← FAT Research Library
📅 Published February 18, 2026
✍️ Dirk Adams
21 min read

← FAT Research Library

📅 Prepared February 18, 2026  |  🔄 Revised May 2, 2026

✍️ Dirk Adams, with the assistance of AI

⌛ 17 min read

FSIS ENFORCEMENT DATA SERIES | PAPER 1 OF 6

USDA-FSIS Recall Data: Collection, Classification, and Significance for Consumer Transparency

How FSIS collects, classifies, and disseminates the federal meat and poultry recall record — and why a comprehensive public dataset remains functionally out of reach for ordinary consumers.

Prepared for publication by Farm Animal Transparency | February 2026 (revised May 2026) | farmanimaltransparency.com

The challenge is not one of data availability but of data usability. The information exists. It is public. It is detailed. But it is locked behind technical formats, bureaucratic nomenclature, and a volume of records that no individual consumer can reasonably navigate.

Abstract

The United States Department of Agriculture’s Food Safety and Inspection Service (FSIS) maintains a comprehensive public record of all recalls and public health alerts involving meat, poultry, and processed egg products sold in the United States. This dataset, spanning from 1996 to the present, represents one of the most complete longitudinal records of food safety enforcement actions available anywhere in the world. Despite its public availability, the recall archive remains functionally inaccessible to the average consumer due to the technical format in which it is published, the volume of records involved, and the absence of consumer-oriented analytical tools. This paper examines how FSIS recall data is collected, classified, and disseminated; describes the regulatory framework that governs the recall process; provides illustrative examples drawn from recent recall patterns; analyzes the scope and duration of available records; and discusses the significance of this data for consumers seeking to make informed decisions about the safety of their food supply. The Farm Animal Transparency (FAT) project integrates this data into searchable, consumer-facing tools designed to bridge the gap between public information and public understanding.

1. Introduction

Food safety in the United States is regulated primarily by two federal agencies: the Food and Drug Administration (FDA), which oversees approximately 78% of the food supply, and the Food Safety and Inspection Service (FSIS) within the U.S. Department of Agriculture (USDA), which regulates meat, poultry, and certain egg products. FSIS derives its food safety authority from three foundational statutes: the Federal Meat Inspection Act (FMIA), the Poultry Products Inspection Act (PPIA), and the Egg Products Inspection Act (EPIA). Together, these laws mandate continuous federal inspection of slaughter and processing operations and provide the statutory basis for enforcement actions when products are found to be adulterated or misbranded.

When an FSIS-regulated product enters commerce and is subsequently found to be adulterated or misbranded, a recall may be initiated. Unlike FDA-regulated products, for which the agency can mandate recalls under the Food Safety Modernization Act (FSMA), FSIS recalls are technically voluntary—they are initiated by the manufacturer or distributor, sometimes at the explicit request of FSIS. However, FSIS retains significant enforcement leverage: if a company refuses to recall a product, the agency has the legal authority to detain products in commerce and, through the USDA Office of the General Counsel, to pursue seizure actions through the Department of Justice.

Each recall event generates a detailed public record. Over nearly three decades, these records have accumulated into a substantial dataset that chronicles the types, causes, and magnitudes of food safety failures across the American meat and poultry supply chain. This paper examines the structure of that dataset, the methodology behind its collection, and the reasons it matters for consumers—even though most Americans have never encountered it in a usable form.

2. How FSIS Recall Data Is Collected

2.1 Triggering Events

FSIS becomes aware of potentially adulterated or misbranded products through several channels. The producing establishment itself may self-report a potential hazard, as required by regulation: official establishments must notify FSIS within 24 hours of becoming aware that adulterated or misbranded product has entered commerce. Additionally, FSIS inspection personnel may identify issues during routine in-plant activities, or the agency’s own sampling programs—including microbiological testing for pathogens such as Salmonella, Listeria monocytogenes, and Shiga toxin-producing Escherichia coli (STEC)—may yield positive results that trigger an investigation. Consumer complaints reported through the FSIS Consumer Complaint Monitoring System represent another important detection mechanism. Finally, epidemiological data from local, state, federal, or international health authorities may identify illness clusters linked to FSIS-regulated products.

2.2 The Event Assessment Committee

When a potential food safety concern is identified, FSIS convenes its Event Assessment Committee (EAC), a multidisciplinary body comprising agency scientists, technical experts, field inspection managers, enforcement personnel, policy experts, and communications specialists. The EAC evaluates all relevant information, including applicable statutory requirements, laboratory results, distribution records, and epidemiological data, to determine the agency’s best course of action.

If the EAC recommends a recall, it also classifies the recall based on the relative health risk the product poses to consumers. This classification is a critical element of the public record and determines the level of public notification that follows.

2.3 Recall Classification System

FSIS employs a three-tier classification system to categorize recalls based on the severity of the associated health risk:

Classification Risk Level Definition
Class I High A health hazard situation where there is a reasonable probability that the use of the product will cause serious, adverse health consequences or death.
Class II Medium A health hazard situation where there is a remote probability of adverse health consequences from the use of the product.
Class III Low A situation where the use of the product will not cause adverse health consequences, or the risk is negligible.

Peer-reviewed analysis of FSIS recall data from 2012 through 2023 reveals that approximately 76% of the 1,001 recall incidents during that period were classified as Class I, 20% as Class II, and only 4% as Class III. The predominance of Class I classifications reflects the fact that most recalls involve biological contamination—the presence of pathogens capable of causing serious illness or death—rather than labeling discrepancies or other lower-risk issues.

2.4 Data Elements Recorded

Each recall event in the FSIS system generates a detailed record containing multiple data elements that are preserved in the public archive. These include the recall number and date of issuance, the recall classification (Class I, II, or III), the name and location of the responsible establishment along with its FSIS establishment number, a description of the recalled product or products including package sizes and identifying markings, the reason for the recall (whether biological contamination, undeclared allergens, foreign material, or other causes), the estimated quantity of product subject to the recall measured in pounds, the geographic distribution of the affected product, and information about whether illnesses have been reported in connection with the recalled product. For Class I recalls, FSIS also develops and publishes a list of retail consignees—the specific retail locations that received the recalled product for sale—including each consignee’s name, street address, city, and state.

3. Illustrative Examples from the Recall Record

To demonstrate both the detail and the practical significance of FSIS recall data, the following three examples are drawn from patterns commonly observed in the recall archive. Company names and establishment numbers have been anonymized; all other details reflect the type and scale of events that appear regularly in the public record.

Example A: Pathogen Contamination in Raw Ground Beef

A beef packing establishment in the western United States initiated a Class I recall of approximately 22,900 pounds of raw ground beef products after FSIS sampling detected the presence of Shiga toxin-producing E. coli (STEC) serotype O145. The affected products—packaged as 10-pound chubs of coarse ground beef labeled at 73% lean—had been distributed to retail outlets in multiple states. Each package bore specific case codes and “Use/Freeze By” dates that allowed consumers to identify the recalled product. No illnesses had been reported at the time of the recall announcement.

This example illustrates several key features of the recall process. First, the recall was triggered by FSIS’s own microbiological sampling program—a proactive detection mechanism rather than a reactive response to reported illness. Second, the level of product detail in the recall notice (package size, case codes, date stamps, time stamps) reflects FSIS’s requirement that consumers be given sufficient information to identify affected products in their own refrigerators and freezers. Third, the classification as Class I—the highest risk level—reflects the serious health consequences associated with STEC infection, which can cause hemolytic uremic syndrome and, in vulnerable populations, death.

Example B: Listeria monocytogenes in Ready-to-Eat Poultry

A poultry processing establishment in the southeastern United States recalled approximately 13,700 pounds of ready-to-eat grilled chicken breast fillet products after FSIS routine testing of finished product detected Listeria monocytogenes (Lm). The products had been distributed nationwide to restaurants, institutional food service operations, and retail locations. Because the products were fully cooked and intended for consumption without further preparation, the presence of Lm in finished product represented an acute public health risk, particularly for pregnant women, elderly consumers, and immunocompromised individuals.

This case exemplifies the particular danger posed by pathogen contamination in ready-to-eat (RTE) products. Unlike raw ground beef, which consumers are expected to cook to a safe internal temperature before consumption, RTE products may be consumed directly from the package. The nationwide distribution of the product and its presence in institutional settings—including schools and hospitals—amplified the urgency of the recall. FSIS classified the event as Class I and issued the recall announcement through all available notification channels.

Example C: Undeclared Allergen in Frozen Prepared Foods

A Midwestern establishment producing frozen tamale products initiated a Class I recall of approximately 2,700 pounds of frozen pork tamales after FSIS inspectors observed during production monitoring that the products contained soy—a major food allergen as defined by the Food Allergen Labeling and Consumer Protection Act—that was not declared on the product label. The products had been distributed to regional retail outlets with a nearly one-year shelf life, meaning potentially affected packages could remain in consumers’ freezers for months after the recall announcement.

This example highlights a recall category—misbranding due to undeclared allergens—that has become increasingly prominent in the FSIS recall record. While pathogen contamination dominates the recall archive by weight of product recalled, allergen-related recalls represent a significant and growing share of total recall incidents. For the estimated 32 million Americans with food allergies, the failure to accurately declare allergens on product labels poses a direct and potentially life-threatening health risk. This particular recall was detected not through sampling or consumer complaint, but through direct observation by FSIS in-plant inspection personnel—underscoring the continued importance of the agency’s physical presence in producing establishments.

Comparative Summary of Illustrative Examples

Note: Company names and establishment numbers have been anonymized. Details reflect patterns typical of the FSIS recall record.

Data Element Example A Example B Example C
Recall Class Class I – High Class I – High Class I – High
Product Type Raw ground beef (73% lean) Ready-to-eat chicken breast fillets Frozen pork tamales
Reason E. coli O145 contamination Listeria monocytogenes contamination Misbranding; undeclared allergen (soy)
Quantity ~22,900 lbs ~13,700 lbs ~2,700 lbs
Detection Method FSIS sampling program FSIS routine testing of finished product FSIS inspector production observation
Distribution Multiple states via retail Nationwide to restaurants and institutions Regional retail distribution
Illnesses Reported None at time of recall None at time of recall None at time of recall

4. How FSIS Recall Data Is Analyzed and Disseminated

4.1 Public Notification

FSIS disseminates recall information through multiple channels to reach the broadest possible audience. For Class I and Class II recalls, the agency issues press releases to media outlets in the areas where the affected product was distributed. All recall press releases, including those for Class III events, are posted on the FSIS website, distributed via the agency’s email subscription service, published on the social media platform X (formerly Twitter), posted on the interagency FoodSafety.gov website, and made available through the FSIS FoodKeeper mobile application.

4.2 Annual Recall Summaries

In addition to individual recall announcements, FSIS compiles annual recall summaries that aggregate data across all recall events for a given calendar year. These summaries provide an overview categorized by recall class, reason for recall, and product type. The annual summary datasets, which are published on Data.gov under a Creative Commons CCZero public domain license, cover the period from 2012 to the present. They enable year-over-year comparison and trend analysis across the entire federally inspected meat and poultry system.

4.3 The Recall API

In September 2023, FSIS launched its first public Application Programming Interface (API) for recall and public health alert data, marking a significant advancement in the agency’s data transparency infrastructure. The API provides programmatic access to recall records in JSON format and supports attribute-based querying, enabling software developers to build applications that integrate FSIS recall data into consumer-facing tools and services. The API was built on cloud infrastructure that FSIS began migrating to in January 2020, when the agency moved several mission-critical applications from on-premises legacy systems to cloud-based platforms.

The launch of the recall API aligns FSIS with the broader U.S. open government initiative, which aims to ensure open access to federal data, research, and information. As then-USDA Under Secretary for Food Safety Dr. Emilio Esteban stated at the time, the goal of the API was to allow third-party developers to extend the reach of food safety information far beyond what the agency could accomplish on its own. The Farm Animal Transparency project is precisely the type of consumer-oriented application that the API was designed to support.

5. Duration and Availability of the Data

The FSIS recall archive represents one of the longest-running continuous records of food safety enforcement actions maintained by any federal agency. The archive contains press releases and notification reports issued by FSIS in conjunction with recalls of meat and poultry products from 1996 to the present—a span of approximately thirty years. Quantity-of-product-recovered data has been tracked and published for recalls since September 2005.

The structured annual summary datasets available on Data.gov currently cover 2012 to the present, providing over a decade of machine-readable data suitable for computational analysis. The API, launched in 2023, provides real-time access to current and archived recall records. Taken together, these overlapping data products offer researchers and application developers multiple access points into the same underlying information at varying levels of granularity.

It is important to note that FSIS has continued to expand both the depth and accessibility of its data holdings over time. The agency has stated that it is developing APIs for additional critical datasets beyond recalls, suggesting that the trajectory of data availability is toward greater, not lesser, openness. However, the availability of these datasets is subject to the priorities and funding decisions of the administrations in office, and continued public access should not be taken for granted.

6. The Accessibility Gap: Public Information, Private Understanding

The FSIS recall dataset is, in every legal and technical sense, public information. It is published by a federal agency, hosted on government servers, made available under open data licenses, and explicitly intended for public consumption. The agency has invested meaningfully in expanding access, from email subscriptions to social media postings to the 2023 API launch.

Yet functional accessibility—the ability of an ordinary consumer to extract meaningful, actionable insight from this information—remains severely limited. Several structural factors contribute to this gap.

First, the sheer volume of data is daunting. Over a thousand recall events occurred in the twelve-year period from 2012 to 2023 alone, representing more than 205 million pounds of recalled product. Each event generates a multi-page press release with specific lot codes, establishment numbers, and distribution data. No consumer can reasonably monitor this volume of information in real time or conduct retrospective analysis without specialized tools.

Second, the data lacks consumer-oriented context. Recall records are identified by establishment number rather than by the brand names consumers encounter at retail. A consumer who purchases chicken products from a particular brand may have no way of knowing that brand’s products are produced at an establishment with a history of recall events without first cross-referencing the FSIS Meat, Poultry, and Egg Product Inspection Directory—itself a technical dataset.

Third, the analytical tools FSIS provides are designed for researchers and developers, not for the general public. The API returns data in JSON format. The annual summaries are published as downloadable spreadsheets. Neither format is intuitive for a consumer trying to answer a simple question: Is the company that produced my ground beef one that has had safety problems?

It is this accessibility gap—between the legal availability of critical food safety data and the practical ability of consumers to use it—that the Farm Animal Transparency project was created to address.

7. Significance of FSIS Recall Data

7.1 Patterns in Product Contamination

Analysis of FSIS recall data reveals important patterns in the types and frequencies of food safety failures across the regulated industry. Published research examining the 2012–2023 period found that product contamination was the leading cause of recalls, accounting for 68% of all events. Processing issues contributed 13%, and other reasons—19%. Within the product contamination category, four bacterial species were responsible for the majority of recalled product by weight: Listeria monocytogenes accounted for 32%, Shiga toxin-producing E. coli for 12%, and Salmonella serovars for 9%.

By product type, poultry products accounted for 29% of recall incidents, beef for 23%, mixed animal meat products for 23%, and pork for 22%. However, when measured by weight of product recalled rather than number of incidents, mixed animal products dominated at 52%, followed by poultry at 27% and beef at 16%—suggesting that recalls involving mixed and poultry products tend to involve larger production volumes.

7.2 Establishment Accountability

Perhaps the most significant consumer application of recall data is the ability to identify which specific producing establishments have experienced repeat recall events. Because each recall record includes the FSIS establishment number, longitudinal analysis can reveal patterns of recurring food safety failures at individual facilities. This information, when made accessible through consumer-facing tools, enables purchasers to evaluate the safety record of the establishments that produce the products they buy—a level of supply chain transparency that is otherwise unavailable at the retail level.

The recall dataset thus functions as a de facto public accountability record for the meat and poultry processing industry. While individual recall announcements are designed primarily as immediate consumer safety alerts, the cumulative archive serves a broader function: it documents the safety performance of every federally inspected establishment that has experienced a recall event over the past three decades.

7.3 Informing Regulatory Policy

FSIS recall data also plays a central role in the agency’s own regulatory decision-making. The data informs sampling program design, inspection resource allocation, and the development of performance standards for pathogen control. When recall data reveals emerging trends—such as a rise in recalls attributable to undeclared allergens, or an increase in contamination with specific Salmonella serovars—the agency can adjust its surveillance and enforcement strategies accordingly.

For external researchers and public health professionals, the recall dataset provides an empirical foundation for studying the effectiveness of food safety interventions, the epidemiology of foodborne illness, and the economic impacts of product contamination on both the industry and the healthcare system.

8. The Farm Animal Transparency Approach

The Farm Animal Transparency (FAT) project addresses the accessibility gap described above by transforming raw FSIS enforcement data—including recall records—into searchable, consumer-oriented tools. The project collects and normalizes data from the FSIS recall archive, the recall API, and annual summary datasets, and integrates this information into a unified database indexed by establishment number, company name, product type, and recall reason.

Through the FAT website and its associated tools, consumers can look up specific FSIS-inspected establishments and view their complete recall histories, including recall classifications, reasons, product descriptions, and affected quantities. This capability transforms what was previously a fragmented collection of government press releases into a coherent, establishment-level safety record that any consumer can access without technical expertise.

The FAT project does not editorialize or interpret the data. It presents the public record as published by FSIS, organized in a format that answers the questions consumers actually ask: Has this company had a recall? What was it for? How serious was it? How much product was involved? By providing these answers in plain language and through intuitive interfaces, FAT fulfills the transparency objective that FSIS has articulated in its own data-sharing initiatives—extending the reach of critical food safety information to the consumers who need it most.

9. Conclusion

The FSIS recall dataset is a remarkable public resource—nearly thirty years of detailed, establishment-level food safety enforcement records, maintained by the federal agency responsible for the safety of the American meat and poultry supply. It is comprehensive, well-structured, and increasingly accessible through modern data interfaces. It is also, for the vast majority of American consumers, effectively invisible.

The challenge is not one of data availability but of data usability. The information exists. It is public. It is detailed. But it is locked behind technical formats, bureaucratic nomenclature, and a volume of records that no individual consumer can reasonably navigate. The result is an information asymmetry in which the industry and its regulators possess detailed knowledge about the safety performance of individual establishments, while the consumers who purchase their products do not.

The Farm Animal Transparency project is one effort to correct this asymmetry. By collecting, normalizing, and presenting FSIS enforcement data—beginning with the recall archive examined in this paper—FAT makes the public record genuinely public. In doing so, it honors the principle that informed consumers are the foundation of a functioning food safety system.

This paper is the first in a series of six examining the major categories of FSIS enforcement data integrated by the Farm Animal Transparency project. Subsequent papers will address Salmonella verification testing in poultry (Paper No. 2), E. coli and Salmonella sampling in beef products (Paper No. 3), the Quarterly Enforcement Report (Paper No. 4), Chemical Residue Repeat Violators (Paper No. 5), and Humane Handling enforcement data (Paper No. 6).

References

1. Federal Meat Inspection Act, 21 U.S.C. §§ 601–695.

2. Poultry Products Inspection Act, 21 U.S.C. §§ 451–472.

3. Egg Products Inspection Act, 21 U.S.C. §§ 1031–1056.

4. USDA-FSIS. “Understanding FSIS Food Recalls.” Food Safety and Inspection Service, U.S. Department of Agriculture. Available at: https://www.fsis.usda.gov/food-safety/safe-food-handling-and-preparation/food-safety-basics/understanding-fsis-food-recalls.

5. USDA-FSIS. “Recall API.” Food Safety and Inspection Service, U.S. Department of Agriculture. Available at: https://www.fsis.usda.gov/science-data/developer-resources/recall-api.

6. USDA-FSIS. “Annual Recall Summaries.” Food Safety and Inspection Service, U.S. Department of Agriculture. Available at: https://www.fsis.usda.gov/food-safety/recalls-public-health-alerts/annual-recall-summaries.

7. USDA-FSIS. “FSIS Launches New Data Tool: Recall and Public Health Alert API.” Press Release, September 7, 2023.

8. USDA-FSIS. “Summary of Recall Cases in Calendar Year.” Data.gov, Creative Commons CCZero. Available at: https://catalog.data.gov/dataset/fsis-summary-of-recall-cases-in-calendar-year.

9. Journal of Food Protection (2025). “USDA Food Recalls for the Period 2012–2023 Compared with FDA Regulated Food Recalls Over the Past Two Decades.” Published March 24, 2025.

10. National Agricultural Law Center. “Food Foundations: What are Food Recalls and How do They Work?” University of Arkansas System Division of Agriculture.

11. 9 C.F.R. Part 418 – Recalls.

12. 21 C.F.R. § 7.3(m) – Recall Classification Definitions.

Disclosure: The Farm Animal Transparency project is an independent, non-commercial initiative. It receives no funding from the food industry, USDA, or any entity with a financial interest in the data presented. All data referenced in this paper is derived from publicly available FSIS records. This paper does not constitute legal, medical, or dietary advice.

Stay Informed on Meat Labeling

Share this research:𝕏 Postin Share✉ Email

Discover more from Farm Animal Transparency

Subscribe now to keep reading and get access to the full archive.

Continue reading