TECHNICAL ADVISOR'S REPORT TO THE FOOD, DRUG, AND COSMETIC PACKAGING MATERIALS COMMITTEE

December 3, 1999

Lester Borodinsky, Ph.D.*

Ladies and Gentlemen:

Ladies and Gentlemen: It is a great pleasure for me to be with you once again in the role of the Committee's Technical Advisor. This report will describe one of the current technical issues we are dealing with in establishing satisfactory regulatory status for plastic food-contact materials. Specifically, this report will address the need for technical attentiveness when standards are being set and when third-party certifying authorities are being established.

Background

It is often desirable, or even necessary, for standards to be set so that a given material or article is deemed appropriate for a given use. Furthermore, it is also often desirable or necessary for a third party to provide written certification or accreditation for the material or article. While there are many reasons why standards are set and why third-party certifying organizations become established, a desirable result of these practices is that they can be, or even at times must be, used for customer assurance. For example, in demonstrating compliance with a standard for which safety is relevant, it is not only critical that a material or article be safe for its intended use, it is also important that the supplier of the material or article be able to assure the customer that the standard is, in fact, met; a valuable tool to provide such assurance is a written, recognized certification. The bottom line is that setting (and meeting) standards, as well as written certification to these standards, can be a beneficial commercial instrument.

However, although such a customer assurance tool can have great value, it is vital that the standard is not set at such a tight level so that there is an unwarranted impediment to certification. While there are many "knowledge domains" that are required for setting standards, not the least of these is appropriate technical expertise - it is critical that there is close technical scrutiny of the details that go into setting and implementation of the standard. The current relevance of this subject is that, as we have advised the Committee in our update letters, Underwriters' Laboratories, Inc. (UL) and NSF International (NSF) (in conjunction with 3-A, the dairy standards organization) are establishing certification programs for equipment in meat and poultry plants.

Likely Certification Programs for Equipment in Meat and Poultry Plants

As many of you know, the U.S. Department of Agriculture (USDA) has eliminated its former program of prior approval of equipment for use in USDA-inspected meat and poultry plants. USDA's Food Safety and Inspection Services (FSIS) was unwilling to retain the equipment approval program, concluding that the program had more to do with the marketing of equipment than its safety. This action to eliminate the equipment approval program was taken as part of the move to the Hazard Analysis of Critical Control Points (HACCP) approach in which plant operators and their suppliers take over more responsibility for the safe production of meat and poultry, reducing the role of USDA prior approval and inspection. It is our understanding that the meat and poultry industries have been concerned about placing the entire responsibility for the suitability of equipment and other materials used in their plants on the plant operators and their suppliers, without any third-party review. Since FSIS was unwilling to keep the program alive, industry persuaded Congress to pass legislation directing the Agriculture Marketing Service (AMS) portion of USDA to establish a voluntary equipment certification program. AMS published an advance notice of proposed rulemaking on July 16, 1999, and, as you may recall, we filed comments on that proposal on behalf of the Committee. 1

Both UL and NSF/3-A are in the process of establishing programs to certify that equipment (and potentially other substances, such as lubricants) meet appropriate standards for use in meat and poultry plants. UL has drafted a proposed standard, identified as ANSI/UL Standard 2128.2 NSF International and 3-A jointly also have developed a proposed standard to cover meat and poultry plant equipment, with modular belts and hand held tools to be specifically addressed in separate standards.3

It is, therefore, critical that the individuals and parties that will be affected by the establishing and implementation of such programs take the opportunity at the outset to make sure sound and realistic technical concepts are incorporated into the standard programs. The responsibility is not finished at that point, however, as it is our experience that such standard setting and implementation programs are evolutionary. Therefore, technical attentiveness to the system is needed on a continual basis. The following example, relating recent activities with NSF Standard 61, demonstrates the need for such attentiveness.

ANSI/NSF Standard 61

The objective of American National Standards Institute/NSF International (ANSI/NSF) Standard 61, "Drinking Water System Components - Health Effects," is to establish minimum health effects requirements for the chemical contaminants and impurities that are indirectly imparted to drinking water from products, components, and materials used in drinking water systems. The products and materials include process media, protective materials, joining and sealing materials, pipes and related products, mechanical devices used in treatment/transmission/distribution systems, and mechanical plumbing devices. Standard 61 was established in 1988, and has been revised on a regular basis since its inception.

Products or materials are evaluated against the Standard by performing specific extraction tests set forth in Standard 61; the extraction testing procedure varies depending on the end-use of the product or material, i.e., distinct time and temperature procedures are established for pipes and related products, barrier materials, joining and sealing materials, process media, and mechanical devices. Furthermore, the compositional nature of the product or material dictates that the extract obtained using these procedures be analyzed for specific substances or groups of substances related to the chemical nature of the product or material, also set forth in Standard 61. In addition, NSF has recently been employing a new analytical procedure, which is not at this point specifically included in the Standard, with the intention of using it to determine if there are any unexpected materials extracted that are not anticipated from a review of the formulated product. The experimental data obtained from the extraction studies are used to determine exposure levels for the extracted substances.

Our attention was drawn to Standard 61 because several products that had previously received certification under Standard 61 were not meeting the Standard when it was time for recertification. It is important to recognize that the "failures" were not a consequence of changes to the formulation or composition of the products. Rather, the implementation of the Standard had changed. Because of the need for a remedy for the situation, a studious examination of the Standard was required, which turned up several aspects of the Standard that might lead to an exaggerated estimate of exposure. A brief overview of these aspects is set forth below.

Standard 61 is structured so that "worst-case" testing is first performed, followed by more realistic testing, if necessary. Specifically, Standard 61 calls for a conditioning phase for the test article, intended to represent the "break-in" period during which water contacting the article is not consumed; we understand that a characteristic example of the "break-in" period is the one that occurs during home construction. For example, the Section in the Standard that addresses pipe indicates that the test specimens should be preconditioned by exposure at room temperature (23 C) for 14 days (or less if specified by the manufacturer). The exposure water should be changed at least 10 times during the 14-day conditioning period, with at least 24 hours for each exposure period. After the 14-day conditioning phase,4 the test period begins - for pipe, the protocol is for exposure for three successive periods at 23 C, each with fresh water; the first two periods are 24 hours apiece and the third period is 16 hours.5 The water obtained from the third period is tested and the value obtained is the worst-case value determined as an initial matter.

It is readily apparent that this value is an exaggeration over actual anticipated levels. Specifically, the exposure period noted above for pipe is for a period of 16 hours. It is understood that the length of time is intended to represent a "work-day" period in which consumers are not at home to allow water to flow through the system and, instead, the water is static in the system. This is a worst-case, as this paradigm requires the assumption that all of an individual's water consumption for a given day is always drawn from water that has remained static in the system for 16 hours, i.e., it does not account for the fact that the level of any migration from the device would be diluted by flowing water. One aspect that becomes apparent is that Standard 61 currently does not account for the total amount of water that may contact an article over its useful life. Furthermore, it is a significant exaggeration of likely exposure to consumers as (1) the procedure in Standard 61 is to assume that an individual's entire one-day water consumption (i.e., 2 liters) is water that has been in contact with a particular material or device, and (2) it is also assumed that the rate of extraction of a substance remains the same over the lifetime of the device.

In recognition of the second exaggeration noted above, Standard 61 contains an option in the event that the worst-case testing results in extraction levels that do not meet acceptable levels. In such a case, "over-time" testing is initiated. After the 14-day conditioning period, the test samples are exposed for either of two additional periods:

(1) Testing is performed for 5 additional days. As the 14-day conditioning phase precedes the 5-day period, the total test period is 19 days; there are at least 12 water changes, each contacting the sample for a period of at least 24 hours. The results are extrapolated to 90 days. In each case, the results of a given analysis are normalized for the appropriate period (i.e., 24 hours), and the 90-day extrapolated value is expressed on a 24-hour basis.

(2) Testing is performed for 90 days, with sampling at 2 points, at a minimum; the time points are day 1 of 90 and on day 90. As the 14-day conditioning phase precedes the 90-day period, "day 1" is actually the 15th day of cumulative exposure and, similarly, "day 90" is the 105th day. In addition to the minimum of 2 samples (days 15 and 105), samples collected during the 14-day "conditioning" period also are tested. The 90-day (day 105) value is based on a 24-hour exposure period.

In addition, we realized that actual intake levels will be far below those determined either by a measurement on day 90 or by consideration of the daily average approach, such as that included in the calculation noted above. An additional exaggeration arises because the assumption currently employed does not account for the fact that all of the water is not consumed; specifically, the testing is based on the assumption that all of the water obtained in the sequential 24-hour periods is consumed and that none of it is diverted for other purposes, such as washing clothes or dishes. The water used for these other purposes may contain the migrant, but it will not be consumed; the subsequent aliquot of water, the one that will be consumed, will contain a lower concentration, as demonstrated in our calculations. We also presented a calculation to demonstrate the extent of this exaggeration.

Our review of the Standard turned up additional areas needing comment to NSF - validation of the analytical results, toxicology requirements, and use of materials already cleared for contact with food by the Food and Drug Administration (FDA).

With regard to validation, our observation was that Standard 61 should, but presently does not, make any provision for validation of analytical results. Consequently, we made the suggestion that the water extracts be validated, referring to FDA's recommended procedure for validating food-simulating solvent extracts to verify the analytical results.7

Regarding toxicology requirements, we suggested that the party requesting certification be permitted to provide pertinent toxicology information regarding close structural analogs in lieu of toxicology data on the subject substance. In addition, we pointed out that, for the Standard's "threshold of evaluation" (TOE), which parallels, and likely is derived from, FDA's Threshold of Regulation, (a) it is not necessary, as is now set forth in the Standard, to establish separate TOE levels for different applications, and (b) that, on a case-by-case basis, a higher TOE level can be established.8 Finally, we recommended that toxicology data on oligomers generally are not necessary in instances in which (1) there are adequate toxicology data on the relevant monomers (and extraction levels of the monomers meet the required limits established by Standard 61) and (2) the overall (i.e., total non-volatile) extraction levels do not exceed 50 parts per million (ppm).

In addition to the above, we recommended that the toxicology data requirements be deemed satisfied for a substance that has been granted clearance by FDA, as the safety of a material for contact with water has already been established sufficiently in instances in which the material is cleared in a food additive regulation under conditions of use that encompass the intended use in contact with water. As this activity in which we are dealing with NSF is still on-going, we do not yet have a final outcome to report to you at this time. Nonetheless, it is important to recognize several aspects of this activity. First, we are not approaching this as an adversarial relationship with NSF. On the contrary, we are working with NSF to amend the process so that the Standard properly reflects a real-world view. As noted above, the certification process has a valuable aspect to it. More importantly, this activity serves as a good example from which to learn how to approach the impending UL and/or NSF/3-a certification programs for equipment in meat and poultry plants, which may, in turn, lead to certification programs for other areas formerly handled by USDA. Specifically, while we are striving to remedy the Standard 61 situation, it would have been preferable to help steer the Standard in a more realistic direction as it was developed. Thus, it is important to closely review the technical details of not only developing standards and programs, but also ongoing programs, to provide practical guidance for their implementation.

Conclusions

The establishment of standards and third-party certifications can be valuable customer assurance tools. However, it is vital that the standards are set and certification programs are established in a valid, rational way. To ensure that such standards and programs that relate to them are implemented properly, it is important that the interested parties be active participants, where it is feasible to do so, not only in the preliminary and initial phases of the establishing of standards and certification programs but also throughout the lifetime of such processes.

* Prepared by Dr. Lester Borodinsky, Keller and Heckman, for the December 2-3, 1999 meeting of The Society of the Plastics Industry, Inc.'s (SPI) Food, Drug, and Cosmetic Packaging Materials Committee, Charleston Place, Charleston, South Carolina.

1 Essentially, we advocated that any certification program adopt or utilize a sanitary standard that provides for adequate flexibility with respect to the basis for establishing FDA compliance of materials for the food-contact surfaces of equipment. FDA compliance has been the standard used by USDA for establishing the compositional suitability of a material for contact with meat and poultry.

2 As you will recall, we had provided comments in August to Underwriters Laboratories (UL) on their draft Standard 2128 for meat and poultry processing equipment similar to the comments we provided to the AMS advance notice of proposed rulemaking.

3 Based on our review of the NSF/3-A proposed standard Hygiene Requirements for the Design of Meat and Poultry Processing Equipment, and the proposed NSF/3-A standard Hygiene Requirements for the Design of Held Tools Used in Meat and Poultry Processing, it is our opinion that the language in these standards sufficiently embraces all the legitimate options for establishing FDA compliance of food-contact materials used in meat and poultry processing plants.

4 The samples collected during the 14-day conditioning period are retained, so that they may be tested, if additional testing is necessary. See "over-time" discussion below.

5 The two initial 24-hour periods are optional.

6 We recognized that there is a procedure contained in Standard 61 that allows the subsequent consideration of a flowing water application in instances in which the worst-case static test results in an unacceptable conclusion. For this reason, we did not address static vs flowing applications in our comments to NSF.

7 See FDA's "Recommendations for Chemistry Data for Indirect Food Additive Petitions," June 1995.

8 We referenced a recently published article by FDA's regulatory reviewers, including those most intimately involved with FDA's administration of the Threshold of Regulation, in which the authors present methods for extending the principle of a single threshold level to a range of dietary concentrations between 0.5 and 15 ppb based on the application of structure-activity relationships, genotoxicity data, and short-term toxicity data. See, M.A. Cheeseman, E.J. Machuga, and A.B. Bailey, "A Tiered Approach to Threshold of Regulation," Food and Chemical Toxicology, 37 (1999), 387-412.

More About SPI:   Vision and Mission . Membership . Business Units . Regional Offices . News and Publications . Calendar of Events . Terms and Conditions of Use