Deibel Laboratories

Science you can trust from concept to consumer.

Deibel Laboratories Articles

Retest Analysis: When Original Sample Results and Retest Results Don’t Correlate

· ·

By Laurie Post, Ph.D.
When undesirable microorganisms are detected in a production lot, root cause analysis often includes testing additional samples from the affected lot or related lots of product. However, upon reanalysis, even when a larger sample size is used, the suspect organism cannot be detected in the retest analysis. Correlation differences between original and retest results can occur for several reasons:

  1. Distribution pattern of the microorganisms
  2. Lack of representative sampling
  3. Organism die-off
  4. Level of contamination and probabilities of detection

Distribution
The distribution pattern of microorganisms in a production lot affects the likelihood of detecting the organism in the product. For instance, distribution patterns can be Random or Non-Random.

Random Distribution: Microorganisms are distributed evenly throughout the entire lot. Microorganisms will likely be detected. This type of microbial distribution is not a factor of time since at any point in the sampling process there is an equal opportunity to detect the organism in question (Example A).


Non-random Distribution: Microorganisms are not distributed evenly throughout the entire lot. Microorganisms may not be detected by random sampling. This type of distribution is dependent on time since the microbial contamination is only represented at certain intervals in the production cycle without equal opportunity for detection (Examples B and C)

Example B shows a non-random distribution pattern of organisms that can be present at the startup of production. In this case, the food actually cleans the line, so that over time the contaminant is reduced to undetectable levels. If a retest sample of this lot of production is pulled from containers made later in the day, then the retest results will be negative for the target organism, even though the organism was present.

Example C is another example of non-random microbial distribution that can occur when condensate drips onto exposed product. Condensate began to drip sometime after startup and continues throughout the rest of the day, increasing toward shutdown. If retain samples taken at the beginning of the day were selected for retest analysis, the retest samples would likely be negative for the target organism even
though the organism was present in that lot of production.


Sampling
Microbiological analyses are only as good as the sampling procedure used to procure the sample. Every effort must be made to obtain a RANDOM SAMPLE. “Spot sampling” or taking a single sample will not yield a representative sample. FDA recommends taking 30 random 25g samples during the entire production run for Ready-to-Eat products that will be consumed without a process lethal to microbial pathogens. These include products such as cereals, confectionary products, cheese, dairy, spices – and many others (FDA Bacteriological Analytical Manual). For fluid and powdered products, most manufacturers employ mechanical or auto-samplers. For discrete products, samples are taken on a time production basis (i.e. one candy bar every half hour) and then the analytical unit (the actual amount tested) is made by compositing. Compositing must be performed carefully and thoughtfully, not only to ensure a representative sample, but also to avoid sample contamination.

Organism Die-off
Organisms contaminating a product may not survive at the same level for long periods of time. For example, Escherichia coli (E. coli) is not capable of long-term survival in an inhospitable matrix, such as chocolate, and will eventually die off. Levels may fall below the ability of an assay to detect the organism.


Levels of Contamination and Probability of Detection
The probability of finding a target organism in a retest sample is based on the probability of finding the target organism in the original sample. If 10 samples are collected for analysis and only one tests positive for a pathogen, the level of contamination is one in ten or 10%. Based on a standard statistical analysis, if there is a 10% chance of finding the target organism in the original sample, then the ability of finding the organism a second time by retest sampling is [10% X 10%] or [0.10 X 0.10 = 0.01] or one in a hundred. Similarly, if the level of contamination is 5%, the probability of finding the positive a second time is one in twenty-five and 1% is one in ten thousand. The short hand version of this is to simply square the level of
contamination, as shown here:

Level of Contamination- Probability of Finding the Target Organism in the Original Sample Probability of Finding the Target Organism in a
Retest Sample
One in two (50%)One in four (25%)
One in four (25%)One in sixteen (6.5%)
One in five (20%)One in twenty five (2.5%)
One in ten (10%)One in a hundred (1%)
One in a hundred (1%)One in ten thousand (0.01%)

There are many factors that can result in correlation differences between original and retest results. But, if a retest is desired, pull many random samples and request a retest as soon as possible.

When To Revalidate A Process Preventive Control

· ·

By Ryan Maus
FSMA legislation (21 CFR §117.160) requires that preventive controls be validated to assure that they are appropriate to control the hazards identified in the food safety plan. Generally, validation occurs:

  • Prior to, or within 90 days of implementation of the food safety plan
  • Whenever a change to a control measure(s) could impact its ability to control the hazard(s)
  • Whenever a reanalysis of the food safety plan reveals the need to do so

Concerning reanalysis of the food safety plan, according to FSMA legislation (21 CFR §117.170), this is required at least once every 3 years or:

  • Whenever new information about potential hazards associated with the food arise
  • Whenever appropriate after an unanticipated food safety problem
  • Whenever you find that a preventive control, combination of preventive controls, or the food safety plan as a whole is ineffective

So what is considered a significant product/process change? Examples from FSPCA’s PCQI training manual include:

  • Raw material changes, including a new supplier (e.g. change in viscosity, moisture levels, particle size; new pathogen hazard identified; increased microbial load)
  • Product or process changes (e.g. reduced water activity, changes in operating parameters, new equipment, equipment is moved)
  • Increasing production volumes that lead to extended run times, changes in bed depth, increased volumes of material in chambers
  • Adverse finding during review or observation of a recurring deviation (i.e. may suggest the validation is no longer adequate)
  • Emerging scientific information on hazards or control measures
  • Changes in consumers profiles and handling (e.g. marketing to children, infants, or an immunocompromised population)

Often, a food producer must consider changes to equipment, such as a heating element, and whether this would impact the equipment’s efficacy in controlling an identified hazard. The most conservative approach to demonstrate the preventive control’s efficacy would be to revalidate the process. However, another approach is to conduct a heat distribution study to verify that the heat distribution did not change significantly in a way that areas of the roaster see lower temperatures than when the roaster was initially validated. This could include collecting data that supports the initial validation, for example, temperature distribution data pre/post heating element replacement. If the analysis of this data indicates the same temperature profile, then you may be able to conclude that the reanalysis does not indicate the need for a full validation study. If the temperature profile has changed, then some additional study would be warranted.

Food Safety Considerations in the Production of Traditional Fermented Products: A Review Article

· ·

By Laurie Post, Ph.D.
A recent article in the Journal of Food Safety by authors at the Federation University in Australia reviewed the processes used in the production of Japanese rice koji and miso and developed a hazard assessment for these fermented foods.

Miso is a traditional fermented soybean paste made for thousands of years. The most common type of miso is made from rice koji, salt, and soybeans using a two stage fermentation process. The first step is the production of koji where rice is fermented with a mold such as Aspergillus oryzae at 30°C for 48 h. The koji and salt are then added to boiled soybeans and fermented by yeasts and bacteria for 2 to 24 months. Koji and miso are increasing in popularity in western countries where contemporary miso varieties are not pasteurized as consumer demand is for more natural products containing live microorganisms. While correctly prepared fermented foods are rarely associated with foodborne illness outbreaks, there have been several reports of illness. These were primarily due to pathogenic microorganisms introduced into the products from external sources such as raw materials or the processing environment. Hygiene and fermentation conditions must be carefully monitored to ensure food safety.


The authors point out that many of the production steps in the manufacture of koji and miso do not fit into contemporary food safety guidelines. Although a reduction in pH is a typical food safety hurdle for fermented foods, this does not apply to non-acidic foods such as koji or miso. Use of a certified safe mold as the starter culture and monitoring appropriate temperatures to ensure prolific growth of the mold (and to minimize growth of unwanted bacteria) are requirements for the safe production of rice koji. An optimized salt content and reduced water activity (Aw) is a food safety hurdle in the production of miso. Focusing on the quality of ingredients and hygienic practices along with the development of standardized
fermentation conditions, would also assist in safe production of these fermented foods. The authors conclude that research on unpasteurized koji and miso is needed to determine optimized temperatures, pH and Aw levels, and salt concentrations during the fermentation process that would assure these fermented foods comply with food safety requirements.

Recent Outbreaks Involving Natural Toxins

· ·

By Ryan Maus
Ingredient hazard assessments often include an assessment of natural toxins. Of these, mycotoxins such as aflatoxin are often first to be assessed. However, other naturally occurring toxins exist and have caused foodborne outbreaks in the past few years. These outbreaks involved toxin containing raw morel mushrooms, undercooked kidney beans, and new ingredients such as tara powder. While consumer
demands can inspire new food trends, it is important to remember the hazard that naturally occurring toxins present and the controls that can
be put in place to reduce the risk of illness.

A recent article describes a 2018 incident in France where an outbreak of acute gastroenteritis occurred at a military base after 200 people ate at a dining facility and became ill. An investigation indicated that chili con carne was the likely source of illnesses. Further analysis indicated that a plant lectin, phytohaemagglutinin, was present in the chili at a level above what is considered a toxic dose. The source of the plant lectin was undercooked kidney beans, a toxin that is destroyed with proper cooking of the beans.


In 2022, numerous customer complaints were made to the manufacturer, the FDA, and the CFSAN Adverse Event Reporting System about illnesses associated with the consumption of Daily Harvest’s French Lentil and Leek Crumbles. The product was eventually recalled and an outbreak investigation undertaken to determine the cause of illnesses. Initially, 393 adverse illness events were reported, including 133 hospitalizations. Illnesses reported involved the gastrointestinal tract, liver, bile ducts, and/or gallbladder. An FDA investigation identified tara protein flour, an ingredient in the product implicated, as an ingredient of interest although non-specific toxin tests, mycotoxin tests, and microbial tests did not identify any results of public health significance. An investigation of the adverse event and the potential role tara protein played in causing illness was published in 2023. The authors found that the original product and tara protein powder ingredient were free from intentional, accidental, and economic adulteration, spiking with synthetics, and toxic compounds. However, three nonprotein amino acids were present at high levels. One of these, baikiain, was present at the highest level and produced a minor metabolite in vivo similar to acetaminophen. When tested using a mouse model at levels relative to consumer consumption, the mice exhibited liver and kidney symptoms similar to an overdose of acetaminophen. The presence of baikiain was theorized to have caused the adverse events seen in consumers, the severity of which depended on the dose of tara consumed and the susceptibility of the individual to the toxin.

In 2023, an outbreak at a sushi restaurant in Montana caused 51 illnesses involving diarrhea, nausea, vomiting, or abdominal pain; and two deaths. An investigation indicated that morel mushrooms were the likely cause of illness. Left over mushrooms tested were found to be true morels and not false morels that contain the toxin gyromitrin, and no significant levels of pesticides, heavy metals, toxins, or pathogens were found. However, documented cases of similar illnesses associated with consumption of raw or undercooked true morel mushrooms have been reported due to low levels of naturally occurring heatlabile hydrazinic toxins in the raw morels. Morels consumed at the restaurant were served raw or lightly cooked.

Food producers need to ensure that proper controls are in place to eliminate toxic hazards, such as cooking. Likewise, new ingredients incorporated into products should be FDA approved and have a history of being safe. More information on natural toxins is available from FDA, WHO, and other sources.

Regulatory Watch Out!

· ·

Use of Unapproved Food Additives – FDA’s Public Inventory

Any substance used or intended for use in food must be authorized by the FDA for use as a food additive under the Federal Food, Drug, and Cosmetic (FD&C) Act, unless the use of that substance is generally recognized as safe (GRAS) by qualified experts or qualifies for an exemption.


As part of its on-going compliance activities, FDA conducts post-market activities to monitor the food supply for chemical contaminants. FDA identifies foods that contain a substance for which there is no authorization as a food additive and then reviews the regulatory status of this substance. FDA scientists analyze whether there is a basis to conclude that the intended use of the substance is GRAS or meets a
listed exception to the food additive definition. When FDA scientists determine that a substance is an unapproved food additive because it is not GRAS for its intended use (and does not meet a listed exception), they deem the additive to be unsafe and any food that contains the additive is adulterated.

On July 12, 2023, FDA released a public inventory of certain food ingredients that do not have GRAS status as determined by the agency and are therefore deemed unsafe. The inventory is a useful tool to assure the use of authorized ingredients.


The FDA Concludes Voluntary Pilot Program to Evaluate Alignment of Third-Party Food Safety Standards
with FSMA Rules


Food manufacturers may find that they are required to comply with one of the Global Food Safety Initiative standards in addition to the FDA’s Preventive Controls for Human Food or Produce Safety Rules. The FDA recently concluded a voluntary pilot program to evaluate alignment of private third-party food safety audit standards including BRC, FSSC22000, SQF and Global G.A.P. with applicable FDA regulations.


The pilot program was launched to help both FDA and industry gain a better understanding of whether these standards align with FDA regulations. Alignment between these standards may allow a company to structure their food safety plan to eliminate redundancy in their documentation and auditing programs.

Buyers and others in the food supply-chain often use third-party audits to assess the quality and safety of a product. Buyers, such as importers and receiving facilities, might stipulate an audit as part of a purchase agreement. In addition, three FSMA regulations – the Preventive Controls for Human Food rule, Preventive Controls for Animal Food (PC Animal Food) rule, and Foreign Supplier Verification Programs (FSVP) rule – allow for third-party audits to be used as supplier verification activities. Points of alignment would provide confidence that, in general, the third-party standards used to audit suppliers adequately address applicable FDA food safety requirements.


The results of the pilot program can be found in The FDA Concludes Voluntary Pilot Program to Evaluate Alignment of Third-Party Food Safety Standards with FSMA Rules. The FDA’s statements regarding alignment of the standards are referenced in the table below and apply only to the specified audit standards and addenda listed.


The reviews listed in the table focused on assessing third-party food safety standards and not the overall quality of the audit programs or qualifications of auditors. The FDA’s review and the findings from this pilot do not constitute an endorsement of any one food safety audit standard, or audits conducted under such standards. The FDA also adds a qualifying statement that third-party audits are not a substitute for FDA or state regulatory inspections for compliance with FDA regulations, including the Preventive Controls for Human Food Rule or the Produce Safety Rule.

Third-Party Food Safety Audit Standard and Applicable
Addendum		Scope Name
BRC Global Standard Food Safety plus the Global Standard Food Safety,
Issue 9, Interpretation Guideline		Preventive Controls for Human Food (PCHF)
FSSC 22000 Scheme 5.1 for Food Manufacturing plus the FSSC 22000, Version 3 FSMA PCHF Report Addendum
Preventive Controls for Human Food (PCHF
SQF Food Safety Code: Food Manufacturing, Edition 9 plus the SQF
Addendum for the Preventive Controls for Human Food		Preventive Controls for Human Food (PCHF)
GLOBALG.A.P. Integrated Farm Assurance – All Farm Base Crops Base – Fruit and Vegetables Checklist. Version 5.4-1- GFS plus the GLOBALG.A.P. Food Safety Modernization Act Produce Safety Rule Add-on Module Version 1.3Produce Safety Rule (PSR)
*Did not include Subpart E, related to Agricultural Water (as applicable to non-sprout produce) or Subpart M, related to sprouts

READY TO WORK TOGETHER?

Request a quote from one of our sales professionals. Contact Us