1

1

International Commission on

MicrobiologicalSpecifications for Foods (ICMSF)

Useful Testing in Food Safety Management

KATHERINE M.J. SWANSON

Ecolab Eagan, MN, USA

Punta del Este, UruguayMonday

5 October 20092

Discussion Topics

�ICMSF Microorganisms in Foods 8 – overview

�Different tests serve different purposes

�Testing for maximum value

3

Microorganisms in Foods 8: Use of Data for Assessing Process Control and Product Acceptance

�Update previously recommended end-product testing criteria

�Add other useful tests for specific product types

� Include microbial food safety and quality

MICRO-

ORGANISMSIN FOODS

8USE OF DATA FOR

ASSESSING PROCESS CONTROL AND

PRODUCT ACCEPTANCE

ICMSF

4

ICMSF Provides AdviceNo official status

�ICMSF recommendations have no official status

�Official recommendations and standards are the province of:�Governments for national standards and regulations, e.g.

�Intergovernmental agencies for international standards, e.g.

5

Book 8 Contents

�Part 1-Principles� Utility of microbial testing for safety & quality

� Validation of control measures� Verification of process control� Verification of environmental control� Corrective action to re-establish control

� Microbial testing in customer-supplier relationships

�Part 2 – Product Categories� Meats � Poultry� Seafood� Feed & pet food� Vegetables� Fruits� Spices, dried soups, flavorings� Cereals� Nuts, oilseeds, dried legumes� Cocoa and confectionery� Oil based foods� Sugar, syrups, honey� Beverages� Water� Dairy products� Eggs� Shelf stable, heat treated foods� Infants and young children� Formulated foods

Some “work in progress”

examples shared today.

6

Different Tests Serve Different Purposes

2

7

Microbial Testing

� “Microbial testing” means

different things to different people� Reams of data

� Detective game to identify unknown

or causative agent

� Presence/absence or qualitative

reaction that’s observed

� Quantitative measurement of the

microbiological status of a sample or

lot

OR

8

Microbiological Testing Can Assess:

�The safety of food

�Validation and verification procedures in HACCP

�Adherence to GMP/GHP

�The suitability of a food or ingredient for a particular purpose

�The keeping quality (shelf-life) of certain perishable foods

9

When & Where to Test

�When there is good evidence that:

�There is a microbiological problem

o Food safety or quality

o Historical or current

AND

�Testing will help to control the problem

10

The purpose of a test determines:

Lot rejection, process adjustment, recall, outbreak investigation, etc.

The action

Investigational sampling, routine sampling, regulatory sampling, etc.The interpretation

Routine (daily, weekly, monthly, quarterly, etc.) or event triggered

The frequency

Environment, line residue, end product, location collected, size/ number of samples

The sample

Time to results, accuracy, repeatability, etc.The method

Indicator or pathogenThe target

11

Target Organism Examples

Life threatening, chronic sequelae, or long duration OR

Designed for sensitive sub-population

Incapacitating, usually not life threatening

Not life threatening, short duration, self limiting, no sequelae

Measure of GMP

Spoilage, reduced shelf life, no health concern

E. coli O157:H7, C botulinumtoxin or E. sakazakii (infants)

Salmonellae, Yersinia enterocolitica, Shigella flexneri, etc.

S. aureus, B. cereus, C. perfringens, Norovirus, etc.

Coliform, Enterobacteriaceae, etc.

Total counts, yeast, mold, etc.

Severe hazard

Serious hazard

Moderate hazard

Indicator

Utility

ICMSF Hazard Categories

From ICMSF Book 7 12

Key ICMSF Sampling Plan Terms

Level above which is unsatisfactory or requires further investigation

M

Level that separates acceptable quality from marginally acceptable or unacceptable quality

m

Maximum number of sample units with unsatisfactory test results

c

Number of sample units analyzedn

3

13

ICMSF Suggested Sampling Plans for Lot Acceptance Testing

Likely Change Before Consumption

Case 15

n=60, c=0

Case 14

n=30, c=0

Case 13

n=15, c=0

Severe

Case 12

n=20, c=0

Case 11

n=10, c=0

Case 10

n=5, c=0

Serious

Case 9

n=10, c=1

Case 8

n=5, c=1

Case 7

n=5, c=2

Moderate

Case 6

n=5, c=1

Case 5

n=5, c=2

Case 4

n=5, c=3

Indicator

Case 3

n=5, c=1

Case 2

n=5, c=2

Case 1

n=5, c=3

Utility

Increase No ChangeReduceHazard Group

An

aly

tica

l u

nit

= 2

5g

14

Sample Size Influence on Probability of Acceptance

0

0.2

0.4

0.6

0.8

1

1.2

0 5 10 15 20

% Defective

Pro

ba

bil

ity

of

Ac

ce

pta

nc

e

n = 15

n = 30

n = 60

m = 086%

74%

55%

1%

15

Testing to Maximize Value

16

Useful Microbial Testing

�Identification of contamination sources

�Environmental monitoring to identify potential pathogen harborage sites

�Utility and indicator organisms to verify effective controls & trends�Effective processing�Effective control of post process contamination

�Investigation sampling for problem solving

17

Process Example

Process 1

Process 2

Process 3

Packaging Line A

Packaging Line B

Ingredients

18

Result Format Influences Information Provided

Quantative

0

1

2

3

4

5

0 10 20

Lot Number

Lo

g (

CF

U/g

)

Presence/Absence

0 10 20

Lot Number

Positive

Negative

4

19

Trend Analysis Can Inform Process Control

0

1

2

3

4

5

0 10 20

Lot Number

Lo

g (

CF

U/g

)

0

1

2

3

4

5

0 5 10 15 20

Lot Number

Lo

g (

CF

U/g

)

0

1

2

3

4

5

0 10 20

Lot Number

Lo

g (

CF

U/g

)

0

1

2

3

4

5

0 10 20

Lot Number

Lo

g (

CF

U/g

)

20

ICMSF Book 8 Testing Considered

�Primary production

�Ingredients

�In-process

�Processing environment

�Shelf life

�End product

21

Primary production

Included for some product types

�Included when production conditions have a major influence on the microbial quality or safety�Fruits, vegetables, spices, meat, poultry, and fish products

�Examples of samples to consider� Irrigation water, fertilizer, feed, and other on-farm practices

22

Ingredient Testing

�May be useful for some applications and not others, e.g. cocoa powder:�Used in chocolate, no heat treatment?Used in ice cream mix that is subsequently pasteurized

�Questions considered� Is control at the ingredient step necessary for safety or quality?

� Is testing necessary to verify the acceptability of the ingredient?

23

In-Process Testing

�Use to verify a kill step or predict potential re-

contamination

�Examples� Intermediate product, line residues, tailings, wash water� Typically indicators with quantitative results

�Questions considered:� Does the process need to be controlled to prevent increase, ensure decrease, maintain current level, or prevent spread of a microbial concern?

� Is testing needed to verify a) the process is functioning as intended or b) contamination is not occurring in the process?

� Are there locations in the process where accumulated product residue may provide a representative or “worst case” sample that predicts the safety or quality of the final product?

24

Processing Environment Testing

�Use to verify that the environment is under appropriate hygienic control

�Examples� Swabs or sponges for sampling sites on equipment or in the environment

� Rapid testing to verify cleaning & sanitation adequacy

�Benefits –� Identify harborage sites that can contaminate end product� Frequently, earlier detection of issues than end product testing

�Questions considered:� Does the environment need to be controlled to prevent contamination of the product with a microbial concern?

� Will testing be beneficial to verify control of the microbial concern in the environment?

5

25

Shelf Life Testing

�Discussed only if microbial activity is relevant to the commodity

�Purpose – verify adequacy of microbial stability for the product life cycle

�Benefits – may predict issue before they are experience in the market place

�Questions considered: � Is shelf life limited by a microbiological safety or quality concern?� Is shelf-life testing feasible?

26

End Product Testing

�Purpose: � demonstrate successful application of controls or � assess the microbiological status of a lot when no other information exists to assess its status.

�Relative importance of end product testing is lower than that for in-process or environmental testing for many product categories.

�Alternative sampling plans may be appropriate, for example:� Reducing the number of samples for on-going surveillance activity� Increasing the number of samples when investigating significant process deviations or outbreaks.

�Questions considered:� Is end product testing necessary to verify the overall manufacturing process?� Is end product testing relied upon for ensuring the safety or quality of the lot?

27

Preliminary Example: Dried Cereal Products

End product

Not relevantShelf-life

Test for Salmonella and Enterobacteriaceae in the processing

environment

Typical guidance levels:

Enterobacteriaceae – 100-1000 cfu/g

Salmonella – absent

HighProcessing

environment

Test appropriate product residues and in-line samples for

Salmonella.

Typical guidance levels:

Salmonella – absent

HighIn-process

Test for mycotoxins if confidence in raw grains is low

Test nuts, cocoa, and other sensitive ingredients with no

subsequent kill step for Salmonella if confidence in supplier is

low.

HighCritical

ingredients

Useful testing

Relative

importance

28

Preliminary Example: Dried Cereal Products (continued)

NA001011ISO 6579 SalmonellaDried

cereals

Low

MmcnCase

Sampling plan & limits/25gAnalytical

method

MicroorganismProduct

Testing for pathogens is not recommended during normal operation

when GHP and HACCP are effective as confirmed by above tests.

When above testing or process deviations indicate a possible safety

issue, test for Salmonella.

10210252ISO 21528 -

1 / IOCCC

Enterobacter-

iaceae

Dried

cereal

MmcnCase

Sampling plan & limits/gAnalytical

methodMicroorganismProduct

Testing for Enterobacteriaceae is recommended to verify process control.HighEnd

product

Useful testing Relative

importance

29

Microbial Sampling Summary

hTesting safety “into” products usually does not work because of sampling probability

hTesting is recommended to generate meaningful data�Impact quality or safety

hFocus on process control preferred�Environmental monitoring�Selected sampling tailored to the line to verify control

30

The ICMSF recommendations in this presentation are not final.

The work is still under development and subject to internal and external peer review.

30

International Commission on Microbiological

Specifications for Foods (ICMSF)