Workplace Safety and Health: Safety in the Meat and Poultry	 
Industry, While Improving, Could Be Further Strengthened	 
(12-JAN-05, GAO-05-96). 					 
                                                                 
Because meatpacking is one of the most dangerous industries in	 
the United States, we were asked to provide the Congress with	 
information on the characteristics of workers in the meat and	 
poultry industry and the conditions in which they work, the types
of injuries and illnesses these workers incur, how injury and	 
illness rates have changed over the past decade, and factors that
may have affected these rates. We were also asked to determine	 
what is known about the effectiveness of the Occupational Safety 
and Health Administration's (OSHA) efforts to improve safety and 
health in the meat and poultry industries.			 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-96						        
    ACCNO:   A15498						        
  TITLE:     Workplace Safety and Health: Safety in the Meat and      
Poultry Industry, While Improving, Could Be Further Strengthened 
     DATE:   01/12/2005 
  SUBJECT:   Accident prevention				 
	     Data collection					 
	     Health hazards					 
	     Inspection 					 
	     Labor force					 
	     Meat packing industry				 
	     Occupational health and safety programs		 
	     Occupational health standards			 
	     Occupational safety				 
	     Population statistics				 
	     Poultry industry					 
	     Safety regulation					 
	     Safety standards					 
	     Working conditions 				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-96

                 United States Government Accountability Office

  GAO 	Report to the Ranking Minority Member, Committee on Health, Education,
                        Labor, and Pensions, U.S. Senate

January 2005

WORKPLACE SAFETY AND HEALTH

Safety in the Meat and Poultry Industry, while Improving, Could Be Further
                                  Strengthened

GAO-05-96

[IMG]

January 2005

WORKPLACE SAFETY AND HEALTH

Safety in the Meat and Poultry Industry, while Improving, Could Be Further
Strengthened

                                 What GAO Found

The largest proportions of workers in the meat and poultry industry,
according to the Bureau of Labor Statistics (BLS), are young, male, and/or
Hispanic. Although the majority of workers are citizens, an estimated 26
percent of them are foreign-born noncitizens. They work in hazardous
conditions involving loud noise, sharp tools, and dangerous machinery.
Many workers must stand for long periods of time wielding knives and hooks
to slaughter or process meat on a production line that moves very quickly.
Workers responsible for cleaning the plant must use strong chemicals and
hot pressurized water.

While, according to BLS, injuries and illnesses have declined over the
past decade, the meat and poultry industry still has one of the highest
rates of injury and illness of any industry. The most common injuries are
cuts, strains, cumulative trauma, and injuries sustained from falls, but
more serious injuries, such as fractures and amputation, also occur.
According to BLS, the injury and illness rate for the industry has
declined from an estimated 29.5 injuries and illnesses per 100 full-time
workers in 1992 to 14.7 in 2001. Injury and illness rates can be affected
by many factors, such as the amount and quality of training, employee
turnover rates, increased mechanization, and the speed of the production
line.

Some evidence suggests that OSHA's efforts have had a positive impact on
the injury and illness rates of workers in meat and poultry plants.
However, while the criteria OSHA uses to select plants for
inspection-which focus on plants with relatively high injury and illness
rates-are reasonable, OSHA could improve its selection process by also
considering trends in plants' injury and illness rates over time. In
addition, it is difficult to assess the effectiveness of OSHA's efforts
because the agency does not assign a unique identifier to each plant,
making it hard to compare the data it collects on specific plants' injury
and illness rates with the information the agency collects on the results
of its plant inspections and other programs.

                     Production Line at a Meatpacking Plant

Source: Gail A. Eisnitz/ Humane Farming Association.

                 United States Government Accountability Office

Contents

  Letter

Results in Brief
Background
Meat and Poultry Workers Tend to Be Young, Male, and/or

Hispanic, and Face Hazardous Working Conditions

Meat and Poultry Workers Are Injured in a Variety of Ways, and
Their Injury and Illness Rates, though Declining, Remain among
the Highest of Any Industry

While OSHA's Programs May Have Improved the Safety and Health
of Meat and Poultry Workers, Programmatic Weaknesses Make
Determining Effectiveness Difficult

Conclusions
Recommendations for Executive Action
Agency Comments

                                       1

                                      3 6

15

21

33 45 46 47

Appendix I Scope and Methodology

Appendix II Interviews with Cleaning and Sanitation Companies 58

Appendix III 	OSHA's Study on Its Impact Using Establishment-Specific
Targeting of Interventions

Appendix IV 	Comments from the Occupational Safety and Health
Administration

Appendix V Comments from the U.S. Department of Agriculture 63

  Appendix VI 	Comments from the U.S. Department of Health and
  Human Services 65

Appendix VII

Comments from the Bureau of Labor Statistics

67

Appendix VIII GAO Contacts and Staff Acknowledgments

69

GAO Contacts 69 Staff Acknowledgments 69

Tables

Table 1: Types of Potentially Hazardous Working Conditions in Meat and
Poultry Plants 20 Table 2: OSHA Inspections in the Meat and Poultry
Industry, 1995 to 2004 34 Table 3: Survey Sample Sizes, Disposition, and
Response Rates 56

Figures

Figure 1: Location of U.S. Meat and Poultry Plants, September 2004 6
Figure 2: Production Line at a Poultry Plant 9 Figure 3: Workers in the
Meat and Poultry Industry, by Race, 2003 16 Figure 4: Location of U.S.
Meat and Poultry Plants with More than

500 Employees, September 2004 18 Figure 5: Nature of Injuries Sustained by
Meat and Poultry Workers and Parts of the Body Affected 24 Figure 6:
Safety and Other Equipment Worn by Meat and Poultry Production Workers 25

Figure 7: Trends in Injury and Illness Rates in the Meat and Poultry
Industry, Compared with Trends in All U.S. Manufacturing, 1992 to 2001 27

Abbreviations

BLS Bureau of Labor Statistics
CPS Current Population Survey
HHS U.S. Department of Health and Human Services
NIOSH National Institute of Occupational Safety and Health
OSHA Occupational Safety and Health Administration
SST site-specific targeting
USDA U.S. Department of Agriculture

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

January 12, 2005

The Honorable Edward M. Kennedy Ranking Minority Member Committee on
Health, Education,

Labor, and Pensions United States Senate

Dear Senator Kennedy:

According to data collected by the U.S. Department of Labor's Bureau of
Labor Statistics (BLS), in 2003, an estimated 527,000 workers were
employed in the animal slaughtering and processing industry.1 According to
the U.S. Department of Agriculture (USDA), in 2004 there were about 5,700
meatpacking and processing plants in the United States. The modern
meatpacking and processing plant is a complex and highly organized
structure, developed for the streamlined slaughter and progressive
disassembly of animals. The industry includes plants in which animals are
slaughtered and cut into pieces, with some facilities also cooking and
packaging the meat for consumption, as well as plants in which meat
products, such as sausage and ham, are produced by adding ingredients to
the meat. The meat processed includes red meat such as beef, veal, pork,
and lamb, and poultry such as chicken and turkey.

Because meatpacking is one of the most dangerous industries in the United
States, you asked us to (1) describe the characteristics of workers in
meat and poultry slaughter and processing plants and the conditions in
which they work; (2) identify the types of injuries and illnesses workers
in meat and poultry slaughter and processing plants incur, how the injury
and illness rates have changed over the past decade, and the factors that
may affect these rates; and (3) determine what is known about the
effectiveness of the Occupational Safety and Health Administration's

1This estimate has a 95 percent confidence interval from 470,783 to
584,003. All demographic estimates for the meat and poultry industry in
this report are based on the March 2004 or the March 1995 Current
Population Survey (CPS) and refers to workers in the animal slaughtering
and processing industry. Unless otherwise noted, CPS percentage estimates
have 95 percent confidence intervals of plus or minus 8 percentage points
of the estimate, and all other estimates have confidence intervals of
within plus or minus 14 percent of the estimate itself. See appendix I for
more information.

(OSHA) efforts to improve safety and health at meat and poultry slaughter
and processing plants.

To respond to your request, we reviewed literature on the industry and
interviewed officials from OSHA and other federal agencies, such as USDA
and the Centers for Disease Control and Prevention's National Institute
for Occupational Safety and Health (NIOSH) within the Department of Health
and Human Services (HHS), as well as individuals from contract cleaning
and sanitation companies, unions, advocacy groups, and key trade
associations. We obtained and analyzed

o  	data on worker demographics from BLS's Current Population Survey
(CPS),

o  	data on workplace injuries and illnesses from BLS's Survey of
Occupational Injuries and Illnesses,

o  data on fatalities from BLS's Census of Fatal Occupational Injuries,

o  	data from OSHA's inspections database (the Integrated Management
Information System),

o  	worksite-specific injury and illness data that OSHA uses to target
specific worksites for inspection (the OSHA Data Initiative),

o  	data on the number and location of meat and poultry plants from USDA's
Performance Based Inspection System, and

o  lists of plants that participate in OSHA's cooperative programs.2

In addition, we visited six meat and poultry plants and two OSHA area
offices and interviewed inspectors at four additional area offices,
selected because the offices had performed the most inspections of meat
and poultry plants. Finally, we conducted a survey of meat and poultry
plants to obtain data on their workers, factors that affect their injury
and illness

2We focused on the three-digit Standard Industrial Classification code
201-the meat products industry. When we refer to the "meat and poultry
industry," we are referring to companies in the meat products industry,
code 201. When we refer to the meatpacking industry, we are referring to
companies in the four-digit code 2011; when we refer to the sausages and
other prepared meat products industry, we are referring to companies in
code 2013; and when we refer to the poultry slaughtering and processing
industry, we are referring to companies in code 2015.

Results in Brief

rates, plants' efforts to improve safety and health, and plants'
interactions with OSHA.3 See appendix I for detailed information on the
scope and methodology for our work. We performed our work in accordance
with generally accepted government auditing standards between January 2004
and September 2004.

According to CPS data, in 2003, the largest proportions of workers in the
meat and poultry industry tended to be young (43 percent under age 35),
male (65 percent), and/or Hispanic (42 percent), with meat and poultry
workers in general laboring in hazardous conditions involving loud noise,
sharp tools, and dangerous machinery. Although the data show that the
majority of workers were citizens, a large proportion, an estimated 26
percent, were foreign-born noncitizens. Generally, meat and poultry
workers are employed in larger plants located in the South and Midwest and
earn a median salary of about $21,320 per year, much less than the typical
pay for workers in all manufacturing industries of about $33,500 per year.
The type of work performed and the plant environment expose workers to
many hazards. The work is physically demanding, repetitive, and often
requires working in extreme temperatures-such as in refrigeration units
that range from below zero to 40 degrees Fahrenheit- and plants often have
high turnover rates. Workers often stand for long periods of time on
production lines that move very quickly, wielding knives or other cutting
instruments used to trim or remove portions of the carcasses. Conditions
at the plant can also be loud, wet, dark, and slippery. Workers
responsible for cleaning the plant must use strong chemicals and hot
pressurized water to clean inside and around dangerous machinery, and may
experience impaired visibility because of steam.

Meat and poultry workers sustain a range of injuries, including cuts,
burns, and repetitive stress injuries, and while, according to BLS,
injuries and illnesses in the meat and poultry industry declined from 29.5
injuries and illnesses per 100 full-time workers in 1992 to 14.7 in 2001,
the rate was among the highest of any industry. Similarly, though not
comparable with these data because of recent changes in OSHA's
record-keeping requirements, statistics for 2002 indicate that injury and
illness rates in the meat and poultry industry remain high in relation to
those of other

3Our survey sample was designed so that we would be able to draw
inferences from the study population. However, because we received an
overall response rate to our survey of only 23 percent, we cannot
generalize the responses we received to the entire population of meat and
poultry producers. See appendix I for more detailed information on the
survey.

industries. While the most common injuries are cuts, strains, cumulative
trauma caused by repetitive cutting motions, and injuries sustained from
falls, more serious injuries, such as fractures and amputation, also
occur. For example, according to OSHA data, a worker died when he
attempted to replace his knife in the scabbard hanging from his belt,
missed the opening, and pushed the knife into his leg, severing his
femoral artery. In addition, some workers become ill because of exposure
to chemicals, blood, and fecal matter, which can be exacerbated by poor
ventilation and extreme temperatures. Because of the many hazards inherent
in meat and poultry plants and the type of work performed, the dramatic
decline in the industry's injury and illness rates has raised a question
about the validity of the data on which these rates are based. Several
factors can affect the rates of injury and illness, such as an emphasis on
safety by employers or employees, the amount and quality of training,
employee turnover rates, and the speed of the production line. However,
the degree to which these factors affect injury and illness rates is
difficult to assess. Some experts believe, for example, that faster line
speeds increase workers' risk of injury. OSHA officials told us that while
they believed that slowing the speed of the production line could help
reduce the number of injuries to workers, they do not have the data on the
effect of line speed on worker safety needed to question, in general
terms, the process of setting line speed or to assess the appropriate
speed at which the lines should operate.

Though certain weaknesses complicate assessments of OSHA's efforts to
improve safety and health at meat and poultry plants, some evidence
suggests that the agency's efforts have had a positive impact on the
injury and illness rates of workers in this industry. For example, in
2003, OSHA conducted inspections of almost 200 meat and poultry plants
that, according to the agency and some plant officials we interviewed,
resulted in many safety and health improvements. Similarly, some evidence
suggests that OSHA's cooperative programs have had a positive impact on
the safety and health of workers. For example, a program initiated by
OSHA's Omaha Area Office, in which it partnered with several meatpacking
plants in the state to share best safety practices, has, according to
OSHA, improved worker safety and health in plants in Nebraska. The agency
has not, however, implemented similar programs in other areas with large
concentrations of meatpacking plants or extended the program to poultry
plants. In addition, the criteria OSHA uses to select plants for
inspection, while reasonable, may not trigger inspection of some at-risk
plants. Currently, OSHA's selection criteria target worksites in
industries with high rates of injury and illness. OSHA also selects a
small number of worksites with low injury and illness rates for inspection
in order to ensure that they are not underreporting injuries and
illnesses, and

randomly selects worksites from high-hazard industries for audits that
verify their injury and illness rates. However, because OSHA's selection
criteria do not require the agency to examine trends in plants' injury and
illness rates over time-and the agency does not attempt to examine these
trends-OSHA may not detect dramatic decreases in these rates that could
raise questions as to the accuracy of the figures. Furthermore, the injury
and illness data on which OSHA bases its selection of plants for
inspection are incomplete, because they do not include injuries and
illnesses incurred by cleaning and sanitation workers not employed
directly by the plants. These workers are not classified by BLS as working
in the meat and poultry industry, although they labor in the same plants
and under working conditions that can be even more hazardous than those of
production workers. Finally, because OSHA does not assign a unique
identifier to each plant for which data are collected, it is difficult to
assess the success of its efforts by comparing information about specific
plants across its databases.

This report contains recommendations for strengthening OSHA's efforts to
improve the safety and health of workers at meat and poultry slaughter and
processing plants by, among other things, adjusting its criteria for
selecting plants for inspection and audits to include those that have had
large reductions in their injury and illness rates over time, and changing
the way it collects data on plants in order to make it easier to measure
the impact of its programs. The report also makes a recommendation jointly
to OSHA and USDA and another to HHS. In their written comments on our
report, OSHA, USDA, and HHS generally agreed with the report's findings,
conclusions, and recommendations. BLS also provided us with written
comments, suggesting several technical corrections that were incorporated
throughout the report, as appropriate.

Background 	According to USDA, there were about 5,700 total meat and
poultry plants in the United States as of September 2004.4 Most of
these-about 4,400- had fewer than 40 employees, and about half of them are
even smaller, with fewer than 10 employees. Figure 1 shows the location of
all meat and poultry plants, regardless of size, in the United States as
of September 1, 2004.

Figure 1: Location of U.S. Meat and Poultry Plants, September 2004

4USDA's primary responsibility in meat and poultry plants is to administer
a comprehensive system of inspection laws designed to ensure that meat and
poultry products moving in interstate and foreign commerce for use as
human food are safe, wholesome, and accurately labeled.

Over the past 25 years, the meat and poultry industry has consolidated, as
today's leading firms built very large plants and some independent firms
disappeared or were bought by larger firms. While many small plants
remain, a few large companies have gained control of the lion's share of
the market. Today, the top four meatpacking companies slaughter, process,
and package about 80 percent of the beef cattle in the United States, and
the top four pork producers control nearly 70 percent of the market. The
poultry industry is nearly as concentrated, with the top five companies
maintaining a market share in excess of 50 percent. Consolidation of the
various meat industries occurred, in large part, because of innovations in
technology and the relocation of plants near the source of livestock.

Industry consolidation has been accompanied by significant changes in the
relations between organized labor and the management of meat and poultry
plants. According to a report by USDA's Economic Research Service, in
1980, 46 percent of workers in the meat products industry were union
members, a figure that had remained stable since the 1970s.5 However, by
the end of the 1980s, union membership had fallen to 21 percent. Declining
rates of unionization coincided with increases in the use of immigrant
workers, higher worker turnover, and reductions in wages. Immigrants make
up large and growing shares of the workforces at many plants. Labor
turnover in meat and poultry plants is quite high, and in some worksites
can exceed 100 percent in a year as workers move to other employers or
return to their native countries. The frequent movement of immigrant
workers among plants and communities limits the opportunities of unions to
organize meat and poultry workers.

Most of today's facilities are designed for an orderly flow from point of
entry of the living animal into the plant to the finished food product.
The animal enters the production facility and proceeds directly to the
kill floor area, where slaughter occurs. The carcass is beheaded,
eviscerated, and chilled for several hours. It is then taken to the
cutting floor, where it is cut into smaller cuts of meat. The new
processing methods-breaking down carcasses into small, vacuum-packed
portions of meat that can be shipped directly to supermarkets-have
transformed the work into an

5Consolidation in U.S. Meatpacking, by James M. MacDonald, Michael E.
Ollinger, Kenneth E. Nelson, and Charles R. Handy. Food and Rural
Economics Division, Economic Research Service, U.S. Department of
Agriculture. Agricultural Economic Report No. 785, Washington, D.C.:
February 2000.

assembly line operation requiring workers to perform an increased number
of repetitive motions.

While slaughterhouses have come to rely on greater mechanization over the
last several decades, much of the work is still done by hand, particularly
when animals vary in size, shape, and weight. The main slaughtering steps
of evisceration and cutting are generally done by hand, using knives.
Figure 2 shows a typical assembly line operation at a poultry plant.

Figure 2: Production Line at a Poultry Plant

Source: Photo used with permission.

OSHA, established after the passage of the Occupational Safety and Health
Act in 1970, is the federal agency within the Department of Labor
responsible for protecting the safety and health of workers in meat and

poultry plants.6 OSHA performs a number of functions, including
establishing safety and health standards, conducting routine inspections,
and conducting investigations in response to complaints from workers and
incidents such as fatalities.

Regional administrators in each of OSHA's 10 regional offices oversee the
enforcement of federal policies within their own regions. Each region is
composed of area offices-of which there are 80 in total-overseen by area
directors. The area directors oversee compliance officers, who are
responsible for conducting inspections and following up on complaints, and
compliance assistance specialists, who provide assistance to organizations
and employers that participate in OSHA's cooperative programs. Compliance
assistance specialists also help employers correct hazards identified
during inspections.

To determine which plants to inspect, OSHA relies on BLS data on injuries,
illnesses, and fatalities by industry. BLS surveys a sample of employers
annually (182,800 worksites were surveyed for 2002) and asks them to
report information on the number of work-related injuries and illnesses
that occur at their worksites. This information comes from injury and
illness records that most private industry employers with more than 10
employees are required by OSHA to maintain. From this information, BLS
calculates industry-level injury and illness rates. BLS also identifies
fatalities from an annual census of all 50 states, the District of
Columbia, and New York City, which report on all work-related fatalities
within their jurisdictions. BLS requires the reporting entities to
corroborate reports of fatalities from multiple sources, such as death
certificates, medical examiners' reports, media reports, and workers'
compensation claims. BLS makes injury, illness, and fatality data
available at the national, as well as at the state, level.

OSHA uses two approaches to ensure general employer compliance with
federal safety and health laws and regulations-enforcement and cooperative
programs. Enforcement, which represents the preponderance of agency
activity, is carried out primarily by using compliance officers to

6Under the terms of the act, states may assume responsibility for
occupational safety and health enforcement through the mechanism of an
OSHA-approved state plan. Twenty-one "state-plan states" operate such
programs with responsibility for most private sector OSHA enforcement in
their states. State plans operate under authority of state law, adopt and
enforce their own standards (which must be "at least as effective" as
federal OSHA's), and set their own goals and priorities for enforcement
and compliance assistance.

inspect employer worksites. Worksites and employers that fail to meet
federal safety and health standards face sanctions, such as paying
penalties for violations of health and safety standards. OSHA's
cooperative approach invites employers to collaborate with the agency
through a number of different programs and uses a variety of incentives to
encourage employers to reduce hazards and institute practices that foster
safer and healthier working conditions.

OSHA's Enforcement Efforts

OSHA selects worksites in selected industries for inspection through its
site-specific targeting (SST) program and through national and local
programs that focus on specific hazards. The SST program focuses on
employers with more than 40 employees, who are required to record all
injuries and illnesses on a log and make this information available to
OSHA.7 Of the almost 40,000 inspections OSHA conducted in 2003, about
2,000 were SST inspections, and about 25,000 were conducted through
national and local emphasis programs.8 The agency also conducts
inspections when fatalities or serious injuries occur and when workers
file complaints about serious safety and health hazards. These inspections
constitute nearly half of the total inspections OSHA conducts annually.

For its SST program, OSHA obtains specific information-such as employer
names and addresses-for all worksites with 40 or more employees,
approximately 140,000 worksites each year. OSHA then selects a portion of
these worksites (approximately 80,000) in the industries with the highest
injury and illness rates,9 and sends them a survey form that requires them
to report (1) the average number of employees who worked for them during
the previous calendar year, (2) the total hours the employees worked
during the previous year, and (3) summary injury and illness data from
their OSHA logs. From this information, OSHA computes the worksites'
injury and illness rates and sends those with relatively high rates a
letter informing them that they may be inspected. Finally, OSHA

7The SST program also focuses on industries outside of manufacturing, with
lost workday case rates above a certain level (5.0 or greater for its 2004
SST program), as reported by BLS. The nonmanufacturing industries included
in the survey for OSHA's 2004 SST program were within the major industry
categories of Agriculture, Transportation, Wholesale and Retail Trade, and
Health Services.

8Worksites in the construction industry are not selected for inspection
under OSHA's SST program. However, through its other inspection
initiatives, 22,724 of OSHA's 39,720 total inspections were of
construction worksites in 2003.

9All manufacturing industries are considered as having high injury and
illness rates.

compiles SST inspection targeting lists containing the names of worksites
with relatively high injury and illness rates for inspection.10

OSHA also has special emphasis programs that focus on a particular safety
or health hazard or the hazards of a specific industry, selected by the
agency's headquarters office for attention. While OSHA's headquarters
provides direction to its area offices in implementing these national
emphasis programs, the area offices have considerable flexibility in
selecting actual worksites for attention. In addition, regional and area
offices use regional and local emphasis programs to highlight industries
or hazards within their jurisdictions that they believe are especially
hazardous.

Because musculoskeletal disorders are prevalent in several industries-
including the meat and poultry industry-but there is no specific standard
that allows OSHA to cite employers for hazards relating to these injuries,
the agency designed a four-pronged approach to address these injuries that
focuses on industries and employers with known high injury and illness
rates.11 The approach includes (1) developing industry or taskspecific
guidelines for a number of industries based on current incidence rates and
available information about effective and feasible solutions; (2)
conducting inspections for ergonomic hazards, issuing citations under the
general duty clause of the Occupational Safety and Health Act,12 and
issuing ergonomic hazard alert letters where appropriate; (3) providing
assistance to businesses, particularly small businesses, and helping them

10OSHA sends primary and secondary lists to its area offices in federal
OSHA states. These offices are expected to visit all worksites identified
on the primary list and inspect worksites on the secondary list as
resources allow. OSHA sends information on additional worksites in
state-plan states to the appropriate state agencies, which are expected to
have an effective high hazard inspection targeting system. All but 4 of
the 21 state-plan states participate in the data gathering program that
would make establishment-level SST-type data available to them for efforts
such as targeting and program evaluation.

11Musculoskeletal disorders include conditions such as tendonitis, carpal
tunnel syndrome, and lower back injuries. Symptoms of these disorders can
include swelling in the joints, limited range of motion, numbness or
tingling sensations, and loss of strength. Events or exposures that can
lead to the injury or illness are bodily reaction/bending, climbing,
crawling, reaching, twisting, overexertion, and repetition.

1229 U.S.C. S: 654(a)(1). This clause is used to cite serious hazards
where no specific OSHA standard exists to address the hazard, as is the
case with ergonomic stressors. According to OSHA, when it uses this clause
to cite an employer, the agency must demonstrate that (1) the employer
failed to keep the workplace free of a hazard to which employees were
exposed, (2) the hazard was causing or likely to cause death or serious
physical harm, (3) the hazard was recognized, and (4) a feasible means of
abatement for the hazard exists.

to proactively address ergonomic issues in the workplace; and (4)
chartering an advisory committee authorized to, among other things,
identify gaps in research about the application of ergonomics and
ergonomic principles in the workplace.

OSHA's Cooperative Programs

OSHA's cooperative programs provide incentives to employers, such as free
consultations, deferrals from SST inspections, and recognition for
exemplary safety and health management systems, for making improvements to
their safety and health management systems. OSHA has implemented these
programs incrementally to reach different employers and worksites in
various ways. OSHA has four primary programs: (1) the On-Site Consultation
Program, (2) the Voluntary Protection Programs, (3) the Strategic
Partnership Program, and (4) the Alliance Program.

The On-Site Consultation Program is a broad network of occupational safety
and health services primarily funded by federal OSHA, but is delivered by
the states. The service, which originated in 1974, focuses on helping
small employers comply with OSHA and state occupational safety and health
standards. The program assigns priority to companies in highhazard
industries and is offered free of charge to eligible employers.13 States
provide consultation visits at employers' requests in order to identify
safety and health hazards and discuss techniques for their abatement.
Small employers receiving consultation services may qualify for
recognition in the Safety and Health Achievement Recognition Program-part
of the On-Site Consultation Program-which exempts them from SST
inspections during the period that their certification is valid, either 1
or 2 years. Participants in this program, who are considered models for
good safety and health practices in their field, must have, at a minimum,
safety and health management systems in place to prevent and control
occupational hazards, as well as illness and injury rates below the
national average for their industry.

The Voluntary Protection Programs, established in 1982, recognize single
worksites with exemplary safety and health management systems. In calendar
year 2003, the average participating worksite had approximately 250
employees, and about 50 percent of the participating sites had 200

13The On-Site Consultation Program defines a small business as one with
fewer than 250 workers at the workplace where the consultation is
conducted and no more than 500 workers companywide.

employees or less. To participate in this program, employers must have
worksites that exceed OSHA standards and must commit to a process of
continual improvement. After receipt of a site's application, OSHA
conducts an onsite review of the site to verify the effectiveness of its
safety and health management system.

The Strategic Partnership Program, established in 1998, was designed to
address specific safety and health management systems in high-hazard
workplaces by promoting collaboration between employers, employees, other
affected organizations, and OSHA.14 Each partnership has a written
agreement that outlines goals (such as the reduction of injuries),
strategies, and measures, and identifies how the partners will work
together to achieve their desired results. Partnerships may focus on
specific hazards or industry issues, or may aim for broader impact through
focus on safety and health management systems. OSHA verifies partner
commitment and success in achieving goals, and while the program does not
offer employers exemption from inspection, it offers other incentives,
such as limiting SST inspections to only the most serious prevailing
hazards and reducing penalties for hazards cited during inspections.

OSHA's Alliance Program targets trade, professional, and other types of
organizations to work collaboratively with OSHA to promote workplace
safety and health issues. In contrast to OSHA's other three cooperative
programs, which typically include safety and health management systems at
specific employer worksites, alliance agreements focus on goals such as
training, outreach, and increasing awareness of workplace safety and
health issues. Alliance participants and their members are not exempt from
OSHA inspections and do not receive any enforcement-related incentives for
being part of an organization participating in an alliance. Instead, OSHA
officials informed us that trade and professional associations have used
the program to address existing and emerging workplace safety and health
issues, such as ergonomics.

In addition to these formal programs, OSHA conducts other compliance
assistance activities, such as outreach and training activities, to aid
employers in complying with OSHA standards and to educate employers on
what constitutes a safe and healthy work environment.

14While OSHA had partnership agreements prior to 1998, the Strategic
Partnership Program was not formalized until that year.

Meat and Poultry Workers Tend to Be Young, Male, and/or Hispanic, and Face
Hazardous Working Conditions

The meat and poultry workforce tends to be young (43 percent under age
35), male (65 percent), and/or Hispanic (42 percent). These
characteristics are more pronounced in the meat and poultry industry than
in the U.S. manufacturing sector overall. Meat and poultry workers
typically earn substantially less than workers in the U.S. manufacturing
sector as a whole. Most large plants are located in the Midwest and South,
and workers in the industry often work in difficult and dangerous
conditions.

Meat and Poultry Workers Are Predominantly Young, Male, and/or Hispanic,
and Most Large Plants Are Located in the Midwest and South

Workers in the meat and poultry industry tended to be younger than workers
in the manufacturing sector as a whole, and almost all of the workers are
employed on a full-time basis. According to CPS data, in 2003, the median
age of workers in the meat and poultry industry was 37 years.15 About 43
percent of all meat and poultry workers were under age 35, compared with
about 29 percent of all U.S. manufacturing workers.16 These workers also
tend to be male. In 2003, men made up 65 percent of the workforce in the
meat and poultry industry. In 2003, in this industry and in U.S.
manufacturing overall, about 95 percent of the employees worked
full-time.17

The racial composition of the meat and poultry workforce is
disproportionately Hispanic. As shown in figure 3, according to the CPS,
in 2003, about 42 percent of meat and poultry workers were Hispanic or
Latino, 32 percent were white, and 20 percent were black.18 These figures
compare with those for U.S. manufacturing as a whole, in 2003, where about
14 percent of the workforce was Hispanic, about 70 percent was white, and
about 9 percent was black. Further, the percentage of Hispanic or Latino
meat and poultry workers in 2003 reflects a 17 percent increase

15The 95 percent confidence interval for this median age is from 35 to 39
years old.

16All percentage estimates describing the workforce in this section are
CPS estimates, and have a 95 percent confidence interval of within plus or
minus 8 percentage points of the estimate itself.

17The percentage estimates for this industry and U.S. manufacturing are 96
and 95 percent, respectively. These percentages are not significantly
different at the 95 percent confidence level.

18In addition, about 2 percent were Asian or other Pacific Islander and 3
percent were American Indian or Alaska native. The CPS is a joint product
of the U.S. Census Bureau and BLS.

from 1994, when about 25 percent of meat and poultry workers were Hispanic
or Latino. In addition, in 1994, a larger percentage of the meat and
poultry workforce-46 percent-was white, and 25 percent was black.

Figure 3: Workers in the Meat and Poultry Industry, by Race, 2003

Foreign-born noncitizens are more highly represented within the meat and
poultry workforce than in manufacturing as a whole. A significant
proportion-about 26 percent-of all workers in this industry are
foreignborn noncitizens, compared with only about 10 percent of all
manufacturing workers in the United States. An even larger percentage of
the production and sanitation workers in the meat and poultry industry- 38
percent-are foreign-born noncitizens.19 In 1994, 28 percent of production
and sanitation workers were foreign-born noncitizens.

19Production and sanitation workers make up about 304,000 of the 527,000
total workers in the meat and poultry industries. The remaining workers in
the industry work in administrative, managerial, engineering, health care,
and transportation-related positions.

In certain areas of the United States, a number of communities have
concentrated groups of immigrant workers-including groups from regions
such as Central America, Southeast Asia, and Eastern Europe- who are
employed in the meat and poultry industry.20 In such areas, employees from
these immigrant groups may make up a relatively large percentage of the
workforce and population in and around meat and poultry plants. For
example, in 2000, the population of one county in Kansas (which, according
to USDA, was one of the largest meat-producing counties in the nation) was
about 43 percent Hispanic, compared with only 7 percent of the population
in the state. According to some industry officials, the increasingly
fragmented nature of the tasks in slaughtering and processing has
diminished the need for a skilled and more highly paid workforce, a fact
that supports the industry's recruitment and employment of unskilled
immigrant labor.

While plants are distributed throughout the United States, larger plants-
those with more than 500 employees-tend to be concentrated in particular
regions and produce the majority of the meat. Of these larger meat and
poultry plants, about 87 percent are located in the South and the Midwest,
54 percent and 33 percent, respectively. Another 9 percent are located in
the West and 4 percent in the Northeast. Figure 4 shows plants with more
than 500 employees.

20GAO, Community Development: Changes in Nebraska's and Iowa's Counties
with Large Meatpacking Plant Workforces, GAO/RCED-98-62, (Washington,
D.C.: Feb. 27, 1998).

Figure 4: Location of U.S. Meat and Poultry Plants with More than 500
Employees, September 2004

Meat and poultry workers tend to earn substantially less than

manufacturing workers in general. In 2003, meat and poultry workers

earned a median salary of about $21,320 per year, while manufacturing

workers earned about $33,500 per year. In addition, the rate of employee

turnover among meat and poultry workers can be high.21 A plant official

with whom we spoke indicated that some workers who are hired have no

intention of staying for a long period of time and approach employment at

meat and poultry plants as a temporary arrangement. According to some

experts, high turnover may benefit plants because they save on some

21The turnover rate is typically calculated by dividing the total number
of employees who left the plant during the most recent year by the total
number of employees.

costs, such as health benefits and vacation pay, while others argue that
high turnover is costly for plants because they must constantly recruit
and train new employees.

Workers Face Several Hazardous Conditions in Meat and Poultry Slaughter
and Processing Plants

The work environment in meat and poultry plants can be risky because of
the current procedures used in the industry. Meat and poultry plants
present risks greater than those faced by workers in many other
manufacturing operations. For example, production lines can require
workers to stand close together while wielding tools necessary for cutting
pieces of meat. Final product processing involves a number of packaging
machines and conveyors that can present a wide range of safety risks to
workers. Workers are also frequently handling or in close proximity to
sources of infectious diseases, such as those carried by animal tissues
and organs. Pathogens can infect workers from open abrasions or through
inhalation. For example, hydrogen sulfide, methane, and carbon dioxide can
be released from decomposing animal manure and waste. In addition, workers
are exposed to many chemicals, including a range of gases, such as
ammonia, and Freon. Table 1 summarizes the hazardous working conditions in
meat and poultry plants.

Table 1: Types of Potentially Hazardous Working Conditions in Meat and
Poultry Plants

Type of hazard

Description of hazard

Animal 	Workers can be injured by animals when they are unloaded and
brought into the plant. Incorrect stunning and slaughtering can result in
unpredictable and violent reactions. The movement of carcasses weighing up
to half a ton or more also poses a possible danger. Bodily fluids from
carcasses, such as blood and fat, can make floors wet and slippery.

Chemicals and pathogens

Workers, especially cleanup crews, are exposed to a number of products
that have strong chemicals, including disinfectants. In addition, workers
are exposed to ammonia used for refrigeration. Workers may also be exposed
to viruses, blood, fecal matter, and bacteria, such as Salmonella.

Temperature 	Some workers are exposed to very hot temperatures, used to
cook or cure meat. Workers are also exposed to very cold temperatures used
to preserve meat and facilitate processing. Frozen meat and poultry
products can require work in even colder temperatures. These problems are
compounded by wet conditions and high humidity. Cleanup crews spray
machinery, floors, and equipment with very hot water, causing steam that
can burn workers and impair vision from fogged safety goggles.

Machine and tool 	Many meat and poultry jobs still require the manual use
of knives, particularly in meatpacking plants where animals vary widely in
size and shape. Increasing mechanization, while reducing the number of
workers exposed to injury on processing lines, can increase the type and
severity of injuries by machines that cut, slice, saw, and grind. Large
objects, such as forklifts, are also a hazard.

Work stress 	Workers on some production lines perform identical motions
for long periods of time. Increasing mechanization can permit faster line
speeds, which in turn can further stress workers, who must keep up with
mechanical equipment.

Noise 	Some workers are exposed to loud machinery for prolonged periods.
Earplugs are required and may reduce ability to communicate warnings.

                             Source: GAO analysis.

Meat and Poultry Workers Are Injured in a Variety of Ways, and Their
Injury and Illness Rates, though Declining, Remain among the Highest of
Any Industry

Meat and poultry workers suffer high rates of many types of injuries and
illnesses, including those affecting the back, trunk, arms, fingers, and
wrists. Although injury and illness rates have declined over the last
decade, according to BLS, those in meat and poultry plants continue to be
among the highest of any industry.

Workers Sustain Many Different Types of Injuries and Illnesses

Workers in the meat and poultry industry, including contract cleanup and
sanitation workers, can suffer a host of serious injuries and illnesses,
most often musculoskeletal disorders. (See fig. 5 for an illustration of
the types of injuries workers suffer.) Many of the injuries-such as those
to the arms, hands, and wrists-are due to the repetitive motions
associated with the meat production process, such as performing the same
cutting motions over time, and can become crippling. For example, carpal
tunnel syndrome, caused by repetitive motion or cumulative trauma, can
severely damage a nerve running through the wrist.22

Workers can also be cut by their own knives during the butchering and
cutting processes. For example, according to an OSHA publication, one
worker in a meatpacking plant was blinded when the knife he was using to
pick up a ham prior to boning slipped out of the ham, striking him in the
eye.23 The report also described an incident in which another worker's
face was permanently disfigured when his knife slipped out of a piece of
meat and struck his nose, upper lip, and chin. In another incident,
according to OSHA, a worker who attempted to replace his knife in the
scabbard hanging from his belt missed the opening and pushed the knife
into his leg, severed his femoral artery, and died. In addition, workers
can be cut by

22Cumulative trauma can be caused by forceful exertions, repetitive finger
or wrist motions, tool vibrations, awkward wrist positions, or specific
repeated motions, and it can be exacerbated by extreme cold or humidity.

23Safety and Health Guide for the Meatpacking Industry, U.S. Department of
Labor, Occupational Safety and Health Administration, 1988, OSHA 3108.

the knives wielded by coworkers. According to OSHA's report, these
"neighbor cuts" are usually the direct result of overcrowded working
conditions.

Other injuries that workers can experience include respiratory irritation
and, in some instances, asphyxiation from exposure to pathogenic
respiratory substances. For example, workers have died from being overcome
by hydrogen sulfide gas and from drowning when they entered manure waste
pits or unknowingly worked near manure waste "lagoons" without taking the
proper precautions, such as conducting an air test and wearing a safety
harness and respirator; such precautions are particularly critical when
workers are in confined spaces. According to OSHA, one worker died from
chemical exposure after being sprayed with 400 pounds of toxic liquid
ammonia while attempting to fix a pipe.

Workers may also suffer injuries and illnesses from contact with animals.
If the animals are still dying when they are hung on the line, they may
struggle and thrash about wildly, resulting in injuries that range from
broken arms to permanent disfigurement and-in the most severe cases-
death. Contact with different bacteria can cause fever, headaches,
vomiting, diarrhea, and kidney damage. In addition, illnesses that can be
contracted from diseased animals continue to raise concerns. According to
USDA officials, the transmission of disease from animals to humans in the
meat and poultry industry is uncommon because of concerted efforts in the
United States and abroad. However, recent outbreaks of bovine spongiform
encephalopathy among cattle (commonly known as "mad cow disease") are
related to incidences of a disease that affects humans, according to the
Centers for Disease Control and Prevention. While beef consumption is
recognized as a mode of infection, it is unknown whether this disease can
be transmitted in other ways, such as exposure to waste or blood.

Workers can also suffer back injuries or other types of injuries from
overexertion, including sprains, strains, tears, hernias, and fatigue.
They can suffer injuries, and even death, from falling or being struck by
an object. For example, workers have been killed by falling ice and
forklift parts, and falls due in part to a lack of functioning safety
devices. In one incident, an employee was killed when a rack of sausage
fell from a manual overhead conveyor system and struck him.

Workers can be burned by heat sealant machines when they wrap meat.
Workers may also sever fingers or hands or even lose limbs on machines
that are either improperly locked or inadequately guarded. For example, in

2002, a sanitation worker at an Alabama plant lost both legs when another
worker activated the meat grinder in which he was standing. In another
incident, an employee dropped his knife into a meat grinder, reached in to
retrieve it, and suffered the amputation of his arm. In yet another
incident, an employee's fingers were amputated when they were caught in
the mixing and blending machine he was operating.

Workers can be injured by falling on slippery floors and exposure to
extreme heat or cold. Such cold temperatures can stress joints and
exacerbate existing conditions such as arthritis and cardiovascular
illnesses. As shown in figure 5, workers may sustain many types of
injuries, and several different parts of the body may be affected.

Figure 5: Nature of Injuries Sustained by Meat and Poultry Workers and Parts of
                               the Body Affected

As shown in figure 6, workers in the meat and poultry industry typically
wear several types of safety and other equipment in an effort to protect
themselves from injury and illness.

Figure 6: Safety and Other Equipment Worn by Meat and Poultry Production Workers

Injury and Illness Rates Have Declined but Remain among the Highest of Any
Industry

Injury and illness rates in the meat and poultry industry fell steadily
from 1992 to 2001, according to BLS data (see fig. 7).24 The meat and
poultry industry's annual rate of incidence of illness and injury in 2001,
at an estimated 14.7 cases per 100 workers, was about half its 1992 rate
of 29.5 cases.25 The incidence rate across all U.S. manufacturing dropped
to about two-thirds of its former rate over the same period, from 12.5
cases to 8.1 cases per 100 full-time workers.

24Injury and illness rates for 2002 are not comparable with 2001 and
previous years' rates because of changes to OSHA's record-keeping
requirements and changes in the way that OSHA requires companies to
categorize injuries and illnesses. These changes took effect January 1,
2002.

25All estimates of injury incidence rates in this report are based on BLS
data and have 95 percent confidence intervals of within plus or minus 14
percent of the estimated incidence rate. Confidence intervals for most
estimates in this report are narrower (more precise) than this. However,
rather than report confidence intervals for every incidence rate estimate
in this report, a broad conservative confidence interval is used to cover
all BLS incidence rate estimates. Additional information about these
estimates is contained in appendix I.

Figure 7: Trends in Injury and Illness Rates in the Meat and Poultry
Industry, Compared with Trends in All U.S. Manufacturing, 1992 to 2001

Total Recordable Case Rate (cases per 100 workers)

                                       35

                                       30

                                       25

                                       20

                                       15

                                       10

                                       5

                                       0

1992 1993 1994 1995 1996 1997 1998 1999 2000 Year

All meatpacking and poultry All U.S. manufacturing Source: GAO analysis of
BLS injury and illness data.

Despite this decrease, injury and illness rates among meat and poultry
plants remain among the highest of any industry. According to BLS data on
injuries and illnesses, in 2002, meatpacking plants recorded an average
annual injury and illness rate of 14.9 cases per 100 full-time workers;
sausages and other prepared meats plants recorded a rate of 10.9 cases;
and poultry plants recorded a rate of 9.7 cases. The average annual injury
and illness rate for all U.S. manufacturing was 7.2 cases.

Within the meat and poultry industry, the incidence rate for specific
injuries and illnesses, as reported by employers, dropped in recent years.
According to BLS data on injuries and illnesses, for example, carpal
tunnel injuries dropped from 24 cases per 10,000 workers in 1992 to 6.8
cases in

2001; strains and sprains dropped from 189.4 cases to 51.9 cases;
tendonitis dropped from 23.6 cases to 3.5 cases; cuts and punctures
dropped from 76.2 cases to 17.9 cases; chemical burns dropped from 9.6
cases to 4.4 cases; and amputations dropped from 5.3 cases to 3.2 cases.

Compared with workers in all U.S. manufacturing industries, meat and
poultry workers sustain a higher rate of certain injuries, such as
chemical burns, amputations, heat burns, tendonitis, and carpal tunnel
syndrome. In 2002, meatpacking workers suffered more of these types of
injuries, but relatively fewer sprains and strains and fractures. The rate
of injuries and illnesses involving repetitive motion in the meat and
poultry industry at 22.2 cases per 10,000 full-time workers was one and a
half times greater than the rate of 14.7 for all U.S. manufacturing in
2002.

A number of injuries sustained by meat and poultry workers are fatal;
according to BLS fatality data, from 1992 to 2002, 229 workers died from
their injuries. Of the 229 worker deaths, almost one-quarter occurred off
plant property, rather than during production, in transportation
accidents. The deaths that occurred in plants over this period included 60
that were caused by contact with objects and equipment (37 of these by
being caught in or compressed by equipment or objects, including running
machinery); 25 by falling; 35 from exposure to harmful substances; 4 from
fires and explosions; and 22 from assaults and violent acts, including
homicides.26

BLS's data on injuries and illnesses, however, may not accurately reflect
plants' incidences of injury and illness. OSHA, researchers, and union
officials have all stated that the underreporting of injuries and
illnesses is a problem in the meat and poultry industry. In the late
1980s, after observing what appeared to be underreporting of worker
injuries, OSHA's offices in region 7 focused their attention on the
meatpacking industry.27 Beginning with an exhaustive review and
reconstruction of a large Nebraska meatpacker's records, OSHA documented
dozens of cases of underreporting and assessed the company more than $2.5
million in

26In 2003, 18 additional deaths were reported as sustained by meat and
poultry workers. BLS and other federal statistical agencies are now
required to use new industry and occupational classifications designed to
reflect the most recent industries and occupations in the economy.
Therefore, the 2003 data are not comparable with prior years' fatality
data and are reported separately because, in some instances, the
occupational definitions in the new classification system are different
from those used previously.

27OSHA's region 7 covers Iowa, Kansas, Missouri, and Nebraska.

penalties. Because of OSHA's findings during this inspection and others
like it, Congress held hearings on the underreporting of occupational
injuries from March to September 1987.28 In 1987, after a National Academy
of Sciences review of the methods BLS used to collect employers' injury
and illness data highlighted several deficiencies, and in response to the
congressional hearings, BLS began a multi-year effort to redesign and test
an improved safety and health statistical system for collecting these
data, which was fully implemented in 1992.29 However, the accuracy of
employers' occupational injury and illness data remains a concern. OSHA
conducted a series of record-keeping inspections of meat and poultry
plants in region 7 throughout the late 1980s and 1990s. As a result of
these inspections, several plants were assessed penalties for
record-keeping violations, including five plants that were assessed
penalties ranging from $290,000 to $998,360. OSHA continues to find some
measure of underreporting of employers' injury and illness information
through the agency's record-keeping audits each year.

In addition, we reported in 1998 that the U.S. Immigration and
Naturalization Service (now the Citizenship and Immigration Services) had
often found illegal aliens employed in meatpacking plants; one agency
official estimated that up to 25 percent of workers in meatpacking plants
in Nebraska and Iowa were illegal aliens. As recently as March 2004, as
the result of an internal audit, one large meatpacking company found 350
undocumented workers employed in one of its plants in the Midwest. Because
large numbers of meat and poultry workers are immigrants-and perhaps
employed illegally-they may fear retaliation or loss of employment if they
are injured and cannot perform their work, and they may be hesitant to
report an injury. Furthermore, according to data from OSHA and academic
researchers published in a BLS periodical, some plants offer employees or
groups of employees incentives, such as money

28Underreporting of Occupational Injuries and Its Impact on Workers'
Safety, (Parts 1, 2, & 3) Hearings Before a Subcommittee of the Committee
on Government Operations, House of Representatives, 100th Congress,
Washington, D.C., March 19, 1987; May 6, 1987; and September 21, 1987.

29E.S. Pollack and D.F. Keimig, eds., Counting Injuries and Illnesses in
the Workplace: Proposals for a Better System, Washington, National
Research Council, National Academy Press, 1987. Beginning in 1992, survey
information on nonfatal incidents involving days away from work was
expanded to profile (1) the occupation and other demographics (e.g., age
and gender) of workers sustaining such injuries and illnesses, (2) the
nature of these disabling conditions and how they occurred, and (3) the
resulting time away from work.

or other prizes, for maintaining low injury and illness rates.30 According
to

the report, while these incentives may improve safety, they also may
discourage workers from reporting injuries that could result in their not
winning the incentive prize or preventing an entire group of workers from
obtaining the prize. In addition, some plants judge the performance of
line supervisors based on the number of days their workers go without an
injury or illness. These supervisors, also influenced by performance
incentives, may underreport injuries or encourage workers not to report
injuries or illnesses.31 Several of the plant officials we interviewed
told us

that they provide incentives and rewards to employees or groups of
employees who work for extended periods of time without injury.

Many Factors Affect Injury and Illness Rates

Injury and illness rates may be affected by many factors, such as employer
or employee emphasis on safety, the amount and quality of training,
employee turnover rates, and the speed of the production line.

Officials from a company, union, or trade association may take steps that
affect worker safety and health. For example, a company may form a plant
safety committee that reviews incidents of injury and illness to identify
safety issues and take steps to address weaknesses. In addition, company
officials may influence worker safety and health by showing their
commitment to safety through actions such as establishing medical safety
management programs at the plants, providing personal protective equipment
to workers, and disciplining workers who do not follow safety
procedures.32 Unions can also play a role in worker safety and health by
negotiating with company officials to take a more proactive approach to
addressing work conditions. Trade associations may offer training courses
and conferences on safety issues, guidance on meeting OSHA

30Hugh Conway and Jens Svenson, Occupational Injury and Illness Rates,
1992-96: Why They Fell, Monthly Labor Review, BLS, November 1998.

31During inspections, OSHA compliance officers ask plant officials if they
utilize incentive programs to reward their employees. An OSHA compliance
officer we spoke to told us that as part of her education and outreach
during an inspection, she suggests alternative ways of rewarding employees
that could minimize underreporting but still reward safe and healthy work
environments, such as providing rewards for consistently wearing personal
protective equipment or using safe work practices.

32A medical safety management program is one that addresses plant safety
and security, emergency management, fire prevention, and the proper
training of employees on the handling and safeguarding of hazardous
materials and medical equipment. It also includes guidelines for creating
and using an incident reporting system, as well as the steps necessary to
educate employees on issues like infection control, personal protective
equipment, ergonomics, and workplace violence.

requirements, and other assistance to companies in improving safety and
health.

Both OSHA and industry officials noted that training is a critical factor
in worker safety and health. Companies provide employee training in a
number of forms, including classroom instruction, on-the-job training, and
written and video training materials (generally in English and Spanish).
Meat and poultry plants typically offer several days of training at the
beginning of a worker's employment covering both job-specific and general
safety training. Plants periodically offer additional training classes or
updates-many of which are mandatory-such as annual refreshers on workplace
safety and health. Many plants also offer or require annual specialized
training on safety issues, such as knife sharpening, which can reduce
strain on line workers, and accident prevention such as "lockout/tag-out"
procedures that ensure that machinery is manually locked or disconnected
from a power source when not in use and tagged to note that it has been
locked or should not be used. In addition, one plant assigns mentors, or
buddies, to new workers to help them work more safely in an introductory
period.

Turnover rates can also affect the safety and health of workers. Turnover
tends to be high in the meat and poultry industry and, according to a
report by USDA's Economic Research Service, turnover rates of 100 percent
a year or more are not uncommon.33 High turnover can affect safety and
health at meat and poultry plants, according to one plant safety official,
because new employees are more likely to sustain an injury or illness than
more experienced workers. In the first few months of employment, an
employee may take shortcuts-because of the lack of familiarity with proper
procedures-that increase his or her vulnerability to injury or illness.
Plant officials often attribute high turnover to difficult working
conditions, extreme temperatures, and the fact that many of the industry's
jobs are physically demanding and stressful.

The speed at which production employees are expected to work, often
determined by the speed of the production line, or line speed, may also be
an important factor influencing their safety and health. The faster the
pace at which the production line moves, the less able workers may be to
perform tasks needed for safety. For example, according to industry
research, at certain line speeds workers may be unable to take the seconds

33Agricultural Economic Report No. 785, Washington, D.C., February 2000.

required to perform certain critical tasks, such as the frequent
sharpening of knives, to ensure that their jobs can be conducted safely.
Some respondents to our survey also noted that line speed is an important
factor affecting worker safety and health. While some trade association
representatives and plant officials told us that the risks associated with
line speed can be mitigated by adding more workers to the line or rotating
workers to other jobs, advocacy group and union representatives have
discounted that argument, stating that some plants may not have either the
additional employee resources to add to the line or the additional space
in the line configuration needed to add more workers.

Line speed is regulated by USDA to permit adequate inspection by food
safety inspectors. According to USDA, when the maximum speeds were
originally set and when they are adjusted by the agency, the safety and
health of plant production workers is not a consideration. OSHA has made
recommendations to companies to slow their line speed, as well as to make
other safety improvements when citing companies for repetitive motion
injury issues, according to an OSHA official. Research is lacking,
however, on the full effects of line speed on worker safety and health.
Industry and OSHA officials told us that the differences across slaughter
and cutting lines prevent systematic comparison, analysis, and regulation
of line speed. According to these officials, because machinery is arrayed
differently on each line, research that might isolate ergonomic limits and
improvements, or examine the incidence of other line-related injuries, is
difficult to accomplish. However, a memorandum issued in 2000 by
Nebraska's Lieutenant Governor recommended that OSHA "undertake a
legitimate study of the speed of the line in meatpacking plants" and that
"the industry should work cooperatively on that study."34 NIOSH officials
and nongovernmental ergonomic experts told us that line speed should be
further researched in order to understand its impact on worker safety and
health.

While USDA has established regulations on line speed, the purpose of the
agency's authority is not to protect workers, but to protect consumers.
USDA sets maximum line speeds based on how quickly its inspectors can
properly inspect the carcasses to ensure the safety of the meat. According
to trade association officials we interviewed, plants set their line
speeds at

34Memorandum from Nebraska's Lieutenant Governor Dave Maurstad to
Nebraska's Governor Mike Johanns entitled, "Review of Working Conditions
in Nebraska Meatpacking Plants," January 24, 2000.

While OSHA's Programs May Have Improved the Safety and Health of Meat and
Poultry Workers, Programmatic Weaknesses Make Determining Effectiveness
Difficult

a rate at or below the maximum while considering such factors as (1) the
speed at which employees can work and still produce a quality product and
(2) the number of animals that need to be processed. While a highranking
OSHA official we spoke to stated that he believed that the agency has the
regulatory authority to set its own line speed maximums, he also said that
it would be a difficult area to regulate.

OSHA has several efforts that target the meat and poultry industry, and
there is some evidence that these efforts have had a positive effect on
worker safety and health. However, the criteria OSHA uses to select
worksites for inspection may allow some plants with high injury and
illness rates to avoid inspection. OSHA's selection criteria do not
require the agency to examine trends in worksites' injury and illness
rates in order to select plants for inspection that have recently reported
significant changes in their rates. In addition, some of the data on which
OSHA bases its selection may be underreported and are incomplete.
Furthermore, OSHA's data collection efforts make determining program
results difficult.

Some of OSHA's Enforcement Efforts Target Compliance in the Meat and
Poultry Industry

Some of the inspections of employer compliance with federal safety and
health standards that OSHA conducts are of meat and poultry worksites. As
shown in table 2, OSHA conducted about 1,900 inspections of plants in the
meat and poultry industry from 1995 to September 15, 2004.35 These
inspections represented less than 1 percent of OSHA's total inspections.

35An additional 189 inspections were made by state occupational safety and
health agencies in state-plan states. In state-plan states, program safety
and health standards, and the enforcement of such standards, must be at
least as effective as federal OSHA programs. See 29 U.S.C. S: 667(c)(2).

Table 2: OSHA Inspections in the Meat and Poultry Industry, 1995 to 2004

                  Number of inspections Number of inspections
                    Year industry manufacturing inspections

in the meat and poultry

in all U.S. Total number of

                  As of Sept. 15,                              
                             2004      154          6,489              29,229 
                             2003      193          8,777              39,718 
                             2002      169          8,913              39,076 
                             2001      160          8,145              36,500 
                             2000      179          8,425              35,110 
                             1999      252          8,985              36,018 
                             1998      289          8,957              34,080 
                             1997      212          9,886              35,916 
                             1996      158          7,281              25,850 
                             1995      168          7,727              26,399 
                            Total     1,934         83,585            337,896 

Source: OSHA's inspections database.

OSHA also has efforts that focus on solutions to injuries prevalent in the
meat and poultry industry, such as repetitive stress disorders. For
example, OSHA's current ergonomics inspection plan uses its
worksitespecific injury and illness database to identify workplaces in
industries with higher than average injury rates. OSHA focuses its
ergonomics inspection resources on industries with relatively high rates
of injuries that appear to be related to ergonomic hazards. In addition,
OSHA's regional or area offices may implement local emphasis programs in
industries with high musculoskeletal disorder or repeated trauma rates and
known ergonomic hazards.36 The agency also responds to employee complaints
about ergonomic hazards.

36In 2003, OSHA implemented local emphasis programs in meatpacking and
three other industries: hospitals, warehousing, and automotive parts
manufacturing.

OSHA Has Involved the Meat and Poultry Industry in Its Cooperative
Programs

In 2002, OSHA formed an alliance with the American Meat Institute to
promote safe and healthful working conditions for meat industry workers.37
The alliance is meant to help reduce ergonomic hazards in the workplace.
It sets specific goals and priorities; key among them is for both OSHA and
the institute to develop and disseminate information and guidance,
particularly through their Web sites. The goal is to provide the
institute's members and others in the meat industry with information to
help protect workers' health and safety, with a focus on reducing and
preventing exposure to ergonomic hazards. The alliance also calls for both
organizations to provide training on ergonomics techniques, program
structure, and applications in the meat industry. Another goal is for OSHA
and the American Meat Institute to promote and encourage the institute's
members to participate in OSHA's cooperative programs such as the
Voluntary Protection Programs and mentor other members in helping them
qualify for participation.

The American Meat Institute also assists OSHA in maintaining and updating
information on safety and health in the meat industry on its Web site. The
institute, along with other stakeholders, provided information to OSHA for
the safety and health topics page on the agency's Web site entitled "OSHA
Assistance for the Meat Packing Industry," and the agency's Web-based
training tool ("eTools") for ammonia refrigeration.38 OSHA provides
information on eTools on many topics pertinent to the meat and poultry
industry, including ammonia refrigeration, machine guarding,
lock-out/tag-out procedures, poultry processing, confined space, and
ergonomic hazards.

Through its Strategic Partnerships Program, OSHA has established national
and regional partnerships within the meat and poultry industry. OSHA has
partnered with companies such as

o  	Tyson Foods. Initiated in 2001, this partnership covers two poultry
processing facilities. The 5-year agreement has a goal of improving and

37The American Meat Institute represents the interests of packers and
processors of beef, pork, lamb, veal, and turkey products and their
suppliers throughout North America. Together, its members produce 95
percent of the beef, pork, lamb, and veal products and 70 percent of the
turkey products in the United States. Headquartered in Washington, D.C.,
the institute provides legislative, regulatory, public relations,
technical, scientific, and educational services to the industry.

38Through its Web site, OSHA offers eTools on several subjects that
provide stand-alone, interactive, training tools on occupational safety
and health topics.

strengthening the company's safety and health management systems, reducing
injuries and illnesses, and serving as a model for improved worker
protection throughout the company.

o  	ConAgra Refrigerated Foods. This multiregional partnership, which
ended in January 2002, was meant to improve safety and health programs and
improve the relationship among OSHA, ConAgra, and the United Food and
Commercial Workers union and to prepare plants working toward
participation in OSHA's Voluntary Protection Programs.

o  	Odom's Tennessee Pride Sausage Inc. Through its regional partnership
with OSHA's region 6, Odom's has committed to reducing its illness and
injury rates and working toward participation in OSHA's Voluntary
Protection Programs.

OSHA's Omaha area office has implemented the following two partnerships:

o  	Nebraska's meat processing industry. In February 2000, members of
Nebraska's meat processing industry and OSHA initiated a voluntary
partnership program intended to address the high fatality, injury, and
illness rates that have plagued the industry. The group meets bimonthly to
learn about current safety and health practices, share safety-related best
practices that have proven successful in their facilities, and discuss
safety issues of concern to participants. Company representatives provide
injury and illness data to OSHA for tracking purposes on a semiannual
basis.

o  	Nebraska cleaning and sanitation companies. Citing the hazardous
working conditions encountered by employees of companies that provide
contract cleaning and sanitation services to meat and poultry plants, in
2003, OSHA's Omaha Area Office decided to establish a partnership with
these companies in order to help reduce injuries and illnesses.
Representatives of five companies have committed to a regional partnership
with OSHA's region 7 in an effort to work cooperatively and
collaboratively to reduce workplace fatalities, injuries, and illnesses
common to cleaning contractors such as strains, lacerations, contusions,
burns, fractures, amputations, dermatitis, and crushing injuries. The
goals of the partnership are to reduce days away from work by 4 percent
and to improve existing safety and health management programs.

OSHA has not, however, implemented programs similar to the Omaha Area
Office's partnerships in other areas of the country with large
concentrations of meatpacking plants or extended this type of program to
poultry plants. A high-ranking OSHA official told us that each area office

develops its own initiatives, which may be directed at other industries or
hazards than those in the meat and poultry industry. In addition,
according to the official, there were individuals in the Omaha office who
had a keen interest in the partnering approach used in the meatpacking
industry and had the entrepreneurial spirit to start these programs. The
office has presented its approach to at least one other office in an
effort to share its experience.

Several of OSHA's Special Efforts Target the Meat and Poultry Industry

Ergonomic Guidelines for the Meat and Poultry Industry

Memorandum of Understanding with USDA

OSHA has other special compliance efforts that target the meat and poultry
industry. For example, several pages of OSHA's Web site are dedicated to
the meat and poultry industry; they list the standards the agency uses to
combat hazards prevalent in this industry.39 OSHA also has several
directives specific to the industry, such as guidance on the acceptable
methods for guarding meat-cutting saws. In addition, OSHA issues
interpretations and compliance letters on issues specific to the meat and
poultry industry.

OSHA has also produced the following two sets of ergonomic guidelines for
the meat and poultry industry:

o  	Ergonomic Program Management Guidelines for Meat Packing Plants. This
document, jointly developed by OSHA and the American Meat Institute, was
developed in 1990 and contains advisory information on management
commitment and employee involvement, including preventive program elements
and detailed guidance.

o  	Guidelines for Poultry Processing. This document, published by OSHA in
September 2004, offers practical recommendations for employers to reduce
the number and severity of musculoskeletal disorders throughout the
industry. In developing the guidelines, OSHA reviewed existing ergonomics
practices and programs, state OSHA programs, as well as available
scientific information. OSHA also consulted with stakeholders, such as the
National Turkey Federation, to gather information on the ergonomic
problems present in the poultry-processing environment and the practices
that have been used successfully in the industry.

Because USDA inspectors are a constant federal presence in plants, OSHA
has established agreements with USDA, the latest of which is meant to

39Some of these standards include process safety management of highly
hazardous chemicals, general requirements for all machines, and guarding
of portable power tools.

improve compliance in meat and poultry plants. In 1994, USDA and OSHA
jointly revised an existing memorandum of understanding between the
agencies that established a process and framework for (1) training USDA
meat and poultry inspection personnel to improve their ability to
recognize serious workplace hazards within the meat and poultry industry,
(2) reinforcing procedures for meat and poultry inspection personnel to
report unsafe and unhealthy working conditions to which they are exposed
to the appropriate authorities, (3) instituting new procedures for USDA's
meat and poultry inspection personnel to refer serious workplace hazards
affecting plant employees to OSHA, and (4) coordinating possible
inconsistencies between OSHA's job safety and health standards and USDA's
sanitation and health standards.

According to the agreement, OSHA's training of USDA inspectors would not
be expected to supplant OSHA expertise in identifying serious workplace
hazards. In addition, USDA inspectors would not be trained to recognize
and refer serious workplace hazards affecting plant employees that tend to
arise only after protracted, cumulative exposure, such as those related to
repetitive motion and noise.

According to USDA officials, the memorandum was revised after a
devastating poultry plant fire in 1991 that killed 25 workers. According
to reports about the fire, the plant's fire doors had been padlocked from
the outside by the factory owner, who had locked the doors as a "loss
control technique" to prevent workers from stealing product. A USDA
poultry inspector was often present at the plant and testified at a
congressional hearing on the fire that he knew the doors were regularly
locked in violation of safety codes and had reported this to plant
officials.40 He did not, however, contact OSHA.

Although the purpose of the revised memorandum of understanding was to
educate USDA inspectors on recognizing and referring workplace hazards,
the agencies' efforts to implement the agreement, such as providing
training to USDA inspectors and evaluating the effectiveness and impact of
the training, have lapsed. According to OSHA officials, although the
agency put together training materials for USDA inspectors, only one
training session was held, and only a small number of individuals

40Review of U.S. Department of Agriculture's Food Safety and Inspection
Service Workplace Safety Regulations, Hearing Before the Subcommittee on
Department Operations, Research, and Foreign Agriculture of the Committee
on Agriculture, House of Representatives, 102nd Congress, Washington,
D.C., November 12, 1991.

were trained. USDA officials we spoke to confirmed this, and stated that
not a lot of effort was made to train inspectors.41 USDA officials told us
that OSHA had proposed a week's worth of training and that it would be
very difficult to pull inspectors from their line duties and send them to
training for that length of time. In their comments on a draft of this
report, USDA officials noted that in-plant inspectors routinely receive
training on topics such as wellness, awareness of infectious diseases, and
worker health and safety.

Evidence Suggests OSHA's Programs Have a Positive Impact on Worker Safety
and Health

Some positive outcomes have resulted from OSHA's efforts directed at the
meat and poultry industry. For example, in 2003, OSHA inspected 193 meat
and poultry plants to determine their compliance with federal safety and
health standards. These inspections produced safety improvements in
several of the plants inspected, according to OSHA, trade association
officials, and some plant officials we interviewed. In addition, according
to OSHA and trade association officials, the widespread use of the
agency's ergonomic guidelines for meatpacking plants has contributed to a
decline in worker illness and injury rates in the last decade.

OSHA's partnerships have also, according to the agency, had positive
outcomes. For example, according to information on its Web site, as a
result of OSHA's partnership with ConAgra Refrigerated Foods, many of the
company's facilities have formed new safety and ergonomics committees with
both management and union participation. According to OSHA, five of the
company's nine participating facilities experienced significant decreases
in workers' compensation costs ranging from 42 percent to 93 percent (with
an average reduction of 62 percent), suggesting a reduction in the injury
and illness rates for these five participating facilities.

Another of OSHA's partnerships, with meatpacking plants in Nebraska, has
shown some positive outcomes. According to officials from OSHA's Omaha
Area Office and some plant officials who participate in the partnership,
the group has made progress toward the goal of making the industry safer.
In addition, according to the participants, the relationship between OSHA
and Nebraska's meatpackers, which had been strained, if not antagonistic,
has improved significantly, and the group has made tremendous progress in
building cooperative, trusting relationships. These

41USDA trains its employees on safety and health issues that affect them
personally.

relationships have developed not only between OSHA and the plants, but
also among the plant officials themselves. According to officials at
OSHA's Omaha Area Office, over the 4-year existence of the partnership,
the members have realized a 23 percent reduction in injuries and illnesses
resulting in days away from work or restricted work activities. There has
also been a 39 percent reduction in total recordable injury and illness
cases, a total that includes cases resulting in days away from work, cases
resulting in restricted work, and cases requiring medical treatment. OSHA
officials told us that they consider these to be noteworthy improvements
over a relatively short period of time in an extremely hazardous industry.
Furthermore, the partnership has allowed OSHA to reach out directly to
meatpacking plants too small to meet the agency's criteria for inspection
(those with fewer than 40 employees) and provide them with information
about improving safety and health at their plants.

OSHA's memorandum of understanding with USDA has also resulted in some
positive outcomes. According to a high-ranking OSHA official, since the
revised memorandum was signed in 1994, USDA inspectors have made 31
referrals to OSHA, 26 of which resulted in an OSHA inspection. USDA does
not track this information and could not verify the number of referrals
made by its inspectors to OSHA. However, we were told by USDA officials
that the department's inspectors rarely make referrals because workplace
hazards are not the focus of their inspections. In addition, we were told
that USDA inspectors are more likely to discuss observed hazards with
plant management before referring them to OSHA, since they have
established relationships with the plants and because the hazards could
affect them as well as plant employees. Finally, OSHA officials said that
because a referral may cause OSHA to inspect the plant, some USDA
inspectors may be reluctant to make such referrals because it could mean
that OSHA would include them in the inspection and cite them for
violations, such as not wearing their personal protective equipment.

Several meat and poultry plants have taken advantage of OSHA's various
cooperative programs. Since 1996, 391 meat and poultry worksites have
received consultation services through OSHA's On-Site Consultation
Program.42 In addition, OSHA has also recognized some meat and poultry
plants as having exemplary safety and health management systems, although
the relatively low numbers of participants from this industry

42The number of visits is actually higher because some worksites have
received multiple visits or an employer can make one request that requires
services at several worksites.

indicates the difficulty in meeting program requirements. As of September
30, 2004, only 8 of the 1,180 Voluntary Protection Programs worksites were
in the meat and poultry industry. Similarly, as of September 1, 2004, only
8 of the 844 worksites participating in OSHA's Safety and Health
Achievement Recognition Program were in the meat industry, and no poultry
plants participated in the program.

Selection Criteria May Not Trigger Inspection of At-Risk Plants, and Data
Collection Makes Assessing Effectiveness Difficult

Selection Criteria

While the criteria OSHA uses to select worksites for inspection focuses
its limited resources mainly on plants with relatively high injury and
illness rates, the agency does not consider trends in worksites' injury
and illness rates over time. As a result, OSHA may not detect dramatic
decreases in these rates that could raise questions as to the accuracy of
the figures. This is of particular concern given the allegations of
underreporting in the industry and weaknesses in the data used to select
plants for inspection. OSHA does, however, select some worksites with low
injury and illness rates in an attempt to ascertain whether worksites with
low rates are underreporting injuries and illnesses. It also randomly
selects some worksites from high-hazard industries for record-keeping
audits designed to verify the injury and illness rates reported to OSHA.
For both of these efforts, however, OSHA selects few meat and poultry
plants. Furthermore, the data it collects on specific worksites-kept in
multiple databases-are not easily tracked, because OSHA does not assign a
unique identifier to each worksite. Therefore, it is difficult to assess
the effectiveness of OSHA's efforts to improve safety and health.

The criteria OSHA uses to select meat and poultry plants for inspection
target worksites that report high injury and illness rates. However,
OSHA's selection criteria do not allow it to detect anomalies in
worksites' reported injury and illness rates, because the agency does not
analyze data on plants' injury and illness rates over time. Although OSHA
surveys meat and poultry plants annually to obtain worksite-specific data
on their injury and illness rates and uses these data to select plants for
inspection, it does not review the data collected from previous years in
order to examine changes in their injury and illness rates. In addition,
these data are incomplete, because OSHA's survey sample varies from year
to year, and because OSHA only asks employers for 1 year of injury and
illness data. In 2002, we reported the problem with OSHA collecting only 1
year's worth of data, concluding that this limited the agency's ability to
effectively identify

hazardous worksites for inspection.43 Area office officials we interviewed
for that report said that, in some cases, the 1-year rate was an outlier
that did not reflect general worksite operations.

The data on which OSHA bases it selections are also incomplete because,
when it surveys worksites in the meat and poultry industry, OSHA does not
ask employers to report injuries and illnesses incurred by contract
cleaning and sanitation workers who work at the plant. Because these
workers are not employees of the plant, their injuries and illnesses are
recorded by the companies for whom they work rather than on the plants'
injury and illness logs.44 As a result, OSHA does not consider all
injuries and illnesses in selecting meat and poultry plants for
inspection. This is a significant oversight because, according to OSHA
officials, experts, and researchers, these workers incur high rates of
injury and illness and often sustain more serious injuries than production
workers. According to information in OSHA's inspections database, between
1998 and 2003, at least 34 contract cleaning and sanitation workers
employed in meat and poultry plants sustained serious injuries or were
killed. However, because these injuries were recorded as occurring in
another industry, none of the injuries were reflected in the meat and
poultry industry's injury and illness rates.45

A large number of workers perform this work under contract for meat and
poultry plants; we interviewed three cleaning and sanitation companies
that employ more than 5,000 workers at 140 plants across the country.46
One contract cleaning company representative reported that the biggest
risk factor affecting the safety of these workers was workers' decisions
to take shortcuts, such as not properly performing lock-out/tag-out
procedures for machinery before cleaning it. Another representative said

43GAO, Workplace Safety and Health: OSHA Can Strengthen Enforcement
through Improved Program Management, GAO-03-45, (Washington, D.C.: Nov.
22, 2002).

44Injuries and illnesses sustained by cleaning and sanitation workers who
are not employees of the plant are recorded in the general industry
category of "Services," which includes maids, janitors, and other workers
employed in cleaning services. Because this industry is not considered
high hazard, OSHA does not collect data from worksites in the industry to
use in selecting worksite for inspection.

45These injuries included fractures, severe chemical exposure, fatal
falls, incidents of crushed or severed limbs or heads, and injuries
necessitating amputation.

46See appendix II for more information on our interviews with the contract
cleaning and sanitation companies.

he felt that the biggest risk factor was the difficulty in communicating
how hazardous the complex and intricate machinery is because of language
or cultural differences. OSHA inspects the cleaning and sanitation shift
during its inspections of meat and poultry plants, whether the workers are
employees of a contract company or the plant. However, plants whose
contract workers have high injury and illness rates may not be selected
for inspection because these injuries are not included in the data OSHA
uses to select meat and poultry plants for inspection.

Because there are allegations of underreporting in the meat and poultry
industry, OSHA attempts to counter such incidences by verifying the injury
and illness rates of worksites it inspects as part of its SST program. In
addition to reviewing employers' logs during SST inspections, OSHA also
randomly selects for inspection 200 worksites each year that report low
injury and illness rates in high-hazard industries to ensure that these
worksites are not underreporting injuries and illnesses.47 In 2003, 5 of
the 200 worksites selected were meatpacking plants; in 2004, 10 were
meatpacking plants. The sausage and other prepared meats industry and the
poultry industry did not have injury and illness rates that met the
criteria for this effort. Therefore, OSHA did not select any worksites in
these industries for inspection that reported low rates.

In a separate effort designed in part to combat underreporting of injuries
and illnesses, OSHA annually conducts a number of comprehensive
record-keeping audits intended to verify the accuracy of the data on
injuries and illnesses that employers submit to OSHA.48 However, the
selection criteria it uses allow the agency to audit the records of only a
few meat and poultry plants. While OSHA's limited resources allow it to
select few worksites in any industry for a record-keeping audit, OSHA is
not doing enough to verify the accuracy of the data that meat and poultry

47OSHA uses BLS's aggregate industry data to determine which industries
are high-hazard. The worksites OSHA selects report a days away from work,
restricted, or transferred rate between 0.0 and 4.0 and a days away from
work injury and illness rate between 0.0 and 2.0, and are selected from
industries that have a days away from work, restricted, or transferred
rate of 8.0 or greater or a days away from work injury and illness rate of
4.0 or greater. OSHA began this effort to inspect 200 low-rate reporting
worksites from high-rate industries in 2002.

48The major difference between the records audits conducted as part of
this program and records reviews performed during other inspections is the
attainment of a medical access order by the OSHA compliance officer prior
to the audit. A medical access order allows OSHA to obtain documents such
as, medical records, state workers' compensation forms, insurer's accident
reports, company safety incident reports, and first aid logs.

                                Data Collection

plants report, considering the dramatic decreases in this industry's
reported injury and illness rates. Of the 250 worksites OSHA selected for
the audits of 2001 and 2002 data, only 3 each year were in the meat and
poultry industry.

While the criteria it uses to select worksites for inspection are rarely
altered, according to OSHA officials, in 2004, they adjusted the criteria
used to select the 200 worksites with low injury and illness rates for
inspection in order to focus on worksites with a large number of
employees. Previously, OSHA selected worksites in high-hazard industries
with a minimum of 40 employees for these inspections; currently, it
selects worksites with a minimum of 200 employees. This change will likely
have the effect of OSHA selecting even fewer meat and poultry plants that
report low injury and illness rates for inspection, because the majority
of plants have fewer than 200 employees. In addition, we were told by a
highranking OSHA official that the agency is considering adjusting the
criteria further to double the number of worksites with low illness and
injury rates for inspection. However, according to OSHA officials,
adjusting the criteria further-for example, to enhance the agency's focus
on a particular industry such as meat or poultry-would require additional
resources and a consideration of the effect on other industries.

OSHA's data do not allow the agency to determine the impact of its
enforcement and cooperative programs on the meat and poultry industry. To
determine the impact of its efforts, OSHA could match the injury and
illness data it collects from employers to data on inspections and
employer participation in its cooperative programs. However, such
matching, which would allow the agency to better relate reductions in
injury and illness rates to its interventions, cannot be easily performed.
The data it collects on specific worksites-recorded in multiple
databases-are not easily tracked because OSHA does not assign a unique
identifier to each worksite. Without such an identifier that can be
tracked across databases, the agency is unable to easily consolidate all
the information associated with each worksite. Without the ability to
compare this information across databases, it is difficult to assess the
success of its efforts.

We attempted to assess the impact of OSHA's programs on the meat and
poultry industry by comparing worksite-specific data across its various
databases to determine an association between changes in a plant's injury
and illness rates and the agency's efforts. However, we encountered
problems because of the lack of a unique identifier for each worksite. To
match data on specific worksites without such an identifier, we relied on
other identifiers-such as the name of the company, address, or zip

code-to find the data associated with a company over time. However,
because of differences in how these identifiers appeared in each of OSHA's
databases, we could not reliably track data for specific companies. For
example, from one year to the next, a company's name may appear
differently in the various databases, or a match on address may not
produce a match for the company name. OSHA officials we spoke to
acknowledged the difficulties involved in this type of effort. In fact,
they recently encountered similar problems in their attempts to evaluate
the impact of the SST program in 2004 (see app. III for more
information).49 OSHA's inability to assess the effectiveness of its
efforts has been a recurring finding.

In November 2002, we recommended that OSHA take steps to assess the impact
of its SST program on workplace injuries and illnesses. Similarly, in
March 2004, we reported that OSHA's lack of comprehensive data on its
cooperative programs-such as their relative impact on worksites' safety
and health-makes it difficult to fully assess the effectiveness of these
programs.50 OSHA agreed with our recommendation but pointed out that the
agency's variety of strategies reach out to different types of industries,
employers, and workers, making it difficult and costly to compare their
relative effects.

The dangerous and repetitive nature of the work in the meat and poultry
industry results in a variety of injuries and illnesses to workers.
Although the efforts by government, employers, and advocacy groups have
helped improve worker safety and health in this industry, and according to
BLS the number and rate of injuries and illnesses have decreased
substantially over the last decade, additional improvements could be made.
For example, the criteria that OSHA uses to select plants for inspection,
while reasonable, do not incorporate consideration of dramatic or sudden
decreases in injury or illness rates in selecting plants for inspection.
In addition, because OSHA lacks complete data on the injuries and
illnesses of meat and poultry workers, particularly those employed by
cleaning and

49Evaluation of OSHA's Impact on Workplace Injuries and Illnesses in
Manufacturing Using Establishment-Specific Targeting of Interventions,
Prepared for OSHA by ERG, Lexington, Mass.: July 23, 2004.

50GAO, Workplace Safety and Health: OSHA's Voluntary Compliance Strategies
Show Promising Results, but Should Be Fully Evaluated before They Are
Expanded, GAO-04-378, (Washington, D.C.: Mar. 19, 2004).

Conclusions

sanitation companies, plants in need of inspection may not be identified
and selected. Further, because OSHA does not track changes in individual
plants' injury and illness rates, or have a method for comparing these
rates with data collected on inspections or plants' participation in its
cooperative programs, it lacks a means of understanding the impact its
programs have on workers in this industry.

OSHA also lacks some of the information needed to participate more fully
in improving worker safety and health. For example, until the effects of
line speed are studied from a worker safety and health perspective to
better understand its effect on injury and illness rates, it will be
difficult for OSHA to provide meaningful input with regard to the process
of regulating the speed of the production line. In addition, OSHA has been
slow in expanding its successful efforts. Because its most successful
program aimed at improving safety and health has not been replicated in
other areas of the country, OSHA is not allowing workers in its other
jurisdictions to realize the benefits, such as the potential for a reduced
number of injuries and illnesses, of this program. Finally, the memorandum
of understanding between USDA and OSHA is not being utilized to the full
extent possible. The efforts called for by the memorandum of understanding
to reinforce and supplement the training of USDA inspectors so they are
able to recognize and refer serious workplace hazards in meat and poultry
plants have lapsed.

Recommendations for In order to strengthen the agency's efforts to improve
safety and health of workers at meat and poultry plants, the Secretary of
Labor should direct Executive Action the Assistant Secretary for
Occupational Safety and Health to consider

o  	adjusting OSHA's criteria for selecting worksites for SST inspections
and for record-keeping audits to consider worksites that have had large
reductions in their injury and illness rates over time;

o  	requiring worksites that are surveyed by OSHA to obtain
worksite-specific data on injuries and illnesses to include (1) multiple
years of data, so that trends in their rates may be analyzed and (2) data
on injuries and illnesses to workers employed by cleaning and sanitation
companies that provide workers to the plant under contract so that these
data can be included in the rates OSHA uses to select plants for
inspection;

o  	requiring that a common identifier for each plant be used in all of
its enforcement and cooperative program databases so that these different

data sets can be more easily compared in an effort to measure the agency's
impact on worker safety and health; and

o  	expanding successful partnerships, such as the Omaha Area Office's
partnership with meatpacking plants in Nebraska to other area offices with
high concentrations of meat and poultry plants.

The Secretary of Labor should direct the Assistant Secretary for
Occupational Safety and Health and the Secretary of Agriculture should
direct the Acting Administrator of the Food Safety and Inspection Service
to

o  	revisit and update their memorandum of understanding to ensure that
USDA inspectors receive training in recognizing and referring workplace
hazards and that the agreement remains current.

In addition, the Secretary of Health and Human Services should

o  direct the Director of the Centers for Disease Control and Prevention
to

Agency Comments

have NIOSH conduct a study of the effect of the speed of the production
line on workers in the meat and poultry industry, a study that would also
include other job-specific features that interact with line speed to
increase the risk of injuries and illnesses to these workers.

OSHA, USDA, HHS, and BLS provided us with written comments on a draft of
this report, which are reproduced in appendixes IV, V, VI, and VII,
respectively. The agencies generally agreed with all of the
recommendations that applied to them.

OSHA noted that it has solicited public comments on its SST program and
will consider our suggestion to expand its selection criteria for SST
inspections. The agency also commented that it will cooperate with USDA to
encourage the revitalization of USDA inspector training and will work with
NIOSH and others to investigate the relationship between line speed and
the risk of injury. Finally, OSHA pointed out that some of the remaining
recommendations, such as expanding its data collection efforts, could have
significant impact on the agency's resources and that it would consider
these recommendations in conjunction with decisions on how best to
allocate the resources it has available.

USDA noted that, because its in-plant employees are a federal presence in
meat and poultry plants, they can help detect and report serious

workplace hazards to OSHA. The agency also noted its responsibility to
enforce the Humane Methods of Slaughter Act, emphasizing that, if inplant
inspectors witness egregious violations of the act-such as hoisted animals
struggling or thrashing-they have the authority to take immediate
enforcement action against such firms including stopping the production
line. USDA commented that compliance with the act indirectly improves
workplace safety.

HHS agreed that there is a need to study the relationship between line
speed and musculoskeletal disorders and other injuries in the meat
industry, and stated that it would direct NIOSH to conduct such a study.
The agency noted, however, the difficulty its staff have had in the past
in gaining access to meatpacking plants to conduct research. HHS also
commented on the resource commitment that would likely be involved for
such a large and detailed, but necessary, study.

BLS noted that, although it conducted a major redesign of its annual
survey of occupational injuries and illnesses in 1992, (in part because of
concerns about the completeness of employer reporting,) there is still
some concern about underreporting of injuries and illnesses among users of
the data. The agency also noted several technical corrections to the
report, as did OSHA, USDA, and HHS, which we incorporated as appropriate.

As arranged with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days
after the date of this report. At that time, we will send copies of this
report to the Secretary of Labor, the Assistant Secretary of Labor for
Occupational Safety and Health, the Secretary of Agriculture, the Acting
Administrator of the Food Safety and Inspection Service, the Secretary of
Health and Human Services, and the Director of the Centers for Disease
Control and Prevention. We will also make copies available to others upon
request. In addition, the report will be available at no charge on GAO's
Web site at http://www.gao.gov.

Please contact me or Revae Moran on (202) 512-7215 if you or your staff
have any questions about this report. Other contacts and staff
acknowledgments are listed in appendix VIII.

Robert E. Robertson Director, Education, Workforce, and Income Security
Issues

                       Appendix I: Scope and Methodology

For this report, we attempted to (1) describe the characteristics of
workers in meat and poultry slaughter and processing plants and the
conditions in which they work; (2) identify the types of injuries and
illnesses workers in meat and poultry slaughter and processing plants
incur, how the injury and illness rates have changed over the past decade,
and the factors that may have affected these rates; and (3) determine what
is known about the effectiveness of the Occupational Safety and Health
Administration's (OSHA) efforts to improve safety and health at meat and
poultry slaughter and processing plants. To address these objectives we

o  	obtained and analyzed relevant data from the Bureau of Labor
Statistics (BLS) on worker demographics and workplace injuries, illnesses,
and fatalities; OSHA's inspection database; the data OSHA uses to target
specific worksites for inspection; and information on plants that
participate in OSHA's cooperative programs;

o  	conducted a survey of meat and poultry plants to obtain data on their
workers, factors that affect their injury and illness rates, plants'
efforts to improve safety and health, and plants' interactions with OSHA;

o  	interviewed officials from OSHA and other federal agencies, such as
the U.S. Department of Agriculture (USDA) and the Centers for Disease
Control and Prevention's National Institute for Occupational Safety and
Health (NIOSH) within the Department of Health and Human Services (HHS),
as well as individuals from contract cleaning and sanitation companies,
unions, advocacy groups and key trade associations; and

o  visited six meat and poultry plants.

Reliability of USDA's, BLS's, and OSHA's Data

To determine the number, location, and regional distribution of plants, we
examined USDA's Food Safety and Inspection Service's Performance Based
Inspection System database. This management system integrates weekly
schedules of inspection tasks; documents inspection findings including
deviations, deficiencies, and corrective actions; and provides a database
for the automatic data-processing support system.

To describe the characteristics of workers employed in the meat and
poultry slaughter and processing industries, we obtained demographic
estimates for meat and poultry workers and for the manufacturing industry
as a whole from BLS's Current Population Survey (CPS) March supplement for
2004 and 1995. The CPS is a monthly survey of households conducted by the
U.S. Census Bureau for BLS. The CPS, a sample of

Appendix I: Scope and Methodology

60,000 households, provides a comprehensive body of information on the
employment and unemployment experience of the nation's population,
classified by age, sex, race, and a variety of other characteristics.

Because the CPS estimates are based on probability samples, they are
subject to sampling error. Slightly different estimates could result from
different samples. We express our confidence in the precision of our
particular sample's results as a 95 percent confidence interval. This is
the interval that would contain the actual population value for 95 percent
of the samples we could have drawn. As a result, we are 95 percent
confident that each of the confidence intervals in this report will
include the true values in the study population.

For the CPS estimates in this report, we estimated sampling error and
produced confidence intervals using the methods provided in the technical
documentation for the 2004 and 1995 March supplements. All CPS percentage
estimates contained in this report have 95 percent confidence intervals
within plus or minus 8 percentage points of the estimate itself. All other
CPS estimates contained in this report have 95 percent confidence
intervals within plus or minus 14 percent of the estimate itself, unless
otherwise noted.

We also reviewed data on injuries and illnesses, collected and published
by BLS through its Survey of Occupational Injuries and Illnesses, for
calendar years 1992 to 2002, as they related to workers in the meat and
poultry industry. BLS's Survey of Occupational Injuries and Illnesses
provides estimates of the number and frequency (incidence rates) of
workplace injuries and illnesses based on logs kept by private industry
employers during the year. Survey estimates are based on a scientifically
selected sample of worksites, some of which represent only themselves but
most of which also represent other employers of like industry and
workforce size that were not chosen in a given survey year. Besides
providing injury and illness counts, survey respondents also are asked to
provide additional information for a subset of the most serious nonfatal
cases logged, namely, those that involved at least 1 day away from work,
beyond the day of injury or onset of illness. Employers answer several
questions about these cases, including the demographics of the worker
disabled, the nature of the disabling condition, and the event and source
producing that condition. BLS calculates relative standard errors for all
estimates it tabulates (see BLS's Web site for more information). These
relative standard errors were used to develop 95 percent confidence
intervals for each estimate. In this report, all estimates of incidence
rates have 95 percent confidence intervals of within plus or minus 14
percent of

Appendix I: Scope and Methodology

the estimated incidence rate. For example, the estimated rate for
tendonitis in 1992 was 23.6 cases per 10,000 full-time meat and poultry
workers. Since 14 percent of 23.6 is 3.3, the confidence interval for this
interval is within 20.3 to 26.9 cases per 10,000 full-time workers.1

We also reviewed data on fatalities, collected and published by BLS
through its Census of Fatal Occupational Injuries for calendar years 1992
to 2003 as they related to workers in the meat and poultry industry. BLS's
Census of Fatal Occupational Injuries is a federal-state cooperative
program that has been implemented in all 50 states and the District of
Columbia since 1992. To compile fatality counts that are as complete as
possible, the census uses multiple sources to identify, verify, and
profile fatal worker injuries. Information about each workplace fatality-
occupation and other worker characteristics, equipment involved, and
circumstances of the event-is obtained by cross-referencing the source
records, such as death certificates, workers' compensation reports, and
federal and state agency administrative reports. To ensure that fatalities
are work-related, cases are substantiated with two or more independent
source documents or a source document and a follow-up questionnaire. Data
compiled by the program are issued annually for the preceding calendar
year. We report the 2003 data in a footnote because the data are not
comparable with data from previous years. According to BLS, the new
industry and occupational classifications the agency is required to use
may, in some instances, have different definitions than the classification
system used previously.

To analyze the extent to which OSHA interacts with meat and poultry plants
through its enforcement programs, we analyzed inspections data for fiscal
years 1996 to 2004 from OSHA's Integrated Management Information System
and worksite-specific injury and illness data collected by OSHA. We
assessed the completeness of these data by reviewing OSHA's documentation
on how the data were collected and performed electronic tests to look for
outliers, missing values, and duplicate records. On the basis of these
reviews and tests, we found the data sufficiently reliable for

1All the rates of occupational injury and illness in this report are based
on BLS data. BLS calculates a relative standard error for each estimate it
tabulates, and the 95 percent confidence intervals for the detailed rates
cited in this report are all within plus or minus 14 percent of the
estimated rates. Rather than report confidence intervals for every
incidence rate estimate in this report, a broad conservative confidence
interval is used to cover all BLS incidence rate estimates. For this
particular estimate, the confidence interval is plus or minus 2.5 percent
of the estimated tendonitis incidence rate estimate used in this report.

                       Appendix I: Scope and Methodology

our purposes. In addition, for OSHA's inspections data, we obtained and
reviewed documentation of internal controls.

We analyzed the data that OSHA uses to target specific worksites for
inspection through its SST program. These data are collected by OSHA
through its annual Data Initiative, which is a nationwide collection of
worksite-specific injury and illness data from approximately 80,000
worksites. OSHA collects data from worksites by using the OSHA
Work-Related Injury and Illness Data Collection Form.

To report on the extent that meat and poultry plants participate in OSHA's
various cooperative programs, we analyzed OSHA's consultation database,
its lists of Voluntary Protection Programs and Safety and Health
Achievement Recognition Program sites, and its lists of current alliances
and strategic partnerships.

We interviewed USDA, OSHA, and BLS officials to establish the reliability
of the data. We found the data to be sufficiently reliable for our
purposes.

Analysis of BLS's CPS and Injury and Illness Data

We explored, for this report, different ways in which the CPS and BLS's
injury and illness data could be used to track changes in injury and
illness rates for various groups of workers and discovered several
limitations. For example, we analyzed CPS data on worker demographics by
industry and data on injuries and illnesses sustained by workers. Using
the two data sets, we attempted to determine whether workers in certain
demographic groups-such as males and females, whites and minorities, and
younger and older employees-were sustaining more injuries or illnesses now
than would be expected, taking into account the number of individuals in
these demographic groups. We also attempted to estimate differences in
injury rates, or in the likelihood of being injured, between certain
worker demographics, such as gender, race, and age. However, a large
percentage of cases in the meat and poultry industry that were reported to
BLS- 24 percent in 2002-lacked data on the race of the injured worker
since race is not a required reporting item.2 Because of this lack of
data, it was not possible to determine whether workers of a certain race
were disproportionately injured.

2Because OSHA does not require employers to record race data on its case
reporting form, BLS cannot require employers to report it as part of the
detailed data it collects for these serious cases. Instead it is a
voluntary item and BLS does not receive the race data for roughly one in
four of the injured or ill workers.

                       Appendix I: Scope and Methodology

First, BLS's data on injuries and illnesses cannot be used by themselves
to estimate injury rates or the likelihood of being injured, since those
data include only information on workers who were injured, but not on
workers who were not injured. While BLS's injury and illness data could be
used to estimate the numbers of workers in the meat products industry at
risk of being injured, overall and in each of the subgroups of interest to
us, its injury and illness data do not provide demographic information
(e.g., data on race, sex, or age) on all workers who were injured, but
only those workers whose injuries were serious enough to have resulted in
the workers requiring time off from work. While we might have merged
information from BLS's injury and illness data and information from the
CPS in order to estimate rates of injuries requiring time off, and
differences in those rates across subgroups, that task was complicated by
the fact that there was considerable information missing on race, which
was one of the factors of greatest interest to us. Because BLS does not
require the various states and industries surveyed to disclose the race of
employees injured, some choose not to, and ultimately race is unknown for
roughly one in every four persons injured.

In addition, we could have estimated differences in the rates of injury
and illness requiring time off across sex and age categories. However, the
lack of detailed information in the CPS on the types of jobs held by
workers employed in the meat products industry would not have made it
possible for us to determine whether differences in injury and illness
rates across age and sex categories was a result of differences in these
demographic characteristics or the result of women and older employees
having different types of jobs than men and younger workers.

Survey of Meat and Poultry Plants

To obtain information about safety and health and the characteristics of
their workforce, we administered a survey to a sample of meat and poultry
plants. Our survey population consisted of plants represented in OSHA's
worksite-specific injury and illness database for years 1999 to 2002. This
database contains annual information on occupational injuries and
illnesses at the worksite (plant) level. The data on worksites, operating
in what are considered high-hazard industries, have been collected since
1995. Since the data for approximately one-third of all existing plants
above a certain size are updated in the database in any particular year,
we included in our sample, all plants included in the database during the
most recently available 4-year period (1999 to 2002). The specific
industries on which we focused were meat and poultry plants in Standard
Industrial Classification code 201-the meat products industry-including
those in meatpacking plants, code 2011; the sausages and other prepared
meat

                       Appendix I: Scope and Methodology

Sample Design

Survey Administration and Response Rates

products industry, code 2013; and the poultry slaughtering and processing
industry, code 2015.

Our survey sample included all plants from the database with more than
1,250 employees. The remaining plants were stratified by industry, using
the three Standard Industrial Classification codes for each of the three
industries that encompass the meat products industry. We drew a random
sample from each of these three industries.

From our total sample of 420 plants, 24 were eliminated for various
reasons, including the fact that the plant had gone out of business, the
plant was not a meat or poultry plant, or the plant was duplicated
elsewhere in our sample.

To develop our questionnaire, we consulted with officials at the American
Meat Institute and the United Food and Commercial Workers union, and
experts at GAO. We pretested a draft of the questionnaire with six
companies in the meat products industry. We mailed the questionnaire,
addressed to the plant safety director (or other appropriate management
personnel), requesting information on the demographic characteristics of
the plant's workforce, the working conditions of the plant, the safety
training and related efforts undertaken within the plant, and the plant's
interaction with (and respondent's opinions on) OSHA. The survey was
conducted between July 2004 and September 2004.

The overall response rate of 23 percent compromises our ability to
generalize the findings across the population of plants and to present
statistically valid results. While the sample was designed to draw
inferences from the study population, we did not produce estimates of the
population of meat producers based on our sample results. We arrived at
this decision both because the response rate was low and because it is
likely that certain key characteristics of respondents differ from those
of nonrespondents. For example, since two major companies refused to
participate, our responses did not include the responses of any plants
from these companies; the experiences of our respondents may differ from
those of plants from these companies. Table 3 summarizes the sample sizes
by industry, their disposition, and our response rates.

                       Appendix I: Scope and Methodology

         Table 3: Survey Sample Sizes, Disposition, and Response Rates

                                              Out of  Responses  Refused to   
      Industry sector     Populationa Sample scopes b received    participate 
                                                                Response rate 
     Meatpacking plants                                         
         with 1,250                                             
     employees or fewer           393  101          4        26        26 27% 
     Sausage and other                                          
       prepared meat                                            
    products plants with                                        
           1,250                                                
     employees or fewer           407  121          5        42        12 36% 
    Poultry slaughtering                                        
       and processing                                           
     plants with 1,250                                          
        employees or                                            
           fewer                  397  100         11        19        20 21% 
     All large meat and                                         
       poultry plants                                           
    with more than 1,250           98   98          4         8         25 9% 
         employees                                              
           Total                1,295  420         24        95        83 23% 

Source: GAO analysis.

aThe population values represent the number of plants in OSHA's
worksite-specific injury and illness database between 1999 and 2002.

b"Out of scopes" include plants that did not slaughter or process meat or
poultry or that were no longer in operation at the time of survey
administration.

Because of our low response rate, we did not use the data obtained from
the survey to draw conclusions about the meat and poultry industry.
Instead, we used the responses to illustrate some of the information
provided from other sources in our report, such as opinions about OSHA as
a factor in the safety and health of workers. We also used the data to
provide examples about the range of responses we found. For example, we
reported that one plant had an employee turnover rate that could reach 200
percent from data obtained from our survey.

Interviews with OSHA Area Offices

To describe the variety and extent of OSHA efforts within the meat and
poultry industry, we interviewed officials from four OSHA regional
offices. We selected these regional offices based on information from
OSHA's inspections database, which contains data on inspections and fines
levied by OSHA. We examined the inspections data to determine the regional
offices that had conducted the highest number of inspections from January
2003 to July 2004 for plants in Standard Industrial Classification codes
2011, 2013, and 2015. From these interviews, we obtained information about
their activities in the meat and poultry industry within their respective
regions, including any regional and local emphasis programs, their
perspectives on factors affecting the safety and health of workers in this
industry, and coordination efforts between their offices and USDA.

                       Appendix I: Scope and Methodology

Visits to Meat and Poultry Plants

During the course of this review, we visited six meat and poultry plants.
Of the six plants, we visited four (two beef, one pork, and one poultry
slaughter and processing plant) to obtain a better understanding of the
work performed by workers in meat and poultry plants and the conditions in
which they work. During these visits, we toured the plants and spoke to
plant officials about worker demographics, plant operations, injury and
illness history, and their experiences with and opinions of OSHA. Three of
these four plants were selected because they were located in an area of
the country where meat production is high; the other was close to our
headquarters office. We visited two additional plants for the purpose of
pretesting our survey instrument; we did not tour these two plants.

Appendix II: Interviews with Cleaning and Sanitation Companies

We interviewed three cleaning and sanitation companies that provided
contracted services to meat and poultry plants. In total, these three
companies employed over 5,000 non-union workers and operated in 140
different plants across the country. One company representative told us
the company had contracts to operate in plants owned by some of the
largest companies in the meat and poultry industry.

In general, all three cleaning and sanitation companies employed workers
who tended to be young and Hispanic. The companies supplied small plants
with crews as small as 1 or 2 employees and large plants with crews as
large as 150 employees. According to representatives from these companies,
cleanup shifts at plants can range from only 2.5 hours to 12 hours, but
typically shifts lasted less than 8 hours. While the majority of time is
spent cleaning the slaughter and process areas, at many plants the crews
are responsible for also cleaning bathrooms and office space. At the end
of the cleanup shift each day, the slaughter and process areas must pass a
USDA inspection in order for the plant to restart its operations. All
three companies' representatives reported that if the plant is not cleaned
within the time allotted for USDA inspection, they must pay some type of
monetary penalty to the plant.

The companies reported providing their workers with all safety and health
training. In addition, the companies employed safety auditors who travel
to various plants to examine safety issues. These examinations may include
interviewing contract workers to see if they are aware of certain safety
regulations or procedures, such as "lock-out/tag-out" procedures, the
issue cited as most important by the companies.

With regard to the incidence of repetitive motion injuries, one company
representative stated that these types of injuries among their workers are
limited because sanitation workers perform tasks that are different from
those performed by plant workers in that they are constantly moving around
and not performing repetitive tasks. He said, however, that conditions
such as working at night, sweating from the steam, freezing from the cold,
and being wet all the time contribute to the high turnover for his
employees.

Appendix III: OSHA's Study on Its Impact Using Establishment-Specific
Targeting of Interventions

In November 2002, we recommended that OSHA take steps to assess the impact
of its SST program on workplace injuries and illnesses.1 OSHA has since
conducted its first major evaluation of the effectiveness of the SST
program and, in 2004, issued a report detailing the results of its
evaluation.2 The report states that worksites experienced statistically
significant cumulative 3-year reductions in the number of injuries and
illnesses as a result of OSHA's efforts.3 However, while these results
indicate positive outcomes, the study did not attempt to isolate the
impact of OSHA's efforts from other factors-such as employers' own safety
programs-that may have as much or more of an influence on injuries and
illnesses. In fact, several employers who responded to our survey stated
that the reduction in their injury and illness rates could be attributed
to other factors that they claimed worked in addition to, or in place of,
OSHA's efforts, and that these other factors were as important in
achieving safety and health improvements.

In performing this evaluation of its SST program, OSHA encountered
difficulties in identifying worksites across databases and dealing with
the incompleteness of certain data fields.4 While not invalidating the
conclusions in its report, the difficulties OSHA encountered suggest that
the data collection for its worksite-specific injury and illness database
could be improved. In particular, OSHA did not consistently assign a

1See GAO-03-45.

2Evaluation of OSHA's Impact on Workplace Injuries and Illnesses in
Manufacturing Using Establishment-Specific Targeting of Interventions,
Prepared for OSHA by ERG, Lexington, Mass.: July 23, 2004.

3We did not evaluate documentation related to the development of the
models presented, so we did not determine whether the work was done
correctly. As described, however, we believe that OSHA's approach was
reasonable and in line with current methodological approaches.

4OSHA used complex data cleaning and matching algorithms to prepare and
combine information within its establishment-specific injury and illness
database and between it and other databases. The establishment-specific
injury and illness database, in particular, presented many challenges
since it is derived from annual surveys of business establishments. For
instance, some of the surveys received were rejected from the analysis
because of missing information, some were duplicate entries, and about
half were rejected because they could not be matched to an establishment
in the prior year. OSHA used similar matching procedures to combine the
injury and illness data with the intervention records housed within OSHA's
inspections database. We believe that OSHA's description of analysis
difficulties with enterprise-level data is fair and, as described, OSHA's
efforts to work through such difficulties seemed thorough. The analysis
file OSHA constructed, though, most likely contains some unknown measure
of mismatch.

Appendix III: OSHA's Study on Its Impact Using Establishment-Specific
Targeting of Interventions

unique identifier to each plant, which made it difficult to compare
information across databases. This issue, combined with a lack of
information in the 2004 report concerning how the model was developed and
tested, points to a need to use caution in interpreting the report's
results.

Appendix IV: Comments from the Occupational Safety and Health
Administration

Appendix IV: Comments from the Occupational Safety and Health
Administration

        Page 63 GAO-05-96 Safety and Health of Meat and Poultry Workers

        Page 64 GAO-05-96 Safety and Health of Meat and Poultry Workers

Appendix VI: Comments from the U.S. Department of Health and Human
Services

Appendix VI: Comments from the U.S. Department of Health and Human
Services

Appendix VII: Comments from the Bureau of Labor Statistics

Appendix VII: Comments from the Bureau of Labor Statistics

Appendix VIII: GAO Contacts and Staff Acknowledgments

GAO Contacts 	Revae E. Moran, (202) 512-3863 Monika R. Gomez, (202)
512-9062

Staff Acknowledgments

(130310)

David G. Ehrlich and Friendly M. Vang-Johnson made significant
contributions to this report throughout the review. In addition,
Luann M. Moy helped develop our data collection instrument and our
overall design and methodology; Margaret L. Armen and
Richard P. Burkard provided legal support; Avrum I. Ashery,
Jennifer R. Popovic, and Melba Edwards designed our graphics;
Paula J. Bonin, Mark F. Ramage, Douglas M. Sloane, and Beverly A. Ross
provided technical assistance; and Corinna A. Nicolaou assisted in report
and message development.

GAO's Mission

Obtaining Copies of GAO Reports and Testimony

The Government Accountability Office, the audit, evaluation and
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.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site (www.gao.gov). Each weekday, GAO posts newly
released reports, testimony, and correspondence on its Web site. To have
GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

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. Government Accountability 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, Contact:

Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

Gloria Jarmon, Managing Director, [email protected] (202)
512-4400Congressional U.S. Government Accountability Office, 441 G Street
NW, Room 7125 Relations Washington, D.C. 20548

Public Affairs 	Susan Becker, Acting Manager, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

                           PRINTED ON RECYCLED PAPER
*** End of document. ***