This is the accessible text file for GAO report number GAO-02-902 
entitled 'Meat and Poultry: Better USDA Oversight and Enforcement of 
Safety Rules Needed to Reduce Risk of Foodborne Illnesses' which was 
released on September 19, 2002.



This text file was formatted by the U.S. General Accounting Office 

(GAO) to be accessible to users with visual impairments, as part of a 

longer term project to improve GAO products’ accessibility. Every 

attempt has been made to maintain the structural and data integrity of 

the original printed product. Accessibility features, such as text 

descriptions of tables, consecutively numbered footnotes placed at the 

end of the file, and the text of agency comment letters, are provided 

but may not exactly duplicate the presentation or format of the printed 

version. The portable document format (PDF) file is an exact electronic 

replica of the printed version. We welcome your feedback. Please E-mail 

your comments regarding the contents or accessibility features of this 

document to Webmaster@gao.gov.



Report to the Committee on Agriculture, Nutrition, and Forestry U.S. 

Senate:



United States General Accounting Office:



GAO:



August 2002:



Meat and Poultry:



Better USDA Oversight and Enforcement of Safety Rules Needed to Reduce 

Risk of Foodborne Illnesses:



GAO-02-902:



Contents:



Letter:



Results in Brief:



Background:



FSIS Is Not Ensuring that Plants’ HACCP Plans Meet Regulatory 

Requirements:



FSIS’s Lack of Consistent Identification and Documentation of 

Repetitive HACCP Violations Weakens Enforcement:



FSIS Is Not Ensuring That Plants Take Prompt and Effective Corrective 

Action to Return to Compliance with HACCP Requirements after Violations 

Have Been Identified:



Conclusions:



Recommendations for Executive Action:



Agency Comments and Our Response:



Appendix I: Scope and Methodology:



Appendix II: Comments from the U.S. Department of Agriculture:



Appendix III: GAO Contacts and Staff Acknowledgments:



Tables:



Table 1: Percentage of Plants with No Documented HACCP Violations 

during Fiscal Year 2001:



Table 2: Plant Size, Number of Noncompliance Records, and Number of 

Repetitive Violations in 16 Plants during Fiscal Year 2001:



Table 3: Elapsed Time--From Failure of First Set to Passing Third Set-

-at Plants with Second-Set Failures of the Salmonella Performance 

Standard and In-Depth Verification Reviews:



Table 4: Time Line of Events between In-Depth Verification Review and 

Third Set of Salmonella Tests at One Plant:



Abbreviations:



FSIS: Food Safety and Inspection Service



GAO: General Accounting Office



HACCP: Hazard Analysis and Critical Control Point



USDA: U.S. Department of Agriculture:



Letter:



August 30, 2002:



The Honorable Tom Harkin

Chairman

The Honorable Richard G. Lugar

Ranking Minority Member

Committee on Agriculture, Nutrition, and Forestry

United States Senate:



Every year, some meat and poultry products are contaminated with 

microbial pathogens, such as Salmonella and E. coli, that cause 

foodborne illnesses and deaths. To improve the safety of meat and 

poultry products, and in response to recommendations from GAO and the 

National Academy of Sciences, the U.S. Department of Agriculture’s 

(USDA) Food Safety and Inspection Service (FSIS) implemented additional 

regulatory requirements for meat and poultry plants. These requirements 

are intended to ensure that plants operate food safety systems that are 

prevention-oriented and science-based. These systems, called Hazard 

Analysis and Critical Control Point (HACCP) systems, were phased in 

from January 1998 through January 2000 at all meat and poultry 

slaughter and processing plants. As the foundation of the HACCP system, 

plants are responsible for developing HACCP plans that, among other 

things, identify all of the contamination hazards that are reasonably 

likely to occur in a plant’s particular production environment, 

establish all of the necessary steps to control these hazards, and have 

valid scientific evidence to support their decisions. As a result of 

implementing HACCP systems over the past 5 years, plants have accepted 

significant new responsibilities for producing safe products, and FSIS 

has made major changes to the roles and responsibilities of its 

inspection workforce.



FSIS, through its 15 district offices across the country, oversees the 

activities of about 7,500 federal inspectors who review the operations 

of about 5,000 plants subject to the HACCP requirements 

nationwide.[Footnote 1] About 3,400 inspectors are stationed in plants 

along slaughter lines to provide traditional carcass-by-carcass 

inspections using sight, touch, and smell. The remaining 4,100 FSIS 

inspectors oversee HACCP systems in plants. As a part of their 

oversight, inspectors determine if plants are complying with HACCP 

requirements, including the requirement that their plans include the 

following specific components:



* A hazard analysis that identifies all the food safety hazards--

biological, chemical, and physical--that are reasonably likely to occur 

and measures to control them.



* Critical control points in plants’ processes where controls can be 

applied to prevent, eliminate, or reduce a food safety hazard to an 

acceptable level.



* Critical limits (maximum or minimum values) at which the hazard 

is controlled.



* Monitoring requirements to ensure that the measured values are within 

critical limits.



* Corrective actions to be taken if critical limits are violated.



* Verification procedures to ensure that the plants’ HACCP systems 

result in safe products.



* Record keeping procedures for documenting HACCP requirements.



To help verify that plants’ HACCP systems are effectively controlling 

food safety hazards, FSIS inspectors test for the presence of the 

pathogen Salmonella on raw meat and poultry in a series of samples--

referred to as a “set.” FSIS established limits for Salmonella, known 

as “performance standards,” which vary depending on the type of 

product. For example, no more than 1 percent of steer carcasses sampled 

in a set of tests may contain Salmonella. In addition to limits on 

Salmonella, FSIS has established a “zero tolerance” for visible feces 

on carcasses at slaughter plants and considers the disease-causing 

pathogen E. coli 0157:H7 an adulterant that is not allowed in ground 

beef. FSIS also considers Salmonella, E. coli 0157:H7, and Listeria 

monocytogenes adulterants in ready-to-eat products such as hot dogs and 

luncheon meats.



When FSIS inspectors find a violation of the HACCP requirements, they 

document the violation on a “noncompliance record” and advise the 

plant. FSIS writes noncompliance records to document HACCP process 

violations, such as a plant’s failure to document its monitoring of 

temperatures for a cooked product, as well as for violations of the 

rules regarding pathogens. If the plant does not correct the violation, 

FSIS may take an enforcement action, such as detaining the affected 

meat or poultry product; slowing one or more production lines; 

withholding the marks showing that a product has passed USDA 

inspection; or suspending inspection services for one or more products 

or the entire plant. While inspection services are suspended, a plant 

cannot operate. However, FSIS may place a suspension on hold--referred 

to as “in abeyance”--to allow the plant to continue operating while it 

corrects the violation.



In 1999, we reported that weaknesses in FSIS’s training for its 

inspectors affected its ability to ensure consistent and effective 

oversight of HACCP.[Footnote 2] The following year, a USDA Inspector 

General report identified shortcomings in plants’ HACCP plans and 

deficiencies in FSIS’s oversight of HACCP’s implementation.[Footnote 3] 

To help address these problems, FSIS stepped up its inspector training 

and initiated two new review mechanisms:



* Food safety systems correlation reviews, which examine a range of 

inspector practices within FSIS districts to improve the effectiveness 

of inspections.



* In-depth verification reviews of HACCP plans in plants with serious 

safety problems to identify weaknesses in the scientific soundness of 

the plans.



FSIS also introduced consumer safety officers into its workforce with 

the scientific and technical expertise to, among other things, review 

the scientific soundness of HACCP plans.



As you requested, this report (1) assesses whether FSIS is ensuring 

that plants’ HACCP plans meet regulatory requirements, (2) determines 

whether FSIS is consistently identifying repetitive violations of HACCP 

requirements, and (3) assesses whether FSIS is ensuring that plants 

take prompt and effective action to return to compliance after the 

agency has identified HACCP violations. As part of our review, we 

analyzed 1,180 noncompliance records from 16 judgmentally selected meat 

and poultry slaughter and processing plants where FSIS frequently found 

HACCP-related violations in fiscal year 2001. Our sample included 

different sizes of plants located in 10 different FSIS districts across 

the country. We also analyzed files for the 68 HACCP enforcement cases 

that were active in fiscal year 2001 in three FSIS districts, including 

the two districts with the most plants in the country. In addition, we 

analyzed data from the in-depth verification reviews that FSIS 

conducted through the end of calendar year 2001 and the food safety 

systems correlation reviews that it completed by May 2002. Appendix I 

contains the details of our scope and methodology.



Results in Brief:



FSIS is not ensuring that all plants’ HACCP plans meet regulatory 

requirements and, as a result, consumers may be unnecessarily exposed 

to unsafe foods that can cause foodborne illnesses. In particular, 

FSIS’s inspectors have not consistently identified and documented 

failures of plants’ HACCP plans to comply with requirements. For 

example, FSIS’s food safety systems correlation reviews in three FSIS 

districts found that, in about 91 percent of the plants sampled, 

inspectors had failed to document deficiencies in basic requirements 

such as the requirement that plants have adequate documentation to 

support the analysis of hazards in their HACCP plans. In addition, 

although sound science is the cornerstone of an effective HACCP plan, 

FSIS does not expect its inspectors to determine whether HACCP plans 

are based on sound science because inspectors lack the expertise to do 

so. FSIS has made limited progress in reviewing the scientific 

soundness of plants’ HACCP plans. While FSIS’s in-depth verification 

reviews have been useful in identifying numerous scientific weaknesses 

in HACCP plans, only about 1 percent of plants have undergone these 

time-and resource-intensive reviews. Similarly, consumer safety 

officers will improve FSIS’s ability to assess the scientific adequacy 

of plants’ HACCP plans. However, only about 6 percent of the officers 

that FSIS needs are on board, and FSIS managers in two large districts 

expressed concern that it may take years to assess the plans for all 

plants in their districts. Finally, we found that inspectors had not 

documented any HACCP violations in 55 percent of all plants during 

2001; yet, when we showed these data to FSIS officials, they were 

surprised at the large numbers and said the absence of violations was 

unusual. For example, one field official said that if inspectors are 

finding no HACCP violations for an entire year, they may not understand 

their HACCP oversight responsibilities. In August 2002, FSIS told us it 

has developed, and would soon release, a new directive to clarify 

inspectors’ responsibilities and new guidance for its supervisors to 

use to verify that inspectors are, among other things, applying 

appropriate inspection methods. FSIS also told us that it had 

introduced an interactive computer tool for inspectors and others to 

use to strengthen their knowledge of HACCP requirements.



FSIS is not consistently identifying repetitive violations, according 

to our review of 1,180 noncompliance records for fiscal year 2001. This 

has occurred in part because FSIS has not established specific, 

uniform, and clearly defined criteria for its inspectors to use in 

determining when a violation is repetitive. Furthermore, at the 

district level, FSIS officials’ understanding of the criteria to 

consider in determining if a violation is repetitive varied. Also, in 

several instances, inspectors have not fully documented the basis for 

their decisions about repetitive violations on noncompliance records. 

Identifying repetitive violations, and maintaining accurate 

documentation on those decisions, is critical in deciding whether a 

HACCP plan is flawed and/or an enforcement action is needed. Moreover, 

we found that FSIS’s inspection database did not provide summary 

information on repetitive violations, which could help FSIS’s managers 

oversee inspectors’ performance and plants’ compliance with HACCP 

requirements. Summary information should also help FSIS identify common 

problems that may be better addressed by advising the industry to take 

corrective actions instead of plant-by-plant enforcement. FSIS 

officials agreed on the need for consistent criteria for identifying 

repetitive violations and expect to issue a directive with those 

criteria by the end of the calendar year. FSIS told us it has begun 

testing software that will allow its managers to extract summary data 

from the inspection database to help them better identify repetitive 

violations.



Finally, FSIS is not ensuring that plants take prompt and effective 

action to return to compliance after a HACCP violation has been 

identified. For example, FSIS has not consistently ensured that the 

actions that plants have taken were effective in eliminating repetitive 

violations, particularly those relating to “zero tolerance” for visible 

feces. Although plants are required to take corrective action each time 

a violation is cited, the number of repetitive violations in various 

plants--109 in one plant alone--shows that FSIS has not ensured that 

recurring violations were eliminated. FSIS also has not ensured that 

plants have taken immediate action, as required under HACCP rules, to 

meet the Salmonella performance standard. At the plants that failed two 

consecutive sets of tests for Salmonella, an average of 20 months 

elapsed from the date of the failure of the first set until the plants 

completed and passed a third set. Finally, when FSIS suspended 

inspections at a plant, it generally placed those suspensions in 

abeyance--often on the same day. This allowed the plants to operate 

while they took corrective actions. According to FSIS guidance, 

suspensions should not be held in abeyance for more than 90 days. 

However, nearly all the plants that were suspended in the three FSIS 

districts we examined had their suspensions placed in abeyance and were 

allowed to remain in abeyance for an average of 10 months, during which 

time they continued to operate. Moreover, we were generally unable to 

verify the time frames in which plants were expected to complete 

corrective actions or the actual time elapsed before the corrective 

actions were completed because the enforcement case files did not 

contain this information. The longer that FSIS allows plants to remain 

out of compliance with regulatory requirements, the greater the risk 

that unsafe food will be produced and marketed.



We are making several recommendations to the Secretary of Agriculture 

to ensure that (1) FSIS inspectors better ensure that plants’ HACCP 

plans fully meet regulatory requirements, (2) FSIS inspectors and 

district officials have consistent criteria for identifying repetitive 

violations, and (3) plants act promptly and effectively to correct 

violations. In commenting on a draft of this report, USDA agreed with 

our recommendations but believes the report does not fully acknowledge 

FSIS’s progress and efforts to ensure that all plants meet regulatory 

requirements. USDA described a number of actions that FSIS has recently 

taken or is planning to take that are consistent with our 

recommendations. We believe that our report reflects the status of 

FSIS’s ongoing and planned actions. If fully carried out and given 

diligent management attention, these actions could go a long way toward 

addressing the problems we found in FSIS’s oversight and enforcement of 

HACCP in U.S. meat and poultry plants and reducing consumers’ risk of 

foodborne illnesses.



Background:



According to the Centers for Disease Control and Prevention, 

contaminated foods cause an estimated 76 million illnesses in the 

United States each year, including 325,000 hospitalizations and 

5,000 deaths. Illnesses stemming from contaminated meat and poultry 

are responsible for an unknown portion of these illnesses and deaths. 

The Federal Meat Inspection Act and the Poultry Products Inspection Act 

give USDA responsibility for ensuring the safety and wholesomeness of 

meat and poultry products that enter interstate commerce.[Footnote 4] 

There are about 5,000 meat and poultry slaughter and processing plants 

nationwide. According to the American Meat Institute, total meat and 

poultry production in 2000 exceeded 80 billion pounds and sales were 

estimated at more than $100 billion.



In January 1998, FSIS began phasing in HACCP regulatory requirements 

for meat and poultry slaughter and processing plants. Large plants--

those with 500 or more employees--were required to have HACCP systems 

in place by January 1998; small plants--those with 10 to 499 employees-

-by January 1999; and very small plants--those with fewer than 10 

employees or annual sales of less than $2.5 million--by January 2000.



As part of its oversight efforts to verify that plants effectively 

control food safety hazards, FSIS established standards for Salmonella 

in raw meat and poultry products and for visible feces on carcasses in 

slaughter plants.[Footnote 5] For Salmonella, FSIS established separate 

standards for the carcasses of cows/bulls, steers/heifers, market hogs, 

and broiler chickens, as well as for ground beef, ground chicken, and 

ground turkey. When a Salmonella test is scheduled, the FSIS inspector 

should take one sample each day the plant produces the product until a 

set is complete, according to FSIS guidance. The number of samples in 

the set depends on the product and ranges from a low of 51 samples for 

broiler chickens to a high of 82 samples for steers and heifers. On 

July 25, 2002, FSIS issued new, more-detailed guidance on actions the 

agency and plants will take after Salmonella set failures. When a plant 

fails a first set of Salmonella tests, FSIS will, among other things, 

notify the plant in writing of the failure and assess the plant’s HACCP 

procedures. After a second consecutive Salmonella set failure, FSIS 

will notify the plant that it must reassess its HACCP plan to determine 

if changes are needed. After the plant completes its reassessment, FSIS 

will conduct an in-depth verification review, among other things.



With regard to the zero tolerance standard for visible feces, the FSIS 

inspector checks a prescribed number of carcasses on each production 

shift to verify that the plant has successfully eliminated visible 

fecal contamination.[Footnote 6] FSIS also requires plants to test meat 

and poultry carcasses for generic E. coli--bacteria that occur 

naturally in animals’ intestinal tracts--to help ensure that the plants 

are minimizing the risk that harmful bacteria may be on the carcasses.



In December 1999, we reported that FSIS inspectors were confused about 

their authority to request changes to HACCP plans and recommended that 

FSIS clarify and provide FSIS inspectors with additional training on 

their roles, responsibilities, and authorities for reviewing and 

verifying HACCP plans. We also recommended that FSIS review all plants’ 

HACCP plans to verify that plants identify and control, through their 

HACCP plans, all food safety hazards that are reasonably likely to 

occur. In June 2000 USDA’s Office of Inspector General reviewed 57 

HACCP plans from 15 plants nationwide and reported that 14 of the 

plants had at least one incomplete HACCP plan.[Footnote 7] The report 

made recommendations to ensure that hazard analyses were complete and 

all critical control points identified.



In response to these reports, FSIS implemented the following additional 

oversight mechanisms:



* Food safety systems correlation reviews to improve the effectiveness 

of FSIS inspection activities. The reviews examine a range of FSIS 

plant inspection practices using a randomly selected sample of 

10 percent or a minimum of 40 plants (whichever is greater) in an FSIS 

district. From April 2001 to May 30, 2002, FSIS had completed reviews 

in 7 of its 15 districts. Subsequent to each review, FSIS provides 

targeted training to inspectors on the basis of the review’s findings.



* In-depth verification reviews to examine plants’ compliance with 

HACCP plan design and implementation requirements. The reviews examine 

elements of plants’ HACCP plans, such as hazard analysis, critical 

control points, and critical limits, and their implementation of these 

plans. From February 2000 to June 30, 2002, FSIS had completed reviews 

in 57 plants.



* Consumer safety officers have been trained by FSIS in microbiological 

hazards, HACCP plan design, epidemiology, and statistics to, among 

other things, review the scientific basis of HACCP plans. FSIS had 32 

consumer safety officers in its district offices as of May 30, 2002.



FSIS plant inspectors have front-line responsibility for reviewing 

HACCP plans to ensure that they meet basic regulatory requirements. 

They use a “noncompliance record” to document violations of HACCP 

requirements and actions taken by plants to correct the violations. 

These records include the following information:



* A unique record number.



* The date of the violation.



* The oversight procedure that the inspector was performing (e.g., 

assessing the process for grinding meat) when the violation was 

discovered.



* The element of the HACCP system--monitoring, corrective action, 

record keeping, or plant verification--where the violation occurred.



* A written description of the violation.



* The plant management’s written response stating both the immediate 

action to correct the violation and any subsequent action to prevent 

its recurrence.



If actions taken by a plant to correct a problem fail to prevent the 

violation from recurring, the plant is said to have a “repetitive 

violation”--a recurring inability to maintain compliance with HACCP 

requirements. According to HACCP regulations, repetitive violations 

indicate that a plant’s HACCP system is inadequate and that an 

enforcement action may be warranted. An enforcement action can also be 

taken for a single serious violation. FSIS may take the following types 

of enforcement actions:



* A regulatory control action, which includes the retention of product, 

rejection of equipment or facilities, slowing or stopping of production 

lines, or refusal to allow the processing of specifically identified 

product. This action is considered the least burdensome type of 

enforcement action and can be initiated by an FSIS inspector to quickly 

respond to violations that can be easily remedied.



* A withholding action is the inspector’s or district officials’ 

refusal to allow the USDA marks of inspection to be applied to the 

product. This action is used for more serious HACCP violations, such as 

repeated failure to maintain HACCP records adequate for inspectors to 

determine whether or not a product was adulterated. This action may 

affect all products in the plant or only those products produced by a 

particular process. When only a particular process is involved, the 

plant may continue with its other operations, but it may not distribute 

the affected product.



* A suspension is an interruption in the assignment of FSIS inspectors 

and, hence, production, in all or part of a plant. An FSIS district 

manager may suspend inspections when the violation cannot be resolved 

through a withholding action or there is an immediate threat to public 

health. The district manager may put a suspension on hold--”in 

abeyance”--to give the plant time to execute a plan to correct the 

violation and prevent its future recurrence. FSIS guidance recommends 

that suspension not be held in abeyance for more than 90 days.



* A withdrawal of the grant of inspection is the removal of FSIS from 

the plant. Under this rarely used action, taken only by the FSIS 

Administrator, a plant’s products cannot enter interstate or foreign 

commerce.



According to the June 2000 Inspector General’s report, while some 

plants had received numerous noncompliance records for the same 

deficiency, FSIS’s inspectors had no understanding of what number, 

frequency, or nature of deficiencies would constitute a breakdown in 

the system requiring an enforcement action. The report further found 

that FSIS inspectors were unsure when to declare a plant’s corrective 

actions unworkable--a critical step in taking further enforcement 

action.



FSIS Is Not Ensuring that Plants’ HACCP Plans Meet Regulatory 

Requirements:



According to FSIS’s food safety systems correlation reviews, inspectors 

are not consistently identifying and documenting failures of plants’ 

HACCP plans to meet regulatory requirements. Furthermore, FSIS does not 

expect its inspectors to determine whether HACCP plans are based on 

sound science--the cornerstone of an effective plan. While in-depth 

verification reviews examine the scientific aspects of HACCP plans, 

they have been conducted in very few plants, and consumer safety 

officers hired to review the scientific soundness of HACCP plans may 

take several years to assess the plans at all plants. Moreover, 

inspectors in 55 percent of the 5,000 plants nationwide did not 

document any HACCP violations during fiscal year 2001. When we brought 

this information to the attention of FSIS officials, they were 

surprised that so many plants had no HACCP violations for an entire 

year.



Food Safety Systems Correlation Reviews Show That FSIS Inspectors Did 

Not Identify Violations in Plants’ HACCP Plans:



FSIS’s food safety systems correlation reviews show that plants have 

deficiencies in their HACCP plans that FSIS’s in-plant inspectors did 

not identify and document in noncompliance records.[Footnote 8] Through 

May 2002, FSIS conducted food safety systems correlation reviews for 

seven districts and completed reports for six of those districts; it 

plans to complete reviews in the remaining districts by the end of 

fiscal year 2003. These reviews, which examine a random sample of 

plants in a district, compare inspection practices within a district 

to, among other things, better target inspector training.



While examining the findings of the six completed district review 

reports, we found that a significant number of plants had deficiencies 

in their HACCP plans that FSIS inspectors had not identified and 

documented. For example, in about 91 percent of the plants sampled in 

three districts, inspectors had failed to issue noncompliance records 

for deficiencies in basic requirements such as the requirement that 

plants have adequate documentation to support the analysis of hazards 

in their HACCP plans.



* In 26 of 27 plants in one of the three districts, inspectors had not 

issued noncompliance records for HACCP plans that failed to include 

supporting documentation on the food safety hazards that were likely to 

occur. Inspectors also had not issued noncompliance records for 15 

plants with plans that failed to address the three categories of 

hazards (biological, chemical, and physical) at each step of their 

production processes.



* In 10 of 14 plants in another district, inspectors had not issued 

noncompliance records for HACCP plans that had either not sufficiently 

documented decisions, not included all likely hazards, or not addressed 

specific pathogens.



* In the third district, inspectors had not issued noncompliance 

records for any of the 15 plants with HACCP plans that did not 

sufficiently document their hazard analyses.



The food safety system correlation reports did not elaborate on the 

reasons for the lack of documented violations in noncompliance records. 

However, all the reports included the general observation that “a 

number of inspection personnel” were unclear about some of the basic 

requirements of the HACCP program.



The Few In-Depth Verification Reviews Completed to Date Have Found 

Potentially Serious Problems:



FSIS conducted in-depth verification reviews during calendar years 2000 

and 2001 at 47 plants that it considered as having potentially serious 

food safety risks.[Footnote 9] The 47 plants included 31 that had 

failed to meet the Salmonella performance standard in two consecutive 

sets of tests; 8 that tested positive for Listeria monocytogenes on 

ready-to-eat meat or poultry products; 4 that tested positive for 

E. coli 0157:H7 on ground beef; and 4 for other concerns.[Footnote 10]



We found that at 44 of these 47 plants, FSIS identified significant 

violations of regulatory requirements in HACCP plans. In 42 of these 44 

plants, the HACCP plans did not include a complete hazard analysis to 

identify the biological, chemical, or physical food safety hazards that 

were reasonably likely to occur.



* One plant did not have the required documentation to substantiate the 

hazards that were identified in the hazard analysis. Instead, according 

to the FSIS review team, the hazard analysis was based on the personal 

experience and general knowledge of plant personnel.



* Another plant’s hazard analysis addressed some but not all parts of 

its production process where hazards could be introduced. Areas not 

addressed included returned products, packaging materials, and nonmeat 

ingredients.



* A third plant’s hazard analysis failed to identify a biological 

hazard as reasonably likely to occur for one product even though plant 

personnel were checking the product for pathogenic bacteria. Because 

the hazard was not identified in the plan and a critical control point 

was not designated for the hazard, no HACCP documentation was generated 

and no control measures were implemented to control the biological 

hazard throughout the process.



In 35 of the 44 plants, the HACCP plans did not adequately identify 

critical control points in their processes where controls could be 

applied to prevent, eliminate, or reduce a food safety hazard to an 

acceptable level.



* One plant’s hazard analysis identified several biological and 

chemical hazards as reasonably likely to occur at various steps in the 

production process but did not establish critical control points to 

address those hazards.



* Another plant’s HACCP plan for its slaughter activities identified a 

biological hazard at the animal-receiving step. The plant did not 

establish a critical control point to address that risk. Instead, it 

identified USDA inspection activities as the control measure. FSIS 

regulations require the plant itself to have control measures for all 

hazards that it identifies as reasonably likely to occur.



The in-depth verification reviews also showed that some plants may 

have a fundamental misunderstanding of what constitutes a hazard that 

is reasonably likely to occur. For example, when a plant fails a second 

set of Salmonella tests or a single test for Listeria monocytogenes or 

E. coli 0157:H7, it must reassess its HACCP plan to determine whether 

any changes are needed to prevent the problem from reoccurring. 

However, 15 of the 47 plants had not identified the pathogen that 

prompted the in-depth verification review as a hazard that was 

reasonably likely to occur. For example, the hazard analysis at one 

plant that had failed two consecutive Salmonella sets did not identify 

Salmonella as a hazard that was reasonably likely to occur--an 

“egregious” omission, according to FSIS’s report on the review for that 

plant.



The in-depth verification reviews are useful in identifying 

deficiencies in the scientific aspects of plants’ HACCP plans. However, 

according to FSIS officials, they are too time-and resource-intensive 

to be implemented on a broader scale. As of June 30, 2002, FSIS had 

conducted reviews at 57 plants--or about 1 percent of the more than 

5,000 plants nationwide.



Consumer Safety Officers Improve FSIS’s Ability to Assess the 

Scientific Adequacy of HACCP Plans, but Reviews of All Plans May Take 

Years to Complete:



FSIS has recognized since HACCP’s inception that its inspection 

workforce did not collectively possess the skills needed to evaluate 

the scientific validity of HACCP plans. As a result, FSIS does not 

expect its inspectors to evaluate the scientific soundness of HACCP 

plans. In October 2001, FSIS introduced consumer safety officers with 

the scientific and technical skills to, among other things, assess the 

scientific soundness of plants’ HACCP plans. Initially, according to 

district office officials, the consumer safety officers were reviewing 

HACCP plans at plants where there was an indication of problems and 

were finding significant violations of HACCP regulatory requirements. 

In response to those findings, FSIS is taking enforcement actions to 

address potential food safety risks. For example, in one district, as 

of June 2002, consumer safety officers reviewed HACCP plans at 59 

plants. As a result of these reviews, FSIS suspended 3 plants, sent 

letters to 17 plants indicating its intention to take enforcement 

action if changes were not made, and sent letters to 24 plants advising 

them to reassess their HACCP plans within 30 days to correct 

deficiencies.



FSIS plans to have 352 consumer safety officers by September 2005: 

32 consumer safety officers on-board as of May 30, 2002; 50 more in 

2002, plus 25 veterinary medical officers who will perform consumer 

safety officer duties on a part-time basis; 105 new consumer safety 

officers in fiscal year 2003; and 140 in fiscal year 2004. An FSIS 

official estimated, however, that FSIS might need a total of 500 

consumer safety officers, including cross-trained veterinary medical 

officers, to carry out HACCP oversight responsibilities. Initially, 

FSIS had planned to bring consumer safety officers on board more 

quickly and in greater numbers. According to FSIS officials, Congress 

did not approve FSIS’s fiscal year 2000 budget request to hire consumer 

safety officers and FSIS’s efforts to retrain existing staff to fill 

these positions have taken longer than FSIS anticipated.



Officials in the two largest district offices (Alameda, Calif. and 

Albany, N.Y.) told us that it could take several years to ensure that 

all HACCP plans at all of the nation’s meat and poultry plants have a 

sound scientific basis if FSIS cannot bring consumer safety officers on 

board as quickly as expected. If this were to occur, an Alameda 

District official told us it would take at least 4 years for the two 

consumer safety officers it has now to review the more than 500 plants 

in the district. Similarly, an Albany District official estimated that 

its five consumer safety officers will need from 2 to 5 more years to 

review HACCP plans at the district’s more than 800 plants.



FSIS’s Inspection Data Raise Further Concerns About Whether FSIS’s 

Inspectors Are Ensuring That All Plants Meet HACCP Regulatory 

Requirements:



FSIS’s headquarters and district officials told us that finding plants 

with zero HACCP-related noncompliance records for an entire year would 

be unusual. HACCP requirements are numerous, and FSIS inspects plants 

on a daily basis, which creates many opportunities to identify and 

document HACCP violations. And as one district official told us, there 

are no perfect plants. That notwithstanding, our analysis of 

information in FSIS’s inspection database showed that 55 percent of all 

plants had no documented HACCP-related violations during fiscal year 

2001. (See table 1.) FSIS officials were surprised at the large 

percentage without violations. FSIS had not analyzed these data for an 

entire year. An FSIS field official told us that if inspectors are 

finding no HACCP violations for an entire year at so many plants, they 

may not understand their HACCP oversight responsibilities.



Table 1: Percentage of Plants with No Documented HACCP Violations 

during Fiscal Year 2001:



FSIS district: Lawrence, Kans.; Number of HACCP plants: 218; Number of 

plants with no violations: 144; Percentage of plants with no 

violations: 66.



FSIS district: Philadelphia, Pa.; Number of HACCP plants: 390; Number 

of plants with no violations: 249; Percentage of plants with no 

violations: 64.



FSIS district: Beltsville, Md.; Number of HACCP plants: 166; Number of 

plants with no violations: 105; Percentage of plants with no 

violations: 63.



FSIS district: Pickerington, Ohio; Number of HACCP plants: 248; Number 

of plants with no violations: 154; Percentage of plants with no 

violations: 62.



FSIS district: Madison, Wis.; Number of HACCP plants: 277; Number of 

plants with no violations: 168; Percentage of plants with no 

violations: 61.



FSIS district: Des Moines, Iowa; Number of HACCP plants: 195; Number of 

plants with no violations: 116; Percentage of plants with no 

violations: 59.



FSIS district: Boulder, Colo.; Number of HACCP plants: 216; Number of 

plants with no violations: 127; Percentage of plants with no 

violations: 59.



FSIS district: Salem, Oreg.; Number of HACCP plants: 267; Number of 

plants with no violations: 150; Percentage of plants with no 

violations: 56.



FSIS district: Atlanta, Ga.; Number of HACCP plants: 379; Number of 

plants with no violations: 211; Percentage of plants with no 

violations: 56.



FSIS district: Dallas, Tex.; Number of HACCP plants: 243; Number of 

plants with no violations: 132; Percentage of plants with no 

violations: 54.



FSIS district: Jackson, Miss.; Number of HACCP plants: 221; Number of 

plants with no violations: 115; Percentage of plants with no 

violations: 52.



FSIS district: Chicago, Ill.; Number of HACCP plants: 330; Number of 

plants with no violations: 170; Percentage of plants with no 

violations: 52.



FSIS district: Raleigh, N.C.; Number of HACCP plants: 165; Number of 

plants with no violations: 85; Percentage of plants with no violations: 

52.



FSIS district: Albany, N.Y.; Number of HACCP plants: 838; Number of 

plants with no violations: 425; Percentage of plants with no 

violations: 51.



FSIS district: Alameda, Calif.; Number of HACCP plants: 530; Number of 

plants with no violations: 268; Percentage of plants with no 

violations: 51.



FSIS district: Springdale, Ark.; Number of HACCP plants: 183; Number of 

plants with no violations: 84; Percentage of plants with no violations: 

46.



FSIS district: Minneapolis, Minn.; Number of HACCP plants: 191; Number 

of plants with no violations: 73; Percentage of plants with no 

violations: 38.



FSIS district: Total; Number of HACCP plants: 5,057; Number of plants 

with no violations: 2,776; Percentage of plants with no violations: 55.



Note: In May 2002, an organizational realignment consolidated FSIS’s 17 

district offices into 15. In the realignment, the Pickerington, Ohio, 

office became a satellite office in the Chicago District, and the 

Salem, Oregon, office became a satellite in the Boulder, Colorado, 

District.



Source: GAO’s analysis of FSIS’s data.



[End of table]



As table 1 shows, the percentage of plants, by district, where no HACCP 

violations were documented on noncompliance records ranged from 38 to 

66 percent. Officials in several districts acknowledged that they had 

reviewed reports that showed that some plants in their districts had no 

documented HACCP violations. According to FSIS officials, they had not 

analyzed the data for an entire year and were not aware of the extent 

to which no violations had been documented. Two officials told us that 

they had asked their circuit supervisors--who oversee FSIS’s in-plant 

inspectors--to investigate plants with no documented violations but 

that they had not asked the supervisors to report back to them with the 

results.



In fact, at 10 of the 43 plants in which FSIS’s in-depth verification 

reviews found serious HACCP implementation problems, FSIS inspectors 

had not documented any HACCP violations on noncompliance records during 

the 12 months preceding the review. For example, at one of those 

plants, the in-depth verification review found that (1) corrective 

actions taken by the plant were not documented, (2) monitoring records 

did not show the time that product temperatures were checked, and (3) 

the required annual reassessment of the HACCP plan had not been done. 

Inspectors had not documented any of these violations.



Although FSIS has implemented a variety of inspector training 

activities in response to our earlier report recommendations, it is 

clear that some FSIS inspectors remain uncertain about their roles, 

responsibilities, and authorities for reviewing and verifying plants’ 

compliance with HACCP requirements. Following a meeting we had with 

FSIS officials in June 2002 to alert them of our preliminary findings, 

FSIS informed us, in a letter dated August 2, 2002, that it was taking 

a number of actions aimed at addressing the problems we identified. 

With regard to inspector activities, FSIS stated that it is developing 

a directive explaining the responsibilities of inspectors under HACCP 

and has introduced an interactive computer tool for inspectors and 

others to use to strengthen HACCP problem solving using fictional 

scenarios.



In its August 2, 2002 letter, FSIS also told us that the agency had 

developed supervisory guidelines that will be a reference for managers 

to use to verify that FSIS inspectors are carrying out their 

responsibilities, including “applying appropriate inspection methods; 

using effective regulatory decision-making; documenting findings 

appropriately; and when warranted, implementing enforcement actions 

properly.” The agency expects to train all field supervisors and 

implement the new guidelines by October 1, 2002, and believes it will 

then be better able to hold supervisors accountable for overseeing 

inspectors’ performance. The letter also stated that FSIS field offices 

are evaluating the results of the food safety system correlation 

reviews to determine how FSIS can improve the reviews and how it can 

use the reviews to strengthen inspectors’ effectiveness.



FSIS’s Lack of Consistent Identification and Documentation of 

Repetitive HACCP Violations Weakens Enforcement:



According to our review of 1,180 noncompliance records from 16 plants 

for fiscal year 2001, plant inspectors have not consistently identified 

and documented repetitive violations of HACCP requirements. The lack of 

consistency occurs, in part, because FSIS has not established specific, 

uniform criteria for identifying repetitive violations. Moreover, even 

at the district level, officials’ understanding of the factors that 

should be considered in determining repetitive violations varied. 

Furthermore, we found that FSIS’s recently revised inspection data 

system lacks important summary information that managers need to 

oversee inspectors’ identification of repetitive violations and 

enforcement decisions. If FSIS does not consistently identify and 

document repetitive violations, it cannot properly and equitably 

enforce HACCP requirements.



The Factors That Inspectors Used to Identify Repetitive Violations 

Differ:



FSIS has not established specific criteria for its inspectors to use 

for determining when violations are repetitive. According to the 

noncompliance records we analyzed, inspectors did not use the same 

factors to identify repetitive violations of HACCP requirements. Table 

2 shows the plant size and the number of noncompliance records and 

repetitive violations for the 16 plants we examined.



Table 2: Plant Size, Number of Noncompliance Records, and Number of 

Repetitive Violations in 16 Plants during Fiscal Year 2001:



Plant: Plant 1; Plant size: Very small; Number of noncompliance 

records: 9; Number of repetitive

violations: 2.



Plant: Plant 2; Plant size: Very small; Number of noncompliance 

records: 12; Number of repetitive

violations: 3.



Plant: Plant 3; Plant size: Very small; Number of noncompliance 

records: 15; Number of repetitive

violations: 8.



Plant: Plant 4; Plant size: Small; Number of noncompliance records: 15; 

Number of repetitive

violations: 4.



Plant: Plant 5; Plant size: Small; Number of noncompliance records: 23; 

Number of repetitive

violations: 0.



Plant: Plant 6; Plant size: Small; Number of noncompliance records: 26; 

Number of repetitive

violations: 16.



Plant: Plant 7; Plant size: Small; Number of noncompliance records: 60; 

Number of repetitive

violations: 34.



Plant: Plant 8; Plant size: Small; Number of noncompliance records: 

100; Number of repetitive

violations: 69.



Plant: Plant 9; Plant size: Small; Number of noncompliance records: 

112; Number of repetitive

violations: 74.



Plant: Plant 10; Plant size: Small; Number of noncompliance records: 

145; Number of repetitive

violations: 44.



Plant: Plant 11; Plant size: Large; Number of noncompliance records: 

12; Number of repetitive

violations: 10.



Plant: Plant 12; Plant size: Large; Number of noncompliance records: 

70; Number of repetitive

violations: 9.



Plant: Plant 13; Plant size: Large; Number of noncompliance records: 

109; Number of repetitive

violations: 85.



Plant: Plant 14; Plant size: Large; Number of noncompliance records: 

132; Number of repetitive

violations: 69.



Plant: Plant 15; Plant size: Large; Number of noncompliance records: 

167; Number of repetitive

violations: 113.



Plant: Plant 16; Plant size: Large; Number of noncompliance records: 

173; Number of repetitive

violations: 19.



Plant: Total; Plant size: [Empty]; Number of noncompliance records: 

1,180; Number of repetitive

violations: 559.



Source: GAO’s analysis of FSIS’s documents.



[End of table]



We found the following variations:



* Inspectors differed in determining when a violation was repetitive, 

which led to inconsistent action on similar violations. For example, in 

one plant, two inspectors issued noncompliance records documenting 

three violations over a 15-day period that had the same inspection 

procedure (slaughter), element of the HACCP system (monitoring), and 

violation (fecal contamination). One inspector issued the first and 

second records 5 days apart; the other inspector issued the third 

record 10 days later. The first inspector did not link the first and 

second records and determine that the second record was repetitive. 

However, the inspector who wrote the third record linked it to the two 

earlier records, and determined that it was repetitive.



* Noncompliance records that contained the same information were not 

always identified as repetitive, which led to inconsistent action on 

the same violations and understating the potential seriousness of the 

problem. For example, on 23 occasions, an inspector in one plant wrote 

a noncompliance record followed within 2 days by another record 

containing the same information. However, the inspector linked the 

first record to the second and determined that the second was 

repetitive on only one occasion.



* Inspectors used different time limits in which violations could be 

linked as repetitive, which may misstate the seriousness of the 

problem. For example, inspectors at one plant used a calendar month. 

Violations occurring within the same month could be linked to one 

another as repetitive, but violations in a subsequent month could not 

be linked back to incidences in the previous months. Inspectors at 

another plant used a period of 4 years. At this plant, noncompliance 

records for some violations were linked as repetitive to as many as 225 

other violations. FSIS has not given guidance on an appropriate length 

of time for linking violations, but for training purposes, it uses a 

rolling 90-day period as the time limit for linking violations as 

repetitive, meaning that an inspector would look back 90 days from the 

date of a violation.



We also found that when inspectors identified violations as repetitive, 

they did not consistently document the basis for their decision--the 

record identification numbers of the previous violations. For example, 

at one plant, inspectors did not record the identification numbers for 

75 percent of the previous repetitive violations, while at another 

plant, 46 percent of repetitive violations did not have this 

information. When documentation is lacking, FSIS cannot determine with 

any confidence the number of times a violation has been repeated and 

whether it warrants further enforcement action.



At the district level, we found that officials in the 10 districts 

where we sampled noncompliance records used varying factors, such as 

the type of violation (e.g., fecal contamination) or element (e.g., 

record keeping), that they said should be considered important in 

determining whether violations were repetitive. District officials also 

disagreed on the time period in which violations could be considered 

repetitive; only one stated that they usually used the rolling 90-day 

period.



FSIS’s Inspection Database Provides FSIS Officials with Limited 

Information on Repetitive Violations:



FSIS’s Performance Based Inspection System, a database that captures 

the results of inspection activities, generates reports on the total 

number of HACCP inspections conducted and the percentage that resulted 

in violations. This information is reported by the type of inspection 

procedure (e.g., slaughter or processing) and the element of the HACCP 

system where the violation occurred (e.g., monitoring or record 

keeping). For example, the reports can identify the number and 

percentage of various HACCP inspection procedures at a plant that 

resulted in violations related to monitoring. FSIS’s managers told us 

they considered these reports adequate to identify the potential for a 

repetitive problem and trigger the need to explore individual 

noncompliance records to determine if a repetitive problem exists. 

Consistently and accurately linking repetitive violations is important 

because FSIS uses repetitive violations as a factor in assessing 

whether a plant has an effective HACCP system, whether an enforcement 

action is warranted, and whether the meat or poultry products from that 

plant are safe for consumers.



However, the inspection database does not capture summary information 

on the number of repetitive noncompliance records an inspector issues 

to a plant, the nature of the repetitive deficiencies, or plant 

managers’ success or failure in taking effective preventive action. 

This type of summary information could assist managers in both 

overseeing inspectors’ performance and plants’ compliance with HACCP 

requirements. In addition, it could serve as one indicator for 

considering further enforcement action or for advising industry on the 

need to address a common problem. For example, managers could oversee 

inspectors’ performance by comparing inspectors’ rates of identifying 

repeat violations. If an inspector identified high rates, then a 

manager could investigate to determine if the inspector proposed or 

took further enforcement action. If the inspector identified low rates 

of repetitive violations, but the data showed high numbers of a 

particular type of violation, then managers could investigate to 

determine why these inspectors did not identify these as repetitive.



Managers have some information on repetitive violations, but not in 

summary format. Inspectors enter information into the database on an 

electronic version of the noncompliance record. Once entered, FSIS 

inspectors, supervisors, and managers from almost any location 

nationwide can review these individual noncompliance records. However, 

to assess the extent of repetitive violations that a particular 

inspector identified, a manager or supervisor would need to view every 

noncompliance record--a cumbersome process. FSIS officials maintain 

that the individual inspector is responsible for assessing the extent 

of repetitive violations, making a professional judgment on the need 

for further enforcement action, and bringing this matter to the 

attention of managers. However, access by managers to summary 

information on the repetitive violations that inspectors already 

identified would facilitate management’s oversight of HACCP 

implementation.



Although FSIS headquarters and district officials told us throughout 

our work that they had sufficient data on repetitive violations, in its 

August 2, 2002 letter, FSIS stated that it now recognizes that it needs 

to improve and strengthen its data systems to support management 

decision making on repetitive violations. The letter stated that FSIS 

has implemented several systems over the past 6 months in an effort to 

address its need for better inspection information from its data 

systems and is testing software that will enable district officials and 

managers to extract information and summary reports to help identify 

problem areas. The letter also stated that FSIS is pilot testing an 

early warning system for district officials that draws on data from 

various FSIS databases.



FSIS Is Not Ensuring That Plants Take Prompt and Effective Corrective 

Action to Return to Compliance with HACCP Requirements after Violations 

Have Been Identified:



FSIS is not ensuring that all plants take prompt and effective 

corrective action to return to compliance with regulatory requirements 

after violations have been identified in three areas. First, FSIS does 

not consistently ensure that plants quickly take effective action to 

eliminate repetitive violations, particularly those of the zero 

tolerance standard for visible fecal contamination. Second, FSIS does 

not ensure that plants take prompt action to meet the Salmonella 

performance standard after a second consecutive failure. Finally, when 

FSIS suspends inspection services at a plant because of serious 

violations, it generally places those suspensions in abeyance, allowing 

the plants to continue operating. However, it rarely identifies a time 

frame for the plant to take corrective actions, and it does not track 

the actual time the plant takes to make the correction. The longer that 

FSIS allows plants to remain out of compliance with regulatory 

requirements, the greater the risk that unsafe food will be produced 

and enter the marketplace.



FSIS Does Not Consistently Ensure That Plants Quickly Take Effective 

Action to Eliminate Repetitive Violations:



According to FSIS regulations, enforcement action is warranted when 

plants fail to demonstrate that their HACCP systems adequately prevent 

multiple or recurring violations. However, FSIS inspectors do not 

consistently initiate enforcement actions in such cases. For example, 

in the 1,180 noncompliance records we examined at 16 plants, the 

violation of FSIS’s zero tolerance standard for fecal contamination was 

the most common type of repetitive violation. Each time an inspector 

documents this violation, FSIS regulations require the plant to take 

corrective action to remove the contamination. However, FSIS did not 

take withholding or suspension enforcement actions at any of the 11 

plants where repetitive fecal contamination was identified. For 

example, FSIS issued 96 noncompliance records to one plant for these 

violations and although 88 percent of these records were linked as 

repetitive, FSIS did not initiate a withholding or suspension 

enforcement action. District officials stated that FSIS did not 

initiate an enforcement action because the plant had “done a good job” 

of addressing violations that were brought to its attention previously 

and the number of noncompliance records issued to the plant for fecal 

contamination was “not out of line” for a large plant. In addition, the 

officials said that violations of the fecal standard are bound to occur 

and most of the fecal contamination was minuscule--about one-eighth to 

one-quarter of an inch in diameter.



In another case, inspectors issued 154 noncompliance records to a plant 

for fecal contamination during the fiscal year, and they identified 109 

of the deficiencies (71 percent) as repetitive, yet took no withholding 

or suspension enforcement action. For this plant, an FSIS official told 

us that the inspector did not recommend enforcement action because, in 

the inspector’s professional opinion, the findings did not warrant it. 

However, because fecal matter can harbor serious contaminants, 

including E. coli 0157: H7, any fecal matter is a potentially serious 

health risk. FSIS’s zero tolerance standard for visible feces 

recognizes this risk.



At plants where FSIS conducted an in-depth verification review, the 

noncompliance records also indicate that FSIS is not ensuring that 

plants quickly take effective action to eliminate repetitive 

violations. For example, at one slaughter plant, over 9-months, FSIS 

inspectors documented 27 instances of animals presented for slaughter 

that had levels of antibiotic drug residues that exceeded the amounts 

allowed by FSIS. On the earliest noncompliance record we reviewed, 

dated August 2000, the FSIS inspector wrote that the finding of 

excessive drug residue indicates, “that there may be an inadequacy in 

[the plant’s] HACCP plan to control food safety hazards identified in 

[its] hazard analysis.” However, it was not until April 2001, 8 months 

later, that the plant implemented a program designed to prevent drug 

residues from entering its products.



FSIS has not established consistent criteria for inspectors to consider 

when assessing whether repetitive violations warrant a withholding or 

suspension enforcement action. According to FSIS officials, the 

decision to pursue an enforcement action is left to the professional 

judgment of each plant inspector. However, some district officials and 

inspectors we interviewed indicated that they would benefit from having 

FSIS identify a uniform set of factors for them to assess when 

considering whether an enforcement action is warranted. These officials 

suggested, for example,



* the length of time between repetitive violations,



* the number of repetitive noncompliance records issued,



* the nature of the violations, and:



* the plant’s efforts and/or success in implementing preventive 

actions.



FSIS officials told us they recognize the need to establish criteria 

for assessing whether repetitive violations warrant enforcement action 

and are in the process of updating a policy directive to include such 

criteria. FSIS expects to implement this directive in early 2003.



In addition, for repetitive violations, FSIS does not require 

inspectors to document whether they had considered and recommended or 

decided against an enforcement action. Such documentation would assist 

other inspectors at the same plant in determining whether enforcement 

actions are warranted when they issue additional noncompliance records 

for similar violations. This documentation could also assist 

supervisors and district office officials in overseeing plants’ 

implementation of HACCP requirements.



FSIS Does Not Ensure That Plants Take Prompt Actions to Meet the 

Salmonella Performance Standard after a Second Consecutive Failure:



FSIS does not ensure that plants take prompt corrective actions after 

they fail to meet the Salmonella performance standard. The HACCP 

regulations require that plants take immediate action to meet this 

standard if they have failed a set, but they do not explain what is 

meant by “immediate action.” However, our analysis of elapsed time 

shows that plants are not taking prompt corrective actions in these 

instances.



After a plant fails a second consecutive set of Salmonella tests, FSIS 

requires the plant to reassess its HACCP plan to determine if it should 

make any changes. In addition, FSIS has conducted in-depth verification 

reviews to evaluate the design and implementation of HACCP plans at 

plants that failed a second set of Salmonella tests. In 2000 and 2001, 

FSIS conducted in-depth verification reviews at 31 of these plants. In 

one case, FSIS conducted an in-depth verification review at one plant 

that had failed three of the past four sets of tests. However, the 

failure that triggered the review was the first set in a new series. 

Following the in-depth verification review, 14 plants passed a third 

set of tests and 5 failed. As of April 2002, for the remaining plants, 

sampling was in process or FSIS was waiting to begin the third set of 

tests.[Footnote 11]



As shown in table 3, our analysis of time frames for each step in the 

process, from first-set failure to passing a third set of Salmonella 

tests, shows that considerable time elapsed. These time frames are not 

consistent with ensuring that plants immediately meet the performance 

standard after a failure, as the regulations require.



Table 3: Elapsed Time--From Failure of First Set to Passing Third Set-

-at Plants with Second-Set Failures of the Salmonella Performance 

Standard and In-Depth Verification Reviews:



Event: First-to second-set Salmonella failure; Average number of days 

elapsed: 198; Range of days elapsed: 113 to 337; Number of plants: 

30[A].



Event: Second-set failure to in-depth verification review; Average 

number of days elapsed: 96; Range of days elapsed: 21 to 322; Number of 

plants: 26[B].



Event: In-depth verification review to reassessment letter; Average 

number of days elapsed: 76; Range of days elapsed: 11 to 284; Number of 

plants: 28[C].



Event: FSIS’s reassessment letter to plant’s written response; Average 

number of days elapsed: 37; Range of days elapsed: 12 to 82; Number of 

plants: 26[D].



Event: In-depth verification review to third-set result; Average number 

of days elapsed: 320; Range of days elapsed: 155 to 543; Number of 

plants: 16[E].



Event: First-set Salmonella failure to third-set pass; Average number 

of days elapsed: 608; Range of days elapsed: 340 to 929; Number of 

plants: 14[F].



[A] Subsequent to the in-depth verification review, one of the 31 

plants had its second-set Salmonella failure overturned on appeal. 

Consequently, it was not included in this calculation.



[B] FSIS conducted an in-depth verification review at one plant that 

had failed three of the past four sets of tests. However, the failure 

that triggered the review was the first set in a new series. Therefore, 

the time from second-set failure to the review could not be calculated. 

Also, in three cases FSIS began the third set of tests prior to the in-

depth verification review. Therefore, these plants are not included in 

this calculation.



[C] FSIS did not send reassessment letters to 2 of the 30 plants 

because the in-depth verification reviews were conducted before the 

agency required reassessment letters in these situations.



[D] Two plants did not receive reassessment letters and so did not send 

a response to FSIS. One plant’s response to the reassessment letter was 

not dated, and FSIS did not provide us with the plant’s response letter 

in one case.



[E] As of April 4, 2002, FSIS had completed the third set of Salmonella 

tests at 19 plants. However, in three cases the third set of tests was 

begun prior to the in-depth verification review. Therefore, the time 

from the review to third-set result could not be calculated for these 

plants.



[F] Of the 19 plants for which a third set of Salmonella tests were 

completed as of April 4, 2002, 14 passed and 5 failed.



Source: GAO’s analysis of FSIS’s data.



[End of table]



Our analysis showed that, on average, it took FSIS about 3 months from 

the date of the second-set failure to begin an in-depth verification 

review. Once the in-depth verification review was complete, an average 

of 2½ months elapsed before FSIS sent its “reassessment” letter to the 

plant listing all of the deficiencies in the design and implementation 

of the HACCP plan found during the review. In one case, the period from 

review to FSIS letter was 284 days--or more than 9 months. The 

reassessment letter instructs the plant to respond to FSIS in writing 

within 30 days stating the changes that will be made to meet the 

Salmonella performance standard.



We found that plants came close to meeting this time frame--replying 

within 37 days on average. On the other hand, FSIS does not 

consistently require that plants quickly take steps to correct the 

deficiencies identified by the in-depth verification reviews. As shown 

in table 3, FSIS has allowed plants to take an average of 608 days--or 

over 1½ years--from their first-set Salmonella failure until the 

successful completion of a third set of tests. Over half that time, or 

about 11 months, elapsed from the date of the in-depth verification 

review until the successful completion of the third set of tests. For 

example, at one plant in the Dallas district, 18 months elapsed from 

the date of the in-depth verification review until the completion of 

the third set of tests. (See table 4.):



Table 4: Time Line of Events between In-Depth Verification Review and 

Third Set of Salmonella Tests at One Plant:



Date: Oct. 30 to Nov. 2, 2000; Action: FSIS team conducts the in-depth 

verification review..



Date: Apr. 3, 2001; Action: FSIS sends the reassessment letter to the 

plant..



Date: Apr. 25, 2001; Action: The plant responds to FSIS about what 

changes it will make to its operations..



Date: May 21, 2001; Action: The plant asks for a delay in conducting 

the third set of tests..



Date: Aug. 23, 2001; Action: FSIS writes the plant that the agency has 

not yet received written notification of the completion dates for the 

plant’s improvement projects..



Date: Oct. 31, 2001; Action: The plant writes to FSIS that the 

improvements would be completed by mid-November and asks that the third 

set of tests not begin until after December 1, 2001..



Date: Dec. 3, 2001; Action: Sampling for the third set of tests 

begins..



Date: Apr. 26, 2002; Action: The third set is completed and the plant 

passes..



Source: GAO’s analysis of FSIS’s documents.



[End of table]



Officials in the Dallas District Office told us that 5 months elapsed 

between the date of the in-depth verification review and FSIS’s 

reassessment letter largely because of the amount of time needed to 

incorporate all of the in-depth verification review team’s comments and 

complete the report of the review’s findings. About 8 months elapsed 

from the date of the plant’s response to the reassessment letter until 

the third set of Salmonella tests began because, according to the 

district officials, the district did not closely monitor the progress 

of the changes the plant was making and because the plant requested and 

received FSIS’s permission to delay the third set until those changes 

were made.



Because plants may continue to produce products that could pose a 

Salmonella risk from the first-set failure until they pass, it is 

important that the second-and third-set tests be scheduled and 

completed as soon as possible. An FSIS headquarters official 

acknowledged that, in some cases, the time between the second and third 

set of Salmonella tests has been too long. However, the official stated 

that plants sometimes make significant changes to their operations 

after an in-depth verification review and the time needed to reassess, 

modify and validate HACCP plans can be considerable. Nonetheless, in 

the Dallas district case described above, at least a portion of the 

delays were due to inattentive oversight by FSIS. In its August 2, 

2002, letter to us, FSIS said it has designed and is ready to test a 

tracking system for the in-depth verification reviews to assist the 

agency in keeping those reviews timely and to allow for trend analysis 

of review results over time.



FSIS further stated in that letter that USDA’s Under Secretary for Food 

Safety, in consultation with the Secretary of Agriculture, established 

a new office--the Office of Program Evaluation, Enforcement and Review-

-to ensure that its programs and policies are implemented and monitored 

correctly. The new office, which began operating on July 15, 2002, will 

conduct in-depth examinations of FSIS policies to determine if the 

policies are adequate or if additional actions are needed. The new 

office began by looking at the Salmonella testing program to determine 

whether it is accomplishing all it should to protect human health. It 

expects to report its findings by mid-September 2002. In addition, FSIS 

noted, “using [the GAO preliminary findings] as a guide, [the new 

office] has begun to assess the adequacy of the field staff’s 

implementation of HACCP.” This preliminary report is also due in mid-

September.



FSIS Does Not Always Ensure That Plants Whose Suspensions Have Been 

Placed on Hold Take Prompt Action to Return to Compliance:



FSIS is also not ensuring that plants take prompt corrective actions 

when it places plants’ suspensions in abeyance. When a suspension is 

in abeyance, FSIS inspection services resume, and the plant continues 

operating while it makes corrections. In analyzing 60 HACCP 

administrative enforcement case files for 2001 for plants in the 

Albany, Alameda, and Madison districts, at which FSIS had suspended 

inspection services, we found that 57 of the 60 suspensions were placed 

in abeyance. In half of the suspensions that were in abeyance (28 of 

the 57 cases), FSIS placed the suspension in abeyance on the same day 

the suspension was issued. Because there was either no or very limited 

interruption in inspection services at the plant, the effect on the 

plant, in terms of economic loss or disruption, was negligible.



On January 24, 2001, FSIS published policy guidance stating that no 

plant’s suspension should remain in abeyance for more than 90 days 

without a specific “operational reason,” such as a violation that 

involved a process that the plant operates intermittently. Limiting the 

time that suspensions can remain in abeyance was also FSIS’s practice 

prior to the January 2001 policy notice establishing a specific time 

frame. Our sample of the 57 enforcement cases included 30 plants that 

had suspensions in abeyance that were closed after the 90-day policy 

went into effect in January 2001. Of these 30 enforcement cases, only 1 

was closed within 90 days. The average number of days from suspension 

in abeyance to case closure was 316 days, or over 10 months. According 

to FSIS officials in charge of enforcement in the three district 

offices where we reviewed enforcement cases, the 90-day time frame for 

holding suspensions in abeyance was not met because (1) the district 

office did not require inspection personnel to report to them on 

whether the plants had completed their corrective and preventive 

actions within the 90-day period, (2) it often took longer than 90 days 

for inspection personnel to inform the district office that plants had 

completed their corrective and preventive actions and the cases could 

be closed, and (3) closing these cases was a low priority for the 

district office.



Further, none of FSIS’s “Notice of Suspension of Inspection” documents 

or other correspondence in the enforcement files for the 30 cases we 

examined specified a date by which corrective actions were expected to 

be implemented and their effectiveness verified. As long as FSIS does 

not establish specific deadlines for plants with suspensions in 

abeyance to correct their problems, plants have little incentive to 

quickly implement and verify the effectiveness of their corrective 

actions.



In addition, we were generally unable to verify that violations were 

corrected before FSIS issued the letter ending the suspension in 

abeyance and closing the case. The enforcement case files frequently 

did not contain evidence showing how and when the district offices 

determined that the plants had completed corrective and preventive 

actions. However, in the Albany District Office, 6 of the 21 closed 

enforcement files contained correspondence from inspection personnel to 

the district office documenting how and when the plant had corrected 

the violation and recommending closing the case. An FSIS headquarters 

official responsible for enforcement activities told us that he would 

expect to see documentation showing how decisions were reached in all 

case files.



Moreover, in the Alameda district office, two of the four plants we 

reviewed continued to have violations of the same requirements for 

which the plant was suspended, according to the inspectors’ 

documentation. Nevertheless, the suspension remained in abeyance, and 

FSIS did not take any further enforcement action. For example, one 

suspension that was placed in abeyance on the day it was issued in 

October 2000 was the result of the plant’s repeated failure to keep 

adequate HACCP records to verify that critical control points were 

being properly monitored to ensure food safety. In March 2001, 

inspectors documented three more instances of the plant’s failure to 

keep the same type of HACCP records and two more instances in the 

following months until the case was closed in September 2001.



FSIS also allows plants to have multiple suspensions in abeyance in 

effect simultaneously, each for a separate production process, or to 

have sequential suspensions in abeyance that last for extended periods 

of time. As a result, a plant can continually struggle to meet 

requirements and require special regulatory oversight for an extended 

period of time and yet remain in business. For example, the plant in 

the Alameda District mentioned above had sequential suspensions placed 

in abeyance. In October 2001, just 1 month after the earlier suspension 

was lifted, the district office sent the plant a letter indicating 

FSIS’s intention to take enforcement action because of the plant’s 

failure to collect product samples for E. coli testing, as required by 

the HACCP regulations. On the basis of the plant’s response to the 

letter, the district office deferred any enforcement action for 90 

days. Recognizing that this was a problem plant, in November 2001, the 

district had a consumer safety officer conduct a comprehensive review 

of the plant’s HACCP system. Owing to the HACCP design and 

implementation problems discovered by the consumer safety officer, such 

as deficiencies in the hazard analysis and record keeping, inspection 

was again suspended in early December 2001 and the suspension was 

placed in abeyance 2 days later. In January 2002, because of the 

plant’s failure to adhere to its October 2001 plan to improve its 

E. coli sampling procedures, inspection was again suspended and then 

placed in abeyance 1 day later. Alameda district officials told us they 

recommended to FSIS headquarters that FSIS withdraw this plant’s grant 

of inspection and were told that there was insufficient cause to take 

this action.



In its August 2, 2002, letter to us, FSIS stated that an administrative 

enforcement data system, which it implemented in January 2002, “for 

distributing copies, tracking status, and querying for information on 

administrative actions” provides, among other things, “status reports 

[that] show suspensions in abeyance to assist District Managers in 

assuring proper follow up at these establishments.” It further stated 

that it sets “very specific timelines for the plant to meet with 

respect to corrective or preventive measures” and that its “in-plant 

personnel conduct verification activities to ensure they are meeting 

the conditions outlined in the timeline. If [plants] do not follow 

through on the timeline/plan, the suspension is reinstated.” FSIS’s 

letter went on to note that “[t]he average time for the closure of 

suspension actions placed in abeyance in [fiscal year] 2002 has been 

105 days.” However, unlike our analysis, this average time is based on 

FSIS enforcement cases for violations of sanitation standards as well 

as HACCP enforcement cases and, as the letter points out, does not 

include an unspecified number of cases that remain open.



Conclusions:



Meat and poultry plants have many incentives to operate safely and 

certainly many appear to be doing so under HACCP. Nevertheless, FSIS’s 

oversight and enforcement needs to be improved to ensure that it is 

achieving its intended food safety objectives. FSIS inspectors are 

currently not consistently identifying and documenting violations of 

HACCP regulatory requirements, and FSIS has not assessed the scientific 

soundness of all HACCP plans in a timely manner. Moreover, some FSIS 

inspectors still do not have a clear understanding of their roles, and 

FSIS managers have not been diligent in overseeing inspectors’ 

activities. Finally, until consumer safety officers complete their 

assessments, some plants may be operating with unsound HACCP plans. 

These weaknesses limit the effectiveness of the HACCP system in 

reducing the risks posed by pathogens and contaminants on meat and 

poultry.



With regard to identifying repetitive violations--signs that a plant 

may be struggling to fully meet HACCP requirements--FSIS’s inspectors 

and managers are confused about the factors that should be considered. 

Until FSIS establishes clear, consistent criteria for determining and 

documenting repetitive violations--and ensures that its inspectors and 

managers understand these criteria--serious problems may go 

unrecognized. The extraction of summary information on repetitive 

violations from FSIS’s inspection database would help determine, among 

other things, when repetitive violations might indicate problems common 

to an industry sector or an FSIS district.



Finally, the longer that FSIS allows plants to remain out of compliance 

with HACCP requirements, the greater the risk that unsafe food will be 

produced. When plants do not take actions that promptly and 

successfully prevent repetitive violations--such as multiple recurring 

violations of the zero tolerance standard for visible fecal 

contamination--FSIS managers and officials must take enforcement 

actions to compel plants to revise their HACCP plans to address these 

problems. The system that FSIS has in place to address plants that fail 

Salmonella performance standards--allowing plants to operate while 

increased food safety risks persist for months and months--needs to be 

reexamined. Similarly, FSIS’s practice of placing a plant in suspension 

but then immediately putting the suspension in abeyance for protracted 

periods of time negates an important incentive for plants to quickly 

correct problems. While some corrective actions could take a 

significant period of time to implement--and placing a suspension in 

abeyance might be warranted when FSIS is sure that interim actions will 

provide for food safety--the circumstances should be clearly 

established and progress closely monitored and documented to ensure 

that plants are returning to compliance as soon as possible.



Recommendations for Executive Action:



To ensure that all HACCP plans fully meet regulatory requirements, we 

recommend that the Secretary of Agriculture direct FSIS to:



* provide inspectors with additional training on their roles and 

responsibilities under the HACCP system and use data, such as the 

results from the food safety system correlation reviews, to help target 

training to address specific weaknesses;



* develop procedures for its field supervisors and district managers to 

use to monitor inspector activities, including, among other things, 

ensuring that FSIS inspectors are consistently applying HACCP 

requirements;



* develop a risk-based strategy and time frames for consumer safety 

officers to complete their reviews of HACCP plans at all plants; and:



* develop a strategy for its supervisors, managers, and officials to 

systematically use data, including annual data on noncompliance records 

by districts, to help oversee plants’ compliance with HACCP 

requirements.



To ensure that FSIS inspectors and district officials use consistent 

criteria for identifying repetitive violations of HACCP regulatory 

requirements, we recommend that the Secretary of Agriculture direct 

FSIS to:



* establish specific, uniform criteria for identifying repetitive 

violations;



* ensure that inspectors consistently document repetitive violations;



* modify data management systems to capture the extent to which 

inspectors are identifying repetitive violations at plants; and:



* develop a strategy for its supervisors, managers, and officials to 

systemically use available data, including summary information, to help 

identify repetitive violations.



To ensure that plants take prompt actions to correct violations, we 

recommend that the Secretary of Agriculture direct FSIS to:



* establish clear, consistent criteria for inspectors to use when 

considering whether to recommend suspension because of repetitive 

violations;



* require its inspectors to document the basis for their decision on 

whether or not to recommend further enforcement action based upon 

documented repetitive violations;



* develop guidance with specific time frames for actions to be taken at 

plants that fail a second set of Salmonella tests, including time 

frames for FSIS to initiate an in-depth verification review, report the 

results of the review, and initiate a third set of tests;



* establish, and document in enforcement case files, time frames for 

plants with suspensions in abeyance to implement and verify the 

necessary corrective actions; and:



* document in the enforcement case file how and when the district 

office determined that the plant had completed its corrective actions 

and, if the suspension is allowed to remain in abeyance for more than 

90 days, the reason for the extension.



Agency Comments and Our Response:



We provided USDA with a draft of this report for review and comment. 

USDA concurred with our recommendations but believes the report does 

not fully acknowledge FSIS’s progress and continuous efforts to ensure 

that all plants meet regulatory requirements. Noting that FSIS has 

placed significant resources into the processes and systems that 

provided the data for our study, USDA states that FSIS has efforts 

ongoing to evaluate the same data and has been open about addressing 

these and other similar concerns and the need for associated program 

improvements. USDA describes a number of actions that FSIS has recently 

taken or is planning to take that are consistent with our 

recommendations. We believe that our report accurately reflects the 

status of FSIS’s ongoing and planned actions. If fully carried out and 

given diligent management attention, these actions could go a long way 

toward addressing the problems we found in FSIS’s oversight and 

enforcement of HACCP in U.S. meat and poultry plants and helping to 

reduce the risk of foodborne illness for American consumers. USDA’s 

comments are presented in appendix II. USDA also provided technical 

suggestions, which we incorporated into the report as appropriate.



We conducted our review from August 2001 through August 2002 in 

accordance with generally accepted government auditing standards.



As agreed with your offices, unless you publicly announce the contents 

of this report earlier, we plan no further distribution until 30 days 

from the date of this letter. At that time, we will send copies of this 

report to the congressional committees with jurisdiction over food 

safety issues; the Secretary of Agriculture; the Administrator, Food 

Safety and Inspection Service; the Director, Office of Management and 

Budget; and other interested parties. We also will make copies 

available to others upon request. In addition, the report will be 

available at no charge on the GAO Web site at http://www.gao.gov:



If you have any questions about this report, please contact me or 

Erin Lansburgh at (202) 512-3841. Key contributors to this report are 

listed in appendix III.



Lawrence J. Dyckman

Director, Natural Resources

 and Environment:



Signed by Lawrence J. Dyckman:



[End of section]



Appendix I: Scope and Methodology:



Our review of the Food Safety and Inspection Service’s (FSIS) 

enforcement of the Hazard Analysis and Critical Control Point (HACCP) 

requirements focused exclusively on domestic meat and poultry slaughter 

and processing plants subject to federal oversight. To assess whether 

FSIS is ensuring that plants’ HACCP plans meet regulatory requirements, 

we analyzed data from several sources. These sources included the 

following:



* The findings on the adequacy of HACCP plans from the food safety 

systems correlation team review reports in six districts.



* The findings concerning HACCP plans from FSIS’s reports for the 47 

in-depth verification reviews of plants’ HACCP systems completed in 

calendar years 2000 and 2001.



* The approximately 700 HACCP-related noncompliance records written at 

plants where FSIS conducted an in-depth verification review to 

determine the type of deficiencies inspectors identified in the year 

prior to the review.



* FSIS’s Performance Based Inspection System database to identify the 

number of plants in each district where no HACCP-related violations 

were documented during fiscal year 2001.



We also visited 17 plants in California to view their HACCP systems in 

operation and observe FSIS’s oversight. We visited plants in each of 

FSIS’s three size categories--large (500 or more employees), small (10 

to 499 employees), and very small (fewer than 10 employees or annual 

sales of less than $2.5 million). The plants we visited were engaged in 

the slaughter and/or processing of chickens, turkeys, hogs, and cattle. 

We reviewed HACCP-related documents and discussed the plant’s HACCP 

system with FSIS officials at each of the plants we visited. In 

addition, we obtained information from headquarters and district office 

officials on the goals, training, and responsibilities of consumer 

safety officers. We also reviewed the HACCP regulations, FSIS 

directives and other policy documents, and interviewed FSIS officials 

from headquarters and the Technical Service Center regarding the 

agency’s approach to reviewing HACCP plans.



To determine whether FSIS is consistently identifying repetitive 

violations of HACCP requirements, we judgmentally selected 16 plants 

located in 10 FSIS districts. We selected these plants from among those 

where FSIS frequently found violations in fiscal year 2001. 

Specifically, we selected our sample from among those plants where 

8 percent or more of the time, when FSIS conducted a HACCP inspection, 

it found a violation. Only about 4 percent of the total plants subject 

to HACCP inspection met this criterion. We excluded from our sample any 

plants that are participating in FSIS’s HACCP-Based Inspection Models 

Project or those that had been subject to an enforcement 

action.[Footnote 12] Of the 16 plants we selected, FSIS considers 6 to 

be large, 7 small, and 3 very small. Ten of the plants we selected were 

poultry plants and 6 were beef plants. We analyzed all of the HACCP-

related noncompliance records provided by FSIS for fiscal year 2001 at 

these 16 plants--in total 1,180 records--to determine the extent to 

which inspectors had identified repetitive violations. We interviewed 

district officials from the 10 district offices to determine the 

factors they consider important in determining whether violations were 

repetitive. The district officials we spoke to included district 

managers; assistant district managers for enforcement; HACCP inspection 

coordinators; managers of inspection staff at a group of plants in a 

district, known as “circuit supervisors;” and veterinarians and an 

inspector at plants. We also interviewed FSIS headquarters officials 

and district officials to discuss issues raised from our review of the 

noncompliance records. We reviewed FSIS’s policy documents and training 

materials related to determining repetitive violations.



To assess whether FSIS is ensuring that plants take prompt and 

effective action to return to compliance after the agency has 

identified HACCP violations, we analyzed data from three different 

sources. First, we analyzed the type and rate of repetitive violations 

identified by inspectors in our sample of 16 plants and spoke to FSIS 

officials in district offices to discuss why no further action was 

taken. Second, we analyzed FSIS’s data on the 31 plants that failed a 

second set of Salmonella tests and, at which FSIS conducted an in-depth 

verification review, to identify the time that elapsed between the 

first, second, and third sets of Salmonella tests. Third, we analyzed 

the case files for 68 HACCP administrative enforcement cases that were 

active in fiscal year 2001 from three FSIS districts (Albany, Alameda, 

Madison). We selected these three districts because they represented 

large, medium and small enforcement caseloads and were located in three 

different regions of the country. Albany and Alameda are the districts 

with the largest number of plants. We interviewed the assistant 

district managers for enforcement in each of the three districts 

regarding the actions taken on these cases. We also spoke to FSIS 

headquarters officials about enforcement issues and reviewed FSIS’s 

policy documents related to HACCP enforcement.



We conducted our review from August 2001 through August 2002 in 

accordance with generally accepted government auditing standards.



[End of section]



Appendix II: Comments from the U.S. Department of Agriculture:



Note: GAO’s comments supplementing those in the report’s text appear at 

the end of this appendix.



United States Department of Agriculture:



Food Safety and Inspection Service:



Washington, D.C. 20250:



AUG 26 2002:



Lawrence J. Dyckman:



Director, Natural Resources and Environment Food and Agriculture 

Issues:



U.S. General Accounting Office 441 G Street, N.W. Room 2T23 Washington, 

DC 20548:



Dear Mr. Dyckman:



Thank you for the opportunity to review and provide comments on the 

U.S. General Accounting Office (GAO) Draft Report, Meat and Poultry - 

Better USDA Oversight and Enforcement of Safety Rules Needed to Reduce 

Risk of Foodborne Illnesses, No. GAO-02-902. Although we agree with the 

GAO recommendations, we believe that the report does not fully 

acknowledge our progress and continuous efforts to fully ensure that 

all plants meet regulatory requirements. Further, it should be 

recognized that FSIS has placed significant resources into the 

processes and systems that generate the data and information used by 

GAO to evaluate our program. FSIS has ongoing efforts to routinely 

evaluate the same type of data. FSIS has been open about addressing 

these and other similar concerns and the need for associated program 

enhancements.



Our comments will highlight a number of actions that we have taken to 

ensure that regulatory requirements are met. Many of these actions have 

been implemented over the last few months and are reflected in Agency 

Directives, Notices and other guidelines. FSIS has used much of the 

same data as GAO in conducting its internal assessments that have led 

to many actual and planned improvements of our internal policies, 

procedures and guidance. Further, we believe that the GAO report 

validates, through its recommendations, the course of action that the 

Agency is taking to enhance the implementation of the Hazard Analysis 

and Critical Control Point (HACCP) program and continue to ensure that 

plants are meeting our regulatory requirements.



GENERAL COMMENTS:



1.We believe that the title of the report is a bit misleading. We 

suggest a different title such as “Continued USDA Oversight and 

Enforcement of Safety Rules Needed to Reduce Risk of Foodborne 

Illnesses”.



2. The GAO report noted what appears to be inconsistent implementation 

of HACCP. Similar observations made earlier by FSIS officials resulted 

in several directives and notices being prepared to provide 

supplemental guidance and clarification of existing policies and 

requirements.



a. In response to decision by the United States Court of Appeals for 
the 

Fifth Circuit, No. 00-11008, USDA v. Supreme Beef Processors, Inc., the 

Agriculture Secretary Ann M. Veneman announced several initiatives that 

FSIS would undertake to address issues related to implementation and 

enforcement of the Salmonella performance standards. These initiatives 

include the development of procedures to ensure that inspection 

activities continue to adequately verify an establishment’s total food 

safety program and revisions in the Agency’s response to Salmonella 

sample set failures.



b. FSIS has developed FSIS Notice 28-02, Action To BE Taken In 

Establishments Subject to Salmonella Testing, dated July 25, 2002. This 

notice provides program personnel with verification activities that 

they are to carry out in slaughter operations subject to the Salmonella 

performance standards and grinding operations that are subject to 
Salmonella 

testing. Carrying out these activities is a way to ensure the proper 

performance of verification activities with respect to an 
establishment’s 

total food safety program. This notice also provides new steps the 
Agency 

will take should an establishment fail to meet the standards.



c. FSIS also issued FSIS Notice 36-01, Rules of Practice, dated 

September 5, 2001. The directive was issued to ensure that all 

inspection program personnel are knowledgeable about the enforcement 

actions that the Agency may take (generally) in inspected 

establishments, the circumstances under which the various types of 

enforcement actions are appropriate and can be taken, and the 

procedures that the Agency will follow in doing so. The rules of 

practice provide a key link between inspection and enforcement 

activities.



d. On August 9, 2002, we issued FSIS Notice 29-02, HACCP Verification 

Procedures and the 30-day Reassessment Letter, which clarifies how 

inspection program personnel should respond when they find evidence of 

a deviation from a critical limit or a HACCP noncompliance while 

performing an 01 and 02 HACCP verification procedure. This notice also 

clarifies issues regarding when inspection program personnel should 

issue 30-day reassessment letters.



e. The Under Secretary for Food Safety, Dr. Elsa A. Murano, in 

consultation with Secretary Veneman, established within FSIS an office 

that is designed to ensure our programs and policies are implemented 

and monitored correctly. This new office, Office of Program Evaluation, 

Enforcement and Review (PEER), will be headed by two veterans of FSIS - 

one with vast Headquarters experience and the other with extensive 

field experience. PEER, was established to provide focus to internal 

review of FSIS programs, as well as an additional set of eyes for the 

Administrator in assessing effectiveness of regulatory activities. PEER 

began functioning as a unit on Monday, July 15, 2002, will conduct in-

depth examinations of policies established by FSIS to determine if 

those policies are adequate or if additional steps are necessary. For 

example, as its first review, the PEER staff began on Monday, July 15, 

a review of the FSIS Salmonella testing program to determine if that 

program is accomplishing all that it should in protecting public 
health. 

Recent reports by consumer activists are being studied to highlight 
areas 

of concern to be examined.



Review of the implementation of policies by Field staff is also a 

function of the PEER office. The PEER staff has also begun to assess 

the adequacy of the field staff’s implementation of HACCP. This initial 

review will include an assessment of Food Safety System Correlation 

(FSSC) reports, conversations with District Managers and other 

officials. This staff, which includes investigators located in the 

field, will also be able to conduct quick investigations when problems 

arise in the field, although compliance officers are also on site for 

such investigations.



f. We are also updating FSIS Directive 5000.1, Enforcement of 
Regulatory 

Requirements in Establishments Subject to the HACCP System Regulations 

(including regulations on Sanitation Standard Operating Procedures 

(SSOP’s), E. coli Testing and Criteria, and Salmonella Performance 

Standards). This directive provides instructions to supervisory 

personnel, consumer safety officers, circuit supervisors, and 

inspection program personnel for reviewing an establishment’s HACCP 

plan and otherwise enforcing the HACCP system regulations (9 CFR part 

417). It also updates previous instructions to inspection program 

personnel regarding the regulations on sanitation performance standards 

(SPS) and SSOP’s (9 CFR part 416). This directive fully explains to our 

inspectors their responsibilities under SPS, SSOP, and HACCP and 

provides them questions to consider in performing their 

responsibilities.



3. The GAO report recognized our FSSC reviews for identifying trends/

weaknesses in our verification and enforcement activities under HACCP. 

The FSSC’s also identify areas in which the industry needs to improve 

in terms of the quality of industry HACCP plans. The FSIS is continuing 

these reviews and the correlation/training sessions it holds for its 

inspection force and for industry following completion of the reviews 

in each District. Correlation in all the district offices will be 

completed by the end of FY 2003. FSIS is evaluating the FSSC’s on an 

ongoing basis to determine what additional improvements can be made and 

how the trends identified can be used to strengthen the knowledge and 

effectiveness of the inspection force and industry.



4.FSIS is improving its systems for ensuring frontline accountability 

by putting in place a formal process and tools that will be mandatory 

for frontline supervisors to use to ensure that inspection program 

personnel carry out their assigned job responsibilities as required 

under the Pathogen Reduction/Hazard Analysis and Critical Control Point 

(PR/HACCP) regulations.



FSIS Directive 4430.3, In-Plant Performance System (IPPS) Reviews, 

dated June 17, 2002, and comprehensive Supervisory Guidelines have been 

distributed through the Agency directives system. These documents 

provide specific instructions and guidance on how to assess the 

performance of non-supervisory inspection program personnel in PR/HACCP 

system environment. The Supervisory Guidelines are a reference for 

verifying that inspection program personnel are carrying out their 

responsibilities. These include applying appropriate inspection 

methods; using effective regulatory decision-making; documenting 
findings 

appropriately; and when warranted, implementing enforcement actions 
properly.



The IPPS directive requires that each field supervisor conduct as many 

as three formal, on-site reviews of each non-supervisory, in-plant 

employee focusing on specific aspects of the individual’s job 

requirements. The supervisor will be required to document the results 

of each review by indicating the specific job requirements that were 

reviewed, and by documenting any deficiencies in knowledge or execution 

identified. The supervisor must also document the follow-up activities 

taken to address the deficiencies. This tool not only targets 

accountability of non-supervisory personnel but also targets the 

accountability of the field supervisors who are responsible for 

monitoring the activities of inspection personnel and ensuring 

consistent and effective verification of regulatory requirements by 

those personnel.



All field supervisors will receive training in the IPPS system by 

October l, 2002, when the system will be fully implemented. FSIS will 

also audit the training and the implementation of the system, once in 

place, to ensure that it is being carried out as intended.



5. In a continuing effort to enhance and improve the scientific basis 
of 

our inspection activities, FSIS has implemented a new service to 

facilitate greater sharing of knowledge and information with and among 

FSIS field personnel. This new service is identified as the 

“Interactive Knowledge Exchange (IKE),” and is described below.



The IKE is a new tool available to all field employees. It helps keep 

employees current, and provides correlation, on regulatory 

requirements, directives, notices, HACCP, S SOP, and/or agency sampling 

programs. Through the use of scenarios IKE allows, an employee to 

review FSIS requirements and hopefully to correlate with his or her 

work-group and supervisor regarding the application of those 

requirements:



A scenario pertaining to an inspection related subject is sent out 

electronically on a bi-weekly basis to all field employees. Each 

scenario includes a “what would you do?” or “what is the correct action 

to take?” section, with suggested solutions to problems or issues 

raised in the scenario.



Any employee may ask questions regarding these scenarios. It is 

believed, however, that the greatest benefit may be gained from 

discussions of the scenarios and questions within individual work 

groups and with supervisors. The scenarios, and any subsequent 

questions and answers, are intended to aid FSIS employees through 

increased understanding and correlation of regulations, policies, and 

standards for the inspection of meat, poultry, and egg products. If 

questions pertaining to any of the scenarios or answers given on IKE 

are not resolved through discussions within the work-group or with the 

supervisor, they can be submitted via electronically for further review 

and resolution to the Technical Service Center.



6. GAO’s report recognizes that FSIS knew it needed a more 

scientifically trained and capable workforce to ensure that it could 

identify problems with the design of HACCP plans. GAO accurately relays 

FSIS’ plan to add several hundred Consumer Safety Officers (CSO) to its 

workforce. However, budget constraints have meant that hiring and 

training CSOs can happen only incrementally over a period of years. 

FSIS plans to hire up to 50 CSO per year. In addition to hiring and 

training CSOs for each district, FSIS has trained its District 

Veterinary Medical Specialists and will be training additional members 

of its workforce in the CSO methodology to increase the cadre of 

employees capable of assessing the design and scientific aspects of 

establishments’ HACCP plans.



7. GAO also concludes that FSIS’ inspection related data systems 
provide 

FSIS managers with limited information on repetitive noncompliance. 

FSIS has implemented several systems over the past 6 months in an 

effort to address this need which we have also recognized as an issue. 

Following is a brief description of each system:



* PBIS 5.0 - GAO’s report outlines the capabilities of PBIS 5.0 but 
also 

points out that trend analysis and summaries are not generated by an 

automated system to help managers identify problem areas. FSIS is 

testing software that will enable field managers to use PBIS 5.0 to 

extract needed information and summary reports for this purpose.



* District Early Warning System - This system is being pilot tested. 

This system will extract and compile data from various FSIS databases 

relating to inspection, sampling, and enforcement and generates an 

“early warning” for District Managers when the combination of certain 

events or findings reach pre-determined thresholds.



* Administrative Enforcement Data System - This electronic system for 

distributing copies, tracking status, and querying for information on 

administrative actions was implemented in January 2002. Among other 

things, status reports show suspensions in abeyance to assist District 

Managers in assuring proper follow up at these establishments.



* FSIS Directive 5000.1 - This directive will set out factors to 

consider in deciding whether there has been repetitive noncompliances.



8. FSIS Notice 5-01, District Managers Responsibilities in assessing an 

Establishment’s Response to a Notice of Intended Enforcement (NOIE), 

dated January 24, 2001, establishes deferral and abeyance processes 

after a NOIE is issued. Under this notice, a District Manager (DM) can 

defer action on a NOIE or hold a suspension action in abeyance if the 

establishment, in response to the NOIE, submits corrective actions that 

the District Manager has substantial reason to believe are adequate. 

The District Manager is expected to make a decision on the adequacy of 

the action as soon as sufficient information becomes available. The 

notice indicates deferrals and abeyance should be for no more than 90 

days without cause. This allows us to put plants on notice and then 
allow 

the plant to implement the plan and demonstrate the plan is effective.



FSIS does not believe that plants are given too much time to comply 

with regulatory requirements. For example, when a suspension is put in 

abeyance, the plant establishes specific timelines for corrective or 

preventive measures, and our in-plant personnel conduct verification 

activities to ensure they are meeting the conditions outlined in the 

timeline. If they do not follow through on the timeline/plan, the 

suspension is reinstated.As of July 10, there have been 13 

reinstatements offisuspension in FY 2002 compared to 11 in FY 2001.



FSIS policy provides for a 90-day timeframe for monitoring and 

verifying effective plant implementation of corrective and preventive 

actions proposed in response to an enforcement action. Timeframes may 

be shortened in appropriate cases.



The Agency implemented enhanced electronic and management systems for 

tracking the status of administrative enforcement actions in order to 

ensure effective oversight and follow-up actions by Districts to close 

or initiate enforcement action. When FSIS verification shows plant 

implementation has been effective and resulted in compliance with 

regulatory requirements, the enforcement action is closed. If FSIS 

verification shows continued noncompliance or failure of the plant to 

implement its proposed corrective actions, the agency takes enforcement 

action, such as the suspension of operations. When plants are unable to 

restore sanitary conditions and assure future compliance with food 

safety requirements, the agency proceeds with additional action, which 

may include formal legal action to continue the suspension or withdraw 

the grant of inspection.



9. An analysis of current enforcement data shows substantial 

improvement in the enforcement arena. The average time for the closure 

of suspension actions placed in abeyance in FY 2002 has been 105 days. 

This average is based upon violations of HACCP and SSOPs. This data has 

been provided to GAO auditors. Other cases remain open for reasons such 
as 

voluntary suspensions, ongoing criminal or other investigations, 
voluntary 

cessation of operations, or for other reasons.



The GAO report also sites a concern regarding multiple suspensions in a 

single plant. Many federal establishments conduct both slaughter and 

processing operations or may have processing operations under various 

production categories. Enforcement actions may be initiated based on 

problems in different segments of plant operations. Thus, multiple 

actions may be appropriate in some cases.



10. The GAO concluded that the time between a Salmonella set failure 

and an In-Depth Verification (IDV), time between the on-site IDV work, 

the final report and 30-day reassessment letter are too long. We agree 

that improvements in the timeliness of the IDV process are needed. The 

timeliness between IDV and reporting of results has steadily improved. 

However, we realize further improvement is needed. The FSIS Technical 

Service Center (TSC) is revamping the IDV process, to require the 

report of findings be issued immediately after the conclusion of the 

review. An IDV tracking system is being developed to ensure timeliness 

and to allow for trend analysis of IDV results over time. The system is 

designed and ready for testing.



11. FSIS announced recently that it would notify all suppliers of 

ground beef grinders of positive Exoli 0157:H7 results at the time a 

positive sample result is obtained on ground beef. Notification will be 

both orally and in writing, and will state that although FSIS is unsure 

which supplier may have provided the product in question, FSIS 

encourages all of these suppliers to conduct their own investigations 

and testing until such time as the cause of the positive is determined. 

This will serve to decrease response time on suspect positives.



12. Lastly, the Agency has moved to establish “End Dates” for the 

completion of third Salmonella sets. It is clear that too much time 

elapse not only between sample sets but in some cases, within a sample 

set. PEER is specifically looking at this issue as part of its review 

of the Salmonella testing program and will make recommendations 

quickly. Additionally, a trend analysis data center is being 

established at the FSIS TSC in Omaha, NE. The function of this office 

will be to review test results and other data concerning the 

effectiveness and safety of a plants operations and provide an early 

warning system of problems.



Enclosed you will find additional technical and editorial comments. If 

you have any questions, please contact Ronald F. Hicks, Assistant 

Administrator for Program Evaluation, Enforcement and Review.



Sincerely,



William J. Hudnall Acting Administrator:



Signed by William J. Hudnall:



Enclosure:



[End of comments]



The following are GAO’s comments on the August 26, 2002, letter from 

the U.S. Department of Agriculture.



GAO’s Comments:



1. USDA believes the report does not fully acknowledge FSIS’s progress 

and continuous efforts to ensure that all plants meet regulatory 

requirements. It noted that FSIS has placed significant resources into 

the processes and systems that provided the data for our study and that 

FSIS has efforts ongoing to evaluate the same data. Our draft report 

acknowledges that in response to reports from GAO and USDA’s Office of 

Inspector General, FSIS initiated the food safety systems correlation 

reviews and in-depth verification reviews and recognizes that these 

reviews have been useful in identifying problems with HACCP 

implementation. USDA describes a number of actions that FSIS has 

recently taken or is planning to take that are consistent with our 

recommendations. Many of these actions, if fully carried out, may go a 

long way toward addressing the problems we found in FSIS’s oversight 

and enforcement of HACCP. However, diligent management attention will 

be needed to ensure this.



2. USDA believes that the title of the report is misleading. We 

disagree. We believe the title accurately reflects the concerns 

detailed throughout the body of the report.



3. The July 25, 2002, FSIS notice 28-02--Action to Be Taken in 

Establishments Subject to Salmonella Testing--does not establish 

specific time frames for actions to be taken at plants that fail a 

second set of Salmonella tests as we recommend.



4. While the September 5, 2001, notice 36-01--Rules of Practice--

identifies FSIS’s various enforcement tools and general circumstances 

in which each type would be appropriate, it does not establish clear, 

consistent criteria for inspectors to use when considering whether to 

recommend a suspension as an enforcement action in response to 

repetitive violations. We are recommending that FSIS do so.



5. FSIS’s August 9, 2002, notice 29-02--HACCP Verification Procedures 

and the 30-day Reassessment Letter--addresses our recommendation for 

the need for additional training for inspectors on their roles and 

responsibilities. However, FSIS’s efforts to provide training for 

inspectors on their roles and responsibilities, in response to our 

previous report recommendation, were not fully effective. We are 

keeping this recommendation because it is too early to tell whether 

this effort will be effective.



6. Our draft report acknowledged that FSIS had established a new office 

to ensure that its programs and policies are implemented and monitored 

correctly. While the review of the Salmonella testing program does not 

directly address any of our recommendations, it may help FSIS to 

provide inspector training and procedures for field supervisors and 

district managers to oversee inspector activities--which we do 

recommend--with respect to Salmonella requirements.



7. FSIS directive 5000.1--Enforcement of Regulatory Requirements in 

Establishments Subject to HACCP System Regulations (including 

regulations on Sanitation Standard Operating Procedures, E. coli 

Testing and Criteria, and Salmonella Performance Standards)--should 

help FSIS address several of our recommendations, including those 

related to ensuring that all HACCP plans fully meet regulatory 

requirements and ensuring that plants take prompt actions to correct 

violations.



8. FSIS directive 4430.3--In-Plant Performance System (IPPS) Reviews--

should respond to our recommendation regarding procedures to monitor 

inspector activities. FSIS is planning to implement the new directive 

and begin training supervisors in October 2002. We are keeping this 

recommendation to ensure the directive is implemented as planned.



9. This new interactive computer tool that inspectors, supervisors, and 

managers can use to strengthen HACCP problem solving using fictional 

scenarios. It should help FSIS address our recommendations regarding 

inspector training and supervisor oversight of inspector activities.



10. The draft report acknowledged the budget constraints regarding the 

hiring of consumer safety officers.



11. FSIS is still testing the PBIS 5.0 and district early warning 

systems, which should provide useful data for FSIS managers to carry 

out their HACCP responsibilities.



12. FSIS does not believe that plants are given too much time to comply 

with regulatory requirements. We disagree. Our review of enforcement 

case files found that, on average, these cases were closed in 10 months 

rather than the 3 months (90 days) recommended by FSIS. In addition, 

while it may be FSIS’s policy to establish specific time frames for 

plants to make corrective actions, none of the 30 enforcement case 

files we examined for plants with suspensions in abeyance that were 

closed after the 90-day policy went into effect contained this 

information.



13. We were not able to determine whether there has been improvement in 

the average amount of time it takes FSIS to close HACCP-related 

suspensions in abeyance because the data FSIS provided us with included 

suspensions for both sanitation violations as well as those for HACCP 

violations. In addition, when they are closed, it is not known how the 

unspecified number of cases that currently remain open will affect the 

average closure time.



14. We did not question the appropriateness of plants having multiple 

suspensions. Rather, we questioned the appropriateness of placing 

suspensions in abeyance at plants that have had repeated problems or 

multiple problems. The examples cited in our report involved (1) a 

plant suspension that remained in abeyance even though the plant 

continued to violate the same requirements for which it had been 

originally suspended and (2) a plant that had multiple sequential 

suspensions placed in abeyance.



[End of section]



Appendix III: GAO Contacts and Staff Acknowledgments:



GAO Contacts:



Lawrence J. Dyckman (202) 512-3841

J. Erin Lansburgh (202) 512-3017:



Acknowledgments:



In addition to those named above, Leo G. Acosta, Judy K. Hoovler, 

James L. Ohl, and Stephen D. Secrist made key contributions to this 

report.



[End of section]



FOOTNOTES



[1] According to FSIS, the approximately 5,000 HACCP plants include 

about 1,200 plants that slaughter and/or process meat, 300 that 

slaughter and/or process poultry, and 3,500 plants that slaughter and/

or process both meat and poultry. FSIS also inspects sanitation at, 

among others, plants that store or ship meat or poultry products. These 

plants are not required to have HACCP plans because they do not produce 

products.



[2] See U.S. General Accounting Office, Meat and Poultry: Improved 

Oversight and Training Will Strengthen New Food Safety System, GAO/

RCED-00-16 (Washington, D.C.: Dec. 8, 1999).



[3] See U.S. Department of Agriculture, Office of Inspector General, 

Food Safety and Inspection Service: Implementation of the Hazard 

Analysis and Critical Control Point System, Report No. 24001-3 

(Washington, D.C.: June 2000).



[4] See 21 U.S.C. 601 et seq. for meat and 21 U.S.C. 451 et seq. for 

poultry.



[5] According to FSIS, the agency selected Salmonella for testing 

because, among other reasons, it is the most common bacterial cause of 

foodborne illness and intervention strategies to reduce the presence of 

Salmonella on raw products should be effective against other pathogens.



[6] The zero tolerance standard also applies to contamination from the 

contents of the animal’s digestive system and its mammary glands 

(milk).



[7] Plants that operate more than one production process, such as 

animal slaughter and the preparation of a cooked product, must 

establish a HACCP plan for each process. As a result, some plants have 

multiple HACCP plans.



[8] We did not talk to inspectors to find out if they identified 

violations but did not document them in noncompliance records. But 

because they are required to do so, we assumed that they documented all 

violations they found.



[9] From January 1, 2002 to June 30, 2002, FSIS conducted an additional 

10 in-depth verification reviews. These reviews fell outside of the 

time period we used for analysis. 



[10] Following the in-depth verification review, one plant had the 

second set Salmonella failure overturned on appeal. In another 

instance, FSIS conducted an in-depth verification review at a plant 

that failed three of its past four sets of Salmonella tests; the 

failure that triggered the review was the first set of a new series of 

tests. 



[11] As of April 4, 2002, of the five plants that failed a third 

consecutive set of tests, one had failed a fourth set, two had passed, 

one had sampling in progress, and one had not yet begun further 

sampling.



[12] At plants participating in this project, plant personnel, instead 

of FSIS inspectors, examine each carcass to make an initial 

determination whether it is unacceptable and should be removed from the 

slaughter line. A reduced number of FSIS inspectors are still at each 

plant to ensure that safety and quality standards are met.



GAO’s Mission:



The General Accounting Office, the investigative arm of Congress, 

exists to support Congress in meeting its constitutional 

responsibilities and to help improve the performance and accountability 

of the federal government for the American people. GAO examines the use 

of public funds; evaluates federal programs and policies; and provides 

analyses, recommendations, and other assistance to help Congress make 

informed oversight, policy, and funding decisions. GAO’s commitment to 

good government is reflected in its core values of accountability, 

integrity, and reliability.



Obtaining Copies of GAO Reports and Testimony:



The fastest and easiest way to obtain copies of GAO documents at no 

cost is through the Internet. GAO’s Web site (www.gao.gov) contains 

abstracts and full-text files of current reports and testimony and an 

expanding archive of older products. The Web site features a search 

engine to help you locate documents using key words and phrases. You 

can print these documents in their entirety, including charts and other 

graphics.



Each day, GAO issues a list of newly released reports, testimony, and 

correspondence. GAO posts this list, known as “Today’s Reports,” on its 

Web site daily. The list contains links to the full-text document 

files. To have GAO e-mail this list to you every afternoon, go to 

www.gao.gov and select “Subscribe to daily E-mail alert for newly 

released products” under the GAO Reports heading.



Order by Mail or Phone:



The first copy of each printed report is free. Additional copies are $2 

each. A check or money order should be made out to the Superintendent 

of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 

more copies mailed to a single address are discounted 25 percent. 

Orders should be sent to:



U.S. General Accounting Office

441 G Street NW, Room LM

Washington, D.C. 20548:



To order by Phone: Voice: (202) 512-6000 

TDD: (202) 512-2537

Fax: (202) 512-6061



To Report Fraud, Waste, and Abuse in Federal Programs:



Contact:



Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470



Public Affairs:



Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800

U.S. General Accounting Office, 441 G Street NW, Room 7149 

Washington, D.C. 20548: