Pem Book

pathogen environmental
monitoring program (PEM)
Presented by the Almond Board of California

Preventing
Salmonella
Recontamination:
Pathogen
Environmental
Monitoring
Program Guidance
Document
2
INTRODUCTION
Introduction
A
review of the history of the microbiological hazards as-
sociated with tree nuts and nut products, including al-
monds, shows that Salmonella spp. is one of the primary
target pathogens of concern at all stages of production
and handling. The Hazard Analysis Critical Control Point (HACCP)
approach is the most widely accepted food safety management
framework around the world and is endorsed by the Almond Board
of California for use by almond handlers for the production of safe
and wholesome products. HACCP is a systematic science-based ap-
proach that identifies, assesses and controls the risk of biological,
chemical and physical hazards in the product (1). The HACCP ap-
proach consists of seven principles, each of which must be followed
for successful implementation. The first HACCP principle states that
the HACCP team should conduct a hazard analysis that assesses
the food safety hazards that are reasonably likely to occur and that
must be controlled for the production of safe product. Since history
has revealed Salmonella spp. to be one of the reasonably likely food
safety hazards to occur within the almond production environment,
it is critical that an almond facility’s HACCP plan identifies, assesses,
and seeks to control and mitigate that risk.
An effective mechanism for controlling the risk of Salmonella in the

almond production environment is the implementation of a Patho-
gen Environmental Monitoring (PEM) program. As an integral com-
ponent of an almond facility’s HACCP plan, a well developed PEM
program will control and mitigate the risk of Salmonella spp. con-
tamination proactively in both the production and post-production
environment. This document is intended to outline the necessary
tools and steps to develop and implement such a program.
3
INTRODUCTION
Characteristics of Salmonella
The genus Salmonella consists of rod-shaped,

gram-negative, non-sporeforming, predomi-
nately motile cells that can grow either aerobi-
cally or anaerobically and are members of the
family Enterobacteriaceae. In nature, they are
found in both warm- and cold-blooded animals,
humans, and in the general environment, includ-
ing the soil and water (2). Salmonella spp. can
grow in the temperature range of 41°F (5°C) to
113°F (45°C) with the optimum growth range of 95°F (35°C) to 109.4°F (43°C). Growth is
slow at temperatures below 50°F (10°C), and most strains do not grow at temperatures
44.6°F (less than 7°C). Salmonella spp. has a pH range for growth of pH 4 to 9 with an
optimum of 7 to 7.5. Almonds fall within this optimum pH range. Salmonella spp. is classi-
fied according to serotype, and over 2,400 serotypes have been described for the genus.
Since Salmonella spp. are vegetative bacteria, they are not as heat resistant as bacte-
rial sporeformers. Strains of Salmonella do vary, however, in their ability to resist heat.
For example, the strain Salmonella Senftenberg 775W is about 10 to 20 times more heat
resistant than the average strain of Salmonella at high water activity (Aw). Aw is a mea-
surement used to describe the water that is available in a food for the microorganisms
to grow. Under favorable conditions, Salmonella spp. can grow in the Aw range of 0.94 to
more than 0.99, with 0.99 being the optimum Aw for growth. Typically the water activity
of almonds under proper storage conditions is below the level required for Salmonella
growth. However, the addition of moisture or water can create conditions that allow Sal-
monella to grow if it is present on the nuts. Of particular relevance to almonds and other
nut products is the ability for Salmonella spp. to survive for long periods under dry condi-
tions. Scientific studies have shown, for example, that Salmonella Enteriditis Phage Type
30 can survive for up to 550 days on almond kernels held under a variety of common
storage conditions (3). Numerous studies have shown that Salmonella spp. can survive
for long periods of time in foods and in farm/food plant environments, as well, when they
become desiccated (4, 5). Although low water activity will inhibit growth, Salmonella spp.
can also survive for extended periods of time in a low Aw environment. For example, one
study has shown that Salmonella spp. can survive for up to 24 weeks in peanut butter,
with a higher incidence of survivors in product stored at 41°F (5°C) versus 69.8°F (21°C)
(6). Furthermore, the Aw of a food product can impact the heat tolerance of Salmonella
spp. One investigation found that low Aw was detrimental to Salmonella survival at 131°F
(55°C) or 140°F (60°C), but that temperatures greater than 158°F (greater than 70°C)
were always protective—meaning that the pathogen was harder to inactivate at the high-
er temperature (7). Research has shown that once Salmonella spp. contaminates peanut
butter, it is not realistically possible to eliminate it with heat treatment—the reduced Aw
environment of peanut butter is highly protective of the pathogen (8).
4
INTRODUCTION
Salmonella and human illness
All serotypes of Salmonellae are considered human pathogens. However, the severity of
illness varies greatly depending on the strain involved and the susceptibility of the host.
Salmonellosis is the leading cause of food-borne illness in the United States, with a re-
ported 16.2 cases per 100,000 population in 2008 reported by the Centers for Disease
Control and Prevention (9). Salmonellosis is an infection with symptoms consisting of
diarrhea, abdominal pain, chills, fever, nausea, vomiting, dehydration and headache (2).
The most susceptible individuals are infants, followed by the elderly and immunocompro-
mised, and then the general population. Symptoms usually appear within 12 to 36 hours
with a range of 5 to 72 hours after ingestion of the contaminated food. Symptoms usually
resolve in healthy individuals within 2 to 5 days, and the disease is generally self-limiting.
In a small percentage of cases, complications can occur, including systemic infection and
reactive arthritis, which can have long-term, disabling effects on the patient. Overall, the
human mortality rate for salmonellosis is low (less than 1 percent), but can be as high as
20% depending on strain.
The infective dose for salmonellosis is dependent on a number of variables including the
strain of Salmonella spp. ingested, the susceptibility of the individual and the type of
food consumed. However, small numbers, as low as 15 to 20 cells, have been shown to
cause illness in humans; therefore, great care must be taken to prevent recontamination
of product once it has been pasteurized (10).
Persistence of Salmonella in the almond growing and production environments
Research has shown that Salmonella

Enteriditis Phage Type 30 has persisted
in a single almond orchard for over a
five-year period and that the same or-
ganism can persist and grow in almond
orchard soils for extended periods
of time (11, 12). Salmonella spp. can
survive for weeks in aquatic environ-
ments, including irrigation water used
for crops. And surveys have shown that
the organism can survive for months in
soils and sediments (13). Research has
also shown that Salmonella Enteriditis
Phage Type 30 can rapidly grow to
high levels in wet almond hull and shell
slurries and can survive drying of the hulls, which can become a source of recontamina-
tion during almond processing if not properly controlled (14). Salmonella spp. has been
found to colonize and persist on conveyor belts (dependent on type of belt material) and
5
INTRODUCTION
on various types of fabrics for weeks

(15, 16). Various serotypes of Salmonel-
lae were found in the production envi-
ronment of an oilmeal processing plant,
including the processing floor, dust
samples, employee shoes, brooms, con-
veyors and other sites (17). Research
has also shown that pests and vermin
such as rodents, birds, cockroaches and
houseflies can harbor, transmit and am-
plify the presence of Salmonella spp. in
the environment (18, 19, 20, 21).
Prevalence of Salmonella in almonds and other raw agricultural commodities
Almonds, as with most agricultural commodities such as grains, spices and raw cocoa,
have been found to harbor Salmonellae and other pathogens. Salmonella spp. has been
isolated from and shown to survive in pecans, peanuts, pistachios, dried edible seeds
(sesame, alfalfa, melon, sunflower, flax), Brazil nuts, hazelnuts, macadamia nuts, walnuts
and almonds (22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33). One study showed a Salmonella
incidence of 1.7% in raw almonds, and another study showed an overall isolation rate of
0.87% + 0.2% from raw California almonds sampled over a five-year period (28, 29). The
latter study showed that Salmonella spp. was present at low levels in positive samples.
Cases of human salmonellosis associated with almonds and other nuts
Given the existence of naturally occurring Salmonella spp. in the environment it is not
surprising to find the pathogen in raw nuts and similar products. These products need to
be handled by the processor as if they are contaminated with Salmonella spp., and mea-
sures must be taken to prevent recontamination of treated product.
Three documented salmonellosis outbreaks traced to the consumption of peanut butter

made with contaminated peanuts have occurred. The first outbreak occurred in 1996 in
South Australia and was traced to a single peanut butter manufacturer that used peanuts
from an outside supplier that were recontaminated after roasting with Salmonella Mban-
daka, a relatively rare strain (34, 35). The other two peanut butter outbreaks occurred in
the United States in 2006 – 2007 and 2008 – 2009 (36, 37). Over 600 people were sick-
ened in each of these outbreaks. The common theme to all three of these peanut butter
outbreaks were multiple major deficiencies in the manufacturing plant that led to recon-
tamination of finished product prior to packaging.
Almonds have recently been the cause of several salmonellosis outbreaks. The first out-
break linked to the consumption of raw almonds occurred in 2000 – 2001 and caused ill-
6
INTRODUCTION
nesses in Canada and the United States due to a rare strain, Salmonella Enteriditis PT 30
(27). The second outbreak traced to the consumption of raw almonds occurred in 2003
– 2004, with illnesses again occurring in Canada and the United States, this time due
to Salmonella Enteriditis PT 9C (38). Product was recalled from more than 10 different
countries. This second outbreak led to the promulgation of the rule for the mandatory
treatment of raw almonds to achieve a minimum 4-log reduction of Salmonella (39). In
2005 - 2006, a cluster of illnesses caused by rare subtype Salmonella Enteriditis occurred
in Sweden that was epidemiologically linked to the consumption of raw almonds (40).
While the pathogen was not isolated from any of the almonds tested in the investigation,
statistically there was a very high matched-odds ratio in the case control study conduct-
ed by the Swedish authorities.
In early 2009, the U.S. Food and Drug Administration found multiple samples of pistachio
nuts and pistachio-containing products from one specific company to be contaminated
with multiple serotypes of Salmonellae, including Salmonella Montevideo, Salmonella New-
port, and Salmonella Senftenberg (41, 42). While no definitive cases of salmonellosis have
been linked to pistachios, the Centers for Disease Control and Prevention reported that
one patient in Connecticut with a matching Salmonella strain DNA fingerprint reported
consuming a pistachio-containing product.
The issue of product recontamination
As discussed, there are many other food-borne illness out-

breaks linked to the consumption of low moisture, low Aw foods,
including various edible nut and seed products, chocolate and
confections, dried dairy powders, and spices. It is indisputable
that raw agricultural commodities such as raw almonds occasion-
ally contain various levels of Salmonellae. Salmonellae can persist in
dry products or in food production environments for long periods
of time. It is incumbent on the processor to ensure that lethality
steps such as steam or PPO treatments for almonds are validated to
achieve a minimum 4-log kill of Salmonellae. However, equally as im-
portant as the validated kill step is the prevention of product recontamina-
tion prior to packaging. It is wasted effort to implement a validated minimum
4-log kill step only to have the treated or pasteurized product recontaminated with
Salmonella prior to packaging.
Data indicates that post-lethality recontamination is a major cause of food-borne illness

and product recalls. A World Health Organization (WHO) survey conducted in Europe
found cross-contamination during processing to be the most important factor relating
to the presence of pathogens in prepared foods (43). A survey of food-borne outbreaks
in the United Kingdom found cross-contamination to be a significant contributing factor
in 32.1% of the cases (44). An investigation of Salmonella spp. cross-contamination in an
7
INTRODUCTION
oilmeal processing plant found that controlling traffic flow (personnel and materials), ro-
dents and airborne dust were key factors in reducing contamination rates (17). In two ma-
jor peanut butter salmonellosis outbreaks, uncontrolled dust in the production facilities
was thought to be a major contributing factor (46). Published guidelines for minimizing
microbial cross-contamination in poultry feed mills indicate that dust control in the feed
milling facility is essential for controlling Salmonellae (45). Other key factors include the
control of employee traffic patterns to minimize the possibility of cross-contamination
and the control of rodents and wild birds. The International Commission on Microbiologi-
cal Specifications for Foods (ICMSF) recognizes that, while it is not possible to prevent
the introduction of pathogens into food processing facilities, it is crucial to minimize their
presence (47). The ICMSF stresses a number of potential sources in food-processing ar-
eas that must be controlled in order to minimize the potential for product recontamina-
tion:
@OeOU`WQcZbcÒZQ][[]RWbWSaacQVOaÒec\ÔabSc`WhSROZ[]\Ra`S_cW`S^VgaWQOZ
separation through plant design and layout in order to minimize entry of pathogens
into processed product areas.
4]]RVO\RZSàO\R[OW\bS\O\QS^Sà]\\SZQO\PSa]c`QSa]TQ]\bO[W\ObW]\O\R[cab
be trained in proper hygiene principles.
>Sà]\OZQZ]bVW\UW\Ô`bWQcZOàV]SaQO\bÒ\aTS`ÔbV]US\aT`][]\SO`SOb]O\]bVS`
and must be controlled.
/WÒ\ReObS`[cabPSQ]\b`]ZZSR Compressed air filters, often used for blowing down
and cleaning processing equipment, can be a source of contamination if not properly
maintained. Water aerosols can disperse microorganisms throughout the facility if not
controlled.
7\aSQbaO\R]bVS`^SabaQO\OQbOadSQb]à]TÔbV]US\bÒ\a[WaaW]\W\bVST]]R[O\c-
facturing plant if not properly controlled.
BÒ\a^]`bS_cW^[S\bacQVOaÒQYab`]ZZSgaQO`baT]`YZWTbaO\RaW[WZO`S_cW^[S\bQO\
he important vectors for transferring microorganisms throughout a facility and should
be limited to use in specific areas.
All of these issues are relevant to the production of almonds and almond products. Dust
control is a critical factor that must be addressed. The introduction of moisture into the
environment should be minimized to the greatest extent possible. The combination of
dust and water can lead to the growth of Salmonellae and other pathogens to high levels
in the environment that can then be subsequently spread throughout the facility. Careful
thought and planning must be done in order to control dust and water in the environ-
ment.
8
SALMONELLA CONTROL ELEMENTS IN THE ALMOND PRODUCTION ENVIRONMENT
Salmonella Control Elements in the

Almond Production Environment
E
ach element in the Salmonella control equation must be addressed to minimize
the potential for product recontamination. Each one of these elements should key elements
have a detailed, documented plan to address them. A lapse in any one element The key elements
increases substantially the risk of product recontamination. required for control
of Salmonella
recontamination in an
The Grocery Manufacturers Association (GMA) has developed a document for food almond production
manufacturers titled the “Control of Salmonella in Low-Moisture Foods.” This document facility can be
conceptualized in the
and its companion annex document Salmonella control
provide guidance on the control of equation:
Salmonella when manufacturing low-
TRAFFIC CONTROL
moisture foods such as almonds (48, (PERSONNEL &
49). The GMA guidance document EQUIPMENT)
+
outlines seven elements that should DUST CONTROL
be applied to control Salmonella in +
low-moisture products. These seven WATER CONTROL
+
elements are consistent with the ele- SEPARATION OF
ments of the Salmonella control equa- RAW & PASTEURIZED
PRODUCT
tion and include: +
EFFECTIVE
1. Prevent ingress or spread of Salmonella in the processing facility. CLEANING &
SANITATION
A hazard analysis should be conducted by a cross-functional team to determine the SALMONELLA

CONTROL
potential sources of Salmonella spp. in the plant. Example of potential sources include
incoming raw materials (almonds and other materials), utensils and equipment, person-
nel and traffic flow, rodents and birds,
airflow, and overall facility design and in-
tegrity. Ingredients known to be contami-
nated with Salmonella should be segre-
gated. Sanitation and cleaning proce-
dures should be developed that limit the
use of water in the production environ-
ment. Employees must be educated on
the potential sources of contamination,
the need to adhere to traffic patterns,
and proper hygienic practices and pro-
cedures needed to prevent the spread of
Salmonella spp. in the facility.
9
2. Enhance the stringency of hygiene

practices and controls in the primary
Salmonella control area.
The Primary Salmonella Control Area

(PSCA) in a low-moisture product fa-
cility is the area where the handling
of ingredients and product requires
the highest level of hygiene control.
Examples within an almond process-
ing facility include, but are not limited
to, sorting and packing lines, finished-
goods packaging, and pasteurization
equipment and the immediate surrounding areas. The key concept is to, protect product
open to the environment prior to packaging. The PSCA should be physically separated
from the rest of the facility. Movement of personnel and materials should be controlled
between the PSCA and the remainder of the facility. Facility layout should be such that
the PSCA is protected from recontamination from the rest of the facility to the greatest
extent possible.
3. Apply hygienic design principles to building and equipment design.
Building layout and equipment design should be based on solid hygienic principles (50).
In particular, the layout and design of equipment and processes in the PSCA should be
well thought out. All efforts should be made to minimize the accumulation of dust and to
exclude moisture from the processing environment through the use of proper dry-clean-
ing practices.
4. Prevent or minimize growth of Salmonella within the facility.
The control of moisture in the manufacturing environment is absolutely crucial to the pre-
vention of Salmonella contamination in low-moisture products such as almonds. Dry con-
ditions must be maintained at all times in the PSCA, except where controlled wet clean-
ing is deemed necessary in special cases such as a product contamination issue. If water
ingress occurs within the PSCA such as through a drain back-up, roof or wall leak, leaky
steam valves, or leaking water pipe, procedures must be implemented to immediately ad-
dress the problem. Once the leak is repaired, the area must be returned to hygienic con-
dition through proper cleaning and sanitation. If wet-cleaning procedures are used, the
area must be thoroughly dry before being put back into operation. Intensive sampling of
the environment for Salmonella and indicator organisms should be conducted to verify
that the cleaning and sanitation procedures were effective in returning the area back to
hygienic condition.
10
5. Establish a raw materials/ingredients

control program.
An approved ingredient supplier program should be

implemented, particularly for those ingredients that
are considered “Salmonella sensitive.” These ingre-
dients are those that have a known history of asso-
ciation with Salmonella spp. “Salmonella sensitive”
raw materials that have been pasteurized or treated
should be segregated upon receipt from other raw
ingredients that have not been treated. On-site au-
dits of ingredient suppliers should evaluate food
safety procedures in their establishment. This means
an implemented and valid HACCP plan with proper
prerequisite programs including GMPs, an aggres-
sive Pathogen Environmental Monitoring (PEM)
program, sanitation practices, raw materials/ingredi-
ents storage, process validation, employee training,
a finished-product hold-and-release program (if finished-product testing is performed),
traceability, and a corrective action plan if positive Salmonella results are found.
6. Validate control measures to inactivate

Salmonella
By government regulation, almonds must be processed to achieve a minimum 4-log in-

activation of Salmonella spp. Proper protocols must be followed in order to validate the
process. The Almond Board of California has issued several guidance documents on the
validation of processes for almonds (51, 52, 53, 54, 55). An external process authority The most
should be used if internal expertise is not available to conduct proper validation proto- effective
cols of the process. Once a process is validated to inactivate Salmonella, the challenge to verification tool
the processor is to prevent recontamination of that product through subsequent handling for determining
the effectiveness
and packaging.
of a facility’s
Salmonella
7. Establish procedures for verification of Salmonella controls and corrective actions. controls is the
implementation
The most effective verification tool for determining the effectiveness of a facility’s Salmo- of an aggressive
nella controls is the implementation of an aggressive Pathogen Environmental Monitoring Pathogen
(PEM) program. A PEM program is an ongoing measure of the effectiveness of the overall Environmental
Salmonella control program in the plant. A PEM program, in itself, is not a Salmonella con- Monitoring
trol program, but it provides feedback on where efforts need to be directed for the over- program (PEM).
all control program. The focus of an aggressive PEM program is in the PSCA, but areas
remote from the PSCA should be included and will be discussed in further detail.
11
Each manufacturer should decide on the value of testing finished product for Salmonella
spp. Producers need to understand the statistics behind finished-product testing. In gen-
eral, finished-product testing as a sole measure of product safety is a poor approach in
the absence of a robust Salmonella control program. If finished-product testing is per-
formed, it is critical that the lots be segregated, placed on hold, and then released into
commerce only after a negative result is obtained. If a sample tests positive for Salmo-
nella spp., the lot is considered adulterated and should not be released into commerce. In
no case should the lot be retested for Salmonella spp. with the intention of negating the
initial positive result. Corrective actions must be taken and documented when Salmonel-
lae are detected in either environmental or finished product samples.
Focusing on the seven Salmonella control elements discussed in the GMA guidance
document and the elements of the Salmonella control equation will significantly reduce
risk to the product and the consumer. Conversely, ignoring these principles will greatly
increase the risk of a Salmonella recontamination event and present increased risk to
the business and the consumer. Food recall costs can be astronomical, leading to bank-
ruptcy and closure of the business. It has been estimated that the cost of the 2008–2009
peanut butter salmonellosis outbreak caused by the now defunct Peanut Corporation of
America could reach as high as $1 billion in lost production and sales by U.S. peanut pro-
ducers (56). Clearly, adherence to the guidelines outlined in this document could have
avoided such a widespread disaster.
12
P R I N C I P L E S O F A PAT H O G E N E N V I R O N M E N TA L M O N I TO R I N G P R O G R A M
Principles of a Pathogen
Environmental Monitoring Program
T
he International Commission of Microbiological Specifications for Foods (IC-
MSF) has recognized that even with the optimal application of a HACCP plan or
a Good Hygienic Practices (GHP) program, it is no guarantee that recontamina-
tion within the processing environment will not occur (47). Despite strict control
of all critical control points (CCPs) in a process to ensure destruction of pathogens in
raw materials, foods can subsequently become contaminated through one of two ways:
1) the addition of a contaminated ingredient after a kill step, or, 2) recontamination from
the processing environment. In the almond-processing environment, recontamination of
the product can occur through myriad ways, including contact with raw and untreated
materials, processing equipment that is not properly cleaned, manufacturing activities,
maintenance activities, sanitation practices, workers, waste, product rework, pests, and
microbial pockets embedded in equipment and the structure of the building. Control of
recontamination is dependent upon a combination of factors such as those described in
the Salmonella control equation, and includes the following (57):
Q Hygienic design, construction and maintenance of the facility

Q Hygienic operation and maintenance of the processes and equipment
Q Application of appropriate (e.g., dry vs. wet) cleaning and disinfection procedures
Q Training of personnel in food safety and hygienic practices
Microbiological monitoring of the food-processing environment can be performed to

meet a number of objectives:
Q Verifying the effectiveness of cleaning and sanitation practices

Q Determining the frequency required for cleaning and sanitation
Q Determining the presence of food-borne pathogens or their indicators in the
environment
Q Discovering environmental sources of spoilage organisms
Q Determining the frequency required for special maintenance procedures (e.g.,
changing air filters)
Q Evaluatiing the hygienic design and fabrication of food-processing equipment and
facilities
13
The focus of this guidance document is on determining the presence of food-borne

pathogens including Salmonellae or their indicators in the environment of the facility
through the development and implementation of an aggressive Pathogen Environmental
Monitoring (PEM) program. An effective PEM program is a measure of the effectiveness
of the Salmonella controls that the facility has in place—proof of how well the facility is
managing all of the elements of the Salmonella control equation. A PEM program should
be thoughtful and aggressively applied. Employees should never be discouraged from
finding a positive result. If Salmonella is present in the manufacturing environment, find-
ing it through an aggressive PEM program can allow you to do something about it. You
want to encourage employees to find it if it is there.
Getting started
Implementing a PEM program may, at first, seem like a daunting undertaking. However,
through the use of a systematic approach, an effective program can be developed in fair-
ly short order. If your company does not have a food safety professional experienced in
the development and implementation of a PEM program, it is strongly recommended that
you make use of an experienced outside expert or process authority to guide you through
it. A PEM program is specific to the individual facility under consideration and specific
to the individual operations within the facility. There is no “one size fits all” program that
can be used outside of the common principles discussed in this guidance document. The
first order of business is to assemble your team that will develop and implement the PEM
program. It is desirable to have several individuals familiar with the operation help in iden-
tifying potential areas of risk and concern. Examples of such individuals include the plant
quality manager, the plant or corporate microbiologist, line supervisors or operators, and
sanitation supervisors or workers. If you are using an external process authority or expert
to help develop and implement your program, you still should make these individuals
available to work with the expert.
Sampling locations: The PEM zoning concept
Once your PEM team is assembled it is important to understand your process flow and
emphasize identifying potential points of product recontamination. As discussed in the
case of almond and nut processing, Salmonella spp. is the target organism of concern.
A flow diagram of the process can be very helpful, but it is absolutely essential that you
walk the plant floor to determine areas where the product may be vulnerable to recon-
tamination after the lethality step. One useful tool that can help you in site selection and
managing your PEM program data is the PEM zoning concept. In the zoning concept,
plant operations are divided into four zones based on level of risk. These four zones are
illustrated in Figure 1 on page 19 and are defined as follows:
14
Zone 1—Areas in the plant that are direct product contact surfaces after the lethality or mi-
crobial reduction step (e.g., roaster) and before the product is sealed in the primary pack-
age. If there is no lethality step in the process, Zone 1 areas are those where the product is
exposed to the plant equipment and environment prior to sealing in the primary packag-
ing. Examples of Zone 1 surfaces in the almond production environment include:
Q Conveyor belts and buckets

Q Utensils
Q Employee hands (if touching product)
Q Slicers and dicers
Q Product hoppers, bins and bin liners
Q Discharge chutes
Q Fillers
15
Zone 2—Non-product contact areas in the plant that are closely adjacent to product con-
tact surfaces. Examples of Zone 2 surfaces in the almond production environment
include:
Q Equipment framework
Q Drip shields and housings
Q Control panels and buttons
Q Overhead pipes directly over Zone 1 surfaces
Q Computer screens
Q Maintenance tools
16
Zone 3—Nonproduct contact surfaces that are in open post-lethality product-processing

areas, but not closely adjacent to Zone 1 surfaces. Zone 3 surfaces, however, have the
possibility of leading to product cross-contamination. Examples of Zone 3 surfaces in the
almond production environment include:
Q Floors, walls, ceilings Q Pallets

Q Hoses Q Foot mats
Q Air handling units Q Foot baths
Q Condensate drip pans Q Drains
Q Trolleys, forklifts, walk-alongs, carts Q Brooms, mops and squeegees
Q Trash containers Q Toolboxes
17
Zone 4—Areas remote from post-lethality product-processing areas. Zone 4 areas, if not
maintained in good hygienic condition, can lead to cross-contamination of Zones 1, 2 and
3. Examples of Zone 4 surfaces in the almond production environment include:
Q Hallways Q Coolers and freezers

Q Loading docks Q Maintenance shop
Q Warehouses Q Office areas
Q Bathrooms
Q Locker rooms
Q Cafeteria and break rooms
18
Figure 1. Zone Concept to Illustrate Areas of Highest Risk (Zone 1) to Lowest Risk (Zone
4) for Product Contamination.
ZONE 4
locker rooms, cafeteria, halls
warehouse, loading dock
ZONE 3
phones, hand trucks, forklifts, walls,
floor and drains
ZONE 2
nonproduct contact surfaces in close
proximity to product (exterior of
equipment, chill units, framework,
equipment housing)
ZONE 1
product contact
surfaces (slicers,
conveyors, peelers,
strip tables, utensils,
racks, work tables,
employee hands,
dicers, pumps)
Following the principles of zoning allows you to take a rational approach to sample site
selection and managing the overall PEM program. It can also be used as an effective
teaching aid for plant personnel and senior management. It is important for you to define
what constitutes Zone 1 to 4 areas in your specific facility and be consistent. Once you
have determined Zones 1 to 4 in your facility, you then need to give careful consideration
to what specific test methods you are going to employ before you begin testing.
19
PEM sampling in raw product areas
Raw product areas are not the primary focus of PEM monitoring, since it is assumed that
these areas will be contaminated with salmonellae from time to time. However, there is
value in monitoring these areas because it is not desirable to have high levels of salmo-
nellae build up in raw product. The mandatory 4-log Salmonella lethality process, as re-
quired in the United States, would be inadequate if high levels of the pathogen (greater
than 104 CFUs) were present in raw almonds. It is not unexpected to occasionally find
salmonellae in raw product areas. However, a frequent incidence might suggest that
cleaning procedures for those areas are not adequate or that a niche or moisture in the
environment may be allowing the pathogen to persist and grow. One recommended
approach would be to use total Enterobacteriacae (TEB) counts as a quantitative indi-
cator of the potential presence of salmonellae. The area should be monitored for both
salmonellae and TEB counts after cleaning to ensure effectiveness of those areas. If
TEB counts exceed 102 CFU per area sampled or per sponge, that suggests cleaning
procedures were not effective and need to be repeated (a 102 CFU action limit provides
a 2-log margin of safety for almonds that are going to be treated using a 4-log lethal-
ity process). The PEM zoning nomenclature can still be used for raw product areas with
slight modification to the definitions. Zone 1 areas are those where product comes into
contact with processing equipment. Zone 2 areas are those sites that are closely adja-
cent to Zone 1 areas. Zone 3 sites are those in the open product processing areas, and
Zone 4 sites are those remote from Zones 1, 2 and 3 areas. It is imperative that the re-
sponse team carefully consider what actions need to be taken in the event of a positive
result in raw product areas.
Types of testing for pathogen environmental

monitoring
There are myriad methods that can be used for

your PEM program. The choice of method de-
pends on a number of considerations. The first
consideration is to determine which elements
you want to include in your PEM program. It is
recommended that you include the following
components in your PEM program:
Q Surface sampling using sponges or swabs

Q Product residue scrapings/fines/dust samples
Q Water samples
Q Air samples
The second consideration is to determine the type of testing you are going to employ.
Generally there are two categories of methods you can use: 1) testing for indicator organ-
20
isms, and 2) testing for the specific target pathogen (Salmonella spp.). Indicators are used
to measure the potential presence of pathogens and to assess the effectiveness of cleaning
and sanitation (58). Food safety indicators should meet the following important criteria:
1. Be easily and rapidly detectable

2. Be easily distinguishable from other members of the food microflora
3. Have a history of constant association with the pathogen whose presence it indi-
cates (e.g., Salmonella spp.)
4. Always be present when the pathogen of concern is present
5. Be an organism whose numbers ideally should correlate with those of the pathogen
of concern
6. Possess growth requirements and a growth rate equaling or exceeding that of the
pathogen
7. Have a die-off rate that at least parallels that of the pathogen and ideally persists
slightly longer than the pathogen of concern
8. Be absent from foods that are free of the pathogen except perhaps at certain
minimum numbers
There are a number of indicator tests that can be used for PEM programs in almond-pro-
cessing operations. One common indicator test is the coliform and Escherichia coli group,
which is commonly used in the food industry as sanitation and process integrity indica-
tors and for HACCP verification (59). Another highly recommended indicator test is the
total Enterobacteriaceae (TEB) count, which has been widely used in Europe as a food
safety and process integrity indicator test. The Enterobacteriaceae group is superior to the
coliform group as an indicator of sanitation because this group, collectively, has greater
resistance to the environment than the coliforms, can colonize areas where sanitation and
cleaning have been insufficient, and members of this group are sensitive to sanitizers.
While the coliform/E. coli and the total Enterobacteriaceae groups are not perfect indica-
tors of the presence of Salmonella spp. in the processing environment, they nevertheless
are good indicators of cleaning and sanitation practices. Although there is a lack of data
correlating TEB counts and Salmonellae in environmental samples from almond and nut
processing operations, data does exist showing the existence of a loose correlation be-
tween TEB counts and the occurrence of Salmonellae in environmental samples from a
dried-milk processing plant (60):
Total Enterobacteriaceae cfu/g Salmonella positive in 50 g percentage

<2 0.5
2 - 100 0.9
100 - 500 8.7
> 500 9.0
21
Note that in the above data Salmonellae were detected when TEB counts were at the
limit of detection of the method (less than 2 CFU/g). It must be emphasized that it is
possible for Salmonellae to be present when the TEB counts are negative (the same is
true for coliform/E. coli counts, as well). This is because enrichment methods used for
the detection of Salmonellae are more sensitive than quantitative TEB or coliform/E. coli
counts. However, the loose correlation in the previous table shows that the percentage
of positive environmental samples for Salmonellae dramatically increase with increas-
ing TEB counts. Therefore the risk of having Salmonellae in your processing environment
increases with increasing TEB counts. Quantification of TEB and coliform/E. coli can be
conducted by standard plating or cultural techniques, including Most Probable Number
(MPN) methods (57). One convenient method that can be used for quantifying either
the TEB or coliform/E. coli groups is the 3M PetrifilmTM method (www.3M.com/product/
information/Petrifilm-plate.html) as shown in Figure 2 (61).1 These plates are provided by
the manufacturer in sealed foiled pouches that are stored under refrigeration until use or
until the end of the expiration date on the label.
A third indicator test that is widely used as a quality indicator for evaluating foods and
food-processing operations is the aerobic plate count (APC). APCs cannot be used as
safety indicators for pathogens (Salmonella spp.) because in almost all cases there is no
correlation between APCs and the presence of pathogens or their toxins. There are ap-
plications, however, where an APC count can be used as an indication of sanitation effec-
tiveness of a process. APC data from dry processing environmental samples can be dif-
ficult to interpret because dry-cleaning procedures will not remove the entire microflora
present on equipment, including sporeforming bacteria. Consequently, APCs can vary
widely depending on the quality of ingredients or product processed on the line. APCs
can be conducted using standard pour-plate methods, the 3M PetrifilmTM methods, or by
the Most Probable Number (MPN) technique, among others (62).
Figure 2. 3M PetrifilmTM Plate Method
Add sample Roll down film Distribute sample

with spreader
1
Mention of commercial names does not constitute an endorsement by the Almond Board of California.
22
Environmental sampling techniques
Sampling procedures and techniques should be conducted by properly trained person-

nel, consistent with standard industry practice outlined in this guidance document. Test-
ing of environmental samples, including fines, debris, sweepings, sponges, and swabs,
provide critical information and feedback on how effective your measures are in control-
ling Salmonellae in your almond manufacturing operations.
Procedures sampling equipment and environmental surfaces for salmonellae
When testing equipment and environ-

mental surfaces for Salmonellae, it is
important to sample as large an area as
reasonably possible. The use of sterile
sponges or the 3MTM Sponge-Stick (Fig-
ure 3) (http://solutions.3M.com/wps/
Microbiology/FoodSafety/) is a very
useful means of sampling large areas
for Salmonellae. Prepared hydrated or
dry sponges in sterile Whirl-Pak® bags
are available from a variety of vendors.
Sterile swabs such as the 3MTM Quick
Swab, the 3MTM Enviro Swab or cul-
turette swabs can be used for sampling small areas such as cracks, crevices, holes, and
other hard-to-reach areas. If sanitizer is used as part of the normal sanitation procedure
in the plant, then sponges or swabs should be placed in sample with neutralizing buffer
(for example, D/E neutralizing buffer). This is essential to recovery of sub-lethally injured
Salmonellae and to ensure that they are able to grow out during culturing of the sponges
or swabs in the laboratory. Sponges/swabs with neutralizing buffer are commercially
available from many vendors.
Dust, fines, floor sweepings and vacuum canister debris can be collected with sterile Figure 3. 3MTM
utensils such as scoops, scrapers and spatulas and placed in sterile Whirl-Pak® bags. Such Sponge-Stick or
sampling utensils may be purchased from a variety of venders, including eNasco (http:// equivalent for Sam-
www.enasco.com/page/wp_index). Sample collection should, where possible, proceed pling Environmental
throughout the almond-processing plant from Zone 1 to Zone 2 to Zone 3 to Zone 4 us- Surfaces
ing the following procedure as a guide:
1. Prelabel the sponge sample bags using a predetermined coding or numbering sys-
tem. Make sure site descriptions indicate from which zone each sample is taken.
2. Thoroughly wash and dry hands. Sanitize hands with an appropriate hand sanitizer.
Put on sterile gloves.
3. Using sterile gloves, remove the sponge or 3MTM Sponge-Stick from the Whirl-Pak®
23
bag or container.
4. Grasping the sponge and using constant pressure, sponge an area as large as reason-
ably possible. Typically, a range of 40 square inches to 400 square inches is used for
testing environmental surfaces. Replace the sponge back into the Whirl-Pak® bag or
container and seal it.
5. Small areas such as cracks, crevices and screw holes may be more appropriately sam-
pled using sterile swabs. Using sterile gloves, remove the swab from its container and
swab the sampling site. Return the swab back to its container.
6. Change gloves between sponge samples and use an alcohol-based sanitizer to mini-
mize the potential for cross-contamination.
7. When Zone 1 sites are sampled with
premoistened sterile sponges the site should be wiped down with an alcohol-based
sanitizer after the sampling is done. Eco-WipeTM FCS single-use quaternary/alcohol-
based wipes (http://www.ecolab.com) are particularly useful for this purpose. These
are approved for nonporous food contact surfaces and are useful in returning the
area swabbed back to hygienic condition, including removing any small amounts of
residual liquid left by the premoistened sponge.
8. Place the collected, sealed sponge samples into a clean container for transport to the
laboratory. Other disposable items such as used gloves, 3MTM Sponge-Stick handles,
Whirl-Pak® bag tear strips, used Eco-WipeTM FCS wipes, and other items should be dis-
carded in appropriate trash containers or another bag or container designated for that
purpose. These items should not be placed in the same container used to collect the
sealed sponge samples.
9. After sampling, immediately transport the samples to the laboratory and refrigerate
until they are tested. Samples should ideally be tested on the same day that they are
collected. In the case of shipping samples to an external laboratory, samples should
be tested no more than 48 hours after they have been taken. If samples are shipped
to an external testing laboratory, they should be appropriately packed with ice packs.
The receiving laboratory should check the temperature upon receipt to make sure
the samples did not warm up during shipment. Receiving temperatures should be no
higher than 45°F upon receipt.
10. A negative control sample should be included with each batch of environmental sam-
ples taken. This is done by removing the sponge from the Whirl-Pak® bag or contain-
er using sterile gloves and then replacing it back into the bag or container. It should
be coded such that the testing laboratory does not know that it is a negative control
sample.
Procedures for sampling equipment and environmental surfaces for total Enterobacteria-
ceae counts (indicator organisms)
It is recommended that preoperational Zone 1 surfaces be routinely tested for total En-
terobacteriaceae (TEB) counts as part of your PEM program in lieu of Salmonella testing.
Zone 1 surfaces could be tested for Salmonella spp. However, if tested, the product made
24
on that line must be held (if there is no documented sanitation break) until testing results
are available. The use of TEB counts obviates the need for hold-and-release testing of
product but can yield extremely valuable information regarding the hygienic status of the
almond-processing lines and equipment. The following procedure for sampling environ-
mental and equipment surfaces for TEB counts should be used as a guide:
1. Prelabel the sponge sample bags or swab sample containers using a predetermined
coding or numbering system. Make sure site descriptions indicate from which zone
each sample is taken.
2. Thoroughly wash and dry hands. Sanitize hands with an appropriate hand sanitizer.
Put on sterile gloves.
3. Using sterile gloves, remove the sponge, the swab or 3MTM Sponge-Stick from the
Whirl-Pak® bag or container.
4. Grasping the sponge or swab and using constant pressure, sponge an area of 200
square inches. If swabs are used, swab an area of 40 square inches. To facilitate ac-
curate coverage of the area, a nonporous plastic template may be used. The template
should be sanitized using a suitable sanitizer such as Eco-WipeTM FCS wipes between
sampling sites. Replace the sponge back into the Whirl-Pak® bag or container and
seal it. Smaller areas may be sampled with the sponge method if it is not possible to
sample a 200 square inch area. The counts per unit area must be adjusted if that is
the case.
5. Change gloves between sponge samples and use an alcohol-based sanitizer to mini-
mize the potential for cross-contamination.
6. When Zone 1 sites are sampled with premoistened sterile sponges or swabs, the site
should be wiped down with an alcohol-based sanitizer after sampling is done. Eco-
WipeTM FCS single-use quaternary/alcohol-based wipes (http://www.ecolab.com) are
particularly useful for this purpose. These are approved for nonporous food contact
surfaces and are useful in returning the area swabbed back to hygienic condition,
including removing any small amounts of residual liquid left by the premoistened
sponge.
7. Place the collected, sealed sponge or swab samples into a clean container for trans-
port to the laboratory. Other disposable items such as used gloves, 3MTM Sponge-
Stick handles, Whirl-Pak® bag tear strips, used Eco-WipeTM FCS wipes and other items
should be discarded in appropriate trash containers or another bag or container
designated for that purpose. These items should not be placed in the same container
used to collect the sealed sponge samples.
8. After sampling, immediately transport the samples to the laboratory and refrigerate
until they are tested. Samples should ideally be tested on the same day that they are
collected. In the case of shipping samples to an external laboratory, samples should be
tested no more than 48 hours after they have been taken. If samples are shipped to an
external testing laboratory, they should be appropriately packed with ice packs. The
receiving laboratory should check the temperature upon receipt to make sure the sam-
ples did not warm up during shipment. Receiving temperatures should be no higher
25
than 45°F upon receipt.

9. A negative control sample should be included with each batch of environmental sam-
ples taken. This is done by removing the sponge from the Whirl-Pak® bag or contain-
er using sterile gloves and then replacing it back into the bag or container. It should
be coded such that the testing laboratory does not know that it is a negative control
sample.
10. Samples should be quantitatively sampled for TEB counts per the procedures out-
lined in the Compendium of Methods for the Microbiological Examination of Foods
(57). One convenient method is to employ the use of the 3M PetrifilmTM method for
conducting TEB counts.
11. Sponge samples are analyzed by adding
100 ml of sterile diluent (e.g., letheen neutralizing broth) to the sample bag. Vigor-
ously massage the sponge for one minute or more to release the microorganisms
and plate according to the TEB method used. If swab samples are taken, vigorously
shake the container holding the swabs by making 50 complete cycles of 15 cm in 10
seconds, striking the palm of your other hand at the end of each cycle. Plate accord-
ing to the TEB method used. Counts should be calculated and reported per unit area
sampled (e.g., TEB count per 200 square inches or TEB count per 40 square inches).
12. When unmeasured surface areas such as pipe interiors, nozzles, valves or gaskets have
been swabbed, the results should be reported on the basis of the entire sampling site
and reported as TEB count per swab or sponge.
Conduct a hygiene zone assessment
A team knowledgeable with the layout of the manufacturing operation should walk the
plant floor to determine sampling locations. As discussed in section 2, the plant should
be segregated into hygiene zones based on the Primary Salmonella Control Area (PSCA)
concept. In almond-processing facilities, the PSCA is the area where the post-lethality-
treated product is exposed to the environment, such as sorting and packing lines, and
final packaging areas. These areas are sometimes referred to as the ready-to-eat (RTE)
area, the critical side of the operation, or the high-hygiene or high-risk area. Examples
of almond plant layout with different levels of hygiene zones are shown in Figure 4. The
objective of hygiene zones is to identify areas of high and low risk within the manufactur-
ing operation. Once these hygiene zones have been identified, specific pathogen control
measures and monitoring programs can be developed. The focus is to prevent the spread
of Salmonellae into the PSCA where protection of the exposed post-lethality treated-
product is critical.
Depending on the type of operation, an almond-processing or manufacturing facility can be

divided into one, two or three hygiene zones in addition to the nonprocessing areas. These
areas would typically be the PSCA, the basic GMP area and a transition area between the
two. An example of a transition area might be between a PSCA and pre-lethality step pro-
cessing as a basic GMP zone.
26
Figure 4. Examples of Almond Plant Layouts layout With Different Levels of Hy-
giene Control. Primary Salmonella Control Area (PSCA) in yellow and basic GMP
area in gray. White areas are non-process areas (adapted from Reference 48).
Brown Skin Facility With Huller/Sheller
warehouse 1
warehouse 2
receiving raw
finished goods
product
brown skin box

packing area
brown
skin bin transitional electronic
packing area sorting,
area metal hulling
detection and
area shelling
hand sorting room incoming

brown
skin sizing
staging
office area
hall
office rest office

room
employee area
locker rooms/rest rooms
Primary Salmonella Basic GMP Area Nonprocess Area

Control Area (PSCA)
27
Handling Facility With Treatment/Pasteurization
employee
roaster
break
room
line 1
hand brown area
sorting skin
line packing finished
goods
rooms
area area
rest
hall
roaster warehouse
line 2
area
office
incoming
transitional
blanching product
area
line 2 warehouse
area
raw
materials
incoming
product roasted/ Primary Salmonella
staging blanched Control Area (PSCA)
packing
area Basic GMP Area
Nonprocess Area
Custom Pasteurization Facility
pasteur-
ization
raw unit
product packing
receiving cooling area
staging chamber
area
conveying unit
rest
room
break room
employee
finished product
staging area
transitional
area 2
rest transitional rest
room room office
area 1
main entrance
Primary Salmonella Basic GMP Area Nonprocess Area

Control Area (PSCA)
28
The team should conduct a hygiene zone assessment of the entire facility and create a col-
or-coded map of the facility using the following procedure:
Q Survey the entire facility, including all production areas, storage, receiving, warehousing
and loading docks, as well as employee facilities such as cafeterias, break rooms, locker
rooms, washrooms, maintenance areas, offices and conference rooms, and others.
Q Designate the PSCA, basic GMP areas, transition areas (if any) and nonprocessing areas.
Q Pay particular attention to areas within the facility where ingredients, products or the
environment could be a potential source of pathogen contamination and have a high
risk to cross-contaminate post-lethality-treated product. Also pay attention to non-
process areas such as refuse and recycling, restrooms, forklift battery charging sta-
tions, boiler rooms, and others that could impact the PSCA.
PEM sampling site selection and frequency of monitoring
Once the team has mapped out the hygiene zones within the manufacturing facility, it
is now time to select the specific sampling sites within each area using the zoning con-
cept discussed in section 3.2. Environmental sampling for Salmonellae is routinely con-
ducted in Zones 2, 3 and 4. Zone 1 surfaces are normally tested for indicators such as TEB
counts. Only under special situations are Zone 1 surfaces sampled for Salmonellae, such
as investigational sampling due to a potential contamination events such as a roof leak or
a finished-product Salmonella-positive result. Zone 1 surfaces could be routinely sampled
for Salmonella; however, any Zone 1 testing for Salmonella or other specific pathogens
necessitates a stringent product hold program until results are received. Testing Zone 1
surfaces for TEB counts or other appropriate indicators obviates the need for a finished-
product hold program.
Table 1 summarizes examples of sampling sites, type of microbiological test and minimum
recommended frequency of testing by zone. Also note that a Zone 1 designation may be
given to equipment surfaces and building structures (e.g., beams, catwalks, overheads,
ceilings, covers, conduit, HVAC units, pipes, etc.) that are directly above product-contact
surface. The assessment team, working together, should determine whether a surface
above a direct product contact surface constitutes a Zone 1 surface. This determination
will depend on a number of factors including the likelihood the surface will contaminate
the product immediately below it (e.g., and condensate formation, dust collection, etc.),
the ability to effectively clean and sanitize the surface on a regular basis, among others.
The number and location of environmental samples taken is determined by the risk levels
inherent to the product and process. Areas with water use, high traffic patterns, a history
of positive pathogen results, and areas where microbiologically sensitive raw materials are
handled or stored should be sampled at a higher frequency. Focus should be given to post-
lethality-treated open product areas (PSCAs) since this is where the risk of product recon-
tamination is highest. In general, this will equate to a greater number of samples collected in
29
Zone 2 and Zone 3 areas than are taken in Zone 4 areas. Sample sites should be identified
and rotated weekly according to shift and the day of the week. A rotation schedule should
be developed to allow all locations within Zones 1, 2 and 3 to be covered within a month.
Zone 4 sites should be rotated so that all sites are covered within a quarter. The sampling
plan must be flexible to allow for additional samples to be taken, as determined by the team.
The overall number of PEM samples taken each week depends on the size of the facility and
on the historical data from the facility.
Establishing your baseline: investigational sampling
Once potential areas for sampling are identified, it is useful to conduct preliminary inten-
sive investigational sampling with the purpose of finding the target pathogen if it is there.
In the preliminary investigational phase, environmental samples are collected at a higher
frequency and number than is done for the ongoing PEM program. The selection of
samples is typically based on the experience of the investigator and the type of process
under consideration. Depending on the size and complexity of the operation, it is not
uncommon to take 25 to 50 samples or more per zone per day for the first month (rotat-
ing shifts in a multiple shift operation), then moving to a weekly schedule with the same
number of samples for the next two to five months. It is highly recommended that you
use a combination of indicator organisms and Salmonellae testing as part of your PEM
program. There are a number of indicator tests that you can use for your PEM program.
As previously discussed, it is recommended that you use total Enterobacteriaceae (TEB)
counts as your indicator test for evaluating Zone 1 and other surfaces. Whether you use
TEB counts or coliform counts as your quantitative indicator method, it is very important
to determine the baseline counts that would be expected under normal operating condi-
tions and what counts that would be unacceptable. This entails doing work to establish
your baseline counts and action levels for counts that deviate from the baseline.
Zone 1 sites are normally tested for TEB counts preoperationally, before sanitizing and pri-
or to start-up of the production line. Sampling after cleaning but before applying sanitizer
is a good measure of cleaning effectiveness. If Zone 1 sites are sampled after the sanitizer
step, then be sure to use a neutralizing buffer for the sample sponges or swabs, as previ-
ously discussed. Zone 1 sites should be tested individually and never composited. Zone 1
sites may be sampled during production, but this will require careful analysis and estab-
lishment of baseline data. This requires collecting TEB count data on your post-lethality-
treated product to determine the expected baseline TEB level. One approach would be
to sample Zone 1 sites intensively for six months to establish baseline levels (preliminary
investigational sampling). Any significant deviation (e.g., 1 log) above the baseline level
constitutes a special cause for corrective action. The team needs to take into consider-
ation variables such as seasonality, geographic differences and supplier source that can
impact the baseline. In general, depending on the degree of post-lethality treatment, the
TEB levels should be very low in the product. The team needs to decide if it adds value to
sampling Zone 1 sites during production, versus taking more Zone 2 and 3 samples during
30
TABLE 1: PEM sampling site examples, type of microbiological test, minimum sampling
frequency, and typical number of samples by zone.
Minimum Typical
Microbiological Frequency of Number of
Zone Examples of Sampling Sites Test Sampling Samples1
I Direct or indirect product contact surfaces2, e.g., sort- Indicator organ- Weekly, post- Line
ing lines, product conveyors product discharge chutes, isms, e.g. Total cleaning prior Dependent
pipeline interiors, storage hoppers, filler hoppers, nozzles, Enterobacte- to sanitizer ap-
product scrapers/utensils, employee hands handling prod- riaceae Counts plication and
uct (TEB), total coli- start-up. Also,
form counts. Sal- as needed for
monella testing investigational,
normally only validation and/
under special or verification
situations purposes
II Environmental surfaces immediately adjacent to product Salmonella Weekly 10 - 15

contact surfaces, e.g., equipment framework/supports,
outside of tunnels or filling cabinets, below filling equip-
ment, control panels, motor housings, catwalks, scales,
scrap containers, drains near zone 1 surfaces, HVAC vents
III Environmental surfaces further removed from product Salmonella Weekly 10 - 15

contact surfaces in open product areas, e.g., hand trucks,
forklifts, walls, ductwork, drains, floors, ceilings, equipment
legs, tools, brooms, squeegees, floor scrubbers, trash con-
tainers, pallets, floor debris, ceiling drain pipes, wash sta-
tions, ingredient storage areas, wall/floor junctures
IV Areas remote from the processing area, e.g., warehouses, Salmonella Monthly 5 - 10
bathrooms, locker rooms, maintenance areas, cafeteria/
break rooms, loading docks, boiler room, offices, plant en-
trance, refuse/recycle areas
1
In general, the same or greater numbers of samples are taken in zone 2 than zone 3, and in zone 3 than zone 4. Larg-
er or more complex operations may require more samples taken per zone than shown here.
2
Direct product contact surfaces (PCS) are surfaces exposed to product during normal equipment operation. Indirect
product contact surfaces are surfaces from which liquids or dust or other material may drain, drop, diffuse, or be
drawn into the product or into the container, and surfaces that touch product contact surfaces or the product con-
tainer
production, combined with a finished product testing program. In some cases, process-
ing equipment such as sorting lines and conveying systems can contain harborage niches
that can be detected only while in operation. If this is a concern, an alternative would be
to run operational Zone 1 testing as part of a periodic verification program rather than
part of the routine PEM program. Also, consideration must be given to weighing the po-
tential risk of inadvertent cross-contamination of Zone 1 sites via sampling during produc-
tion against the insights gained from the data through collecting operational Zone 1 sam-
ples. Factors such as ease of collecting the sample, implications of temporarily shutting
down the line for sample collection and others must be considered before implementing
Zone 1 sampling during production.
Zones 2 and 3 samples should be collected both preoperationally and operationally for
Salmonella. Operational samples should be taken throughout the production run (for ex-
ample, just after start-up, three or four hours after start-up, and at the end of the run).
31
These sampling times and sites can be rotated from week to week.
Zone 4 samples should typically be collected monthly. The focus should be on sites that
are adjacent to open product areas or where traffic (people and materials) flows into
or out of open product areas (PSCAs). Remote locations such as locker rooms, loading
docks, warehouses, cafeterias and break rooms, and other areas should also be included.
Employee lockers, if not properly cleaned and maintained, have been shown to be a
source of Salmonella contamination. The intent of Zone 4 sampling is to identify potential
harborage sites of the pathogen that could ultimately become a source for spreading it
throughout the production facility.
Once swab locations are selected, a master list can be compiled by zone throughout the fa-
cility. Each zone can be mapped for tracking purposes and entered into a master database.
A generator of random numbers can be used to select swab locations to be sampled each
week; however, you should ensure that each location is sampled rotationally so that they
are sampled at least four times, minimum, within a year. The same exact location within a
zone should not be sampled each time samples are collected unless data has shown it to be
a chronic problem location. The sampling plan needs to be flexible, allowing for additional
sample collecting based on the data obtained. The concept of “follow the data” should be
practiced at all times. A PEM program is dynamic and should be responsive to the data
generated by sampling.
Microbiological methods available for testing PEM samples
There are myriad microbiological methods available for the analyses of PEM samples.
Whatever method is selected, it is absolutely imperative that you validate the method on
your samples for your specific applications. It is recommended that you use an official or
industry-recognized method for testing samples. In the United States, the methods in the
FDA’s Bacteriological Analytical Manual (BAM) are considered official methods for the
testing described in this guidance document (63). Other official and/or industry-recog-
nized methods include:
Q ISO 6579 methods, which are considered official methods in Europe, but are increas-
ingly recognized around the world (64)
Q American Public Health Association’s Compendium of Methods for the Microbiological
Examination of Foods (57)
Q AOAC International’s Official Methods of Analysis (65)
There are also country-specific and industry-specific methods published, but most of the
above-cited method references are universally recognized and accepted. Methods that
have been validated and found to be equivalent in specificity and sensitivity to these of-
ficially recognized methods may be used; however, you need to make sure the methods
are properly validated.
32
There are a number of rapid methods available for the detection of Salmonellae in envi-
ronmental, ingredient, and in-process or finished-product samples. Many of these meth-
ods are immunological-based Enzyme Linked Immunosorbent Assays (ELISAs) that utilize
specific antibodies for capture of the Salmonellae, modified cultural methods that often
utilize selective and differential media for the isolation and identification of Salmonellae,
and genetic-based methods such as the Real Time Polymerase Chain Reaction (RT-PCR)
assays that target gene sequences specific to Salmonellae. All of these methods have ad-
vantages and disadvantages and must be carefully considered and thoroughly validated
before being used for your samples. A validated rapid method is generally considered a
screening method, where negative results are accepted as such, but positive results need
confirmation (either cultural or by some other recognized method). It’s advisable to pro-
ceed based on a presumptive Zone 1 positive result from an environmental sample from
a cleaning, sanitation and product disposition perspective. This approach is the most
conservative, gaining you time for cleaning and sanitation and other interdictive steps if,
indeed, it turns out to be a confirmed positive result.
Never composite environmental samples by combining multiple sponges or swabs into

one pre-enrichment. This practice may make it difficult to detect low levels of Salmonel-
lae present in one sample because of increased background competitive microflora lead-
ing to a false negative result. Also, a positive result on a composited sample does not
allow you to identify the specific location(s) that are positive for Salmonellae. This makes
troubleshooting more difficult and results in broader corrective actions than otherwise
swabs
would be needed.
H A N D L E R
Environmental samples may be pooled by combining up to 10 post-enriched samples into
one sample to run a rapid method such as RT-PCR or ELISA. The Pathatrix® Auto is a com-
sterile
sterile
sterile
mercially available system from Matrix MicroScience (www.matrixmsci.com) and approved
bag
bag
bag
by the Association of Analytical Communities Research Institute (AOAC RI). It’s based on
recirculating immunomagnetic separation and allows for the efficient pooling of environ-
pre-enrichment
mental samples. If pooled samples yield a Salmonella-positive result, the individual pre-en-
richment samples comprising the pooled sample are run individually to identify the positive
sample(s). Sample pooling has the advantage of significantly reducing the per-assay cost of
running samples. It is recommended that only samples from the same zone be pooled for
L A B O R A T O R Y
testing purposes. As with most rapid methods, the ability to pool samples is method-depen-
dent and must be thoroughly validated for your specific applications.
selective enrichment
It is highly recommended that all Salmonella isolates be serotyped and characterized by
a genetic typing method such as pulsed-field gel electrophoresis (PFGE), ribotyping, or
another validated and recognized method. Genetic typing methods are very useful in
troubleshooting and tracking data from your PEM program. Genetic typing maps can be
petri dish
developed showing “hot spots” or problem areas in the plant. These maps are usually set
up using blueprints or diagrams of the relevant areas according to zones. It should be
understood that it is possible for multiple strains of Salmonella to be isolated from an en- Sample pooling
33
vironmental sample. Multiple strains of Salmonella have been isolated from raw nuts and
from production areas (33). Therefore, the presence of one strain of Salmonella in prod-
uct and a different strain in the production environment does not necessarily mean they
have no commonality.
If your operation does not have its own microbiological testing laboratory, you should use
a reputable accredited independent testing laboratory. There are a number of resources
available to help you choose a properly accredited independent food microbiology testing
laboratory. The AOAC’s Analytical Laboratory Accreditation Criteria Committee (ALACC)
has published “Laboratories Performing Microbiological and Chemical Analyses of Food
and Pharmaceuticals” (ALACC Guide) which are guidelines based on the ISO 17025 re-
quirements. Up-to-date information on these requirements may be found at http://www.
aoac.org/accreditation/faq2.htm. Other valuable resources that can be used in helping
you find accredited food microbiology testing laboratories include the American Associa-
tion for Laboratory Accreditation (A2LA) (http://www.a2la.org/appsweb/food.cfm) and
the American Council of Independent Laboratories (ACIL) (http://www.acil.org/). The
value of using accredited laboratories is to ensure that they are producing accurate, reli-
able and consistent results using properly validated methods. There are also a number of
independent expert consultants that can be used to help you find and qualify a properly
accredited laboratory.
Data interpretation and corrective actions
Once preliminary investigational data is collected, it must be analyzed and interpreted. In-
tensive data from the preliminary investigational phase is used to set up the ongoing PEM
program. Data, during the preliminary investigational phase, should be continually moni-
tored and used to guide ongoing sampling during that phase. If an area shows repeated
positives, then that area should be considered a potential harborage or problem area that
warrants continued attention. Once the ongoing PEM program is initiated, based on the
intensive data analyzed from the preliminary investigational phase, the frequencies and
typical number of samples per zone outlined in Table 1 may be implemented. It is critical
that corrective actions be implemented and documented whenever a Salmonella positive
occurs. With most environmental samples it is recommended that corrective actions be
initiated when a presumptive Salmonella result occurs. As discussed previously, this is the
most conservative route, gaining time in the event the sample is confirmed positive for
Salmonella. Confirmation can take up to a week; therefore, taking action on a presump-
tive positive minimizes your risk exposure while you wait for confirmed results.
The following are key considerations relative to taking proper corrective actions:
Q Your facility should have a predetermined action plan that would be implemented in
the event of a Salmonella-positive environmental sample result. The action plan should
be specific for each of the four zones and include protocols for:
- The type of immediate corrective actions to be taken by zone
34
- Actions taken to verify that Salmonella has been eliminated from the area in ques-
tion
- An analysis to find the root cause of the contamination so that it can be prevented
in the future
Q All corrective actions, including additional sample results, need to be properly docu-
mented. It is extremely useful to have a computer-based spreadsheet for tracking re-
sults and documenting corrective actions.
Q If a positive result is found in any sampling zone, the area needs to be thoroughly ex-
amined both visually and through vector swabbing to determine the extent of the con-
tamination and to ascertain potential causes of the problem. Vector swabbing entails
taking additional multiple environmental samples around the initial positive site. Vector
sampling is usually done in a typical “star-burst” pattern around the initial positive site
as shown diagrammatically in Figure 5. Typically, 10 to 15 additional sponge or swab
samples are taken around the initial positive sites. Sampling, where possible, should ra-
diate out from the initial positive site in all directions, including up and down, if appro-
priate. Troubleshooting samples are usually run as separate samples and not pooled as
discussed in section 3.8.
Q The specific corrective actions taken are based on an assessment of the likelihood
of finished-product contamination based on the location of the initial positive site. A
positive finding in Zones 2, 3 or 4 does not necessarily implicate finished product. That
decision should be made by the team and appropriate management personnel.
Figure 5. Vector Sponge/Swab

Sampling Starburst Pattern Hot
spot
Around the Initial Salmonella
Presumptive Positive Site.
initial
positive site Starburst
sampling
pattern
35
Table 2 provides some examples of corrective actions following the initial positive Salmo-
nella result in the plant environment. You should have a predetermined cross-functional
response team in place to conduct follow-up investigations on Salmonella positive findings.
The response team should consist of members from Microbiology/Food Safety, Quality
Assurance, Sanitation, Operations, Engineering, Maintenance, Management and other dis-
ciplines, as appropriate. All key personnel that can help troubleshoot find the root cause
of the problem and correct it should be part of the response team. The focus is on finding
and eliminating all potential sources of environmental contamination to the greatest extent
possible.
The response team should conduct a preliminary investigation into any positive Salmo-
nella finding to determine potential sources of contamination. A report should be com-
piled that details all maintenance disruptions and activities, in-plant construction, un-
planned line downtime or other nonstandard production activities (e.g., plant
R & D trials) in the area from the last full microbiological cleanup or sanitation to the cur-
rent positive finding. Immediate actions should be taken to correct any obvious GMP or
other deficiencies based on the findings including:
Q Quarantine the suspect area and limit access to minimize spreading the contamination
to other parts of the plant.
Q Reinforce hygienic practices among employees, outside contractors and others, and re-
train in GMPs and principles of food safety, if necessary.
Q Assess and adjust the type and frequency of cleaning and sanitation procedures, if
needed.
Q Eliminate sources of water and water accumulation, if found.
Q Repair structural damage (e.g., on floors, walls and other structures) as necessary.
Q Reexamine traffic patterns (both personnel and materials) and redirect them,
if feasible.
Q Audit handling practices (production, sanitation, maintenance and material handling)
and make adjustments where necessary.
Q Redesign and/or perform equipment maintenance as necessary.
Q Conduct interdictive cleaning such as floor scrubbing and sanitation, or cleaning of
overhead pipes and equipment.
36
TABLE 2: Examples of corrective actions by zone that should be taken after initial Salmo-
nella positive environmental result
Zone I Corrective Actions

Product should always be placed on hold if zone 1 Salmonella testing is to be done
Quarantine the suspect area and limit access to that area
Break down the line from the initial positive site on for visual inspection, additional vector
sponge/swab sampling, and cleaning and sanitation activities
Conduct vector sampling in zones 1, 2, and 3 around the area of the initial positive result prior to
cleaning. Precaution should be taken not to spread contamination to other areas of the plant
Thoroughly clean and sanitize the line and surrounding area using dry, controlled wet, and/or wet
cleaning procedures appropriate for low moisture environments (47, 49)
Conduct pre-operational inspections on the line equipment and area prior to start-up and take
additional vector samples of the area prior to start-up. It is highly advisable not to start-up the
line until all vector sampling results are obtained (if the line is started prior to obtaining all vector
sampling results, then product must be put on hold until negative results are obtained)
Increase the frequency of intensive sampling of the line and adjacent areas from weekly to daily
(zones 1 – 3). After three consecutive days of negatives are obtained, the normal routine PEM
sampling plan may be reinstituted
The response team should make a careful decision on disposition of finished product that is put
on hold as a result of a zone 1 positive Salmonella finding. All finished product from full microbio-
logical clean-up/sanitation to full microbiological clean-up/sanitation must be addressed by the
team. Product should be re-worked, if possible, or condemned according to all legal and regula-
tory statutes. It is not an acceptable practice to test lots of finished product for Salmonella in re-
sponse to a confirmed zone 1 result for the purposes of releasing product.
Zone II Corrective Actions

Stop production and prepare the system for cleaning and sanitation
Quarantine the suspect area and limit access to that area
Break down the line from the initial positive site on for visual inspection, additional vector
sponge/swab sampling, and cleaning and sanitation activities
Conduct vector sampling in zones 2, and 3 around the area of the initial positive result prior to
cleaning. Precaution should be taken not to spread contamination to other areas of the plant
Thoroughly clean and sanitize the line and surrounding area using dry, controlled wet, and/or wet
cleaning procedures appropriate for low moisture environments (47, 49)
additional vector samples of the area prior to start-up. Do not restart the line until satisfactory
vector swab results have been obtained.
(zones 2, 3). After three consecutive days of negatives are obtained, the normal routine PEM
37
Zone III Corrective Actions

The response team should make the decision whether or not to stop production based on the
proximity of the initial positive site to product contact areas
Quarantine the suspect area and limit access to that area, if feasible
Visually inspect the area and conduct additional vector sponge/swab sampling prior to cleaning
and sanitation activities
Conduct vector sampling in zones 2, and 3 around the area of the initial positive result prior to
cleaning (zone 2 sampling is done to ensure that contamination has not spread closer to open
product areas). Precaution should be taken not to spread contamination to other areas of the
plant
Thoroughly clean and sanitize the area (at least a 50 foot radius, if possible) using dry, controlled
wet, and/or wet cleaning procedures appropriate for low moisture environments (47, 49)
additional vector samples of the area prior to start-up. Do not restart the line until satisfactory
vector swab results have been obtained.
(zones 2, 3). After three consecutive days of negatives are obtained, the normal routine PEM
Zone IV Corrective Actions

A Salmonella positive finding in a zone 4 location does not implicate finished product, but it
does provide information on non-production areas and the potential for spread of contamination
throughout the facility
Quarantine the suspect area and limit access to that area, if feasible
Visually inspect the area and conduct additional vector sponge/swab sampling prior to cleaning
and sanitation activities
Conduct vector sampling in selected zone 3 areas adjacent to the location of the initial zone 4
positive location, if appropriate, and zone 4 sites around the area of the initial positive result prior
to cleaning (selected zone 3 sampling is done to ensure that contamination has not spread closer
to open product areas). Precaution should be taken not to spread contamination to other areas
of the plant
Thoroughly clean and sanitize the area (at least a 50 foot radius, if possible) using dry, controlled
wet, and/or wet cleaning procedures appropriate for low moisture environments (47, 49)
Take additional vector samples of the area after cleaning and sanitation to verify effectiveness of
those procedures
Increase the frequency of intensive sampling of the areas from monthly to daily (zone 4 and se-
lected zone 3 areas adjacent to the location of the initial zone 4 positive). After three consecutive
days of negatives are obtained, the normal routine PEM sampling plan may be reinstituted.
38
It would not be unexpected to occasionally find a Salmonella-positive result in Zone 4

areas such as high-traffic hallways and employee locker rooms. However, a positive find-
ing needs to be aggressively addressed in order to minimize the potential for spread of
the pathogen to other parts of the facility. A Zone 3 positive in the absence of any Zone
2 positives should be considered as an early indication of the need to make the cleaning
and sanitation program more robust or to address other potential issues in the plant like
traffic flow, and structural or maintenance issues.
If vector samples test positive for Salmonella in any zone, then additional samples must
be taken to define the scope of the problem. Additionally, aggressive actions must be un-
dertaken to eliminate the problem. The finding of persistent problem areas or “hot spots”
over time is an indication that the primary contamination source may be a harborage site
where the pathogen has established itself and may be multiplying. In the case of repeat-
ed or persistent problem areas, aggressive corrective actions must be taken to contain
and correct the problem. The following steps should be taken as part of the root cause
analysis by the response team:
Q Map the locations of positive samples on a facility diagram to help define the scope of
the problem.
Q Implement daily vector sampling of the environment until the situation is corrected.
Q Restrict traffic flow in these areas to the extent possible.
Q Visually inspect areas for potential harborage sites and intensify cleaning efforts in these
areas
Q Reinforce GMP and food safety practices with line operators and other personnel.
Q Visually monitor handling practices (production, sanitation, maintenance, material han-
dling) and make adjustments where necessary.
Q Scrutinize equipment cleaning and preventive maintenance practices, then modify if
necessary.
Q Repair structural damage (e.g., floors, walls, other structures) as necessary.
Q Redesign and/or perform equipment maintenance as necessary.
Q Zone 1 sampling or finished-product testing may need to be implemented or intensi-
fied in the event of persistent Zone 2 positives.
In extreme cases where a hot spot cannot be eliminated or contained, serious consid-
eration must be given by the response team to taking that production line or piece of
equipment out of service and physically restricting or segregating that area from the rest
of the plant until a permanent solution can be found.
When using a quantitative indicator such as TEB or coliform counts, none of these or-
ganisms should be present on equipment after cleaning and sanitation activities. Table 3
lists recommended guidelines for aerobic plate count coliform and TEB counts on clean
equipment surfaces before and after application of sanitizer. Typically, preoperational
sampling is done after cleaning but before application of sanitizer. This gives a better in-
39
dication of the effectiveness of cleaning. Samples that are taken after application of sani-
tizer must include the use of proper neutralizing solutions as discussed in section 3.4.1 to
ensure that residual sanitizer does not inhibit recovery of injured cells, if present.
If operational samples are collected for TEB or coliforms, it is critical that a baseline be
established under normal operating conditions, as discussed in section 3.7. An upward
trend or a sudden deviation from the established baseline would be cause to initiate an
investigation and corrective actions. The response team must carefully evaluate trend
data over time to establish what constitutes a significant trend.
Table 3. Recommended Microbiological Indicator Limits for Equipment Cleaning Before

and After Application of Sanitizer
Quantitative Post-Heat Treatment Post-Heat Treatment

Microbiological Indicator Target/ Taken Before Sanitizer – Pre-op Taken After
Test Acceptable Limits (cfu/40 in2) Sanitizer (cfu/40 in2)
Target < 100 < 10
Aerobic Plate Count
Acceptable < 500 < 100
Target < 10 < 10
Coliforms
Target < 10 < 10
Total Enterobacteriaceae
Plant construction, equipment installation, and major repairs
It is well documented that activities such as plant construction, equipment installation

and major repair work can lead to increased recontamination risk to the product if not
properly managed. In the event of such activities, increased control procedures are re-
quired, including:
Q Setting up temporary control barriers within the plant, as appropriate. This may include
physically separating the area through the use of temporary walls, ceiling-to-floor plas-
tic curtains or other suitable containment barriers.
Q Modifying traffic flow in the area to minimize the risk of spreading contamination
throughout the rest of the facility.
Q Increasing the amount of cleaning and sanitation during construction or equipment-
related activities.
Q Reinforcing GMP and hygiene practices with plant personnel and, especially, outside
contractors.
Q If used equipment from outside the plant is installed, it is highly recommended that the
equipment be cleaned and sanitized before it enters the plant. The effectiveness of the
cleaning and sanitation should be verified through sponge or swab sampling before
40
THE ROLE OF AUDITS IN MANAGING A PEM PROGRAM
installation.
Q Airflow and air pressure in the area should also be evaluated and adjusted if necessary
to minimize airborne transmission of dust and contaminants.
Sampling the environment for Salmonella should be performed during construction or

other major activities at an increased frequency and number to ensure that no prob-
lems are being created. Sampling sites and frequency should be determined by the team
based on an evaluation of:
Q Location of construction or other activities.

Q Type of construction or activity (e.g., demolition, installation, major repair, material
removal).
Q Time duration of the activities.
Q Types of environmental controls implemented.
Once construction or major equipment-related activities are completed, the area needs to
be thoroughly cleaned and sanitized. Verification of cleaning and sanitation effectiveness
must be performed by intensively sampling the area for Salmonella contamination before
the area is released for production activities. An aggressive PEM protocol should be fol-
lowed during commissioning of the area or equipment. Depending on the magnitude of
the construction or major activities, this often entails taking hundreds of sponge or swab
samples of the environment. If positive results are obtained, then the area or equipment
must be recleaned/resanitized and resampled until negative results are obtained.
41
The Role of Audits in Managing

a PEM Program
T
he food industry has become increasingly reliant on independent third-party
audits and certification to ensure that their processes, personnel and establish-
ments conform to food safety and other standards. However, third-party audits
have recently come under fire due to several large-scale food-borne outbreaks
linked to establishments that have received high scores from third-party auditing firms
including the 2008–2009 peanut butter outbreak caused by the Peanut Corporation of
America (37, 66). There are a number of factors that have led to this variability in the
quality and consistency of food safety audits:
Q Differences in Experience Levels of Individual auditors—Even though auditors may be

specific to a segment of the food industry (e.g. almond handling and processing), their
experience level can vary greatly. This can lead to difficulties in asking the right ques-
tions or in focusing on the right issues.
Q Auditors Are Not “Jack-of-All-Trades”—It is very rare to find an auditor experienced in
all the various aspects of engineering that are germane to food processing, including
process, packaging, mechanical, electrical, and design or system engineering. Most
auditors are not experts in product formulation, food processing, sanitary design, mi-
crobiology and food safety, or other relevant disciplines. Many auditors come from a
quality-assurance background rather than a food safety background.
Q Time and Cost—Many audits are done in one day or less. This is definitely not enough
time to conduct a comprehensive food safety audit. Some companies look for the
cheapest alternative, which can lead to glaring mistakes.
Q Deceit on Part of the Facility Being Audited— It is extremely difficult for auditors to
address withheld information or data that is pertinent to the audit or to pick up on fab-
ricated or falsified data.
Q Too Much “Paper Review” and Not Enough Time Spent on the Plant Floor—Auditors
must strike a balance between documentation review (which is important) and time
spent in the plant observing infrastructure and practices (which is crucial).
Q Lack of Follow-Up—Often, there is no or little followup by auditors to ensure that defi-
ciencies have been addressed in a timely manner.
42
Q Lack of Audit Depth—Virtually all audits do not encompass on-site audits of suppliers
providing ingredients and materials to the facility being audited. The same is also true
for external testing laboratories that the facility being audited may be using for micro-
biological and other critical testing.
Q Over-Reliance on Audits by the Facility—Some companies believe that if they “pass” an
audit, everything must be in order, and they become complacent—until the next audit.
Q Announced vs. Unannounced Audits—Most third-party audits are announced, giving
the facility time to prepare for the audit. Some certification bodies and auditors pro-
vide advice to the facility on how to prepare for the upcoming audit. The audit is not
reflective of the true operating conditions in the facility.
Q Implicit Versus Explicit Conflict of Interest—Both accreditation bodies that set the au-
dit standard and the auditing firm that certifies the facility conforms to the auditing
standard go to great lengths to try to ensure impartiality and prevent explicit conflict-
of-interest issues with auditors. However, most accreditation bodies and auditing firms
are “for-profit” entities; therefore, issues with implicit conflict of interest may persist,
since the facility undergoing the audit pays for the audit. Some auditors may tend not
to be as tough in their evaluation as they otherwise might be for fear of losing the
business.
Q Companies Do Little or Nothing With Audit Results—Some companies do not react with
a sense of urgency in addressing audit results, particularly on those items that are char-
acterized as “minor deficiencies.” Some companies even address major deficiencies such
as roof leaks with only temporary or stop-gap measures and do not devote the proper
investment and resources to the correct long-term solution.
Some skeptics believe that third-party audits are not truly independent and, by nature, are
suspect and of minimal or no value. In fact, however, third-party audits do provide value,
but you must recognize they are not infallible. The value your company gets out of third-
party audits is directly proportional to your commitment to aggressively following through
on audit findings, including minor deficiencies. You must recognize that minor deficiencies,
if not addressed in a timely manner, at some point will become major issues. It is much bet-
ter to address them when they occur rather than let them become major problems for you.
There are a number of resources available to you to use in developing or strengthening
your third-party certification program. The U.S. FDA has issued a guidance document on
voluntary third-party certification programs for foods and feeds (67). This document de-
scribes the general attributes that FDA believes a certification program should include to
provide high-quality verification of food safety. This document outlines the attributes of a
strong third-party certification program. including:
Q Authority of the certification body to perform audit activities

- Authority to examine and gather records and other information
- Authority to collect and analyze samples
- Authority to assess and report on compliance with certification criteria
Q Qualification and training of auditors
43
Q Effective audit program elements

- Risk-based
- Written policies and procedures
- Verification that the establishment meets certification criteria
- Process for addressing establishment complaints about audits
- Documentation and recordkeeping
Q Quality assurance program for audits and
auditors
- Field evaluation of audits to verify audits are consistent
- Audit report evaluation
- Sample report evaluation
- Individual auditor performance
Q Compliance and corrective actions
- Apply a risk-based approach to determine when an investigation, followup and re-
audit are necessary
- Evaluate whether the establishment has executed proper corrective actions
- Withdraw certification if the establishment fails to take corrective actions
Q Industry relations
- The certification body (auditing company) should provide establishments seeking
certification with information about current FDA requirements and guidance
- It is preferable that the certification body be actively involved in regulatory, scien-
tific, industry and other external activities
Q Resources
- The third-party certification body should have sufficient resources to accomplish the
elements of the certification program
Q Self-assessment of the overall certification
program
- Assesses performance and identifies strengths and weaknesses
Q Laboratories
- The certification body should have access to the appropriate laboratory services
needed to support the audit
Q Ability and willingness to notify the FDA
- Product safety issues
- Certification withdrawal
- Changes to the certification program
Q Attention to conflict of interest
- The certification body and its auditors should be free of any conflicts of interest that
threaten impartiality
An in-depth evaluation of your PEM program and data should always be a part of an
overall independent third-party audit. This also includes an in-depth review of all docu-
mented corrective actions, procedures and other information discussed in this guidance
document. The PEM program is a major tool that is critical to demonstrating the effec-
44
P E R S O N N E L E D U C AT I O N A N D T R A I N I N G A N D M A N AG E M E N T C O M M I T M E N T
tiveness of your facility’s GMPs, hygienic practices and food safety plan. Therefore, it is
crucial it be part of every third-party food safety audit that is conducted.
A strong third-party auditing and certification program, if conducted properly, can be a

meaningful investment and tangible evidence of your company’s commitment to food
safety. In the end, the rewards you get from a strong program are directly related to the
time, diligence and commitment you put into the program.
45
Personnel Education and Training

and Management Commitment
I
t is incumbent upon the company to ensure that employees are properly trained in
effective GMPs, hygienic practices, food safety principles, and other practices and
procedures that enable them to do their job in an effective manner that will not
jeopardize the product or the consumer. Plant personnel can have a direct impact
on the safety of the foods produced in the facility. The risk of product recontamination
within the facility can be significantly reduced with effective training and monitoring of
ongoing employee practices.
The FDA’s GMP regulations stipulate that “Personnel responsible for identifying sanita-
tion failures or food contamination should have a background of education or experience,
or a combination thereof, to provide a level of competency necessary for production of
clean and safe food. Food handlers and supervisors should receive appropriate training in
proper food handling techniques and food-protection principles and should be informed
of the danger of poor personal hygiene and insanitary practices” (68). The FDA also
states, as part of its GMP modernization initiative, that ineffective training of employees
is a problem at the food manufacturer level relative to controlling microbiological haz-
ards in foods (69). In their analysis, the FDA states that it is not clear that current training
methods are sufficient and that the training many companies conduct may be too gener-
ic. They also believe that other impediments to effective training may include training the
wrong people not training enough people or not providing enough training. The FDA has
articulated the following beliefs about training (70):
Q Food production workers should have appropriate training in the principles of

food hygiene and food protection, and this training should include the impor-
tance of employee health and personal hygiene.
Q Training must be delivered in a form that is readily understandable to all person-

nel and delivered in a manner that is easily understood by the trainee.
Q Food processors must maintain a record of this training for each employee.
Q Certain core principles of food safety, equipment sanitation and regulatory com-
pliance must be included in the training of all food workers and supervisors.
46
The FDA’s focus on the importance of employee health and personal hygiene is support-
ed by the fact that many food-borne outbreaks and illnesses, including those caused by
Salmonella, have been traced to recontamination of the food directly by food handlers.
The Center for Disease Control (CDC) publishes an annual list of infectious and commu-
nicable diseases that are transmitted through handling foods (Table 4) (71). Stressing the
importance of complying with good personal hygiene and hygienic practices, including
proper hand washing, is critical in any employee GMP training program.
Table 4. CDCs List of Pathogens Transmitted by Food Contaminated by Infected Handlers

or From Food That is Cross-Contaminated During Processing or Preparation
Pathogens Occasionally Transmitted by Food

Pathogens Frequently Transmitted by Food Through
Through Infected Handlers or Through Cross-
Infected Handlers
Contamination During Processing/Preparation
Noroviruses Campylobacter jejuni
Hepatitis A virus Cryptosporidium spp.
Salmonella Typhi Entamoeba histolytica
Sapoviruses Enterohemorrhagic
Escherichia coli
Shigella spp. Enterotoxigenic Escherichia coli
Staphylococcus aureus Giardia intestinalis
Streptococcus pyogenes Nontyphoidal Salmonella
Taenia solium
Vibrio cholerae
Yersinia enterocolitica
Effective training is also a key part of a successful PEM program. Employees must under-
stand that effective monitoring of the environment is a critical measure of the success of
the company’s food safety commitment. Employees should never be discouraged from
aggressively trying to find the pathogen in the plant’s environment. If Salmonellae are
present in the plant environment, you want to find it. Only if you can find it can you con- Effective training
trol it and reduce the risk to the franchise and your consumers. The FDA also believes is also a key part
that third-party validation of test results can be useful to further establish confidence in of a successful
your environmental sampling results (69). It is highly recommended that your company PEM program.
Employees
have a qualified outside expert validate your PEM program.
should never be
discouraged from
You must also understand that education and training go hand-in-hand. Food safety edu- aggressively
cation focuses more on the why food safety is important, and food safety training focus- trying to find
es more on the how to do food safety (72). The reason food safety education and train- the pathogen
ing are stressed so much is that they focus on influencing the employee’s behavior. Re- in the plant’s
search has shown that if education and training materials can be personalized, then they environment.
are much more effective in influencing behavior than by just showing facts or statistics.
47
REFERENCES
Overall, the success of any PEM program is directly dependent on support and commit-
ment throughout the organization, starting with top management and cascading down.
If management does not provide sufficient resources, both in terms of capital and per-
sonnel to do an effective job, then it will ultimately fail. If management is more worried
about “meeting the numbers” than supporting a robust food safety culture throughout
the organization, then it will fail. In high-performing, successful companies, food safety
is not considered just a “program”; it is a pervasive part of the company’s culture. The
leaders in successful companies take a systems-based, behavior-based approach to cre-
ating a food safety culture throughout the organization (72, 73). Food safety should not
be viewed as a cost. Sure, there are costs associated with a commitment to food safety,
but senior management must realize that investing in food safety, including investing in a
solid PEM program, is a smart investment. It is an investment no different than in invest-
ing in sales and advertising, production and distribution, or new product development.
Research shows that companies that are dedicated to food safety and those that instill
a culture of food safety throughout the organization are big winners in the marketplace.
The company wins, the employees win, and the consumer wins.
48
REFERENCES
References
(1) US FDA, USDA, NACMCF. Hazard analysis and critical control point principles and application
guidelines. Adopted August, 1997. Available at: http://www.fda.gov/Food/FoodSafety/HazardAn-
alysisCriticalControlPointsHACCP/ucm114868.htm. Accessed: 15 July 2009.
(2) D’Aoust, JY. Chapter 9: Salmonella. In Doyle, MP, editor, Foodborne Bacterial Pathogens, New
York, NY: Marcel Dekker, 1989, p. 327 – 445.
(3) Uesugi, AR, MD Danyluk, and LH Harris. 2006. Survival of Salmonella Enteriditis phage type 30
on inoculated almonds stored at -20, 4, 23, and 35°C. J Food Protect 69: 1851 – 1857.
(4) Bale, MJ, PM Bennett, JE Beringer and M Hinton. 1993. The survival of bacteria exposed to des-
iccation on surfaces associated with farm buildings. J Appl Bacteriol 75: 519 – 528.
(5) Hiramatsu, R, M Matsumoto, K Sakae, and Y Miyazaki. 2005. Ability of shiga toxin-producing
Escherichia coli and Salmonella spp. to survive in a desiccation model system and in dairy foods.
Appl Environ Microbiol 71: 6657 – 6663.
(6) Burnett, SL, ER Gehm, WR Weissinger, and LR Beuchat. 2000. Survival of Salmonella in peanut
butter and peanut butter spread. J Appl Bacteriol 89: 472 – 477.
(7) Mattick, KL, F Jørgensen, P Wang, J Pound, MH Vandeven, LR Ward, JD Legan, HM Lappin-
Scott, and TJ Humphrey. 2001. Effect of challenge temperature and solute type on heat tolerance of
Salmonella serovars at low water activity. Appl Environ Microbiol 67: 4128 – 4136.
(8) Shachar, D and S Yaron. 2006. Heat tolerance of Salmonella enterica serovars Agona, Enteridi-
tis, and Typhimurium in peanut butter. J Food Protect 69: 2687 – 2691.
(9) US Centers for Disease Control and Prevention. April 10, 2009. Preliminary FoodNet data on
the incidence of infection with pathogens transmitted commonly through food – 10 states, 2008.
MMWR Weekly 58: 333 -337.
(10) US FDA. Bad Bug Book: Salmonella spp. Available at: http://www.fda.gov/Food/FoodSafety/
FoodborneIllness/FoodborneIllnessFoodbornepathogensNaturalToxins/BadBugBook/ucm069966.
htm. Accessed 29 July 2009.
(11) Uesugi, AR, MD Danyluk, RE Mandrell, and LJ Harris. 2007. Isolation of Salmonella Enteriditis
phage type 30 from a single almond orchard over a 5 – year period. J Food Protect 70: 1784 – 1789.
(12) Danyluk, MD, M Nozawa-Inoue, KR Hristova, KM Scow, B Lampinen, and LJ Harris. 2007. Sur-
vival and growth of Salmonella Enteriditis PT 30 in almond orchard soils. J Appl Microbiol 104: 1391
– 1399.
(13) Winfield, MD and EA Groisman. 2003. Role of nonhost environments in the lifestyles of Salmo-
nella and Escherichia coli. Appl Environ Microbiol 69: 3687 – 3694.
(14) Uesugi, AR and LJ Harris. 2006. Growth of Salmonella Enteriditis phage type 30 in almond hull
and shell slurries and survival in drying almond hulls. J Food Protect 69: 712 – 718.
(15) Stock, SL, CB Annett, CD Sibley, M McLaws, SL Checkley, N Singh, MG Surette and AP White.
2007. Persistence of Salmonella on egg conveyor belts is dependent on the belt type but not the
rdar morphotype. Poultry Sci 86: 2375 – 2383.
(16) Wilkoff, LJ, L Westbrook and GJ Dixon. 1969. Persistence of Salmonella typhimurium on fabrics.
Appl Microbiol 18: 256 – 261.
(17) Morita, T, H Kitazawa, T Iida and S Kamata. 2006. Prevention of Salmonella cross-contamina-
tion in an oilmeal manufacturing plant. J Appl Microbiol 101: 464 – 473.
(18) Kapperud, G, H Stenwig and J Lassen. 1998. Epidemiology of Salmonella typhimurium O: 4 – 12
49
REFERENCES
infection in Norway. Am J Epidemiol 147: 774 – 782.

(19) Shimi, A, M Keyhani and K Hedayati. 1979. Studies on salmonellosis in the house mouse, Mus
musculus. Laboratory Animals 13: 33 – 34.
(20) Kopanic Jr., RJ, BW Sheldon, and CG Wright. 1994. Cockroaches as vectors of Salmonella:
laboratory and field trials. J Food Protect 57: 125 – 132.
(21) Meerburg, BG and A Kijlstra. 2007. Role of rodents in transmission of Salmonella and Campylo-
bacter. J Sci Food Agricul 87: 2774 – 2781.
(22) Beuchat, LR and EK Heaton. 1975. Salmonella survival on pecans as influenced by processing
and storage conditions. Appl Microbiol 29: 795 – 801.
(23) Kirk, MD, CL Little, M Lem, M Fyfe, D Genobile, A Tan, J Threfall, A Paccagnella, D Lightfoot, H
Lyi, L McIntyre, L Ward, DJ Brown, S Surnam and IST Fisher. 2004. An outbreak due to peanuts in
their shell caused by Salmonella enterica serotypes Stanley and Newport – sharing molecular infor-
mation to solve international outbreaks. Epidemiol Infect 132: 571 – 577.
(24) Vural, A and ME Erkan. 2008. The research of microbiological quality in some edible nut kinds.
J Food Technol 6: 25 – 28.
(25) Little, CL, W Jemmott, S Surman-Lee, L Hucklesby and E de Penna. 2009. Assessment of the
microbiological safety of edible roasted nut kernels on retail sale in England, with a focus on Sal-
monella. J Food Protect 72: 853 – 855.
(26) Willis, C, CL Little, S Sagoo, E de Penna and J Threlfall. 2009 [In Press]. Assessment of the mi-
crobiological safety of edible dried seeds from retail premises in the United Kingdom with a focus
on Salmonella spp. Food Microbiol, doi 10.1016/j.fm.2009.05.007.
(27) Isaacs, S, J Aramini, B Ciebin, JA Farrar, R Ahmed, D Middleton, AU Chandran, LJ Harris, M
Howes, E Chan, AS Pichette, K Campbell, A Gupta, LY Lior, M Pearce, C Clark, F Rodgers, F Jamie-
son, I Brophy and A Ellis. 2005. An international outbreak of salmonellosis associated with raw al-
monds contaminated with a rare type of Salmonella Enteriditis. J Food Protect 68: 191 – 198.
(28) Eglezios, S, B Huang and E Stuttard. 2008. A survey of the bacteriological quality of pre-
roasted peanut, almond, cashew, hazelnut, and brazil nut kernels received into three Australian nut-
processing facilities over a period of 3 years. J Food Protect 71: 402 – 404.
(29) Little, CL, N Rawal, E de Pinna, J McLauchlin, and the Food, Water, and Environmental Surveil-
lance Network. 2009. LACORS/HPA co-ordinated food liaison group studies: assessment of the
microbiological safety of edible nut kernels on retail sale in the UK with a focus on Salmonella.
Available at: http://www.lacors.gov.uk/lacors/ContentDetails.aspx?id=22118. Accessed: 12 December
2009.
(30) US FDA. December 17, 2009. Recall – firm press release: Willamette shelling recalls shelled
hazelnuts because of possible health risk. Available at: http://www.fda.gov/Safety/Recalls/
ucm194806.htm. Accessed: 20 December 2009.
(31) US FDA. November 11, 2009. Safety: enforcement report for November 11, 2009. Raw macada-
mia nuts contaminated with Salmonella. Available at: http://www.fda.gov/Safety/Recalls/Enforce-
mentReports/ucm190285.htm. Accessed: 20 December 2009.
(32) Riyaz-Ul-Hassan, S, V Verma, A Malik and GN Qazi. 2003. Microbiological quality of walnut ker-
nels and apple juice concentrate. World J Microbiol Biotechnol 19: 845 – 850.
(33) Danyluk, MD, TM Jones, SJ Abd, F Schlitt-Dittrich, M Jacobs and LJ Harris. 2007. Prevalence
and amounts of Salmonella found on raw California almonds. J Food Protect 70: 820 – 827.
(34) Scheil, W, S Cameron, C Dalton, C Murray and D Wilson. 1998. A South Australian Salmonella
Mbandaka investigation using a database to select controls. Australian New Zealand J Pub Health
22: 536 – 539.
50
REFERENCES
(35) Scheil, W, C Dalton, S Cameron and C Murray. 1997. A multistate Salmonella Mbandaka out-
break associated with peanut butter: the South Australian experience. J Clin Epidemiol 50 (supple-
ment 1): S18.
(36) US Centers for Disease Control and Prevention. June 1, 2007. Multistate outbreak of Salmonella
serotype Tennessee infections associated with peanut butter – United States, 2006 – 2007. MMWR
Weekly 56(21): 521 – 524.
(37) US Centers for Disease Control and Prevention. January 29, 2009. Multistate outbreak of Sal-
monella infections associated with peanut butter and peanut butter – containing products – United
States, 2008 – 2009. MMWR Weekly 58 (Early Release): 1 – 6.
(38) US Centers for Disease Control and Prevention. June 11, 2004. Outbreak of Salmonella sero-
type Enteriditis infections associated with raw almonds – United States and Canada – 2003 – 2004.
MMWR Weekly 53(22): 484 – 487.
(39) 7 CFR Part 981. Almonds grown in California; outgoing quality control requirements. Final rule.
72(61): 15021 – 15036
(40) Müller, LL, M Hjertqvist, L Payne, H Pettersson, A Olsson, L Plym-Forshell and Y Andersson.
2007. Cluster of Salmonella Enteriditis in Sweden 2005 – 2006 – suspected source: almonds. Euro-
surveillance 12: 153 – 155.
(41) US FDA. 2009. Update on pistachio product recall. Available at: http://www.fda.gov/Safety/Re-
calls/MajorProductRecalls/Pistachio/Update/default.htm. Accessed: 04 August 2009.
(42) US Centers for Disease Control and Prevention. April 14, 2009. Salmonella in pistachio nuts,
2009. Available at: http://www.cdc.gov/print.do?url=http percent3A//www.cdc.gov/Salmonella/pis-
tachios/update.html. Accessed: 06 August 2009.
(43) Rocourt, J, G Moy, K Vierk and J Schlundt. 2003. The present state of foodborne disease in
OECD countries. WHO, Geneva, Switzerland. Available at: http://www.who.int/foodsafety/publica-
tions/foodborne_disease/en/OECD percent20Finalfor percent20WEB.pdf. Accessed: 06 August
2009.
(44) Powell, SC and RW Atwell. 1998. An evaluation of the collection and analysis of epidemiologi-
cal data for support of food safety control systems. J Food Protect 61: 1170 – 1174.
(45) Butcher, GD and RD Miles. May, 1995. Minimizing microbial contamination in feed mills produc-
ing poultry feed. University of Florida IFAS Extension Bulletin. Available at: http://edis.ifas.ufl.edu.
Accessed: 29 July 2009.
(46) Jargon, J and J Zhang. January 15, 2009. Peanut – butter probe focuses on Georgia plant. Wall
Street Journal. Available at: http://online.wsj.com/article/SB123194586477481479.html. Accessed: 01
August 2009.
(47) International Commission on Microbiological Specifications for Foods. 2002. Chapter 11: Sam-
pling to assess control of the environment. In ICMSF, Microorganisms in Foods 7: Microbiological
Testing in Food Safety Management. Kluwer Academic/Plenum, New York, NY.
(48) Grocery Manufacturers Association. February 4, 2009. Control of Salmonella in low – moisture
foods. Available at: http://www.gmaonline.org/science/SalmonellaControlGuidance.pdf. Accessed
01 August 2009.
(49) Grocery Manufacturers Association. February 4, 2009. Annex to control of Salmonella in low
– moisture foods. Available at: http://www.gma.online.org/science/SalmonellaGuidanceAnnex.pdf.
Accessed: 01 August 2009.
(50) Cramer, MW. 2006. Food plant sanitation: design, maintenance, and good manufacturing prac-
tices. CRC Taylor & Francis, Boca Raton, FL.
51
REFERENCES
(51) Almond Board of California. October 1, 2008. Guidelines of validation of propylene oxide pas-
teurization, v.3.0. Available at: http://www.almond-board.com/files/PPO percent20Pasteurization
percent20Validation percent20Guidelines percent20and percent20SOPs.pdf. Accessed: 01 August
2009.
(52) Almond Board of California. April 13, 2007. Guidelines for validation of blanching processes,
v1.0. Available at: http://www.almond-board.com/files/Blanch percent20Process percent20Valida-
tion percent20Guidelines.pdf. Accessed: 01 August 2009.
(53) Almond Board of California. October 23, 2007. Guidelines of validation of dry roasting pro-
cesses, v.1.2. Available at: http://www.almond-board.com/files/Dry percent20Roast percent20Vali-
dation percent20Guidelinesv1_2.pdf. Accessed: 01 August 2009.
(54) Almond Board of California. October 23, 2007. Guidelines for process validation using Entero-
coccus faecium NRRL B-2354, v1.2. Available at: http://www.deibellabs.com/pdf/Enterococcus_vali-
dation.pdf. Accessed: 01 August 2009.
(55) Almond Board of California. May, 2007. Almond action plan direct verifiable (DV) user informa-
tion. Available at: http://www.almond-board.com/files/DV percent20User percent20Info.pdf. Ac-
cessed: 01 August 2009.
(56) Fedix, E. 11 March 2009. Peanut butter recall could cost $1B. Associated Press Manufactur-
ing.net. Available at: http://www.manufacturing.net/News-Peanut-Butter-Recall-Could-Cost-
s1B-031109.aspx?menuid=252. Accessed: 07 August 2009.
(57) Evancho, GM, WH Sveum, LJ Moberg and JF Frank. 2001. Chapter 3: Microbiological monitor-
ing of the food processing environment. In: FP Downes and K Ito (Eds), Compendium of Methods
for the Microbiological Examination of Foods. American Public Health Association, Washington, DC.
(58) Jay, JM, MJ Loessner and DA Golden (Eds). 2005. Chapter 20: Indicators of food microbial
quality and safety. In: Modern Food Microbiology, 7th Ed. Springer, New York, NY.
(59) Totorello, ML. 2003. Indicator organisms for safety and quality – uses and methods for detec-
tion: minireview. J AOAC Internat 86: 1208 – 1217.
(60) Mossel, DAA, JEL Corry, CB Struijk and RM Rand. 1995. Chapter 8: Evaluation of the efficacy
of measures to ensure wholesomeness and quality of food by assessing compliance with reference
values (standards). In: Essentials of the Microbiology of Foods: A Textbook for Advanced Studies.
John Wiley & Sons, West Sussex, England.
(61) Kornacki, JL and JL Johnson. 2001. Chapter 8: Enterobacteriaceae, coliforms, and Escherichia
coli as quality and safety indicators. In: FP Downes and K Ito (Eds), Compendium of Methods for
the Microbiological Examination of Foods. American Public Health Association, Washington, DC.
(62) Morton, RD. 2001. Chapter 7: Aerobic plate count. In: FP Downes and K Ito (Eds), Compendium
of Methods for the Microbiological Examination of Foods. American Public Health Association,
Washington, D.C.
(63) US FDA. Bacteriological Analytical Manual. Available at: http://www.fda.gov/Food/ScienceRe-
search/LaboratoryMethods/BacteriologicalAnalyticalManual/BAM/default.htm. Accessed 12 De-
cember 2009.
(64) International Organization for Standardization. ISO 6579: 2002. Microbiology of food and
animal feeding stuffs – horizontal method for the detection of Salmonella spp. Available at: http://
www.iso.org/iso/iso_catalogue/catalogue_detail.htm?csnumber=29315.aoac.org/. Accessed: 12 De-
cember 2009.
(65) AOAC International. Official Methods of Analysis. 18th Edition. Available at: http://www.eoma.
aoac.org/. Accessed: 12 December 2009.
52
REFERENCES
(66) Moss, M and A Martin. March 6, 2009. Food problems elude private inspectors. The New York
Times. Available at: http://www.nytimes.com/2009/03/06/06food.html. Accessed: 12 December
2009.
(67) US FDA. January, 2009. Guidance for industry – voluntary third-party certification pro-
grams for foods and feeds. Available at: http://www.fda.gov/RegulatoryInformation/Guidances/
ucm125431.htm. Accessed: 12 December 2009.
(68) 21 CFR 110: Current good manufacturing practice in manufacturing, packing, or holding hu-
man food. Available at: http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.
cfm?CFRPart=110&showFR=1. Accessed: 12 December 2009.
(69) US FDA. August 9, 2004. Good manufacturing practices (GMPs) for the 21st century – food
processing. Available at: http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/
CurrentGoodManufacturingPracticesCGMPs/ucm110877.htm. Accessed: 12 December 2009.
(70) US FDA. November, 2005. Food cGMP modernization – a focus on food safety. Available at:
http://www.fda.gov/ohrms/dockets/dockets/04n0230/04n-0230-rpt0001-vol4.pdf. Accessed: 12
December 2009.
(71) CDC. November 23, 2009. Diseases transmitted through the food supply. Federal Register
74 (224). Available at: http://www.the federalregister.com/d.p/2009-11-23-E9-28093.
(72) Yiannas, F. 2009. Food Safety Culture: Creating a Behavior-Based Food Safety Management
System. Springer, New York, NY.
(73) Keener, L. December 2002/January 2003. The total plant food safety audit: rating your over-
all system. Food Safety Magazine.
53
Almond Board of California, 1150 Ninth Street, Suite 1500, Modesto, CA 95354 USA

Pem Book

Uploaded by

Copyright:

Available Formats

Pem Book

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pem Book

Uploaded by

Copyright:

Available Formats

pathogen environmental

monitoring program (PEM)

Presented by the Almond Board of California

An effective mechanism for controlling the risk of Salmonella in the

The genus Salmonella consists of rod-shaped,

Salmonella and human illness

Persistence of Salmonella in the almond growing and production environments

Research has shown that Salmonella

on various types of fabrics for weeks

Prevalence of Salmonella in almonds and other raw agricultural commodities

Cases of human salmonellosis associated with almonds and other nuts

Three documented salmonellosis outbreaks traced to the consumption of peanut butter

The issue of product recontamination

As discussed, there are many other food-borne illness out-

Data indicates that post-lethality recontamination is a major cause of food-borne illness

Salmonella Control Elements in the

A hazard analysis should be conducted by a cross-functional team to determine the SALMONELLA

2. Enhance the stringency of hygiene

The Primary Salmonella Control Area

3. Apply hygienic design principles to building and equipment design.

4. Prevent or minimize growth of Salmonella within the facility.

5. Establish a raw materials/ingredients

An approved ingredient supplier program should be

6. Validate control measures to inactivate

By government regulation, almonds must be processed to achieve a minimum 4-log in-

Q Hygienic design, construction and maintenance of the facility

Microbiological monitoring of the food-processing environment can be performed to

Q Verifying the effectiveness of cleaning and sanitation practices

The focus of this guidance document is on determining the presence of food-borne

Sampling locations: The PEM zoning concept

Q Conveyor belts and buckets

Zone 3—Nonproduct contact surfaces that are in open post-lethality product-processing

Q Floors, walls, ceilings Q Pallets

Q Hallways Q Coolers and freezers

PEM sampling in raw product areas

Types of testing for pathogen environmental

There are myriad methods that can be used for

Q Surface sampling using sponges or swabs

1. Be easily and rapidly detectable

Total Enterobacteriaceae cfu/g Salmonella positive in 50 g percentage

Figure 2. 3M PetrifilmTM Plate Method

Add sample Roll down film Distribute sample

Environmental sampling techniques

Sampling procedures and techniques should be conducted by properly trained person-

Procedures sampling equipment and environmental surfaces for salmonellae

When testing equipment and environ-

than 45°F upon receipt.

Conduct a hygiene zone assessment

Depending on the type of operation, an almond-processing or manufacturing facility can be

Brown Skin Facility With Huller/Sheller

brown skin box

hand sorting room incoming

office rest office

Primary Salmonella Basic GMP Area Nonprocess Area

Handling Facility With Treatment/Pasteurization

Custom Pasteurization Facility

Primary Salmonella Basic GMP Area Nonprocess Area

PEM sampling site selection and frequency of monitoring

Establishing your baseline: investigational sampling