HACCP Programs and Practices Evolve

“The final step, risk characterization, is the component that risk managers will use to inform their decisions, so it is critical that the preceding components—hazard identification, exposure analysis, and dose response analysis, if included—be completed with the highest accuracy and diligence,” he added.

Christy’s current HACCP plan has some elements of QRA, because he has always strived to be ahead of the curve when implementing food safety procedures. When he first got into the food business, Christy began to read up on the Code of Federal Regulations and started to develop his own internal food safety program. “When the HACCP regulations were implemented, we already had our framework in place, it was just random. We had the pieces to the puzzle; we just had to put it into a plan. Many small companies thought HACCP would be just a new road bump regulation like wearing a hairnet or gloves. They didn’t realize it would be a new global approach to handling food products,” said Christy.

“Even today, many new, smaller food companies don’t realize it’s not enough to have a great recipe. That’s the easy part of the business. Food safety is the critical part of the business,” added Christy, who often directs new companies to use the USDA-FSIS Web site (www.fsis.usda.gov). “Food safety can be a very confusing thing, but it doesn’t need to be. The FSIS Web site has excellent templates and is a valuable tool for setting up a HACCP framework.”

Predictive Microbiology and Modeling

Further supporting HACCP decision-making and risk evaluation, predictive microbiology/modeling is an emerging tool for protecting the integrity of the food supply. It uses mathematical models or equations that predict behaviors of foodborne microorganisms when exposed to different environments, taking into account temperature, water activity, pH, and so on. Models can be used to estimate microbial growth, inactivation, or survival in culture media or food products.

“Models for inactivation have been around for almost a century, dating back to the development of the ‘12 D bot cook’ used to assure the safety of canned foods. Models for growth and survival, on the other hand, have been available for about 20 years,” Dr. Schaffner said.

Commonly used modeling tools include the USDA Agricultural Research Service’s predictive modeling program and the predictive modeling information portal, the ComBase predictor developed by the Institute for Food Research, and the American Meat Institute’s integrated lethality spreadsheet.

“USDA-FSIS recognizes the benefits of using predictive modeling, but it recognizes there are some drawbacks: Models may not reflect all variables, may not include the effect of natural flora, usually do not account for the effects of heat shock and adaptation of microorganisms, and model variability may not be known or not correct,” said Dr. Schaffner. “While the FDA does not have a formal policy, it has an extensive history of model use for thermal processing of LACF [low acid canned foods]. Also, FDA has used and is using modeling to make other policy decisions.”

Recognized by many national governments and the Codex Alimentarius, the HACCP system has become a principal food safety program around the world, particularly during the past decade. HACCP’s preventative approach to food safety makes it especially appealing; companies need to implement all prudent strategies that protect reputations, public health, and their bottom lines.

New techniques such as predictive microbiology and QRA will support the further implementation of HACCP by improving HACCP decision making. Informal polling of webinar attendees supports this belief. Half of webinar attendees who responded to polls during the events said that companies in their industry were starting to use or already regularly use predictive modeling, and 66% of the respondents said they planned to learn more about predictive modeling. The majority indicated that microbial modeling and quantitative risk assessment were relevant to their industry, with 48% saying it was very important, 24% saying it was somewhat important, and 26% saying it was something to consider for the future.