may be present). For example, § 118.4(a)(2)(i) of our shell egg safety regulation requires an environmental test for

Salmonella

Enteritidis when the pullets are 14 to 16 weeks of age. If the environmental test is positive, § 118.4(a)(2)(iii) requires shell egg testing to commence within 2 weeks of the start of egg laying (unless the eggs are diverted to treatment, see § 118.6(a)(2)). We tentatively concluded that the followup shell egg testing would be covered by the rule, but the initial environmental testing would not. Section 422(b)(1)(A)(i) of the FD&C Act is implemented in § 1.1107(a)(1) of this final rule. For a discussion of FDA's interpretation of “identified and suspected food safety problem,” see Response 35.

Section 422(b)(1)(A)(ii) of the FD&C Act requires testing to be conducted under this subpart, “as required by the Secretary, as the Secretary deems appropriate, to address an identified or suspected food safety problem.” Section 422(b)(1)(A)(ii) of the FD&C Act is implemented in § 1.1108 of this final rule, which addresses the directed food laboratory order. (For discussion of the directed food laboratory order, see Comment 41 through Comment 56 and Responses, below.) Section 422(b)(1)(A)(ii) of the FD&C Act also authorizes § 1.1107(a)(3) of this final rule, which requires that food testing be conducted under this program when it is conducted to address an identified or suspected food safety problem and is presented to FDA in three administrative procedural settings: As part of evidence for a hearing under section 423(c) of the FD&C Act prior to the issuance of a mandatory recall order, as part of a corrective action plan under section 415(b)(3)(A) of the FD&C Act submitted after an order suspending the registration of a food facility, or as part of evidence submitted for an appeal of an administrative detention order under section 304(h)(4)(A) of the FD&C Act.

Section 422(b)(1)(B)(i) of the FD&C Act requires testing to be conducted under this subpart, “in support of admission of an article of food under section 801(a).” Section 422(b)(1)(B)(i) of the FD&C Act is implemented in § 1.1107(a)(4) of this final rule. Section 422(b)(1)(B)(ii) of the FD&C Act requires testing to be conducted under this subpart when such testing is to support removal from an import alert through successful consecutive testing, and is implemented in § 1.1107(a)(5) of this final rule.

We appreciate those aspects of comments that express support for the proposed testing provisions.

(Comment 30) Some comments note that there have been foodborne illnesses associated with shell eggs produced at farms with less than 3,000 laying hens. These comments also note that food safety recalls associated with shell eggs, including from cage-free and free-range egg farms that have less than 3,000 laying hens, affect all egg farms. In the view of these comments, FDA's egg safety rule should therefore not exclude shell egg producers with less than 3,000 laying hens, and all egg farms regardless of size should be subject to this rule for the testing described in § 1.1107(a)(1)(ii).

(Response 30) This final rule requires use of a LAAF-accredited laboratory for certain followup tests that already are required by other food safety regulations (§ 1.1107(a)(1)). Because shell egg farms that have less than 3,000 laying hens are exempt from the egg safety rule, such farms are not subject to this final rule for the egg safety rule testing that falls within the scope of this subpart.

(Comment 31) Some comments opine that our use of the term, “corrective action testing” with respect to followup testing in response to an identified or suspected food safety problem appears to mean something different than it does in the world of conformity assessment. These comments assert that for conformity assessment purposes, “corrective action” means that a laboratory takes an “action to eliminate the cause of a nonconformity and to prevent recurrence;” these comments cite ISO/IEC 9001.

(Response 31) In the proposed rule, we used the term, “corrective action” to refer to actions taken by a conformity assessment entity in response to a deficiency (see,

e.g.,

84 FR 59452 at 59491 (“the probation notice would either inform the laboratory that the laboratory has a specified time period to take corrective actions specified by FDA[,] or request that the laboratory submit a corrective action plan to FDA for FDA's approval that identifies the corrective actions it will take to address deficiencies identified”). In the proposed rule, we also used the term, “corrective action” to describe followup activities undertaken by a food manufacturer or processor after product or environmental testing indicates the presence of a pathogen or indicator organism (84 FR 59452 at 59455).

We understand why comments express the view that it may have been confusing for the term, “corrective action” to mean two different things in the proposed rule. In addition, in the proposed rule, we could have been more precise in our use of the term, “explicit corrective action testing” to describe testing covered by section 422(b)(1)(A)(i) of the FD&C Act. Section 422(b)(1)(A)(i) directs this program to cover testing “in response to a specific testing requirement under [the FD&C Act] or implementing regulations, when applied to address an identified or suspected food safety problem.” Not all the testing described by this statutory language may be properly categorized as corrective action testing, (

e.g.,

the sprouts environmental tests at 21 CFR 112.146(c) are considered verification tests within the sprouts regulatory framework; see § 1.1107(a)(1)(i)).

3

To improve clarity and precision, we use the phrase, “explicit followup testing” in the final rule to mean the testing that we have determined will be subject to this subpart under our section 422(b)(1)(A)(i) authority.

3

For more information on sprouts environmental testing, see the “Compliance with and Recommendations for Implementation of the Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption for Sprout Operations” draft guidance, available at

https://www.fda.gov/regulatory-information/search-fda-guidance-documents/draft-guidance-industry-compliance-and-recommendations-implementation-standards-growing-harvesting.

For the foregoing reasons, including to minimize risk of confusion and to improve the final rule, we generally reserve use of the term, “corrective action,” to the conformity-assessment context, in this document. Exceptions include discussion related to the preventive controls regulations; see Comment and Response 37. For clarity we have added the following definition of “corrective action” to § 1.1102: “

Corrective action

means an action taken by an accreditation body or laboratory to investigate and eliminate the cause of a deficiency so that it does not recur.” Relatedly, in §§ 1.1121, 1.1131, and 1.1161 of the final rule, we have added references to the specific sections of the relevant ISO/IEC standard to clarify the process a recognized accreditation body or LAAF-accredited laboratory must take to address deficiencies through corrective action.

(Comment 32) In the proposed rule, we described the circumstances under which testing of imported food would be subject to the requirements of this final rule. In brief, we proposed that an owner or consignee whose entry has been detained because the food is or appears to be adulterated or misbranded must use a LAAF-accredited laboratory to conduct the food testing used as testimonial evidence supporting admission to the United States. The

other import testing that we proposed to cover in this final rule is testing to support the removal of food from import alert through successful consecutive testing. Import alerts inform FDA's field staff and the public that the Agency has enough evidence to allow for DWPE of products that appear to be in violation of FDA's laws and regulations.

Some comments express appreciation that the proposed rule included information on when imported foods would need to be tested. Some comments support our proposal to require the use of a LAAF-accredited laboratory for testing conducted to support removal from import alert. These comments endorse the portion of the proposed rule preamble that discussed the importance of reliable testing of imports and indicate that in the past, food commodities subject to import alert have caused multiple foodborne illness outbreaks. These comments state that although it will take many tools and approaches to ensure the safety of imported foods, reliable testing is a critical component of a successful strategy.

(Response 32) With appreciation for these supportive comments, we confirm that the import-related circumstances under which food testing is required by this subpart in the proposed rule remain unchanged in the final rule: Testing in support of admission of an article of food under section 801(a) of the FD&C Act (§ 1.1107(a)(4)) and testing to support removal from an import alert through successful consecutive testing (§ 1.1107(a)(5)).

(Comment 33) Some comments express confusion about when this final rule would apply and asked when the requirements of the final rule would apply to regulatory feed testing laboratories.

(Response 33) A regulatory feed testing laboratory may choose to seek LAAF-accreditation to conduct testing under this subpart. If animal food were the subject of testing required to be conducted under this program (

i.e.,

the subject of food testing under § 1.1107(a)(2) through (5)), then an owner or consignee would need to use a LAAF-accredited laboratory to conduct the test. For a discussion of program implementation, see Response 14.

(Comment 34) Some comments express the erroneous understanding that the laboratory accreditation final rule would apply only when food testing is conducted in a food manufacturing or processing facility. These comments express the concern that adulteration may occur after the food leaves the production facility, in which case testing conducted during production is outdated and inaccurate, and potentially masks a food safety problem.

(Response 34) We first clarify that the testing covered by this rule is not limited to testing in a food manufacturing or processing facility. Certain testing at farms is also covered; for example, § 1.1107(a)(1)(ii) describes shell egg testing, and those eggs originate on a poultry farm. In addition, this rule covers a significant number of tests of imported food (§ 1.1107(a)(4) and (5)). Because FDA agrees that adulteration may occur while food is in transit, the final rule generally requires imported food products subject to this final rule to be sampled and tested after the food has arrived in the United States. (See § 1.1107(c) and Response 40 for more on this topic.) Thus, testing of imported food subject to this final rule generally will occur at or near the U.S. border.

FDA also has other tools to address adulteration that occurs outside of production establishments, including another FSMA regulation, the “Sanitary Transportation of Human and Animal Food” regulation (part 1, subpart O), which requires shippers, carriers by motor vehicle or rail vehicle, receivers, and other persons engaged in the transportation of food, to use sanitary transportation practices to ensure that the food is not transported under conditions that may render the food adulterated.

(Comment 35) In the preamble to the proposed rule, we discussed considerations in our interpretation of the phrase, “identified or suspected food safety problem,” which appears in section 422(b)(1)(A)(i) and (ii) of the FD&C Act and is therefore important in determining which testing will be covered by this subpart. Among other things, we explored other uses of similar phrases elsewhere in FSMA. We tentatively concluded that an “identified food safety problem” could be present when a specific article of food violates a provision of the FD&C Act that relates to food safety. We tentatively concluded that a “suspected food safety problem” typically would have a basis in fact about a particular article of food (

e.g.,

a lot or batch) or food production environment (

e.g.,

a specific facility). We reasoned that the requisite suspicion would not be satisfied by the common or usual characteristics of food (

e.g.,

whether a food is considered “high risk”) or the manner in which the food is typically produced. We tentatively concluded that the routine product testing and environmental monitoring requirements required by the preventive controls for human food regulation (see § 117.165(a)(2) and (3), respectively), are not conducted to address a suspected (or identified) food safety problem, because this testing is conducted to verify the implementation and effectiveness of preventive controls (“verification testing”) and not because a food safety problem is suspected or identified. 84 FR 59452 at 59462. This same tentative conclusion would apply to the routine product testing and environmental monitoring requirements required by the Current Good Manufacturing Practice, Hazard Analysis, and Risk-based Preventive Controls for Food for Animals (preventive controls for animal food) regulation (§ 507.49(a)(2) and (3) (21 CFR 507.49(a)(2)) and (3), respectively).

In the proposed rule we explained that, in the preventive controls for human food regulation, FDA indicated that an “unanticipated food safety problem” could occur where a preventive control is not properly implemented, including where a pathogen or indicator organism is detected during routine product or environmental testing (verification testing). In the proposed rule we tentatively concluded that, depending on the circumstances, a routine test that indicated the presence of an indicator organism would not necessarily constitute a suspected food safety problem. 84 FR 59452 at 59462.

Some comments dispute our interpretation of “identified or suspected food safety problem.” From their perspective, there is no need for the problem to be particularized to an article of food or a facility. These comments state that the statute does not direct that “an identified or suspected food safety problem,” could only be present in relation to a specific article of food or facility. The comments argue that the appearance of the phrase, “food safety problems” in two FSMA titles that cover multifaceted approaches to food safety (Title I: “Improving Capacity to Prevent Food Safety Problems” and Title II: “Improving Capacity to Detect and Respond to Food Safety Problems”) supports the position that Congress did not intend for the same terms to be read narrowly in the context of section 422 of the FD&C Act. These comments indicate that the economic analysis accompanying the proposed rule estimated that far fewer tests would be subject to the LAAF program under section 422(b)(1)(A) than under section 422(b)(1)(B) of the FD&C Act.

(Response 35) The phrase, “identified or suspected food safety problem,” appears twice in section 422(b)(1)(A) of

the FD&C Act and therefore helps demarcate which testing will be covered by this subpart. The statute does not define either “identified or suspected food safety problem,” or “food safety problem,” nor do those phrases appear elsewhere in the body of FSMA. As referenced above, the phrase, “food safety problem” appears in the FSMA titles: Title I, “Improving Capacity to Prevent Food Safety Problems,” and Title II, “Improving Capacity to Detect and Respond to Food Safety Problems.” Comments urge us to infer from the breadth of the various provisions within each of those two titles, that when Congress used the same phrase in section 422(b)(1)(A) of the FD&C Act, it intended the phrase to be broadly interpreted. However, we cannot impute such an intention to Congress without some indication of that intent in section 422 of the FD&C Act or the legislative history. Indeed, one could reasonably infer the opposite—that from the breadth of the provisions within FSMA Titles I and II, Congress must have intended for the phrase, “food safety problems” to have different meanings in different contexts. In sum, “food safety problem” is not defined in the statute, and thus it falls to FDA to elaborate on its meaning.

In the proposed rule, we looked at other FSMA standards and other FSMA regulations, before making the tentative conclusions described above in Comment 35. We finalize those conclusions without change.

In this vein, we observe that the purpose of routine product and environmental testing under the preventive controls regulations is to verify that preventive controls are consistently implemented and are effective (§§ 117.165(a) and 507.49(a)). Accordingly, such testing does not address an identified or suspected food safety problem, and is not covered by this subpart.

(Comment 36) In the proposed rule, we tentatively concluded that although section 422(b)(1)(B)(i) of the FD&C Act requires testing, “in support of admission of an article of food under section 801(a)” to be conducted under this subpart, it was reasonable not to apply section 422(b)(1)(B)(i) to food testing related to FSVP. We explained that under section 801(a)(3) of the FD&C Act, FDA may refuse admission of an article of food if the food is, or appears to be, adulterated or misbranded. When FDA determines that an article of food is, or appears to be, adulterated or misbranded, we must notify the owner or consignee of our determination, and state the reason(s) for such determination (§ 1.94(a)). FDA must also specify a period of time during which the owner or consignee may introduce testimony relevant to the admissibility of the article of food. Id. Owners or consignees often engage laboratories to test the food and then introduce the test results (along with associated data and analysis) as evidence that the food is admissible. If FDA determines that the sampling methods and testing results are valid and indicate that the article of food does not appear to violate the FD&C Act, FDA will determine that the article of food is admissible, release it from detention, and permit its entrance into the United States. Thus, the focus of section 422(b)(1)(B)(i) of the FD&C Act is the characteristics of an article of food that is pending at the border. Under § 1.1107(a)(4) of this final rule, the testing obtained by the owner or consignee and submitted as testimony to support release of the article of food from detention, must be conducted under this subpart.

FSMA amended the FD&C Act to add section 805, “Foreign Supplier Verification Program,” to require persons who import food into the United States to perform risk-based foreign supplier verification activities for the purpose of verifying that imported food meets applicable U.S. safety requirements. The FSVP regulation, codified in §§ 1.500 through 1.514, specifies the foods and importers to which the FSVP regulation applies and establishes requirements related to supplier verification. Depending on the circumstances, sampling and testing of a food may be an appropriate supplier verification activity. See § 1.506(d)(1)(ii)(B). If an FSVP importer fails to comply with the FSVP regulations for a particular food, that food may be refused admission under section 801(a)(3) of the FD&C Act.

4

However, such refusal is not because the article of food pending at the border is, or appears to be, adulterated or misbranded. Instead, the refusal is a consequence of the importer's failure to comply with its FSVP obligations. Testing the article of food detained at the border in this instance would have no impact on its admissibility under section 801(a)(3) of the FD&C Act, because the detention is due to the characteristics of the importer. In the proposed rule we tentatively concluded that, because the focus of the FSVP provision in section 801(a)(3) of the FD&C Act is entirely different than the focus of the circumstances addressed by section 422(b)(1)(B)(i) of the FD&C Act, it is reasonable not to apply the latter subpart to the testing of food conducted under FSVP.

4

For more information on FSVP, see

https://www.fda.gov/food/food-safety-modernization-act-fsma/fsma-final-rule-foreign-supplier-verification-programs-fsvp-importers-food-humans-and-animals.

Several comments agree with our reasoning regarding testing under FSVP and our proposal that such testing not require use of a LAAF-accredited laboratory. However, other comments disagree, expressing the perspective that as the proposed rule would cover testing to support removal from import alert, it seems more consistent with the FSMA framework to also require testing related to FSVP to be conducted under this subpart. We understand these comments to mean that, because FSVP addresses the safety of food imports, and testing related to import alerts also addresses the safety of food imports, FDA is being inconsistent in covering import alert testing under this subpart, but not testing related to FSVP. These comments further suggest that we not require test results related to FSVP to be sent directly to FDA. The comments do not explain why FSVP tests, which they argue should be subject to this subpart, should nevertheless be excepted from the requirement that all test results under this subpart be submitted directly to FDA.

(Response 36) We disagree that our determinations regarding testing related to FSVP are inconsistent with covering testing to support removal from import alert under this subpart. As an initial matter, the section of the statute authorizing the LAAF program explicitly directs that testing to support removal from import alert be subject to this program, and does not mention FSVP. Further, for the reasons discussed in the proposed rule and briefly described in the comment summary above, we conclude that it is reasonable not to apply section 422(b)(1)(B)(i) of the FD&C Act to food testing related to FSVP. These comments do not explain why FSVP test results would warrant an exception from the § 1.1152(b) requirement to submit all tests results under this program directly to FDA, and as the final rule will not cover testing related to FSVP, the suggestion is inapplicable.

(Comment 37) Some comments agree with our tentative conclusion in the proposed rule that the routine product and environmental testing that occurs pursuant to a preventive controls food safety plan should not require the use of a LAAF-accredited laboratory. Some of these comments encourage FDA to make explicit in the final rule that routine product testing under the preventive control regulations is performed to verify that applied controls have been

effective, and not to address an identified or suspected food safety problem, and therefore is not covered by the laboratory accreditation final rule. Some comments also request that FDA clarify that environmental testing conducted in response to routine environmental monitoring results indicating the presence of a pathogen or indicator organism would not typically be considered testing conducted to address an identified or suspected food safety problem, and would therefore typically fall outside the scope of the laboratory accreditation final rule. According to these comments, facilities should have an opportunity to perform an analysis of the root cause for the environmental positive, take corrective actions and conduct additional testing as needed, before FDA determines that an identified or suspected food safety problem exists and possibly warrants testing by a LAAF-accredited laboratory.

On the other hand, some comments urge FDA to include within the purview of this final rule all food testing required by our regulations, and at a minimum the verification testing and followup testing conducted under the preventive controls and FSVP regulations.

5

Some of these comments contend that FDA has misinterpreted the statute, and claim that section 422(b)(1)(A) of the FD&C Act grants broad discretion to FDA to require use of a participating laboratory in such circumstances.

6

Some comments highlight the language in section 422(b)(1)(A)(ii) of the FD&C Act, which states in relevant part, “as the Secretary deems appropriate, to address an identified or suspected food safety problem,” and argue that such language grants FDA “expansive” authority for the final rule to cover circumstances where either FDA or facilities themselves have identified a food safety hazard and are using testing as part of the approach to address the hazard. Such comments express the view that if FDA does not require more domestic food testing to be conducted under this program, FDA is failing to address food safety problems as Congress intended. Comments encourage the Agency to adopt a broader statutory interpretation of section 422(b)(1)(A) of the FD&C Act even if we do not expand the testing subject to the final rule, so that we may preserve the authority to add more testing to § 1.1107 in the future.

5

Some comments refer to “corrective action testing;” we have changed the phrase to “explicit followup testing.” See Response 31.

6

Some comments imply that the testing required under section 422(b)(1)(A) of the FD&C Act is limited to domestic food production circumstances. However there is nothing in the statute that limits section 422(b)(1)(A) to testing of food produced domestically, and accordingly § 1.1107(a)(1)-(3) of this final rule also refrains from imposing that limitation.

In support of their contentions, some comments offer an example of a Georgia food processing facility that was conducting environmental testing as required by the preventive controls for human food regulation but whose products (boiled eggs) nevertheless caused an outbreak, which, according to the comments, calls into question the accuracy of the test results and the quality of the facility's testing program.

These comments posit that perhaps FDA did not propose to include testing related to the preventive controls or FSVP regulations within the scope of this subpart because testing under those regulations is not always required; depending on the circumstances the facility or importer may find other actions sufficient. These comments find such reasoning unpersuasive because in their view, whenever testing is required as a verification or followup activity under the preventive controls or FSVP regulations, the testing is being conducted “in response” to a regulatory requirement and so is covered by section 422(b)(1)(A) of the FD&C Act.

These comments alternatively posit that perhaps FDA did not propose to cover preventive controls and FSVP testing because this approach might be burdensome for industry. According to these comments, if that is the case, then such concerns could be addressed by providing additional time for implementation; further, any such concerns would be offset by the positive health and economic benefits that they suggest testing would create by preventing outbreaks.

(Response 37) Some comments contend that section 422(b)(1)(A) of the FD&C Act grants FDA broad discretion to require testing to be conducted under this subpart. We address the two subparagraphs of section 422(b)(1)(A) in turn.

Section 422(b)(1)(A)(i) of the FD&C Act

Section 422(b)(1)(A)(i) of the FD&C Act provides that testing must be covered by this program when the testing is conducted, “in response to a specific testing requirement under this Act or implementing regulations, when applied to address an identified or suspected food safety problem.” We discussed our interpretation of “identified and suspected food safety problem” in Response 35, above, and concluded that routine product and environmental testing that occurs pursuant to a preventive controls food safety plan (§§ 117.165(a) and 507.49(a)) is not covered by this subpart. We turn now to our interpretation of the phrase, “in response to a specific testing requirement under this Act or implementing regulations.”

In the proposed rule, we tentatively interpreted, “specific testing requirement under this Act or implementing regulations” to mean that this subpart would cover food testing explicitly required by a statutory or regulatory provision. 84 FR 59452 at 59462. We identified nine testing requirements in FDA regulations that were both explicit and address an identified or suspected food safety problem: Five testing requirements in the egg safety

This text is long and has been trimmed here. Open the source document for the complete record.