L. Imports
 
1. Background
    The majority of seafood consumed in the United States is imported.
FDA's surveillance system for imports largely consists of reviewing the
customs entries for fish and fishery products being offered for entry
into the United States, engaging in wharf examinations and sample
collections for laboratory analysis, and placing products with a
history of problems on automatic detention. As with domestic
inspections, this method is basically a ``snapshot'' approach that
places a significant burden on the government to uncover problems. It
has failed to result in full compliance or consumer confidence in the
safety of imported seafood. Consequently, the agency tentatively
concluded that HACCP controls should apply to imported fish and fishery
products as well as to domestic products. Among other things, FDA
proposed that the definition of ``processor'' explicitly include those
who process seafood in foreign countries.
    In addition, FDA tentatively concluded that the importer should
share some responsibility with the foreign processor for safety. More
often than not, it is an U.S. importer, rather than the foreign
processor, who actually offers imported fish and fishery products for
entry into the United States. The preamble noted that, while many
importers are conscientious about the safety of the products that they
import, others have little understanding of the potential hazards
associated with their products. Thus, the agency tentatively concluded
that the existing system of import controls had not promoted a sense of
responsibility in the import industry.
    Therefore, in addition to proposing to require that foreign
processors that export to the United States comply with part 123, FDA
proposed that importers of fish and fishery products take steps to
ensure that their shipments are obtained from such processors.
Specifically, FDA proposed that importers: (1) Have and implement a
HACCP plan that describes how the product will be processed while under
their control; (2) maintain a copy of the foreign processor's HACCP
plan; and (3) take affirmative steps to ensure that the imported fish
or fishery product was produced in conformance with the foreign
processor's HACCP plan and with the proposed sanitation requirements.
The agency also proposed that importers need not take affirmative steps
if the fish or fishery product was imported from a country with which
FDA has a MOU documenting the equivalency of the foreign inspection
system with the U.S. system.
2. Should Imports Be Subject to These Regulations?
    115. Approximately 70 comments addressed various aspects of the
proposed requirements for imports. Approximately half of the comments
that addressed the import provisions argued that it is necessary to
subject imported products to the same regulatory requirements as
domestically processed products. These comments were submitted by
processors, trade associations, State and foreign government agencies,
professional associations, and individuals. Many of these comments
argued that exempting foreign processors from the requirements of these
regulations would put the domestic industry at an unfair economic
disadvantage. Other comments stated that the import requirements would
increase consumer confidence in seafood because they would ensure that
imported fishery products have been produced under the same HACCP
requirements and held to the same sanitation standards as domestically
produced product. A few comments suggested that imported products are
more likely to present safety hazards than domestically-produced
products because of a lack of understanding of CGMP's on the part of
foreign processors. One comment asserted that a number of countries,
including Canada, the EU, Iceland, and Thailand are in varying stages
of establishing HACCP programs for their own domestic seafood
processors.
    Most of the remaining comments (approximately one-half) did not
comment on whether HACCP controls should be required for imported fish
and fishery products but discussed aspects of the agency's proposed
approach. These comments will be addressed later in this section.
    FDA did not receive any comments that persuaded it that imports
should be exempt from the requirements of these regulations. On the
contrary, the comments reflect a nearly universal recognition that the
safety of seafood cannot be adequately ensured if the majority of
products (that is, imports) are not subject to the same controls as
domestic products.
    Therefore, the agency has not modified the regulations' basic
approach for imports.
    116. Only two comments objected to the concept that imported fish
or fishery products should meet the same requirements as those for
domestic products. One of these comments argued that FDA should be
tolerant of a foreign processor that may not have the knowledge or time
to develop a HACCP plan before its product is ready for export and
urged the agency to develop a temporary waiver system to accommodate
such firms.
    FDA is convinced that a 2-year implementation period, as discussed
in the ``Effective Date and Compliance'' section of this preamble, will
provide sufficient time for processors, both within and outside the
United States, to develop and implement HACCP plans and otherwise come
into compliance with the provisions of these regulations. The comment
provided no basis for treating foreign processors any differently than
domestic processors in this regard.
    117. Another comment suggested that raw material fish and fishery
products imported for further processing in the United States should be
exempt from the requirements of the regulations but provided no reason
to support that position.
    The exemption requested by the comment would make it difficult, if
not impossible, to control environmental hazards that may be associated
with these products. This preamble and the preamble to the proposed
regulations fully discuss the conclusions of the NAS, which identified
raw material hazards, such as microbiological contamination in
molluscan shellfish and natural toxins in both shellfish and finfish,
as among the most pressing problems that must be addressed to ensure
seafood safety. For the most part, these hazards are best addressed at
the time of harvest and by primary processors, through HACCP, at the
time of receipt. In many cases, there is little opportunity for control
beyond the latter point. Raw material fish and fishery products for
further processing comprise a substantial portion of fish
 
  Page 65153
and fishery products imported into the United States (Ref. 212, p. 49).
Thus, to exempt foreign processing of such products from the
requirements of these regulations would be to greatly diminish the
scope and, therefore, the overall effectiveness of these regulations.
    118. One comment that supported the need for equitable treatment of
imported and domestically produced products urged the agency to provide
the same opportunities for processors abroad to familiarize themselves
with the requirements of these regulations as it does the domestic
industry. The comment argued that just printing the regulations in the
Federal Register would not fulfill that responsibility. The comment
further suggested that FDA send copies of guidance materials to all
known foreign seafood processors, preferably in their native language.
    FDA acknowledges the difficulty in reaching foreign processors with
information about the requirements of these regulations. However, mass
mailings to, and multiple translations of, these regulations and the
Guide for all foreign seafood processors that export to the United
States would not be practicable for FDA.
    The agency intends to reach foreign processors primarily by
briefing foreign embassy staffs and by communicating with U.S.
importers during public and trade association meetings. Based on
experience in disseminating information about U.S. requirements to the
import community, the agency expects that these two groups will provide
the necessary information and guidance materials (in the appropriate
languages) to the foreign processors that they represent. This same
approach was used in disseminating information about the proposed
regulations. In fact, FDA became aware of a Japanese translation of the
proposal shortly after it issued.
    In addition, FDA traditionally has provided training and technical
assistance for foreign processors and government officials on a variety
of food control topics, within the constraints of budget and manpower.
These projects have principally been conducted in developing countries,
often those in which the agency has become aware of a particular
problem that threatens the safety of products offered for entry into
the United States. FDA anticipates that these kinds of projects will
continue, and that they will focus more closely on HACCP. FDA also
expects that HACCP training, performed in accordance with the
standardized training materials under development by the Alliance (see
the ``Training'' section of this preamble), will provide further
opportunity for foreign processors to be exposed to the requirements of
these regulations.
3. Should Importers Be Subject to These Regulations?
    119. Approximately half of those who commented on the import
provisions addressed whether the importer should be required to take
steps to ensure that its shipment originates from a foreign processor
that operates under HACCP. Approximately half of these comments favored
the concept and half opposed it, with both groups being diverse in
their representation.
    Of those who opposed it, many argued that these requirements should
be the responsibility of the government, and that FDA should not
require that importers enforce them. A number of these comments further
argued that equivalent foreign government inspection systems cannot be
presumed to be in place, and that the only way to achieve a ``level
playing field'' is for FDA to perform inspections of foreign processors
at the same frequency, and using the same standards, that the agency
applies to domestic processors. One comment suggested that it may be
necessary to obtain legislative authority to perform foreign
inspections, as a condition of importation. Another comment suggested
that FDA auditing of foreign processor compliance would give importers
assurance that the products that they obtain from such sources had been
produced in accordance with appropriate U.S. standards.
    One comment, while not opposed to mandatory importer
responsibilities, nonetheless argued that FDA should spend as much time
and effort inspecting foreign processors as it does on domestic
processors because over 50 percent of the seafood consumed in the
United States is imported. The comment continued that, ``to do any less
would be an unfair burden to domestic processors and would not
accomplish the stated goal to significantly improve the safety of
seafood consumed in the U.S.''
    One comment argued that there is no real cost savings in assigning
importers the responsibility of verifying foreign processor compliance
rather than assigning that responsibility to FDA, because importers
will merely pass along the additional costs to the consumer. Another
comment noted that many small importers obtain products from over 25
countries, and that they cannot afford to provide the surveillance
necessary to ensure compliance.
    Another comment argued that many importers function simply as
brokers, connecting a buyer with a seller, and that they lack the
expertise, manpower, and facilities to evaluate the adequacy of a
processor's HACCP controls. One comment stated, ``Many of the people
involved in importing never see the product and know nothing about
fish--these are people in a small room with a battery of phones!''
Another comment argued against placing reliance for assuring the safety
of imported seafood on persons who have a financial interest in the
product but lack the required knowledge about seafood safety.
    One comment argued that requiring importers to exercise control
over their suppliers has no parallel in the proposed domestic HACCP
scheme. The comment stated that domestic processors must control the
hazards that are introduced during their processing operations but need
not be involved in verifying the control of those hazards associated
with their supplier's operations. Some comments argued that the
responsibility for controlling hazards that are reasonably likely to
occur should be assigned to the foreign processor, while others argued
that it should be assigned to the U.S. processor to whom the importer
sells the product. One comment asserted that importers are not in a
position to exercise control over the processing of products in foreign
plants any more than they are in a position to exercise control over
how the products are handled by their customers.
    Most of those comments that supported the concept of importer
responsibility provided no reason. However, one comment stated that
requirements on importers would ensure that someone in the United
States would be legally responsible for the safety and wholesomeness of
each imported product.
    FDA recognizes that requiring importers to take steps to ensure
that foreign processors from whom they purchase seafood products are in
compliance with these regulations could necessitate significant changes
in the operations of importers who have limited their activities to
matching buyers with sellers based on product specifications that may
have had little to do with safety. However, for two reasons, FDA cannot
agree that responsibility with regard to safety is inappropriate for
importers.
    First, it has always been the importer's responsibility to offer
for entry into this country products that are not adulterated under
U.S. law. It is a prohibited act, under section 301(a) of the act, to
introduce into interstate commerce an adulterated food. Thus, an
importer would be committing a prohibited act if it failed to ensure
that
 
  Page 65154
the food that it is offering for import into the United States is not
adulterated under section 402 of the act, including section 402(a)(4),
one of the principal provisions on which these regulations are based.
    Currently, however, the importer is not required to operate in a
proactive manner to ensure that it is meeting this responsibility.
Rather, the importer need only offer products for entry into commerce
and thereby place the burden on the government to find a problem. Many
importers traditionally have purchased ``FDA rejection insurance'' to
hedge against that possibility. The government can shift the burden to
the importer by placing the importer's products on automatic detention
if it finds problems that warrant such a step, but in most instances
the burden remains on the government.
    Second, responsible importers understand the issues related to the
safety of the seafood products that they import and customarily require
that foreign suppliers conform to their product specifications and
applicable U.S. regulations relating to safety. These importers take
various measures to ensure that a foreign processor can comply with
their specifications and safety requirements before they agree to
purchase products from the foreign processor.
    Thus, it is feasible for importers to take steps to ensure that
they are not offering adulterated products for entry into U.S.
commerce. Requiring such measures will not be a significant added
burden for many importers, particularly as HACCP principles become more
widely used and understood in international commerce. Foreign
processors that want to participate in the export market, not only to
the United States but to the EU, Canada, and an increasing number of
other countries, will implement HACCP and sanitation control programs
and will be prepared to address an importer's needs for verification.
    FDA does not agree that there is no parallel in the domestic scheme
to the importer's responsibility to ensure that the goods it is
offering were produced under HACCP. Domestic processors, like importers
must work with their suppliers (e.g., fishermen) to ensure that all
reasonably likely hazards (e.g. natural toxins and agricultural and
industrial chemical contaminants) are controlled. FDA is confident that
importers, like processors, will realize that ensuring that foreign
processors institute preventive control systems is a cost effective
means of ensuring that the products that they offer for entry into the
United States will consistently meet FDA's entry requirements and will
be safe for consumption. FDA also disagrees with those comments that
suggested that a requirement that importers take steps to ensure that
the products they offer for entry have been produced under a HACCP plan
is an abrogation of FDA's responsibilities. As stated previously, the
industry has a responsibility to ensure that the food that it
introduces into interstate commerce is not adulterated. FDA has a
responsibility to verify that industry is meeting its obligation and to
take remedial action if industry fails to do so. Importers, who are
usually the owners of the products that they are offering into
commerce, are a part of that industry. FDA cannot accept that importers
have no responsibility to ensure that their products are not
adulterated.
    The agency recognizes that probably the most effective way for a
regulatory agency to evaluate a processor's compliance with the HACCP
and sanitation requirements is through onsite inspection of facilities,
practices, and records. FDA has performed a limited number of
inspections of foreign processors and, within its budgetary
limitations, will continue to do so to enforce these regulations.
However, such inspections are costly, and any attempt to significantly
increase their number would require additional resources.
    FDA will continue its traditional import surveillance role,
utilizing entry document review, wharf examinations, sample
collections, and automatic detentions as screening tools. These tools
indirectly evaluate the adequacy of HACCP and sanitation controls and
will continue to be useful in detecting significant problems. While
end-product testing and evaluation are not adequate substitutes for
preventive controls in ensuring the safety of a product, they can
provide verification where appropriate (Ref. 34, pp. 201-202).
    FDA has concluded that requiring HACCP controls, together with
import surveillance and periodic inspections of importers to ensure
their compliance with the requirements of Sec. 123.12, will better
ensure the safety of imports than the current system.
    In a related matter, Sec. 123.3(g) makes clear that, under ordinary
circumstances, freight forwarders, custom house brokers, carriers, or
steamship representatives will not be required to fulfill the
obligations of an importer. It is possible, although FDA has no way to
know with any certainty, that some of those that objected to being
required to fulfill those obligations would, as a result of these
clarifications, find that they would not be expected to do so.
4. Memoranda of Understanding (MOU's)
    120. Many of the comments that objected to the importer
responsibility provisions of the proposal on the grounds that the
government is the appropriate entity to ensure foreign processor
compliance, stated that the most effective means of ensuring such
compliance would be for FDA to enter into MOU's with the governments of
exporting nations. Approximately one-third of those that commented in
any way on the importer provisions urged FDA make the negotiation of
MOU's a high priority. Only one comment objected to the development of
MOU's.
    Several comments argued that FDA should develop MOU's with all
countries from which seafood is imported. One of these comments pointed
out that to do otherwise would unfairly cause the obligations of
importers to vary considerably. A few comments argued that the
existence of an MOU should be a prerequisite for the importation of
seafood products from a country. One of these comments stated that
mandatory MOU's would reduce the complexity of the present import
surveillance situation, reduce the number of countries exporting
seafood to the United States, and encourage the development of improved
food safety programs in exporting countries. Another comment asserted
that MOU development is appropriate because government-to-government
relationships and audits can be free of influence from packers and
importers, whereas foreign suppliers may be prone to provide false
assurances about their programs to prospective importers.
    One comment urged FDA to fully describe the process and criteria
for developing and evaluating MOU's and expressed concern about the
process because of the varying level of sophistication of foreign
seafood control programs. One comment stated that the foreign
government should be responsible for evaluating the foreign processor's
HACCP plan, inspecting the foreign processor, periodically analyzing
products produced by the foreign processor, and issuing health
certificates. A few comments stated that FDA should monitor the
effectiveness of the foreign government's control program in a manner
that is authorized in the MOU. These comments stated that, under the
MOU's, the foreign government should provide FDA with periodic lists of
processors that meet the requirements of these regulations, or,
alternately, that all seafood processors
 
  Page 65155
in the country would be required to meet the requirements.
    One comment urged FDA to publish periodic reports on the status of
MOU's on seafood products and to make them available to all importers.
This comment and others argued that it should be FDA's responsibility
to notify importers about changes in the status of MOU's, rather than
be the responsibility of the importer to find out about any changes.
One of these comments noted that, because a change in the status of an
MOU could be very detrimental to importers, there must be sufficient
lead time to allow importers to develop alternate verification
procedures when changes do occur.
    Another comment urged FDA to coordinate with U.S. importers and
exporters in developing a schedule for MOU development. The same
comment urged FDA to assign more resources to the development of MOU's.
    On the other hand, one comment stated that the MOU development
process is overly open-ended and could result in inconsistencies
between domestic and foreign requirements. The comment argued that this
inconsistency could result in an economic disadvantage for domestic
processors.
    FDA agrees with those comments that urged that the agency give high
priority to the establishment of MOU's with U.S. seafood trading
partners. In the absence of significant numbers of agency inspections
of foreign processing facilities, FDA acknowledges that an MOU can be
the most efficient and effective mechanism for ensuring that foreign
processing plants are operating in compliance with the requirements of
these regulations. FDA also agrees that the potential for signing an
MOU with FDA is likely to serve as an incentive for the improvement of
regulatory food control programs and processing conditions in seafood
exporting countries, especially where the existence of an MOU serves to
excuse the importer of products from that country from certain
verification activities.
    FDA has concluded that the development of MOU's or similar
agreements with foreign regulatory agencies is an appropriate method
for ensuring that foreign processors that export to the United States
are in compliance with the requirements of these regulations. Moreover,
as suggested by several comments, the agency has determined that, where
an MOU exists, there is no need for the importer to perform any
independent verification procedures for purposes of these regulations.
In this situation, the importer should be able to rely upon the foreign
regulatory authority to ensure compliance by foreign processors.
    FDA is therefore retaining the provision on MOU's from the proposal
but modifying it to provide that, where an importer elects to obtain a
fish or fishery product from a country with which FDA has an active MOU
or other similar agreement, the importer need not engage in any
independent verification activities.
    The agency has developed an internal protocol for developing MOU's
and is negotiating agreements with several countries. FDA is committed
to negotiating as many MOU's as possible. Also in the Federal Register
of June 15, 1995 (60 FR 31485), FDA published the notice of
availability of a new Compliance Policy Guide on MOU's.
    However, it is not reasonable to expect that an agreement could be
reached with all countries from which seafood is imported into the
United States. The barriers to achieving such a result include the
inadequacy of foreign regulatory programs and the lack of interest on
the part of some foreign governments in entering into an agreement. The
availability of FDA resources also can affect at least how long it
takes FDA to enter into a particular MOU.
    For these reasons, the existence of an MOU or similar agreement as
a requirement of entry of fish or fishery products into the United
States would result in an enormous negative economic impact to a major
segment of the U.S. seafood industry. Moreover, such a restriction is
not warranted from a public health perspective given the alternative
means of verifying the existence of HACCP controls that are provided in
these regulations.
    Experience obtained in part in the international portion of the
FDA/NMFS seafood HACCP pilot project has demonstrated that foreign
seafood regulatory programs vary considerably, both in their
capabilities and in their structures. Likewise, foreign seafood
processing conditions are highly variable. Thus, FDA cannot simply
follow a boiler plate format in negotiating MOU's. Rather, they must be
tailored to the specifics of the situation presented by a particular
country. It is possible that some agreements can provide simply for the
submission of lists of approved processors to FDA at regular intervals;
others may require much more extensive FDA involvement before and after
goods flow under the agreement. Some agreements may cover all of a
country's seafood processors, while others may be targeted to specific
species or product forms, depending on factors such as the capability
of the foreign regulatory authority.
    In any case, all agreements can be expected to provide for FDA
verification of the effectiveness of the foreign programs, including
onsite visits. FDA is principally interested in two- way agreements,
that is, agreements that acknowledge the acceptability of the U.S.
regulatory system to the foreign government as well as the
acceptability of the foreign regulatory system to the U.S. government.
    The agency will make every reasonable effort to communicate with
the industry about changes in the status of MOU's through Federal
Register notifications and other means. FDA is open to suggestions
about the best ways to communicate in this regard.
    Nevertheless, it will ultimately be the importer's responsibility
to keep appraised of any changes in the status of MOU's.
    The agency is also receptive to the views of the seafood industry
and others about how countries should be prioritized for the purpose of
negotiating MOU's. Any information that the agency receives on this
topic will be coupled with existing information concerning the
likelihood of negotiation success and the types and quantity of
products typically offered for entry from the country in question.
5. Importer Verification Procedures
    121. The remaining comments discussed specific aspects of the
proposed importer requirements. Some of these comments argued that the
responsibilities that were proposed for importers are onerous,
unworkable, and inefficient but offered nothing in support of these
assertions.
    A number of comments objected to the proposed requirement that all
importers have and implement a HACCP plan. Several of these comments
contended that an importer's plan can only address the hazards that
occur during the time that products are under the importer's control
(i.e., from the time the importer takes possession of the product until
it is shipped to its customer), and that requiring that the plan cover
this point is inconsistent with the principles of HACCP. One comment
agreed that an importer should be required to develop a plan if it also
processes the product, as in the case of an importer who stores the
product. The comment asserted that, in such a case, however, the
importer's HACCP plan would be minimal. The comment further asserted
that the foreign processor should be the party responsible for
developing a HACCP plan that addresses the hazards
 
  Page 65156
introduced during processing in the foreign plant. The comment
recommended that, as an alternative to having a HACCP plan, an importer
should be able to develop SOP's that outline the steps that it will
take to determine whether to purchase the product from a foreign
supplier.
    A number of comments supported the proposed requirement for
importer HACCP plans but provided no reasons for their support.
    The agency agrees that it would be inappropriate to require that
importers have and implement a HACCP plan regardless of whether they
process the products they import. As stated elsewhere in this preamble,
HACCP is a system that provides immediate feedback, through the
monitoring of CCP's, as to whether a process is under control. Unless
an importer is also a processor, there are no CCP's in the classic
sense for the importer to monitor, and from which to obtain real-time
feedback. Consequently, only where importers also process in accordance
with the definition of that term at Sec. 123.3(k) will they be required
to have and implement a HACCP plan that meets the requirements of
Sec. 123.6. Those food safety hazards that can be controlled by the
foreign processor must be addressed in the foreign processor's HACCP
plan in accordance with Sec. 123.6.
    Consequently, FDA has revised the regulations to limit the
responsibilities of importers. Instead of having to maintain their own
HACCP plan, under Sec. 123.12(a), in the absence of an MOU or similar
agreement, importers only need to maintain and implement written
verification procedures for ensuring that the fish and fishery products
that they offer for import into the United States have been processed
in accordance with the requirements of these regulations. The only
exception to this rule would be if the importer itself engages in
processing, such as holding food, in which case the importer would, as
stated above, also be a processor and subject to Sec. 123.6.
    122. In determining the nature of the verification procedures that
an importer must have and implement, FDA considered the comments that
addressed the appropriate functions and existing procedures of
importers. Several comments noted that importers routinely purchase
products according to specifications and observed that these
specifications could be the basis for reasonable control measures for
importers. The NACMCF recommended that importers be required to
maintain product specifications that are relevant to product safety for
fish and fishery products that they import. The NACMCF listed water
activity, pH, histamine content, and, perhaps, pathogen limits as
examples of specifications that importers might set in an effort to
ensure product safety.
    The agency agrees with the comments that product specifications can
be useful tools with which importers can exercise some control over the
products that they purchase and offer for entry into the United States.
In fact, FDA stated in the preamble to the proposed regulations that
the purpose of an importer's plan was, in part, to include criteria for
how the importer will decide to purchase seafood. FDA is also
encouraged by the fact that the comments generally agreed that having
product specifications would not constitute a new burden for many
importers.
    For these reasons, the agency in Sec. 123.12(a)(2)(i), is requiring
that the importer's written verification procedures include product
specifications that are designed to ensure that the product is not
adulterated under section 402 of the Federal Food, Drug, and Cosmetic
Act because it may be injurious to health or have been processed under
insanitary conditions. These are the adulteration sections that relate
to the safety of fish.
    In many cases, importers will find existing Federal food safety
standards, including tolerances and guidelines, to be useful
specifications. In other cases, specifications may need to be tailored
to the circumstances. For example, the importer might need to ensure
that the temperature of a modified atmosphere packaged product, when it
comes off a ship, is 38  deg.F (in such foods there is a risk of C.
botulinum type E growth and toxin development which can occur at
temperatures as low as 38  deg.F), although the CGMP's at
Sec. 110.80(b)(3)(i) state only that refrigerated foods should be
stored at 45  deg.F or below. The importer is encouraged to seek the
advice of qualified experts, as needed, in setting specifications. The
same resources available to help domestic processors in setting CL's
are available to importers, including the Guide; Sea Grant Institution
and other academics; Federal, State, and local food safety regulatory
agencies; consultants; the Alliance and other training courses; and
published literature.
6. Affirmative Steps: General
    As a second part of the importer's verification procedure, FDA is
essentially retaining from the proposal, in Sec. 123.12(a)(2)(ii), the
requirement that the importer take affirmative steps to ensure that the
products being offered for entry are actually being produced under
controls that meet the requirements of these regulations. In order for
product specifications to be meaningful, importers must take steps to
establish that their suppliers are in fact operating in a manner that
can reasonably be expected to produce a product that meets those
specifications. Effective verification involves scrutinizing the
standard, much like evaluating whether the HACCP plan continues to be
appropriate, and scrutinizing performance to determine whether the
standard is consistently reached, much like reviewing monitoring
records (Ref. 34, p. 201). FDA is adopting this approach in
Sec. 123.12(a)(2)(ii).
    Among the affirmative steps that FDA proposed that a processor take
were: (1) Obtaining the foreign processor's HACCP-monitoring records;
(2) obtaining a certificate from a foreign government inspection
authority certifying that the firm is operating under a valid HACCP
plan or certification on a lot-by-lot basis; (3) regularly inspecting a
supplier's facilities; (4) periodic end-product testing by the importer
or a private laboratory hired by the importer; or (5) other such
verification measures as appropriate. FDA listed these affirmative
steps as examples of the types of measures that would be acceptable to
the agency. FDA does not wish to predetermine all the possible ways
that an importer could perform affirmative steps.
    123. A number of comments objected to each of the affirmative steps
that FDA listed in the proposed regulations, and a few expressed
support for each. However, few of the comments provided any reasons for
their positions.
    One comment suggested that the best means by which an importer can
ensure that the conditions at a foreign processing facility are at
least equivalent to those required of domestic processors is for the
importer to verify the adequacy and implementation of the foreign
processor's HACCP plan during a visit to the facility. Another comment
stated that, ``without both audits and HACCP records, foreign plants
(possibly domestic facilities too) will not adhere to the letter of the
FDA regulation and assure safe product in the market.''
    Conversely, a number of comments argued that it would be unworkable
for importers to conduct inspections of foreign processors. One of
these comments stated that to justify the expense of such an
undertaking would necessitate that a highly trained, competent
individual perform the function.
    As stated earlier, FDA remains convinced that importers must
exercise
 
  Page 65157
sufficient control over the fish and fishery products that they offer
for entry into their country to ensure that the products are produced
pursuant to the requirements of these regulations. The agency
recognizes that any one of the affirmative steps may not be appropriate
or feasible for a particular importer or foreign processor. The
regulations allow importers to select an affirmative step that is
workable for their circumstances and to develop appropriate affirmative
steps other than those listed in the regulations (see
Sec. 123.12(a)(2)(ii)(F)). However, such measures must provide at least
an equivalent level of assurance of foreign processor compliance as
that provided by the listed affirmative steps.
    Additionally, FDA has modified the importer requirements to allow
for the performance of any of the affirmative steps by a competent
third party (Sec. 123.12(b)). This provision provides even greater
flexibility to importers in meeting the requirements of these
regulations.
    Thus, FDA is not persuaded that the affirmative steps are not
feasible or appropriate and has included them in these final
regulations.
    124. A comment argued that government certificates should not be
acceptable unless they are issued by countries with which FDA has
signed an MOU or similar agreement. The comment asserted that,
especially in developing countries, there may be different
interpretations of the regulations, and differences in competency,
credibility, infrastructure, intent, and uniformity that might bring
the utility of such certificates into question.
    FDA acknowledges that it is likely to have a higher level of
confidence in certificates received from a government entity with which
it has signed an agreement than with one with which no agreement
exists. However, as discussed above, it is unlikely that the agency
will be able to negotiate an MOU with every country that exports
seafood to the United States. Thus, there may be countries that have
excellent certification programs with which FDA, for a variety of
reasons, simply does not have an opportunity to enter into an
agreement. Moreover, if the agency learns, either through its own
routine surveillance activities, consumer complaints, or other means,
that there is evidence that a country is routinely issuing certificates
inappropriately, the agency will try to inform firms that import fish
or fishery products from that country that it will expect them to use
other means of verification if they want to avoid the appearance that
those products are adulterated under section 402(a)(4) of the act (see
Sec. 123.12(d)).
    125. One comment urged that certification be permitted on a
continuing basis rather than requiring lot-by-lot certification.
    FDA agrees that continuing certification is appropriate and notes
that the language and intent of the proposed regulations would have
allowed for it. Nonetheless, in an effort to further clarify this
situation, the agency provided in Sec. 123.12(a)(2)(ii)(B) that:
``Obtaining either a continuing or lot-by-lot certificate * * *'' will
be one way to satisfy the requirement that an importer take affirmative
steps to ensure that the product is produced in accordance with the
requirements of this part.
7. Foreign Processor HACCP Plans
    126. Approximately 15 comments addressed whether importers should
be required to have on file copies of the HACCP plans of each of their
foreign processors. Approximately half of these comments supported such
a requirement, although for the most part they provided no reasons for
their support. The other half objected to the requirement. One of these
comments argued that possession of a foreign processor's HACCP plan
would be cumbersome for the importer and would provide no assurance
that product shipped by that processor was processed in accordance with
the plan. One comment cautioned that it would be unrealistic to expect
that importers could make any but a rudimentary judgment as to the
adequacy of foreign processors' HACCP plans. Such judgments, these
comments asserted, should be reserved for the regulator when the plans
are assessed during inspections of importers' records.
    One comment cited the possibility of breaches in confidentiality
because commercially sensitive material would be supplied to importers.
A related comment suggested that, to solve the confidentiality problem,
the foreign processors' HACCP plans should be filed directly with FDA
rather than with importers.
    Although the agency continues to believe that a foreign processor's
HACCP plan provides a useful basis for verification, FDA is persuaded
by the comments that there are logistical and other issues that could
render the retention of HACCP plans by importers unmanageable in some
cases. FDA has also concluded that, in most cases, affirmative steps
such as those listed in Sec. 123.12(a)(2)(ii) (e.g., onsite inspection
by the importer and certification by a foreign government agency) will
be adequate to enable an importer to verify that the products being
imported are safe in accordance with the requirements of these
regulations.
    As described previously, the NACMCF recommendations describe two
primary goals of verification: (1) Ensure that the plan is adequate to
address the hazards that are likely to affect the product; and (2)
ensure that the plan is being consistently implemented. The affirmative
steps listed in Sec. 123.12(a)(2)(ii) are designed to address both of
these functions. For example, obtaining HACCP and sanitation monitoring
records from the foreign processor (Sec. 123.12(a)(2)(ii)(A)) enables
the importer to confirm that the foreign processor has addressed the
relevant hazards and sanitation concerns (i.e., those for which there
are monitoring records), and that it is monitoring to ensure that these
concerns are under control during the production of lots that are
shipped to the importer. Similarly, obtaining governmental or third
party certification of foreign processor compliance with the
requirements of these regulations (Sec. 123.12(a)(2)(ii)(B)) or
inspecting the foreign processor directly (Sec. 123.12(a)(2)(ii)(C))
enables the importer to confirm that the foreign processor has an
adequate HACCP plan and SSOP, and that the relevant sanitation and
safety concerns are being controlled for those lots that are shipped to
the importer. The affirmative step options provided for by
Sec. 123.12(a)(2)(ii)(D) and (a)(2)(ii)(E) are discussed later in this
section.
    Consequently, FDA has not included a requirement that importers of
fish and fishery products have on file the HACCP plans of each of their
foreign suppliers in these final regulations.
    Nonetheless, FDA points out that maintaining copies of these plans
could be one of several measures that an importer could incorporate
into its affirmative steps. Therefore, these final regulations in
Sec. 123.12(a)(2)(ii)(D) incorporate the concept as one of the
affirmative steps that an importer may choose to use for verification
purposes.
    127. One comment noted that the plans of foreign processors would
normally be prepared in the native language of the country of origin
and asked whether FDA would require that these documents be translated
into English. On the other hand, another comment recommended that HACCP
plans be maintained in both the language of the native country and in
English.
    FDA agrees with the comment that argued that a copy of a
processor's HACCP plan would not, by itself,
 
  Page 65158
provide adequate assurance that a given shipment of imported product
was processed in compliance with that HACCP plan or that the sanitation
requirements of Sec. 123.11 were met. One additional thing is needed to
provide such assurance: a written guarantee from the foreign processor
that the products shipped to the importer are processed in accordance
with these regulations. The guarantee is necessary to demonstrate that
the HACCP and sanitation control systems are being implemented for
products shipped to the importer. An importer should be able to make a
reasonable judgement about the validity of the guarantee through a
rudimentary review of the plan, as described below. Therefore, FDA is
including these requirements in Sec. 123.12(a)(2)(ii)(D).
    FDA is also providing in Sec. 123.12(a)(2)(ii)(D) that the foreign
processors' HACCP plans that are maintained by importers be written in
English, so that they will be meaningful to the importer and will allow
for regulatory review.
    128. As stated above, one comment cautioned the agency about the
ability of many importers to evaluate the adequacy of HACCP plans that
they might retain.
    FDA acknowledges that many importers may not have the technical
expertise to evaluate the adequacy of seafood HACCP plans. However, the
agency is convinced that, as a result of the importers' assessment of
the food safety hazards that are reasonably likely to be presented by
the product, the importer should have developed some general
expectations about the content of the HACCP plan (e.g., which hazards
should be addressed). The importer should be able to spot any obvious
shortcomings and to discuss them with the foreign processor. It is not
enough that importers simply file away the documents upon receipt.
Importers may find it advantageous to make a judgment about the
likelihood that their product specifications will be met and to insist
that they be given a guarantee that contains assurances that the
specifications will be met.
    129. Regarding the comment that complained about the potential loss
of confidentiality of foreign processor HACCP plans that are provided
to importers, since the agency has eliminated the requirement that all
importers retain copies of foreign processor plans, the significance of
this issue has been minimized. In the case where a foreign processor
does not wish to share its plan with the importer, the processor and
the importer would need to agree upon another means of providing for
importer verification.
    130. Regarding the comment that suggested that all foreign
processors file their plans with FDA, the resource demands on the
agency that would come with such an undertaking would be prohibitive.
FDA cannot accept this suggestion.
8. Other Affirmative Steps
    As a related matter, FDA has determined that, in the absence of a
requirement that importers maintain a copy of the foreign processor's
HACCP plan, finished product tests alone are insufficient as an
importer affirmative step to ensure that the foreign processor is
operating in accordance with these regulations. Finished product
testing alone has a small statistical likelihood of detecting defects
in a product, especially when the occurrence of such a defect is an
uncommon event, as is the case with most seafood hazards (Ref. 213).
The proposed requirement for the importer to obtain a copy of the
foreign processor's HACCP plan, in addition to performing finished
product testing, would have provided indirect evidence that HACCP
controls are in place and would have lent support to a conclusion,
based upon the analytical findings, that the relevant hazards are under
control. In the absence of such evidence, the importer cannot
reasonably conclude that the hazards are being controlled based solely
on a negative analytical finding. For this reason FDA has required in
Sec. 123.12(a)(2)(ii)(E) that such sampling be accompanied by a written
guarantee from the foreign processor that products being shipped to the
importer are processed in a manner consistent with the requirements of
these regulations. The guarantee provides the importer with reasonable
assurance that HACCP and sanitation controls are in place and are being
implemented, in a manner similar to the way that the foreign
processor's HACCP plan would have under the requirements of the
proposed regulations. Under this alternative, the importer would not
have to maintain a copy of the HACCP plan.
    For clarification and consistency within the document, FDA has
revised the language of two of the affirmative steps to include
reference to the sanitation provisions of the regulations. In both the
proposed regulations and these final regulations the stated purpose of
the affirmative steps is to enable the importer to verify that the fish
or fishery product was processed under conditions that meet both the
HACCP and sanitation requirements of these regulations. However, the
formulations of two of the affirmative steps in the proposal did not
make specific reference to sanitation. To avoid confusion over what the
affirmative steps should cover, Sec. 123.12(a)(2)(ii)(A) now reads
``Obtaining from the foreign processor the HACCP and sanitation
monitoring records * * *'' and Sec. 123.12(a)(2)(ii)(B) reads ``* * *
certifying that the imported fish or fishery product is or was
processed in accordance with the requirements of this part.''
    131. Several comments asked the agency to specify the frequency
with which the importer affirmative steps must be taken. A few comments
suggested that the frequency should be no greater than the frequency of
equivalent FDA verification activities.
    It would not be practical for the agency to specify frequencies for
affirmative steps that would be appropriate in all circumstances.
Consistent with the frequency of monitoring by processors, importers
should take affirmative steps to monitor their suppliers with
sufficient frequency to accomplish its purpose--that is, to provide the
importer with reasonable assurance that the foreign processor is
operating in compliance with these regulations.
    It would be inappropriate to tie importer affirmative step
frequencies to average FDA sampling and inspection frequencies. FDA
sample collection and inspection frequencies are determined, in part,
by the compliance history of individual firms, agency priorities, and
overall agency resources, not simply on a desired average minimum rate
of verification. Thus, FDA's rate of inspection has no bearing on how
frequently an importer should monitor a supplier.
    132. A number of comments urged that the agency permit importers to
contract with third parties to perform verification activities on their
behalf. Two comments opposed such a provision but did not provide
reasons for their position.
    Several comments urged that certificates by nongovernmental third
parties be accepted as an affirmative step. One of these comments, from
a trade association, suggested that an equivalent arrangement has been
accepted by FDA in controlling the importation of canned mushrooms from
the Peoples Republic of China. This same comment argued that a system
where individual importers inspect each of their suppliers is highly
inefficient. The comment suggested that a single, technically competent
party should perform the inspections. The trade association offered to
serve as a
 
  Page 65159
clearinghouse for the reports of such inspections. Likewise, the
association offered to serve as a clearinghouse for finished product
sample results for imported products, reducing the number of samples
needed when the same product is imported by a number of importers. The
comment further suggested that the association be permitted to hold
foreign processor HACCP plans for its members, and perhaps for
nonmembers. The comment argued that acceptance of this suggestion would
reduce the number of duplicate records for the same product stored by
various importers.
    The agency accepts that third party verification can be an
appropriate and efficient control mechanism. Such a system is
consistent with the use of third parties by processors for plan
development, record review, and CL deviation evaluation. Therefore, FDA
has added a new provision at Sec. 123.12(b), that reads, ``An importer
may hire a competent third party to assist with or perform any or all
of the verification activities specified in paragraph (a)(2) of this
section, including writing the importer's verification procedures on
the importer's behalf.'' It is worth pointing out that where an
importer uses the services of a third party, the importer remains
responsible for the verification procedures that are performed. The
importers must be able to demonstrate that appropriate verification
measures have been performed. This step may involve providing an FDA
investigator with a copy of the foreign processor's HACCP plan, results
of end-product sampling, results of an onsite inspection, the foreign
processor's monitoring records, or the foreign processor's written
guarantee. Third parties must, of course, be competent to perform the
duties in question, and FDA reserves the right to challenge such
competency. The agency has no objection to the use of clearinghouses
for importer verification activities, as long as the forgoing
requirements are met.
9. Importer Records
    As previously mentioned, the proposed regulations would have
required that importers develop and implement a HACCP plan. One effect
of such a requirement would have been that importers would have had to
maintain appropriate records. As has been explained, FDA is adopting
only those essential components of the proposed approach that the
agency considers to be practicable for importers. One such component is
recordkeeping. Recordkeeping is essential in documenting for the
benefit of importers and the agency the affirmative steps of importers,
in the same way that it is essential in documenting the monitoring,
corrective action, and verification activities of processors. For this
reason, the agency has retained the recordkeeping aspect of the
proposal for importers, in a manner that is consistent with the overall
approach for importers in these final regulations. Section 123.12(c),
which treats importer records identically to processor records, reads,
``The importer shall maintain records, in English, that document the
performance and results of the affirmative steps specified in paragraph
(a)(2)(ii) of this section. These records shall be subject to the
applicable provisions of Sec. 123.9.''
    133. FDA proposed that importers encourage foreign processors to
obtain HACCP training. A few comments urged the agency to make it clear
that foreign processors must comply with the same training requirements
as are applicable to domestic processors. One comment urged the agency
to permit HACCP- training courses for foreign processors to be
conducted in the country of origin by ``an official agency.''
    FDA agrees that the need for training is the same for foreign
processors as it is for domestic processors. The intended benefits of
the training requirements are fully discussed in the ``Training''
section of this preamble. Nonetheless, the agency finds that the
proposed requirement that importers encourage foreign processors to
obtain training is unnecessary. Foreign processors that ship seafood
products to the United States are advised of the training requirement
of these regulations in the same way that they are advised of the other
requirements of these regulations, through publication of the
regulations. In addition, as mentioned elsewhere in this preamble, FDA
intends to provide the embassies of seafood exporting countries with
information concerning these regulations in order that they may in turn
provide it to the processors in their countries. Consequently, FDA is
not adopting this provision.
    FDA has no objection to HACCP training being performed in the
country of origin by ``an official agency'' or other entity, as long as
the course of instruction is at least equivalent to that provided by
the standardized course under development by the Alliance.
10. Determination of Compliance
    FDA proposed to require that there be evidence that imported fish
and fishery products were processed under conditions that comply with
the requirements of these regulations, and that if assurances that this
was the case did not exist, the product would appear to be adulterated
and would be denied entry. This section of the proposed regulations
provided five types of evidence that the agency would consider as
adequate to provide such assurance.
    134. A few comments supported these provisions. However, a few
comments suggested that, if the importer is unable to provide assurance
that a HACCP system is in place, the importer should be permitted to
conduct finished product testing rather than having the product denied
entry. One comment urged that importers be held only to a ``best
efforts'' standard in determining whether their suppliers are in
compliance with these regulations. This comment suggested that if an
importer cannot determine that such compliance exists after using its
best efforts, the importer's product should not be banned from the
United States.
    The purpose of these regulations is to cause processors of fish and
fishery products, both domestic and foreign, to develop and implement
HACCP systems of preventive controls to ensure the safety of their
products. The importer requirements are designed to impose an
obligation on importers to ensure that, like domestic products, the
products that they are importing are not adulterated within the meaning
of section 402(a)(4) of the act. This requirement means that importers
must be able to satisfy themselves, and ultimately FDA, that the fish
and fishery products that they are offering for import were produced
subject to a HACCP system and sanitation controls designed to prevent
insanitary processing conditions that may render the food injurious to
health. If an importer does not have evidence that shows that the
products were produced subject to such controls, it should not offer
the product for import into this country. The lack of such evidence
creates the appearance of adulteration that cannot be overcome by the
collection and analysis of a finished product sample by an importer.
Given the problems that can arise in seafood processing if HACCP and
sanitation controls are not in place, under sections 402(a)(4), 701(a),
and 801(a) of the act, FDA is adopting Sec. 123.12(d), which provides
that if evidence does not exist that an imported fish or fishery
product has been processed under conditions that are equivalent to
those required of domestic processors, the product will appear to be
adulterated.
    Section 123.12(d) derives from proposed Sec. 123.12 (a) and (b).
FDA has combined these provisions and, as
 
  Page 65160
suggested by a comment, has used terminology consistent with the rest
of the regulation in Sec. 123.12(d). While proposed Sec. 123.12 (a)(1)
through (a)(5), which described the types of evidence that could be
used to demonstrate compliance with the proposed regulations, reflected
important principles for the importation of fish, based on the
comments, FDA finds that these provisions were causing confusion, and
that the statute can appropriately be implemented without including
them in the final rule. For this reason, FDA has not adopted these
provisions.
    135. One comment asked what documents, if any, would have to be
presented to FDA at the time of entry concerning the status of the
foreign processor. Another comment suggested that importers should note
on the entry documents that a HACCP plan is available for the foreign
processor. This comment stated that FDA would have an opportunity to
review the plan as part of its determination of whether to allow entry
of the product.
    FDA is not requiring that evidence of the importers' affirmative
steps be presented along with the existing U.S. Customs Service entry
documents as a matter of routine practice. It is possible that, in some
circumstances, such a step will be necessary (e.g., where the agency
has reason to believe that inappropriate conditions exist in the
foreign processing facility). However, typically, the importer will be
able to retain such evidence in its files and to make it available to
the agency when FDA performs an inspection at the importer's place of
business. Such a system is necessary because of the time that is
necessary for the agency to properly review the importer's
documentation of its affirmative steps and of the foreign processors'
HACCP plans. Nonetheless, the agency is willing to explore alternate
methods of implementing the import requirements of these regulations,
such as that suggested by the comment. FDA welcomes a continuing public
dialog about this matter.
    136. One comment asked whether FDA would maintain an approved list
of foreign processors.
    The agency has no plans to maintain such a list, nor is it apparent
upon what basis such a list would be prepared. A possible exception
would be as part of an MOU arrangement, where the foreign country would
agree to provide a list of ``approved'' firms to FDA. In such a
situation, FDA would use reasonable means to inform the import industry
of the purpose and contents of the list and update them as rapidly as
possible when changes are made.
    137. One comment expressed concern that the same foreign processor
HACCP plan might be reviewed by different FDA investigators in
different ports of entry, and that these investigators might reach
different conclusions as to its adequacy. The comment urged that the
agency coordinate such reviews, as well as reviews of importers'
affirmative steps, in a way that would minimize inconsistencies.
    FDA acknowledges that the situation might well arise where
different investigators review the same foreign processor HACCP plan as
a part of different importer inspections. To minimize inconsistencies
in such reviews, the agency intends to train its inspectional staff in
the requirements of these regulations and the application of HACCP
principles to seafood processing, including training on the Guide. The
agency also intends to develop guidance relative to importer
verification activities.
 
M. Guidelines or Regulations?
 
1. Background
    FDA recognizes that many processors will need guidance in the
preparation of HACCP plans, and that HACCP plans will vary in
complexity. The agency is committed to providing the industry with
technical assistance that includes general guidelines for HACCP plans
and the contents of plans for specific types of products and processes.
    As part of FDA's seafood HACCP proposal, the agency included
guidelines, in the form of appendices, on how processors of cooked,
ready-to-eat products and products involving scombrotoxin-forming
species could meet various provisions of the proposed regulations
relating to the development and implementation of HACCP plans. FDA
regards these products as being high-risk relative to other seafoods.
They involve special considerations or special hazards for which
additional guidance would likely be useful.
    Cooked, ready-to-eat fishery products present an elevated risk of a
microbiological hazard compared to most other seafood products. They
are cooked as part of processing and might not receive additional
cooking by consumers before consumption. Consequently, to be safe,
these products must not contain pathogens at a level that will cause
disease and must not be subjected to time-temperature abuse that would
allow any existing pathogens to grow to unacceptable levels.
    Scombrotoxin-forming species are fish that can form a toxin if
exposed after death for significant periods to temperatures that permit
the growth of certain bacteria. Scombrotoxin can result in a mild to
severe allergic response in humans.
    The guidelines for these products contained advice about hazards
that are reasonably likely to occur and on details for HACCP plans for
the control of these hazards. In addition to asking for comments on the
substance of the guidelines, the agency asked for comment on whether
these guidelines should remain as guidelines, or whether some or all of
them should be adopted as regulations. As regulations, they would,
essentially, tell processors that certain hazards must be controlled in
their HACCP plans, identify in advance critical points in the
processing of these products that processors must control to minimize
these hazards, and tell processors what they would have to do, at a
minimum, to maintain proper control of those critical points.
    In another appendix to the proposed regulations, FDA published
excerpts from the draft Guide, mentioned earlier in this preamble, for
the stated purposes of publicizing the existence of that draft Guide
and of providing processors with information about the types of
guidance that the agency expected would be available in it.
    One of the excerpts that FDA published was guidance on the
processing of smoked and smoke-flavored fish. These products represent
a significant hazard relative to contamination with C. botulinum,
especially when packaged in reduced oxygen atmosphere packaging. FDA
requested comment on whether this guidance should remain solely within
the Guide, whether it should be provided an appendix to the
regulations, or whether it should be adopted as regulations. The effect
of adopting these materials as regulations would be the same as for the
appendices described above.
    If these materials remained in the form of guidelines, processors
would be free to adopt them or not, so long as measures that provide an
equivalent or superior degree of safety are implemented.
    138. Approximately 55 comments responded to FDA's request for
comment on whether these materials should remain as guidelines or be
adopted as regulations. The majority of comments preferred guidelines.
A few comments suggested that FDA initially issue guidelines, then
possibly convert them to regulations after gaining experience with them
as adjuncts to a functioning HACCP system or after pilot testing them.
A few comments preferred to retain some of the materials as
 
  Page 65161
guidelines and convert others to regulations.
    Over one-third of those who commented on this subject supported
guidelines in general, without distinguishing among the three
guidelines. They argued that guidelines are in keeping with the general
philosophy of HACCP that processors assume responsibility for the
safety of their products. Some stated that detailed regulations for
processors to follow would not provide an adequate incentive to
processors to develop a full understanding of the hazards associated
with their products or processes. The result could be the development
of rote HACCP plans that might be inadequate for safety in specific
situations.
    Some comments pointed out that, while guidelines can assist
processors to identify controls, guidelines probably could never
properly identify the CCP's and limits for all processors given the
uniqueness of individual processing methods. In the case of
regulations, processors would be obliged to adhere to the presented
limits regardless of their appropriateness to the operation. Many of
these comments preferred the flexibility that guidelines provide in
permitting HACCP controls to evolve with a changing knowledge base and
new technologies. Some expressed concern that if the guidelines were
adopted as regulations, the industry would bear an unnecessary burden
of having to petition FDA for amendments in order to accommodate new
products or processes. Modifications to the regulations could take
considerable time.
    Several comments specifically objected to adopting either the
guidelines for cooked, ready-to-eat products or the guidelines for
scombroid toxin-forming species, or both, as regulations. The reasons
were generally the same as those given by those comments that supported
the use of guidelines generally.
    One comment did express the concern that adopting the scombroid
guideline as regulations would have the effect of adopting a policy
action level for histamine as a defacto regulation without a formal
notice and comment rulemaking.
    Several comments requested that guidelines only appear in the
Guide, and not in appendices to the regulations, to alleviate
confusion.
    However, FDA did receive a number of comments that urged the agency
to adopt these guidelines as regulations. These comments cited a need
for minimum enforceable standards for these products to ensure the
protection of the public health. The comments argued that minimum
standards would avoid confusion about what is enforceable, and what is
not. They pointed out that as regulations, these provisions could be
more readily enforced.
    FDA believes that all of these comments have merit. Guidelines can
provide flexibility that regulations sometimes lack. Moreover, because
they are advisory in nature, guidelines are less likely to be followed
by rote.
    FDA thus agrees that, ideally, HACCP should serve as a catalyst for
processors to develop a full understanding of the relationships between
their products and processes and human food safety and to devise
controls for ensuring safety. There may well be more than one way to
reach an appropriate safety endpoint. Regulations might not always take
such alternatives into account.
    On the other hand, in those cases involving high-risk products
where adherence to scientifically established minimum standards or
procedures is necessary to ensure a safe product by design, and those
minimums are not likely to change, there is good reason to make those
minimums something more than advisories. In those types of situations,
it makes no sense to act as if the work that scientifically established
the minimum processing conditions had not been done.
2. Cooked, Ready-To-Eat Products and Scombroid Species
    These, then, are the considerations that FDA has weighed. In the
case of cooked, ready-to-eat products and products made in whole or in
part from scombroid toxin-forming species, FDA is persuaded that the
guidelines should remain as guidelines, at least until there is enough
experience with them to determine whether a change to regulations is
warranted. The agency has concluded that processors should be given
maximum flexibility, at least initially, to identify the reasonably
likely hazards and the CCP's and CL's for those hazards that are most
appropriate for their manufacturing processes. FDA will examine over
time whether processors are achieving an adequate degree of preventive
control for these products under the guidelines, and whether they are
doing so by following the guidelines exactly or partially or by relying
on alternative approaches.
    FDA acknowledges that many comments objected to the details of the
appendices. These comments will be addressed when the first edition of
the Guide is published. FDA recognizes that these materials will be
more easily modified and improved if they remain as guidelines, at
least for the time being.
    FDA agrees that all of these guidelines should appear solely in the
Guide. There are no appendices to these final regulations.
3. Smoked and Smoke-Flavored Fishery Products
    The guidance for smoked and smoke-flavored fish contained specific
processing parameters (i.e., time and temperature of smoking and
finished product salt and nitrite concentrations) to be met in the
processing of such products, and control mechanisms for ensuring that
they are met. C. botulinum toxin production is prevented in smoked and
smoke-flavored fish by controlling these interrelated variables, as
well as by controlling the temperature of the product throughout the
chain of distribution.
    139. Approximately 25 comments addressed whether these materials
should be regulations or guidelines. About half of the comments,
representing State and Federal regulatory agencies, professional
associations, and others, urged that the materials be codified as
regulations. The remainder, representing processors and trade
associations, requested that the guidelines remain as guidelines.
    A number of the comments that urged that the smoked and smoke-
flavored fish guidelines be issued as regulations asserted that
regulations are more easily enforceable, would provide clear direction
to the industry, and would provide much needed nationwide uniformity in
the processing of smoked fish. One comment from a State regulatory
agency observed that processors are not adhering to existing
guidelines, such as the 1991 recommendations for these products by
AFDO, and are unlikely to change their operations in response to
another guideline. Several comments argued that the States need Federal
regulations to support their own efforts to regulate the industry and
to foster uniformity among the various existing State requirements. One
of these comments also stated that such regulations are needed to
ensure the safety of smoked fish, because the product has a history of
involvement in botulism outbreaks, is handled more than most other
products, increasing the risk of microbiological contamination, and is
frequently not cooked prior to consumption. One comment suggested that
the guidelines be tested in pilot programs before making them
mandatory, and that research information on smoked fish be disseminated
to industry through
 
  Page 65162
technical bulletins, workshops, and meetings.
    Several of the comments that suggested that the proposed guidelines
remain as guidelines argued that FDA has not demonstrated that present
practices in the smoked fish industry are causing risks that would
justify regulations, and that there have been no recent incidents of
botulism attributable to smoked fish. Several comments stated that most
of the problems with smoked fish in the past have resulted from abuse
of the product at retail or by the consumer.
    A few comments objected to FDA's contention that large portions of
the industry do not conduct final product testing and to the inference
that all smoked fish processors do not monitor the composition of their
products. The comments stated that responsible companies do conduct
product testing on a regular and routine schedule, have scheduled
processes, and are aware of what they are doing.
    Other comments recommended that FDA enforcement of the current
CGMP's, coupled with State and local enforcement of the Food Code for
smoked products that are produced in restaurants, retail, and food
service establishments, would make it unnecessary to treat smoked fish
products any differently than other products under these HACCP
regulations. One comment suggested that guidelines would have the same
impact as regulations because HACCP plans would be rejected by FDA if
they do not contain the recommended controls, and because States would
adopt the guidelines as regulations.
    One comment argued that the issuance of prescriptive regulations
would eliminate the diversity in the types of smoked fish products
available and result in a ``homogeneous'' market. Another comment
counseled that the issuance of a regulation would cause Alaskan native
salmon processors to abandon their traditional trade.
    The agency remains convinced that smoked and smoke-flavored fish is
a potentially hazardous food. While cases of botulism have not been
attributed to commercially prepared smoked or smoke-flavored fish in
over 30 years, the outbreaks of the 1960's clearly demonstrate the
potential for such occurrence. Virtually all the research that has been
conducted establish that processors need to control time, temperature,
and salinity (T-T-S) parameters and other matters for these products in
order to provide adequate barriers to toxin production (Ref. 214).
    As the preamble to the proposed regulations pointed out, FDA and a
number of States have longstanding concerns that the actions of a
significant portion of the smoked fish industry do not demonstrate a
full appreciation for the nature of the risks. FDA and New York State
surveys of the smoked fish industry in the late 1980's, for example,
showed that many processors did not routinely control their T-T-S
parameters.
    The comments have not persuaded FDA that, even without regulations,
processors will employ preventive controls to ensure the safety of
these products as a matter of design and not of chance. Botulism
derives from one of the most dangerous toxins known to exist. Controls
to prevent the formation of this toxin cannot be left to chance. HACCP
controls for this hazard are highly appropriate because HACCP requires
that the processor analyze its operation to determine how hazards
affecting its product can arise, and that it institute specific
controls to prevent those hazards. The majority of comments that
addressed smoked and smoke flavored fish products either supported the
concept of HACCP controls or did not argue against them.
    140. The question, therefore, is whether, in addition to requiring
HACCP plans for these products, FDA should mandate specific CCP's,
minimum CL's, monitoring frequencies, and other matters that processors
would have to include in their HACCP plans. If the agency were to
codify draft guidelines as regulations, the agency would be answering
that question in the affirmative. The preamble to the proposed
regulations identified the T-T-S parameters in the draft guidelines as
being scientifically established minimums for ensuring that toxin
produced by C. botulinum will not be produced over the shelf life of
the product under refrigerated conditions and under conditions of
moderate temperature abuse. FDA has been urged for years to mandate
such T-T-S parameters for these products. In 1988 and 1989, for
example, AFDO passed resolutions asking FDA to expedite the development
of regulations for the safe processing of smoked fish. The comments to
this rulemaking that supported regulations over guidelines support the
mandating of specific T-T-S parameters.
    However, a significant number of other comments challenged whether
some of the parameters in the guidelines were actually minimums, as FDA
had contended. They specifically objected to the minimum water-phase
salt levels in the draft guidelines for air packaged smoked and smoke-
flavored fish. Generally, these comments stated that there is little
safety concern with air-packaged smoked or smoke-flavored fish (hot or
cold smoked) containing as little as 2.5 percent water phase salt (less
than the minimum stated in the guidelines), and requested that FDA
reexamine the existing scientific data. A few comments stated that air-
packaged smoked fish has a limited shelf life in the refrigerated state
and that NMFS research has shown that spoilage occurs before toxin
production. One comment stated that NMFS, New York State Department of
Agriculture and Markets, and AFDO all consider a minimum water-phase
salt content of 2.5 percent to be acceptable for air-packaged products.
    A few comments suggested that an alternative to specifying T-T-S
parameters would be to require that all processors have a scheduled
process for air-packaged products. The comment stated that this
requirement has been successful in the State of New York and has
enabled industry to produce products with water-phase salt
concentrations that are lower than those proposed by FDA. A few
comments suggested that the high salt levels proposed by FDA for smoked
and smoke-flavored products would be counterproductive to those
government programs aimed at reducing salt in the human diet and would
be unacceptable, or only marginally acceptable, to consumers. Other
comments suggested that the necessary minimum salt levels for smoked
and smoke-flavored fish might be reduced by shortening the shelf life
of the product or by storing and distributing the product frozen.
    The comments have persuaded FDA that it may be possible for
processors to use parameters other than those in FDA's draft guidelines
and still produce a safe product. Moreover, the NACMCF has recently
endorsed AFDO's recommended parameters for smoked and smoke-flavored
fish. Most notably, these recommendations differ from those in FDA's
draft guidelines in that they provide for a minimum finished product
water phase salt content in air-packed product of 2.5 percent, whereas
the FDA proposal provided for a range of minimum values of from 2.5
percent to 3.5 percent, depending upon other processing parameters.
    The agency acknowledges, therefore, that some recommended T-T-S
parameters differ from those in FDA's draft guidelines. FDA
acknowledges the possibility that other safe T-T-S parameters exist as
well. It is reasonable to suppose that there is more to be learned
about how the development of C. botulinum toxin is controlled in these
products, given the lack of reported illnesses in recent years. Thus,
while
 
  Page 65163
FDA strongly believes that the T-T-S parameters in its draft guidelines
provide effective controls for botulism, the agency accepts that they
are not necessarily the only effective controls, or that all effective
controls have been identified.
    Consequently, the agency has concluded that, at least for now, the
most appropriate place for such guidance on T-T-S parameters and
related matters is the Guide, and that it would not be appropriate to
adopt specific parameters for the processing of smoked fish by
regulation. However, because of the extreme nature of the hazard, and
in response to comments, FDA has chosen to codify a rudimentary
performance standard for the control of botulism in these products from
the draft guidelines (item number 11). As incorporated at subpart B,
Sec. 123.16, the performance standard reads:
 
    In order to meet the requirements of subpart A of this part,
processors of smoked and smoke-flavored fishery products, except
those subject to the requirements of part 113 or 114 of this
chapter, shall include in their HACCP plans how they are controlling
the food safety hazard associated with the formation of toxin by C.
botulinum for at least as long as the shelf life of the product
under normal and moderate abuse conditions.
 
    This requirement responds in part to the comments that proposed
that FDA require that all processors scientifically establish scheduled
processes for smoked and smoke-flavored fish, rather that mandate
specific T-T-S parameters and other matters. It requires processors to
establish CL's that are both appropriate to their operation and
scientifically sound. Because botulism is undoubtedly a hazard that
must be controlled in the production of these products, subpart B of
part 123 does not impose a requirement that would not exist in its
absence. It has been included for emphasis and as a reminder to
processors. The Guide will provide processors with assistance with
regard to specific T-T-S parameters and related matters.
    141. FDA proposed that, with certain exceptions, fish to be smoked
or salted be eviscerated and free of residual viscera. The preamble to
the proposed regulations explained that salted or smoked uneviscerated
fish present a potential hazard for the development of C. botulinum
toxin production. The viscera of fish contain the majority of C.
botulinum normally present.
    One comment argued that the entire evisceration provision should be
moved to the Guide. Another comment suggested that specific
evisceration requirements be codified but not as part of sanitation.
    These regulations require in subpart B of part 123 that the
processors of smoked and smoke-flavored fish describe in their HACCP
plans how they are controlling the food safety hazard associated with
the formation of toxin by C. botulinum. Specific types of controls will
be provided in the Guide. Because evisceration is one form of control
for this toxin, it will be covered in the Guide as well and need not be
included in the regulations. Consequently, FDA has not included this
proposed provision in these final regulations.
 
N. Molluscan Shellfish
 
1. Background
    In addition to the general HACCP provisions in subpart A of part
123, FDA proposed subpart C of part 123--``Raw Molluscan Shellfish,''
which set forth specific requirements for the processing of fresh or
frozen molluscan shellfish. Proposed subpart C of part 123 described
certain types of controls that processors of these products must
include in their HACCP plans in order to meet the requirements of
subpart A of part 123.
    Specifically, FDA proposed to require that processors of raw
molluscan shellfish identify in their HACCP plans how they are
controlling the origin of the molluscan shellfish that they process.
FDA proposed to require that these controls include accepting only
molluscan shellfish that originated from growing waters that are
approved by a shellfish control authority, that are from harvesters
that are licensed or from processors that are certified by a shellfish
control authority, and that are properly tagged or labeled. In
addition, FDA proposed to require that processors maintain records to
document that each lot of raw molluscan shellfish meets these
requirements. FDA also proposed to amend Sec. 1240.60 (21 CFR 1240.60)
to provide for a system of tagging for shellstock and labeling for
shucked molluscan shellfish as a means of source identification.
    It is important to note that shellfish control authorities in the
United States are generally agencies of State governments, and that the
tagging of molluscan shellfish is an important aspect of State
shellfish control programs. As discussed below, reference to aspects of
existing State programs in these Federal regulations is not intended to
supplant or override the State programs in any way. Rather, these
provisions are intended to strengthen the Federal system in ways that
will complement, and thereby better support, State programs.
    Molluscan shellfish consumed raw or partially cooked pose unique
public health risks. As the preamble to the proposed regulations noted,
they probably cause the majority of all seafood-related illnesses in
the United States. This situation is not unexpected, given the nature
of the product and the way that it is consumed. The preamble documented
a relationship between the microbiological quality of molluscan
shellfish growing waters and the incidence of molluscan shellfish-borne
disease. It also noted that naturally occurring toxins may accumulate
in molluscan shellfish because they are filter-feeding animals.
    The NSSP was established as a cooperative program among FDA, State
regulatory agencies, and the molluscan shellfish industry, relying on
section 361 of the PHS Act (42 U.S.C. 264), to provide for the
classification and patrol of shellfish growing waters and the
inspection and certification of shellfish processors. The preamble to
the proposal reaffirmed FDA's support for the NSSP but noted the
difficulties that are associated with ensuring the safety of these
uncooked products. As the preamble stated, FDA tentatively determined
that it could strengthen and provide additional support for the
cooperative program through these regulations.
2. Should There Be Specific Requirements for Raw Molluscan Shellfish?
    FDA received approximately 45 comments about the proposed
requirements for raw molluscan shellfish. The responses were from
processors, trade associations, State and Federal government agencies,
individuals, consumer advocacy groups, and a foreign country.
Approximately half of these comments urged FDA to eliminate proposed
subpart C of part 123 and the proposed amendment to Sec. 1240.60, while
the other half acknowledged the advisability of including these kinds
of provisions but commented on, or questioned, various specifics of
them.
    The comments that generally supported the need for specific
requirements for raw molluscan shellfish were from trade associations,
molluscan shellfish industry members, consumer advocacy groups, Federal
and State government agencies, individuals, and a professional
organization. A number of comments noted that special requirements for
molluscan shellfish are warranted because of the association of these
products with illness. One
 
  Page 65164
comment in particular stated that, with respect to seafood, molluscan
shellfish ``serve as the primary source of illness due to ingestion.''
One comment noted that Federal regulations relating to source of origin
controls for raw molluscan shellfish would enable FDA to lend support
to the States in the administration of the NSSP. Another comment
suggested that the proposed regulations would improve FDA's regulatory
effectiveness with regard to molluscan shellfish control. The comment
from the ISSC stated that ``The Conference has long recognized and
supported expansion of FDA authority to assist States in assuring the
safety of molluscan shellfish.''
    The comments that suggested that subpart C of part 123 and the
amendment to Sec. 1240.60 be deleted were from State government
agencies and seafood processors. A number of the comments that
suggested deletion of the proposed provisions stated that the tagging
and labeling requirements of the NSSP are designed not to serve as a
control to prevent harvesting from closed areas but to assist States in
tracing shellfish that are implicated in illness outbreaks back to the
harvest area. The comments went on to state that harvesters who
illegally harvest from closed waters do not identify the shellfish as
originating from the closed area. The comments maintained that
preventing illegal harvesting is the key to reducing the incidence of
illness, and that the only known method to achieve this goal is through
effective law enforcement, including the patrol of closed waters.
    A number of these comments argued that increased FDA funding and
support for State molluscan shellfish control and patrol efforts would
do more than the proposed rule to deter illegal harvesting, to increase
States' compliance with the NSSP, and reduce the number of illnesses
caused by molluscan shellfish. The comments went on to state that the
proposed regulations unnecessarily duplicate the requirements now in
place in the Manual of Operations for the NSSP. They contended that
formal adoption of NSSP requirements into Federal regulations would
release State agencies from their cooperative relationship with FDA
under the NSSP.
    One comment noted that the weaknesses in State molluscan shellfish
control programs are in areas not addressed by the proposed
regulations, such as improperly classified growing waters; the ability
of State growing water classification programs to respond to breakdowns
at waste water treatment facilities or unexpected climatic events that
affect the quality of molluscan shellfish growing waters; and improper
handling by caterers and consumers. The comment concluded that the
proposed HACCP provisions for molluscan shellfish will, therefore, not
reduce the incidence of illness attributable to such products.
    As previously mentioned, FDA is a partner with State and foreign
regulatory authorities and with industry in the NSSP. The NSSP Manual
of Operations provides the standards for State and foreign molluscan
shellfish regulatory programs that belong to the cooperative program,
as well as for processors. The participating States routinely adopt
those standards as law or regulations, but the NSSP itself has neither
Federal nor State regulatory stature.
    Each participating State and foreign nation classifies and monitors
its molluscan shellfish growing waters, controls harvesting, inspects
molluscan shellfish processors, and issues certificates for those that
meet the shellfish control authority's criteria. FDA evaluates State
and foreign molluscan shellfish control programs and publishes monthly
the ``Interstate Certified Shellfish Shippers List,'' which lists the
molluscan shellfish processors that are certified under the cooperative
program. States that are in the program are not willing to receive
shellfish from noncertified shippers.
    FDA disagrees with the comments that suggest that establishment of
the proposed source controls in Federal regulations would supplant the
similar and, in some cases more stringent, requirements of
participating States and foreign nations or the standards set forth in
the NSSP. Rather, the agency is convinced that they will reinforce and
support these requirements and standards.
    The molluscan shellfish industry is subject to significant
regulatory oversight in those States that participate in the NSSP.
However, the quality and effectiveness of State laws and enforcement
activities can vary considerably as a function of the financial and
administrative support available to the responsible State units (Ref.
7, p. 15). For example, FDA documented discrepancies in State
enforcement practices during its 1994 evaluation of State programs to
determine compliance with the NSSP standards (Ref. 215). Moreover,
although all harvesting States participate in the NSSP, many other
States do not.
    Based on these factors, FDA proposed, and is now adopting, subpart
C of part 123 and amendments to Sec. 1240.60 to support and strengthen
the shellfish program in two ways. First, these provisions will
complement the efforts of the States. FDA recognizes that while States
are making significant and important efforts to ensure that all
shellfish harvested in their jurisdiction are taken only from open
waters and then properly tagged, some shellfish that do not meet these
requirements inevitably escape State control. The new provisions will
allow FDA to take action against shellfish that are not harvested from
open waters or that are not properly tagged if it encounters such
shellfish in interstate commerce and make the gravamen of such action
the origination from unopen waters or the lack of proper tagging
itself, rather than evidence that the shellfish are injurious to
health.
    Second, the regulations require that processors only use shellfish
that originate from growing waters that have been approved for
harvesting and that have been properly tagged. Failure to do so can
result in Federal regulatory action against the product or against the
processor itself. This fact should provide a significant incentive to
processors to ensure that they are not receiving shellfish that do not
meet these requirements.
    Taken as a whole, rather than diminishing in any way the importance
of State programs, FDA's regulations elevate the importance of those
programs. These regulations make proper origin and tagging--concepts
that derive directly from the NSSP--keys to the unimpeded movement of
shellfish in interstate, as well as intrastate, commerce.
    Moreover, these requirements extend these control measures to
imported products, enabling FDA to more efficiently and effectively
ensure the safety of imported raw molluscan shellfish. At present, the
agency must resort to advising State regulatory authorities of the
prospective entry of raw molluscan shellfish from an uncertified source
(Ref. 216, part V, p. 5). While States normally take action against
uncertified imported raw molluscan shellfish, FDA is aware that
uncertified imports enter interstate commerce (Ref. 107).
    FDA acknowledges that uniform Federal tagging and record-keeping
requirements will not completely eliminate illegal harvesting. The
agency agrees with the comments that rigorous enforcement of closed
area restrictions by State regulatory agencies will always be needed.
Unquestionably, increased funding would help State efforts to classify
and patrol growing areas. However, FDA does not have resources for this
purpose. Nonetheless, the agency remains convinced that there are
 
  Page 65165
significant, positive steps that can be taken to strengthen source
controls as part of HACCP, and thereby to support the cooperative
program.
    A processor's most significant safety control for raw molluscan
shellfish is at the point of receipt. If processors refuse to accept
molluscan shellfish for which there is no assurance that they have been
legally harvested, the incentive for illegal harvesting would be
eliminated. FDA participation in a number of covert investigations into
illegal molluscan shellfish harvesting in recent years has convinced
the agency that, in many cases, processors are aware of the illegal
harvesting activity of their suppliers (Ref. 217). If the provisions of
these regulations can help foster a culture change in that respect,
shellfish safety will be significantly enhanced.
    Based on these considerations, the agency proposed that, as a
universal aspect of the HACCP plans for these products, molluscan
shellfish processors engage in certain activities to ensure that the
products that they receive originate only from waters that have been
approved by a shellfish control authority (e.g., checking tags on
containers of shellstock, licenses of fishermen, and certification of
suppliers). Molluscan shellfish that are clearly improperly tagged or
from questionable sources must be rejected by processors as a
requirement of their HACCP plans. It is reasonable to conclude that, as
more processors adopt HACCP and exercise greater control over their
suppliers, the amount of illegally harvested shellfish offered for sale
will decrease, because the market for such product will decline.
    While it is true that the tagging requirements of the NSSP were
primarily designed as a means of tracing back molluscan shellfish
involved in incidences of illness to their harvest area, they have also
served as a key component in efforts by FDA and State regulators and
industry to ensure that molluscan shellfish that are placed in commerce
originate from areas that are approved by a shellfish control
authority. It is certainly true that the tags on containers of
molluscan shellfish that are harvested from closed waters are often
falsified to disguise their true origin. However, such falsification
carries potential Federal and State penalties and is a focus of current
molluscan shellfish control programs.
    Regarding the comments that pointed to weaknesses in State
programs, at retail, in the classification of molluscan shellfish
growing waters, and elsewhere, which are not directly addressed by
these regulations, the agency acknowledges that HACCP plans and
specific source control requirements cannot serve as a substitute for
improvements in the food safety system that directly address these
weaknesses. Regulatory systems will always have their strengths and
weaknesses, and research to better understand and control hazards will
always be needed. Nonetheless, these comments provide no reason for FDA
to abandon its efforts to remedy existing agency weaknesses and, in
particular, to lend support to the States in those areas to which these
regulations do relate.
    141. One comment stated that references cited in the preamble to
the proposed regulations in support of the tagging requirements (Refs.
102 through 109) do not provide convincing evidence of a need for such
a measure. The comment stated that, for the most part, the references
that FDA cited document corrective actions taken by State regulatory
agencies that would likely be the same measures that FDA officials
would take under the proposed regulations. In addition, the comment
stated that a failure to have properly tagged shellfish does not always
mean that the shellfish were harvested illegally. The comment pointed
out that the absence of a tag could mean simply that the tag was lost.
    The references in the question contain examples of problems
associated with molluscan shellfish tagging, recordkeeping, and
harvesting. FDA cited these references to demonstrate that, in some
cases, the deterrent effect of existing State tagging requirements and
sanctions is inadequate to prevent problems from arising (Refs. 102,
103, and 109). The problems documented in the references helped
persuade FDA to propose Federal source control requirements to help
deter the interstate shipment of shellfish from unapproved harvest
areas. FDA did not intend to imply that the State actions that were
documented in these references were incorrect, or that FDA would have
responded in a different manner. FDA continues to believe that the
references are relevant and supportive to its intended assertion.
    A few comments maintained that a better strategy for decreasing
illness from the consumption of molluscan shellfish would be to
increase the education efforts of FDA and of the ISSC that are directed
toward consumers and the medical community to alert susceptible
individuals to the risks associated with the consumption of raw
molluscan shellfish.
    The agency agrees that consumer education can play a vital role in
reducing illnesses associated with the consumption of raw molluscan
shellfish, particularly in medically compromised individuals. During
the period of 1984 through 1994, the agency expended nearly $1 million
to alert the public to the risks of raw molluscan shellfish consumption
by distributing brochures to consumer groups, groups that represent
those with special medical conditions, and consumers; developing a
video news release; issuing press releases; and establishing the toll-
free, FDA Seafood Hotline. Included in this expenditure is the agency's
efforts to inform the medical community about the illnesses associated
with the consumption of raw molluscan shellfish by providing
informative articles to medical bulletins and journals and by mailing
brochures and news articles to target professional groups. The agency
will continue its consumer education efforts, but such efforts alone
will be insufficient to address the hazards posed by the consumption of
raw molluscan shellfish harvested from unapproved growing waters. The
existing and planned consumer education efforts are geared toward
individuals in high-risk consumer groups, advising them to avoid
molluscan shellfish that have not been fully cooked. The risks posed by
viruses, toxins, and many bacteria are to the population as a whole.
There is little advice that the agency could provide that would enable
consumers to protect themselves from these kinds of hazards in
molluscan shellfish.
    143. Several comments questioned the validity of FDA's statement
that molluscan shellfish consumed raw or partially cooked pose unique
public health risks and probably cause the majority of all seafood-
related illnesses in the United States.
    The comments provided no data upon which to conclude that either
the NAS or FDA is wrong in this regard. FDA remains convinced that the
statements made in the preamble to the proposed regulations are valid,
and that the references support these statements.
3. Cooked Versus Raw Molluscan Shellfish
    144. Comments from a number of State agencies, trade associations,
seafood processors, and the ISSC objected to the use of the terms
``raw'' and ``fresh or frozen'' in the title of part 123 subpart C and
the text of the proposed regulations on shellfish. These comments were
concerned because these terms would have the effect of exempting canned
and any other heat- processed molluscan shellfish from the source
control, recordkeeping, and
 
  Page 65166
tagging provisions of subpart C of part 123 and Sec. 1240.60(b).
    The comments stated that limiting these provisions to raw products
would allow foreign firms to continue to heat-treat or can molluscan
shellfish that are harvested from foreign waters that do not meet NSSP
standards and to export them to the United States. The comments stated
that this situation was not in the best interest of the public health
because of the potential for the presence of heat-stable natural
toxins, such as paralytic shellfish poison or amnesiac shellfish
poison, as well as chemical contaminants. The comments also complained
that, because State laws and regulations require that all molluscan
shellfish harvested in the United States come from waters approved by a
shellfish control authority regardless of whether they are to be
consumed raw or cooked, continuing to allow foreign processors who
export cooked shellfish to the United States to use molluscan shellfish
from unapproved growing waters places the domestic shellfish industry
at a competitive disadvantage. Other comments requested that FDA
clarify whether canned shellfish were included in subpart C of part 123
but did not suggest that canned and other heat-processed shellfish be
included.
    FDA has responded to these comments generally in response to
comment 34, supra. The agency adds the following points:
    It is important to recognize that foreign processors who export
cooked molluscan shellfish to the United States now will have to have
HACCP systems through which they identify and control hazards that are
reasonably likely to occur. These hazards include heat stable toxins
and chemical contaminants that would cause these products to be
adulterated under U.S. law.
    To further clarify that the requirements of subpart C of part 123
apply only to the processing of molluscan shellfish that are not heat
treated or treated in some other manner by the processor to eliminate
microorganisms of public health concern, FDA has modified the language
at Sec. 123.20 to read, ``This subpart augments subpart A of this part
by setting forth specific requirements for processing fresh or frozen
molluscan shellfish, where such processing does not include a treatment
that ensures the destruction of vegetative cells of microorganisms of
public health concern.''
4. Shellfish Control Authorities
    FDA proposed to require that processors only process molluscan
shellfish that originate from waters approved for harvesting by a
shellfish control authority. The term ``shellfish control authority''
is defined at Sec. 123.3(o) to include foreign government health
authorities that are legally responsible for the administration of a
program that includes classification of molluscan shellfish growing
areas.
    145. Two trade associations questioned how a processor could
evaluate the competency of a foreign shellfish control authority. They
stated that FDA should require that a foreign country that exports
shellfish to the United States have an agreement with the agency that
establishes that a competent shellfish control authority exists in that
country, and that the foreign shellfish program meets NSSP standards.
One comment from a seafood processor argued that it would be
unreasonable to require processors to verify that molluscan shellfish
from all over the world are caught or cultivated in waters that meet
NSSP standards. The comment stated, moreover, that a processor could
not keep abreast of which countries have current shellfish agreements
with FDA and which countries do not.
    FDA acknowledges the merits of requiring that a foreign country
that exports shellfish to the United States have an agreement with the
agency but has concluded that, given the significance of such a
requirement and the agency's failure to raise the possibility of
imposing it in the proposal, it is beyond the scope of this rulemaking.
Even though FDA is not imposing such a requirement, it is the case that
the only means by which a processor can ensure that the molluscan
shellfish of foreign origin that it receives are in compliance with the
requirements of subpart C of part 123 of these regulations is by
determining whether the foreign shellfish control authority is formally
recognized by FDA. It is not likely that the processor could employ any
other process that would give it assurance that molluscan shellfish
harvesting waters that are approved by the shellfish control authority
are properly classified. Such a determination is appropriately
performed through government to government audit.
5. Shellfish From Federal Waters
    146. Comments from a significant number of trade associations and
seafood processors stated that a requirement that shellfish originate
only in waters ``approved for harvesting by a shellfish control
authority'' would preclude harvesting in Federal waters unless the
Federal government introduced a formal approval process for waters
under its purview through a Federal shellfish control authority.
    Under the current system, State agencies are responsible for
approving molluscan shellfish growing waters. However, State
jurisdiction extends only to waters that are within three miles of the
shore. Waters beyond that point but up to 200 miles offshore are under
the jurisdiction of the Federal government. The comments pointed out
that the harvesting of molluscan shellfish is permitted in all of the
oceanic waters under Federal control unless there is a specific Federal
action to declare an area unsafe under the provisions of the Magnuson
Fishery Conservation and Management Act. The comments further noted
that large volumes of molluscan shellfish are harvested in Federal
waters.
    How Federal waters will be classified, and by whom, has not been
fully resolved. The comments are correct that the proposed requirement,
if incorporated into the final rule, would pose significant problems
for molluscan shellfish processors who receive product harvested from
Federal waters. Therefore, FDA has modified Sec. 123.28(b) to allow for
the receipt of molluscan shellfish that are harvested in U.S. Federal
waters except where such waters are specifically closed to harvesting
by an agency of the Federal government. This provision is consistent
with the provisions of the Magnuson Act.
    It is worth noting that, by allowing Federal waters to be open
unless they are specifically closed, this system is the opposite of the
State system, under which waters are closed unless they are
affirmatively classified so as to be open. This difference is
reasonable from a public health standpoint because there is less
likelihood that Federal waters will be affected by pollution than will
near shore State waters.
    Furthermore, because there is no Federal authority to license
shellfish harvesters who fish in Federal waters, FDA has modified
Sec. 123.28(c) to require only that a harvester be in compliance with
such licensure requirements as may apply to the harvesting of molluscan
shellfish, rather than specifically requiring licensure.
6. Tagging and Recordkeeping Requirements
    147. FDA proposed recordkeeping requirements for processors to
follow with respect to shellstock and shucked molluscan shellfish in
Sec. 123.28 and requirements for the information to be included on the
shellstock tag in Sec. 1240.60. A few comments stated that
 
  Page 65167
the proposed molluscan shellfish tag and record requirements were too
specific, and that placing such requirements in the form of regulations
would make it difficult to make timely changes to these requirements as
future needs may dictate. The comments asserted that FDA or the ISSC
may wish to modify the content or form of molluscan shellfish tags or
records to improve product traceability. They suggested that FDA write
the tagging and recordkeeping requirements at Sec. 123.28 and
Sec. 1240.60 in general terms and allow the specific form and
information required on the tags to be addressed by the NSSP. The
comments went on to urge that, if the agency were determined to include
specific tagging and recordkeeping requirements as part of the final
regulations, they should be fully consistent with current NSSP
guidelines.
    It is not the agency's intent that the provisions of Sec. 123.28
and Sec. 1240.60 would preclude the ISSC or State agencies from
adopting additional recordkeeping or tagging requirements. The
recordkeeping and tagging requirements in these regulations are the
minimum necessary to ensure that shellfish can be traced through
distribution channels, back to the harvest source. FDA explained why
each of the specific requirements is necessary in the preamble to the
proposed regulations, and the comments did not take issue with the
agency's explanation with respect to any of the particulars. Therefore,
FDA disagrees with the comments that the recordkeeping or tagging
requirements are more restrictive than they need to be, or that they
would interfere with the NSSP tagging program.
    Moreover, as stated previously, the agency has drafted the
regulations broadly enough so as not to conflict with any improvements
that the ISSC may wish to make in the form that a tag may take or in
how information on tags is to be stored. The definition of the word
``tag'' at Sec. 123.3(t) (added at Sec. 1240.3(u) for consistency)
reads, in part, ``a record of harvesting information attached to a
container of shellstock * * *.'' This definition is sufficiently broad
to include such systems as bar codes, embossed plastic, or other
nontraditional methods of identification that may be used by the
industry in the future. The agency is supportive of efforts to improve
the existing methods of recording harvesting information.
    Nonetheless, it is important for the regulations to identify the
minimum specific information that must appear on a tag. During past
illness outbreaks, FDA, State regulatory agencies, and industry have
had difficulty tracing the implicated shellfish to their sources,
especially after they have been in the possession of several different
processors (Refs. 99; 100; 102-106; 109; 218; and 219, pp. 37-39).
These difficulties in tracing the shellfish have occurred because the
shellfish were not in compliance with the tagging and recordkeeping
provisions of the NSSP. The requirements at Sec. 123.28 will enable FDA
to help the States to enforce minimum tagging and recordkeeping
requirements for imported and domestic products. Moreover, the agency
believes that placing the tagging and recordkeeping requirements in
Federal regulations will emphasize the nationwide importance of this
information in protecting the public health, as described earlier.
    148. One comment noted that the NSSP does not specify that the name
of the harvester must be on a molluscan shellfish tag, but that the
proposed regulations would require this information.
    The NSSP specifies that the number assigned to the harvester by the
shellfish control authority must be listed on the tag. The agency
recognizes that there may be a variety of effective ways to identify
the harvester of the molluscan shellfish, depending on the method of
harvest, State requirements, and local tradition. For this reason, the
agency has modified Sec. 1240.60(b) to read that the tag shall
disclose:
 
* * * by whom they were harvested (i.e., the identification number
assigned to the harvester by the shellfish control authority or, if
such identification numbers are not assigned, the name of the harvester
or the name or registration number of the harvester's vessel).
 
For consistency, FDA has made a similar change in Sec. 123.28(c)(5).
     149. A significant number of comments recommended that FDA modify
Sec. 1240.60(b) to allow bills of lading or other shipping documents to
accompany bulk shipments of shellstock in lieu of tags, as long as they
provide the same information. A few comments suggested that bills of
lading or other shipping documents be used in lieu of tags on
individual containers of shellstock when a shipment consists of a large
volume of shellstock in sacks or boxes. Several comments asked for
clarification of the impact of the proposed requirements on current
repacking operations that commingle shellstock from various harvesters
into one container.
    FDA recognizes that an inconsistency existed between proposed
Sec. 123.28 and proposed Sec. 1240.60 because the former would have
allowed shipping documents to provide the required information for bulk
shipment, and the latter would not. FDA agrees with the comments that
recommended providing for the use of shipping documents and has
modified Sec. 1240.60(b) to provide the needed consistency. Under
existing industry practice the truck, cage, or vessel hold serve the
same purpose as a container for the shellstock, making tagging
impractical. In that case, the shipping document serves the same
function as the tag.
    However, the agency does not agree with the suggestion that
containers of shellstock in large shipments be allowed to be covered by
shipping documents in lieu of tags. FDA cannot justify treating
shellstock in large shipments differently than shellstock in smaller
shipments, nor could the terms ``large'' or ``small'' be readily
defined. Large shipments can be subdivided, perhaps many times, or
commingled with other lots of molluscan shellfish. The source
information would, therefore, be necessary on each container to ensure
proper identification. Without tags, the identity of individual
containers could be lost. FDA is requiring that all shellstock, even
after repacking, bear a tag that identifies the prescribed information,
including the identification of the harvesters to ensure that all
shellstock is readily traceable (Sec. 1240.60).
7. Other Considerations
    150. Comments from a few trade associations and from seafood
processors stated that FDA should require a production code on each
container of shucked molluscan shellfish. The comments suggested that
the code consist of an identifying mark that allows the processor to
determine where the remainder of the lot was shipped, and where and
when the relevant shellstock was harvested.
    FDA agrees that production codes can be useful on containers of
shucked molluscan shellfish to facilitate trace back of questionable
product. The agency encourages the use of codes by molluscan shellfish
processors. However, such a requirement is not within the scope of the
proposed regulations. The agency will consider whether such a
requirement should be pursued in a separate rulemaking.
    151. Comments from several consumer groups stated that if a warning
label is not mandated by FDA on raw molluscan shellfish to alert at-
risk consumers of the danger to health posed by the product, FDA should
require that Gulf Coast oyster processors adequately cook the product
to
 
  Page 65168
eliminate risks from Vibrio vulnificus during periods when shellfish
cannot be harvested free from this bacterium. They further stated that
cooking should not be required when the shellfish are free from this
bacterium.
    FDA agrees that effective controls are needed to protect consumers
from the hazard posed by V. vulnificus in Gulf Coast oysters during
certain times of the year. The agency is evaluating the potential
effectiveness of a variety of control mechanisms. Mandating specific
mechanisms, however, would be outside the scope of this rulemaking. FDA
is therefore taking these comments under advisement.
    152. A few comments urged that the word ``processor'' be changed to
``certified dealer and licensed harvester'' throughout Sec. 123.28 to
make the terminology consistent with the NSSP and to clarify that these
requirements apply to everyone who buys and sells shellfish before the
shellfish reach the retail marketplace. The comments also recommended
changing the word ``shipper'' to ``processor or packer'' in the
provision that is codified at Sec. 123.28(d)(3) in these final
regulations to include the shucker, repacker, shipper, reshipper, or
depurator.
    As mentioned in the ``Definitions'' section of this preamble, the
agency has concluded that the definition for ``processor'' covers all
NSSP classifications of shellfish dealers, without specifically naming
each one. For consistency throughout the regulations, use of the term
will remain unchanged.
    FDA agrees, however, that the word ``shipper,'' as the agency used
it in the proposed regulations, could cause confusion because that term
has a different meaning in the NSSP Manual of Operations. Therefore,
FDA has changed the language of the final regulations to read ``packer
or repacker.'' The certification number of the packer or repacker is
readily available to the processor since it is required, under the NSSP
standards, on each label of shucked product. For consistency, FDA has
modified Sec. 1240.60(c) to also read ``packer or repacker'' where it
had previously read ``processor.''
    153. A few comments stated that proposed Sec. 123.28(b), which
referred to molluscan shellfish that ``originate from growing areas
that are approved for harvesting,'' appears to prohibit processors from
using products from harvest waters classified under the NSSP as
``conditionally restricted.'' Under the NSSP, shellstock taken from
such waters cannot be directly marketed for human consumption but can
be ``relayed'' to an open growing area for harvest at a later date. In
addition, the comments pointed out that the word ``approved'' is a
formal classification designation used in the NSSP. These comments said
that harvesting is permitted from areas with other than ``approved''
classifications but would appear to be prohibited under the provisions
of proposed Sec. 123.28(b). The comments suggested use of the term
``open'' growing waters, which would apply to several different NSSP
harvest area classifications.
    FDA agrees that the word ``originate'', as used in proposed
Sec. 123.28(b), is too restrictive and has revised this section to say
``harvested from.'' With regard to the word ``approved,'' FDA concludes
that no change is warranted. The proposal neither adopted nor
referenced the NSSP growing water classification system. The phrase
``approved for harvesting'' should not be construed to be limited to
those waters classified by a State under the NSSP as ``approved''
areas. The phrase ``approved for harvesting'' is used in the final
regulations for its common sense meaning (i.e., those areas from which
harvesting is allowed), which FDA believes is appropriate.
    154. For clarification purposes, definitions for the terms
``certification number'', ``shellfish control authority'', and ``tag''
have been added at Sec. 1240.3(s), (t), and (u), respectively. These
definitions are taken directly from Sec. 123.3.
    155. One comment urged that the regulations be modified to
specifically state that a HACCP plan for raw molluscan shellfish that
contains the controls specified in subpart C of part 123 is deemed to
meet the requirements of Sec. 123.6.
    The agency disagrees with this comment. The requirements of subpart
C of part 123 are designed to control the environmental hazards
associated with the harvest waters from which molluscan shellfish are
taken, particularly those relating to sewage- related pathogens,
chemical contaminants, and natural toxins. For this reason, they must
be included in the HACCP plans of all raw molluscan shellfish
processors. However, they are not intended to control process-related
hazards resulting from, for example, time-temperature abuse of the
product, improper use of food additives, or metal fragments. To control
these hazards, the processor needs to follow the general approach set
out in subpart A of part 123. The agency has developed the two subparts
to be complementary and has strived to eliminate any redundancy in
their provisions. Thus, it is theoretically possible that a HACCP plan
that contains the controls specified in subpart C alone of part 123
still might not meet all the requirements of Sec. 123.6.
    FDA has made two modifications in Sec. 1240.60(b) for clarity only.
Where the proposed regulations required that the tag identify the ``* *
* place where harvested * * *,'' FDA has added, ``(by State and
site).'' This change makes Sec. 1240.60(b) consistent with
Sec. 123.28(c)(2). Additionally, where the proposed regulations stated
that improperly tagged or labeled product would be ``subject to seizure
and destruction,'' FDA has amended the language to read, ``subject to
seizure or refusal of entry, and destruction.'' This change is to make
clear that, for imported products, the appropriate regulatory procedure
is refusal of entry, rather than seizure.
 
O. Compliance and Effective Date
 
1. Effective Date
    FDA proposed that these final regulations be effective and enforced
1 year after the date that they are published in the Federal Register.
The purpose of this proposed effective date was to provide processors
with enough time to develop and implement HACCP plans. The agency
invited comment on whether 1 year would be adequate.
    156. FDA received more than 60 comments about the proposed
effective date. Virtually all comments agreed that the agency should
provide some period before the regulations become effective. The
comments either agreed with a 1-year implementation period or requested
a longer period of 2 years or more. There were also a number of
comments that responded to the agency's question about whether
implementation dates should be staggered based on such factors as size
of firm and level of risk.
    A minority of comments stated that 1 year for implementation is
adequate. These comments argued that after 1 year, the industry would
have had 3-years notice of the requirements. The comments argued that 3
years was sufficient total time to be informed about impending
regulations. Another comment stated that one year might be sufficient
for the seafood industry, but other food industries could need
considerably more time.
    Several comments recommended that FDA provide an implementation
period of longer than 1 year but did not recommend a specific
alternative. These comments were concerned that HACCP training would
not be completed in time for a 1-year implementation date; that foreign
processors would need more
 
  Page 65169
time to implement HACCP; and that 1 year after Canada required HACCP
for its seafood industry, only half of its firms had complied.
    The largest number of comments on this topic recommended that FDA
make the regulations effective 2 years after publication. The reason
most often cited was that it will be more than 1 year before most of
the affected firms can complete HACCP training. The next most
frequently cited reason was that firms and trade associations needed
more time for HACCP plan development. Several comments also discussed
the time required to modify equipment and raise capital; to respond to
initial, voluntary reviews of HACCP plans by regulatory agencies; and
for Federal, State, and local officials to understand HACCP and how to
enforce it and to arrange for cooperative enforcement. A few comments
stated that FDA needs sufficient time to develop agreements with
foreign countries.
    Several comments contended that more than 2 years should be allowed
to implement the regulation. These comments mentioned the cultural
change that HACCP will require and concern about the impact that the
regulations will have on small firms as well as on large firms with
multiple products and lines. They also mentioned the time needed for
training.
    Over 20 comments recommended that FDA stagger effective dates. A
majority of these comments stated that such a phased-in start-up should
be based on product risk. The remainder of the comments split in favor
of either considering both the size of a firm and the risk from the
products it makes or just the size of a firm.
    A smaller number of comments argued against a staggered start.
These comments expressed the view that small firms and foreign products
should not be treated differently and pointed out that all firms will
already have had 3 years of notice. Some of these comments stated that
it would be hard to justify staggering implementation based on risk
when the illness data are so incomplete. Others expressed the view that
administering a staggered start would use up valuable resources and
only result in confusion; that staggering would put some firms at a
competitive disadvantage; and that it might encourage pro