|
USE OF MICROBIAL PATHOGEN COMPUTER
MODELING IN
HACCP PLANS
- What is the purpose of this Notice?
This provides inspection program personnel with
information about Microbial Pathogen Computer Modeling (MPCM)
programs and appropriate verification activities when an
establishment uses MPCM programs in validating and
maintaining its HACCP plans. FSIS has also developed
guidance material about MPCM programs for establishments
and inspection program personnel (Attachment). This
information is available on
www.usda.fsis.gov.
NOTE: This notice is a reissuance of FSIS
Notice 50-04.
- What is an MPCM program?
An MPCM program is computer-based software that, based
on such factors as growth, lethality, and survival in
culture broth and food products, estimates the growth or
decline of foodborne microbes in food samples in
production.
- How can the MPCM programs be used?
MPCM programs can be valuable tools for establishments to
use in supporting hazard analyses, developing critical
limits, and evaluating the relative severity of problems
caused by process deviations. They can also be used to help
predict the expected effectiveness of corrective actions.
- What are the limitations of MPCM programs?
It is not possible or appropriate to rely solely upon a
predictive modeling program to determine the safety of
foods and processing systems. Determining pathogen growth
or survival and controlling it in food products requires
complete and thorough analysis by an independent
microbiology laboratory, challenge studies, and surveys of
the literature. The limitations are discussed further in
the attached guidance material. MPCM programs do not
replace these types of activities or the
judgment of a trained and experienced microbiologist.
- How should inspection program personnel verify
the use of MPCM programs?
- Establishments are responsible for validating their HACCP plans and must justify the use of the conclusions reached by the use of MPCM programs. Inspection program personnel should verify that establishments document the use of MPCM programs as specified in 9 CFR 417.5. Generally, an MPCM program would not be the only documentation relied upon to support an element of a HACCP plan. However, in certain circumstances, a microbiologist or other trained process authority professional may determine the MPCM program is the most appropriate source of data to support HACCP decision making. For example, the control of Clostridium botulinum in low acid canning technology has long been established and documented in scientific and other technical reference literature. Provided that the control parameters for C. botulinum are incorporated into an MPCM program and accurately reflect the process under review, then the MPCM program may be relied upon as the sole source for decision making for a HACCP element. In such cases, the microbiologist or other trained professional on the HACCP Team must document their decision to use the MPCM as part of the HACCP records.
- Inspection program personnel should verify that the parameters used in the predictive model match the ones used by the establishment in its process, and that the data produced by the MPCM program were taken into account by the establishment in its decision making process during the HACCP plan development or implementation.
(NOTE: Inspection program personnel should not use or place on Agency computers an establishment’s MPCM program. In the future, inspection program personnel may have access to an Agency issued MPCM program.)
-
If inspection program personnel have questions regarding an
establishment’s use of an MPCM program, they should
contact the Technical Service Center. If necessary, a
Enforcement Investigation Analyst Officer may respond to
the concerns about the establishment’s use of the MPCM
programs.
Philip S. Derfler /s/
Assistant Administrator
Office of Policy, Program, and Employee Development
| DISTRIBUTION:
Inspection Offices; T/A
Inspectors; Plant Mgt; TRA; ABB; TSC; Import Offices
|
NOTICE
EXPIRES: 6/1/06
|
OPI:
OPPED |
Attachment 1
GUIDELINES FOR THE USE OF MICROBIAL PATHOGEN COMPUTER
MODELING IN HAZARD ANALYSIS AND HACCP PLANS
Introduction
Microbial Pathogen Computer Modeling (MPCM) programs
are innovative tools based on predictive microbiology.
These modeling programs are used for predicting
bacterial growth, decline, or inactivation, and
survival in culture broth or food substrates. Questions
from program and industry personnel have indicated
confusion as to whether MPCM programs can be used as
part of the validation of a HACCP plan under 9 CFR
417.2 (a) and 417.4 (a) (1).
What Is an MPCM Program?
MPCM programs are computer-based software programs to
estimate microbial growth, lethality, or survival in
culture broth or food products. These programs can also
approximate the growth or decline of foodborne microbes
under specific conditions of an experimental protocol.
Predictions are based on a bacterial species and its
response in broth culture or actual food substrates, to
environmental agents, including such intrinsic product
characterizations as pH, salt, phosphates, nitrites,
and water activity, and such extrinsic influences to
the product as temperature and culture atmosphere (such
as aerobic or anaerobic). The programs can quantify the
effects of two or more such agents on the pathogen and,
in some cases, allow extrapolations to be made outside
the model’s data range.
The Role of MPCM Programs
MPCM programs can be valuable tools for establishments
to use in supporting Hazard Analyses, estimating
Critical Control Point (CCP) limits and ranges, and
evaluating the relative severity of problems caused by
process deviations. They are initial estimators of
microbial behavior and may be useful as a tool for
evaluating potential problems.
A food manufacturing company may use an MPCM program to
estimate the influence of each limiting agent or
combination of agents during processing. MPCM programs
should not be used by themselves though, in lieu of
necessary laboratory work and literature searches. They
do not necessarily reduce the need for challenge tests,
storage trials, and other conventional techniques to
assess pathogen growth, decline, or survival. MPCM
programs can be used as a measure in evaluating the
significance of a specific process deviation.
These programs are useful tools for:
- Predicting bacterial growth potential or decline in a particular food after specific time-temperatures are reached under specified conditions.
- Identifying potential CCPs where the model indicates that at a certain level controllable factors will either permit or suppress microbial growth. Allowable limits can be partially based on quantitative predictions of bacterial behavior at different levels under varying conditions.
- Reformulating product based, at least in part, on conditions that influence microbial growth.
-
Determining product disposition as a consequence of a
process deviation; providing information that can be used
in decision making; modeling information along with other
considerations that may result in a decision to rework,
utilize, or condemn product.
- Providing graphical modeling tools that can be used as instructional aids for demonstrating to employees the impact of following HACCP requirements, maintaining and monitoring equipment, or the benefits of high-quality raw material with initial low microbial populations versus those raw materials with higher populations.
The Limitations of MPCM Programs
It is not yet possible to rely solely upon any predictive
modeling program to determine the safety of all foods and
processing systems. Determining pathogen growth or survival
in food products requires a complete and thorough validation
from an independent microbial laboratory, challenge studies,
and a survey of the literature. MPCM programs do not replace
microbial validation, experimental challenge studies, or the
judgment of a trained and experienced microbiologist.
Current MPCM models do not usually include all influences on
bacterial growth and survival. Such excluded influences
include bacteriocins or anion effects from acidulents,
phosphates, sorbates, and humectants other than sodium
chloride. Programs do not typically consider the protective
buffering effects of various food components when converting
predictions from experimental broth cultures to different
and more solid food matrices. The programs do not include
the inhibitory or supporting effect that the native
bacterial population may have on pathogenic bacteria. These
effects will vary in different seasons, locales, climates,
and other unforeseen events in an establishment. These
models do not usually account for increased resistance of
bacteria to certain treatments because of prior conditioning
of cells, e.g., from heat shock or induced acid tolerance.
The programs are based on a few specific pathogens and the
behavior of these pathogens under controlled laboratory
conditions. Many of the current models do not consider
enrichment growth or competitions among different organisms
typically present in foods. Most models do not consider what
phase of growth or the physiological state that the modeled
bacteria were in when exposed to each of the limiting
environmental factors. Because the initial predictive phase
of growth of the bacterial cell can be different from the
actual conditions, additional laboratory testing may still
be needed.
Risk assessments developed in establishments cannot be based
on MPCM programs alone. In order to estimate risk, a useful
model should be based on complete exposure assessments,
dose-response assessments, and risk
characterizations to predict the overall risk with attendant
uncertainties. Extension of the predictive models to account
for the full structure of a risk assessment would require
more investment for establishments to develop and use.
Finally, from a statistical point of view, the nature and
magnitude of variability and uncertainty associated with
predictive models is not completely understood. In some
models, the upper and lower confidence intervals may be
inaccurate or even missing. In summary, there is no
guarantee that predicted values will match those that would
occur in any specific food system unless the model was
developed for that specific food.
MPCM Programs Used as Part of a Hazard Analysis, the
Validation of the HACCP Plan, or a Corrective Action
MPCM programs are support tools that can be used in hazard
analyses, identifying critical limits, and validating the
HACCP plan. Plant management must realize that MPCM programs
are only predictive tools for estimating potential growth,
lethality, or survival potential for pathogens. As such,
predicted values will likely not correspond to those that
would occur in any specific food system. These programs can
suggest the level of a hazard likely to occur. MPCM programs
also help determine whether controllable factors can reduce
or eliminate that hazard or correct the process deviation.
As a consequence of the uncertainty of the predictive
capabilities when applied to a particular set of
circumstances, establishments should be extremely careful
about how they use these programs, if they choose to use
them at all. They should be used in a conservative manner
and other factors should play a role in making critical
decisions about a process or deviation.
MPCM programs, as stated above, cannot be relied upon as the
sole means of ensuring food safety and producing effective
process systems for all products. Before the models could be
used in such a manner, the user would have to validate the
models for each specific food of interest. Initial
validation, as defined in 9 CFR 417.4, requires repeated
testing of the CCPs and critical limits. Ultimately,
predictive models can serve only as a tool for approximating
critical limits but not as a replacement for testing for
these limits.
MPCM programs can be used as predictive models to ascertain
the effects of process deviations or as an analysis tool to
assist in determining the relative severity of a deviation.
If all factors leading to the deviation can be determined,
the MPCM program could be used to show the effects from the
deviation and to develop a working plan for corrective
action. All other requirements in analyzing the deviation
and implementing the corrective action as stated in 9 CFR
417.3 must then be followed.
MPCM Programs and FSIS
FSIS does not approve nor advise industry on the proper use
of specific MPCM programs. Industry may wish to contact a
process authority or other professional about the use of
specific MPCM programs. For particular case use, contact the
Technical Service Center.
|
