In the past 25 years, there has been a great evolution in the food and feed manufacturing industries, from a reliance on finished product testing, to intensive environmental monitoring programs (EMPs). These programs can take many forms, ranging from very generic indicator monitoring (testing for ATP or aerobic plate counts), to specific pathogen EMPs, with the most common being Listeria spp. and Salmonella—organisms that are uniquely qualified to take up residence in food manufacturing plants. While these programs can look very different, they all generate data that can be used to improve food safety protocols.
Most EMPs that look for broad indicator groups are quantitative. With ATP testing, most detection platforms provide quantitative data in the form of relative light units. If using aerobic plate count data, this too is quantitative, yielding counts typically reported as colony forming units per sample. Other quantitative indicator groups used in EMPs include yeast and mold, coliforms, and Enterobacterieacea. Over time, statistical process controls can be used to determine when counts are trending upward or when there is a sharp spike in counts from the established baseline data, indicating a potential problem.
In contrast, most pathogen EMPs are based on qualitative testing. In other words, results are reported as either positive or negative. One of the limitations of this type of testing is that it is unknown whether there was a single cell present on the sample, or if there were thousands, as both can result in a positive result.
EMPs have gotten much bigger and more complicated, often testing for multiple pathogens in multiple areas of a facility. For example, in the early days of Listeria EMPs, many programs sampled only drains in the ready-to-eat (RTE) area. If a drain was positive, it was re-sanitized and re-tested. Today, it is not uncommon for a food or pet food manufacturing facility to test for quantitative indicator groups at pre-op to verify the effectiveness of sanitation (mainly on product contact surfaces), and also test for specific pathogens in multiple sampling zones (see Figure 1) and in multiple hygienic zones of the facility (see Figure 2). This increasing complexity of EMPs also means that the data evaluation becomes more complex.
In addition to these complexities, EMP data can come from several different types of testing. Routine monitoring is the backbone of most programs. It is the daily, weekly, or monthly testing that is conducted in a relatively steady manner to verify the efficacy of multiple prerequisite programs, such as sanitation, employee practices, hygienic design, and hygienic zoning. Quarterly sampling is not recommended since the time between testing is too great should a positive be found and a corrective action implemented. If the routine monitoring indicates that there is a problem, another form of testing kicks in, the investigational sampling. This testing is conducted to try to find the root cause or origination of the contamination. Investigations can be completed with just a few samples, but more realistically, hundreds or thousands of samples may need to be processed to solve a pathogen contamination issue. Another type of testing that can generate valuable data is special cause sampling. This includes the “extra” testing that is done to help verify control during unusual events, such as construction projects, inclement weather, power interruptions, or even periods of unusually high production volumes.
It is now nearly universally acknowledged that a good EMP will sometimes generate positive results and should be viewed as a win/win situation. Most regulators and auditors understand this, and the focus of their attention often becomes the appropriateness of the investigation and the corrective actions. This is where a good data management system is especially critical. The first important component of an investigation is understanding what should trigger an investigation. With a quantitative program, it may be triggered when a gradual but real increase in counts has occurred or when a spike above the usual background level occurs. For a qualitative pathogen EMP, the trigger may be a single positive on a Zone 1 site (direct product contact), or a repeat positive in this same area. It could also be a general increase in the number of positives in an area or zone.