 |
|
| Progress Report
on Salmonella Testing of Raw Meat and Poultry Products, 1998–20081 |
Executive Summary
In calendar year 2008, FSIS analyzed 36,203 verification samples across eight meat
and poultry product classes with the following percent positive rate of Salmonella per
product class: broilers (7.3%), market hog (2.6%), cow/bull (0.5%), steer/heifer (0.2%), ground beef (2.4%),
ground chicken (25.5%), ground turkey (15.4%) and turkey (6.2%). The Agency believes that the higher the
percent positive rate, the greater the potential for the public to consume a product that can be
attributed to foodborne illness. FSIS schedules approximately 75 sample sets monthly using a risk-based
algorithm to target establishments demonstrating variable or poor process control (i.e, establishments
with high percent positive rates). Previous Salmonella Annual Reports can be found on the FSIS web site.
The Agency has determined that by 2010, 90% of establishments eligible
for Salmonella testing should be in Category 1, i.e. demonstrate
consistent process control. Since 2006, FSIS has focused sampling
on raw carcass product classes believing that improvement in process
control in these product classes will result in improvement in process
control in raw ground product classes. As of December 2008, 81% of
establishments are in Category 1. Based on eight product classes,
the following percentages of establishments are in Category 1: broilers
(82%), market hog (86%) cow/bull (81%), steer/heifer (86%), ground
beef (76%), ground chicken (50%), ground turkey (71%), and turkey
(97%).
Three new policies were introduced in CY 2008:
- A subcategory (2T) was created to represent establishments with a most recent set at < 50% of the performance standard or guidance and a prior set at >50 % of the performance standard or guidance. The Agency considers these establishments to be transitioning to Category 1.
- FSIS started rounding down the allowable positives for Category 1 for product classes having odd numbered performance standards or guidance (turkey, ground turkey, and ground beef). The Agency believes that this more stringent policy is necessary and an important incentive to improve performance in controlling Salmonella.
- To better allocate testing resources towards higher
risk establishments, beginning in the second quarter, FSIS began
sampling very low volume producing raw ground beef establishments
for Salmonella when they were scheduled for E. coli
O157:H7 testing (one sample, two tests). FSIS no longer schedules
these establishments under the traditional verification sampling
program.
2008 Annual Report
The Food Safety and Inspection Service (FSIS) uses data from its regulatory testing programs to monitor the
effectiveness of its Pathogen Reduction: Hazard Analysis and Critical Control Point (PR/HACCP) rule and
to assess process control in individual establishments. The U.S. Department of Health and Human Services
has incorporated the target of 6.8 cases of salmonellosis/100,000 persons into the Healthy People 2010
objectives, which FSIS recognizes as appropriate guidance for the Agency's strategic planning to strengthen
public health protection.
In 1996, FSIS established PR/HACCP to verify that establishments demonstrate consistent process control for
preventing, eliminating, or reducing the contamination of raw meat and poultry products with disease-causing bacteria,
by setting Salmonella performance standards that slaughter establishments and establishments that
produce raw ground products should meet. Raw products with established performance standards or guidance include:
carcasses of cows/bulls, steers/heifers, market hogs, broilers, and turkeys. Processed products measured
by performance standards include: ground beef, ground chicken, and ground turkey. The performance standards
are based on the prevalence of Salmonella as determined from the Agency's nationwide microbiological
baseline studies conducted prior to PR/HACCP implementation. The performance standards and guidance are
expressed in terms of the maximum number of Salmonella-positive samples acceptable per sample set.
The number of samples in a sample set varies by product, and the maximum number of positive samples acceptable
in a set provides an 80% probability of an establishment passing when it is operating at the standard.
Prior to 2006, there were two phases of the FSIS regulatory program for Salmonella in raw products:
non-targeted and targeted testing. Non-targeted or "A" set tests were collected at establishments randomly
selected from the population of eligible establishments, with a goal of scheduling every eligible establishment
at least once a year. Other codes (such as "B", "C", and "D") represented sample sets collected from establishments
targeted for follow-up testing following a failed set. Beginning with the third 2006 quarterly report, reports
have data summarized from all sets collected. (Note: Earlier reports have not been retrospectively changed).
Consequently, under Table A1, the
annual reports from 2005 and earlier reflect results from only the "A" set samples.
In February 2006, FSIS issued a federal
register notice: "Salmonella Verification Sample Result
Reporting: Agency Policy and use in Public Health Protection" announcing
how FSIS would report and use results from its Salmonella
verification sampling program for meat and poultry establishments.
In this Notice, the Agency announced its intention to redirect its
Salmonella verification sampling program and announced 11
new initiatives to encourage establishments to reassess their food
safety systems to achieve and maintain consistent process control.
As one of these new initiatives, FSIS increased testing frequency
in establishments with variable or highly variable process control
compared to those showing consistent process control.
Since June 2006, establishments have been scheduled based on risk-based
criteria designed to focus FSIS resources on establishments with
the most samples positive for Salmonella and the greatest
number of samples with serotypes
most frequently associated with human salmonellosis, (relative
to each product class) as defined by CDC.2
Furthermore, establishments are grouped into one of three categories.
Category 1 includes establishments whose two most recent Salmonella
set results are equal to or less than 50% of the performance standard
or guidance. The Agency considers these establishments to demonstrate
consistent process control. Category 2 includes establishments where
at least one of their two most recent set results was greater than
50% of the performance standards or guidance without exceeding it,
or they have passed their most recent set but failed the one prior
to that one. These establishments are considered to have variable
process control. An additional subcategory 2T was created in 2008
under Category 2 (reported in 2008 second quarterly report and onwards).
An establishment with its last set ≤ 50% of the performance
standard or guidance and the prior set at > 50% of the performance
standard will now be sub-categorized as 2T, with T standing for transitioning
to Category 1. Category 3 includes establishments whose most recent
Salmonella set result has exceeded the performance standard
for its product class. The Agency considers these establishments to
display highly variable process control.
In 2008, in addition to creating subcategory 2T, two additional policy
changes were introduced:
- FSIS is now rounding down the allowable positives for Category 1 status for those product classes having
odd numbered standards or guidelines. Therefore the current performance standards will be: Turkey
(Category 1: ≤6 vs ≤7 Salmonella positive results), Ground Beef (Category 1: ≤2 vs ≤3
Salmonella positive results), and Ground Turkey (Category 1: ≤14 vs ≤15 Salmonella
positive results). FSIS recognizes that an establishment with highly effective process control should strive to operate
over time at a relatively low level of Salmonella positive samples if it is to maintain its Category 1 status.
The Agency views this more stringent policy as a necessary and important incentive to improve performance in
controlling Salmonella.
- In the second quarter CY 2008, FSIS began testing very low volume
establishments (producing < 1,000 pounds/day) for Salmonella
in the same manner as they are scheduled and tested for E.
coli O157:H7. In this way, the Agency can better allocate
testing resources towards higher risk establishments.
Results
This report presents percent positive Salmonella sample results and percent sample sets meeting the
Salmonella performance standards listed by product class and PR/HACCP establishment size for 2008
and previous years (Table A1 and A2).
Data is presented in aggregate form for the periods 1998-2003 and 1998-2008.
Individual sample results (Table
A1) are counted in the year the sample is collected. Sample sets
(Table
A2) are counted in the year they are completed. For example, samples
from a broiler set collected in 2007 are counted in the 2007 results
(Table A1). However if the sample set was completed in 2008, the set
was counted in 2008 (Table A2). In Table
3, establishments are listed by product classes and percentage
within categories 1, 2T, 2, and 3 for each product class.
Figure 1 (PDF Only) compares Salmonella
percent positive results from 1998 to 2008 with baseline prevalence by product class and year. Percentages of
sample sets meeting Salmonella performance standards by product class from 1998 to 2008 are presented in
Figure 2 (PDF Only).
Broilers
FSIS continues to direct resources toward testing broiler establishments. In 2008 (Table A1),
6,514 Salmonella samples were analyzed from broiler establishments. The total percentage of positive
samples was 7.3% in 2008, down from 8.5%, 11.4% and 16.3% in 2007, 2006, and 2005, respectively. In addition,
of the 149 sets completed in 2008 at broiler establishments, 97.3% met the Salmonella performance standard
compared to 97.2% (215 sets completed) in 2007 and 88.6% (184 sets completed) in 2006 (Table A2).
By the end of 2008, 144 broiler establishments eligible for federal testing were in Category 1, comprising 82% of
all broiler establishments (Table A3)
compared to 74% in 2007. Furthermore in 2008, thirteen broiler establishments (7%) were in Category 2T. Seventeen
broiler establishments were in Category 2 (10%). Two broiler establishments were in Category 3, making up 1% of all
broiler establishments. In 2007, 74% of broiler establishments eligible for Salmonella sampling were in
Category 1.
Market Hog
In 2008, 4,244 samples were analyzed. The total percent of Salmonella positive sample tests was the
same as 2007 (2.6%) and was down from 4.0% in 2006 (Table A1).
Of the 90 sets completed in 2008, 93.3% met the Salmonella performance standard for this product class.
Eighty-six percent of market hog establishments were in Category 1, representing 181 establishments. Eight percent
of the establishments (17) were in 2T. Six percent establishments were in Category 2, representing 12 establishments.
One establishment exceeded the performance standard (Category 3). In 2007, 69% of all market hog establishments
eligible for Salmonella testing were in Category 1.
Cows/Bulls
In 2008, 2,301 samples were analyzed at eligible establishments with 0.5% of the samples positive for Salmonella.
This compared to 3,969 samples analyzed in 2007, (1.1% positive) and 2,246 samples analyzed (0.8% positive) in 2006.
Forty-eight sets were completed in 2008 with 97.9% meeting the Salmonella performance standard for this product class.
Eighty-one percent of cow bull establishments (69 establishments) were in Category 1, 14% (12 establishments) were
in Category 2T, 5% (4 establishments) in Category 2, and no establishments were in Category 3. In 2007, 65% of
cow/bull establishments eligible for Salmonella testing were in Category 1.
Steers/Heifers
During 2008, 4,965 samples were analyzed with 0.2% positive for Salmonella. In 2007, 4,355 samples were
analyzed (0.2% positive) and in 2006, 3,674 samples were analyzed (0.3% positive). In 2008, 64 sets were completed,
with 92.2% of these meeting the performance standard. This compares to 2007, when 57 Salmonella sets were
completed and 94.7% met the performance standard and to 2006 when 33 sets were completed and 93.9% met the
performance standard.
Eighty-six percent (75 establishments) were in Category 1 at the end of 2008, 6% (5 establishments) were in
Category 2T with 7% (6 establishments) in Category 2 and 1% (1 establishment) in Category 3. In 2007, 56% of
steer/heifer establishments eligible for Salmonella testing were in Category 1.
Ground Beef
In 2008, 16,763 ground beef samples were analyzed and 2.4% tested positive for Salmonella. This compares
to 2007, when 13,695 samples were analyzed (2.7% positive) and to 2006 when 17,849 samples were analyzed
(2.0% positive). Three hundred sixty three sets were completed, 87.3% meeting the performance standard,
compared to 2007 when 253 sets were completed (96.0% met the performance standard) and to 2006 when 371 sets
were completed (97.6% met the performance standard).
Seventy-six percent (223 establishments) were in Category 1, seven percent (21 establishments) were in Category 2T,
13% (38 establishments) in Category 2, and 4% (10 establishments) in Category 3. In 2007, 77% of ground beef
establishments eligible for Salmonella testing were in Category 1.
NOTE: Some of the samples and completed sets reported here are from establishments currently
in the low volume sampling program described earlier in this report. These samples were collected and analyzed
prior to initiation of the low volume sampling program which began in April.
Ground Chicken
In 2008, 411 samples were analyzed, with a total of 25.5% percent positive. The total percent of sample sets meeting
the performance standard was 88.9% with nine sets completed.
Fifty percent (4 establishments) of ground chicken establishments were in Category 1 at the end of 2008, 25%
(2 establishments) were in Category 2T with 25% (2 establishments) in Category 2 and no establishments in Category 3.
In 2007, 29% of ground chicken establishments eligible for Salmonella sampling were in Category 1.
Ground Turkey
In 2008, 876 samples were analyzed with a total percent at 15.4%, compared to 17.4% in 2007 (820 sample) and
20.3% in 2006 (444 samples). Eighteen sets were completed and all 18 passed for 2008, similar to 2007, when
all 15 sets completed passed and to 2006 when all 12 sets completed passed.
Seventy-one percent (17 establishments) of ground turkey establishments were in Category 1, 21% (5 establishments)
were in Category 2T, with the remaining 8% (2 establishments) in Category 2 and none in Category 3. In 2007,
64% of ground turkey establishments eligible for Salmonella sampling were in Category 1.
NOTE: The sampling frames for ground chicken and turkey are being re-examined. FSIS is
considering redefining ground poultry product eligible for testing as all raw comminuted chicken, turkey, or
chicken/turkey/beef/lamb/pork mix including mechanically separated poultry produced at federally-regulated
establishments that is not further processed into RTE product.
Turkeys
In 2008, 129 turkey samples were analyzed, with a total percent at
6.2%, and total percentage positive was same as that in 2007. Of the
five sets completed in 2008, all sets (i.e. 100% of establishments)
met the Salmonella performance standard.
Ninety-seven percent (31 establishments) were in Category 1, with one establishment (3%) in Category 2T.
In 2007, 87% of turkey establishments eligible for Salmonella testing were in Category 1.
This report reflects data collected during 2008. The Agency has reviewed the progress made since 2006 and will
issue a report this year.
Note: Restructuring how Salmonella sets are scheduled
means that comparison of results from 2006 onwards to previous years will be less meaningful in terms of trends.
Similarly, the changes to the verification program will prevent valid comparisons of testing results over time
(e.g., quarter-to-quarter or year-to-year trends).
Tables and Figures for this report
|
Last Modified:
June 8, 2009 |
|
|
|
 |
 |
|
|
|
|
|
|
 |
|
 |
|
|
|
|
|
To view PDF files you must have Adobe Reader installed on your computer.
|
|
|
|
|
 |
|
|
|
