6.1 The Need for In-process Control
6.2 The Hazard Analysis Critical Control Point (HACCP) Concept
6.3 Identification of Critical Control Points
6.4 Critical Control Point Specifications
6.5 Checking for Compliance with End Product Specifications
6.6 Incubation Tests
Fatal errors in low-acid canned food manufacture are rare, which, given the volume of production, suggests that traditional process control measures (achieved through staff education and training, inspection of facilities and operations and testing or examinations) are effective. This comes as no surprise; for ultimately it is in the canners` interests to assure that their products are not only safe to eat, but also that they are of the expected quality. At the worst, failure to regulate end product quality will lead to outbreaks of food poisoning and expensive recalls; at best, it will gradually undermine the image of the product, and it will limit the ability of the manufacturer to supply to an agreed specification.
The collective experience of the international fish canning industry is generally sound; nevertheless, manufacturers can ill afford to overlook the outbreaks of botulism which in 1978 and 1932 led to the death of three consumers of commercially canned Alaskan salmon. In both these cases spoilage was the result of post-process contamination by C. botulinum (type E). The first outbreak involved only one can from a production lot of 14 600 units, yet the manufacturer inspected (visually and with a dud detector) some 14 million units. Reportedly, 3 515 cans were screened for botulinal toxin and all were negative. The second outbreak was attributed to a single can from a production lot of some 24 000 cans, and led to the recall of 60 million containers from nine canneries. During this investigation approximately 1 000 cans were tested for the presence of botulinus toxin and none was found. In addition to the logistical difficulties of implementing extensive recalls, these incidences demonstrate the impracticability of relying on large scale product recalls and quality audits as a means of detecting unsafe finished product.
In 1978 the International Commission on Microbiological Specifications for Foods (ICMSF), stated that microbiological sampling methods are inappropriate for assessing the safety of low-acid canned foods; they said "...experience demonstrates that, if present, C. botulinum would be expected to occur at such low frequency that no conceivable sampling plan would be adequate as a direct measure of its presence." Theoretically, the probability of a single C botulinum spore surviving a "botulinum cook" (Fo = 2.8 min) is estimated at 10-12 (see section 1.3). The probability of botulism arising through C botulinum entering containers, via post-process contamination, has been estimated to be from 10-7 to 10-10; while that due to botulism being caused by the containers failing to receive a thermal process has been estimated at between 10-6 and 10-8. Notwithstanding that these figures are estimates and difficult to validate, it is clear that when the probabilities of C. botulinum (or its toxin) being present in a can of low-acid canned food are so low, the chances of detecting it by terminal analyses are remote - even with a 100% inspection procedure. Should the method of testing be reliable, which is not necessarily the case (e.g., when the presence of vacuum is taken to be the indicator of safety from non-proteolytic type E C botulinum), many of the test procedures that are available are destructive and therefore not feasible. Questions as to the value of terminal analyses were revealed at the twelfth session of the Codex Committee on Processed Meats and Poultry Products when it was reported that "...there was general agreement within ICMSF that indirect implant control and hygienic post-processing handling were better measures...(for protecting public health)...than extensive end product examination." (Codex Alimentarius 1982). It is recognized that traditional end product sampling procedures can indicate a gradual deterioration in performance, but they ought not be relied upon to detect manufacturing defects which may compromise the safety of the product.
The two botulism outbreaks cited, and the prevailing attitudes toward traditional terminal analyses (microbiological and/or physical), underscore the desirability of alternate methods to assure the safety of canned fishery products. Hence the attraction of a process control system that minimizes the chances of manufacturing defects, while providing permanent records which demonstrate that the canned product was prepared according to generally recognized standards of good manufacturing practice. It is against this background that application of the Hazard Analysis Critical Control Point (HACCP) concept for in-process control in the manufacture of canned fishery products warrants the attention of canners and regulatory agencies.
There are three elements to the HACCP approach for in-process control:
- the environment within the can is suitable for toxin formation, and
- it is conceivable that under some circumstances the finished product is not likely to be treated (e.g., heated prior to consumption) in a manner which can be relied upon to render harmless any toxin that may be present.
The distinction between traditional control mechanisms and the HACCP approach to in-process control is clear. The former relies on terminal analyses to assess the adequacy of each operation, while the latter relies on in-process testing to demonstrate that all factors critical to the safety (and marketability) of the product have been adequately controlled. Given that improperly manufactured canned fishery products present a potential health risk, and since the safety of high volume canned food production cannot be assessed solely by terminal analyses, the error prevention techniques of the HACCP concept are both rational and potentially more cost effective.
Central to the implementation of the HACCP concept for in-process control is the identification of the critical control points. Since manufacturing techniques for canned fishery products vary greatly between plants, it follows that there will be different CCPs for different production lines. The CCPs for a canning process can be identified with the aid of a process flow diagram which should be constructed for the entire operation from the receipt of the raw and packaging materials through to transport and storage of the finished product. The standard symbols used in process flow diagrams are shown in Table 12.
Process flow diagrams provide a visual means of summarizing the entire sequence of production, and from this it is a simple step to identify CCPs. A typical list of CCPs for manufacture of canned fishery products (preserved by heat alone) is shown in Table 13. This list identifies those points in production where the manufacturer must establish specifications, a monitoring system complete with records, and a follow up system to confirm that any adjustments made achieve the desired effects. The frequency of monitoring and the test procedures to be used must also be specified. As the list may not be appropriate for all canning operations, it is important that manufacturers construct their own process flow diagram so that they can be sure no CCPs have been omitted - or included unnecessarily.
In commercial practice there are only a few specifications which are constant for all fish canners. Instead, manufacturers must select specifications which are relevant to each of their particular operations, while taking care that they fulfill the absolute requirement of product safety. Specifications should be set to reflect the desired sensory qualities of the product, they should be realistic and they should be geared to the ability of the plant to match them while remaining profitable. Although it may be difficult to define standards which are applicable to, and accepted by, an entire industry, it is possible to speak in terms of compliance with generally recognized standards of good manufacturing practice (the GMP guidelines referred to throughout. this text, and in particular in sections 1.5, 2.1.4, and 3.2).
For further information regarding GMP guidelines for cannery operations, readers should consult the following publications:
The major objective in seeking compliance with GMP guidelines is protection of public health; and because of this many of the articles referred to above tend to neglect those factors affecting the sensory and the physical properties of the product. Where this information is sought, in the first instance, reference should be made to the Codex Standards for Fish and Fishery Products (CAC/VOL. V- Ed. 1. 1981).
It has been argued that it is unrealistic to suppose that terminal analyses of finished product can assure the safety of the canned product, and that a more rational method of achieving this aim is to implement a system of in-process control at process CCPs. However, this is not to imply that there is no need for terminal analyses of any description, for without doubt, the manufacturer must be confident that the finished product has the desired sensory characteristics. This means that some assessment (subjective and/or objective) of end product quality is necessary.
Table 12 Symbols for use in process flow diagrams
Delay or temporary storage
Permanent or controlled storage
Combined operation and inspection
The functions listed in Table 12 are described as follows:
|Operation:||Any change to the chemical, physical or microbiological characteristics of the food material is defined as an operation. Pre-cooking tuna, or steam flow closing filled cans are examples of operations. Operations require energy, through manual or automatic means, and take the material one step closer to the finished product.|
|Inspection:||Inspection, usually requiring labour and/or equipment is required to maintain in-process control but it does hot take the product any nearer to being finished. Inspection in fish canning operations includes checking for the removal of bones and viscera prior to filling, measuring product fill weight and temperature, and checking that the retort is at scheduled operating temperature.|
|Transportation:||Transportation requires labour and/or equipment to move the product. An example is moving filled un-processed cans packed in retort baskets at the end of the filling line to the retort.|
|Delay:||A delay or temporary storage occurs when there is an unscheduled interruption to the process which is the result of a constraint other that imposed by the method of production. Examples include delays while retort baskets are held awaiting entry to the next available retort.|
|Storage:||Storage occurs when the material is held under controlled or permanent storage, as when bright stacked cans are held for labelling.|
Table 13 Checklist of critical control points(CCPs) for manufacture of canned fishery products
|Raw and packaging material quality|
|Product temperature and delay during preparation|
|Filling weight (liquid to solid ratio if applicable)|
|Container size and adequacy of the hermetic seal|
|Retorting (venting; process time, temperature and Fo value; cooling technique)|
|Plant sanitation and cooling water chlorination|
|Transport and storage conditions|
In addition to the assessment of colour, flavour and texture, manufacturers should monitor (and record the results of their evaluations), for several other factors which contribute directly and indirectly to end product quality. These include the following:
Although they should not be used as the sole criterion of product safety, incubation tests can provide valuable information as to the adequacy of the thermal process and also a means of monitoring (indirectly) the microbiological 1 quality of in-coming raw materials.
Should a process be of marginal severity, so that a measurable proportion of the population of spore-forming thermophilic bacteria survive, it may be possible to detect changes in the incidence of spoilage after thermophilic incubation tests on the production samples which have been collected as part of routine quality control. It is difficult to predict the level of spoilage in the trade which correlates with a known incidence of spoilage arising from incubation of test samples; however, it was reported by Stumbo (1973) that a thermophilic spoilage level of 1% after thermophilic incubation was found in commercial practice to give rise to a spoilage rate of 0.001% (i.e., 1 in 100 000 units) in the trade. Should a fish canner be able to collect sufficient data to draw their own conclusions concerning the relation between spoilage induced by thermophilic incubation and trade spoilage, the value of these incubation tests becomes clear; particularly for those manufacturers whose products are expected to be marketed in warm climates.
Under normal circumstances there would be little point in routinely monitoring spoilage arising from mesophilic incubation as all the spores which might lead to growth under these conditions should have been eliminated by thermal processes in which target Fo values are 10 to 15 min. However, if there are incidences of mesophilic spoilage detected by these test measures, it is reasonable to conclude that there has been either a significant lapse in the microbiological quality of the raw materials, or a gross failure in the delivery of the scheduled thermal process (see Table 3 for a summary of factors that could lead to this phenomenon). In such circumstances corrective remedial action should follow immediately and, suspect stock should be isolated pending a detailed examination.
Incubation test may be carried out in the laboratory or with bulk samples. With the former, the validity of the results must; be verified before any conclusions as to the suitability of the test sample (and by implication, the suitability of the population from which the samples were drawn). Factors to be considered when selecting testing procedures for laboratory incubation include:
With bulk incubation the factors to be considered include: