Ways To Improve Food Safety

By Amanda Earing, News Editor, Manufacturing.net After peanut products were linked to a national salmonella outbreak that produced the largest recall in history, many are wondering could this have been prevented in some way. Is more testing the answer?

Food safety is again in the national spotlight. After peanut products were linked to a national salmonella outbreak that sickened hundreds, left nine dead, produced the largest recall in history, sparked a congressional hearing and spurred a criminal investigation, many are wondering could this have been prevented in some way. Is more testing the answer?

Witnesses from the FDA at recent Senate hearings said that the key to stop having these types of recalls is prevention. They mention using the same methods that the USDA and U.S. Seafood Inspection Program use, a process called HACCP. Steve Wilson, American Society for Quality's food safety expert, sheds light on how the HACCP process and other preventive measures can help improve food safety.

Mnet: What is the HACCP process?

Wilson: HACCP stands for Hazard Analysis Critical Control Point. It is an analysis that determines what your potential food safety hazards are, what could possibly go wrong and where in the process it could happen. You then identify critical control points to limit those hazards in the process. You also would set a critical limit that allows you and others to know when to take action -- whether it means rejecting a product or reprocessing. HACCP is designed to prevent hazards from occurring in the final product.

For instance, this level of process control helps identify what steps can be used to kill a pathogen and where in the process those steps are. You want to be sure that you have records in place that document those steps because that’s a critical control point. Line workers at those locations know what their job is and follow that plan. Records reviewed by internal quality control and regulators when they come into the facility will show that you followed the plan throughout the process. In the meat, poultry, seafood and juice industry, HACCP is already widely used. It’s not a new concept, but is recently being applied to other areas.

Mnet: Besides HACCP, are there other quality control procedures that should be in place?

Wilson: There are a number of quality principals that can be used to enhance the safety of foods or the systems used to produce foods. For example, ISO has put out a standard called ISO 22000 for food safety management. It incorporates HACCP, but also includes management principals and their role in food safety. HAACP requires a level of assurance for food safety, which includes verification by management. ISO 22000 states in very clear language the roles and responsibilities of management that help during internal audits to enhance the HACCP plan in place.

While the standard isn’t required, I personally recommend it because it’s a very good standard to work with. Much of the standard is common sense and is universally accepted and developed by the international community.

Mnet: What should food manufacturers be doing in their facilities to keep U.S. food supplies safer?

Wilson: Manufacturers need to be diligent to their operating systems. Most of these systems by now have had some level of approval or review by the regulatory authorities. Mistakes happen when there is a lack of attention to detail on following the plan that has been agreed to. Sanitation and cleaning the plant on a daily basis is where most of the problems occur. Very few, if any problems, are intentional -- it’s just a question of large plants trying to keep business going. But they need to keep it as clean as possible and have the diligence and oversight within their own operations. The ultimate responsibility of the law is on the processors.

Mnet: How can manufacturers ensure the integrity of their supply chain?

Wilson: The supply chain is one of the biggest issues in food safety management. Most HACCP regulations indicate control one step up and one step down from the chain. This includes communications and a number of issues along those lines. However, regulators are considering whether to go further up and down the chain. For example, if you’re a seafood processor, should you be responsible for the integrity of the product that was on the shipping vessel and throughout the chain before it arrives at your door? It’s a difficult concept and relies heavily on good traceability which is not an easy task. As an example, if a raw commodity like seafood comes into the U.S. at the port, each lot is a certain size. From there, it gets broken up into several lots and as it’s transported across the country it gets broken up even further. Tracing something like that is not an easy proposition.

Mnet: What can manufacturers do to help aid recalls?

Wilson: There are already guidelines and regulations regarding recalls. Manufacturers need to know where the product is to a certain point. The problem again is the actual traceability and when you find problems. Some products are being produced and continuously going out the door and by the time you find the problem, it could be anywhere at that point. The idea of recalls is to get as much back as you possibly can. One of the key things manufacturers can do is a mock recall each year with their shippers. See how far they can go back and how far they can trace their products without actually calling it back. The mock recall really tests -- if you do it properly -- the integrity of your system and tells you where your weaknesses are and what needs strengthened.

Mnet: It took months to find the source of the salmonella outbreak. Why do these investigations to trace contaminated products back to their point of origin take so long?

Wilson: One of the problems with investigating contaminated products is first you have to find out where the contamination occurred. This is integral. If the contamination occurred in the facility, you need to only go somewhere in the facility to find the contamination. But if the contamination came into the facility, possibly through the growers, that significantly complicates the traceability of the product. If it came from the growers, it could have entered dozens of facilities and could have gone further and further out, resulting in a massive problem. Or, there might not have been anything wrong with the lot originally. It could have been during transportation from the grower to the manufacturer. This is why records are so important. There are actually a limited number of manufacturers, but lots of suppliers and a lot of buyers. So the manufacturers’ records are a good starting point to any investigation.

Mnet: Is more testing and inspections the answer to preventing food safety issues?

Wilson: No. Food is not a piece of steel. When you test a specific lot in a raw agricultural commodity such as fish or peanuts, it will only give results for what you tested. It isn’t a good indicator of the entire lot. With peanuts, there are a number of issues that can go wrong when testing. If you’re a manufacturer of peanut butter, you can test a batch of 12 jars and all that does is give you a general indicator of what might be possible. All you’re really able to tell is whether those 12 jars tested positive for salmonella, but it doesn’t tell you if the whole lot has salmonella. So, imagine you take a sample out of the 12 and you have a zero tolerance for salmonella. You test those jars and only one jar is positive for salmonella. That is difficult for you because you have zero tolerance, but only one jar tests for salmonella. It doesn’t indicate the whole lot is the problem. At the same time, if all 12 jars test for salmonella, you know you have a big problem, but you still don’t know how big that problem is. All it does is give you an indicator that some level of action is needed, but it doesn’t tell you the susceptibility of the entire lot.

Mnet: How important is it to test again if you receive a positive sample?

Wilson: Anytime you take statistically valid samples and you sample again, you increase the possibility of accepting a rejectable lot. In other words it is easier to accept a lot that is a problem based on a statistical basis. So you don’t want to do that. But what you want to do is investigate why there was a positive in the first place. It is possible -- entirely possible -- the positive was analytical error. In general terms, when you find a positive, it indicates an action is required on your part and that action is rarely, if ever, to let the lot go.

Steve Wilson is a member on the American Society for Quality’s board of directors and ASQ's Food Safety Expert. He is chief quality officer for the U.S. Department of Commerce’s ‘Grade A’ seafood inspection program and has worked with the U.S. government on implementing food safety procedures, primarily HACCP and ISO 22000. He was a member on the U.S. Technical Advisory Group to create the International standard on food safety.

More in Home