Implementing Preventive Controls

By David Anderson and Karina Martino

September 2016 marked the date that many food establishments were required to implement aspects of the Food Safety Modernization Act (FSMA). The Preventive Controls (PC) for Human Food rule (21 CFR part 117) and Animal Food PC rule (21 CFR part 507) are similar in many ways, with some deadline differences. The Human Food PC rule deadline was September 19 for most companies; for Animal Food, that deadline was applicalbe for compliance to GMPs, with the remaining compliance having an additional two years. Small and very small businesses may have more time to comply.

On August 23, FDA issued new draft guidance to help the food industry comply with the GMPs and the Human PC rule. Comments started August 24. The guidance included the first five of 14 chapters, and explained FDA’s current thinking on how to comply with the rule. (For more information, visit http://bit.ly/2bzMA8F.)

Two additional draft guidances were released August 24 to assist industry with the implementation of the Animal Food PC rule.

FOOD SAFETY PLANS. Both PC rules require that food processing facilities adhere to GMPs as defined in the FSMA regulations and develop food safety plans for any foods for which a preventive control is required by the hazard analysis. FSMA food safety plans share many similarities with HACCP plans. Both require hazard analysis, monitoring procedures, verification procedures, and records. HACCP prerequisite programs may be viewed as preventive controls in FSMA; and critical limits in HACCP may be defined as process controls in the food safety plan. A comparison of HACCP principles and FSMA requirements is shown in Table 1.

SCIENTIFIC/TECHNICAL BASIS. Scientific information may be used to conduct a hazard analysis or determine process controls and preventive controls. But what sources can be used for these? Some are noted in the preamble to the FSMA rules, published in the Federal Register. Others have been listed by Codex Alimentarius. Briefly, here are a few:

• Regulatory documents (such as performance standards). Processors familiar with USDA requirements for meat and poultry know this one. A required minimum reduction of pathogens is stated in the law.
• Published “safe harbor” processes. Some process times and temperatures have been well established and are known for food safety. The requirements for milk pasteurization are an example.
• Generally held historical knowledge. The ability of pH and water activity to control pathogen growth is well-known, generally held knowledge. When harmful bacteria are controlled by formulation, this knowledge could be helpful in a food safety plan.
• Published processing guidelines. Industry-group processing guidelines may be used by a processor. The GMA Science Operations group can assist with processing guidelines; the Almond Board of California published times and temperatures for dry roasting almonds to destroy pathogens.
• Peer-reviewed scientific literature or technical data. Microbiological and technical journals may be a source for data to support a food safety plan. Articles abound about process times and temperatures for pathogen destruction, limits to microbiological growth, and effects of process treatments.
• Scientifically valid experimental data, e.g., microbiological challenge study. A processor may wish to conduct its own study in the lab or plant.
• Previous validation studies. Perhaps a company has a library of prior studies that can be applied to a new product. If a prior study is substantially similar to a prior one, the determination may be made to use prior study data instead of conducting a new study.
• Scientific pathogen models; mathematical modeling. Models may be helpful in understanding pathogen growth, inactivation or destruction. Models may be used as guidelines to help determine the limits for experiments.
• Operational data and surveys. A plant may apply operational data or survey information to food safety plans. For example, microbiological sampling may help support a plant’s assertion that a “kill-step” process control is adequate. However, sampling data alone, in the absence of other controls, may not to be enough to show that a hazard is being adequately controlled.

NEXT: THE PLANT FLOOR. With the scientific and technical data in hand, the work moves to the processing area. That work may be described as an “in-plant validation.” The in-plant procedures and processes must be demonstrated to meet the requirements of the science. FSMA regulations allow 90 days for this in-process work, after a process has been introduced into a facility.

Imagine that a scientific article has shown that a given process time, process temperature, product water activity, and pH are sufficient to destroy pathogens in a product. Our job, then, is to determine that the plant’s formulation procedures give us a product that meets the water activity and pH limits. Further, we will confirm that the process meets the stated retention time in the equipment at the required temperature; and we will be sure that the plant has a calibration program in place for equipment used to monitor those formulation and process steps. We also may add a “safety margin” to each of the limits to use as plant operating limits. We will be sure that processing records clearly demonstrate how the limits are met, and determine how record review occurs. Corrective actions for an operator to take if a process does not meet the defined limits are also described. We will write a brief report of the activities of this in-plant validation, and include it with the food safety plan.

Implementation of food safety (and HACCP) plans in facilities require a knowledge of our products; an understanding of how scientific and technical information can be applied; and attention to preventive controls, process controls and recordkeeping. The requirements of FSMA regulations are causing some companies to think more thoroughly in each of these areas and to refine recordkeeping.

The authors: Anderson is senior science advisor; Martino is processing specialist science operations, GMA.